Acute Abdomen in Systemic Lupus Erythematosus: The Importance of Early Laparotomy

100 Q1997 by Excerpta Medica, Inc. 0002-9343/97/$17.00All rights reserved. PII S0002-9343(97)00020-X

/ 2210 5020 Mp 100 Wednesday Jul 23 10:51 AM EL–AJM (v. 103, no. 1) 5020

Acute Abdomen in Systemic Lupus Erythematosus:The Importance of Early LaparotomyFrancisco Medina, MD, Alonso Ayala, MD, Luis J. Jara, MD, Magdalena Becerra, MD,Juan M. Miranda, MD, Antonio Fraga, MD, Mexico City, Mexico

BACKGROUND: Acute abdomen (AA) in systemiclupus erythematosus (SLE) is a challengingdiagnostic and therapeutic problem. Mostpatients are on steroid and/orimmunosuppressive treatment and mortality ishigh.

METHODS: We assessed the relationship betweenthe causes of AA in SLE and the SLE diseaseactivity index (SLEDAI).

RESULTS: Of 51 patients with SLE and AA, 36had active disease (Group 1) and 15 inactivedisease (Group 2). Group 1 included 19 patientswith vasculitis (mean SLEDAI 15.4, range 13 to24). Three patients with intraabdominalthrombosis and high titers of anticardiolipinantibodies (mean SLEDAI 18.3) and 14 patientswith non–SLE-related AA (SLEDAI 8.2, range 5to 11). Group 2 consisted of 15 inactive SLEpatients (mean SLEDAI 1.7, range 0 to 4).Mortality was high in the active group (14patients) compared with inactive SLE (2 cases).A delay in surgical exploration (39.3 vs 178.6hours) had a negative influence on theprognosis.

CONCLUSIONS: In SLE patients with AA, a SLEDAIscore below 5 is indicative of non–SLE-relatedAA. Elevated aCL were found in patients withintraabdominal thrombosis. AA in inactive SLE isnon–SLE-related and has low mortality, providedan appropriate surgical treatment is given.Early laparotomy influences positively theprognosis of SLE patients with AA. Am J Med.1997;102:100–105. Q 1997 by ExcerptaMedica, Inc.

Acute abdomen (AA) is a severe and often lifethreatening event that requires early diagnosis

and appropriate treatment. The occurrence of AA

From the Rheumatic Diseases Unit (FM, LJJ, JMM, AF), Departments ofSurgery (AA) and Pathology (MB), Hospital de Especialidades Centro Med-ico La Raza, Instituto Mexicano del Seguro Social, Mexico City, Mexico.

Presented in part at the 53rd Annual Scientific Meeting of the AmericanCollege of Rheumatology. Cincinnati, Ohio.

Requests for reprints should be addressed to Francisco Medina, Per-goleros 166-1103, ISSFAM, 14620, D.F., Mexico.

Manuscript submitted December 4, 1996 and accepted in revisedform March 12, 1997.

in SLE is a challenging diagnostic and therapeuticproblem that carries a high morbidity and highmortality rates. Most patients with AA are treatedwith steroids and/or immunosuppressive therapy.1

However, this can increase mortality, especiallywhen surgery is delayed and the cause of abdom-inal pain is not identified early.2 – 4 Previous reportshave emphasized the presence, treatment and out-come of intestinal vasculitis. However, non–SLE-related causes of AA in active and inactive SLEpatients have not been examined in detail.5 – 8

Several indices have been devised for measuringSLE activity,9 but none has been used in studies ofAA in SLE. The decision for surgical intervention inSLE patients with AA is often difficult, particularlyin patients with severe disease activity.10 These pa-tients are prone to perforation and infection, espe-cially if they are taking corticosteroids, immunosup-pressives or nonsteroidal anti-inflammatory drugs,as the usual signs of peritoneal irritation may notfully develop and this situation may delay surgery.11

In addition, AA from various causes may occur inpatients with inactive disease. The purpose of thisarticle is to report our experience with a large seriesof patients with SLE and AA and emphasize the needfor early surgical intervention.

PATIENTS AND METHODSAll patients with SLE and AA admitted to the

Rheumatic Diseases Unit at La Raza Medical Centerentered into an observational study during an 8-yearperiod. All patients fulfilled the revised AmericanCollege of Rheumatology criteria for SLE,12 and allwere hospitalized with an acute event unrelated totrauma. Patients were examined for signs and symp-toms of lupus activity, abdominal pain, fever, abdom-inal distention, rebound tenderness, nausea, vomit-ing, diarrhea and abnormal peristalsis. SLE activitywas quantified on admission using the SystemicLupus Erythematosus Disease Activity Index(SLEDAI).9 Plain abdominal radiographs, obtained inall patients, were evaluated for the presence of ileus,diffuse bowel distention with air–fluid levels, pseu-doobstruction pattern, mucosal abnormalities(edematous haustra and/or valvulae conniventes),thumb-printing and intraabdominal free air. Surgicalconsultation were obtained in all patients to deter-mine the need for surgical intervention. All patients

ACUTE ABDOMEN IN SLE/MEDINA ET AL

August 1997 The American Journal of MedicineT Volume 103 101


TABLE IPatients with Active SLE and Acute Abdomen

Final Diagnosis No. Surgery Mortality

Vasculitis (Subgroup 1A,n Å 19)

Intestinal ischemia 4 3 0Intestinal necrosis 5 4 2Ileal perforation 2 2 1Colonic perforation 2 1 0Necrotizing pancreatitis 2 1 1Acalculous cholecystitis 2 2 1Subtotal 19* 13 9†

Abdominal thrombosis(Subgroup 1B, n Å 3)‡

Mesenteric thrombosis 2 1 1Hepatic arterial thrombosis 1 1 0Subtotal 3 2 2§

Non–SLE-related(Subgroup 1C n Å 14)

Acute appendicitis 3 3 1Lithiasic cholecystitis 3 3 0Perforated duodenal ulcer 2 2 1\

Acute pancreatitis 2 2 0Pyocholecyst 1 1 0Retroperitoneal hematomaØ 1 1 0Pancreatic abscess 1 0 1Negative laparotomy 1 1 0Subtotal 14 13 3

Total 36 28 14* Two patients did not undergo surgery due to good response to steroidtreatment.† Surgery was not performed in four patients because of ASA V score.‡ aCL titers ú5 SD.§ Surgery not performed because of ASA V score.\ Without evidence of vasculitis.Ø Anticoagulant therapy.

were treated with intravenous hydrocortisone 60 mg4 times daily during the first 12 to 24 hours, afterwhich they underwent surgical examination unlessa complete resolution of the abdominal event wasobserved. Laboratory tests obtained on admission in-cluded complete WBC count, ESR by a modifiedWintrobe technique; prothrombin time, kaolin-acti-vated thromboplastin time; and serum creatinine,BUN, electrolytes and amylase. Antinuclear antibod-ies (ANA) were determined by immunofluorescencewith HEp-2 cells. Antibodies to double stranded-DNA (ds-DNA) were measured by the Crithidia lu-

ciliae method. Anticardiolipin antibodies (aCL) weredetermined by ELISA and reported in Arbitrary Units(AU).13 Serological tests for syphilis (VDRL and FTA-ABS) were also performed. Rheumatoid factor wasmeasured by nephelometry and latex fixation test.C3 and C4 were determined by radial immunodiffu-sion and total hemolytic complement (CH50) wasmeasured by the method of Mayer, modified.14 Oralcholecystography, upper gastrointestinal series, ar-teriography, ultrasonography and computerized ax-ial tomography of the abdomen were recorded whenavailable.

The SLEDAI consists of a weighted score for 24specific clinical manifestations and laboratory pa-rameters. Values are higher for potentially life-threatening organ and system involvement. PossibleSLEDAI scores range from 0 to 82. Because AA is asevere event, for the purpose of this study, and al-though it may signify low degree activity, we defineda patient as inactive until a SLEDAI score of 4 wasreached.

For histopathologic studies, paraffin-embeddedblocks were cut and stained with hematoxylin-eosin.All biopsy specimens were examined for the pres-ence of inflammatory vascular disease. Special stainsincluded periodic acid-Schiff, reticulin, phospho-tungstic acid-hematoxylin, Mason’s trichrome, acid-fast and Gram stains. Differences in nonparametricdata were tested by the Mann-Whitney U test, and bythe chi square test. A P value ofõ0.05 was consideredsignificant.

RESULTSFifty-one SLE patients (50 female, 1 male) with

signs and symptoms of AA were studied. Mean agewas 29.41 { 11.00 years (range 15 to 61), and meandisease duration at the time of the acute abdominalevent was 31 months (range 0 to 216).

Patients were divided in two groups according toSLEDAI scores: Group 1: patients with active disease(n Å 36), and Group 2: patients with inactive disease(n Å 15). Patients in Group 1 were further dividedinto three subgroups according to histopathologicalfindings: 1A) patients with vasculitis (n Å 19), 1B)

patients with intraabdominal thrombosis (n Å 3)and, 1C) patients with non-SLE related AA (n Å 14).The 1A subgroup consisted of 19 patients with amean SLEDAI score of 15.4, (range 13–24). In twocases, AA was resolved promptly with 60 mg hydro-cortisone 4 times daily and they did not need sur-gery. Thirteen cases underwent laparotomy due topersistent AA despite steroid treatment. AA was theinitial manifestation of SLE in two cases. In four pa-tients surgery was not performed due to critical con-dition classified as American Society of Anesthesi-ology (ASA) score V (mortality over 90% in the next24 hours with or without surgery)15 (Table I). Eightpatients had successful recovery, and 5 died. Sub-group 1B (abdominal thrombosis without vasculitis),consisted of three active patients with a SLEDAIscore of 16, 18 and 21. Of 2 patients that underwentsurgery, 1 died from massive intestinal thrombosis,and 1 with hepatic arterial thrombosis improvedwith thrombectomy and 100 mg of prednisone perday. The third patient had ASA V score and died be-fore laparotomy. Mesenteric thrombosis was re-vealed at autopsy. These three patients had high lev-


102 August 1997 The American Journal of MedicineT Volume 103


TABLE IINon–SLE-related Acute Abdomen in Inactive SLE

(Group 2: n Å 15 patients; events Å 18)

Final diagnosis No. Surgery Mortality

Acute appendicitis 3 3 0Acute pancreatitis 3 1 0Lithiasic cholecystitis 2 2 0Mucinous ovaric cystadenoma 1 1 0Policystic ovarian 1 1 0Perirenal abscess 1 1 0Retroperitoneal abscess 1 1 0Hepatic amebic abscess 1 1 0Tubo-ovarian abscess 1 1 0Tubarian pregnancy 1 1 0Perforated colonic diverticulum* 1 1 0Perforated gastric ulcer* 1 1 1Appendicitis followed by

Retroperitoneal abscess 1 1 1Total: 18 16 2* Without evidence of vasculitis.

els of aCL, defined as more than 5 SD above normal(mean titer IgG 18.3, IgM 10.4 A.U.), and 2 had thelupus anticoagulant.

Subgroup 1C (AA unrelated to SLE), consisted of14 patients with a mean SLEDAI score of 8.2 (range5 to 11). Twelve of these patients underwent surgicalexamination that revealed diverse abdominal dis-eases and primary intra-abdominal infections (TableI). Three patients died, 1 due to perforated duodenalulcer, 1 due to perforated appendicitis with perito-nitis and another due to pancreatic abscess. Two ofthese patients had delayed surgery (4 and 8 days),and in 1 patient surgery was not performed (nec-ropsy finding). Group 2 had 15 inactive SLE patientswith a mean SLEDAI score of 1.7 (range 0 to 4).There were 18 abdominal events in this group (3 pa-tients had two AA each). Sixteen exploratory lapa-rotomies were carried out, which revealed a diverseabdominal pathology (Table II). Two patients dieddue to delayed surgery (7 and 30 days, respectively).One had a perforated peptic ulcer, and the other hadappendicitis and a retroperitoneal abscess.

Eight of the 51 patients had an exudative sterileascites at laparotomy. Three cases had a positiveANA and low complement levels in the ascitic fluid.Intestinal vasculitis was shown in 7 of these patients.One patient lacked relevant features at laparotomybut improved with an increased steroid dose.

As a diagnostic tool, paracentesis and peritoneallavage was performed in another 8 patients; in 5 ofthem it was positive (vasculitis in 2 cases, intraab-dominal infection in 2 and hematoma in 1). However,in the 3 with negative paracentesis, 1 patient had in-testinal vasculitis and other had retroperitonealabscess.

Once intraabdominal infection was ruled out, 9 pa-tients with vasculitis shown at surgery were treatedwith methylprednisolone pulse therapy 1 g per dayfor 3 days, followed by cyclophosphamide 1 g/m2 in-travenously in 4 cases. Eight patients improved, and1 patient died due to widespread ileal necrosis.

Table III shows the clinical and laboratory find-ings of AA patients. All patients with abdominalthrombosis had high levels of aCL (ú5 SD). On theother hand 66% of patients with vasculitis, 50% ofpatients with non-SLE and 44% of patients with in-active SLE and non–disease-related AA had slightlyelevated, but not significant, aCL. Table IV showsthe results of a logistic regression correlating centralnervous system involvement (OR 22.28), thombocy-topenia (OR 11.57), and cutaneous vasculitis (OR8.85) with AA due to vasculitis. Lymphopenia wasassociated with global activity disease but not withAA due to vasculitis and/or thrombosis. Other cor-relations lacked significance.

The overall mortality rate of AA in SLE was31.37%, however, the mortality of SLE-related ab-dominal complications (Subgroups 1A, 1B) was 50%.In contrast, the mortality of non–SLE-related ab-dominal pathology in active SLE (Subgroup 1C) was21%, and in the inactive SLE patients (Group 2) itwas 13%. A comparison was made between early (24to 48 hours) and late surgical exploration (ú48hours), showing that early laparotomy increased thesurvival rate (P õ0.001) (Table V). A delay in sur-gical examination had a negative influence in the sur-vival rate of all SLE patients with AA. The mean de-lay from onset of the abdominal event to surgery was39.3 hours in the 34 patients who survived and 178.6hours in the 10 patients who died (P õ0.001). Of 16patients who died (14 with active and 2 with inactiveSLE), an initial diagnosis of vasculitis was made in14, mesenteric thrombosis in 1 and pancreatitis inanother. In the 14 active SLE patients, vasculitis wassuspected in 12 and confirmed in 8. The 4 misdiag-nosed cases had intraabdominal abscesses. In the 2inactive SLE patients, the suspected diagnosis wasvasculitis, but both had peritonitis secondary to ap-pendicitis in 1 case and perforated gastric ulcer inthe other. Other findings associated with a high mor-tality rate included an elevated ureic nitrogen in 13of 14, severe lymphopenia, below 1,000 per mm3 in12 of 14, thrombocytopenia (less than 100,000 permm3) in 12 of 14 and sepsis in 10 of 14.

DISCUSSIONGastrointestinal manifestations of SLE were first

noted by Osler, who reported in 189516 11 cases oferythema exudativum multiforme with abdominalcrises. Acute abdomen in SLE is a severe and life-threatening event that requires early diagnosis and




TABLE IIIClinical and Laboratory Findings in Acute Abdomen

Group 1: Active SLE

1A Vasculitis

n Å 19 %

1B Thrombosis

n Å 3 %

1C Non–SLE-related

n Å 14 %

Group 2: Inactive SLE

n Å 18 %

Thrombocytopenia 16 84 1 33 6 42 2 11CNS involvement 15 78 1 33 6 42 2 11Hemolytic anemia 9 47 0 0 3 21 1 5Cutaneous Vasculitis 14 73 2 66 8 57 1 5Lymphopenia 18 94 3 100 13 92 4 22Anticardiolipin ab. 13 68 3 100 7 50 8 44Rheumatoid factor 7 36 1 33 4 30 4 22Osteonecrosis 5 14 1 33 3 21 3 16

TABLE IVOdds Ratio and Logistic Regression of Clinical and Laboratory Features

Present in SLE Patients with AA due to Intestinal Vasculitis

Odds Ratio (95% CI) f P

CNS involvement 22.28 (2.89–171.1) 3.104 0.003Thrombocytopenia 11.57 (1.067–125.5) 2.449 0.044Cutaneous vasculitis 8.857 (1.626–125.5) 2.181 0.012Rheumatoid factor 1.00 (1.00–1.00) 0.1458 E-04 0.048Lymphopenia 0.9983 (0.99–1.00) 0.1748 E-02 0.053

TABLE VImportance of Early Surgical Treatment of Acute Abdomen in 44 Patients with SLE*

24–48 Hours†

Operated Survived Died

ú48 Hours

Operated Survived Died

Overall patients 33 33 — 11 1 10Active SLE 19 19 — 9 1 8Inactive SLE 14 14 — 2 — 2* Surgery was not performed in 6 patients because of ASA V score. Four patients did not undergo surgery due to good response to medical treatment (2vasculitis and 2 pancreatitis).† From established acute abdomen symptomatology to onset of surgery.

treatment. Different studies have pointed out thatvasculitis is the cause of AA in about 60% of SLEpatients.8,17–19 In the present study, we found vas-culitis as causing AA in only 35% of all patients. Thispercentage increases to 53% in active SLE patients.These cases may require surgery due to necrosis, in-farction or perforation. Moreover, active SLE pa-tients may also have non–SLE-related AA that re-quires surgical treatment. The present series showsthat nonvascular conditions such as cholecystitis,appendicitis and abscesses are almost as frequent asabdominal vascular events in active SLE patients.The assumption that, in active SLE patients, most ofthe abdominal problems are due to vasculitis maylead to serious treatment errors, as these patientsmay be treated with high steroid doses and surgicalexploration may be further delayed.

An abdominal emergency due to vasculitis may bethe first manifestation of SLE,20 this occurred in 2 of

our patients. The association of abdominal vasculitiswith cutaneous vasculitis, thrombocytopenia, cen-tral nervous system involvement, rheumatoid factorand osteonecrosis has been previously empha-sized.7,8 In our study, abdominal vasculitis was as-sociated with thrombocytopenia, neurological in-volvement and cutaneous vasculitis, but not withlymphopenia, osteonecrosis and rheumatoid factor.

Antiphospholipid antibodies have been associatedwith an increased risk for venous and/or arterialthrombosis.21 In this series, all patients with abdom-inal thrombosis had high titers of aCL (ú5 SD). Oneof these patients had hepatic arterial thrombosisand, fortunately, with thrombectomy and high dosesof steroids, this complication was resolved. Theother 2 patients died due to massive intestinal throm-bosis, as has been reported.22 In this connection, adevastating noninflammatory vasculopathy may oc-cur abruptly in antiphospholipid syndrome, and may


104 August 1997 The American Journal of MedicineT Volume 103


be the clinical presentation of this severe and oftenlethal disease.23

The radiologic signs of AA are often discreet andnonspecific. As observed in the present and in previousstudies,24 many of the macroscopic lesions seen at sur-gery or autopsy do not lend themselves to recognitionby roentgenography, until the diagnosis has been de-layed and the patient is critically ill.

An association between pancreatitis and SLE wasnot mentioned in early descriptions of SLE2,6,17; how-ever, Reynolds et al25 reported that 20 (37%) of 63SLE patients with abdominal complaints had pancre-atitis. In our study, of 8 patients with pancreatic in-volvement (16%), pancreatitis was found in 7 casesand pancreatic abscess in the other patient. Fourevents of pancreatitis were present in patients withactive disease. In most of them the process resolvedwith an increase in the steroid dosage and medicalmeasures, suggesting that, in some cases, pancrea-titis was related to SLE. Two patients died from nec-rotizing hemorrhagic pancreatitis, with histologic ev-idence of pancreatic vasculitis. In contrast,pancreatic biopsy performed at laparotomy showedonly edematous changes in 2 cases (1 with activeSLE), indicating that non–disease-related pancrea-titis is also present in SLE patients.

As reported by Lang,26 colonic perforation, dif-fuse intestinal ischemia and ileal necrosis evolvedduring intensive immunosuppressive treatment in4 of our patients. These cases emphasize the needfor early exploratory laparotomy in SLE patientswith AA on high doses of steroids or immunosup-pressive therapy. Persistence or progression ofclinical findings should be followed promptly byexploratory laparotomy.

The non–disease-related causes of AA in SLEhave not been studied adequately. Flanigan et al11

reported 63 patients with AA and collagen dis-eases. They did not separate the different diseases,but found vasculitis in only 9% of the acute abdom-inal complications, implying that there are manycauses of AA in addition to vasculitis. We alsofound a variety of etiologies in our SLE patientswith AA. All of the inactive patients had a non–SLE-related abdominal event and, similar to the ICactive SLE subgroup, appendicular and vesicularpathology, as well as primary intraabdominal in-fections, were responsible for AA.

In our study primary abscesses were the secondmost frequent cause of AA, and all occurred in pa-tients on immunosuppressive therapy. Musher et al30

recommended paracentesis and peritoneal lavage todifferentiate perforation from SLE peritonitis. How-ever, a negative result does not rule out an intraab-dominal complication. In our study there were threenegative paracentesis, two of them in patients with

significant abdominal pathology (intestinal vasculi-tis, retroperitoneal abscess). Peritoneal inflamma-tion has been described in autopsy studies of SLE,and ascites was found in 8% to 11% of patients.2,24

Clinically, ascites may express as abdominal pain,even inducing signs of AA,31 as in 1 case of this se-ries, or it may be painless.32 Sterile ascites was foundin 14% of our patients with positive ANA and lowcomplement, as previously reported.31

Abdominal vasculitis and thrombosis have beenassociated with SLE activity; however, a validatedmeasure of overall disease activity has not been em-ployed. In our study a SLEDAI score below 5 was astrong indicator that the abdominal complicationwas not due to SLE. We expect that, with the iden-tification of these risk factors, the complications andmortality related to AA in SLE patients may be an-ticipated.

Pollack et al emphasized in 195833 that the traumaof exploratory laparotomy frequently leads to furtherclinical deterioration. More recently, it has been re-ported that SLE patients do not tolerate surgery welland have a propensity to form adhesions.34 In con-trast, our study and other recent reports8–10,15,35 rec-ommend early surgery. Indeed, a delay in surgicalexploration negatively influences prognosis.

In conclusion, our study indicates that AA has adifferent significance in active and inactive SLE pa-tients, and suggests that a systematic measurementof activity and early laparotomy may improve theprognosis. In patients with AA and active SLE, theprobabilities of having vasculitis or non–SLE-relatedcauses are even. High titers of aCL should suggestintraabdominal thrombosis, and portend high mor-tality. Immunosuppressive therapy is associatedwith intraabdominal infections and may mask the ab-dominal symptomatology, and thus be another factorto delay laparotomy. We recommend early surgeryin all SLE patients with AA.

ACKNOWLEDGMENTSThe authors gratefully acknowledge the helpful suggestions of Dr. Juan Canosofor critical review of the manuscript.

REFERENCES1. Hoffman BI, Katz WA. The gastrointestinal manifestations of systemic lupus ery-thematosus: a review of the literature. Semin Arthritis Rheum. 1980;9:237–247.2. Dubois EL, Tuffanelli DL. Clinical manifestations of systemic lupus erythe-matosus. Computer analysis of 520 cases. JAMA. 1964;190:104–111.3. O’Neill PB: Gastrointestinal abnormalities in the collagen diseases. Am J DigDis. 1961;6:1069–1083.4. Harvey AM, Shulman LE, Tumulty PA. Systemic lupus erythematosus: review ofthe literature and clinical analysis of 128 cases. Medicine. 1954;33:291–437.5. Shapeero LG, Myers A, Oberkircher PE, Miller WT. Acute reversible vasculitisof the gastrointestinal tract. Radiology. 1974;112:569–574.6. Couris GD, Block MA, Rupe CE. Gastrointestinal complications of collagendiseases: surgical implications. Arch Surg. 1964;89:695–700.7. Zizic TM, Shulman LE, Stevens MB. Colonic perforations in systemic lupuserithematosus. Medicine. 1975;54:411–422.




8. Zizic TM, Classen JN, Stevens MB. Acute abdominal complications of systemiclupus erythematosus and polyarteritis nodosa. Am J Med. 1982;73:525–531.9. Bombardier C, Gladman DD, Urowitz MB, Caron D, Chang CH, and the Com-mittee on Prognosis studies in SLE: Derivation of the SLEDAI. A disease activityindex of lupus patients. Arthritis Rheum. 1992;35:630–640.10. Rothfield NF. Systemic lupus erythematosus. Clinical aspects and treat-ment. In: McCarty DJ, Koopman WJ, eds. Arthritis and Allied Conditions, 12thed. Philadelphia, PA: Lea & Febiger, 1993:1169–1170.11. Flanigan RC, McDougal WS, Griffen WO. Abdominal complications of col-lagen vascular disease. Am Surg. 1983;49:241–244.12. Tan EM, Cohen AS, Fries JF, et al. Special article. The 1982 revised criteriafor the classification of systemic lupus erythematosus. Arthritis Rheum. 1982;25:1271–1277.13. Alarcon-Segovia D, Deleze M, Oria CV, et al. Antiphospholipid syndrome insystemic lupus erythematosus. a prospective analysis of 500 consecutive pa-tients. Medicine. 1989;68:353–365.14. Mayer MM. Complement and complement fixation. In: Kabat EA, Mayer MM,eds. Experimental Immunochemistry, 2nd ed. Springfield, Illinois: Charles C.Thomas, 1971:133–137.15. Owens WD, Felts JA, Spitnagel EL. American Society of Anesthesiologyphysical status classification: a study of consistency of ratings. Anesthesiology.1978;49:239–243.16. Osler W. On the visceral complications of ‘‘erythema exudativum multi-forme’’. Am J Med Sci. 1895;110:629–646.17. Brown CH, Shirley EK, Haserick JR. Gastrointestinal manifestations of sys-temic lupus erythematosus. Gastroenterology. 1956;31:649–666.18. Mayer LF. Gastrointestinal manifestations of systemic lupus erythemato-sus. In: Lahita RG, ed. Systemic Lupus Erythematosus, 1st ed. Churchill Living-stone Inc, 1987:747–760.19. Wallace DJ. Gastrointestinal manifestations and related liver and biliarydisorders. In: DJ Wallace, BH Hahn, eds. Dubois’ Lupus Erythematosus, 4th ed.Philadelphia, PA: Lea & Febiger, 1993:410–417.20. Stoddard CJ, Simms JM, Kennedy A, Hughes P. Acute abdominal compli-cations in systemic lupus erythematosus. Br J Surg. 1978;65:625–628.21. McNeil HP, Chesterman CN, Krilis SA. Immunology and clinical importanceof antiphospholipid antibodies. Adv Immunol. 1991;49:193–280.

22. Asherson RA, Morgan SH, Harris EN, Gharavi AE, Krauz T, Hughes GR.Arterial occlusion causing large bowel infarction—a reflection of clotting dia-thesis in systemic lupus erythematosus? Clin Rheumatol. 1986;5:102–106.23. Weissman SG, Thoyoparan RS, Goodwin TA, Lockshin MD. Occlusive vas-culopathy in systemic lupus erythematosus: association with anticardiolipin an-tibody. Arch Intern Med. 1991;151:389–392.24. Matolo NM, Albo D Jr. Gastrointestinal manifestations of collagen diseases:surgical implications. Am J Surg. 1971;122:678–682.25. Reynolds JC, Inman RD, Kimberly RP, Chuong JH, Kovacs JE, Walsh MB.Acute pancreatitis in systemic lupus erythematosus: report of 20 cases and areview of the literature. Medicine. 1982;61:25–32.26. Lang TJ. Gastrointestinal vasculitis and Pneumatosis Intestinalis due tosystemic lupus erythematosus: successful treatment with pulse intravenous cy-clophosphamide. Am J Med. 1988;85:555–558.27. Rosner S, Ginzler E, Diamond H, et al. A multicenter study of outcome insystemic lupus erythematosus. II. Causes of death. Arthritis Rheum. 1982;25:512–517.28. Dujoune CA, Azarnoff DL. Clinical complications of corticosteroid therapy.Med Clin North Am. 1973;57:1331.29. Hellmann DB, Petri M, Whiting-O’Keefe Q. Fatal infections in systemic lupuserythematosus. The role of opportunistic organisms. Medicine. 1987;66:341–348.30. Musher DR. Systemic lupus erythematosus. A cause of ‘‘medical peritoni-tis’’. Am J Surg. 1972;124:368–372.31. Bitran J, McShane D, Elman MH. Ascites as the major manifestation ofsystemic lupus erythematosus. Arthritis Rheum. 1976;19:782.32. Jones PE, Rawcliffe P, White N. Painless ascites in systemic lupus ery-thematosus. Br Med J. 1977;1:1513.33. Pollack VE, Grove WJ, Kark RM, Muehrcke RC, Pirani CL, Steck IE. Systemiclupus erythematosus simulating acute surgical condition of the abdomen. NEJM.1958;259:258–266.34. Gore RM, Marn CS, Ujiki GT, Craig RM, Merquardt J. Ischemic colitis as-sociated with systemic lupus erythematosus. Dis Colon & Rectum. 1983;26:449–451.35. Jovaisas A, Kragg G. Acute gastrointestinal manifestations of systemiclupus erythematosus. Can J Surg. 1987;30:185–188.

Documents

Acute Abdomen in Systemic Lupus Erythematosus: The Importance of Early Laparotomy