SOME RADIOGRAPHIC MANIFESTATIONS OF SCURVY...
6
SOME RADIOGRAPHIC MANIFESTATIONS OF EARLY SCURVY BY JAMES F. BRAILSFORD From the Royal Orthopaedic Hospital, Birmingham (RECEIVED FOR PULBUCATION OCTOBER 28, 1952) Though scurvy is associated with characteristic radiographic appearances it is important to realize that it can manifest itself clinically before radio- graphy detects any change from the normal in the bones. The latent negative radiographic period may be of several months. In most conditions it varies with the individual and the factors which are associ- ated with the deficiency. There is some evidence that the deficiency factors, as in rickets, act upon the foetus in utero and in the early days of separate life, though scurvy is commonly regarded as a deficiency disease which shows itself between the eighth and eighteenth months. The initial lesion which usually attracts clinical attention is the haematoma. Hutchinson and Moncrieff (1941) have pointed out that -the chief changes in infantile scurvy are in the neigh- bourhood of the bones. A section made across a limb at the site of a swelling shows that the peri- osteum is hypervascular, thickened and separated from- the subjacent bone by a layer of partially organized blood clot. There is no sign of inflam- mation and no hard bone is formed in the periosteum except in longstanding cases.' Therefore, in these early cases we see no sign of change in the bones. The first radiographic evidence may be a subperiosteal haemorrhage (Fig. 2) or fractures at the growing extremity of one or more of the long bones (Fig. 1), but usually such lesions have been present for a week or more before radiographic examination has been made, for we often see that already some calcium has been deposited in an amorphous form within the associated haematoma. and is most evident at the periphery of the haema- toma, leaving a translucent zone between it and the periphery of the shaft of the bone (Fig. 2). Infants, apparently healthy in other respects, have been seen at birth with considerable calcium in haematomata associated with fracture, indicating intra-uterine damage of some weeks' duration. Massive haematomata with much calcium surround- ing both femora of an unusual infant at birth have been illustrated (Brailsford, 1948). Certainly, within the first few months of life radiographs may show amorphous calcium deposition in haematomata enveloping one or more long bones, which appear to have normal structure and density, and, though no other definite indication of scurvy may be present, organization, ossification and complete restitution to normal bone takes place readily on the adminis- tration of vitamin C. These features are illustrated by the sequence of radiographic changes in Fig. 2. Fig. 2 illustrates the case of an infant. aged 3 months, who developed a swelling in the left thigh. A radiograph of the thigh on August 15, 1947 (Fig. 2A), showed amorphous calcium deposited in a large subperiosteal haematoma which enveloped most of the shaft of the femur. A further radiograph (Fig. 2B) on August 28, 1947, showed an increase in the size of the lesion, and its definition suggested that rapid organization and ossification was taking place. The radiograph of September 8, 1947 (Fig. 2c), shows that within a month most of the haematoma had been converted into bone. On the clinical and radiographic evidence this was interpreted as sar- coma and drastic treatment was contemplated. The films were submitted to me by Dr. Patricia Franklyn, and I suggested that the condition was due to scurvy and advised the administration of appropriate amounts of vitamin C. With its exhibition the patient rapidly showed signs of recovery and by January 8, 1948 (see Fig. 2D), considerable resolution had taken place. By 1952 the bone had returned to normal (Fig. 2E). Obviously in this case the clinical signs of scurvy were not present and the radiographic evidence had been misinterpreted but the therapeutic test was both positive and spectacular. In 1943 in a paper on osteogenesis imperfecta, I published an account of four patients with this dystrophy who developed an unusual complication, namely, the development around one or more bones of a swelling, probably as the result of trauma, since it was observed that a similar sequence of changes 81 copyright. on 20 April 2018 by guest. Protected by http://adc.bmj.com/ Arch Dis Child: first published as 10.1136/adc.28.138.81 on 1 April 1953. Downloaded from
Transcript of SOME RADIOGRAPHIC MANIFESTATIONS OF SCURVY...
SOME RADIOGRAPHIC MANIFESTATIONS OF EARLYSCURVY
BY
JAMES F. BRAILSFORDFrom the Royal Orthopaedic Hospital, Birmingham
(RECEIVED FOR PULBUCATION OCTOBER 28, 1952)
Though scurvy is associated with characteristicradiographic appearances it is important to realizethat it can manifest itself clinically before radio-graphy detects any change from the normal in thebones. The latent negative radiographic period maybe of several months. In most conditions it varieswith the individual and the factors which are associ-ated with the deficiency. There is some evidence thatthe deficiency factors, as in rickets, act upon thefoetus in utero and in the early days of separate life,though scurvy is commonly regarded as a deficiencydisease which shows itself between the eighth andeighteenth months. The initial lesion which usuallyattracts clinical attention is the haematoma.Hutchinson and Moncrieff (1941) have pointed outthat
-the chief changes in infantile scurvy are in the neigh-bourhood of the bones. A section made across alimb at the site of a swelling shows that the peri-osteum is hypervascular, thickened and separatedfrom- the subjacent bone by a layer of partiallyorganized blood clot. There is no sign of inflam-mation and no hard bone is formed in the periosteumexcept in longstanding cases.'
Therefore, in these early cases we see no sign ofchange in the bones. The first radiographic evidencemay be a subperiosteal haemorrhage (Fig. 2) orfractures at the growing extremity of one or more ofthe long bones (Fig. 1), but usually such lesions havebeen present for a week or more before radiographicexamination has been made, for we often see thatalready some calcium has been deposited in anamorphous form within the associated haematoma.and is most evident at the periphery of the haema-toma, leaving a translucent zone between it and theperiphery of the shaft of the bone (Fig. 2).
Infants, apparently healthy in other respects, havebeen seen at birth with considerable calcium inhaematomata associated with fracture, indicatingintra-uterine damage of some weeks' duration.Massive haematomata with much calcium surround-ing both femora of an unusual infant at birth have
been illustrated (Brailsford, 1948). Certainly, withinthe first few months of life radiographs may showamorphous calcium deposition in haematomataenveloping one or more long bones, which appearto have normal structure and density, and, thoughno other definite indication of scurvy may be present,organization, ossification and complete restitution tonormal bone takes place readily on the adminis-tration of vitamin C. These features are illustratedby the sequence of radiographic changes in Fig. 2.
Fig. 2 illustrates the case of an infant. aged 3months, who developed a swelling in the left thigh.A radiograph of the thigh on August 15, 1947 (Fig.2A), showed amorphous calcium deposited in alarge subperiosteal haematoma which envelopedmost of the shaft of the femur. A further radiograph(Fig. 2B) on August 28, 1947, showed an increase inthe size of the lesion, and its definition suggested thatrapid organization and ossification was taking place.The radiograph of September 8, 1947 (Fig. 2c), showsthat within a month most of the haematoma hadbeen converted into bone. On the clinical andradiographic evidence this was interpreted as sar-coma and drastic treatment was contemplated. Thefilms were submitted to me by Dr. Patricia Franklyn,and I suggested that the condition was due to scurvyand advised the administration of appropriateamounts of vitamin C. With its exhibition the patientrapidly showed signs of recovery and by January 8,1948 (see Fig. 2D), considerable resolution had takenplace. By 1952 the bone had returned to normal(Fig. 2E). Obviously in this case the clinical signsof scurvy were not present and the radiographicevidence had been misinterpreted but the therapeutictest was both positive and spectacular.
In 1943 in a paper on osteogenesis imperfecta, Ipublished an account of four patients with thisdystrophy who developed an unusual complication,namely, the development around one or more bonesof a swelling, probably as the result of trauma, sinceit was observed that a similar sequence of changes
81
copyright. on 20 A
pril 2018 by guest. Protected by
http://adc.bmj.com
/A
rch Dis C
hild: first published as 10.1136/adc.28.138.81 on 1 April 1953. D
ownloaded from
FIG. I.-Radiographs of limbs of an emaciated and neglected infantaged 11 months.
A.-On admission on January 19, 1948, fractures at the lower ends of the femoraldiaphvses and a subtrochanteric fracture of the upper third femur are seen.
C-Almost complete resolution of haemorrhages.
took place at the site of surgical injury. The traumawas such that it escaped the attention it demanded.Radiographs of the recent swelling of the inner sideof the right femur showed the rounded outline of a
large, soft-tissue mass. Its almost sudden appearance,its extent and its clearly defined rounded borderindicated that it was probably a haematoma. Othertumour masses in the same and adjacent limbssuggested masses of longer duration presenting theappearances which, after a sequence of changes,developed in the most recent (Fig. 3A). The serialradiographic appearances of these lesions were
identical with those seen in haematomata, that is,large, rounded, soft-tissue swelling around the boneof an acute haematoma which gradually took on a
cloudy and later a granulated appearance. Thegreater density indicated calcium deposits in thehaematoma but in this condition the calcium depositis denser than in haematomata associated with nor-
mal bone. With the passage of time organization of
B-After anti-scorbutic feeding.
the calcium and ossification was indicated by thediminishing mass and the corrugated appearance ofits shrunken periphery, and gradually by the defini-tion and density of the bone. The new bone persistedin its irregular form and the details of the envelopedfemoral shaft were completely absorbed (Fig. 31).
This is a very unusual sequence of changes forosteogenesis imperfecta, in which are usually seenfractures which heal without any undue callus forma-tion or around which there may be little evidence ofcallus, the fragments remaining disunited for manyyears or even throughout a subsequent long life.There is only one other condition which resemblesit, the haematoma which forms at the site of a frac-ture of a bone of a paralysed limb of an otherwisenormal skeleton. I have no doubt about the haema-toma being due to injury in such cases, and theresulting bone deformity does not undergo the readymoulding seen in the normal limb (Figs. 4A and B)but there was no indication of paralysis in thesepatients.
I was led to consider the factor which could beresponsible for multiple haemorrhages on relativelyslight trauma, and the possibility of the added com-plication of scurvy suggested itself. Scurvy is theonly condition, apart from haemophilia and leukae-mia and other conditions, with their characteristicclinical signs and symptoms, in which are seenmultiple sites with subperiosteal haematoma aroundbones which may be apparently normal radio-graphically. But in scurvy the haematomata arereadily organized on the administration of vitaminC. They follow a definite sequence of changes. Theycalcify and become ossified, the new bone is grad-ually absorbed, and finally the affected bone becomesand remains indistinguishable from the normal in
copyright. on 20 A
pril 2018 by guest. Protected by
http://adc.bmj.com
/A
rch Dis C
hild: first published as 10.1136/adc.28.138.81 on 1 April 1953. D
ownloaded from
shape and size and structure (Figs. 1 and 2). In osteo-genesis imperfecta we see the initial stages as far asossification, but not the gradual restitution of thebone to the normal. We must however rememberthat osteogenesis imperfecta is what its name implies,an imperfect ossification, seen in the disturbedgrowth at the metaphyses (Fig. 3B), and it seemsreasonable to suggest that this disturbance in growthis responsible for the bone of unusual character,organized from haematomata, seen in osteogenesisimperfecta. There had never been any suspicion inmy mind of sarcoma; the radiographs illustrated abenign lesion which resolved.Baker (1946) described two similar cases under the
title 'Hyperplastic Callus Simulating Sarcoma inTwo Cases of Fragilitis Ossea'. One of these childrenfalling had sustained a fracture through the femoralshaft around which a large mass formed. Five weekslater it was extensively explored for a completebiopsy. This was considered to show chondro-sarcoma, but amputation was regarded as too riskyand of doubtful value. Although after operation thethigh swelled excessively so that the tumour appearedto fungate through the skin, it subsided and the bonylesion resolved in the manner I have indicated. AsBaker commented, 'Had the limb been removed ...
FiG. 2.-A-Radiographs of the limbs of an infant aged 3 months on
August 15, 1947, with early scurvy. The right thigh shows a fusiformtumour with cakifiction in a large haematoma. The calium is in an
amorphous form, is ill-defined, and is separated from the cortex of thefemur by a parallel translucent zone. The diaphysis and the epiphyses of
the left femur and tibia are normal in appearance.
.dw
B-The right thigh onAugust 28. 1947, show-ing that some organiza-tion of the calcium has
already taken place.
C-Same limb on Sep-tember 8, 1947, afteradministration of vita-min C had been started.The mass has now beenconverted into bone with
good definition.
D-Same limb on January8, 1948, when consider-able resolution has taken
place.
E-Same limb on August 30, 1949.The bone has now almost normaL-characteristics. The epiphyses show
no sisn of nast scurvy.
copyright. on 20 A
pril 2018 by guest. Protected by
http://adc.bmj.com
/A
rch Dis C
hild: first published as 10.1136/adc.28.138.81 on 1 April 1953. D
ownloaded from
ARCHIVES OF DISEASE IN CHILDHOOD
A
No adequate explanation has been put forth. It isstated by many that there is no evidence of scurvy,syphilis, or rickets and that the lesions invariablyresolve without any specific treatment. It may bethat the observers have looked for gross evidence andhave not appreciated that scurvy may exist in certaincases before such gross evidence has been revealed.It has been indicated that it can occur without the
B
bones showing any evidence of departure from thenormal texture and density, one or more fracturesbeing the only radiographic evidence to indicateabnormality even though clinically the infant mayshow profound malnutrition. Fig. 1 shows the
84j,.
copyright. on 20 A
pril 2018 by guest. Protected by
http://adc.bmj.com
/A
rch Dis C
hild: first published as 10.1136/adc.28.138.81 on 1 April 1953. D
ownloaded from
SOME RADIOGRAPHIC MANIFESTATIONS OF EARLY SCURVYappearance of the bones of such an infant aged 11months. There is no radiographic indication ofscurvy though the patient was markedly emaciatedand had obviously been seriously neglected. Thefractures had been present for 10 to 14 days beforethere was any radiographicevidence of the sub-periosteal FIG. 4.-Radiograp
oq v u n~1__k..+ ;+ w 1__iilcIIiurfrI7Ige out it was cicarthat other fractures had occur-red and had resolved withouttheir presence being suspected.The rapid improvement in theclinical condition followingadequate food containing freshfruit juices and codliver oil ledto the development of a finehealthy infant of nearly twicethe weight within a fewmonths. The clinical improve-ment was associated with cal-cification and ossification ofthe haemorrhagic foci, andwithin six months even themore massive periosteal massesof irregular new bone had beenabsorbed.That deficiency and endo-
crine imbalance can be demon-strated in infants at birth weknow from studies of foetalrickets and hyperparathyroid-ism. We know also that one ofthe earliest signs of scurvyis subperiosteal haemorrhagewhich may be so large as to bedetected clinically, yet theradiographs may show appar-amntlw l%r%m%lhnnfX for:z'or more, and only whensufficient calcium has beendeposited to give contrast tembeAorap7, shoniSeabove that of the soft tissues calification and ossifica-can we see the size and posi- i aupperaturmofttion of haematomata. Cer- tibia and fibula onetainly not until then can we month old.detect clinically or fromthe radiographs subperiostealhaemorrhage of a minordegree such as could lead to the production of theradiographic appearances of cortical hyperostoses.The statement that the hyperostoses resolve with nospecific medication tends to underestimate the atten-tion which would be given to the feeding of theinfants in whom these lesions have been found. Thisresponse could be regarded as the positive thera-
peutic test of scurvy. Such children would certainlybe given adequate vitamins in their food; it would beunreasonable to deprive them to prove that lack ofvitamins checked resolution. As will be seen fromFig. 2 the radiographic appearances of the bones are
hs of a paralysed limb.
B-Radiograph on June1, 1949, showing theunusual ossification ofthe upper third of thetibia at the site of thehaematoma, also morerecent fractures of thelower thirds of the tibiaand fibula with sub-periosteal haemorrhages.
only disturb and
normal; there is no indicationin their shape, texture, densityor growth centres of scurvy;only the signs of resolution ina large haematoma indicatedthis. Clinically that lesion wasinterpreted as a sarcoma, andthe radiographic evidence wasthought to support this andamputation considered, butfollowing the author's inter-pretation of scurvy the lesioncompletely resolved on theadministration of vitamin C.What is difficult to under-
stand is the relatively suddenoutburst of so many cases.Many, it would seem, developin the U.S.A. Some isolatedcases have been reported else-where. In Birmingham I didthe radiography for all theinfant welfare clinics of thecity for over 20 years andnever met with one such case.The suggestion could be madethat such cases were not sentfor radiography, but, with theprominence of the local signs,such suspicion of neglect ofinvestigation on the part of theclinicians concerned I shouldregard as unreasonable, bear-ing in mind the triviality ofthe clinical signs in patientswho were sent for radiography.
Multiple biopsies have beenperformed on many cases ofinfantile cortical hyperostosesbut no evidence of any valuefor the understanding of thecondition has been so ob-tained. Indeed biopsies could
delay complete resolution oflesions which, untouched, have been shown to re-solve completely within the minimum of time.Explorations are not made before the lesions haveshown considerable resolution: the evidence pre-sented then indicates that even the nature of thelesion could not be determined.
85
copyright. on 20 A
pril 2018 by guest. Protected by
http://adc.bmj.com
/A
rch Dis C
hild: first published as 10.1136/adc.28.138.81 on 1 April 1953. D
ownloaded from
86 ARCHIVES OF DISEASE IN CHILDHOODIt may be that some deleterious factor for the
foetus and infant is operating through the mother.I have evidence which indicates that the mother cantransmit a 'resistance' to ordinary doses of vitaminD and calcium; further, not all the physical andchemical factors exhibited, with or without the know-ledge of the doctor, during the care of the motherare necessarily beneficial to the welfare of the infant.And in any discussion of the conditions which areassociated with the radiographic appearances ofperiosteal irregularities the effects of the adminis-tration of vitamins must be considered. I saw oneinfant with the typical radiographic evidence ofscurvy, to whom, under medical supervision, massivedoses of vitamin D had been given over a longperiod. The skeleton gradually becam denser anddenser until it became of the density of Albers-Schonberg's disease. The haematomata did not showthe healing response seen after vitamin C but per-sisted and developed an even greater density beforethe infant died.The radiographs of infants between the ages of 6
months and 3 years, who have had excessive vitaminA, show some periosteal reaction and increaseddensity at the growth extremities of the diaphyses,but, unlike cases showing infantile cortical hyper-ostoses, the metatarsals show periosteal accretions.Arena, Sarazen and Baylin (1951) reported hyper-vitaminosis A in one infant of 62 months. Radio-graphs showed craniotabes and periosteal accretionswhich readily resolved when administration ofvitamin A ceased. Chronic poisoning due to excessvitamin A has been discussed and recorded byCaffey and Silverman (1945). Two of their casesshowed periosteal accretions resembling the lesssevere cortical hyperostoses at the late age of 25months, but no facial swelling or hyperostoses of themandible. Seven of the patients had been givenexcessive amounts of vitamins A and D over longperiods. The limbs were swollen and painful. Theribs, clavicles and long bones showed periostealchanges, but those bones which were unaffectedseemed to be normal in appearance. Improvementwas noted within a few days of stopping the intakeof vitamins and the hyperostoses gradually resolved.(The vitamins are usually administered in oleumpercomorphum which is said to contain 60,000 unitsof vitamin A and 8,500 units of vitamin D per g.)Such preparations are readily available in the drugstores without any need for medical supervision.
During my tour of the U.S.A. during 1948 and 1949in the popular magazines and in the parlour cars ofthe trains I could not but overhear from membersof the public of their common use, Indeed through-out that tour I 'learnt' more in this way of allergies,infections, isotopes and the need for shots, vitaminsand periodic changes in babies formulae than in anyperiod of my medical career.
Su*The well recognized clinical and radiographic signs
of scurvy are usually not seen until the infant is 8 to18 months of age, but the earliest radiographic signsmay be those of fracture with a surrounding haema-toma. In some cases such lesions may be seen with-out any definite clinical evidence of scurvy; in othercases emaciation and other signs of neglect may bepresent and yet the bones may have normal radio-graphic density and regularity.The recent haematoma can be suspected from the
presence of a fusiform soft-tissue swelling; theradiograph will reveal no more at this stage, butwithin seven to 14 days radiographic signs of thedeposition of amorphous calcium can be detectedtowards the periphery of the haematoma. This isfollowed by a denser granular deposition and laterits organization into bone when it shrinks and losesits regular peripheral outline. The lesion may bemistaken clinically, histologically and radiographic-ally for sarcoma, but with vitamin C medication theossified haematoma is gradually absorbed and thebone is left with normal features as distinct from thescurvy of the later age period. The radiographicfeatures of unusual haematoma seen in certain casesof osteogenesis imperfecta and paralysed limbs andthe lesions of infantile cortical hyperostoses are con-sidered in relation to the sequence of radiographicchanges in the lesions in scurvy. The best proof ofscurvy may be the response to the therapeutic test-complete resolution of the lesion when adequatevitamin C is administered. The effect of vitamins onthe bones is described.
REFERENCESArena, J. M., Sarazen, P. and Baylin, G. J. (1951). Pediatrics. 8. 788.Baker, S. L. (1946). J. Path. Bact., 58. 609.Brailsford, J. F. (1943). Brit. J. Radiol., 16. 129.
(1948). The Radiology of Bones and Joints. 4th ed. London.Caffey, J. and Silverman, W. A. (1945). .Amer. J. Roentgenol., 54. 1.Fairbank, H. A. T. and Baker. S. L. (1948). Brit. J. Surg., 36. 1.Hutchison, R. and Moncrieff, A. (1941). Price's Textbook ofMedicine.
6th ed., p. 454.Watson-Jones, R. (1952). Fractures and Joint Injuries, 4th ed.
Edinburgh.
copyright. on 20 A
pril 2018 by guest. Protected by
http://adc.bmj.com
/A
rch Dis C
hild: first published as 10.1136/adc.28.138.81 on 1 April 1953. D
ownloaded from
