Journal of Clinical Virology 36 (2006) 13–16

Disseminate and fatal cytomegalovirus disease with thymitis in a naiveHIV-patient after early initiation of HAART:

Immune restoration disease?

Sonia Gutierrez a,1, Silvia Alconchel b, Ezequiel Ruiz-Mateos c,1, Miguel Genebat a,1,Alejandro Vallejo c,1, Eduardo Lissen a,1, Jorge Fernandez-Alonso b, Manuel Leal a,∗,1

a Department of Internal Medicine, Virgen del Rocio University Hospital, Seville, Spainb Department of Anatomy Pathology, Virgen del Rocio University Hospital, Seville, Spain

c Department of Biochemistry, Virgen del Rocio University Hospital, Seville, Spain

Received 5 October 2005; received in revised form 10 December 2005; accepted 16 December 2005

Abstract

da©

K

1

tittvimthiO(c

IS

1d

We describe a naıve HIV-infected patient who developed a Pneumocystis carinii pneumonia and disseminate and fatal cytomegalovirusisease within 3 months after initiation of HAART, suggesting due to coincidence in time, an immune restoration disease. We propose anlternative hypothesis.

2005 Elsevier B.V. All rights reserved.

eywords: Immune restoration disease; Thymitis cytomegalovirus; AIDS

. Introduction

HIV-patients under HAART may experience severe sys-emic inflammatory reactions that have been defined asmmune restoration disease (IRD). It has been communicatedhat IRD has two different patterns; an earlier pattern duringhe first 3 months of HAART as an immune response againstiable opportunistic pathogens, and a later pattern as anmmune response against non-viable opportunistic pathogens

onths to years after HAART (French et al., 2004). Duringhe IRD, a baseline CD4 cell count below 100 cells/mm3

as been reported among patients, while after HAART anncrease above 200 cells/mm3 is reached (Price et al., 2001).utcomes range from minimal morbidity to fatal progression

French et al., 2004; Hirsch et al., 2004). In this way, atypi-al presentations of mycobacterial, cytomegalovirus (CMV),

∗ Corresponding author at: Viral Hepatitis and AIDS Unit, Department ofnternal Medicine, Virgen del Rocio University Hospital, Seville, PC 41013,pain. Tel.: +34 955012396; fax: +34 955012390.

E-mail address: mleal@cica.es (M. Leal).1 Viral Hepatitis and AIDS Unit.

hepatitis B virus, hepatitis C virus and JC virus have beendescribed after initiating HAART (French et al., 2004; Safdaret al., 2002).

We here report the case of a naıve HIV-infected patientwho developed Pneumocystis carinii pneumonia as well asdisseminated and fatal CMV infection coinciding with theinitiation of HAART.

2. Case report

In 1993, a 32-year-old woman was diagnosed of HIVinfection in our unit. Then, her CD4 T cell count was 840cells/mm3, and HIV plasma viral load (pVL) was above75,000 copies/mL. The moment that primoinfection occurredin the past was unknown because it was asymptomatic. Sinceshe declined to receive antiretroviral therapy, a progres-sive CD4+ T cell count decrease was taking place duringthe following 9 years. In July 2002, she began HAARTwith zidovudine, lamivudine, and abacavir, having HIV pVLabove 75,000 copies/mL, CD4+ T cell count of 90 cells/mm3,and a thymic volume (measured by mediastinic computed

386-6532/$ – see front matter © 2005 Elsevier B.V. All rights reserved.oi:10.1016/j.jcv.2005.12.007

14 S. Gutierrez et al. / Journal of Clinical Virology 36 (2006) 13–16

tomography) of 1.2 cm3. In the next 3 months after the ini-tiation of HAART, the patient developed several infectiousevents such as P. carinii pneumonia after 10 days, which wassuccessfully treated with cotrimoxazole. Seven days later,the patient showed low level of conscience, convulsions, andplasmatic Na+ concentration of 102 mequiv./L, recoveringfrom the neurological events after hiponatremia treatment.Three weeks after the initiation of HAART, a colonoscopywas performed due to the development of mucosanguineousdiarrhoea. The biopsy of the colon revealed cytomegalicinclusion bodies that allowed the diagnosis of CMV coli-tis, which was treated with ganciclovir and foscarnet, whileHAART was interrupted after 6 weeks of its initiation dueto an apparent IRD. She had not previously suffered a CMV-related disease, thus, she did not take ganciclovir or foscarnet.Despite of this new treatment, symptoms as fever, cough,and disnea turned up, as well as bilateral interstitial infiltrate,which was evidenced by chest radiography. Since the patienttreated with ganciclovir and foscarnet showed favourableevolution, empiric CMV pneumonia was diagnosed. Ten dayslater, the patient referred lost of vision and fundus ophthal-moscopic examination confirming CMV retinitis, despite thetreatment she had taken. At that moment, CMV plasma viralload was above 738,000 copies/mL, it was not possible toachieve the sequence of CMV strain to study the develop-msd

Iisctwf

3

(ticu(tiDf

oppd

Fig. 1. Evolution of total CD4 cells counts and TRECs.

TRECs generated during the rearrangement of T cell receptorgenes as one proposed molecular marker for the determina-tion of recent thymic cell emigrants. A PCR-based methodfor quantifying �Rec-�J� TREC number has been described(Douek et al., 1998).

4. Results

Evolution of total CD4 cell counts and TRECs are showedin Fig. 1. Thymic volume was 1.2 cm3 (score 2), which hasbeen related with an attrofic thymus and fat infiltration.

Evolution of plasma viral load VIH and CMV are showedin Fig. 2. Plasma viral load CMV was measured in blood sam-ples since year 1993 and it revealed that CMV was detectablein plasma 2 months before starting HAART. It increaseduntil 738,000 copies/mL after initiating first HAART reg-imen. Despite of second HAART and therapy with ganci-clovir, foscarnet and cidofovir were started, CMV was alwaysdetectable in plasma.

4.1. Necropsy

It was procedure as a standard autopsy, with speciallooking for thymic remnants. Immunohistochemical thymusstudies were performed by streptavidine-biotine techniquefae

FC

ent of drug related resistance to ganciclovir and foscarnet,o, due to the progression of CMV disease with this drugs weecided to switch to cidofovir.

Since the patient progressed with adverse clinical events,RD was ruled out and a new HAART regimen includ-ng zidovudine, lamivudine, and lopinavir/ritonavir was pre-cribed. Finally, although CD4 T cell count increased to 432ells/mm3 and CMV viral load decreased to 1000 copies/mL,he patient developed fatal meningoencephalitis after 12eeks of initiation of HAART. Necropsy was performed to

urther studies.

. Material and methods

Plasma HIV-1 RNA was measured by a quantitative PCRHIV Monitor Test kit version 1.5, Roche Molecular Sys-em Inc., Branchburg, NJ) according to the manufacturer’snstructions. This assay has a detection limit of 50 HIV-1opies/mL. Frozen plasma samples stored since 1993 weresed to measure CMV viral load by a quantitative PCRCobas Amplicor CMV Monitor test, Roche Molecular Sys-ems Inc., Branchburg, NJ) according to the manufacturer’snstructions. This assay has a detection limit of 400 CMV-NA copies/mL. Total CD4 cell count was determined in

resh samples by conventional flow cytometry.Since the patient was enrolled in a study about the role

f the thymus in T cell repopulation, the following measuredarameters were available: mediastinic computed tomogra-hy that was performed with a modified method as previouslyescribed (Choyke et al., 1987); and the quantification of

or CD3, CD20 (Dako-labs), CD4, CD8, CD5 (Vitro labs),nd pan-cytokeratins (Menarini lab) antibodies, on paraffinmbedded sections. The scarce lymphocytes presents were

ig. 2. HIV and CMV plasma viral load evolution. (�) Plasma viral loadMV; (�) plasma viral load HIV.

S. Gutierrez et al. / Journal of Clinical Virology 36 (2006) 13–16 15

Fig. 3. Cytomegalic inclusion bodies in the medullar epithelial cells of thy-mus.

stained with CD3, CD4, CD8, and CD5, with sparse CD20positive lymphocytes placed between them. Pan-cytokeratinsstained most of the cells with intranuclear cytomegalic inclu-sions.

Macroscopic findings were no relevance. Microscopicstudy showed generalized infection with cytomegalovirusinclusions, mainly in lungs, with lymphocytic interstitialpneumonitis, and adrenal glands, which showed massivenecrotizing adrenalitis. Both serious adrenalitis as the pneu-monitis explained the cause of death. Other affected organswere the liver, brain, and heart.

Thymus showed diffuse cortical atrophy with scarce lym-phocytes, and frequent cytomegalic inclusion cells in the coreof the medulla (Fig. 3). Neither necrosis nor plasma cell infil-tration were presented. Microscopical changes found wereaccording with changes reported in later stages of the thymicprocess in AIDS patients (Davis, 1984). Cytomegalic inclu-sion bodies in the medullar epithelial cells were a relevantfinding.

5. Discussion

Our report describes a serious thymic dysfunction in anaıve HIV-infected patient who developed P. carinii pneumo-n3c

twi2Ipvi

On the other hand, it has been reported that the loss ofCMV-specific CD4 T cell response cause recurrences ofCMV retinitis despite of the increase of CD4 cell counts afterHAART (Johnson et al., 2001; Komanduri et al., 2001). Tothe best of our knowledge, there are no reported cases of deathfor disseminated CMV disease in HIV-positive patients afterHAART.

Initially, we thought that the patient developed IRDbecause the presence of P. carinii pneumonia and CMV col-itis, coinciding with the introduction of HAART. Moreover,CD4 cell count at baseline was low but increased after treat-ment. Subsequently, CMV disease was hard to be explainedas a IRD due to the increase of CMV plasma viral load 2months before initiation of HAART, the sudden clinical evo-lution and the disseminate infection by viables pathogensfound after necropsy. Thus, an alternative hypothesis wasneeded to formulate.

In this way, as revealed by the presence of a later AIDSstage thymus with cytomegalic inclusion bodies, there was aserious thymic dysfunction. This might impair the regenera-tion of new complete peripheral T cell receptor repertoire thatcould enhance opportunistic infectious diseases, as P. cariniiand CMV infection, despite antiretroviral treatment.

Despite thymic impairment, thymic function-relatedmarkers as, total CD4 cell counts and TREC levels, increasedafter HAART. We suggested that, CD4 cell counts could haveiwt

ebdm

toHn

A

S

R

C

D

D

ia and disseminate and fatal cytomegalovirus disease withinmonths after initiation of HAART suggesting, since their

oincidence in time, an immune restoration disease.Nowadays, most of the atypical manifestations of infec-

ious diseases after HAART are included as IRD. In thisay, CMV IRD has frequently been reported as eye disease,

ncluding recurrent CMV retinitis or uveitis (Deayton et al.,000; Stone et al., 2002). Other less common forms of CMVRD have been reported, such as, pancreatitis, submaxillitis,seudotumoral colitis, and isolated fever with positive CMViraemia (Gilquin et al., 1997). Resolution has been observedn all patients after therapy with ganciclovir and foscarnet.

ncreased by redistribution of the cells within lymph nodes,hile decreased T-cell turnover after HAART could explain

he increment in TREC levels (Lempicki et al., 2000).Cytomegalic inclusion bodies found in thymic medullar

pithelial cells is a relevant finding since these cells haveeen reported to be prominent targets of viral replication inifferent cultures, although it has not been studied in animalodels (Mocarski et al., 1993; Numazaki et al., 1989).In conclusion, this case report stands out the difficulty

o make an IRD diagnosis. In this way, the developmentf opportunistic infectious coinciding early after initiatingAART should be considered in the differential IRD diag-osis since they have different management and prognosis.

cknowledgements

This work was supported by Red de Investigaciones enida (RIS) from the Ministerio de Sanidad y Consumo, Spain.

eferences

hoyke PL, Zeman RK, Gootenberg J, Greenberg JN, Hoffer F, FrankJA. Thymic atrophy and regrowth in response to chemotherapy: CTevaluation. Am J Roentgenol 1987;149:269–72.

avis Jr AE. The histopathological changes in the thymus gland inthe acquired immune deficiency syndrome. Ann NY Acad Sci1984;437:493–502.

eayton J, Wilson P, Sabin C, et al. Changes in the natural history ofcytomegalovirus retinitis following the introduction of highly activeantiretroviral therapy. AIDS 2000;14:1163–70.

16 S. Gutierrez et al. / Journal of Clinical Virology 36 (2006) 13–16

Douek DR, McFarland P, Keiser E, Gage EA, Massey JM, Haynes BF,et al. Changes in thymic function with age and during treatment ofHIV infection. Nature 1998;396:690–5.

French MA, Price P, Stone S. Immune restoration disease after antiretro-viral therapy. AIDS 2004;18:1615–27.

Gilquin J, Piketty C, Thomas V, Gonzalez-Canali G, Belec L, KazatchkineMD. Acute cytomegalovirus infection in AIDS patients with CD4counts above 100 × 106 cells/L following combination antiretroviraltherapy including protease inhibitors. AIDS 1997;11:1659–60.

Hirsch H, Kaufmann G, Sendi P, Battegay M. Immune reconstitution inHIV-infected patients. Clin Infect Dis 2004;38:1159–66.

Johnson S, Benson C, Johnson D, Weinberg A. Recurrences ofcytomegalovirus retinitis in a human immunodeficiency virus-infectedpatient, despite potent antiretroviral therapy and apparent immunereconstitution. Clin Infect Dis 2001;32:815–9.

Komanduri K, Feinberg J, Hutchins R, et al. Loss of cytomegalovirus-specific CD4 T cell responses in human immunodeficiency virus type1-infected patients with high CD4 cell counts and recurrent retinitis.J Infect Dis 2001;183:1285–9.

Lempicki RR-A, Kovacs J-A, Baseler M-W, et al. Impact of HIV-1 infec-tion and highly active antiretroviral therapy on the kinetics of CD4

and CD8 T cell turnover in HIV-infected patients. Proc Natl Acad SciUSA 2000;97:13778–83.

Mocarski ES, Bonyhadi M, Salimi S, McCune JM, Kaneshima H. Humancytomegalovirus in a SCID-hu mouse: thymic epithelial cells areprominent targets of viral replication. Proc Natl Acad Sci USA1993;90(1):104–8.

Numazaki K, Goldman H, Bai XQ, et al. Effects of infection by HIV-1cytomegalovirus and human measles virus on cultured human thymicepithelial cells. Microbiol Immunol 1989;33(9):733–45.

Price P, Mathiot N, Krueger R, Stone S. Immune dysfunction and immunerestoration disease in HIV patients given highly active antiretroviraltherapy. J Clin Virol 2001;22:279–87.

Safdar A, Rubocki R, Hovarth J. Fatal immune restoration disease inhuman immunodeficiency virus type 1-infected patients with progres-sive multifocal leukoencephalopathy: impact of antiretroviral therapy-associated immune reconstitution. Clin Infect Dis 2002;35:1250–7.

Stone S, Price P, Tay-Kearney M, French MA. Cytomegalovirus (CMV)retinitis immune restoration disease occurs during highly activeantiretroviral therapy-induced restoration of CMV-specific immuneresponses within a predominant Th2 cytokine environment. J InfectDis 2002;185:1813–7.