We report two boys with juvenile dermatomyositis (JDM) complicated by pancreatitis. One also had hepatitis and probably mild bowel vasculitis, while the other ...
British Journal of Rheumatology 1997;36:912–916 PAEDIATRIC RHEUMATOLOGY SERIES EDITOR: P. WOO
SEVERE JUVENILE DERMATOMYOSITIS COMPLICATED BY PANCREATITIS Y. SEE, K. MARTIN, M. ROONEY and P. WOO Department of Rheumatology, Great Ormond Street Hospital for Sick Children, London SUMMARY We report two boys with juvenile dermatomyositis (JDM) complicated by pancreatitis. One also had hepatitis and probably mild bowel vasculitis, while the other had catastrophic bowel vasculitis with multiple perforations. Both were on corticosteroids, but had features of active vasculitis. The former improved with high-dose i.v. pulsed methylprednisolone, while the latter improved only after immunosuppression with i.v. methylprednisolone, cyclophosphamide and plasmapheresis. Although bowel vasculitis is a known complication of severe JDM, pancreatitis and hepatitis are extremely rare. We have found in a literature search only three other reports of pancreatitis complicating JDM. We wish to alert physicians that pancreatitis may develop in JDM. It should be considered as a differential diagnosis in the child with active disease who develops abdominal pain. Control of vasculitis with adequate immunosuppression, as well as general supportive measures, may be valuable in the treatment of pancreatitis in JDM. K : Juvenile dermatomyositis, Pancreatitis, Adequate immunosuppression.
J dermatomyositis (JDM) is an autoimmune disease characterized by rash and proximal myopathy. It is different from the adult form in that it is not associated with malignancy. Well-recognized lifethreatening complications include gut vasculitis with haemorrhage and perforation [1, 2]. We describe here two boys with active JDM who in addition developed pancreatitis.
was focally raised, but normal overall. In the specimen taken, no inflammatory cell infiltrate was seen. The changes, though not typical of, were considered consistent with JDM on review at Great Ormond Street Hospital (GOSH). EMG done previously showed small spiky and polyphasic units of patchy distribution in the vastus medialis and rectus femoris, consistent with the myopathy of JDM. MRI performed at GOSH showed high signal in the buttocks and thigh muscles, especially the quadriceps and vastus lateralis, consistent with the proximal myopathy of JDM. At referral, his liver enzymes were markedly elevated (see above), prothrombin time (PT) and partial thromboplastin time (PTT) were prolonged. Amylase was also elevated to 147 IU/l (33–135). The erythrocyte sedimentation rate (ESR) was 20 mm/h (0–10) and C-reactive protein (CRP) Q 0.3 (Q1). Blood glucose levels were normal and there was no glucosuria. Table I shows his investigations sequentially with his initial results in the first column. Screening tests for infection were negative: viral screen for hepatitis [hepatitis A, B, C, Epstein–Barr virus (EBV), cytomegalovirus], as well as other organisms reported to be associated with JDM (influenza A, coxsackie B, measles, mumps, rubella, varicella zoster, mycoplasma). Blood cultures were negative. Because of the rather atypical muscle biopsy appearances, we checked for metabolic and muscle diseases. Blood film showed no vacuolated lymphocytes, no excess glycogen nor evidence of acid maltase deficiency. Metabolic screen was negative for total, acyl and free carnitine. Urine was negative for myoglobin. Antinuclear antibodies (ANA), rheumatoid factor (RF), antibodies to extractable nuclear antigen (ENA), antineutrophil cytoplasmic antibodies (ANCA) and liver kidney microsomal (LKM) antibody were absent.
CASE 1 A 14-yr-old Caucasian boy developed classical features of JDM over an 8 month period. He presented first with blistering eruptions over his face, oral ulcers and a photosensitive red rash over the knuckles, elbows, interphalangeal joints, shoulders and face (initially diagnosed as psoriasis). He subsequently developed proximal muscle weakness, periorbital oedema and mild alopecia. He deteriorated with progressive ill health, loss of weight and episodes of abdominal pain. On referral to our service 8 months after onset of disease, he appeared ill, cachectic and jaundiced. He had multiple areas of skin ulceration, as well as elbow contractures and restricted hip movement. There was no hepatomegaly. A clinical diagnosis was made of JDM with hepatitis. Muscle enzymes were found to be raised: creatinine kinase (CK) 409 U/l (normally up to 120 U/l), ALT 490 U/l (10–45), AST 2550 U/l (15–45), LDH 1690 IU/l (360–730). Muscle biopsy previously performed at the local hospital showed focal areas of vacuolation in the muscle fibres associated with loss of ATPase activity, cytoplasmic body formation and fibre necrosis, as well as a hint of perifascicular atrophy. Glycogen content Submitted 24 April 1997; revised version accepted 2 May 1997. Correspondence to: Patricia Woo, UCL Medical School, Department of Molecular Pathology, The Windeyer Building, 46 Cleveland Street, London W1P 6DB.
= 1997 British Society for Rheumatology 912
913
SEE ET AL.: SEVERE JDM WITH PANCREATITIS TABLE I Blood investigation results of Case 1 Parameter
Reference range
CK LDH ALT AST Serum bilirubin Conjugated serum bilirubin Alkaline phosphatase PT PTT Amylase Calcium Albumin
(up to 120 U/l) (360–730 IU/l) (10–45 U/l) (15–35 U/l) (up to 18 mmol/l) (110–440 U/l) (12–16) (27–37) (33–135 IU/l) (2.22–2.58 mmol/l) (35–55 g/l)
On referral prednisolone* started
2 days later
409 1690 490 2550 98 61 337 22.5 63 147 2.16 35
365 142 111 324 12 37.7 407 2.27 34
3 days later IVMP† started 209 q 2250 753 4705 158 110 285 12 35 805 2.27 30
1 week later 20 833 179 302 133 69 262 16.5 37 219 2.30 37
2 weeks later 21 614 121 160 71 264 146 2.27 35
*Prednisolone 50 mg daily. †IVMP, i.v. methylprednisolone 0.5 g, three doses given on alternate days.
Lupus anticoagulant was negative, and the levels of Ig G, A and M showed no immunodeficiency. Anticardiolipin antibody titres were mildly elevated: IgG 14.4 AEU (up to 7), IgM 24.9 AEU (up to 10) and C3 slightly low at 0.65 (0.75–1.65), C4 was normal at 0.37 (0.2–0.65). The lymphocyte count was initially low at 0.71 × 109/l (1.5–4). However, repeat values of these were normal and there were no other markers for lupus. He was commenced on oral prednisolone at 50 mg daily (01 mg/kg/day) after his first clinic visit, but continued to deteriorate. He developed severe abdominal pain with radiation to the back and vomiting. On review 3 days later, he looked very ill, had abdominal distension, guarding, rebound tenderness and deepening jaundice. Bowel sounds were absent. Ultrasound of the abdomen showed a bulky pancreas with increased echogenicity and surrounding fluid consistent with pancreatitis and ascites. There was a small pleural effusion. The intrahepatic biliary tree was prominent, but the common bile duct was not dilated. The gall bladder contained some sludge but there were no stones or thickened gall bladder wall. Abdominal X-ray (AXR) showed some loops of distended bowel, but excluded free air under the diaphragm. CT scan confirmed ascites surrounding the pancreas which enhanced uniformly, but did not demonstate any other underlying cause for the pancreatitis. Blood markers had worsened with rising enzyme levels, especially amylase, ALT and AST (see Table I). Our clinical impression then was of severe JDM with worsening hepatitis, pancreatitis and possible gut vasculitis. A white cell scan performed later showed increased uptake in the small bowel consistent with vasculitis. As his clinical deterioration and raised enzymes were present before the onset of steroids, we attributed his deterioration to inadequate control of vasculitis, i.e. it occurred in spite of rather than secondary to steroids. He was pulsed with i.v. methylprednisolone, three doses of 1 g given on consecutive days. In addition, ileus and pancreatitis were managed conservatively with a nil-by-mouth
regime, nasogastric suction, careful fluid and electrolyte balance, and colloid support. The patient also received i.v. octreotide (somatostatin analogue) to decrease pancreatic secretions, omeprazole to decrease gastric acid secretion, antibiotic cover and self-controlled analgesia via morphine infusion. With this immunosuppression and supportive treatment, he showed dramatic improvement within days, both clinically and in laboratory markers. Within weeks, amylase, muscle and liver enzymes approached normal. He needed total parenteral nutrition for a short period, but was eating heartily and participating in physiotherapy within 2 weeks. He was subsequently maintained on prednisolone and cyclosporin A. CASE 2 An Irish boy first developed proximal muscle weakness at the age of 9 yr in December 1992 when he noticed difficulty getting up from a squatting position and was slow in football. He got progressively weaker, his gait deteriorated and he developed redness over the lateral malleoli, malar prominences, eyelids and metacarpal-phalangeal joints. Muscle biopsy confirmed the diagnosis of JDM and he initially responded to oral steroid treatment at 30 mg (1 mg/kg/day). ANA, RF and antibodies to DNA and ENA were negative. Complement and immunoglobulin levels were normal. His dose of steroids was gradually decreased after the first 3 months. However, he had intermittent episodes of severe abdominal pain over the next 2 yr. Investigations performed locally resulted in a diagnosis of left-sided pelvi-ureteric junction obstruction (PUJO) with hydronephrosis and pyelonephritis which necessitated nephrostomy and subsequent nephrectomy in 1994. The kidney specimen showed chronic inflammation and segmental scarring, but no evidence of vasculitis nor autoimmune aetiology. Over the next year, he developed progressive deterioration with increasing weakness, rash and dysphagia. This was unresponsive to increased oral
914
BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 8
steroid dose (40 mg/kg), i.v. immunoglobulin infusions and cyclosporin A. He also developed recurrent episodes of severe abdominal pain, sometimes with coffee ground vomits, attributed to oesophagitis and gastritis secondary to steroids and non-steroidals. Endoscopy 1 week prior to referral showed oesophageal erosions and a large erythematous area within the antrum of the stomach. He was referred to us aged 13 yr having developed grand mal fits. He was thin and pale, and had classical features of JDM with skin changes (heliotrope rash, Gottron’s papules and nailfold capillary changes), marked muscle weakness and wasting, contractures and calcinosis. MRI of his thigh showed florid i.m. high signals on T2-weighted images compatible with active JDM as well as areas of low signal likely to be due to necrosis. Muscle enzymes were not elevated at this time, which could be attributed to lack of muscle bulk due to long-standing disease. He was given immunosuppression in the form of pulsed i.v. methylprednisolone (three daily doses of 0.5 g with a second pulse given a week later). However, he remained ill and developed features of an acute abdomen after the second dose of i.v. methylprednisolone with vomiting, severe pain, guarding, rebound tenderness and absence of bowel sounds. His fits were more frequent. Serum amylase was normal at 88 U/l (range 33–135 IU/l). Laparotomy and endoscopy at this time (6 November 1995) revealed severe oesophagitis as well as severe duodenal and jejunal vasculitis with two perforations in the duodenum and three in the jejunum which were repaired by oversewing. He was started on daily plasmapheresis, i.v. hydrocortisone and 0.75 g of i.v. cyclophosphamide in place of cyclosporin A. However, 5 days later, he again developed severe abdominal pain radiating to the back with redness over the lower abdominal wall. Ultrasound of the abdomen showed distended small bowel with ileus and ascites, but the pancreas was obscured by bowel gas. A second laparotomy (10 November 1995) performed because of persistent features of an acute abdomen revealed an inflamed pancreas. Serum amylase was noted to be elevated postoperatively at 208 IU/l and subsequently rose to a high of 273 IU/l on 2 January 1996. He did not develop any hypocalcaemia or hyperglycaemia. A screen for infections associated with pancreatitis
(influenza A, coxsackie B, measles, mumps, rubella, mycoplasma, hepatitis A, varicella zoster) was negative and blood culture was negative. IgG for EBV was positive, but IgM negative with no difference on a repeat specimen. Isotope scan for perfusion using unlabelled pertechnetate postoperatively on 17 November 1995 demonstrated good perfusion in the region of the bowel supplied by the superior mesenteric artery. However, he remained extremely ill with a distended abdomen and absent bowel sounds, and a repeat scan on 29 November 1995 demonstrated isotope in the region of the stomach, but very little in the remainder of the gut, and the prognosis was felt to be poor. Intravenous methylprednisolone and i.v. pulsed cyclophosphamide initially 750 mg 3 weekly and then 500 mg monthly, were continued. Daily plasmapheresis (10 times) with fresh frozen plasma replacement was given. Prostacyclin and dopexamine were given to improve gut perfusion, and i.v. omeprazole to reduce gastric acid secretion. Antimicrobial cover was given to prevent secondary infection. Pancreatitis was managed with a conservative regime including i.v. octreotide to decrease pancreatic secretions and i.v. fentanyl via patient-controlled analgesia pump. The course of the illness was protracted and the patient required intensive care for a week. It was 6 weeks before bowel sounds returned. Barium meal and follow through showed no perforation, and so he was started slowly on oral feeds but still required additional parenteral nutrition for 4 months. A subsequent graduated intensive physiotherapy regime helped him regain the strength, coordination and range of movement necessary to stand again after 8 months, and then to walk independently within a year. On recovery from the acute episode of gut vasculitis and pancreatitis, he was maintained on oral steroids, cyclosporin A and monthly i.v. cyclophosphamide. DISCUSSION Pancreatitis is relatively rare in children. It is uncommon in connective tissue disease, but has been reported in SLE in adults and children [3–15], and also in Crohn’s disease in adults [16, 17]. However, there have only been three other reports of pancreatitis in JDM [18–20]. Some of the causes of pancreatitis are listed in Table II. Many cases are of unknown aetiology or are
TABLE II Some causes of acute pancreatitis in children Causes 1. Drugs and toxins [15, 21, 22] 2. Infections [15] 3. Hereditary pancreatitis [15, 23, 24] 4. Obstructive [15, 24] 5. Systemic disease [3–17, 24] 6. Traumatic [15, 25]
corticosteroids, azathioprine, cimetidine, non-steroidals, tetracycline, erythromycin, sulphamethiazole, frusemide, thiazides, valproate mumps, measles, mycoplasma, malaria, rubella, EBV, coxsackie B, influenza A, hepatitis A, varicella zoster biliary tract obstruction by stones, sludge, worms, cysts, tumour, malformation SLE, HSP, PAN, Kawasaki, Crohn’s, hyperlipidaemia, hypercalcaemia, uraemia, refeeding after starvation, carnitine palmitoyl transferase II deficiency blunt injury, surgery, child abuse
SEE ET AL.: SEVERE JDM WITH PANCREATITIS
secondary to a systemic disease process. While available data may not be consistent to support a definite disease association with pancreatitis, drugs may be a co-factor in inducing pancreatitis [21]. Steroids have been implicated as a possible cause of pancreatitis from animal experiments and clinical studies [7, 12, 26–29]. Pancreatitis associated with cortisone therapy has been described in children by Baar and Wolff [18], as well as Marczynska [30]. Possible mechanisms include increased volume/viscosity of pancreatic secretions [31], hyperlipidaemia [29], fluid and electrolyte imbalance [32], hypersensitivity [18] and intravascular coagulation [33]. In the numerous reports of pancreatitis complicating SLE in adults as well as children [3–15], some patients were on treatment (e.g. steroids, azathioprine, antimetabolites, antibiotics, thiazides) or had other conditions (e.g. infection) which may have caused pancreatitis, but in some no other cause was present but active SLE. Reynolds et al. [11] reported 20 cases of SLE, of whom 19 recovered with minimum complications despite increased or continued steroids. The pancreatitis occurred in the absence of other precipitants including corticosteroids in four of their patients (who presented with pancreatitis) and in three previously reported cases [8, 10, 34]. Pancreatic inflammation was typically associated with multiple organ involvement. Autopsy and experimental studies [9, 34] suggest that the vasculitis of SLE with subsequent obliterative arterial and venous thrombosis is the likely pathophysiological basis for pancreatitis. Pancreatitis associated with JDM is very rare. Baar and Wolff [18] reported a 2-yr-old boy with JDM and congestive heart failure who developed abdominal pain and an upper abdominal mass, and who subsequently died. The post mortem showed necrosis of the pancreas and peripancreatic fatty tissue. In this child, pancreatitis could have been due to inadequately controlled systemic disease. Petrou et al. [19] reported an 11-yr-old girl with JDM who developed recurrent pancreatitis. She had been treated with methylprednisolone, cyclophosphamide, azathioprine, cyclosporin, plasmapheresis and thiazides. The pancreatitis may have been due to active JDM or its therapy prior to the development of pancreatitis. Finally, in 1989 Heckmatt et al. [20] reported two out of 14 children with JDM entering a trial of cyclosporin A who had developed pancreatitis previously, although it is not clear whether this was related to drugs, disease activity or some other reason. All 14 children had been on steroids at some time, 10 had received azathioprine. In this report, we have described two boys with chronically active JDM who developed pancreatitis. One child had overt bowel vasculitis with multiple perforations. The other had widespread cutaneous ulceration and hepatitis as well as probable low-grade bowel vasculitis as noted on white cell scan. Pancreatitis was manifested by severe abdominal pain, nausea, prostration, guarding and rebound tenderness. Pancreatitis was confirmed by raised amylase as well as ultrasound and CT scan in one boy, and at laparotomy
915
in the other. Both were treated with corticosteroids, but one had a raised amylase even before starting steroids. They were not at the time on other drugs known to cause pancreatitis and did not have evidence of the infections associated with pancreatitis. They both responded to aggressive measures to suppress vasculitis as well as general supportive measures. One responded dramatically to i.v. pulsed methylprednisolone even though it is a drug that could cause pancreatitis, while the other improved after aggressive immunosuppression. Therefore, we suggest that the pancreatitis in these two boys was part of widespread vasculitic activity due to JDM similar in pathogenesis to SLE patients with pancreatitis. Of some interest are the moderately raised anticardiolipin antibodies in Case 1. A report by Font et al. in 1989 [35] suggests that these antibodies are present in many multisystem autoimmune diseases: 6% in adult dermatomyositis patients. It is important to take abdominal pain seriously in children with JDM. Causes include life-threatening bowel vasculitis with ulceration, haemorrhage, necrosis and perforation, as well as pancreatitis, hepatitis, ascites and renal vessel occlusion. Although pancreatitis is extremely rare, it can be fatal from cardiovascular collapse, respiratory distress or renal failure. It should be considered early in the child with abdominal pain and differentiated from other causes. The diagnosis of pancreatitis is largely clinical. Ultrasound and CT scan help in identifying an enlarged hyperechoic pancreas with surrounding fluid. Amylase values are often raised, but may not be [36]. There are also other causes of a raised amylase in a child with abdominal pain besides pancreatitis, including penetrating or perforated peptic ulcer, mesenteric thrombosis, intestinal obstruction, appendicitis and opiates [25]. However, it was only at laparotomy that pancreatitis was diagnosed in one of our patients as the pancreas was not visualized on ultrasound at the time because of a distended fluid-filled gut. Treatment of pancreatitis should include general supportive measures which include colloid, fluid and electrolyte support, adequate analgesia, and prevention of secondary infection. The use of somatostatin or its analogue octreotide should be considered. Complications such as hypocalcaemia and hyperglycaemia should also be sought and managed where possible. Drugs known to precipitate pancreatitis should be discontinued. If underlying widespread vasculitis is considered the likely cause of pancreatitis, then adequate immunosuppression is necessary to bring this under control.
R 1. Bohan A, Peter JB. Polymyositis and dermatomyositis. N Engl J Med 1975;292:344–7, 403–7. 2. Banker BQ. Dermatomyositis of childhood, ultrastructural alterations of muscle and intramuscular blood vessels. J Neuropathol Exp Neurol 1975;34:46–75. 3. Goldberg BH, Bergstein JM. Acute respiratory distress in
916
4.
5. 6. 7. 8. 9. 10. 11. 12. 13.
14. 15.
16.
17. 18.
BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 8 a child after steroid induced pancreatitis. Pediatrics 1978;61:317. Harvey AM, Shulman LE, Tumulty PA, Conley CL, Schoenrich EH. Systemic lupus erythematosus: a review of the literature and clinical analysis of 138 cases. Medicine 1954;33:291. Kaplan MH, Dreiling DA. Steroids revisited II. Was cortisone responsible for the pancreatitis? Am J Gastroenterol 1977;67:141–7. Mekori YA, Schneider M, Yarelzky A et al. Pancreatitis in SLE: A case report and review of the literature. Postgrad Med 1980;56:245–7. Nelp WB. Acute pancreatitis associated with steroid therapy. Arch Intern Med 1961;108:102–10. Netto AP, Dreiling DA. Pancreatitis in disseminated lupus erythematosus. A case report. J Mt Sinai Hosp 1960;27:291. Pollack VE, Grove WJ, Kark RM et al. SLE simulating acute surgical conditions of the abdomen. N Engl J Med 1958;259:258. Reifenstein EC, Reifenstein EC Jr, Reifenstein GH. A variable symptom complex of undetermined etiology with fatal termination. Arch Intern Med 1939;63:553. Reynolds JC, Inman RD, Kimberley RP et al. Acute pancreatitis in SLE: report of 20 cases and a review of the literature. Medicine 1982;25–32. Riemenschneider TA, Wilson JF, Vernier RL. Glucocorticoid induced pancreatitis in children. Paediatrics 1968;41:428–37. Simons-Ling N, Schachner L, Penneys N, Gorman H, Zillereulo, Strauss J. Childhood SLE associated with pancreatitis, subcutaneous fat necrosis and calcinosis cutis. Arch Dermatol 1983;119:491–4. Sparburg M. Recurrent acute pancreatitis associated with SLE. Am J Dig Dis 1967;12:522–6. Werlin SL. Exocrine pancreas. In: Nelson WE, Behrman RE, Kliegman RM, Arvin AM, eds. Nelson textbook of paediatrics. Philadelphia: WB Saunders Co., 1996:1122–4. Malka D, Levy P, Chemtab A, Bernades P. Acute relapsing pancreatitis revealing Crohn’s disease and regressing under corticotherapy. Gastroenterol Clin Biol 1994;18:892–4. Omata F, Arakawa S, Takahashi, Ueno F. Crohn’s disease associated with pancreatitis of unknown etiology; a case report. Gastroenterol Jpn 1993;28:420–3. Barr HS, Wolff OH. Pancreatic necrosis in cortisone treated children. Lancet 1957;i:812–5.
19. Petrou P, Balagh Z, Rosdy B, Gomer B. Recurrent pancreatitis in Juvenile Dermatomyositis. Orv Hetil 1989;180:1937–9. 20. Heckmatt J, Saunders C, Peters AM et al. Cyclosporin in juvenile dermatomyositis. Lancet 1989; May 13:1063–6. 21. Frick TW, Speiser DE, Bimmler D, Largiader F. Drug induced acute pancreatitis: further criticism. Dig Dis 1993;11:113–32. 22. Mallory A, Kern F. Drug induced pancreatitis. Baillie`re’s Clin Gastroenterol 1988;2:293–307. 23. Elistur Y, Hunt J, Chertow B. Hereditary pancreatitis in children in West Virginia. Pediatrics 1994;93:528. 24. Weizman Z, Durie PR. Acute pancreatitis in childhood. J Pediatr 1988;113:24. 25. Geokas MC, van Lancker JL, Kadell BM, Machleder HI. Acute pancreatitis. Ann Intern Med 1972;76:105–17. 26. Bencosme SA, Lazarus SS. The pancreas of cortisone treated rabbits. Arch Pathol 1956;62:285–95. 27. Stumpf HH, Wilens SL, Somoza C. Pancreatic lesions and peripancreatic fat necrosis in cortisone treated rabbits. Lab Invest 1956;5:224–35. 28. Carone FA, Liebow AA. Acute pancreatic lesions in patients treated with ACTH and adrenal corticoids. N Engl J Med 1957;257:690–7. 29. Oppenheimer EH, Boittnott JK. Pancreatitis in children following adrenal corticosteroid therapy. Bull Johns Hopkins Hosp 1960;107:297. 30. Marczynska RM. Pancreatic necrosis in a case of Still’s disease. Lancet 1957;i:815–6. 31. Nelp WB, Banwell JG, Hendrix TR. Pancreatic function and the viscosity of pancreatic juice before and during cortisone administration. Bull Johns Hopkins Hosp 1961;109:292. 32. Stein AA, Powers SR Jr. Pancreatic acinar ectasia. Arch Pathol 1956;62:494. 33. McKay DG. Disseminated intravascular coagulation. New York: Harper and Row, 1965:43. 34. Bywaters EGL. Peripheral vascular obstruction in rheumatoid arthritis and its relationship to other vascular lesions. Ann Rheum Dis 1957;16:84. 35. Font J, Cerrera R, Lopez-Soto A, Pallares L, Bosch X, Ampurdanes S et al. Anticardiolipin antibodies in patients with autoimmune diseases: isotype distribution and clinical associations. Clin Rheumatol 1989;8:475–83. 36. Clavien PA, Robert J, Mayer P et al. Acute pancreatitis and normoamylasemia: not an uncommon combination. Ann Surg 1989;210:614.
