Seasonal variation in pulmonary thromboembolism

Monthly incidence of significant pulmonary thromboembolism (PTE) at Queen Mary Hospital, Hong Kong, 1987-1992

J7an Feb Mar Apr May J7un J7ul Aug Sep Oct Nov Dec

No. of necropsies 310 259 289 277 260 287 338 286 274 267 288 311No. of cases of significant PTE 14 14 16 11 12 6 7 13 11 14 7 5Significant PTE (% of total necropsies) 4 5 5-4 5-5 4-0 4.6 2-1 2.1 4 5 4-0 5 2 2 4 1 6

of the year, no significant differences couldbe found. When the individual seasons werecompared, the rate of pulmonary throm-boembolism in summer (26/911) was lowerthan that in spring (39/826), p<005. Whenindividual months were compared with the restof the year, a significant difference was foundonly for December, where the rate was lower(p = 005 with Yates's correction). When therates in two consecutive months were comparedwith those of the rest of the year, the troughs

16 - | mUK|D Hong Kong

14

12

Month

Monthly variation in the rates of necropsies with significant pulmonary thromboembolism(Hong Kong) or massive pulmonary embolism (UK).

in both summer (June and July) and winter(November and December) had significantlylower rates (p<0-02 for each period).

DiscussionThe main feature of the pattern of seasonalvariation in Hong Kong is the presence of twotroughs in early summer and early winter. Themain feature of the Birmingham report is thepresence of two peaks in spring and autumn.In fact the two graphs show many similarities,and this pattern of variation of pulmonarythromboembolism is similar throughout thesubtropical and the temperate regions.'4 Thepostulations cited in the Birmingham reportto explain the effect of weather on pulmonaryembolism do not seem to hold for Hong Kong.The temperature in Hong Kong is moderate.There is very little variation and no extremes.Activities are not limited in any way by theweather.'4 The true reasons for this seasonalvariation in the incidence of pulmonary throm-boembolism in Hong Kong, as for the UnitedKingdom, remains unknown.

1 Green J, Edwards C. Seasonal variation in the necropsyincidence of massive pulmonary embolism. J Clin Pathol1994;47:58-60.

2 Chau KY, Yuen ST, Ng THK, Ng WF. An autopsy studyof pulmonary thromboembolism in Hong Kong Chinese.Pathology 1991;23:181-4.

3 Bergqvist D, Lindblad B. A 30-year survey of pulmonaryembolism verified at autopsy: an analysis of 1274 surgicalpatients. BrJ'Surg 1985;72:105-8.

4 Schamoff JC, Rosenberg M, Mistica BA. Seasonal variationin fatal thromboembolism and its high incidence in thesurgical patient. Surg Gynecol Obstet 1963;116:11 -4.

Clin Pathol 1995;48:579-582

Hospital Comarcaldel Noroeste,Caravacade la Cruz,30400 Murcia,Spain:Internal MedicineG F Alguacil-GarciaM Martinez-AlbadalejoB Gonzalez-PinaM de Paco-Moya

Pathology ServiceJ Moreno-Requena

Gastroenterology UnitH Hallal-Hachem

Correspondence to:Dr G F Alguacil-Garcia,Gran Via 12 6°-B,Caravaca de la Crux,30400 Murcia,Spain.Accepted for publication18 July 1994

Idiopathic granulomatous vasculitis: response to

immunosuppresslve therapy

G F Alguacil-Garcia, J Moreno-Requena, M Martinez-Albadalejo, H Hallal-Hachem,B Gonzalez-Pina, M de Paco-Moya

AbstractA case of idiophatic granulomatous vas-

culitis (disseminated visceral giant cellarteritis) is described in an old woman,

the seventh case of this rare disorder re-

ported to date. The main organ affectedwas the liver and, to our knowledge, thisis the first patient to be diagnosed whilestill alive and the only case to have receivedmedical treatment. It is also the first time

that muscular involvement has beendocumented in this condition. Cyclophos-phamide treatment resulted in disap-pearance of symptoms and increase inweight. The patient died of an unrelatedcondition.(J7 Clin Pathol 1995;48:579-582)

Keywords: Visceral giant cell arteritis, cyclophos-phamide.

20C.)

C0

0a2)

579

on Septem

ber 2, 2020 by guest. Protected by copyright.

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.48.6.579 on 1 June 1995. D

ownloaded from

Alguacil-Garcia, Moreno-Requena, Martinez-Albadalejo, Hallal-Hachem, Gonzalez-Pina, de Paco-Moya

e . ..4.:-F..'..o

*~ 4-i......'J_

4.r

Figure 1 Granulomatous vasculitis of the liver. Acute and chronic inflammatinfiltrate of lymphocytes and neutrophils in small and medium size hepatic artfibrinoid necrosis and narrowing of vascular lumen. Haematoxylin and eosin,

.-g H"* -st } b s S*ssX~~oowss!j« ~~~~~~~~~~~~.* -l.~Jt4 9 < 0.

4.

.4.

W~~~~~~k

44 , -----

44

'~ ~~~~~~~~~~~, t

gian "Ml(shor anw.Heaoyi ndesn 2.()Grnlmtu eznzthe~~~~~~~fwallof a eimszdhpti reyjihpesneoeea uliulaein

~ ~ ~

Figure 2 Granulomatous vasculitis of the liver. (A) Hepatic portal artery showingperivascular granuloma, peripheric lymphocytic infiltrate, histiocytes (long arrow), and

giant cells (short arrow). Haematoxylin and eosin, x 325. (B) Granulomatous lesion in

the wall of a medium sized hepatic artery, with presence of several multinucleated giantcells of Langhans type (arrows). Haematoxylin and eosin, x 200.

Within the heterogeneous group of systemiivasculitides, there are some in which the mos

distinguishing histological characteristic is thi

existence of granulomatous lesions in the vas

cular wall. Though the most representativientities of this group are the granulomatou

lung vasculitides (Wegener granulomatosis andChurg-Strauss disease), there are other rareconditions such as granulomatous vasculitis ofthe central nervous system, temporal arteritis,and some cases ofvasculitis associated with theuse of drugs which show similar histologicallesions.'2 In 1973 Lie described the necropsyfindings in four patients who presented withgranulomatous arteritis and giant cells in thesmall and medium sized vessels of several ex-tracraneal organs. He considered that thesepatients had a new illness, closely related totemporal (giant cell) arteritis, which he called"disseminated visceral arteritis of giant cells".Two further cases were subsequently described(also in necropsies),4 and the condition wasrenamed "idiopatic granulomatous vasculitis".4

In this paper we describe the seventh case ofthis rare disorder so far reported, and documentthe clinical response to the immunosuppressivetherapy in this, the only patient with the illnesswho has been diagnosed while still alive andgiven medical treatment.

Case reportA 76 year old white woman with non-insulin-dependent diabetes mellitus, who had beentreated with irregular doses of tolbutamide fortwo years, arrived at our emergency ward be-cause of asthenia, anorexia, fever, ankle oed-ema, diffuse non-specific abdominal pain, andmarked deterioration in her general condition.On physical examination cachexia, skin pallor,hepatomegaly, and moderate malleolar oedemawas observed; there were no cutaneous lesions,palpable lymphadenopathy, signs of neuro-pathy, or other physical abnormalities on exam-ination except for weakness and mild generalmuscular atrophy. Among the laboratory tests,the following results were found: packed cellvolume 30%, white cell count 6-9 x 10' cells/l(neutrophils 77%, lymphocytes 21%, bands2%), platelet count 157 x 109/l, MCV 92 fl,blood glucose 26 mmol/l, alkaline phosphatase440 U/1, gammaglutamyl transferase 86 U/1,serum iron 6 j.mol/l, serum ferritin 40 jg/l,plasma creatinine 98 jimol/l; aspartate trans-aminase, alanine transaminase, sodium, potas-sium, calcium, chloride, bilirubin, and uricacid showed normal plasma values. Total pro-tein concentration was 56 g/l (albumin 24 g/l,al1 globulin 3 g/l, at2 globulin 7 g/l, ,B globulin9 g/l, y globulin 13 g/l). HBsAg and rapidplasma reagin test was negative. Antinuclearantibodies were not found. Prothrombin timeand partial thromboplastin time were normal.Urine results were as follows: protein 0 15 g/24 h, glucose >20 g/1, 10-15 white cells and2-3 red cells per high power field withoutgranular or red cell casts, and Ziehl stainingwas negative. ECG showed non-specific STsegment and T wave abnormalities and chestradiography was normal. An ultrasonographic

ic examination of the abdomen showed mildst homogeneous hepatomegaly and abdominalLe computed tomography (CT) revealed doubtfuls- enlargement of the head of the pancreas.re Barium examination of the colon, oeso-is phagus, stomach, and duodenum showed no

A.*

.9

580

on Septem

ber 2, 2020 by guest. Protected by copyright.

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.48.6.579 on 1 June 1995. D

ownloaded from

Immunosuppressive therapy and idiopathic granulomatous vasculitis

abnormalities. On suspicion of a neoplasm inthe head of the pancreas, a laparotomy was

done, but this did not confirm the findings ofthe abdominal CT scan. A wedge biopsy ofthe hepatic parenchyma was performed and a

peripancreatic lymphatic ganglion and a smallsupernumerary spleen were removed. Histo-pathological examination of the lymphatic gan-glion was normal, but both in the hepaticparenchyma and in the supernumerary spleenmany foci of granulomatous vasculitis were

seen involving several small and medium sizedarteries, with focal lesions of fibrinoid necrosis,severe endarteritis, and acute and chronic peri-arteritis (fig 1). In places these lesions pro-

gressed to granulomas destroying the vascularwall (fig 2A and B), and containing multi-nucleated giant cells of the Langhans type (fig2A and B). Extravascular intralobular gran-

ulomas were also observed in the liver. Lefttemporal artery biopsy showed only arterio-sclerotic changes. Right deltoid muscle biopsyshowed endarteritis and periarteritis with no signof fibrinoid necrosis, together with perivascularinflammatory infiltration. These findings were

compatible with vasculitis of non-specific char-acter.

Since the patient's condition continued todeteriorate, with anorexia, loss of weight, andelevation of serum alkaline phosphatase, it wasdecided to begin treatment with cyclo-phosphamide, 100 mg/day for two weeks andthen 50 mg/day. The initial tolerance was ex-

cellent and she was discharged from hospital.Two months later her symptoms had dis-appeared, her weight had increased by 3 kg, thepacked cell volume was 34%, and the alkalinephosphatase was 112 U/1. The patient did notattend for follow up and we could not contacther.Two and half years later, she came again to

our emergency ward with anorexia, asthenia,arthralgia and itching which had started a

month previously; she was still taking cyclo-phosphamide and a moderate pancytopeniawas evident (packed cell volume 25%, whitecell count 2-9 x 1i09, platelet count 65 x 109/1).Creatinine, uric acid, alkaline phosphatase, as-

partate transaminase, alanine transaminase,lactate dehydrogenase and gammaglutamyltransferase were normal. The patient did notwant to undergo further tests, so the cyclo-phosphamide was stopped and she was givena packed cell transfusion and discharged. Amuscle biopsy from the right deltoid did notshow vasculitis. She died 10 months later, be-cause of cardiogenic shock after an acute myo-cardial infarction; necropsy was not authorised.

DiscussionThis patient's histopathological findings were

those of a granulomatous vasculitis withoutrenal, pulmonary, cutaneous, or neurologicalinvolvement. The pathological features of pul-monary granulomatous vasculitis are similar tothose found in this case (fibrinoid necrosis,vascular and extravascular granulomas, giantcells),'5 but the absence of lung, renal, cut-

aneous, and neurological involvement, eosin-ophilia, asthma, or sinusitis rules out thesepossibilities, in agreement with newly reviseddiagnostic criteria.5 Likewise the presence ofgranulomas and giant cells (without eosin-ophilic or polymorphonuclear infiltrates), inaddition to the absence of renal, cutaneous,neurological, or gastrointestinal involvement,would be very unusual in polyarteritisnodosa.'5 There are several drugs which cancause hepatic granulomas6 or vasculitis7 8; how-ever, in such cases there is commonly eosin-ophilia and cutaneous involvement in theform of leucocytoclastic vasculitis, with littleevidence of fibrinoid necrosis and only rarelygranulomatous vasculitis.278 Our patient wasunder treatment with tolbutamide and thoughthis drug may produce hepatic granulomatosisor cholestatic hepatitis,9 we have not found anycase of visceral vasculitis following its use. Twocases of granulomatous vasculitis with hepaticand visceral involvement have been describedafter treatment with glibenclamide, another sul-phonylurea, and though one of these cases'0had some similarities with ours, both had thetypical characteristics of drug related vasculitis(tissue eosinophilia, cutaneous involvement,and absence of fibrinoid necrosis and giantcells),78 which makes them very different fromthe histological findings in our patient.The clinical and histological characteristics

observed in our patient are similar to those ofother patients described so far with idiopaticgranulomatous angiitis34 (disseminated visceralgranulomatous angiitis), that is, non-specificclinical manifestations, involvement of smalland medium calibre arteries in several organs(liver, pancreas, heart, lymph glands), and thepresence of necrotising granulomatous arteritis(figs 1 and 2), multinucleate giant cells (fig 2),and extravascular granulomas, with absence ofeosinophilia (either in the tissues or in peri-pheral blood) or involvement of skin, centralnervous system, and temporal arteries. Thoughfibrinoid necrosis is a relatively minor featureof this entity, nevertheless in one of the casesdescribed it was prominent,4 as it was in ourpatient (fig 1). It was also striking that therewas marked though non-specific muscular in-volvement in our patient, which was not presentin the other cases described so far.The patient was treated with 50 mg/day of

cyclophosphamide (1 mg/kg/day) for two anda halfyears and though, for reasons beyond ourcontrol, her treatment could only be reviewed at3 and 30 months, she tolerated it well andthere was striking clinical improvement andprobably also histological improvement sincein the second muscle biopsy lesions ofvasculitiswere not found.Though, unfortunately, necropsy could not

be performed, which would have completedthe case description and made clear the causeof death, we believe that the findings reportedhere extend our knowledge about this raredisorder, and above all about its therapy, sincethis is the first time that satisfactory responseto treatment with an immunosuppressive agenthas been described.

581

on Septem

ber 2, 2020 by guest. Protected by copyright.

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.48.6.579 on 1 June 1995. D

ownloaded from

Alguacil-Garcia, Moreno-Requena, Martinez-Albadalejo, Hallal-Hachem, Gonzalez-Pina, de Paco-Moya

1 Cupps TR, Fauci AS. The vasculitides. In: Smith LH, ed.Major problems in internal medicine, vol 21. Philadelphia:WB Saunders, 1981: 26-49 and 72-87.

2 Gaffey CM, Chun B, Harvey JC, Manz HJ. Phenytoin-induced systemic granulomatous vasculitis. Arch PatholLab Med 1986;110:131-5.

3 Lie JT. Disseminated visceral giant cell arteritis. Histo-pathologic description and differentiation from othergranulomatous vasculitides. Am J Clin Pathol 1978;69:299-305.

4 Petersen OMF, Frankel SR. Idiopatic granulomatous vas-culitis. In Churg A, Churg J, eds. Systemic vasculitides.New York: Igaku-Shoin, 1991;181-91.

5 Hunder GG, Arend WP, Bloch DA, Calabrese LH, FauciAS, Fries JF, et al. The American College ofRheumatology1990 criteria for the classification of vasculitis. ArthritisRheum 1990;33:1065-144.

6 Berk JE, Cohen M. Granulomas hepaticos. In: Berk JE, ed.Bockus: Gastroenterlogia, 4th Spanish edition. Barcelona:Salvat, 1987:3475-90.

7 Mullick FG, McAllister HA, Wagner BM, Fenoglio JJ. Drugrelated vasculitis. Clinicopathologic correlations in 30patients. Hum Pathol 1979;10:313-25.

8 Haber MM, Marboe CC, Fenoglio JJ. Vasculitis in drugreactions and serum sickness. In: Churg A, Churg J,eds. Systemic vasculitides. New York: Igaku-Shoin, 1991:305-13.

9 Bloodworth JMB, Hamwi GJ. Histopathologic lesions as-sociated with sulfonylurea administration. Diabetes 1961;10:90-9.

10 Ingelmo M,-Vivancos J, Bruguera M, Sierra J, Balcells A.Angiitis por hipersensibilidad y hepatitis granulomatosainducida por glibenclamida: a proposito de un caso. MedClin (Barc) 1980;75:306-8.

J Clin Pathol 1995;48:582-584

Histological features of the thyroid gland in apatient with lithium induced thyrotoxicosis

Y Mizukami, T Michigishi, A Nonomura, S Nakamura, M Noguchi, E Takazakura

KanazawaUniversity Hospital,Kanazawa, JapanY MizukamiA Nonomura

Departmentof Nuclear MedicineT Michigishi

Departmentof Internal MedicineS Nakamura

Departmentof SurgeryM Noguchi

Kurobe CitizensHospital, Kurobe,Japan: Departmentof Internal MedicineE Takazakura

Correspondence to:Dr Yuji Mizukami MD,Pathology Section,Kanazawa UniversityHospital,13-1 Takara-machi,Kanazawa 920, Japan.Accepted for publication13 October 1994

AbstractA 26 year old woman with lithium in-duced thyrotoxicosis is reported. Thethyrotoxicosis was associated with a non-tender diffuse goitre and a low radio-iodine uptake by the gland. The thyro-toxicosis was reversible and remittedon withdrawal of the drug. The histo-pathological alterations of the thyroidglad were characterised by extensive fol-licular cell disruption with no lympho-cytic infiltration. It is postulated thatlithium might directly damage thyroidfollicular cells and that subsequent re-lease ofthyroglobulin into the circulationmight be a cause of transient thyro-toxicosis.(_7 Clin Pathol 1995;48:582-584)

Keywords: Thyrotoxicosis, lithium, thyroid histology.

Lithium is increasingly used for the treatment ofmanic depressive illness. Thyroid disturbancesduring lithium treatment have been noted.These commonly present as a goitre with orwithout hypothyroidisml'3; more rarely there isa goitre with hyperthyroidism.24 There have,however, been few reports describing the histo-logical alterations of the thyroid gland duringlithium treatment.23 5 7 8

In the present report, we describe the histo-pathological features of the thyroid gland ina patient with lithium induced thyrotoxicosiswhich remitted on withdrawal of the drug.

Case reportA 26 year old woman was referred to us becauseof a two week history of sweating, palpitations,and anorexia. She had a three year history ofmanic depressive psychosis and had been takinglithium carbonate, 800 mg/day, for the pre-ceding two years. She had no history of thyroiddisease and there was no family history ofthyroid disorders. She was clinically euthyroidsix months before this episode, and serum val-ues at that time including a thyroxine (T4)of 90 nmol/I (normal 59-142), a free T4 of19 6 pmol/I (normal 9 0-27-0), and a triiodo-thyronine (T3) of 1-7 nmol/l (normal 1 2-2-9). Physical examination revealed warmmoist hands and fine finger tremor, but ex-ophthalmos was not evident. The thyroid glandwas diffusely enlarged and was non-tender.No nodules were palpable. The serum thyroidhormone concentrations were raised: free T487-5 pmol/l, T3 3-5 nmol/l, and free T313 6 pmol/I (normal 3-4-8 2). TSH was sup-pressed at <20 mU/l (normal <100). Radio-active iodine uptake was markedly reduced toonly 1-0% at 24 hours (normal 10-40%).Serum antithyroglobulin and antimicrosomeautoantibodies were negative. TSH bindinginhibitor immunoglobulin (TBII) was also neg-ative. These clinical and laboratory findingssuggested a diagnosis of silent (painless) thy-roiditis. Large needle biopsy (Silverman) ofboth lobes of the thyroid gland was performed.Lithium therapy was stopped and the patientwas given propranolol for the control of thyro-toxic symptoms. The values of the serumthyroid hormones continued to rise for two

582

on Septem

ber 2, 2020 by guest. Protected by copyright.

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.48.6.579 on 1 June 1995. D

ownloaded from