Chapter 33 Gastrointestinal and Hepatic Manifestations

Gastrointestinal Involvement

Gastrointestinal (GI) manifestations are common in patients with systemic lupus erythematosus (SLE), and the prevalence of their various manifestations is listed in Box 33-1. Abdominal symptoms and signs may be the result of SLE, medications that are used to treat SLE, or intercurrent processes. Historically, William Osler was impressed with the frequency of GI crises in those with lupus and labeled lupus as the new disease that could mimic any other disease.¹

Box 33-1 Key Points

Systemic Lupus Erythematosus and the Gastrointestinal Tract

1. Gastrointestinal (GI) symptoms are common in systemic lupus erythematosus (SLE). Secondary causes, such as concurrent disease, stress, and medication, must be ruled out.

2. Sore throat and oral ulcers are common.

3. Dysphagia is present in 1% to 7.3% of patients, especially in association with Raynaud phenomenon.

4. Anorexia, nausea, vomiting, or diarrhea may be prominent in one third of patients when the disease is active. Chronic intestinal pseudoobstruction (CIPO) causes these symptoms and is a disturbance of the enteric nervous system. Inflammatory bowel disease, infection, and concomitant drug administration must be ruled out as other causes.

5. Peptic ulcer disease is found in 6% of patients with acute abdominal pain and is usually caused by antiinflammatory medication.

6. Ascites is found in 8% to 12% of patients. If a result of nephrosis, cirrhosis, or congestive heart failure is present, it is a painless transudate. Exudative causes might be painful and include serosal inflammation. Patients with lupus peritonitis are often responsive to steroids.

7. Pancreatitis is a serious complication of SLE. Pancreatitis is associated with pancreatic vasculitis, activity of SLE in other systems, and, rarely, with subcutaneous fat necrosis. Mild elevation of pancreatic enzyme levels may occur in SLE without pancreatitis; high levels suggest pancreatitis. Steroids are the treatment of choice, but steroids (along with thiazide diuretics, and azathioprine) can induce pancreatitis.

8. Abdominal pain, distention, and tenderness warrant a search for ischemia or bowel ulceration, especially in patients with a SLEDAI score of 4 or higher. In the outpatient setting, abdominal pain, distention, and tenderness may suggest the presence of small intestinal bacterial overgrowth (SIBO).

9. Malabsorption syndromes are rare but do occur.

10. Mesenteric or intestinal vasculitis is a life-threatening complication of SLE, usually associated with multisystem activity. High doses of steroids are required. To reduce mortality, early surgical intervention is indicated if prompt improvement does not occur with steroids. Patients may die from complications of obstruction, perforation, or infarction if surgical exploration is not performed within 48 hours in appropriate patients.

SLEDAI, Systemic Lupus Erythematosus Disease Activity Index.

Prevalence

GI complaints were the initial presentation in 10% of Dubois’ patients ²; 25% to 40% had protracted symptoms. Haserick and colleagues³ divided the GI symptoms of SLE among 87 patients into three groups: none (63%), minor (29%), and major (8%). Subclinical involvement of the GI tract is also common; chronic mucosal infiltration with inflammatory cells was found in 96% of 26 autopsied children with SLE.⁴ Younger patients with lupus are more susceptible. Among 272 patients with SLE, the prevalence of GI manifestations ranged from 10% in children to none in patients over the age of 50 years.⁵ A review on GI manifestations in SLE published by Sultan and colleagues⁶ found anorexia to be the most common reported manifestation with a prevalence of 36% to 71% in published studies. A recent publication found the cumulative prevalence of GI symptoms in patients with lupus who are of Asian descent to be 3.8% to 18%.⁷ In one recent study, patients with lupus who required hospitalization had a higher prevalence of GI symptoms with 39 out of 177 patients being admitted for a GI complaint.⁸

Pharyngitis, Dysphagia, and Esophagitis

Recurrent sore throat is not an infrequent finding, especially in children ⁹ (see Chapter 23 for discussions on mucous membrane lesions and other features of oral pathologic conditions). Dysphagia and heartburn occur in 1% to 7.3%^2,^9,¹⁰ and 11% to 50%⁹ of patients, respectively. Chong and colleagues¹¹ found that patients with lupus in comparison with patients with rheumatoid arthritis (RA) had significantly more heartburn. In a literature review, Zizic¹² related that although only 5% of patients with SLE complained of dysphagia, 25% had impaired esophageal peristalsis, compared with 67% of patients with scleroderma. Several studies using esophageal manometry noted aperistalsis or hypoperistalsis of the esophagus in approximately 10% of patients with SLE.¹³^–¹⁵ Aperistalsis is sometimes correlated with the presence of Raynaud phenomenon. Gutierrez and colleagues¹⁶ compared esophageal motility in 14 patients with SLE and 17 patients with mixed connective-tissue disease (MCTD). A definite correlation was found between Raynaud phenomenon and hypoperistalsis, with the latter being more common in MCTD. The patients in the SLE group had only a slightly decreased lower esophageal sphincter pressure. Esophageal motor dysfunction in SLE can also produce diffuse spasm and result in symptoms of chest pain.

Ramirez-Mata and colleagues ¹⁷ performed esophageal manometric studies in a group of unselected patients with SLE and noted abnormalities in 16 patients. An absence of or abnormally low contractions were found at the upper one third in seven patients, at the lower two thirds in three patients, in the entire esophagus in two patients, at the lower esophageal sphincter in two patients, and at the lower two thirds plus the lower sphincter in the remaining two patients. No relationship was found among the presence of esophageal dysfunction and activity, duration, and therapy of SLE. Interestingly, five of the 34 patients who had normal studies complained of dysphagia and heartburn. Esophageal imaging with Gastrografin, computed tomographic (CT) scanning, or endoscopy is required to make the diagnosis of esophageal ulceration or perforation from systemic vasculitis.¹⁰

The treatment of esophageal symptoms include small and frequent meals, the avoidance of postprandial recumbency, and the administration of antacids, proton-pump inhibitors, histamine-2 (H₂) antagonists, or parasympathomimetic agents. Fungal esophagitis from the use of antibiotics and steroids may need treatment with fluconazole.

Anorexia, Nausea, Vomiting, and Diarrhea

The most common cause of anorexia, nausea, vomiting, and diarrhea in patients with SLE is related to the use of salicylates, nonsteroidal antiinflammatory drugs (NSAIDs), antimalarial drugs, corticosteroids, and cytotoxic agents. Anorexia, nausea, vomiting, and diarrhea symptoms can continue to occur for weeks after therapy is stopped. When caused by the disease, manifestations are persistent and are not explained by other factors.

Anorexia occurs in 49% to 82% of patients,^2,⁹ especially if untreated. Nausea has been reported in 11% to 38% of patients.^2,⁹ When medications are excluded as a cause, however, the incidence is approximately 8%.¹⁸ Vomiting and diarrhea can be prominent in patients who are hospitalized with lupus and GI symptoms^2,^9,¹⁸ and are observed in up to 56.4% and 30.8% of patients, respectively, who are admitted. Children appear to have an increased incidence of all these symptoms.⁵

Motility Disorders

Chronic intestinal pseudoobstruction (CIPO) reflects a dysfunction of the visceral smooth muscle or the enteric nervous system.^19,²⁰ Symptoms and signs of CIPO in patients with SLE include a subacute onset of abdominal pain and distention associated with vomiting and constipation and a distended tender abdomen with hypoactive or absent bowel sounds or a complete lack of bowel sounds. Radiologic examination reveals dilated, fluid-filled bowel loops and occasional bilateral ureteral dilation with a reduced bladder capacity. Antroduodenal manometry demonstrates intestinal and esophageal hypomotility. Nojima and colleagues²⁰ described two patients with CIPO who had antibodies to proliferating cell nuclear antigen (PCNA) but no other specific antibodies or clinical manifestations of SLE. Treatment of CIPO usually involves high doses of steroids, broad-spectrum antibiotics, and promotility drugs. Perlemuter and colleagues²¹ reported the use of octreotide at a dose of 50 µg twice a day subcutaneously in CIPO in SLE and scleroderma. The symptoms of CIPO resolved in the three patients receiving treatment within 48 hours. Recurrence of symptoms responded to increasing the dose of octreotide.

Small intestinal bacterial overgrowth (SIBO) may be the result of disordered motility caused by the lack of duration or propagation of migrating motor complex or from the lack of IgG class antibacterial antibodies to indigenous bacteria in the GI tract to neutralize bacteria.^22,²³ Albano and colleagues²⁴ investigated the presence of SIBO in 14 patients with SLE using a lactulose hydrogen breath test. Symptoms of SIBO such as bloating (50%), diarrhea (64%), constipation (42%), and abdominal pain (42%) were present in these patients with SLE without any clear identifiable cause after the history and physical examination were completed. Breath hydrogen above 20 million parts per million (ppm) with two distinct peaks of hydrogen production was diagnostic for SIBO. Twelve patients (86%) were found to have SIBO by predefined criteria.

Abdominal Pain and Acute Abdomen

Abdominal pain is found in 8% to 37% of patients with SLE,^2,^7,^9,^18,²⁵ with the lowest incidence being reported in series that excluded medication-related symptoms. Abdominal pain may be the first symptom of a catastrophic lupus-related complication such as mesenteric vasculitis or may be a non-lupus–related cause such as gluten-sensitive enteropathy or irritable bowel syndrome. A thorough history and physical examination with special attention paid to the activity of lupus in other organs and appropriate imaging can help differentiate between lupus- and non-lupus–related complications. Patients with abdominal pain, even without tenderness, need an aggressive and comprehensive evaluation, including a complete blood count, amylase-level determination, blood-chemistry profiles, and abdominal radiography. If free air, a moderate amount of free fluid, acidosis, and/or hyperamylasemia without pancreatitis are present, then diagnostic laparoscopy should be performed. If pseudoobstruction and/or thumbprinting of the bowel are seen without free peritoneal fluid, then specialized tests, such as an upper GI series, barium enema, CT, magnetic resonance imaging (MRI), gallium and indium white-cell scanning, and visceral angiography, may be necessary.

Intravenous fluids should be administered to patients suspected of having an intraabdominal crisis while undergoing these initial diagnostic evaluations. If peritonitis is suspected, broad-spectrum antibiotics should be administered. Aggressive fluid replacement, antibiotics, and steroid stress dose coverage precede laparoscopic or open surgical exploration. Steroid therapy can mask bowel ischemia and perforation. The best application of diagnostic laparoscopy is in the evaluation of a patient with equivocal findings. In 412 consecutive admissions to Cleveland hospitals for collagen vascular diseases,²⁵ 63 patients had abdominal complaints; of these, 48 had SLE. Pain was present in 85% of patients, fever was noted on examination in 76% of patients, and peritoneal signs were recorded in 10% of patients. Corticosteroids were administered to 64%. Acute causes, including duodenal or gastric ulcer, gastritis, and pancreatitis, were determined in 33 patients. Mesenteric vasculitis was present in 3 patients, and the pain was of undetermined cause in 16 patients. Surgery was performed on 21 patients; in 11, it was exploratory. Al-Hakeem and colleagues²⁶ identified 13 patients with a principal diagnosis of abdominal pain out of 88 patients with SLE who were admitted to the hospital during a 15-year period. Diagnoses accounting for abdominal pain included adhesions,³ diverticulitis,³ cholecystitis,² perforated ulcer and colon, gastroenteritis, duodenitis, and inflammatory bowel disease (1 each). Of the 13 patients in the study, 9 required surgery. In another survey of 63 procedures,²⁷ 16% morbidity and 6% surgical mortality rates were recorded. Rojas-Serrano²⁸ evaluated the causes of emergency department consultation for patients with SLE and found that abdominal pain was the third most-frequent reason accounting for 18 out of 180 patients with lupus visiting the emergency department.

Min and colleagues ²⁹ studied the causes of acute abdominal pain in patients visiting the emergency department. Twenty-six patients with SLE and abdominal pain made 44 visits to the emergency department. Twenty-seven (59.1%) of these visits were for ischemic bowel disease. Other diagnoses included pancreatitis, serositis, splenic infarction, angioedema, renal vein thrombosis, pelvic inflammatory disease, upper GI bleeding, and ectopic pregnancy. CT scanning and ultrasound help establish the diagnosis of ischemic bowel disease. Kwok and colleagues³⁰ studied 706 lupus admissions to the hospital in South Korea from 1990 to 2006; 87 (12.3%) of these patients were admitted for abdominal pain as the main complaint, 41 (47.1%) out of these 87 patients had lupus enteritis as the cause of their abdominal pain. The Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) score and the titers of antiendothelial cell antibodies of patients with enteritis were higher than those admitted for abdominal pain without enteritis.

Peptic Ulcer Disease

The incidence of peptic ulcers in patients with SLE has been reported as being between 4% and 21%,^30,³¹ but these studies antedate the present era of endoscopy and gastroprotective therapy. In a more recent report, perforated ulcers have been found in 3 (5.8%) out of 55 patients with SLE and acute abdomen.³² Luo and Chang³³ found that NSAIDs and aspirin were more predictive of developing peptic ulcers compared with Helicobacter pylori seropositivity or steroid use in a cohort of 65 patients with lupus who had endoscopies and biopsies performed before and after undergoing pulse steroid therapy.

Helicobacter pylori Infection in Systemic Lupus Erythematosus

Sawalha and colleagues ³⁴ studied the prevalence of seropositivity against H. pylori and four other control antigens in 466 patients with SLE and compared them with 466 patients in a control group matched for age (+/− 3 years), sex, and ethnicity. Most of the patients in the control group were from the same pedigree multiplex for SLE. The frequency of seropositivity to H. pylori was only lower in the SLE cohort, compared with the control group, and this difference could be explained by the lower prevalence in patients who were African American (38.1 versus 60.2; OR = 0.4, P = 0.0009, 95% confidence interval [CI] 0.24-0.69). The mean age of onset of illness in the H. pylori group was significantly older (34.4 versus 28, P = 0.039), compared with the patients with H. pylori seronegative with SLE. Direct biopsies of the gastric antrum were not performed in this study.

Junca and colleagues ³⁵ investigated the prevalence of intrinsic factor and pernicious anemia in 30 patients with SLE and 45 patients in a control group. Pernicious anemia was characterized by the presence of low serum cobalamin concentration and macrocytic anemia; the presence of intrinsic factor antibody was found in only one patient (3.3%), although 23% of patients had low cobalamin levels and 10% patients had intrinsic factor antibody out of the 30 patients with SLE.

Inflammatory Bowel Disease

Ulcerative Colitis

Ulcerative colitis and lupus may occur concurrently in a small number of patients with lupus. Two patients each were found in the large case series of over 400 patients with lupus. Dubois and Wallace ^2,¹⁰ and Kurlander and Kirsner³⁶ elegantly documented the clinicopathologic correlations and remarked that lupus colitis and ulcerative colitis can be indistinguishable. In 1965, Alarcón-Segovia and Cardiel³⁷ reviewed the literature extensively concomitant SLE and ulcerative colitis, adding additional patients in detail from their Mayo Clinic experience. Additionally, 100 patients with ulcerative colitis were evaluated for SLE, which was found in 3 patients. Folwaczny and colleagues³⁸ found an increased prevalence of positive antinuclear antibody (ANA) in patients with Crohn disease and ulcerative colitis, compared with first-degree relatives and normal control participants. Eighteen percent of patients with Crohn disease and 43% of patients with ulcerative colitis had a positive ANA test, whereas 13% of relatives of patients with Crohn disease and 24% of relatives of patients with ulcerative colitis had a positive test. Of the healthy patients in the control group, 2% had a positive test. Both sulfasalazine and olsalazine have been associated with the development of drug-induced ANA and SLE.

Regional Ileitis

Concurrence of SLE and regional ileitis (i.e., Crohn disease) is surprisingly rare and has been reported in about ten patients.^35,³⁹

Collagenous Colitis

Collagenous colitis is a distinct disorder that is characterized by colonic lymphocytic infiltration of the surface epithelium. Heckerling and colleagues ⁴⁰ reported that patients with collagenous colitis have watery diarrhea but a normal endoscopic appearance and radiographic findings. Collagenous colitis rarely overlaps with lupus and may be treated with corticosteroids instead of sulfasalazine when it coexists with lupus.

Celiac Disease in Association with Systemic Lupus Erythematosus

Gluten-sensitive enteropathy is one of the most common autoimmune conditions with a prevalence of 1 in 100 patients. An overlap of this condition with SLE has not been frequently cited. Out of the 246 biopsy-proven patients with celiac disease, 6 fulfilled the American College of Rheumatology (ACR) criteria for lupus. The diagnosis of lupus was made before the diagnosis of the celiac disease in all the patients, and the symptoms of lupus did not improve with the dietary elimination of gluten.⁴¹ Mader and colleagues⁴² screened 61 patients fulfilling the ACR criteria for SLE for the presence of antiendomysial antibodies (AEAs) and antigliadin antibodies (AGLAs) and compared the prevalence of seropositivity with 35 healthy controls. None of the patients with lupus or the controls had AEA; however, 27 (44.3%) of the patients with lupus and 6 (17.1%) of the patients in the control group had AGLAs. A positive correlation was found between the presence of AGLAs and arthritis in lupus. The authors concluded that the presence of AGLAs is an epiphenomenon, although small intestinal biopsies were not performed to rule out celiac disease. Mader and colleagues recommend AEA as the preferred screening test for celiac disease in association with SLE.

Protein-Losing Enteropathy and Malabsorption

The presence of severe diarrhea and significant hypoalbuminemia (reported to be as low as 0.8 g/dL) without proteinuria should raise the suspicion of protein-losing enteropathy. A case series of 15 patients with lupus admitted to Peking Union Medical College Hospital from November 2001 to April 2006 showed that the mean age was 40.1 ± 15.4 years (range from 19 to 71 years) with a female-to-male ratio of 4 : 1.⁴³ Although all 15 patients had various degrees of peripheral edema, the number of patients with ascites, pleural, and pericardial effusions were 73%, 60%, and 47%, respectively. Fifty-three percent had protein-losing enteropathy as the initial manifestation of lupus and only 40% of the cohort had symptoms of abdominal pain and diarrhea. Patients had significant hypoalbuminemia (100%), hypocomplementemia (80%), hyperlipoproteinemia (67%), and hypocalcemia (40%), but proteinuria and positive double-stranded DNA (dsDNA) antibody status was low. Endoscopy and biopsy showed bowel mucosal edema and chronic inflammation, and technetium-99m (Tc-99) albumin scintigraphy was the diagnostic test used for all the patients. The radiolabeled albumin excretion in the gut resolved in 60% of the patients when they were treated with corticosteroids or immunosuppressive agents. Kim and colleagues⁴⁴ showed that hyperlipoproteinemia helps differentiate protein-losing enteropathy in lupus from lupus enteritis and that the level of serum albumin is significantly reduced in lupus-related versus idiopathic protein-losing enteropathy. Hizawa and colleagues⁴⁵ described the radiologic findings in protein-losing enteropathy. Lupus enteritis was associated with irregular spiculation and thickening, as well as thumbprinting, which are suggestive of ischemia on double-contrast radiography of the small intestine. Protein-losing enteropathy, by contrast, had thickened folds with nodules that, at biopsy, were shown to be lymphangiectasia. Increased fecal excretion of intravenous radiolabeled albumin is the best quantitative study for following disease activity, although one report⁴⁶ has suggested that alpha-1 antitrypsin (α₁-antitrypsin) clearance also can monitor response to therapy.

Response to corticosteroids is nearly universal, but resistant patients may require cyclosporine or intravenous pulse cyclophosphamide. Octreotide, a somatostatin antagonist, is useful in treating protein-losing enteropathy by decreasing intestinal blood flow and by modulating activated inflammatory cells by binding to the surface of the somatostatin receptor. Medium-chain triglycerides are also useful, first because they are carried through the portal system, thereby decreasing lymphatic flow, and second by virtue of being absorbed via the large bowel, thus correcting lipid deficits in this condition. Some patients also may require a gluten-free diet.⁴⁷

Mader and colleagues ⁴⁸ investigated a cohort of 21 patients with SLE for malabsorption with a screening d-xylose absorption test, examination of the stool for fat droplets, and a histologic examination of a specimen of the duodenum obtained during endoscopy. Two patients (9.5%) had evidence of malabsorption manifested by an abnormal d-xylose absorption and excessive fecal fat excretion. Two other patients showed excessive fecal fat excretion. One of the patients with malabsorption had abnormal small bowel histologic findings of flattened villi and an inflammatory infiltrate. No excessive deposition of immunoglobulins was revealed in the mucosa on immunoperoxidase staining. The cause of the malabsorption remains uncertain.

Ascites and Peritonitis

Ascites can be the initial presentation of SLE. Ascites occurs in 8% to 12% of adult patients with lupus, often as a manifestation of the nephrotic syndrome, and in 36% of children with SLE.⁴⁹^–⁵¹ In an excellent review of ascites in SLE, Schousboe and colleagues⁵² classified ascites as either acute or chronic. Acute causes include lupus peritonitis, infarction, perforated viscus, pancreatitis, mesenteric vasculitis, and hemorrhagic and bacterial peritonitis. Chronic causes of ascites include lupus peritonitis, congestive heart failure, pericarditis, nephrotic syndrome, Budd-Chiari syndrome, protein-losing enteropathy, underlying malignancy, cirrhosis, and tuberculosis. Ascitic fluid can be inflammatory or noninflammatory. Jones and colleagues⁵³ reported that noninflammatory lesions are always painless and associated with transudative fluid, and most patients have nephrotic syndrome. Peritonitis is usually inflammatory, painful, and exudative. Schocket⁵⁴ showed that peritoneal tissue can contain immune complex deposits and inflammatory infiltrates. ANA, anti-DNA, and low complement levels can be present in peritoneal fluid. Low and colleagues⁵⁵ reported the characteristics of lupus serositis on barium radiographic examination and CT imaging. The small-bowel barium series showed segments of spiculation with tethering, angulation, and obstruction. CT scanning demonstrated ascites and asymmetric thickening of the small bowel wall.

Ascites caused by lupus peritonitis is usually responsive to steroids. Other causes may require additional interventions including azathioprine.⁵⁶ Gentle dieresis and paracentesis are important adjunctive measures that often provide symptomatic relief, provided that renal function is not impaired by this approach.

Pancreatitis

Prevalence

Pancreatitis can be the initial manifestation of SLE and may also be caused by non-lupus–related causes. The Johns Hopkins lupus cohort reported that 72 of their 1842 patients with lupus had a diagnosis of acute pancreatitis.⁵⁷ Campos and colleagues⁵⁸ found acute pancreatitis in 11 out of 263 (4.2%) pediatric patients with lupus seen in their clinic in São Paolo over a period of 26 years.

Clinical Presentation and Etiopathogenesis

Abdominal pain is present in over 80% of patients with pancreatitis associated with SLE and is often accompanied by nausea, vomiting, and fever, although the pain frequently does not radiate to the back.⁵⁹ The diagnosis is usually established using amylase and lipase measurements, as well as imaging using ultrasound or CT scanning.⁶⁰ The cause was found to be related to lupus in the majority of patients; however, other non-lupus–related causes, such as biliary tract disease, hypertriglyceridemia, alcohol, and drug exposure, each play a role in approximately one half of the patients with lupus-associated pancreatitis.

Corticosteroids are not considered to play a role in causing pancreatitis in patients with SLE. Hernandez-Cruz and colleagues ⁶¹ reviewed a database of patients with SLE and found 18 patients with 26 episodes of pancreatitis with an average SLEDAI score of 6.5 at the time of the acute pancreatic episodes. Out of the 26 episodes, 11 were severe and 4 patients died—3 of pulmonary hemorrhage and 1 from septicemia. The most common cause was thought to be medication use (8 episodes), and hypertriglyceridemia, alcohol, and cholelithiasis were thought to be the cause in 4, 2, and 2 patients, respectively. Pascual examined a database of patients with SLE from July 1984 to July 2001 and found 49 separate acute pancreatitis episodes in 35 patients with lupus, giving a prevalence of 3.5%.⁶² Of the 49 episodes, 14 (28.5%) patients were considered to have biliary disease. Alcohol, increased triglycerides, or uremia were considered to be the cause in 10 (20.4%) patients. The remaining 17 (34.7%) patients were considered to be idiopathic or SLE was considered the cause. Steroids and azathioprine did not cause a relapse of the symptoms when patients were challenged with these medications. The Mexican Systemic Lupus Erythematosus Disease Activity Index (Mex-SLEDAI) scores were significantly higher in the idiopathic group (median of 9 [3 to 19] in the idiopathic group versus 5 [0 to 23] in the toxic metabolic group, and 3 [0 to 18] in the biliary group). A case control analysis with a controls-to-cases ratio of 4 : 1 showed an increased prevalence of acute pancreatitis in patients with SLE (46% versus 14%). The odds ratio for the severity of pancreatitis and mortality was significantly higher in pancreatitis associated with SLE than in non-SLE controls (odds ratio [OR] 8.6 versus 7.5). Similar conclusions were drawn by Derk and DeHoratius⁶³ in their review of all hospital admissions for patients with lupus to Thomas Jefferson University Hospital between 1982 and 2002. Of the 2947 hospitalized patients with SLE, 25 (0.85%) had acute pancreatitis. The majority (76%) of patients had active SLE at the time of admission with an average involvement of 4.4 organs. Of the 25 patients with acute pancreatitis, 18 had an increase in their corticosteroid doses with improvement in their clinical and laboratory parameters. In conclusion, pancreatitis is more often a severe and often fatal manifestation of lupus than an illness from other factors such as biliary disease or from medications including corticosteroids or azathioprine.

Pancreatitis in childhood-onset SLE appeared to coincide with the development of macrophage activation syndrome in 10 out of 11 pediatric patients with lupus.⁶⁴ However, the diagnosis was confirmed by bone marrow aspiration in only 3 out of these 10 patients. Yeh and colleagues⁶⁵ reported that pancreatitis may occur as a result of thrombi in pancreatic arteries, because of antiphospholipid antibodies.

Mild elevations of serum amylase levels may be noted in patients with SLE in the absence of pancreatitis. Hasselbacher and colleagues ⁶⁶ studied 25 patients with SLE but without pancreatitis and 15 patients in a non-SLE control group. Amylase levels were elevated in 5 patients, and 6 patients had macroamylasemia, compared with none in the control group. The mean amylase level in the SLE group was 161.7 mg/dL, compared with 116.4 mg/dL in the control group; this difference was statistically significant. Macroamylasemia results from decreased renal clearance of an immunoglobulin-amylase complex. The presence of a pathogenic autoantibody to amylase was proposed. A correlation is present between active SLE and elevated amylase levels without abdominal pain.

Only gold members can continue reading. Log In or Register to continue