Annals of the Rheumatic Diseases, 1985; 44, 349-353

Case report

Still's disease and the virus-associatedhaemophagocytic syndromeJ A MORRIS,' A R ADAMSON,2 P J L HOLT3 AND J DAVSON4

From the 'Lancaster Moor Hospital; the 2Royal Lancaster Infirmary; the 3Manchester Royal Infirmary; andthe 4Department of Pathology, University of Manchester, Stopford Building, Oxford Road, ManchesterM13 OPT

SUMMARY A 15-year-old male developed features of Still's disease. He was treated, withincomplete response, with aspirin and corticosteroids. Some 10 months after the onset hecontracted chickenpox and became gravely ill. A diagnosis of histiocytic medullary reticulosis wasmade because erythrophagocytic histiocytes were detected in marrow aspirate films. Necropsystudies showed systemic infiltration by benign-appearing histiocytes and led to the final diagnosisof the virus-associated haemophagocytic syndrome due to varicella-zoster infection in animmunocompromised patient.

The virus-associated haemophagocytic syndrome(VAHS)l is a rapidly progressive disorder charac-terised by systemic proliferation of erythrophago-cytic histiocytes resulting in a clinical syndromeclosely resembling that of histiocytic medullaryreticulosis (HMR), now known as malignant his-tiocytosis (MH).2 The importance of distinguishingthe VAHS from HMR/MH is great, since thecytotoxic therapy appropriate for HMRIMH2 wouldhave fatal consequences if given in a case of VAHS,for which supportive measures alone are needed.3VAHS has mostly occurred as a rare complication inpatients undergoing immunosuppressive therapy,and the virus infections believed to have initiatedthe syndrome have usually been members of theherpes group, which includes varicella-zoster virus.In the example of VAHS we now report the patientsuffered from Still's disease (juvenile chronic arthri-tis, JCA). The syndrome followed a varicella-zostervirus infection (chickenpox) eight months after thestart of corticosteroid therapy.

Case report

A 15-year-old male developed a sore throat,diarrhoea, generalised muscular aching, pyrexia,Acccptcd for publication 4 Scptcmbcr 1984.Correspondence to Dr J Davson.

night sweats, stiffness of the limbs, and a macularrash over the whole of his body. He was admitted toLancaster Royal Infirmary with a swinging tempera-ture 37-o040-O0C. The fauces were mildly inflamed,and small lymph nodes were palpable in bothaxillae, but there was neither generalised lym-phadenopathy nor hepatosplenomegaly.

INVESTIGATIONSHaemoglobin 14-1 g/dl. Leucocytes 31-9 x 109/1(neutrophils 98%). ESR (Westergren) 113 mm infirst hour. Antistreptolysin 0 titre 300 Todd units.Rheumatoid and antinuclear factors both negative.Infectious mononucleosis slide test negative. Viralserology: tests for Q fever, influenza, parainfluenza,psittacosis/LGV agent, adenovirus, mumps, herpessimplex, varicella-zoster, cytomegalovirus, rubella,and measles all negative.

Sternal marrow aspirate showed normal haemo-poiesis. Left axillary lymph node biopsy showed apicture typical of dermatopathic lymphadenopathy.On liver biopsy and scan no abnormality wasdetected.

COURSE OF DISEASEOne month after admission a pericardial friction rubdeveloped. Treatment with prednisolone, 60 mgdaily, produced clinical improvement. The

349

on 11 July 2018 by guest. Protected by copyright.

http://ard.bmj.com

/A

nn Rheum

Dis: first published as 10.1136/ard.44.5.349 on 1 M

ay 1985. Dow

nloaded from

350 Morris, Adamson, Holt, Davson

haemoglobin had now fallen to 9-6 g/dl and theleucocyte count was 25-5 x 109/l (neutrophils 90%,lymphocytes 9%, monocytes 1%). Serum albumin31*0 g/l, globulin 58-0 g/l. Immunoglobulins: IgG,36 5 g/l; IgA, 5-4 g/l; IgM, 2*4 g/l. Electrophoresisshowed a polyclonal increase in gammaglobulins.The varicella-zoster titre was <1/10. HLA-B27antigen not detected. In the synovial fluid leucocyteswere 10-2 x 109/l, mainly degenerate monocytes,ragocytes numerous, suggestive of inflammatoryarthritis.Some 60 days after the onset of symptoms he

developed synovitis of his wrists and knees, whichpersisted until his death. His condition was onlypartly controlled by aspirin 1200 mg qds andprednisolone 5 mg qds, and the leucocytosis per-sisted. He was then transferred to Manchester RoyalInfirmary, where a provisional diagnosis of systemicStill's disease (JCA) was made, based on thepersistent arthritis, the typical rash, the high swing-ing fever, the severe neutrophil leucocytosis, andthe pericarditis. This diagnosis was supported by thepolyclonal increase in gammaglobulins and by thenegative rheumatoid factor test.Ten months after onset his condition deteriorated

a few days after his younger sister and brother hadboth developed chickenpox. He started to vomit,had diarrhoea, and on admission to hospital 'lookeddreadful.' He had a swinging pyrexia and aleukaemoid peripheral blood picture: leucocytes60-0 x 109/1 (neutrophils 86%, metamyelocytes 4%,lymphocytes 7%, monocytes 3%). The haemoglobinwas 10.6 g/dl.

Hydrocortisone 100 mg was given four hourly byintravenous infusion. On the third day he improved,but a maculopapular rash had appeared over thetrunk and limbs, and a pericardial friction rub hadagain become audible. Three days later the rashbecame vesicular and the clinical diagnosis ofchickenpox was made. Fluid from a vesicle wasexamined (Professor M. Longson), and herpesgroup virus particles were identified, confirming thediagnosis of chickenpox.

In the next 10 days the leucocyte count fell rapidlyto 3-1 x 109/l and atypical monocytes were reportedin the blood film. A sternal marrow aspirate showedthe presence of occasional erythrophagocytichistiocytes (Figs. I and 2); histiocytic medullaryreticulosis (HMR/MH) was diagnosed. He becamejaundiced, the spleen was now palpable, thrombo-cytopenia (35-0 x 109/l) was present, and deathfollowed a cardiorespiratory arrest.

NECROPSYMild generalised lymphadenopathy and a moderatehepatosplenomegaly (liver 1800 g; spleen 420 g)

Fig. 1 Sternal marrow aspirate. A large histiocyte withhyperchromatic bean-shaped nucleus and abundantcytoplasm filled with ingested erythrocytes, platelets, andone nucleated haemopoietic cell. (Jenner-Giemsa stain,x 950).

Fig. 2 Sternal marrow aspirate. A histiocyte with aneccentrically placed nucleus and cytoplasm stuffed witherythrocytes, showing the low nuclear-cytoplasm ratiocharacteristic of reactive histiocytes. (Jenner-Giemsa stain,x 900).

on 11 July 2018 by guest. Protected by copyright.

http://ard.bmj.com

/A

nn Rheum

Dis: first published as 10.1136/ard.44.5.349 on 1 M

ay 1985. Dow

nloaded from

Still's disease and the virus-associated haemophagocytic syndrome 351

were found. Systemic infiltration by cytologicallynormal histiocytes involved the liver (Fig. 3), thespleen, the lymph nodes (Fig. 4), and the bonemarrow (Fig. 5). Erythrophagocytosis was marked

(Figs. 4 and 5). The final diagnosis was fatalhistiocytosis which, because the cytology of theinfiltrating histiocytes appeared reactive rather thanneoplastic, was VAHS rather than HMR/MH.

Fig. 3 Liver. The hepaticsinusoids are massively infiltratedby histiocytes with bulky granular

;;-v* :Sv 5 cytoplasm that show mostly lownuclear-cytoplasm ratio. (H and E,- .v3- S ~x350).

jr;

Fig 4 Lymph node. A sinus in/ ~~~~theparacortical area contains

several histiocytes witheccentrically placed nuckei andcytoplsm stuffed with ingestederythrocytes. (HandE, x350).

wp*-rs

ioitw I I*. '14. i O.N - -I'll-,-,%,% --V i. I%..

.0. "P.14 -ZIL--I.-.'- 1.1':--.

on 11 July 2018 by guest. Protected by copyright.

http://ard.bmj.com

/A

nn Rheum

Dis: first published as 10.1136/ard.44.5.349 on 1 M

ay 1985. Dow

nloaded from

352 Morris, Adamson, Holt, Davson

11-

7

~k

Discussion

It seems certain that the patient was initiallysuffering from JCA. The onset at under 16 years,the duration of the arthritis (10 months),4 togetherwith the characteristic rash, the high swingingpyrexia, the marked neutrophil leucocytosis, andthe pericarditis all support this diagnosis. Thenegative result for varicella-zoster viral antibodyobtained early in the illness shows that he had nothad chickenpox before the onset of JCA and thussupports the clinical diagnosis of varicella subse-quently. Varicella-zoster virus is a member of theherpes group, and almost all the reported cases ofVAHS have been due to one or other members ofthis group. ' 5That a concurrent varicella infection inchildren receiving steroid therapy may have seriousconsequences is well recognised.6 In some casesdeath has ensued,7 but the necropsy reports havenot described fatal histiocytosis.We have to accept, with the advantage of hind-

sight, that the diagnosis during life of HMRIMHmade in this case because of the presence oferythrophagocytic histiocytes in a marrow aspiratefilm was incorrect. That the distinction betweenreactive and neoplastic histiocytes may be difficultto achieve, especially in marrow films, has beenemphasised.8

This case appears to be the first reported instance

Fig. 5 Bone marrow(necropsy). Scattered amongthe haemopoietic cells areseveral histiocytes showing lownuclear-cytoplasmic ratio andcontaining platelets anderythrocytes in their cytoplasm.(Hand E, x350).

of JCA complicated by VAHS. One case of rheuma-toid arthritis terminated with HMR/MH,9 but atthat date (1977) VAHS had not been separated fromHMR/MH.Our experience with this case of systemic JCA is

in line with observations') that exacerbations in thisdisease are often preceded by an infection, and thatserious infections are likely in patients takingconsiderable doses of corticosteroids. The histolo-gical findings in this case are closely similar to thosefound in a case of VAHS following an Epstein-Barrvirus infection in a youth aged 16 who died fiveweeks after onset5 and provide support for the viewthat a histological distinction between VAHS andHMR/MH can be achieved. But at the same time wewould emphasise that such a distinction cannotsafely be made on the basis of the appearance oferythrophagocytic histiocytes in a sternal marrowaspirate smear.

Rferences

1 Risdall R J, McKenna R W, Nesbit M F, et al. Virus-associatedhemophagocytic syndrome. Cancer 1979; 44: 993-1002.

2 Lampert I A, Catovsky D, Bergier A. Malignant histiocytosis: aclinico-pathological study of 12 cases. Br J Haematol 1978; 40:65-77.

3 McKenna R W, Risdall R J, Brunning R D. Virus-associatedhemophagocytic syndrome. Hum Pathol 1981; 12: 395-8.

4 Munthe E, ed. The care of rheumatic children. Eular Mono-graph No. 3. Basle: Eular, 1978.

on 11 July 2018 by guest. Protected by copyright.

http://ard.bmj.com

/A

nn Rheum

Dis: first published as 10.1136/ard.44.5.349 on 1 M

ay 1985. Dow

nloaded from

Still's disease and the virus-associated haemophagocytic syndrome 353

5 Wilson E R, Mailuh A, Stagno S, Crist W M. Epstein-Barr 8 Wick M R, Ducatman B S, Pierre R V. Histiocytic hemopha-virus-associated hemophagocytic syndrome. J Pediatr 1981; 98: gocytosis. Arch Pathol 1981; 105: 115.260-2. 9 Crow J, Gumpel J M. Histiocytic medullary reticulosis present-

6 Faliiers C J, Ellis E F. Corticosteroids and varicella. Arch Dis ing as rheumatoid arthritis. Proc R Soc Med 1977; 70: 632-4.Child 1965; 40: 593-9. 10 de Vere-Tyndall A, Bacon T, Parry R, Tyrrell D A J, Denman

7 Cheatham W J, Weller T H, Dolan T F, Dower J C. Varicella: A M, Ansell B M. Infection and interferon production inreport of two fatal cases with necropsy, virus isolation and systemic juvenile chronic arthritis: a prospective study. Annserologic studies. Am J Pathol 1956; 32: 1015-35. Rheum Dis 1984; 43: 1-7.

on 11 July 2018 by guest. Protected by copyright.

http://ard.bmj.com

/A

nn Rheum

Dis: first published as 10.1136/ard.44.5.349 on 1 M

ay 1985. Dow

nloaded from