Thalassemia Intermedia

Proceedings from a Conference on Thalassemia IntermediaSponsored by the New England Thalassemia Program

Boston, MA November 14, 1996

THE GENETIC RESOURCE: SPECIAL ISSUE, 1997VOLUME 11, NUMBER 2

Thalassemia IntermediaA R e g i o n I C o n f e r e n c e

Proceedings from a Conference on Thalassemia Intermedia Sponsored by the New England Thalassemia Program

Boston, MANovember 14, 1996

Edited by: Howard A. Pearson, M.D., Lauren C. Berman, M.S.W., Allen C. Crocker, M.D.

THE GENETIC RESOURCE: SPECIAL ISSUE, 1997VOLUME 11, NUMBER 2

Sponsored by: New England Thalassemia Program

With support from: Maternal and Child Health Bureau (grant number MCJ-251005)New England Regional Genetics Group (grant number MCJ-251003)Connecticut Campaign Against Cooley’s Anemia

Published by: National Center for Education in Maternal and Child Health, Arlington, VA

Thalassemia IntermediaA R e g i o n I C o n f e r e n c e

Cite as

Pearson HA, Berman LC, Crocker AC, eds. 1997. Thalassemia Intermedia: A Region I Conference. Arlington, VA:National Center for Education in Maternal and Child Health.

The New England Thalassemia Program, established in 1982, is a collaboration of thalassemia treatment centers atsix major hospitals in New England. The program assures optimal care for patients and families affected by or atrisk for thalassemia. We offer a network of thalassemia services that include education, diagnosis, culturally sensi-tive screening, genetic counseling, comprehensive medical care, and psychosocial support. The hospital centers arebased at Children’s Hospital in Boston, Bay State Medical Center in Springfield, Rhode Island Hasbro Children’sHospital, New England Medical Center, the University of Massachusetts Medical Center in Worcester and YaleNew Haven Hospital.

We are directly affiliated with the New England Regional Genetics Group, one of 10 regional networks of theCouncil of Regional Networks of Genetic Services. We also work closely with both the Asian ThalassemiaScreening Project at the South Cove Community Health Center in Boston and the Joint Center for Sickle Celland Thalassemic Disorders at the Brigham and Women’s Hospital.

The National Center for Education in Maternal and Child Health (NCEMCH) is dedicated to improving thehealth of children and families through providing national leadership in three key areas of maternal and childhealth: information services and systems, continuing education, and policy analysis and development. NCEMCH’smultidisciplinary faculty and staff—experts in such diverse fields as pediatrics, public health, law, public policy,social work, nutrition, library science, communications, publishing, conference management, and systems tech-nology—make it especially effective in developing comprehensive program initiatives. Established in 1982 atGeorgetown University, NCEMCH is part of the Georgetown Public Policy Institute. NCEMCH is funded primarily by the U.S. Department of Health and Human Services through its Maternal and Child Health Bureau.For more information or a free publications catalog, contact: NCEMCH, 2000 15th Street North, Suite 701,Arlington, VA 22201-2617. Phone: (703) 524-7802. Fax: (703) 524-9335. World Wide Web: http://www.ncemch.org. Internet: [email protected].

ISBN 1-57285-041-8. Illustration on cover and interior reprinted by permission of Oxford University Press.

For further information or additional copies of this booklet contact:

Lauren C. Berman, M.S.W.Director, New England Thalassemia ProgramI.C.I. - Judge Baker 2Children’s Hospital300 Longwood AvenueBoston, MA 02115Phone: (617) 355-5943Fax: (617) 355-7940

This publication was made possible through grant number MCJ-251005 from the Maternal and Child HealthBureau, Health Resources and Services Administration, Public Health Service, U.S. Department of Health andHuman Services. Additional funding was provided by the New England Regional Genetics Group, grant numberMCJ-251003.

PrefaceLauren C. Berman, M.S.W. ...............................................................................................................................v

Conference Planning Committee ...........................................................................................................................vi

IntroductionEdwin N. Forman, M.D. ...................................................................................................................................1

Thalassemia Research Activities of the National Heart, Lung, and Blood Institute, NIHAlan S. Levine, Ph.D. ........................................................................................................................................3

The Evaluation of Thalassemia IntermediaHoward A. Pearson, M.D. .................................................................................................................................5

Beta-Thalassemia Intermedia in the United States: A Multicenter StudyPatricia J. Giardina, M.D., Igal Fligman, M.D., Alan R. Cohen, M.D., Haig H. Kazazian, M.D., BeatriceGee, M.D., Edwin N. Forman, M.D., Kenneth R. Bridges, M.D., Howard A. Pearson, M.D. ......................11

Iron Overload and Iron-Chelating Therapy in Thalassemia IntermediaNancy F. Olivieri, M.D. ...................................................................................................................................15

Hemoglobin Switching Protocols in Thalassemia: Experience with Sodium Phenylbutyrate and HydroxyureaGeorge F. Dover, M.D. ....................................................................................................................................21

The Molecular Basis of Thalassemia IntermediaHaig H. Kazazian, Jr., M.D., and Catherine A. Stolle, Ph.D. ..........................................................................27

Questions and Comments .....................................................................................................................................29

Infections in Thalassemia IntermediaAlan R. Cohen, M.D. ......................................................................................................................................31

Panel Discussion ....................................................................................................................................................35

EpilogueAllen C. Crocker, M.D. ...................................................................................................................................43

Conference Program ...............................................................................................................................................45

Conference Participants..........................................................................................................................................47

Table of Contents

Thalassemia intermedia is one of the most

challenging and controversial hematologic

disorders. This report was assembled from

discussions held at a conference in Boston,

Massachusetts, on November 14, 1996. The confer-

ence brought together leaders and researchers in the

fields of hematology and genetics. Presenters included

such well-known experts as Drs. Howard Pearson

(conference chair), Alan Cohen, George Dover, Ed

Forman, Patricia Giardina, Carol Hyman, Haig

Kazazian, and Nancy Olivieri.

We offer this material in the hope that our expe-

rience will stimulate further thinking about the clini-

cal spectrum of thalassemia intermedia and the

genetic corrolations. In these proceedings, you will

find a history and definition of thalassemia interme-

dia, a compilation of data from patients in six north-

eastern medical institutions, a study of the use of

sodium phenylbutyrate and hydroxyurea to stimulate

fetal hemoglobin, and an analysis of the molecular

basis of thalassemia intermedia. There are also pre-

sentations on the assessment of iron overload, the use

of chelation therapy, splenectomy, and infection con-

trol. We close with a lively panel that explores treat-

ment approaches and future questions about the role

of fetal hemoglobin enhancement and bone marrow

transplant.

As the social worker at the Boston Children’s

Hospital thalassemia program, I have developed a

great respect for patients and their families. Over the

past 12 years, I have been impressed by the strength,

courage, and perseverance of those who cope with the

challenges of this illness. I have seen many patients

confront their worst fears and grow to become confi-

dent and healthy adults. Because thalassemia interme-

dia is less common and further complicated by uncer-

tainty and isolation, I am pleased we could devote an

entire conference to this topic. With this project, we

capture the energy and spirit of the New England

Thalassemia Program. I would like to thank the pre-

senters and the planning committee, whose innovative

ideas were the foundation of this event. It is an honor

to work with so many creative and dedicated profes-

sionals.

This conference and publication would not have

been possible without the support of the Maternal

and Child Health Bureau, the New England Regional

Genetics Group, and the Connecticut Campaign

Against Cooley’s Anemia. We are pleased to make

these proceedings available to both the public and

professionals.

Lauren C. Berman, M.S.W.Director New England Thalassemia Program

Preface

THALASSEMIA INTERMEDIA: A REGION I CONFERENCE v

Conference Planning Committee

Howard A. Pearson, M.D., ChairYale-New Haven Hospital, Yale, CT

Lauren C. Berman, M.S.W.Children’s Hospital, Boston, MA

Kenneth R. Bridges, M.D.Brigham & Women’s Hospital, Boston, MA

Allen C. Crocker, M.D.Children’s Hospital, Boston, MA

Edwin N. Forman, M.D.Rhode Island Hospital, Providence, RI

Beatrice Gee, M.D.Children’s Hospital, Boston, MA

Jeanne Harvey, R.N.University of Massachusetts Medical Center, Worcester, MA

Susan Kurth, R.N.Children’s Hospital, Boston, MA

Margaret Lee, M.D.South Cove Community Health Center, Boston, MA

Betty Leef, R.N.Rhode Island Hospital, Providence, RI

Marge MacDowell, M.S.W.Rhode Island Hospital, Providence, RI

Kathy Marson, R.N.New England Medical Center, Boston, MA

Linda Rink, R.N.Yale-New Haven Hospital, New Haven, CT

Molly Schwenn, M.D.University of Massachusetts Medical Center, Worcester, MA

Joyce YinSouth Cove Community Health Center, Boston, MA

vi THALASSEMIA INTERMEDIA: A REGION I CONFERENCE

Good morning and welcome to the NewEngland Thalassemia Program’s second con-ference on thalassemia intermedia. I want to

take this moment to thank all those who have madethis meeting possible: the speakers, members of theNew England Thalassemia Program, and especiallyLauren Berman, our director, and Dr. HowardPearson, our valued and consistent mentor, whobrought this conference into being.

The first thalassemia intermedia conference washeld 11 years ago. At its conclusion, we, like Socrates,achieved the wisdom of understanding how little weknew. We knew neither how to predict the naturalhistory of a given patient nor how to manage thepatient, nor even how to define this condition (orthese conditions). Now, at this second conference,with our panel of esteemed experts, it might not beunrealistic to hope that we can move further towardanswering the following questions:

How can we define thalassemia intermedia,perhaps even at a molecular level, such that wecan increase understanding of the pathophysi-ology and enable accurate prediction of thenatural history of the condition for a givenpatient?

How should a patient needing a splenectomy,transfusions, chelation, growth and/or sex hor-monal therapy, stimulation of fetal hemoglo-bin, bone marrow transplantation, etc., bemanaged?

To start us off, I will give a brief followup reporton the original two children who were the focus ofthe first thalassemia intermedia conference. The pho-

tographs of the children are used with their permis-sion and the permission of their family.

Child 1

Child 1 (C1) was noted to be anemic at about 14months of age, and, upon evaluation, was diagnosedwith thalassemia intermedia, probably doubly heterozy-gous. Aside from dietary recommendations to avoidiron-rich foods and drink tea, no other therapy wasgiven. Over the next few years, she developed progres-sive splenomegaly and mild thalassemic facial features.Her hemoglobin dropped to 5 g%. She received onetransfusion. At age 6, a splenectomy was performed,and thereafter her hemoglobin level has averaged 9 g%.She has not received any further transfusions and hasnever received chelation therapy. She has short stature,like her family, but has reached a height of 5 feet, 1inch, and achievedpuberty, with menarcheat 15 years of age.Aside from a possiblestress fracture (shinsplints in her righttibia), a suspicious testfor hepatitis C at onetime (not confirmed),pneumonia at age 16,and sinusitis at age 20,she has had no seriouscomplications and hasno evidence of diabetesor gallstones. She iscurrently 22, has been employed, and is now back incollege studying for a degree in nursing. Currently, herserum iron is 83 mg%, her total iron binding capacity is222 mg%, and her ferritin is 152 mg%. A cardiac evalu-

Introduction

Edwin N. Forman, M.D.

THALASSEMIA INTERMEDIA: A REGION I CONFERENCE 1

ation reveals mild cardiomegaly but a normal echocar-diogram and electrocardiogram. As you can see from herphotograph, she is an attractive young woman who has anactive social life and a cheerful disposition.

Child 2

Child 2 (C2), brother of C1, was diagnosed withβ-thalassemia intermedia at age 12 months, and theprogression of his condition is roughly similar to that

of his sister. He devel-oped a worsening ane-mia, down to 5 g% bythe time he was 4 yearsof age, and progressivesplenomegaly and mildthalassemic facies. Hereceived three bloodtransfusions at aroundage 4 years withoutsignificantly affectinghis hemoglobin level.A splenectomy wasperformed at age 4-1/2, and since that

time his hemoglobin level has averaged 8.5 g%.Although he has developed secondary male sexualcharacteristics (at age 15 years), he was quite short inhis preadolescent years—below the 5th percentile—and was put on human growth hormone with a goodresponse (currently in the 25th percentile). He has hadone bone fracture (in the wrist at age 11 years) and hasundergone cholecystectomy (at 18 years of age) fortroublesome gallstones. Aside from viral meningitis (4years of age) and salmonella gastroenteritis (age 9years), he has not suffered any unusual infections. Hedoes have bony protuberances at his elbows, which areprobably secondary to marrow expansion. In 1994, inassociation with a flu-like syndrome, his hematocritdropped to 21. He felt severely fatigued, and hereceived one transfusion. Currently (age 19 years), heis hepatitis C and HIV negative, and his serum iron is218 mg%, his total iron binding capacity is 244 mg%,and his ferritin is 176 mg%. A cardiac evaluationreveals mild cardiomegaly, but his echocardiogram andelectrocardiogram are normal.

As you can see from his photograph, he is a nice-looking, slender but muscular young man with a friend-ly and cheerful demeanor. He is employed full time,working for his father doing manual labor.

With those case presentations, let us begin theprogram.

Our first speaker is Dr. Alan Levine, director of theBlood Diseases Program at the National Heart, Lung,and Blood Institute, who will give us an update on theactivities of the institute with regard to thalassemia.

2 THALASSEMIA INTERMEDIA: A REGION I CONFERENCE

At the National Institutes of Health (NIH),hematology research is supported by approxi-mately 12 separate institutes. In fiscal year

1995,1 funding for such research totaled over $500million. For nonmalignant hematologic diseases,however, the National Heart, Lung, and BloodInstitute (NHLBI) supported the vast majority ofhematology research (approximately $263 million).The relative institute expenditures for 1996 areexpected to be much the same.

NIH defines an orphan or rare disease as having aprevalence in the United States of fewer than 200,000patients. β-Thalassemia, with fewer than 1,000patients, is perhaps the “orphan” of orphan diseases.However, NHLBI by no means treats thalassemia asan orphan. NHLBI support for thalassemia researchhas almost doubled since 1988, to an estimated $14.7million in 1996. This amount does not include thesubstantial support for research in areas that are notspecific to thalassemia but nevertheless are extremelyimportant, such as unrelated marrow and stem celltransplantation, including cord blood transplantation,and research on improving the safety of the nation’sblood supply. The research program spans a broad

spectrum of research, from the most basic researchthrough applied research to clinical applications, dis-ease prevention, and knowledge dissemination.Specific areas include the following: Red blood cellmembrane defects in thalassemia; improved prenataldiagnosis for disease prevention; gene replacementtherapy; prenatal treatment and curative therapies;drugs to increase fetal hemoglobin; oral iron chelatordevelopment; improving the safety of the nation’sblood supply; bone marrow, umbilical cord blood, andstem cell transplantation; and education and informa-tion dissemination.

Most thalassemia research is supported as investi-gator-initiated research. The NHLBI also utilizesspecial targeted initiatives to accomplish specificobjectives. In the past few years, these have includedthe following grant- or contract-supported programs:Gene Therapy Strategies for the Treatment of Cooley’sAnemia; Basic Research on Hematopoietic Stem CellBiology; In Utero Stem Cell Transplantation for GeneticDiseases; Stem Cell Sources and Transplantation Biology;Specialized Centers of Research in Hematopoietic StemCell Biology; Human Stem Cell Sources andTransplantation Biology; Hetrovirus EpidemiologyDonor Study; Viral Activation Transfusion Study; andUmbilical Cord Blood: Collection, Storage andTransplantation. In addition, the institute has recently

2

0

4

6

8

10

1214

16

1988 1990 1992 1994 [1996]

Thalassemia Research Activities of the NationalHeart, Lung, and Blood Institute, NIH

Alan S. Levine, Ph.D.


NHLBI Thalassemia Research($ Millions)

50

100

150

200

250

300

NHLBI NCI NIDDK NIAID NCRR NICHD OTHERS

NIH Support of Hematology Research -FY 1995 ($ Millions)


held workshops entitled Butyrate Therapy for Cooley’sAnemia, Cooley’s Anemia Progress Review, and A SpecialEmphasis Panel on New Therapies for Thalassemia.

The NHLBI recently published a new booklet forthe lay public entitled What is Cooley’s Anemia? and a60-page booklet for professionals, Cooley’s Anemia:Progress in Biology and Medicine—1995. The latter con-tains a chapter discussing recommendations forresearch, which include: development of safe and effec-tive orally active iron-chelating agents; development ofapproaches to the accurate, noninvasive assessment ofbody iron burden; development of new approaches toimproving compliance with iron chelation therapy, par-ticularly among adolescents; development of safe andeffective therapies to enhance fetal hemoglobin produc-tion; development of a consensus regarding the optimalmanagement of thalassemia intermedia; support andexpansion of stem cell biology, transplantation, and genetherapy research; enhancement of the safety of theblood supply provided to patients with Cooley’s anemia;development of a safe and effective treatment of viralhepatitis; investigation into the psychosocial needs ofpatients related to increased life expectancy; adaptationof current counseling programs and written materialsfor multicultural application; investigation of the psy-chosocial impact of hormone therapy in adolescents andyoung adults; and establishment of a network of med-ical centers to enable rapid testing of new clinicalmodalities.

In 1996, a special emphasis panel on new therapiesfor thalassemia was convened to focus on the clinicallyrelated recommendations mentioned in the precedingparagraph and to reach a consensus on the high priorityissues in need of additional research support. The panelrecommended clinical studies of more effective ironchelators and drugs to increase fetal hemoglobin; clini-cal studies of endocrine disturbances (including delayedsexual development, osteoporosis, infertility, and long-term effects of hormone therapy); and research toimprove the technology for noninvasive measurementof tissue iron deposits.

NHLBI is committed to continued support of tha-lassemia research. The outlook for the future forpatients with β-thalassemia is bright. Unrelated donor

stem cell transplantation is being improved, and newtherapies for iron chelation, fetal hemoglobin enhance-ment, and gene therapy are on the horizon.

References

1. NIDDK. Eighth Annual Report of the Kidney, Urologic,and Hematologic Diseases Interagency CoordinatingCommittee. NIH, 1996.

Thalassemia was defined as a clinical entity in1925 when Dr. Thomas B. Cooley and his asso-ciate Pearl Lee, pediatricians at the Detroit

Children’s Hospital, presented a paper at the annualmeeting of the American Pediatric Society describingfive young children with severe anemia, splenomegaly,and peculiar bone abnormalities.1 At that time, childrenwith severe anemia and splenomegaly were classified ashaving “Von Jaksch’s Anemia.” This was a clinicalhodgepodge that included many different diseases:infections, malignancies, benign anemias, and doubtlessmany others. Two years later, in 1927, Cooley publishedhis classic paper in American Diseases of Childhood thatdescribed seven children with distinctive features thatindicated they represented a new syndrome.2 Theircommon features included a peculiar facies resemblingthe Mongolian race, with a yellow skin color and thick-ening of facial bones and malar eminences. The thicken-ing of the calvarial and long bones had a unique and dis-tinctive roentgenographic appearance. Using the limitedhematologic tests that were available at the time, Cooleydescribed severe anemia, increased osmotic resistance ofthe red blood cells (RBCs), and the presence of manynucleated RBCs in the blood. He also observed that hispatients were of Italian ethnicity, and he suggested thenames erythroblastic or Mediterranean anemia for the

disease that was subsequently given his name.Following Cooley’s epochal descriptions, other simi-

lar children were reported in North America andEurope. In 1932, Whipple and Bradford in Rochester,New York, described pathological findings in severalchildren.3 Apparently wishing to avoid the eponym“Cooley’s anemia,” they coined the term “thalassemia”from the Greek word thalass, meaning “the sea” (i.e., theMediterranean). Thus, thalassemia became “the sea inthe blood” (Table 1).

The characteristic features of Cooley’s anemia arewell known. Ninety percent of children develop symp-toms of anemia in the first year or two of life. The ane-mia is so severe that treatment with blood transfusions isnecessary.4 Untreated patients survive only a few years.5

The RBC morphology of severe thalassemia is extremeand characteristic and includes hypochromia, microcyto-sis, and target cells, as well as many nucleated RBCs.

Shortly after Cooley’s report was published, a seriesof related papers appeared in the Italian, Greek, andAmerican medical literature (Table 2). Rietti, Greppi,

The Evaluation of Thalassemia Intermedia

Howard A. Pearson, M.D.


Whipple & Bradford 1936“Thalassemia”

ΘαλασσαThe Sea in the Blood

TA B L E 1

1925–35 “Malatti de Rietti - Greppi - Micheli”: Mediterranean hematologic disorder, microcythemia

1936 Caminopetros: Mendelian recessive disease

1940 Wintrobe; Dameshek:Mild, familial disorder, RBC morphology “resembling Cooley’s Anemia”

1942–48 Gatto, Valentine, and Neel:Heterozygous relationship to homozygous Cooley’s Anemia

TA B L E 2 : T H A L A S S E M I A M I N O R

and Micheli described individuals with a mild, familial,microcytic anemia with increased RBC osmotic resis-tance.6–8 Caminopetros in Greece recognized a similarblood condition and showed that it was recessively trans-mitted.9 Across the Atlantic, Wintrobe in Baltimore andDameshek in Boston, apparently unaware of the Italianand Greek literature, described families with a mildinherited anemia.They noted that the red cell morpholo-gy of affected individuals resembled that seen in Cooley’sanemia, though not as extreme.10,11

It was shortly thereafter that Gatto in Italy andValentine and Neel in the United States clearly pointedout the relationship of these mild microcytic anemiasto the severe Cooley’s anemia and suggested the clini-cal terms thalassemia “minor” and “major” for the het-erozygous and homozygous conditions.12,13 It is likelythat all of these early reports were descriptions of whatnow is called β-thalassemia, a genetic disorder thatresults in decreased production of the β chains of adulthemoglobin.

Microcytosis of the RBCs is an important hema-tologic feature of thalassemia minor. The recognitionof microcytosis has been greatly facilitated by the useof electronic cell counters that directly measure themean corpuscular volume (MCV) of the RBC popula-tion. This has been widely and successfully used toscreen populations for thalassemia minor.14 Almost alladult patients with heterozygous β-thalassemia haveMCVs of less than 75 fL. In infants and children, themicrocytosis of thalassemia is superimposed on the“physiologic” microcytosis of infancy and childhood.15

In a patient with microcytic RBCs, a diagnosis of β-thalassemia is usually associated with an increased lev-els of HbA2. Levels of fetal hemoglobin (HbF) arenormal (less than 2.0%) in about 50% of patients, andin the rest rarely exceed 7.0%.

Table 3 lists the salient features of β-thalassemiaminor. These include microcytosis and elevations ofHbA2, defining what has been called “classical tha-lassemia trait.” There are other hematologically identi-fiable variants of β-thalassemia minor. So-called βδ-thalassemia trait is a familial microcytosis with normallevels of HbA2 and HbF levels of 5–10%. The LeporeHb trait is characterized by a familial microcytosis,normal levels of HbA2, and the presence of a 5–10%electrophoretically abnormal hemoglobin componentthat migrates in the position of HbS. Finally, some tha-lassemia pedigrees include individuals who are hemato-logically normal but transmit a genetic factor thatinteracts with and intensifies a thalassemia gene. Suchindividuals are known as “silent carriers.”

We now know that thalassemia is a very heteroge-neous genetic condition. More than a hundred differ-ent mutations have been associated with thalassemiaphenotypes.16 Molecular genetics have been used todiagnose thalassemia prenatally, which has led to amarked reduction in the number of severe homozy-gous infants born throughout the world.17,18

Almost all patients with β-thalassemia syndromescan be clinically classified as having thalassemiaminor or thalassemia major, and the genetic relation-ship of heterozygous and homozygous genotype to


Anemia Microcytosis HbA2 HbF Other(MCV<75 f l)

“Classical” 0 – + + >3.5% 1 – 5%

βδ 0 – + + <3.5% 5 – 10%

Lepore 0 – + + <3.5% 1 – 5% 5 – 10%Lepore Hb

Silent carrier 0 0 <3.5% <2%

TA B L E 3 : F O R M S O F β - T H A L A S S E M I A T R A I T

these clinical conditions is straightforward. In morethan 95% of patients, there is a direct correlation ofmajor and minor phenotypes with homozygous andheterozygous genotypes. However, a small number ofpatients do not fit into this neat dichotomy. Sturgeonfirst identified this group in the American litera-ture.19 He suggested the term “thalassemia interme-dia” to describe patients who were hematologicallytoo severe to be called “minor” and too mild to becalled “major.” Thalassemia intermedia is a clinicaldesignation often used to characterize individualswho are homozygous for β-thalassemia genes butmaintain hemoglobins of 6–9 g/dL without regulartransfusions. They have more severe RBC morpho-logical abnormalities than the trait, as well as varyingdegrees of splenomegaly and skeletal changes. In oneof my own early thalassemia papers, published in1964, I described a number of patients with tha-lassemia intermedia and attempted to reconcile clini-cal and hematologic findings with genotype20 (Table4). Even 30 years later, it doesn’t look too bad! Table 5shows an expanded summary of the features thatdefine β-thalassemia major, minor, and intermedia.

Thalassemia intermedia has an extraordinarilywide clinical spectrum. Some patients have very fewclinical abnormalities. Others experience severe con-sequences when their erythropoetic systems attemptto compensate for ineffective erythropoiesis and ane-mia. These compensatory efforts may be associatedwith massive erythroid marrow hypertrophy inmedullary and extramedullary sites. In the longbones, marrow expansion results in cortical thinningand pathologic fractures. The marrow spaces of thecranial vault markedly expand, producing a “hair-on-end” appearance. Expansion of the facial bones andobliteration of the maxillary sinuses result in protru-sion of the upper jaw and malocclusion. Thesechanges may result in cosmetic abnormalities that cancause emotional distress. Today in the United Statesbony and cosmetic abnormalities are seen only inpatients with thalassemia intermedia because modernhypertransfusion programs prevent this in thalassemiamajor. Expanding paravertebral extramedullary hema-topoiesis may compress the spinal cord. The spleenenlarges, often requiring a splenectomy to relieve themechanical burden. These changes are progressive but


Major Intermedia MinorHemoglobin (g %) <7 7 – 10 >10Reticulocytes (%) 2 – 15 2 – 10 <5Nucleated RBC ++++ – + + – 0 0RBC morphology ++++ ++ +Jaundice ++ +0 0Splenomegaly +++ + 0Skeletal changes +++ – ++ + – ++ 0Transfusion +++ – + + – 0 0

HOMOZYGOUS

HETEROZYGOUS

INTERACTIONS WITHTHALASSEMIA VARIANTS

HEMOGLOBIN HSYNDROMES

INTERACTIONS WITHABNORMAL HEMOGLOBINS

CLI

NIC

AL

GEN

ETIC

Clinical and genetic characteristics of thalassemia syndromes.

TA B L E 4 : T H A L A S S E M I A

preventable, and are at least partially reversible byadequate transfusion therapy. As a group, patientswith thalassemia intermedia live longer than patientswith thalassemia major, although many of themdevelop complications in later life.

The proportion of patients with homozygousthalassemia intermedia is strikingly different amongdifferent ethnic groups. About 10% of patients ofMediterranean ethnicity who are homozygous canbe classified as intermediates. In contrast, more than70% of African-American patients who are homozy-gous may be classified as such.21 This differencereflects the kinds of thalassemia mutations, especial-ly so-called “mild mutations,” that are prevalent inthese ethnic groups.

The purpose of this symposium is to seek theanswer to a number of questions concerning thediagnosis and management of patients with tha-lassemia intermedia. We are fortunate to have theparticipation of experts from a number of NorthAmerican centers, where nearly 100 patients are cur-rently being followed.

Before we can begin, we must first define whatthalassemia intermedia is. We have decided to use arather restrictive definition, so we are all talking

about the same thing. The definition to be used inthis conference is as follows:

The term thalassemia intermedia has beenused to describe the clinical and hematologicfindings in patients whose illness is not assevere as that which characterizes homozy-gous β-thalassemia, but is more severe thanthe heterozygous carrier.

D. Weatherall

For the purposes of this conference, the following,rather arbitrary criteria and exclusions will be used.

Exclusions:

l. Only β-thalassemia syndromes will be included.Alpha chain abnormalities (e.g., HbH disease)will not be included. However, if an α chainabnormality interacts with β-thalassemia andmodifies the usual phenotype, it can be included.

2. Although there may be similar clinical andhematological features, double heterozygotes fora β-thalassemia and a β-hemoglobinopathy (i.e.,HbS-β-thalassemia and especially HbE-β-tha-lassemia) will not be included.


MAJOR INTERMEDIA MINORClinical Onset in Infancy Later Onset AsymptomaticSplenomegaly ++++ +++ – ++++ 0 – +Jaundice +++ + – +++ 0 – +Bone changes ++++ ++ – ++++ 0Facial changes ++ – ++++ 0 – ++++ 0

HematologicAnemia ++++ ++ – +++ 0 – +RBC ↓ ↓ N – ↑Microcytosis + + +NRBC ++ – ++++ + – +++ 0

BiochemicalHbF 10 – 95+% 10 – 95+% N or <10%

HbA2 N or ↑ N or ↑ N or ↑ (>3.5%)

TA B L E 5 :M A N I F E S TAT I O N S O F T H A L A S S E M I A

Criteria:

1. Homozygosity for a β-thalassemia gene of anydefinable sort (β° or β+) or a β-thalassemia (β° orβ+) and a related β ς chain production gene (i.e.,βδ-thal, Hb Lepore, Swiss type HRFH, etc.) canbe indicative of thalassemia intermedia.Homozygosity and double heterozygosity shouldbe established, if possible, by family studies orgenotype.

2. Maintenance of a hemoglobin level >7.0 g/dLand no regular transfusions for an extended peri-od of time after 4 years of age can be indicative ofthalassemia intermedia. There is a small subset ofpatients who are homozygous who maintainhemoglobin levels >7.0 during their first few yearsof life but then have hemoglobins fall lower andrequire regular transfusions.

3. Patients with thalassemia intermedia who beginregular transfusions because of unacceptable, pro-gressive cosmetic changes or poor growth duringthe first few years of life despite hemoglobin lev-els >7.0 g/dL are difficult to categorize. However,for the purposes of this conference, they will beclassified as having thalassemia major, unlessolder affected siblings fulfill criteria 1–3 above.

4. Patients who develop progressively increasinglysevere anemia (<7.0 g/dL) and splenomegaly andwho, after splenectomy, maintain a hemoglobinlevel >7.0 g/dL and do not require regular trans-fusions can be classified as having thalassemiaintermedia.

Using the clinical and hematological data from alarge number of patients from contemporary NorthAmerica, we hope at the end of our discussions today toanswer a number of questions:1. How do we define and diagnose thalassemia

intermedia?2. What are the indications for such interventional

therapy as splenectomy, transfusions, and chela-tion?

3. Can we predict early in life how a given patientwill fare?

4. In view of the major advances in molecular genetics,can we relate thalassemia genotype to phenotype?

5. What are the long-term complications of tha-lassemia intermedia, and what is the prognosis ofthese patients today?These are the issues I hope we can address, if not

answer, today. If we are even partly successful, thelives of these patients and the patients of the futurewill be better, and the understanding of the physi-cians caring for them will be increased.

References

1. Cooley TB, Lee OP. Series of cases of splenomegalyin children with anemia and peculiar bone changes.Trans Amer Pediatr Soc 37:29, 1925.

2. Cooley TB, Witter, ER, Lee OP. Anemia in childrenwith splenomegaly and peculiar changes in the bones:Report of cases. Am J Dis Child 34:347, 1927.

3. Whipple GH, Bradford WL. Racial or familial anemiaof children associated with fundamental disturbances ofbone and pigment metabolism (Cooley-van Jaksch). AmJ Dis Child 44:336, 1932.

4. Logothetis J, Constantioulakis M, Economidou J, et al.Thalassemia major: A survey of 138 cases with empha-sis on neurology and muscular aspects. Neurology22:294, 1972.

5. Silverstroni E, Bianco I. Screening for microcythemia inItaly; analysis of data collected in the past thirty years.Am J Hum Genetics 27:198, 1975.

6. Rietti F. Ittero emolytica primitive. Atti Acad Scient MedNat Ferrara 2:14, 1925.

7. Greppi E. L’iperglobalia microcytica (microcyitosi vera)con aumento della iteroemolitico premitivo. MinervaMed 1:189, 1931.

8. Michili F, Penatal F, Levi GM. Anemia ipocromicasplenomedica con elletto dilosi-ponchillo estos.Hematologica Arcivo 16:5, 1925.

9. Caminopetros J. Recherches sur l’anemic erythroblas-tique inantile, dis periples de lar Mediterranean orien-tale. Ann Med 43:27, 1938.

10. Wintrobe MM, Mathews E, et al. A familial hemato-poietic disorder in Italian adolescents and adults. JAMA114:1580, 1940.

11. Dameshek W. “Target cell” anemia: Anerythroblastictype of Cooley’s anemia. Am J Med Sci 200:445, 1940.

12. Gatto I. Richerche sue familiari di pambine affitti demalathia di Cooley. Arch Ital Pediat Eupericalt 9:128,1942.


13. Valentine WN, Neel JV. Hematologic and geneticstudy of the transmission of thalassemia. Arch Int Med74:185, 1944.

14. Pearson HA, O’Brien RT, et al. Screening for tha-lassemia trait by electronic measurement of mean cor-puscular volume. N Engl J Med 288:351, 1973.

15. Berman RN, Ritchey AK, et al. Hematology of tha-lassemia trait—age-related developmental aspects andintrafamilial correlations. J Pediatr 97:901, 1980.

16. Forget BG, Pearson HA. Hemoglobin synthesis andthe thalassemias. In Blood: Principles and Practice ofHematology, eds. Handin RI, Lux SE, Stosset TP, p.1525. Philadelphia: J. B. Lippencott Co., 1995.

17. Cao A, Rossatilli, et al. The prevention of homozygousβ-thalassemia by carrier screening and prenatal diag-nosis in Sardinia. Clin Genet 36:277, 1989.

18. Pearson HA, Guiliotis DK, et al. Patient age distribu-tion in thalassemia major; changes between 1973 and1985. Pediatrics 80:53, 1987.

19. Sturgeon P, Itano HA, et al. Genetic and biochemicalstudies of intermediate types of Cooley’s anemia. Brit JHaemat 1:264, 1955.

20. Pearson HA. Thalassemia intermedia: Genetic andbiochemical consideration. Ann NY Acad Sci 119:390,1964.

21. Pearson HA, Cohen AR, et al. The changing profile ofhomozygous β-thalassemia: Demography, ethnicityand age distribution of current North Americanpatients and changes in two decades. Pediatrics 97:352,1996.


Aretrospective multicenter review of tha-lassemia intermedia was undertaken by sixnortheastern medical institutions in the

United States: New York Hospital, Cornell UniversityMedical College; Children’s Hospital of Philadelphia,University of Pennsylvania School of Medicine; NewHaven Hospital, Yale University School of Medicine;Boston Children’s Hospital, Harvard UniversitySchool of Medicine; Rhode Island Hasbro Children’sHospital, Brown University School of Medicine; andBrigham and Women’s Hospital, Harvard UniversitySchool of Medicine.

The demographics, age, sex, and ethnic origins ofthe patients were reviewed, as well as their medicalhistories, including transfusions, alloimmunization,facial appearance, splenectomy, puberty, growth, frac-tures, bone mineral density, cholecystectomy,cholelithiasis, fractures, iron overload and chelation,infections, β- and α-globin genotypes, miscellaneousconditions, and use of agents to enhance fetal hemo-globin production.

Seventy-one patients (36 females and 35 males)with β-thalassemia intermedia were eligible for thestudy. Patient data were obtained from medicalrecords, and not all data were available for somepatients.

The majority of patients were of Mediterraneanethnic origin. The ethnic origins of the patients were:48% Italian, 25% Greek, 13% African, 4% MiddleEastern, 3% Armenian, 1% Indian, 4% mixed, and 1%unknown.

The mean age of the population was 26.6 ± 12.3years (range 3.6–65, median 25.4). During the pasttwo decades, two patients were lost to followup at theages of 3.6 and 13 years, respectively, and fourpatients expired at ages 26.7, 27.5, 29.8, and 41.1years. All those who died suffered from congestive

heart failure; related causes of mortality includedhepatic cirrhosis in two patients and complicationsfrom AIDS in one patient. The median survivalextrapolated from a descriptive Kaplan-Meier plotwas estimated at 51 years. The mean age at diagnosisin all 71 patients was 3.8 ± 3.2 years (range 0–19,median 3.5).

The mean age at splenectomy was 11.9 ± 8.3years (range 1.2–41, median 9.0). Fifty-five of 71patients (77%) underwent splenectomy. All but threepatients increased their hemoglobin levels by 0.5–5g/dL postsplenectomy. The mean hemoglobin levelspresplenectomy and postsplenectomy were 6.7 and8.5 g/dL, respectively, representing a mean rise inhemoglobin of 1.8 g/dL. The mean hemoglobin levelof the 16 nonsplenectomized patients was 8.5 g/dL.

Regular transfusions were given to 23 of 71patients (32%) for indications including severeextramedullary hematopoiesis, chronic fatigue, con-gestive heart failure, and stunted linear growth orsevere cosmetic facial changes. Patients who requiredregular transfusions were diagnosed for thalassemiaintermedia at a mean age of 2.6 ± 3.7 years (range0.3–5, median 2.5). Splenectomy was performed in 20of these 23 patients (86%). Patients who ultimatelyrequired a regular transfusion program after undergo-ing splenectomy did so after a 13-year median inter-val (range 0–38 years) postsplenectomy. Patients onregular transfusions received a mean of 23 units peryear postsplenectomy.

Sporadic transfusions were given to 28 of 71patients (40%) perioperatively, during pregnancy,bouts of illness, and acute episodes of leg ulcers.Patients who received sporadic transfusions or whoremained transfusion-independent were diagnosed ata mean age 4.5 ± 3.7 years (range 1–19, median 4). Ofthose patients who were sporadically transfused, 24 of

Beta-Thalassemia Intermedia in the United States:A Multicenter Study

Patricia J. Giardina, M.D., Igal Fligman, M.D., Alan R. Cohen, M.D., Haig H. Kazazian, M.D.,Beatrice Gee, M.D., Edwin N. Forman, M.D.,Kenneth R. Bridges, M.D., Howard A. Pearson, M.D.


28 (86%) underwent splenectomy. Patients whoreceived sporadic transfusions received a mean of twounits per year postsplenectomy. Fifty-five percent ofpatients (11 of 20) who received no transfusions alsounderwent splenectomy.

Transfusion-related alloimmunization was detect-ed in 6 of 51 patients (12%). This occurred with equalfrequency in regularly and sporadically transfusedpatients. The reported antibodies included E, e C, cD, Kell, Cob, and Yya. Clinically significant alloim-munization occurred in three patients with autoim-mune hemolytic anemias: two who were sporadicallytransfused and one who had been placed on a regulartransfusion program.

Facial manifestations were observed in mostpatients. As judged by professional staffs, these cos-metic changes were severe in 23 of 71 patients (33%),moderate in 18 of 71 (25%), and minimal in 30 of 71(42%). Seven patients underwent osteoplasty, radia-tion therapy, and/or transfusion support in order toameliorate the thalassemic facies.

Fifteen of 71 patients (21%) had abnormal lineargrowth (<5% for height), 31 of 71 (44%) had agrowth range of >5% and <50%, and 25 of 71 (35%)had above-average linear growth (≥50%).

Data for the age of attainment of puberty wereavailable for 45 patients (23 males and 22 females).Spontaneous puberty occurred in 38 of 45 patients(84%), but was delayed in 22 of 45 (49%). Delayedpuberty was defined as two standard deviations abovethe mean age of menarche at ≤13 years and the meanage of acquisition of secondary sexual characteristics

in males at ≤16 years. Seven patients required hor-mone therapy to induce puberty (five females and twomales). The mean age of menarche in this series was15.7 ± 2.2 years (range 12.5–22, median 15.0). Themean age of male secondary sexual characteristics cor-responding to Tanner stage III was 15.2 ± 1.5 years(range 12–18, median 15.0).

Various endocrinopathies were reported in 21 of71 patients (30%). Twenty-one percent of patientshad poor linear growth (less than the fifth percentile),10% had primary hypogonadism, 6% had secondaryhypogonadism, 4% had hypothyroidism, and 3% hadhypoparathyroidism.

Thirty of 71 patients (42%) sustained fractures,with 14 of 30 (47%) sustaining multiple fractures. Ofthe 60 fractures that occurred, the long bones of theupper and lower extremities were most commonlyinvolved. Fractures occurred at a mean age of 18.7 ±10.7 years (range 1–60, median 18.0).

Spinal bone mineral densities (S-BMD) weremeasured in 19 patients with thalassemia intermedia.The mean S-BMD of 10 female patients was 0.79 ±0.11 g/cm2—slightly lower than age-matched femalepatients with thalassemia major (0.83 ± 0.15 g/cm2),but significantly lower than normal female controls ofcomparable age (1.20 ± 0.10 g/cm2). Similarly, themean S-BMD of nine male patients with thalassemiaintermedia was 0.88 g/cm2, slightly lower than nor-mal male controls of comparable age (1.26 ± 0.13g/cm2). All patients with thalassemia intermediaexhibited S-BMD values at or near the fracturethreshold of 0.8 g/cm2. Five patients with thalassemiaintermedia with no fractures were age- and sex-matched with five patients with thalassemia interme-dia who had either symptomatic paravertebralextramedullary hematopoiesis or severe facial changes.The mean S-BMD of the fracture-free group was0.86 ± 0.17 g/cm2, versus 0.73 ± 0.14 g/cm2 for thegroup with severe facial changes and/or spinal cordcompression.

Cholelithiasis or sludge was demonstrated withultrasonography or CT scans in 37 of 71 patients(52%) at a mean age of 18.7 ± 7.7 years (range 6–38,median 18.0). Cholecystectomy was performed in 30of 71 (42%) at a mean age of 20.0 ± 7.9 years (range6–40, median 20.0).


β+/β+ 17/39 44% αα/αα 25/33 76%β+/β0 15/39 38% αα/α – 5/33 15%β0/β0 7/39 18% αα/αα αα 2/33 6%

αα/αα α 1/33 3%

The beta and alpha globin synthesis analysis demonstrates that44% of population are compound heterozygous for β+/β0 synthesisand 18% are homozygous for β0/β0 synthesis. 76% have a normalcomplement of alpha globin genes, 15% have a single alpha genedeletion, and 3% to 6% have alpha globin gene duplications.

THALASSEMIA INTERMEDIAMULTICENTER STUDY

β AND α GLOBIN STUDIES

Peak ferritin levels were available in 64 patients.The mean peak ferritin level was 2743 ± 2640 ng/mL(range 121–13000, median 2050). Peak levels werereported at a mean age of 20.1 ± 10.0 years (range3–52, median 18.3). Forty-five patients (70%) hadferritin levels exceeding 1000 ng/mL, and 34 of 45(76%) received subcutaneous desferrioxamine (SCDFO) therapy at a mean age of 19.8 ± 10.3 years(range 4.3–50, median 17.0). The patients whounderwent iron chelation therapy had a mean reduc-tion in their ferritin level of 1966 ± 2505 ng/mL, (i.e.,from a mean of 3971 ± 2799 ng/mL to a mean of2005 ± 2212 ng/mL).

Eighteen of 23 regularly transfused patients (78%)were started on SC DFO. Their mean ferritin levelswere reduced from a mean of 4711 ± 3334 ng/mL to amean of 2534 ± 2732 ng/mL (i.e., a mean reduction of2177 ± 3033 ng/mL). This reduction was accom-plished in a mean interval of 12.4 ± 4.9 years from thestart of chelation therapy. Similarly, 16 of 28 sporadi-cally transfused patients (57%) were started on SCDFO. Their ferritin levels were reduced from a meanof 3226 ± 1631 ng/mL to a mean of 1473 ± 1207ng/ml, for a mean reduction of 1753 ± 1419 ng/mL,and this was accomplished in a mean interval of 12.3 ±5.1 years from the initiation of iron chelation therapy.

Major infectious complications were reported in32 of 71 patients (45%). The most common major

infections reported were pneumonia (21%), fever withsepsis (15%), and fevers of unknown origin (10%).Minor, recurrent infections were upper respiratoryinfections, nasopharyngeal infections, and cellulitis.

A host of medical complications were reported in28 of 69 patients (41%). The most prevalent condi-tions were cardiac disease, including congestive heartfailure, arrhythmia and pericarditis, pulmonary hyper-tension, severe extramedullary hematopoiesis, legulcers, and hyperuricemia. Infrequently reported med-ical conditions included osteoporosis, hepatic cirrho-sis, peptic ulcer disease, and thrombotic episodes.

Fetal hemoglobin enhancement was attempted in10 of 71 patients (14%). Nine of 10 patients receivedbutyric acid analogs, and 4 of these exhibited anincrease in fetal hemoglobin of 1–1.5 g/dL. Onepatient received hydroxyurea but had no significantchange in hemoglobin level.

β-Globin DNA analysis was available in 39patients. The most prevalent alleles were IVS1,6 in 19of 39 (23%), β°39 in 19 of 39 (23%), and IVS1,110 in15 of 39 (18%). With regard to β-globin chain syn-thesis, 17 of 39 (44%) had homozygous β+/β+ allelesand 15 of 39 (38%) had heterozygous β+/β° alleles,while only 7 of 39 (18%) had homozygous β°/β° glo-bin alleles.

α-Globin gene DNA analysis was available in 33patients. The majority of these—25 of 33 (76%)—had


• IVS1,6 19 23% • IVS2,1 2 3%• β039 19 23% • -87 (C-G) 2 3%• IVS1,110 15 18% • -28 (A-C) 2 3%• δβ 7 8% • NORMAL β 2 3%• IVS1,1 6 7% • -88 (C-A) 1 1%• IVS2,745 4 5% • -92 (C-T) 1 1%• LEPORE 3 4% • A-C POLY A SITE 1 1%

T H A L A S S E M I A I N T E R M E D I AM U LT I C E N T E R S T U D Y

β A L L E L E D I S T R I B U T I O N

The distribution of beta allele includes 13 different beta globin mutations of which IVS1,6and β039 are most common at 23% frequency followed by IVS1,110 at 18%, δβ at 8%,IVS1,1 at 7%, IVS2,745 at 5%, Lepore hemoglobin at 4%. The milder mutations includingIVS2,1, -87 (C-G), -28 (A-C), -88 (C-A), -92 (C-T), A-C poly A site are rarer at 1% to 3%.

four α genes, 5 of 33 (15%) had a deletion of a singleα gene, and 3 of 33 (9%) had additional α genes(αααα /αα in two and ααα /αα in one).

Complete α-globin and β-globin genotypes wereavailable in 29 patients. In this subset of patients, themost common β genotype was the compound het-erozygous IVS1,6/β°39 with a normal α-gene com-plement, found in 7 of 29 patients (25%). There wasthe suggestion of a milder phenotype in individualswho had a concurrent α gene deletion, as evidencedby the later age at diagnosis, milder facial cosmeticchanges, no requirement of regular transfusion sup-port, and lower ferritin peak levels. There was also asuggestion that mild β mutations may yield a disorderof less severity.

This retrospective series reviews various clinicaland laboratory parameters in a limited patient popu-lation that is widely distributed over the northeasternUnited States. To enhance our understanding of phe-notypic expression of genotype and optimize themanagement of patients with thalassemia intermedia,it is recommended that complete DNA analysis ofboth α and β genotypes be obtained in all patients.This review found suggestions of a milder phenotyperesulting from mild β gene mutations and/or concur-rent α gene deletions. Regular annual clinical and lab-oratory evaluations would allow for a prospectiveevaluation of phenotypic expression and generate agreater data base to identify predictive parameters ofdisease expression. Management strategies, includingthe use of splenectomy, transfusion, and chelationtherapies, should be optimized. Finally, additionaltherapeutic options need to be further explored. Canfetal hemoglobin enhancement and effective oral ironchelators improve the clinical outcome of patientswith thalassemia intermedia? Should bone marrowtransplantation be offered to those who require regu-lar transfusion programs? These questions must bepursued in prospective clinical research studies.


• Limited population is widely distributed

• Predominantly Mediterranean origins

• Median age of diagnosis 3.5 years

• Median age of population 25.3 years

• Median age of splenectomy 9.0 years

• Median age of cholecystectomy 20.0 years

• Splenectomy increases Hb level by nearly 2 g/dl

• Splenectomy may delay onset of regular TX by a median interval of 13 years

• TX therapy was required in 32% of pts. regardless of splenectomy

• Alloimmunization occurs in 12% of pts.

• Fractures are common; median age 18.0 years

• Bone mineral densities are at the

fracture threshold

• Severe growth retardation occurs in 21%

• Puberty spontaneously achieved in 84%

• Iron overload is responsive to chelation

• Congestive heart failure is the major cause of death

• Median survival is 51 years

• Genotype analysis is required in all pts.

• Genotype may be predictive of

phenotype

• Alpha gene deletion—an ameliorating factor or a better phenotype predictor

• Annual evaluations to optimize therapies and generate prospective data

• Additional therapies to be further pursued—Hb enhancement, oral iron chelation

T H A L A S S E M I A I N T E R M E D I AM U LT I C E N T E R S T U D Y

S U M M A RY

In patients with thalassemia major who are regu-larly transfused, the most important consequenceof life-saving transfusions is the accumulation of

iron within tissues, causing progressive organ dys-function that is fatal without chelating therapy.1 Inpatients with thalassemia intermedia, the requirementfor transfusions is less regular than with thalassemiamajor, and iron loading secondary to increased gas-trointestinal iron absorption is less accelerated thanthat associated with transfusional iron overload inthalassemia major.2,3 Nevertheless, the clinical conse-quences of iron loading and the issues relating toiron-chelating therapy encountered in patients withthalassemia major may also apply to patients withthalassemia intermedia. Assessing iron overload,determining the optimal concentrations of body iron,and determining the time to initiate treatment for ironoverload are difficult in both patients with thalassemiamajor and patients with thalassemia intermedia.

Assessment of Iron Overload

In all patients with thalassemia, both direct andindirect means of assessing body iron are available. Nosingle indicator or combination of indicators is idealfor evaluating iron status in all clinical circumstances.Measurement of hepatic iron stores provides the mostquantitative means of assessing the body iron burdenin patients with thalassemia major,4 and may be con-sidered the reference method for comparison withother techniques. Data accumulated over the past 10years permit a quantitative approach to the manage-ment of iron overload, providing guidelines for thecontrol of body iron burden.

Indirect. The measurement of plasma or serumferritin is the most commonly used method for indi-rectly estimating body iron stores.1 Normally, ferritin

concentrations decrease with depletion of storage ironand increase with storage iron accumulation. A maxi-mum glycosylated plasma ferritin concentration ofabout 4,000 µg/L may represent the upper physiolog-ic limit of the rate of synthesis;5 higher concentra-tions are thought to be due to the release of intracel-lular ferritin from damaged cells. Interpretation offerritin values may be complicated by a variety of con-ditions that alter concentrations independently ofchanges in body iron burden, including ascorbatedeficiency, acute infection, chronic inflammation,hemolysis, and ineffective erythropoiesis, all of whichare common in patients with thalassemia intermedia.In one study of patients with thalassemia major andsickle cell disease, the 95% prediction intervals forhepatic iron concentration, given the plasma ferritin,were so broad as to make determination of plasmaferritin a poor predictor of body iron stores.6

Serum iron, transferrin, transferrin saturation, andtransferrin receptor concentration do not quantita-tively reflect body iron stores, while chelator-inducedurinary iron excretion is also vulnerable to extraneousinfluences of inflammation, the activity and effective-ness of erythropoiesis, extramedullary hematopoiesis,and ascorbic acid deficiency, all of which are commonin patients with thalassemia intermedia.1 Computedtomography, nuclear resonance scattering from man-ganese-56, and magnetic resonance imaging havebeen used to image tissue iron stores in vitro and invivo, but none of these modalities is clinically avail-able for the accurate measurement of hepatic ironconcentrations.1 Indeed, while many studies demon-strate that magnetic resonance imaging can reflect thepresence of and changes in tissue iron in vivo, eventhis method has not been shown to provide measure-ments of tissue iron that are quantitatively equivalentto those determined by tissue biopsy.

Iron Overload and Iron-Chelating Therapy inThalassemia Intermedia

Nancy F. Olivieri, M.D.


Direct. Direct measurement of hepatic iron con-centration is the most quantitative, specific, and sensi-tive method for determining the body iron burden inpatients with thalassemia major.7 Liver biopsy is thebest direct means of assessing iron deposition, permit-ting chemical measurement of the nonheme (storage)iron concentration and histochemical examination ofthe pattern of iron accumulation in hepatocytes andKupffer cells, as well as evaluation of the extent ofinflammation, fibrosis, and cirrhosis. Magnetic sus-ceptometry using a superconducting quantum inter-ference device (SQUID) magnetometer provides adirect measure of hepatic storage iron that is based ona fundamental physical property of ferritin and hemo-siderin.8–11 Only two sites—one in the UnitedStates8 and one in Germany11—have the specializedequipment needed for measuring hepatic magneticsusceptibility.

Optimal Body Iron in Patients withThalassemia Intermedia

Because the magnitude of the body iron burdenseems to be the principal determinant of clinical out-come,12,13 the prime goal of iron-chelating therapy inpatients with thalassemia—both major and interme-dia—is optimal control of body iron. The optimalbody iron should minimize both the risk of adverseeffects from iron-chelating therapy and the risk ofcomplications from iron overload. In a patient treatedwith deferoxamine, therapy to maintain normal bodystorage iron—corresponding to a hepatic iron ofapproximately 0.2–1.6 mg iron per gram liver, dryweight—might abate the likelihood of complicationsof iron overload, but greatly increases the probabilityof dose-related drug toxicity. Guidance about the riskof complications associated with slightly higher levelsof body iron may be derived from the clinical experi-ence with hereditary hemochromatosis, a condition inwhich the route and severity of iron loading may besimilar to those in thalassemia intermedia. In boththese conditions, iron overload is the result of abnor-mal regulation of iron absorption resulting in an inap-propriate increase in iron uptake. The minor degree ofiron loading that develops in about one-quarter ofpatients who are heterozygotic for hereditary hem-

ochromatosis appears to be associated with normallife expectancy.14 By contrast, patients who arehomozygotic for the disorder develop greater ironburdens and have an increased risk of cardiac disease,hepatic fibrosis, diabetes mellitus, endocrine abnor-malities, and other complications of iron over-load.15–18 Finally, body iron burdens correspondingto hepatic iron concentrations at or exceeding 15 mgiron per gram liver, dry weight, greatly increase therisk of cardiac disease and early death.12 Theseconsiderations suggest that a conservative goal foriron-chelating therapy in patients with thalassemiaintermedia is maintenance of hepatic iron concentra-tions of approximately 3.2–7 mg iron per gram liver,dry weight—the range found in patients who are het-erozygotic for hereditary hemochromatosis.

Initiation of Chelating Therapy inThalassemia Intermedia

When should the patient with thalassemia inter-media be treated for iron overload? This decisiondepends not only on the amount of excess iron butalso on the rate of iron accumulation; the duration ofexposure to increased iron; the partition of the ironburden between relatively benign sites in the macro-phage and more hazardous deposits in parenchymalcells; ascorbate status, which helps determine the allo-cation of iron between macrophage and parenchymalcells; the presence of other hepatotoxins, includingalcohol and viral hepatitis; and other factors in indi-vidual patients.1 Even in patients with thalassemiamajor, the optimal age for beginning iron-chelatingtherapy is uncertain. Elevated hepatic iron concentra-tions associated with hepatic fibrosis, which are notuniformly evident by determinations of serum ferritinor laboratory abnormalities of liver function, havebeen observed in transfused children with thalassemiaunder 3 years of age.19,20 These data suggest thatsome modified program of chelating therapy likely isindicated after 1–2 years of regular transfusions inpatients with thalassemia major. Because of theimprecision of indirect measurements, initiation oftherapy in all patients with thalassemia should bebased upon hepatic iron concentration, obtained afterapproximately 1 year of regular transfusions. Reliance


on serum ferritin measurements alone can lead toinaccurate assessment of body iron burden in individ-ual patients.

Rate and Severity of Iron Loading inThalassemia Intermedia

Only a few studies have attempted to define theextent of iron loading in patients with thalassemiaintermedia. One of these examined the absorptionand rate of accumulation of iron in 15 patients withhomozygous β-thalassemia, ages 4–42 years, notreceiving regular transfusions.2 The severity of ironloading in these patients was assessed from transfu-sion history, as well as through determinations ofserum iron, total iron binding capacity, and serum fer-ritin concentration. Some patients had never beentransfused; the maximum iron load received fromtransfusions was estimated to be approximately 10 g(equivalent to the iron that might be administeredover a 2-year period in a regularly transfused, 50-kgpatient with thalassemia major). The absorption of59-labeled ferrous iron was measured with the Oxfordtotal body counter.21 After a fast of 2 hours or more,13 of the patients received a standard dose of 5 mg of59Fe as a freshly prepared solution of ferrous sulfateto which 50 mg of ascorbic acid was added. Totalbody radioactivity was measured immediately afterthe dose and 7–14 days later. The percentage reten-tion of 59Fe was then calculated, with suitable correc-tions for background and isotope decay. In four sub-jects, absorption was also measured from a standardmeal containing approximately 5 mg iron. In fivepatients, total iron balance was calculated.

The values for total iron binding capacity in thepatients varied from 30% to 100%; in nine patients,transferrin saturation exceeded 75%. Concentrationsof serum ferritin increased with the age of thepatients, although an increasing scatter of values wasnoted in the older patients. The absorption of 59Fefrom a 5 mg dose was greatly increased in this groupof patients, all of whom, as noted above, had evidenceof increased body iron burden. Absorption of ironvaried between 17% and 89%, compared with a meanof 15% in normal iron-replete individuals. There wasa highly significant correlation between the percent-

age of 59Fe absorbed and the serum iron: The higherthe serum iron, the greater the absorption of iron. Innormal individuals, the percentage of iron absorbed isinversely proportional to the body iron burden.Therefore, this study indicates that the mechanism(s)regulating iron absorption in the presence of adequateiron stores may be abnormal in patients with tha-lassemia intermedia.

Supporting these observations was the finding ina balance study of a positive iron balance of 3–9 mgiron per day, or between 3 and 10 times normalabsorption of iron.2 This is the only balance studyreported in patients with thalassemia intermedia, andit indicates that iron loading in patients with tha-lassemia intermedia may be on the order of 2–5 g ironper year. By contrast, in a 50-kg patient with tha-lassemia major who may receive a yearly volume of180 mL/kg packed cells, with the hematocrit of trans-fused blood of about 75%, iron accumulation wouldbe around 6.7 g per year. Clearly, the older patientwith thalassemia intermedia might be expected to beat similar risk for iron-induced hepatic, cardiac, andendocrine dysfunction as the patient with thalassemiamajor. Because the studies by Pippard and colleaguesrelied upon the use of transferrin saturation andserum ferritin concentrations without direct measure-ments of tissue iron, iron overload was not quantitat-ed in these patients. Although studies of organ func-tion were not reported in this series, three of thesepatients were noted to be diabetic or prediabetic, sug-gesting that loading of tissue iron induces similarchanges in patients with thalassemia intermedia as itdoes in patients with thalassemia major. More recent-ly, striking elevations of hepatic iron concentrationhave been observed in patients with thalassemia inter-media with only slight increases in serum ferritin con-centration.22 Thus, direct determination of body ironburden is indicated in patients with slightly elevatedserum ferritin concentrations.

Iron-chelating therapy in patients with thalassemiaintermedia should be initiated if the hepatic iron con-centration exceeds 6 mg iron per gram liver, dryweight. Hepatic iron concentration and liver histologyshould be assessed at intervals every 1–2 years inpatients receiving chelating therapy. One report hasindicated that short-term therapy with the orally active


iron-chelating agent 1,2-dimethyl-3-hydroxypyrid-4-one (deferiprone; L1) may have been effective in reduc-ing hepatic storage iron in one patient with thalassemiaintermedia.23 This agent may prove to be effective inthe reduction of tissue iron in conditions such as tha-lassemia intermedia, which involve less severe degreesof iron overload than thalassemia major.24,25

References

1. Olivieri NF, Brittenham GM. Iron-chelating therapyand the treatment of thalassemia. Blood 89:739–61,1997.

2. Pippard MJ, Callender ST, Warner GT, WeatherallDJ. Iron absorption and loading in beta-thalassaemiaintermedia. Lancet ii:819–821, 1979.

3. Cossu P, Toccafondi C, Vardeu F, Sanna G, Frau F,Lobrano R, Cornacchia G, Nucaro A, Bertolino F, LoiA, DeVergillis S, Cao A. Iron overload and desferriox-amine chelation therapy in beta thalassemia interme-dia. Eur J Pediatr 137:267–271, 1981.

4. Pippard MJ. Measurement of iron status. Prog ClinBiol Res 309:85, 1989.

5. Worwood M, Cragg SJ, McLaren C, Ricketts C,Economidou J. Binding of serum ferritin to con-canavalin A: Patients with homozygous β-thalassaemiaand transfusional iron overload. Br J Haematol 46:409,1980.

6. Brittenham GM, Cohen AR, McLaren CE, MartinMB, Griffith PM, Niehuis AW, Young NS, Allen CJ,Farrell DE, Harris JW. Hepatic iron stores and plasmaferritin concentration in patients with sickle cell ane-mia and thalassemia major. Am J Hematol 42:81, 1993.

7. Overmoyer BA, McLaren CE, Brittenham GM.Uniformity of liver density and nonheme (storage) irondistribution. Arch Pathol Lab Med 111:549, 1987.

8. Brittenham GM, Farrell DE, Harris JW, Feldman ES,Danish EH. Magnetic-susceptibility measurement ofhuman iron stores. N Engl J Med 307:1671, 1982.

9. Brittenham GM. Noninvasive methods for the earlydetection of hereditary hemochromatosis. Ann NYAcad Sci 526:199, 1988.

10. Pootrakul P, Kitcharoen K, Yansukon P, Wasi P,Fucharoen S, Charoenlarp P, Brittenham G, PippardMJ, Finch CA. The effect of erythroid hyperplasia oniron balance. Blood 71:1124, 1988.

11. Nielsen P, Fischer R, Engelhardt R, Tondüry P, GabbeEE, Janka GE. Liver iron stores in patients with sec-ondary haemosiderosis under iron chelation therapywith deferoxamine or deferiprone. Br J Haematol91:827, 1995.

12. Brittenham GM, Griffith PM, Nienhuis AW,McLaren CE, Young NS, Tucker EE, Allen CJ, FarrellDE, Harris JW. Efficacy of deferoxamine in prevent-ing complications of iron overload in patients withthalassemia major. N Engl J Med 331:567–74, 1994.

13. Olivieri NF, Nathan DG, MacMillan JH, Wayne AD,Martin M, McGee A, Koren G, Liu PP, Cohen AR.Survival of medically treated patients with homozy-gous β-thalassemia. N Engl J Med 331:574–8, 1994.

14. Cartwright GE, Edwards CQ, Kravitz K, Skolnick M,Amos DB, Johnson A, Buskjaer L. Hereditaryhemochromatosis: Phenotypic expression of the dis-ease. N Engl J Med 301:175, 1979.

15. Niederau C, Fischer R, Sonnenberg A, Stremmel W,Trampisch HJ, Strohmeyer G. Survival and causes ofdeath in cirrhotic and in noncirrhotic patients withprimary hemochromatosis. N Engl J Med313:1256–62, 1985.

16. Bassett ML, Halliday JW, Powell LW. Value of hepat-ic iron measurements in early hemochromatosis anddetermination of the critical iron level associated withfibrosis. Hepatology 6:24, 1986.

17. Niederau C, Fischer R, Purschel A, Stremmel W,Haussinger D, Strohmeyer G. Long-term survival inpatients with hereditary hemochromatosis. Gastro-enterology 110:1107–79, 1996.

18. Loreal O, Deugnier Y, Moirand R, et al. Liver fibrosisin genetic hemochromatosis: Respective roles of ironand non-iron related factors in 127 homozygouspatients. J Hepatol 16:122–127, 1992.

19. Angelucci E, Baronciani D, Lucarelli G, Baldassarri M,Galimberti M, Giardini C, Martinelli F, Polchi P, PosizziV, Ripalti M, Nuretto P. Needle liver biopsy in thalas-saemia: Analyses of diagnostic accuracy and safety in1184 consecutive biopsies. Br J Haematol 89:757, 1994.

20. Berkovitch M, Collins AF, Papadouris D, Wesson D,Sirna JB, Brittenham GB, Olivieri NF. Need for early,low-dose chelation therapy in young children withtransfused homozygous β-thalassemia. Blood 82(Suppl1):359a, 1993.

21. Warner GT, Oliver R. A whole-body counter for clini-cal measurements utilizing the “shadow shield” tech-nique. Phys Med Biol 11:83–93, 1966.

22. Galanello R. Personal communication. January 1996.23. Olivieri NF, Matsui D, Koren G, Liu PP, Blendis L,

Cameron R, McClelland RA, Templeton DM.Reduction of tissue iron stores and normalization ofserum ferritin during treatment with the oral ironchelator L1 in thalassemia intermedia. Blood 79(10):2741–48, 1992.

24. Olivieri NF, for the Toronto Iron Chelation Group.Long-term followup of body iron in patients with tha-lassemia major during therapy with the orally active


iron chelator deferiprone (L1). Blood 88(Suppl 1):310a,1996.

25. Olivieri NF, for the Toronto Iron Chelation Group.Randomized trial of deferiprone (L1) and deferoxam-ine in thalassemia major. Blood 88(Suppl 1):651a,1996.


Homozygous β-thalassemia, a disease inwhich inadequate production of β-globinleads to severe anemia, affects thousands of

individuals worldwide. Current management of thiscondition includes the use of regular red cell transfu-sions and iron chelation therapy. The development ofan effective therapy to increase hemoglobin levels inpatients with homozygous β-thalassemia without theuse of red cell transfusions could allow normal growthand development while decreasing or eliminatingtransfusional iron overload, which remains the majorcause of death, reduced life expectancy, and morbidityin individuals with this disease.1 While bone marrowtransplantation can achieve these aims,2 it is not atherapeutic option for the majority of patients.

For some years, there has been interest in increas-ing γ-globin transcription and fetal hemoglobin(HbF) production in patients with β-hemoglo-binopathies.3,4 For patients with homozygous β-tha-lassemia, increased γ-globin production and a reduc-tion in the ratio of α- to non-α-globin could reason-ably be expected to ameliorate the severity of the ane-mia. To this end, trials of chemotherapeutic agents,including 5-azacytidine3,5–7 and hydroxyurea,3,8,9

have been conducted, but myelotoxicity, fears of long-term carcinogenesis, and only modest responses totreatment have limited the clinical usefulness of theseagents. Erythropoietin has also been used, butresponses to this therapy have been variable.10,11

There is considerable evidence that butyrateanalogs induce erythroid differentiation12–14 andstimulate HbF production in human erythroid prog-enitors in vitro.15–17 In vivo, these agents have alsobeen shown to reactivate embryonic globin produc-tion in an avian model,18 delay the switch from fetalto adult globin in ovine fetuses,19 and increase HbFproduction in adult primates.17,20–22

In humans, several fatty acids, including α-amino-butyric acid,23 arginine butyrate,24,25 isobu-tyramide,26,27 sodium phenylbutyrate,28,29 propionicacid,30 and 2-propylpentanoic (dipropylacetic) acid(unpublished data), have now been demonstrated tostimulate fetal hemoglobin production, suggestingthat they may play a role in the treatment of the β-globin disorders. However, previous clinical trials ofthese agents in β-thalassemia have been limited torelatively short-term trials of the intravenous agentarginine butyrate24,25,31 and oral isobutyramide.26,27

Sodium phenylbutyrate (SPB) is an orally admin-istered agent originally developed to promote wastenitrogen excretion in the treatment of urea cycle dis-orders,32 and it is currently used for this purpose in anFDA-approved Phase III trial. Over 100 patient yearsof experience have revealed no untoward effects ofthis drug. The finding of increased HbF levels inthese patients28 stimulated clinical trials of SPB inpatients with β-hemoglobinopathies.

We have begun a preliminary trial of oral SPB inpatients with homozygous β-thalassemia. This repre-sents the largest clinical trial to date of any hemoglo-bin switching agent used in patients with thalassemia.

Our preliminary data demonstrate that SPB cansafely be administered to patients with homozygousβ-thalassemia and is well tolerated by the majority.Problems include the need to take 40 tablets daily,epigastric discomfort, and the body odor experiencedby some patients. We expected poor compliance withthis regimen based on previous experience with thedrug,29 but surprisingly this was not the case, possiblybecause many of these patients had prior experiencewith other cumbersome medical interventions,including transfusion schedules and iron chelationtherapy. Oral administration has clear advantages overthe intravenous administration required for arginine

Hemoglobin Switching Protocols in Thalassemia:Experience with Sodium Phenylbutyrate and HydroxyureaGeorge F. Dover, M.D.


butyrate, particularly as all available evidence suggeststhat in the management of the β-hemoglo-binopathies, these therapies, if effective, will be need-ed long term.

We found that 36% of all patients (4 of 11) or50% of patients who were not transfused (4 of 8)responded to SPB. A response was defined as a sus-tained increase in hemoglobin of more than 1 g/dLover pretreatment values.33 Clearly, SPB can increasehemoglobin in some patients with homozygous β-thalassemia but is not effective in all. While it seemsevident that β-globin mutation alone does not predictresponse, the fact that two siblings treated in thisstudy both responded to SPB therapy raises the possi-bility that some other genetic factor is involved.Other genetic factors linked and unlinked to the β-globin locus have been shown to affect HbF levels innormal individuals and patients with β-hemoglo-binopathies.34–36

The failure of hemoglobin to increase in patientsshowing decreased levels of lactate dehydrogenase andindirect bilirubin is disappointing and raises interest-ing questions about the cause of these changes if theyare not related to decreased hemolysis. Similarly, wehave observed increased production in F reticulocytesin all patients treated with SPB to date. In somepatients, levels of F reticulocytes have remained high-er than baseline up to a month or more after the ces-sation of therapy, even though the agent is known tobe rapidly metabolized and excreted. Similar observa-tions have been reported following the use of argininebutyrate.24,27 This uniformity of F reticulocyteresponse, the persistence of response in some patientslong after the cessation of therapy, and the lack ofcorrelation between changes in F reticulocytes andincreased total hemoglobin or increased absolute HbFproduction may indicate increases in HbF that areinsufficient to decrease ineffective erythropoiesis.

Those patients who did respond to therapy wereinconsistent in their response. Decreases in traditionalindicators of hemolysis in all nontransfused patientswere not predictive of an increase in hemoglobin, andincreases in hemoglobin were not entirely explained byincreased HbF. This suggests that “classic” hemoglobinswitching—an increase in γ-globin production with aresultant decrease in globin chain imbalance—could

not explain the increases in hemoglobin. Three possi-ble explanations exist. SPB may have: (1) caused non-specific induction of all globin production (α, β, and γ)and not just γ alone, (2) caused nonspecific expansionof red cell mass through the release of thalassemic redcells previously sequestered in the marrow or due to anincrease in production of thalassemic red cells, or (3) prolonged red cell survival without a change in redcell production. There is evidence to support the firstof these explanations (personal communication, G.Stamatoyannopoulos), and the latter two possibilitieslead to testable hypotheses in further patients.

A positive correlation between baseline serumerythropoietin level and the potential response toSPB therapy was observed. This observation—together with the fact that erythropoietin levels inpatients with homozygous β-thalassemia are generallyelevated, but inappropriately so for the degree of ane-mia—suggests that clinical trials of combination ther-apy using erythropoietin with SPB may be of value.However, it must be remembered that in thesepatients erythropoietin levels are related to other fac-tors, such as baseline HbF percent,36 and thereforeerythropoietin may only be a marker of some otherfactor affecting response.

Both oral hydroxyurea (HU) and subcutan-eous/intravenous erythropoietin (EPO) have beenshown to increase hemoglobin levels in some patientswith thalassemia.8–10 Rachmilevitz has shown thatcombinations of these two drugs also increase hemo-globin levels, but it is not clear whether these drugstaken together are additive. Fibach has shown thatSPB and HU in human erythroid cultures have a syn-ergistic effect on increasing HbF.16 Since not all sub-jects with thalassemia respond to SPB, we believe ini-tial trials of combination therapy (HU and SPB, EPOand SPB) are warranted.

Predictors of Increased Hemoglobin inResponse to SPB Therapy

Response to SPB therapy, as defined by a sus-tained increase in total hemoglobin of more than 1g/dL above baseline, did not appear to be predicted byβ-globin mutation; baseline percent HbF, absoluteHbF, or F reticulocyte levels; baseline hemoglobin; or


baseline α- to non-α-globin ratios. Similarly, signifi-cant falls in lactate dehydrogenase and indirect biliru-bin—both traditional measures of hemolysis—couldbe demonstrated in all those nontransfused patients,with no observed differences between responders andnonresponders. Interestingly, those patients withbaseline erythropoietin levels greater than 120mU/mL were significantly (p<0.05) more likely toexperience an increase in hemoglobin (4 of 6) thanthose whose baseline erythropoietin level was below120 mU/mL (0 of 6). A similar trend existed betweenbaseline HbF percent in those patients not receivingregular red cell transfusions and their response to SPBtherapy, although this did not reach statistical signifi-cance. Of the four patients with a baseline HbF per-cent of less than 40, none responded to therapy. Incontrast, four of the five patients with baseline HbFpercent greater than 40 did respond (p=0.075).

Compliance with SPB Therapy

A 25-day supply of SPB tablets was provided tothe patients; a further supply was provided only whenthe patient specifically requested more tablets. In thisway, compliance was calculated for each patient bycomparing the number of tablets dispensed to thatprescribed. Compliance with therapy was a problemin only one patient (#3), who abruptly discontinuedtherapy after 100 days, having been 95% compliant upto that time. For the patients as a group, compliancewith medication was 97 ± 3%.

Adverse Events Occurring on Therapy

The daily dose of 20 g of SPB contributes 2,460mg (107 mmol) of sodium to the diet, a significantproportion of the recommended daily intake. Whilein the hospital, one of the 12 patients (#6) developedankle edema, associated with a 3.5% increase in bodyweight, which resolved spontaneously with dietarymodification. Following discharge from the hospital,one patient (#1) required intermittent treatment witha thiazide diuretic, and one (#8) required an increasein her previous diuretic dose to control peripheraledema. No patient developed hypertension. Epigastricdiscomfort following the ingestion of the tablets, the

most common adverse effect, was reported by 7 of the12 patients. Two patients, both splenectomized andnot on regular penicillin prophylaxis, had nonfatalepisodes of bacterial septicemia. Patient #4 developedstreptococcus pneumonia on day 71 and plesiomonasshigelloides at day 200, and patient #6 developedstaphylococcus epidermidis septicemia at day 24 relat-ed to a indwelling central venous catheter. Patient #8suffered a hemorrhage from a gastric ulcer at day 220,soon after the commencement of aspirin therapy forlong-standing pulmonary hypertension. Patient #1,who had spun hemoglobin levels of between 5.1 and7.5 g/dL associated with marked erythroblastosis,developed spinal cord compression, required irradia-tion at day 323, and ceased SPB therapy. Patient #10developed a deep venous thrombosis at day 28, with ahemoglobin level of 5.9 g/dL. Three patients experi-enced bad body odor while on therapy; in one case,the patient (#12) was unable to tolerate the medica-tion long term even at half the usual dose. Thesecomplaints are probably related to the in vivo β-oxi-dation of phenylbutyrate to phenylacetate, a com-pound with an offensive odor secreted as a defensemechanism by the stinkpot turtle.38

Arginine Butyrate, Then SPB

In collaboration with Dr. Olivieri at the Hospitalfor Sick Children in Toronto, we have treated onepatient with SPB after she was taken off intravenousarginine butyrate. The patient was treated for 30 daysand maintained her hemoglobin level, but wasremoved from therapy when she had a reoccurrence ofneurologic toxicity secondary to expansion of hermarrow, which required irradiation. Dr. Olivieri doesnot attribute this side effect to SPB, but more patientswill have to be treated to determine whether SPB cansafely maintain the hemoglobin levels of patientsalready treated with arginine butyrate.

HU and SPB

We have treated two patients with thalassemiawho are transfusion dependent with combinations ofHU and SPB. Both patients were maintained on theirusual transfusion schedule, and response was moni-


tored by their pretransfusion hemoglobin levels. Thefirst patient showed no response after 60 days and wasdiscontinued. The second patient was treated longerand has shown a steady increase in his pretransfusionhemoglobin level and a decrease in his transfusionrequirements.

Update of SPB/HU

To date, we have enrolled 14 patients onSPB/HU protocols: Four were treated throughBaltimore/Yale/Penn and 10 through Toronto. Allfour from Baltimore/Yale/Penn are off-study (threecompleted SPB/HU trials for more than 200 days,and one stopped therapy without reaching toxicitywith HU). The two patients with thalassemia inter-media showed no synergistic effect on SPB/HU, andthe patient who was transfusion dependent showedonly a questionable effect, according to Dr. A. Cohen.

We originally proposed to treat 10 patients withSPB/HU for more than 200 days. We have completedthe evaluation of five and have started two more.

We have treated the Toronto patients for a totalof 61 patient months on SPB (18 patient months onSPB/HU) on this protocol and have seen no signifi-cant increase in hemoglobin levels. The twoBaltimore/Yale/Penn patients were treated for 11months each on SPB/HU (for a total of 22 patientmonths).

Only 2 of the present 14 patients treated withSPB who are not transfusion dependent have shownany increase in hemoglobin levels with SPB alone. Anadditional patient not in the study has had an increasein hemoglobin after over 2 years of SPB, but she doesnot wish to add HU or EPO.

References

1. Zurlo MC, De Stefano P, Borgna-Pignatti C, DiPalma A, Piga A, Melevendi C, Di Gregorio F,Buratini MG, Terzoli S. Survival and causes of deathin thalassemia major. Lancet 2:27, 1989.

2. Lucarelli G, Galimberti M, Polchi P, Angelucci E,Baronciani D, Giardini C, Andreani M, AgostinelliF, Albertini F, Clift RA. Marrow transplantation inpatients with thalassemia responsive to iron chela-tion therapy. New Engl J Med 329:840, 1993.

3. Nienhuis AW, Ley TJ, Humphries RK, Young NS,Dover G. Pharmacological manipulation of fetalhemoglobin synthesis in patients with severe β-tha-lassemia. Ann NY Acad Sci 445:198, 1985.

4. Stamatoyannopoulos JA, Nienhuis AW. Therapeuticapproaches to hemoglobin switching in treatment ofhemoglobinopathies. Ann Rev Med 43:497, 1992.

5. Ley TJ, DeSimone J, Anagnou NP, Keller GH,Humphries RK, Heller P, Nienhuis AW. 5-Azacytidine selectively increases γ-globin synthesisin a patient with β+-thalassemia. N Engl J Med307:1469, 1982.

6. Dunbar C, Travis W, Kan YW, Nienhuis A. 5-Azacytidine treatment in a β°-thalassemic patientunable to be transfused due to multiple alloantibod-ies. Br J Hematol 74:467, 1989.

7. Lowrey CH, Nienhuis AW. Treatment with azacyti-dine of patients with end-stage β-thalassemia. NEngl J Med 329:845, 1993.

8. McDonagh KT, Orringer EP, Dover GJ, NienhuisAW. Hydroxyurea improves erythropoiesis in apatient with homozygous β-thalassemia. Clin Res38:346A, 1990 (abstr).

9. Fucharoen S, Siritanaratkul N, Winichagoon P,Siriboon W, Chowathaworn J, Muangsup W,Chaicharoen S, Poolsup N, Chindavijak B,Pootrakul P, Piankijagum A, Schechter AN, RodgersGP. Hydroxyurea increases HbF levels and improvesthe effectiveness of erythropoiesis in β-tha-lassemia/HbE disease. Blood 82:357a, 1993 (abstr).

10. Rachmilewitz EA, Goldfarb A, Dover G.Administration of erythropoietin to patients with β-thalassemia intermedia: A preliminary trial. Blood78:1145, 1991 (letter).

11. Olivieri NF, Sheridan B, Freedman M, Dover G,Perrine S, Nagel RS. Trial of recombinant humanerythropoietin in thalassemia intermedia. Blood80:3258, 1992 (letter).

12. Orkin SH, Swan D, Leder P. Differential expressionof α- and β-globin genes during differentiation ofcultured erythroleukemic cells. J Biol Chem250:8753, 1975.

13. Anderson LC, Jokinen M, Gahmberg CG.Induction of erythroid differentiation in the humanleukemia cell line K562. Nature 278:364, 1979.

14. Samid D, Shack S, Sherman LT. Phenylacetate: Anovel nontoxic inducer of tumor cell differentiation.Ca Res 52:1988, 1992.

15. Perrine SP, Miller BA, Faller DV, Cohen RA,Vichinsky EP, Hurst D, Lubin BH, Papayanno-poulou Th. Sodium butyrate enhances fetal globinexpression in erythroid progenitors of patients withHbSS and β-thalassemia. Blood 74:454, 1989.


16. Fibach E, Prasanna P, Rodgers GP, Samid D.Enhanced fetal hemoglobin production by pheny-lacetate and 4-phenylbutyrate in erythroid precur-sors derived from normal blood donors and patientswith sickle cell anemia and β-thalassemia. Blood82:2203, 1993.

17. Stamatoyannopoulos G, Nakamoto B, Josephson B,Li Q, Blau A, Liakapoulou E, Papayannopoulou Th,Brusilow S, Dover G. Acetate, a product of butyratecatabolism, stimulates γ-globin expression in adultcells in vivo and in culture. Blood 82:313a, 1993(abstr).

18. Ginder GD, Whitters MJ, Pohlman JK. Activationof a chicken embryonic globin gene in adult ery-throid cells by 5-azacytidine and sodium butyrate.Proc Natl Acad Sci USA 81:3954, 1984.

19. Perrine SP, Rudolph A, Faller DV, Roman C, CohenRA, Chen S-J, Kan YW. Butyrate infusions in theovine fetus delay the biologic clock for globin geneswitching. Proc Natl Acad Sci USA 85:8540, 1988.

20. Constantoulakis P, Papayannopoulou Th,Stamatoyannopoulos G. Alpha-amino-N-butyricacid stimulates fetal hemoglobin in the adult. Blood72:1961, 1988.

21. Constantoulakis P, Knitter G, StamatoyannopoulosG. On the induction of fetal hemoglobin bybutyrates: In vivo and in vitro studies with sodiumbutyrate and comparison of combination treatmentwith 5-aza C and Ara C. Blood 74:1963, 1989.

22. Blau CA, Constantoulakis P, Shaw CM,Stamatoyannopoulos G. Fetal hemoglobin inductionwith butyric acid: Efficacy and toxicity. Blood81:529, 1993.

23. Perrine SP, Greene MF, Faller DV. Delay in thefetal globin switch in infants of diabetic mothers.New Engl J Med 312:334, 1985.

24. Perrine SP, Ginder GD, Faller DV, Dover GJ, IkutaT, Witkowska HE, Cai S-P, Vichinsky EP, OlivieriNF. A short-term trial of butyrate to stimulate fetal-globin-gene expression in the β-globin disorders.New Engl J Med 328:81, 1993a.

25. Sher GD, Entsuah B, Ginder G, Dover G, Little J,Donsky J, Berkovitch M, Lewis N, Chang L,Perrine S, Olivieri NF. Intravenous infusion of argi-nine butyrate increases γ-globin mRNA expressionand F-reticulocytes in patients with homozygous β-thalassemia and sickle cell disease. Blood82:312a,1993 (abstr).

26. Costin D, Dover G, Olivieri N, Beutler E, WalshCT, Torkelson S, Pantazis C, Brauer M, Faller DV,Perrine SP. Clinical use of the butyrate derivativeisobutyramide in the β-globin disorders. Blood82:357a, 1993.

27. Perrine SP, Olivieri NF, Faller DV, Vichinsky EP,Dover GJ, Ginder GD. Butyrate derivatives: Newagents for stimulating fetal globin production in theβ-globin disorders. Am J Ped Hem Onc 16:67, 1994.

28. Dover GJ, Brusilow SW, Samid D. Increased fetalhemoglobin in patients receiving sodium 4-Phenylbutyrate. New Engl J Med 327:569, 1992a(letter).

29. Dover GJ, Brusilow SW, Charache S. Induction ofHbF production in subjects with sickle cell anemiaby oral sodium phenylbutyrate. Blood 84:339–343,1994.

30. Little J, Tuchman M, Ginder GD. Elevated fetalhemoglobin levels in propionic acidemia. Clin Res42:238A, 1994 (abstr).

31. Perrine S, Dover G, Costin D, Pantazis C, EmburyS, Lazzari G, Vichinsky E, Daftari P, Xin A,Olivieri N. Correction of globin chain imbalance inthalassemia major by arginine butyrate therapy.Blood 82:312a, 1993b (abstr).

32. Brusilow SW. Treatment of urea cycle disorders. InTreatment of genetic disease, ed. Desnick RJ, p. 79.Churchill-Livingstone, 1991a.

33. Collins AF, Pearson HA, Giardina P, McDonaghKT, Brusilow SW, Dover GJ: Oral sodium phenyl-butyrate therapy in Homozygous beta thalassemia:A clinical trial. Blood 85:43–49, 1995

34. Dover GJ, Smith KD, Chang YC, Purvis S, MaysA, Meyers DA, Sheils C, Serjeant G. Fetal hemo-globin levels in sickle cell disease and normal indi-viduals are partially controlled by an X-linked genelocated at Xp22.2. Blood 80:816, 1992b.

35. Thein SL, Sampietro M, Rohde K, Rochette J,Weatherall DJ, Lathrop GM, Demenais F.Detection of a major gene for heterocellular heredi-tary persistence of fetal hemoglobin after accountingfor genetic modifiers. Am J Hum Genet 54:214, 1994.

36. Galanello R, Barella S, Turco MP, Giagu N, Cao A,Dore F, Liberato NL, Guarnone R, Barosi G.Serum erythropoietin and erythropoiesis in high-and low-fetal hemoglobin β-thalassemia intermediapatients. Blood 83:561, 1994.


The molecular basis of β-thalassemia inter-media is complex and puzzling. The molec-ular causes of this condition are myriad.

Individual variation and genetic modifiers that arepoorly understood lead to great variation in geno-type-phenotype correlations. For example, somepatients who are homozygous for the IVS2, nt 1mutation have β-thalassemia major, while othershave β-thalassemia intermedia. Clearly, one canmake generalizations about the molecular basis ofβ-thalassemia intermedia, but these generalizationsfail to hold when one considers the individualpatient.

It is still possible, however, to make some senseof the molecular basis of the condition.Thalassemias are due to an imbalance of globinchain synthesis. In the case of β-thalassemia inter-media, the imbalance is greater than that seen in β-thalassemia trait and less than that of β-thalassemiamajor. In general, most individuals with tha-lassemia intermedia are homozygotes or compoundheterozygotes with either two mild β-thalassemiaalleles or two severe alleles plus a modifier such asα-thalassemia. Uncommonly, two severe alleleswith no known modifier can unexpectedly lead tothe milder intermedia state. Conversely, β-tha-lassemia intermedia with one defective β-globingene is usually associated with unusually severeimbalance in globin synthesis due to a very severeβ-globin mutation or excess α-globin genes. Veryrarely, the genetic modifier altering the phenotypein a patient carrying a single defective β-globingene is not found.

We begin by discussing patients with two defec-tive β-globin genes. In most instances of β-tha-lassemia intermedia, both β-globin genes are defec-tive and carry mild mutations. These mild alleles are:

1. Promoter mutations, the most important of whichare -101(C-T) in Italians, -88 (C-T) in blacks, -87(C-G) in Mediterraneans, and -29 (A-G) inblacks;

2. The cap site mutation, +1 (A-C), in AsianIndians;

3. The HbE mutation (codon 26 G-A);4. The IVS1, nt 6 mutation in Mediterraneans; and5. Deletion mutations such as those that cause δβ-

thalassemia, Hb Lepore, or HPFH.

Homozygosity for these alleles also produces dif-ferent phenotypes. For example, homozygosity for -101 is extremely mild, for -88 leads to β-thalassemiaintermedia, for -29 in blacks can produce a state ofwellness indistinguishable from “normal,” for +1 inAsian Indians can appear as mild β-thalassemia trait,for HbE is usually similar to β-thalassemia trait ormild thalassemia intermedia, and for IVS1, nt 6 isusually classic β-thalassemia intermedia. Many indi-viduals who are genetic compounds with one of thesemild alleles in combination with a second severeallele, such as a nonsense or frameshift mutation, haveβ-thalassemia intermedia.

In patients carrying two severe β-globin muta-tions, the major modifier is concomitant α-globingene deletion, with loss of either one or two α-globingenes. Since chromosome 16s harboring only a singleα-globin gene are fairly common among β-tha-lassemia populations (including Southeast Asians,blacks, and Mediterraneans), individuals carrying only2 or 3 α-globin genes and two severe β-globin allelesare not uncommon in many parts of the world. Theseindividuals usually have a β-thalassemia intermediaphenotype. Presumably, since both α- and β-globindefects protect against falciparum malaria in het-erozygotes, gene frequencies of these defects increased

The Molecular Basis of Thalassemia Intermedia

Haig H. Kazazian, M.D., and Catherine A. Stolle, Ph.D.


in parallel in malarial-infested regions of the world.As mentioned above, a minority of patients with

intermedia can have two severe β-globin gene defects,no α-globin deletion, and a milder, transfusion-freecourse. The etiology of this milder phenotype in thesepatients is unknown.

Occasionally individuals who are merely het-erozygous for a severe β-thalassemia allele have a mildthalassemia intermedia phenotype due to either a verysevere allele producing an unstable β-globin chain orproduction of excess α-globin chains. In the firstgroup are individuals who appear to be either newcases without a family history or autosomal dominantcases of the disorder. These patients usually have amutation in exon 3 of the β-globin gene that pro-duces significant instability of the globin, inclusionbodies in the marrow red cell precursors, and a tha-lassemia intermedia phenotype. Globin chain imbal-ance becomes greater than in β-thalassemia trait dueto a dominant negative effect of the abnormal β-glo-bin which produces precipitation of normal β-globinchains in the hemoglobin tetramer.

In the second group are a small number of indi-viduals in whom triplicated α gene clusters have beenfound along with a single β-globin defect. Twopatients with 6 α-globin genes and a single severe β-globin mutation were seen in the Punjab region ofIndia. Two other patients with 5 and 6 α-globingenes were found in Israel. Other isolated cases ofindividuals carrying 5 α-globin genes and one normalβ-globin gene have been seen in Italy. All of theabove patients had thalassemia intermedia.

Other individuals heterozygous for a common β-thalassemia allele such as nonsense codon 39 may alsohave mild thalassemia intermedia. The reason for thisvariation from the expected β-thalassemia trait inthese individuals is unknown. Speculations about therole of sequence variants in the variable region rough-ly 500 bp 5' to the β-globin gene and in the secondhypersensitivity region (HS2) roughly 10 kb 5' to theε-globin gene are still just that—speculations. Nodefinitive proof that these sequences are genetic mod-ifiers of phenotype has been put forward.

Our workup for thalassemia intermedia in thelaboratory starts with a family history and, if possible,hematological data on the parents. If both parents

appear to have β-thalassemia trait, we expect to findmutations in both β-globin genes of the patient. Ifone parent looks like a silent carrier with perhaps aHbA2 that is slightly elevated and an MCV around78–80, we look for one silent carrier (i.e., mild) allelein the patient, such as -101 or the +1 mutation,depending upon the ethnic group of the patient. Ifone parent has a phenotype similar to the patient, welook for an autosomal dominant mutation in exon 3.Without any clues from the history or hematology ofthe parents, we first look for mutations in both β-glo-bin genes. If two mutations are found that explain thephenotype, we often stop the analysis. If two severealleles are found, we analyze the α-globin genes fordeletions. If only a single β-globin defect is found, welook for excess α genes to explain the intermedia phe-notype. If after this analysis, only the single β-globinmutation has been observed, we then sequence the β-globin gene and its environs to find any new, raremutations and carry out Southern blot analysis todetermine the presence of a gene deletion involvingthe β-globin cluster. This complete analysis provides asatisfactory explanation for the thalassemia phenotypein well over 90% of patients.


Q:How clinically useful is genotypic infor-mation in practice?

Dr. Giardina: Here is an anecdote of aGreek family whose first child had thalassemia majorwith an intervening sequence IVS2, nt 745 and a β°-39 combination. Their second child was a heterozy-gote. Prenatal testing was done in Dr. Kazazian’s lab-oratory, and fortunately that child was an HLA matchwith the older sibling. A successful bone marrowtransplantation took place 2 years ago. The motherbecame pregnant again, and on this occasion the fetuswas found by chorionic villus sampling to havehomozygous β-thalassemia. After being informed ofthis, she decided to keep the pregnancy. The child wasborn, and the lovely part of the coincidences of goodfortune is that this child also has an HLA match withhis older middle sibling. That is good fortune, andthose lovely things don’t always happen in this world.I presented this anecdote to some geneticists whohave been trained in Europe, and they are shockedthat I allowed a pregnancy with a homozygous-affect-ed fetus to continue. I hope that in the United Stateswe hematologists and geneticists never have to reachthat level of withdrawing choice from an affectedcouple.

Dr. Kazazian: The parent’s freedom to choose isclearly the situation in this country, and everything isvoluntary. I don’t personally know of situations wherethe severity of thalassemia is known from a previouschild where in a second pregnancy the couple hasdecided to carry the unborn affected fetus to term. Ido know of times when the fetus’ genotype modifiesthe counseling. There are children of parents withmild alleles or a combination of one mild allele andone severe allele where we informed the couple that

there was a reasonable chance of thalassemia interme-dia and the fetus was carried to term. I know of plentyof individuals with other diseases—for instance, sicklecell anemia—where, after prenatal diagnosis of dis-ease, about 40% of pregnancies are terminated andabout 60% are carried to term. For factor 8 def-iciency, hemophilia, I think it is of the same order ofmagnitude: something like 50% termination, 50%completion of pregnancy. We try to individualizecounseling but believe the decision is up to the coupleas to what happens.

Dr. Forman: The mother spoken of earlier knewwhat she was getting into better than we do, havinglived through the first child’s thalassemia. But I amcurious as to why people want to be tested if theyknow ahead of time that they won’t act on the infor-mation. Do they change their minds afterwards, orare they doing it to prepare for the future if they planto carry the affected fetus to term?

Dr. Giardina: I think they are preparing. I thinkthey may not know what they will do for sure in theevent of another affected child. They are figuring thatthere is a three out of four chance that they won’thave an affected child.

Dr. Dover: I was very impressed earlier todaywith Dr. Giardina’s discussion and the consortiumpulling together the thalassemia intermedia data. Ithas been complemented this afternoon by Dr.Kazazian’s genotyping of patients. However, what isabsolutely essential to really interpret this data is toask those centers who supplied data about these 71patients with thalassemia intermedia to also randomlysend Dr. Kazazian’s lab DNA from patients with thalmajor. It will not be clear to me that we truly under-


Questions and Comments

stand the real phenotype/genotype relationship ofthalassemia intermedia until I see that there really is adifference between thal intermedia and thal majorgenotypes. If one is going to make conclusions aboutdisease severity on the basis of these genotypes, it isnecessary to demonstrate that the people with thalmajor in these same centers have different mutations.

Dr. Kazazian: Although they weren’t randomized,Dr. Giardina has sent us blood from 40 or more otherpatients, and I don’t know whether they had thalmajor or intermedia. She could easily go back andfind out what they had and give you an answer to thatquestion, at least from her series.

Dr. Dover: The homogeneity that we see in thedifferent centers in Europe is very different from whatwe find in the United States. We see so many differ-ent mutations here, and my experience was that whenI looked at patients who had “thal major” or who wereon transfusions and I compared them to the ones with“thal intermedia,” there weren’t major genotypic dif-ferences.

Dr. Kazazian: We looked at them in Sicily withDr. Maggio, and clearly they were very different.There was a big difference in the distributions ofmutations in those with major versus intermedia. Thecases of thalassemia intermedia were loaded withIVS1 nt6, and the majors had very few. In addition,the total number of frequent mutations in the regionwas small—five or six.

Dr. Olivieri: In Milan, it’s not an homogenouspopulation. Milan is a center in northern Italy thatgets a lot of referrals, so it has a widespread number ofmutations. What they found was (although you worrythat physicians don’t uniformly transfuse for the sameindications) there weren’t major differences betweenthal “major” and thal “intermedia” genotypes.

Dr. Dover: That’s my point about this analysis.Although it intuitively makes sense to me, genotype/phenotype relationships may not hold up. That’s veryimportant if you are going to start using this informa-tion for counseling or if you are going to start asking

me to select patients for some form of pharmacologicmanipulation of fetal hemoglobin.

Dr. Kazazian: I really do think that the thalmajors have the severe mutations.


Although thalassemia intermedia does nothave a unique association with infectiouscomplications, it shares features with other

hematologic disorders that predispose patients whoare affected to serious risks of infections. For example,since most patients with thalassemia intermediaundergo splenectomy, the problem of postsplenecto-my sepsis is as relevant to the care of patients withthis disorder as it is to the care of patients with tha-lassemia major, sickle cell disease, and idiopathicthrombocytopenic purpura (ITP). Similarly, ironoverload and iron chelation therapy are often part ofthe clinical course of patients with thalassemia inter-media, as well as patients with transfusion-dependentblood disorders, which raises the possibility of infec-tion with organisms that thrive in an environment inwhich iron or a siderophore is abundant.

Splenectomy

Before discussing the problem of postsplenectomysepsis, it is important to identify the common indica-tions for splenectomy in thalassemia intermedia.Splenectomy is generally undertaken in patients withthalassemia intermedia who experience weakness orchronic fatigue due to their anemia. In addition,patients with impaired growth or development maybenefit from splenectomy. The appearance of earlychanges in the facial bones or the finding of corticalthinning in the long bones due to expansion of thebone marrow may warrant splenectomy as a means ofraising the hemoglobin level, and perhaps decreasingthe amount of erythropoietic activity. A similar ratio-nale for splenectomy may apply to patients with mas-sive splenomegaly or extramedullary hematopoiesis.Finally, patients who experience a steadily falling

hemoglobin level, particularly in the second or thirddecade of life, or patients who require repeated redcell transfusions because of frequent decreases in theirbaseline hemoglobin values may benefit from splenec-tomy. In each of these instances, splenectomy mayonly be of partial benefit, and regular blood transfu-sions similar to those used in thalassemia major maybe required.

In a study of 37 patients with thalassemia inter-media in Great Britain, Modell and Berdoukas foundthat 19 were splenectomized and 18 were nonsplenec-tomized.1 The median age at splenectomy was 8years. The hemoglobin levels rose from a mean valueof 7.1 g/dL prior to splenectomy to 8.9 g/dL aftersurgery. Data collected from 71 North Americanpatients for this conference were remarkably similar tothe British data in regard to median age at splenecto-my (9 years) and mean presplenectomy and post-splenectomy hemoglobin levels (6.8 and 8.5 g/dL,respectively). Of the 71 patients, 56 underwentsplenectomy and 15 had intact spleens.

Postsplenectomy Sepsis

The problem of postsplenectomy sepsis withencapsulated organisms has been recognized for manyyears. These infections are often abrupt in onset andrapidly fatal. For unknown reasons, patients with tha-lassemia were believed to be at particular risk forpostsplenectomy sepsis compared with patientssplenectomized for other reasons. Singer compileddata from 24 series, as well as from his own center,and found that 25% of 109 patients undergoingsplenectomy for thalassemia developed postsplenecto-my sepsis.2 In contrast, only 1.4% and 2.0% ofpatients undergoing splenectomy for trauma and ITP,

Infections in Thalassemia Intermedia

Alan R. Cohen, M.D.


respectively, developed postsplenectomy sepsis. Of the27 patients with thalassemia who developed seriousbacterial infections, 12 died.

Subsequent studies have demonstrated a lowerbut still significant risk of postsplenectomy sepsis inpatients with thalassemia. Issaragrisil et al. studied1,018 patients with hemoglobin E-β-thalassemia inThailand and found a 4% rate of postsplenectomysepsis among 228 splenectomized patients.3 Themortality rate of 89% was unusually high. Among 101patients with thalassemia major and 18 patients withthalassemia intermedia studied by Modell andBerdoukas, 5 patients had serious bacterial infections,including peritonitis, osteomyelitis, and meningitis.4

However, no episodes of sepsis were recorded. Thelower rates of postsplenectomy sepsis found in morerecent studies may reflect the use of prophylacticpenicillin, immunization with pneumococcal vaccine,and delay of splenectomy beyond the age of 3 years.

The effectiveness of prophylactic penicillin inpreventing serious bacterial infection in patients withabsent splenic function has been well demonstrated insickle cell disease.5 However, compliance remains animportant issue. Borgna-Pignatti et al. studied com-pliance with prophylactic penicillin in 42 splenec-tomized patients with thalassemia.6 These investiga-tors collected five random urine samples at the timeof transfusion therapy. In 21% of patients, two orfewer of the five urine samples were indicative ofrecent penicillin intake. The major risk factor for non-compliance with prophylactic penicillin in the studywas a duration of more than 6 years after splenecto-my. Patient age and history of previous infection didnot predict compliance patterns. Most interestingly,physicians involved in the patients’ care were unableto successfully predict which patients were compliantwith penicillin therapy.

Because of the concern about postsplenectomysepsis, partial splenectomy has been considered as analternative for patients with thalassemia intermedia.Investigators in Paris reporting on six patients foundthat the mean hemoglobin level increased in the firstyear after surgery.7 However, the results of a singlepatient with more than 2 years of followup offer apossible warning about the long-term value of this

procedure. This patient had an initial increase inhemoglobin level, from 5.0 to 8.0 g/dL. In the secondyear, the hemoglobin fell to 6.5 g/dL. By the fourthyear, the mean hemoglobin was 7.0 g/dL, but thepatient required three transfusions. In light of theextraordinarily strong erythropoietic drive found inthalassemia intermedia, regrowth of the spleen re-mains a distinct possibility, and the long-term benefitsof partial splenectomy, both in regard to risk of post-splenectomy sepsis and overall improvement in thehematologic condition, must be carefully assessed.

Yersinia enterocolitica Infection

Many bacteria rely on iron as an essential nutri-ent. Most of these organisms have established meth-ods for retrieving iron from the environment, usuallythrough production of a siderophore. Indeed, chela-tors developed for human application, includingdeferoxamine, may be derived from these sidero-phores. Yersinia enterocolitica does not produce its ownsiderophore, and it is usually found in environmentssuch as the gut, where it can take advantage ofsiderophores produced by other bacteria. Experimentsin mice have demonstrated the increased lethality ofYersinia enterocolitica when iron, deferoxamine, or acombination of the two is added to the animal.8 Forexample, in mice, the LD50 of two of the serogroupsof Yersinia enterocolitica that most commonly causeinfection in humans is greater than 10.8 However,when the animals are pretreated with deferoxamine,the LD50 falls as low as 101.1, and pretreatment ofthe mice with iron followed by deferoxamine reducesthe LD50 below 10.

The human counterpart of these animal experi-ences can be found in patients with iron overload. Inan extensive review, Blei et al. found 47 reports ofserious infection with Yersinia enterocolitica in patientswith thalassemia major.9 Thirty-five of the 47patients had septicemia, and an additional 10 patientshad abdominal infections, including mesentericadenitis, peritonitis, or abdominal abscesses. Thirtypatients were receiving deferoxamine, 6 were receivingno chelation therapy, and information was unavailablefor 11 other patients. These data demonstrate the sig-


nificant problem posed by Yersinia enterocolitica inpatients with iron overload, particularly those receiv-ing chelation therapy.

Summary

Infections remain a serious problem for patientswith thalassemia, as they do for other patients whohave undergone splenectomy or who have excessiveamounts of iron and require chelation therapy. Inassessing the overall risks associated with this surgicalprocedure, it would be helpful to be able to character-ize the risk of postsplenectomy sepsis for patientswith thalassemia intermedia receiving presplenectomyimmunizations and postsplenectomy prophylacticpenicillin. As new iron chelators are developed, theirability to enhance bacterial growth should be carefullymonitored. The prevention of infection in patientswith thalassemia may become increasingly importantas the improved overall management of the diseaseextends the length and quality of life of affectedpatients.

References

1. Modell B, Berdoukas V. The clinical approach to tha-lassemia, p. 244. London: Grune & Stratton, 1984.

2. Singer DB. Postsplenectomy sepsis. Perspect PediatrPathol 1:285–311, 1973.

3. Issaragrisil S, Wanachiwanawin W, Bhuripanyo K, etal. Infection in thalassemia: a retrospective study of1018 patients with β-thalassemia/HbE disease. BirthDefects 23:505–511, 1988.

4. Modell B, Berdoukas V. The clinical approach to tha-lassemia, p. 142. London: Grune & Stratton, 1984.

5. Gaston MH, Verter JI, Woods G, et al. Prophylaxis with oralpenicillin in children with sickle cell anemia. A randomizedtrial.N Engl J Med 314:1593–1599,1986.

6. Borgna-Pignatti C, DeStefano P, Barone F, et al.Penicillin compliance in splenectomized thalassemics.Eur J Pediatr 142:83–85, 1984.

7. de Montalembert M, Girot R, Revillon Y, et al. Partialsplenectomy in homozygous β-thalassaemia. Arch DisChild 65:304–307, 1990.

8. Robins-Browne RM, Prpic JK. Effects of iron anddesferrioxamine on infections with Yersinia enterocoliti-ca. Infect Immun 47:774–779, 1985.

9. Blei S, Puder DR. Yersinia enterocolitica bacteremia in achronically transfused patient with sickle cell anemia.Case report and review of the literature. Am J Ped HemOnc 15:430–434, 1993.


Dr. Forman: This has been a very interestingconference. I would like to ask the speakersto come up and sit in front. This is the

moment for the audience to ask their questions andfor the speakers to ask each other questions. Wehave an additional member of our panel: Dr. CarolHyman from California, who has considerableexpertise in the long-term care of patients, will joinDrs. Pearson, Giardina, Cohen, Olivieri, andKazazian.

Q: I have a question for Dr. Pearson. One ofthe discussions earlier today dealt with trying totreat patients with transfusion therapy before bonychanges become pronounced. I have had the expe-rience recently of being in a situation where theparents want to believe that the child looks normaland it is clear that the child does not. I don’t thinkthat is an unusual situation. It’s very hard not onlyto be the bad guy who says “I think we need to goto transfusion therapy” (which is something I thinkmost families dread), but on top of it you are sayingthe reason why you need to do it is that the childlooks funny. What do you do with that situation?Have you encountered it? I see some noddingheads, so I doubt we are the only ones.

Dr. Pearson: I think that serial photographshave been most helpful. Ask them to take a photo-graph of the child at regular intervals. Even a par-ent who lives with the child’s appearance every dayand becomes immune to changes can see the pro-gression. I also found that showing them the x-rays is helpful. If you can show the hair-on-endappearance with a skull that is thick and put itbeside a normal skull, even a doubtful person canrealize there are differences.

Q: How can we prevent the bone problems asso-ciated with bone marrow expansion?

Dr. Olivieri: Dr. Beatrix Wonke has used biphos-phonates in thalassemics in the United Kingdom, butI think that her studies were done primarily in adults.Bone disease doesn’t always reverse, and there are painproblems with that. We await the outcome of thatinteresting study.

Dr. Giardina: We have been impressed that someof the older women with thalassemia intermedia haveessentially had secondary amenorrhea and are hypoe-strogenemic. This adds the problem of low estrogen tothe already sick bones. I am convinced that thisaccounts for some of the fractures we have seen, and itis why we have been very aggressive in providing estro-gen replacement to adolescent women who need it.

Dr. Olivieri: I agree. We did a survey of 58 adultswith thalassemia intermedia. In about 35 of them welooked at calcium bone index, bone indices, as well asquantum digital radiography. The two risk factors fora low bone density were hypogonadism and the irreg-ularity of transfusion. Even people with thalassemiaintermedia and good sexual development had a lowbone density, but not quite as low as a person withthalassemia major or intermedia and hypogonadismas well.

Q: Can transfusion therapy help prevent fracturesin an adult who has severe osteopenia?

Dr. Pearson: I don’t know about adults, but cer-tainly with adolescents transfusions can cause reversaland improvement of the bone structure. I don’t knowwhat the critical age is. If you have a patient with ter-

Panel Discussion


rible cosmetic problems, can you improve his facialappearance? I know that you can in childhood, per-haps early adolescence, but I am not sure about adultsafter their bone growth becomes fixed. I have alsoseen a few patients who had been on transfusionswho, when they went off transfusion, had subsequentprogression of their cosmetic abnormalities in laterlife.

Dr. Cohen: A single case may or may not beinstructive, but we encountered a gentleman who wasprobably in his late 20s when he first came to us withdebilitating back pain. He had terrible osteopenia,was literally unable to stand up straight, was hunchedover and could barely move to get on the examiningtable. Then he went on chronic transfusions, and hewas very, very dramatically improved after about 3months. It was something of a miracle. I think it is atleast worth a shot to see whether there would besymptomatic improvement in adult patients who havesevere bone disease.

Q: Did this patient have vertebral compressionfractures?

Dr. Cohen: Yes.

Dr. Hyman: We have had a couple of patientswith very severe thalassemia intermedia. One youngwoman as a child had broken many of her bones. Shewould be in the hospital most of the year with frac-tures, and every time they would move her from onebed to another she would fracture another bone. Ahypertransfusion program was begun for that patientin order to shut off her hematopoesis. Another of ourpatients with thalassemia intermedia had the mostsevere facial changes I have ever seen. It took threemajor facial surgeries and an aggressive transfusionprogram to maintain the correction of facial bonegrowth. Interestingly, once her face looked “normal,”it took her over a year to adjust emotionally to thechanges.

Dr. Crocker: There has been no comment yettoday from any of the speakers on the issue of self-image or self-identity, especially in this condition that

the caregivers are finding hard to define. If you havethal major, you sort of belong to “the club.” Your ill-ness is acknowledged, being maximally challenged,and there is a whole set of precepts required to getthrough life with the best outcome in terms of sur-vival and control of morbidity. If you are an individualwith thal trait, presumably your agonies are minor andthis can be incorporated with comfort, particularly ina pedigree where you happen to be a winner. Butwhen you have something in between that is almostas serious but not necessarily, and really isn’t wellunderstood by anyone, and in textbooks it is coveredonly extremely vaguely and ambivalently, and we arenot sure whether it is one disease or six diseases.

Dr. Pearson: Are you referring to my chapter?

Dr. Crocker: This is a little reminiscent of ameeting I just came from, an annual convention ofthe MPS (Mucopolysaccharidosis Society). There is asort of strata in which there are very severely affectedchildren, extremely mildly affected ones, and also agroup of in-between ones, such as those with delayedonset Hunter syndrome. These children and theirfamilies are trying to figure out where they stand:Should they be worried about themselves, or are theyreally okay, and what’s up? How do you give a personwith moderate clinical involvement with thal inter-media a sense of who he is and what’s going on?

Dr. Olivieri: I appreciate what you are saying, andit is something that our thal intermedia patients talkabout. In fact, the Diamond Blackman patients thatwe see in the transfusion clinic say the same things.“We can’t get the oral chelator, we don’t have any realsense of the field moving forward, and there are just14 of us in the clinic.” In our program, we do see thethal intermedia patients frequently and hover overthem more than we do the thal majors, so their con-nection to us is just as close as the patients with thalmajor, although perhaps not as frequent. We havetried to make an effort to include them in the pro-gram. In fact, many of those patients we worry moreabout than the standard transfused patient. In ourprogram, we have been making an effort to useBernadette Modell’s survey of quality of life for tha-


lassemia major, and we apply it to everyone. We arejust in the middle of analyzing that now. I don’t reallyhave any idea if any of the attempts to involve themclosely make a difference.

Ms. Kurth: We have a patient in Boston whoseparents were told, “Your child has thalassemia inter-media, but your child also needs transfusions.” Nowthe child is 3 years old after being started on transfu-sions at about age 2. He is doing well on transfusions,is growing well at the 90th percentile, but has an anx-ious mother. As I was talking to her the other day, Ispoke of him as though he had thalassemia major.And she said, “Dr. Forman told me he has thalassemiaintermedia, and now you mean he has thalassemiamajor. Is he worse?” I had quite a bit of backpedalingto do to try to relieve her anxiety. She is more worriedthan most, but it shows the dilemma that you have inexplaining the diagnosis to families and having themunderstand what it predicts and what it doesn’t pre-dict. I tried to stress that although your child needstransfusions, and that’s all we really know, he is doingvery well now and is expected to continue to do well.

Dr. Pearson: I would be very reluctant to make adefinitive diagnosis of thalassemia intermedia under 4years of age, because there are certain children whosehemoglobin falls gradually, and it isn’t clear what isgoing on until they are 4 or 5 years old.

Q: How reliable is ferritin as a measure of ironburden?

Dr. Olivieri: What you are asking is, “In theevent that you cannot do a quantitative assessment ofiron such as a liver biopsy or you are not close toCleveland and you can’t do a SQUID, how effectivean estimate of body iron burden is the serum ferritinconcentration?” It is certainly true that we have allused the serum ferritin to estimate how a patient isdoing. When we have looked within 14 months of aliver biopsy, there is a very good correlation betweenferritin and hepatic iron, but the confidence intervalsare so wide that a ferritin of 2,000 identifies a liveriron of between 2 and 22 micromoles per gram dryweight. You could be 10 times off using ferritin. So I

don’t think it is useful for quantitative assessment, andthe problem is that if the patient has another problemsuch as hepatitis C or alcohol use, then you may havea more complicated situation. If we want to makesome conclusions from this conference, there are cer-tain important things that kill people with tha-lassemia intermedia: infections, complications ofsplenectomy, iron overload. Maybe these things couldbe addressed by the group of people here so the nextconference in less than 11 years could actually answersome prospective questions on this. It is difficult toget liver biopsies, and you really need an infrastructureto do it. We had a lot of trouble with our hepatolo-gists, so we went to our invasive radiologist, and nowthey do two biopsies under ultrasound guidance on aweekly basis. Physicians really could make more useof the SQUID technology, and hopefully NIH willhelp to make that more available. A ferritin in Canadacosts approximately $33 an assessment, and the biop-sy costs us just the radiology time, so we figure if weget that one biopsy every 18 months we are actuallysaving because we have an exact measure, whereas weare doing a ferritin estimate six times in those 18months. Dr. Cohen is now going to comment on howhe agrees with me.

Dr. Cohen: The problem with this question isthat Dr. Olivieri knows exactly what I am going to sayand I know exactly what she is going to say. My sug-gestion to Dr. Olivieri to cut her costs down is to stopdoing ferritins every 3 months. You have to ask thequestion with any test you order, which is, “What amI going to do with the information I get, and howimportant is that versus the risk of the procedure I amabout to do?” If it is a venipuncture, it is less of adebate. If it is a liver biopsy, no matter who is doingit, it certainly causes more of a debate, certainly forthe families as well as physicians. I am not antibiopsyor antiliver iron, but I do think one has to look at thewhole patient and reach a conclusion as to how muchinformation you are going to get. I think there are sit-uations where it is invaluable, and we have seen some,and Dr. Olivieri has gloated when we have. I thinkthere are other situations where we have had verygood information based on knowing the family,knowing the ferritin level over time, knowing the


child, and to have subjected that child to a liver biop-sy might have been considered inappropriate.

Dr. Olivieri: You can’t predict that situation neces-sarily in advance. Are you saying if you knew this fam-ily was compliant, and you knew the relative Desferalconsumption and the relative transfusion load, thatyou could estimate that those patients would have liveriron between 3 and 7? If you don’t do the biopsy, youdon’t know that.

Dr. Giardina: We have all been in this business solong that we have had personal experience with thingsthat were in vogue. Twenty years ago I was doing liverbiopsies, and I was measuring serum ferritins, and Irecognized the fact then that in a population of 88patients or so the serum ferritins were all over theplace and they would jump around by 500 or 700. In1984, when I was trying to compare high respondersand low responders to chelation therapy (based on fer-ritin levels), that ferritin just wasn’t helpful. I treatedmy patients as if ferritin didn’t exist, but I measuredthem. And I told my patients, “Just use your Desferal,don’t worry about your ferritins.” I did liver biopsies,and the liver biopsy percents dry weight of iron wereall over the place as well. This wonderful man I workwith, Bob Grady, said, “Pat, the specimens are toosmall, your patient population is so variable becausethere is cirrhosis, and you can’t get an accurate mea-surement.” So I stopped looking at liver biopsies, and Itold my patients, “Just use your Desferal.” In close to20 years, actually, we have achieved something. We seeit in the fact that the patients’ lives have been pro-longed and that they are living healthier lives and bet-ter quality lives. I guess I am not being very much of ascientist today, but I have a feeling we are just shiftingour horizon. There is something else now in vogue,and it is called a SQUID, but I am not really quitesure that getting SQUID measurements on a yearlybasis is going to afford us anything for the patient. Weare going to have another parameter to look at interms of analyzing data, but is it really going to helpus? Forgive me. I know that we should be seeking anyadditional parameter to evaluate iron overload, andmaybe the SQUID is it, but maybe it isn’t.

Dr. Olivieri: Iron overload is the life-limitingcomplication of thal intermedia. I am not interested indoing SQUIDS because it is in vogue. I happen to livein Toronto, and Cleveland isn’t far away by plane, andI am glad I don’t have to do a liver biopsy every 18months. One should use the most accurate determina-tion of body iron burden. I have just told you that thedata indicate that a ferritin level can identify a liveriron of between 2 and 22. There is a huge variability,up to 10 fold. The bottom line is that determinant ofhepatic iron is not simply “in vogue,” it is the best wayone can quantitatively assess body iron burden. I havethat in mind for the patients—not for studies, but forpatients. We have made mistakes with deferoxaminetoxicity. We have underdosed patients. We have over-dosed patients. I do agree with Dr. Cohen’s view thatit is difficult to convince families to have biopsies.There’s no question he is right, and that’s why SQUIDtechnology is useful. I don’t view that just going andsaying “Do your Desferal” is useful. We still havepatient deaths. The important thing in thal intermediais that there are other complicating factors that mayconfuse the serum ferritin versus hepatic iron concen-tration, such as ascorbate deficiency. So we really, reallydo need a quantitative assessment, and there really isn’tanother one right now.

Dr. Hyman: First let me thank both Dr. Formanand Lauren Berman for asking me to participate in thepanel, especially since I am from the West Coast andhave opposite viewpoints—mainly because I don’t geta chance to talk to anyone there, so I am sort of bymyself. As far as the ferritin is concerned, I have seennumbers of patients who receive ascorbic acid dailyand still run serum ferritins of 200, 300, and 400, andthey pour out iron when you give them Desferal. Myattitude is to beware of the interpretation of low serumferritins. On the other hand, if you have patientswhose ferritins are coming down after starting in thethousands, when you get below 1,000, you have towatch out on the dose that you don’t get Desferal toxi-city.

Dr. Forman: Let me raise a more benign chal-lenge. I will ask Dr. Hyman to speak first. Whenshould splenectomy be performed?


Dr. Hyman: Well, I am being asked the questionbecause everyone here knows that I am a rather anti-splenectomy person. Splenectomy has to be looked atboth short-term and long-term for what it does for thepatients and also what happens to the patients whenthey are adults. One of the things that happens as thepatients grow older is that they get more hypoxic andrequire higher pretransfusion hemoglobins as they agebecause of chronic peripheral hypoxemia. We alsoknow that a lot of them die from cardiac disease,which we say is due to iron overload. There is someevidence from different sources showing that some ofthe cardiac problems in the thalassemics at least arestarting on the right side of the heart, not the left,even though they eventually develop left-sided heartfailure. Why is that? Is it just iron? There are a lot ofpathology data showing there are chronic frequentthromboemboli all over the lungs and that this is caus-ing some of the AV shunting. It causes little spiderhemangiomas under the subendothelium. A largenumber of patients develop pulmonary hypertensionwhen they get older. Dr. Loukopoulos’ group has apaper out where they have shown a large number ofpeople with thalassemia intermedia who developedpulmonary hypertension. The vast majority of themhad their spleens out, and he too began wondering ifsplenectomy is a source of the problem. My objectionto splenectomy is that I think it causes problems downthe line rather than immediately. Yes, without it youget a little more iron because more transfusions areneeded, but I think you get irreversible changes lateron which may kill our patients. This is why I am reallyopposed to splenectomy in most situations. When wehave done it in some of our thal intermedia patients,they have had partial splenectomy. I know it is morecostly, and I know there is a morbidity to it, but Ithink one has to consider a way to avoid the late com-plications of splenectomy. I think partial splenectomyshould be thought of as an option.

Q: But what’s the connection between the emboliand splenectomy and between the pulmonary hyper-tension and splenectomy?

Dr. Hyman: When you take the spleen out youget a higher platelet count. I think we have to consider

whether we should give aspirin to our patients withhigh platelet counts after splenectomy. I am beginningto wonder if we are causing some of the deaths, andwhether we should hypertransfuse them instead oftaking out their spleen for chronic fatigue and some ofthese other symptoms of anemia.

Dr. Pearson: Speaking of pulmonary hyperten-sion, I have one patient who at age 30 developedright-sided pulmonary hypertension. In talking to thecardiologist at Cornell, they have only two patientswith pulmonary hypertension. So it isn’t really com-mon. It appears to be an unusual event.

Dr. Giardina: We have three patients, all with thalmajor, who actually have pulmonary hypertension.

Dr. Forman: I wonder how many institutions arenow doing partial splenectomy? Could you tell us a lit-tle bit about the risk of the procedure and how youassess the patient?

Dr. Cohen: While having never done a partialsplenectomy myself, I would be delighted to answeryour question. Actually Eliot Vichinsky and LoriStyles gave a very beautiful presentation of the tech-nique for doing this. An open laparotomy is per-formed, and the vascular supply to the spleen is tiedoff and the edges turn black segmentally. I had towatch it once so I thought I would share it with every-body here. This was continued until 60–80% of thespleen—it might have been 80% of the spleen—wasgone (necrotic). The idea is basically to leave whatamounts to about half of a normal spleen size behind.At our own institution, with an open splenectomy, theaverage length of stay is 4–5 days and isn’t a managedcare phenomenon. For partial splenectomy, the staymay be longer. There was also a very high incidence ofpneumonia postoperatively in that group. That beginsto worry me. I think you better be pretty sure you aregetting something for your money in order to add tothe potential morbidity of the procedure.

Dr. Hyman: At L.A. Children’s, radiologists do anumber of partial splenectomies by embolectomytechnique. With that technique, they sometimes have


to go in more than once. They embolectomize it seg-mentally, and they try to necrose about 70–80% ofthe spleen. It does require hospitalization of five daysor more. I am personally opposed to completesplenectomy. The partial splenectomy can be done intwo ways, by embolectomy or surgically. I think thepostop course has to focus on prevention of pneumonia.

Dr. Dover: Let me make just one commentabout that. Just to clarify, what Lori Styles and EliotVichinsky presented was primarily the use of thistechnique in patients with sickle cell disease andsequestration crisis. They weren’t talking about tha-lassemia intermedia, and I think one really does haveto be worried about the possibility of splenicregrowth in this condition. In sickle cell disease, it isa very different issue altogether, because the spleenundergoes autoinfarction. We don’t want to go backto have to do another splenectomy. All the surgeonswill say they don’t want to have to reoperate, so Ithink we need some data to prove that they won’thave to.

Dr. Pearson: A couple of clinical points. Splenicembolization produces infarction, which can be avery painful event. Second, a child at our hospitalsome years ago had ligation of his main splenicartery. Apparently there are collateral vessels becausesix months later he had normal splenic function (lowpit count) and a normal radionuclide spleen scan.

Dr. Giardina: Dr. Hyman, your point is verywell taken. There is some amount of obstructivelung disease, pulmonary hypertension, and right-sided heart failure in patients with both thal majorand thal intermedia, and it should be well studied.There have been sporadic reports of protein C, pro-tein S deficiency in thalassemia patients, and therehas been some suggestion that the abnormal ery-thron mass produces some endothelial damage andmight result in the hypercoaguable states. I thinkthat there is a lot of work that we can do on investi-gating and clarifying possible coagulation disordersthat are associated with ineffective erythroporesiswith transfusion and with splenectomy. There maybe multiple factors involved here.

Dr. Hyman: Another question is whether weshould consider transfusing patients with thal inter-media during childhood and early adolescence, andthen stopping transfusions to see how they do.Transfusions will help get them through the compli-cations from their bones, growth and development,and endocrine deficiencies. We could transfuse themthrough the growth spurt and then see what happensoff transfusions. We have to think seriously beforewe take out the spleen. You have to look at the coag-ulation problems and put them on aspirin, or thinkof whether you are going to coumadinize them. Thisseriously has to be thought of and not just forgotten.I think these are real possible complications that arecoming along. If you look at the pathology, what arethese patients dying of and what are their symp-toms? Chronic hypoxia and heart problems.

Dr. Forman: Let’s take a case where you knowthe child has thalassemia intermedia, and the diag-nosis is supported by the family history. The child isnow 3, 4, or 5; the spleen is getting larger; hemoglo-bin is falling under 7, under 6; and you know fromthe family history that if you remove the spleen, thehemoglobin is going to go to 8 and stabilize. Whatwould be your indications for splenectomy, and whatwould be your options? One option is hypertransfu-sion, trying to keep that spleen from getting big;another is partial splenectomy; another is splenecto-my. Can you think of any other options, and whichone would you recommend?

Dr. Hyman: I would recommend partial splenec-tomy.

Dr. Cohen: I would take out the spleen. I willgive a couple of reasons why. I think a partialsplenectomy is fine if you do it as an experiment, butI don’t think we know enough about the long-termresults, which is exactly the point you were getting atwith splenectomy. I think to start that child onhypertransfusion is fine if you think that the childwas not going to be able to survive without hyper-transfusion. I would agree that transfusion is a rea-sonable option if in fact an older sibling had gone onto transfusion therapy anyway, or you wished that he


or she had gone on to transfusion therapy. But ifthat older child, as many thal intermedia patients do,is doing very nicely at a hemoglobin of 8 withoutbony changes or other changes, and you know youcan accomplish that by splenectomy, I am not surequite frankly what the argument would be.

Dr. Olivieri: I would agree. We would tell thefamily exactly that and then most of them, because ofthe fear of blood, would choose for splenectomy,which is what I would recommend also. Dr. Hyman,if this coagulation problem really exists, then it needsto be looked at, as Dr. Cohen is pointing out. It reallyneeds to be studied. I don’t know how to study thegeneration of thromboplastic substances or a hyperco-agulable state or lung function, but you have a lot ofsplenectomized thalassemics in this country, inCanada, and everywhere, and you have a lot of non-splenectomized patients. You can do a control andlook and see if that’s the clinical impression. I thinkwhat Dr. Cohen is really saying is that there is a lotmore evidence that splenectomy really does cause amean hemoglobin rise of at least 2 grams, and you canavoid transfusion. Transfusional iron overload still isgoing to be the thing that causes mortality. You arequite right that this pulmonary hypertension, right-sided heart failure, and diastolic dysfunction are allobserved, but they still haven’t been proven to be dueto splenectomy.

Dr. Forman: Let me respond to that. The youngpeople that I presented this morning benefitedtremendously from splenectomy. Early evidence thatthromboembolism might be associated with splenec-tomy would not be sufficient to tilt the balance tomove away from splenectomy in general.

Dr. Hyman: Well, I still vote for partial splenec-tomy. You have to look at both short-term and long-term. I really think serious consideration should begiven to partial splenectomy, despite its complica-tions and its cost. I think the Thai people may havesome data about their patients with hemoglobin Ethalassemia, though this is a different form of tha-lassemia intermedia. Some of them seem to feelstrongly about the partial splenectomy, and that

there is a splenectomy relationship to their cardiacdeaths.

Dr. Forman: I want to know if Dr. Pearson wouldhave recommended taking that spleen out.

Dr. Pearson: Yes, I would have.

Dr. Forman: Would you have recommendedtransfusions instead?

Dr. Pearson: No.

Q: I have a question about people who are put onDesferal and then followed longitudinally. Is the uri-nary iron excretion a helpful measure of chelation?Are there any data on what happens to the urinaryexcretion iron over time—not when their ferritindrops initially, but when it is high and they are on astable regimen? After you take one value of ironexcretion, can you assume it is going to be the samenext year?

Dr. Olivieri: Our experience is that when youlook at urine iron excretion and correlate it with bodyiron burden as quantitated by hepatic iron, the corre-lation is extremely bad. If you calculate urine creati-nine and weight and determine if the test reflects a24-hour collection, only 43% of our outpatient urinesamples are correctly collected. I suppose if you did itin hospital it might be much better.

Q: In your experience, do you think being able tomore precisely define body iron in your patients hasbeen inspiring them to use their Desferal?

Dr. Olivieri: Yes, but probably because we are abit zealous in telling them that the ferritin level is notthat useful. I have heard patients tell each other,“Don’t tell me your ferritin, what was your liver ironlast time?” So they are very aware that this is not themeasurement we use any longer, but we use the liveriron—just as we used to encourage them with serumferritin. If your physician believes in it and tells youwhat they think, if you are inspirable, you are going tobe inspired. There is nothing different about hepatic


iron versus a serum ferritin. When we used serum fer-ritin alone to estimate body iron burden, we used toalways discuss the value at every visit. Now we discussthe hepatic iron as what we believe is the quantitativeassessment.

Q: What toxicities are associated with the oraliron chelator, deferiprone (L1)?

Dr. Olivieri: In the study we did with patientswith thal major, we saw arthropathy in 3 of the 21patients that were reported. In one, it reversed. Inanother, it went away after interarticular steroids weregiven, and the other still has the problem and contin-ues on L1. There has been an incidence of agranulo-cytosis and neutropenia. There have been 13 reportedcases in the literature. It is difficult to know what thedenominator of that is. Zinc deficiency was describedin four patients by Victor Hofbrand.

Q: Have all the incidences of agranulocytosisreversed when they were taken off L1?

Dr. Olivieri: Yes.

Q: How does cirrhosis affect your interpretationof the liver iron?

Dr. Olivieri: If a person has frank cirrhosis, Idon’t think the quantitation of liver iron will be accu-rate, because cirrhosis is a nodular process. If you lookat the histology, iron isn’t loaded into the nodules. If aperson is frankly cirrhotic, it is difficult to see howiron chelation therapy is going to help this process,because to my knowledge advanced cirrhosis is notreversible. If a person has cirrhosis andhemachromatosis, life expectancy is not extended byiron-chelating therapy. We are already seeing somethalassemia patients who develop hepatoma after cir-rhosis has been established. It’s not to say that if youhave a good cardiac outlook, and since hepatoma isn’t100 percent fatal—you certainly want to continue tochelate them. Your question is, “What value of liveriron is recommended?” In those cases, the liver biopsy,I believe, is not quantitative of iron concentration.


The ideas developed and reported in thismonograph represent the product of anintense and spirited conference day. It is

appropriate to note that the occasion had a gratifyingspirit of collegiality, trust, and collaboration. Therewere also the necessary feelings of passion. As a con-vener and not as a presenter, I am moved to commentthat the nine principal speakers and panelists are veryspecial persons in this field, with a unique history ofcaring and continuity. We have heard from NorthAmerica’s thalassemia legends.

As conclusions are drawn regarding thalassemiaintermedia, it seems reasonable to say that we do notyet have the tiger by the tail. Intermedia is a low inci-dence situation, but with high stakes. It shares muchwith the world of thalassemia major. Important areasof study include genotype/phenotype correlations, thepace of transfusion, utility of liver iron measures, thehygiene of splenectomy, control of infection, and oraltherapies, all of which have had substantial discussionin these proceedings. Significant new data are avail-able now.

Epilogue

Allen C. Crocker, M.D.


Thalassemia Intermedia ConferenceConference Director: Howard A. Pearson, M.D.

Children’s Hospital, Boston, MANovember 14, 1996

8:30–9:00 Registration & Coffee

9:00–9:15 IntroductionEdwin N. Forman, M.D.Director, Division of Pediatric Hematology/OncologyRhode Isalnd HospitalProfessor of PediatricsBrown University School of Medicine, Providence, RI

9:15–9:30 Update on Activity of the National Heart, Lung, and Blood InstituteAlan S. Levine, Ph.D.Director, Blood Diseases ProgramDivision of Blood Diseases and ResourcesNational Heart, Lung, and Blood Institute, Bethesda, MD

9:30–10:00 Thalassemia Intermedia: History and Definition of ThalassemiaIntermediaHoward A. Pearson, M.D.Professor of PediatricsYale University School of Medicine, New Haven, CT

10:00–11:00 Characterization and Findings in Current American and CanadianPatients with ThalassemiaPatricia J. V. Giardina, M.D.Associate Professor of Clinical PediatricsChief, Division of Pediatric Hematology/OncologyThe New York Hospital/Cornell Medical Center, New York, NY

11:00–11:15 Break

11:15–11:45 Iron Status in Thalassemia IntermediaNancy F. Olivieri, M.D., FRCP (C)Hospital for Sick ChildrenDirector, Hemoglobinopathy ProgramProfessor of Pediatrics & MedicineUniversity of Toronto, Toronto, Canada


11:45–12:15 Pharmacological Treatment of Thalassemia IntermediaGeorge F. Dover, M.D.Professor, Pediatric Oncology and MedicineDirector, Department of PediatricsJohns Hopkins University School of Medicine, Baltimore, MD

12:15–1:15 Lunch

1:15–2:00 Genetic Basis of Thalassemia IntermediaHaig Kazazian, M.D.Professor and Chairman of GeneticsUniversity of Pennsylvania School of Medicine, Philadelphia, PA

2:00–2:30 Infections in Thalassemia IntermediaAlan R. Cohen, M.D.Professor of PediatricsUniversity of Pennsylania School of MedicineChief, Division of HematologyChildren’s Hospital of Philadelphia, Philadelphia, PA

2:30–2:45 Break

2:45–3:45 General Panel Discussion & Audience ParticipationDrs. Pearson, Giardina, Cohen, Oliveri, and Kazazian

3:45–4:15 EpilogueAllen C. Crocker, M.D.Program Director, Developmental Evaluation CenterChildren’s HospitalAssociate Professor of PediatricsHarvard Medical School, Boston, MA


Fajiri Abdul-HameedBoston Sickle Cell Center818 Harrison AvenueFGH-2Boston, MA 02118Phone: 617-534-5727Fax: 617-534-5739

Sharon Agnew818 Harrison AvenueBoston, MA 02118Phone: 617-534-5727Fax: 617-534-5739

Richard Ancona, M.D.300 Middle Country RoadSmithtown, NY 11787Phone: 516-979-6466Fax: 516-979-6475

Edward J. Benz, Jr., M.D.Johns Hopkins University1830 E. Monument StreetRoom 9026Baltimore, MD 21205Phone: 410-955-6642Fax: 410-955-0430

Lauren C. Berman, M.S.W.Children’s HospitalInstitute for Community Inclusion300 Longwood AvenueBoston, MA 02115Phone: 617-355-5943Fax: 617-355-7940

John BernardBoston Sickle Cell Center818 Harrison AvenueFGH-2Boston, MA 02118Phone: 617-534-5727Fax: 617-534-5739

Kenneth R. Bridges, M.D.Brigham & Women’s Hospital221 Longwood AvenueBLI 327Boston, MA 02115Phone: 617-732-5842 or 617-732-8490Fax: 617-975-0876

Alan Cohen, M.D.Children’s Hospital of Philadelphia34th & Civic Center BoulevardPhiladelphia, PA 19104Phone: 215-590-3438Fax: 215-590-3525

Allen C. Crocker, M.D.Children’s Hospital300 Longwood AvenueBoston, MA 02115Phone: 617-355-6509Fax: 617-355-7940

Karen CrutchfieldBoston Sickle-Cell Center818 Harrison AvenueFGH-2Boston, MA 02118Phone: 617-534-5727Fax: 617-534-5739

Conference Participants


Alan D. D’AndreaDana Farber Cancer Institute44 Binney StreetBoston, MA 02115Phone: 617-632-2080Fax: 617-632-2085

George Dover, M.D.Johns Hopkins University720 Rutland AvenueTraylor 907Baltimore, MD 21205Phone: 410-955-5976Fax: 410-955-9850

Richard Drachtman, M.D.The Cancer Institute of New Jersey185 Little Albany StreetNewark, NJ 08903Phone: 908-235-7898Fax: 908-235-6462

Kathy Duncan, P.N.P.Boston Floating Hospital755 Washington StreetBox 14Boston, MA 02111Phone: 617-636-5535Fax: 617-636-4632

Igal Fligman, M.D.Cornell Medical Center525 East 68th StreetNew York, NY 10021Phone: 212-746-3400Fax: 212-746-8609

Frances Flug, M.D.Hackensack University Medical Center30 Prospect AvenueHackensack, NJ 07601Phone: 201-996-5437Fax: 201-487-7340

Edwin N. Forman, M.D.Rhode Island Hospital593 Eddy StreetProvidence, RI 02903Phone: 401-444-5171Fax: 401-444-8845

John GaspardBoston Sickle-Cell Center818 Harrison AvenueFGH-2Boston, MA 02118Phone: 617-534-5727Fax: 617-534-5739

Beatrice Gee, M.D.Children’s HospitalEnders 7300 Longwood AvenueBoston, MA 02115Phone: 617-355-7432Fax: 617-355-7262

Patricia Giardina, M.D.Cornell Medical Center525 East 68th StreetNew York, NY 10021Phone: 212-746-3415Fax: 212-746-8609

Jed B. Gorlin, M.D.Children’s Hospital300 Longwood AvenueBoston, MA 02115Phone: 617-355-6268Fax: 617-355-6086

Jeanne Harvey, R.N.UMASS Medical Center55 Lake Avenue, NorthWorcester, MA 01655Phone: 508-856-4225Fax: 508-856-4282


Barkat Hooda, M.D.Rhode Island Hospital593 Eddy StreetProvidence, RI 02903Phone: 401-444-5171Fax: 401-444-8845

Heather Hume, M.D.Ste-Justine Hospital3175 Cote Ste-CatherineMontreal, QuebecCANADA, H3T 1C5 Phone: 514-345-4639Fax: 514-345-4884

Minnie HunterBoston Sickle Cell Center818 Harrison AvenueFGH-2Boston, MA 02118Phone: 617-534-5727Fax: 617-534-5739

Carol B. Hyman, M.D.Cedars-Sinai Medical Center8700 Beverly BoulevardRoom 4310Los Angeles, CA 90048Phone: 310-855-6387Fax: 310-652-0681

Nathaniel W. James, M.D.Director, International ClinicMaine Medical Center/Out-Patient Dept.22 Bramhall StreetPortland, ME 04102Phone: 207-871-4323Fax: 207-871-6308

Haig H. Kazazian, Jr., M.D.University of Pennsylvania475 CRB415 Curie BoulevardPhiladelphia, PA 19104-6145Phone: 215-898-3582Fax: 215-573-7760

Dee Kermack, R.N.Montreal Children’s Hospital2300 Tupper StreetMontreal, QuebecCANADA, H3H 1P3Phone: 514-934-4420Fax: 514-934-4424

Christopher Keuker, M.D.Hasboro Children’s Hospital593 Eddy StreetProvidence, RI 02903Phone: 401-444-5171Fax: 401-444-8845

Naomi KleinHospital for Sick Children555 University AvenueToronto, OntarioCANADA, M5G 1X8Phone: 416-813-7381Fax: 416-813-5327

Jolanta Kulpa, M.D.Long Island College Hospital340 Henry StreetBrooklyn, NY 11201Phone: 718-780-1025Fax: 718-780-2989

Susan Kurth, M.P.H., R.N.Children’s Hospital300 Longwood AvenueBoston, MA 02115Phone: 617-355-8304Fax: 617-738-9857

Margaret Lee, M.D.South Cove Community Health Center885 Washington StreetBoston, MA 02111Phone: 617-482-7555Fax: 617-482-2930


Betty Leef, R.N.Rhode Island Hospital593 Eddy StreetProvidence, RI 02903Phone: 401-444-8391Fax: 401-444-4485

Alan S. Levine, Ph.D.National Institutes of Health6701 Rockledge DriveMSC 7950Bethesda, MD 20892-7950Phone: 301-435-0050Fax: 301-480-0868

Samuel Lux, M.D.Children’s Hospital300 Longwood AvenueBoston, MA 02115Phone: 617-355-7904Fax: 617-355-7262

Marge MacDowell, M.S.W.Rhode Island Hospital 593 Eddy StreetProvidence, RI 02906Phone: 401-444-5711Fax: 401-444-5715

Elyse Mandell, N.P.Brigham & Women’s Hospital221 Longwood AvenueBoston, MA 02115Phone: 617-732-8485Fax: 617-739-3324

Marie Martin, R.N.Children’s Hospital of Philadelphia 34th & Civic BoulevardPhiladelphia, PA 19104Phone: 215-590-2197Fax: 215-590-3694

Lillian McMahon, M.D.Boston Sickle Cell Center818 Harrison AvenueFGH-2Boston, MA 02118Phone: 617-534-5727Fax: 617-534-5739

K. S. Murthy, M.D.Bay State Medical Center299 Carew StreetSpringfield, MA 01104Phone: 413-739-4085Fax: 413-733-3646

David G. Nathan, M.D.Dana-Farber Cancer Institute44 Binney StreetBoston, MA 02115Phone: 617-632-2155Fax: 617-632-2161

Ellis Neufeld, M.D.Children’s Hospital300 Longwood AvenueBoston, MA 02115Phone: 617-355-8183Fax: 617-734-6791

Eric Nisbet-BrownHospital for Sick Children555 University AvenueToronto, OntarioCANADA, M5G 1X8Phone: 416-813-5643Fax: 416-813-4954

Nancy Olivieri, M.D.Hospital for Sick Children555 University AvenueToronto, OntarioCANADA, M5G 1X8Phone: 416-813-6823Fax: 416-813-5346


Linda Ott, M.D.11 Durant StreetNewton, MA 02158Phone: 617-527-5722Fax: 617-527-3452

Howard Pearson, M.D.Yale-New Haven Hospital 333 Cedar StreetNew Haven, CT 06510Phone: 203-785-6662Fax: 203-785-7194

Sandra Perez, R.N., BSNBoston Sickle Cell Center818 Harrison AvenueFGH-2Boston, MA 02118Phone: 617-534-5727Fax: 617-534-5739

Paul F. Poulin, M.D.York Hospital15 Hospital DriveYork, ME 03909Phone: 207-363-0041Fax: 207-351-2235

Linda Rink, R.N.Yale-New Haven Hospital 333 Cedar StreetNew Haven, CT 06520Phone: 203-785-4640Fax: 203-785-7194

Peter Rintels, M.D.Rhode Island Medical Foundation593 Eddy StreetProvidence, RI 02903Phone: 401-444-5395Fax: 401-444-8919

Cathy Rosenfield, M.D.New England Medical Center755 Washington StreetBox 14Boston, MA 02111Phone: 617-636-5535 Fax: 617-636-8388

Molly Schwenn, M.D.UMASS Medical Center55 Lake Avenue, NorthWorcester, MA 01655Phone: 508-856-4225Fax: 508-856-4282

Julianna Shen, R.N.Cornell Medical Center525 East 68th StreetNew York, NY 10021Phone: 212-746-3400Fax: 212-746-8609

Shirley Smith, R.N., M.S.DHS-PHS HRSA-MCH Bureau Room 1826Boston, MA 02203Phone: 617-565-1451Fax: 617-565-4027

David Steele, M.D.759 Chestnut StreetSpringfield, MA 01799Phone: 413-784-3222 x5377Fax: 413-784-3613

Richard Steingart, M.D.BayState Medical Center759 Chestnut StreetSpringfield, MA 01199Phone: 413-784-3316Fax: 413-784-3613


Sharon StewartBoston Sickle Cell Center818 Harrison AvenueFGH-2Boston, MA 02118Phone: 617-534-5727Fax: 617-534-5739

R. Sundaran, M.D.Long Island College Hospital340 Henry StreetBrooklyn, NY 11202Phone: 718-780-1025Fax: 718-780-2989

Diane Tremblay, R.N.Internal Medicine & Oncology Association371 Park StreetW. Springfield, MA 01089Phone: 413-732-2051Fax: 413-734-5555

Mitchell Weiss, M.D.Children’s Hospital300 Longwood AvenueBoston, MA 02115Phone: 617-355-8126Fax: 617-734-6791


NEW ENGLAND REGIONAL GENETICS GROUPOFFERS

Educational Video on Thalassemia for Southeast Asians

This video was produced in three Southeast Asian languages (Khmer, Laotian, and Vietnamese) toassist health care providers, community workers, and other educators to provide information aboutthalassemia to Southeast Asian communities. The conversational format includes medical andgenetic aspects of thalassemia as well as carrier screening, prenatal diagnosis, and genetic counseling.The English translation is included on all language versions of the video.

Price: $15.00Video Length: 10 minutes

To order or for more information, contact:

Mary Castro Aten, Assistant to the CoordinatorNew England Regional Genetics Group28 Clarendon StreetNewton, MA 02160Telephone: (617) 243-3033Fax: (617) 243-3032

Name/Title: ______________________________________________________________________

Clinic/Organization: ______________________________________________________________

Address: ________________________________________________________________________

City: ______________________________________ State: ________ Zip: ______________

Telephone: ______________________________________________________________________

Comments: ______________________________________________________________________

________________________________________________________________________________

MAKE CHECKS PAYABLE TO: MIIRI, INC.(Federal ID #04-2261109)

Maternal and Child Health Bureau

National Center for Educationin Maternal and Child Health

ISBN 1-57285-041-8

Thalassemia Intermedia

Documents

Transcript of Thalassemia Intermedia