Protein-losing enteropathy (PLE) is a complication in some systemic lupus erythematosus (SLE) patients that is often misdiagnosed. With this study, we provide insight into clinical characteristics, laboratory characteristics, diagnostic tests, risk factors, treatment, and prognosis of the disease.
A retrospective, case-control study was performed in 44 patients with SLE-related PLE (PLE group) and 88 patients with active SLE (control group) admitted to our care from January 2000−January 2012. Risk factors for SLE-related PLE were examined, and we analyzed the accuracy of single and combined laboratory characteristics in discriminating SLE-related PLE from active SLE. Serum albumin and C3 levels were measured as outcome during and after treatment with corticosteroids and immunosuppressive agents.
The PLE group had lower mean serum albumin and 24-hour urine protein levels, higher mean total plasma cholesterol levels, and greater frequencies of anti-SSA and SSB seropositivity compared with the control group. Anti-SSA seropositivity, hypoalbuminemia, and hypercholesterolemia were independent risk factors for SLE-related PLE. The simultaneous presence of serum albumin (<22 g/l) and 24-hour urine protein (<0.8 g/24 h) had high specificity, positive predictive value, negative predictive value, and positive likelihood ratio, a low negative likelihood ratio and no significant reduction in sensitivity. High dosage of glucocorticosteroid combined with cyclophosphomide were mostly prescribed for SLE-related PLE.
Citation: Chen Z, Li M-T, Xu D, Yang H, Li J, Zhao J-L, et al. (2014) Protein-Losing Enteropathy in Systemic Lupus Erythematosus: 12 Years Experience from a Chinese Academic Center. PLoS ONE 9(12): e114684. https://doi.org/10.1371/journal.pone.0114684
Editor: Masataka Kuwana, Nippon Medical School Graduate School of Medicine, Japan
Received: June 29, 2014; Accepted: November 12, 2014; Published: December 9, 2014
Copyright: © 2014 Chen et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the paper and its Supporting Information files.
Funding: This study was supported by the Chinese National Key Technology R&D Program, Ministry of Science and Technology (2008BAI59B02), the Chinese National High Technology Research and Development Program, the Ministry of Science and Technology (2012AA02A513), the Research Special Fund for Public Welfare Industry of Health, Ministry of Health (201202004), and the National Major Scientific and Technological Special Project for “Significant New Drugs Development” Ministry of Science and Technology (2012ZX09303006-002). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Protein-losing enteropathy (PLE) is a condition characterized by profound edema and severe hypoalbuminemia that is secondary to excessive serum protein loss from the gastrointestinal (GI) tract . Hypoalbuminemia is an important clinical symptom as it can lead to edema, fluid balance disorders, and heart failure. In addition, patients with hypoalbuminemia are vulnerable to infection. In systemic lupus erythematosus (SLE), hypoalbuminemia is usually ascribed to nephrotic syndrome, disease exacerbation, or liver disease . Most PLE-related GI manifestations are not as common as other organ involvement such as lupus nephritis, and PLE is clinically indistinguishable from nephrotic syndrome . Therefore, PLE can be easily missed by internists, rheumatologists, gastroenterologists, and nephrologists .
Most previous reports on PLE describe isolated cases or small case series of patients –. Because little information exists on SLE-related PLE, we undertook a case-control study to: (1) describe the clinical and laboratory characteristics of this complication in patients with SLE; (2) explore risk factors for PLE in SLE patients and evaluate the discriminative ability of laboratory characteristics; and (3) report our experience on the treatment of SLE-related PLE with a combination corticosteroid and immunosuppressive cyclophosphamide therapy.
All patients were admitted to the Peking Union Medical College Hospital (PUMCH) between January 2000 and January 2012. A total of 44 patients were enrolled in the PLE group. Patients diagnosed with SLE according to the American College of Rheumatology (ACR) 1997 revised criteria were included in the study . PLE diagnoses were based on clinical symptoms and laboratory test results often encountered in patients with PLE, irrespective of Technetium 99 m-labelled (99 m Tc) human serum albumin (HSA) scintigraphy results. For this study, a clinical diagnosis of PLE included hypoalbuminemia that could not be fully explained by urinary protein loss, and reduced protein synthesis in which malnutrition or severe liver disease had been excluded. Patients that had evidence of protein leakage from the GI tract as detected by 99 m Tc-HSA scintigraphy were considered to have definite PLE, whereas patients in which this examination was not performed were considered to have probable PLE. The control patient group comprised 88 active SLE patients that were randomly and consecutively selected by SPSS software among the 4648 contemporaneous SLE patients (without PLE) at the PUMCH. Patients with concurrent Sjögren’s syndrome were identified according to American European Consensus Group (AECG) classification criteria . Written informed consent was obtained from all patients prior to admission to our hospital, and the study was approved by the PUMCH ethics committee.
Clinical and laboratory data
Demographic (gender, age, disease duration) and clinical data for all patients were collected according to Systemic Lupus International Collaborating Clinics (SLICC) 2012 criteria . Laboratory data included: serum albumin and C3 and C4 complement levels; plasma calcium, total cholesterol, and triglyceride levels; and 24-hour urine protein levels. Autoantibody profiles (anti-nuclear antibody [ANA], anti-double-stranded DNA antibody [anti-dsDNA], anti-ENA antibodies [including anti-SSA, anti-SSB, anti-Sm, and anti-RNP antibody]), and Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2K)  were obtained from all patients upon admission to hospital. ANA was detected by indirect immunofluorescence (IIF) with HEp-2 cell substrates. Anti-dsDNA was detected by IIF using flagellate protista substrates and enzyme-linked immunosorbent assay (ELISA). Anti-ENA was detected by immunodiffusion assay.
Treatment and assessment of clinical response
All patients in the PLE group were treated with corticosteroids. This treatment was administered intravenously. Initial intravenous steroid dosage was 1 mg/kg/d prednisone or the equivalent dosage of another corticosteroid. For methylprednisolone (MP) pulse therapy, 1000 mg was administered for three days and then changed to intravenous prednisone 1 mg/kg/d. Immunosuppressive therapy was administered as required. Immunosuppressive agents included cyclophosphamide (CTX), mycophenolate mofetil (MMF), and methotrexate (MTX).
Complete response (CR) to therapy for the PLE group was defined as total resolution of edema and GI symptoms, together with return of serum albumin levels to ≥35 g/dl. Partial response (PR) was defined as partial improvement of edema and GI symptoms with improved serum albumin levels (≥30 g/dl). Non-response (NR) was defined as the failure of serum albumin levels to increase to 30 g/dl or more .
SPSS software v.17 (Chicago, USA) was used for all statistical analyses. To analyze differences between groups, Student’s t test or Wilcoxon rank test were used for numerical data. The Chi square test or Fisher’s exact test was used for qualitative data. Variables with significant P values between groups in the univariate analysis were computed to a multivariate logistic regression model to predict independent risk factors for SLE-related PLE. The discriminative ability of laboratory tests was assessed using either the standard threshold, i.e., the threshold reported in the literature or judged to be clinically meaningful, or the best threshold, i.e., the threshold obtained through the receiver operating characteristic (ROC) curve analysis that produced the most appropriate tradeoff between sensitivity and specificity . The ability of single and combined in series laboratory characteristics to discriminate PLE from active SLE was evaluated by ROC curve analysis. Sensitivity, specificity, positive predictive values (PPV), negative predictive values (NPV), positive likelihood ratios (PLR), and negative likelihood ratios (NLR) were determined. Two-tailed P values <0.05 were considered statistically significant.
A total of 44 patients with SLE-related PLE were included in the PLE group. Twenty-six patients had probable PLE and 18 patients had definite PLE. Patients in this group were predominantly women (84.1%). The mean age was 35.1±2.13 years (range, 16–71 years). These data were comparable with the control group (Table 1). However, patients with PLE had greater frequencies of serositis and hypocomplementemia compared with patients in the control group. Upon admission to the hospital, patients in the PLE group had lower frequencies of acute cutaneous lupus, synovitis, and anti-dsDNA seropositivity compared with the control group. At diagnosis, renal and neurological symptoms, as well as the proportion of anti-Sm–positive patients were similar between groups. The mean SLEDAI-2K score of all patients within the PLE group was significantly lower than in the control group (8.0±0.75 vs. 10.4±0.59, P = 0.015). Eight (18.2%) patients in the PLE group and eight (9.09%) patients in the control group had concomitant Sjögren’s syndrome. The frequency of Sjögren’s syndrome was not different between the two groups (P = 0.131).
Thirty-nine (88.6%) patients in the PLE group had different degrees of peripheral pitting edema. Seven (15.9%) patients had nausea and vomiting; 15.9% and 50.0% of patients had abdominal pain and diarrhea, respectively. Thirty-nine (88.6%) patients had ascites, 33 (75.0%) had pleural effusion, and 22 (50.0%) had pericardial effusion. Upon determination of organ involvement, 12 (27.3%) patients in the PLE group had single GI involvement, and 32 patients (72.7%) had concomitant disease activity in other organs. Twenty-five PLE patients (56.8%) had concomitant lupus nephritis (LN). Of these, seven patients underwent renal biopsy. Histopathological analyses revealed two patients with Class II LN, one patient with Class IV LN, and four patients with Class V LN. Of the 25 concomitant LN patients, seven patients had 24-hour urine protein secretion >0.5 g, including one patient with minor pathologic changes in the renal biopsy, and six patients who demonstrated protein leakage from the GI tract by 99 m Tc-HSA scintigraphy, indicating that the serum albumin decline observed in these patients was mainly caused by loss from the GI tract.
Five (11.4%) patients’ conditions in the PLE group were complicated with neuropsychiatric syndromes of SLE (NPSLE), four had severe hematologic abnormalities, two had lung involvement, and three had angiogenesis with embolization. The distribution of SLEDAI-2K scores between patients in the PLE group was as follows: six patients scored >12, thirty-four patients scored between 4–12, and four patients scored between 1–3 (S1 Table in S1 File).
Mean serum albumin, C3 and C4, plasma calcium, and 24-hour urine protein levels were markedly lower in the PLE group compared with the control group. Mean platelet count and total plasma cholesterol levels were markedly higher in the PLE group compared with the control group. Twenty-nine (65.9%) patients with PLE were anti-SSA-seropositive and this percentage was higher than in the control group. Twenty five percent of patients in the PLE group were anti-SSB–seropositive, which was also higher than in the control group. The percentage of anti-RNP-positive patients and mean plasma triglyceride levels were comparable between groups (Table 2).
Probable PLE vs. definite PLE
The differences between definite PLE and probable PLE patients within the PLE group were examined. The demographic characteristics of patients with definite PLE and patients with probable PLE were comparable, as were mean values or frequencies of laboratory test results and the frequency of all SLICC criteria upon admission to the hospital. However, neurological symptoms were more common in patients with definite PLE. In addition, patients with definite PLE had borderline lower serum complement levels, yet higher SLEDAI-2K scores compared to patients with probable PLE (S1−S3 Tables in S1 File).
Risk factors for SLE patients with PLE
Variables with significant P values between groups in single factor analyses were computed using a multivariable logistic regression model to predict the independent risk factors for SLE-related PLE (Table 3). Anti-SSA seropositivity, hypoalbuminemia, and hypercholesterolemia were independent risk factors for SLE-related PLE, while hypoalbuminemia and hypercholesterolemia were negatively associated with SLE-related PLE. Therefore, the odds of an SLE patient with anti-SSA seropositivity developing PLE were 3.718 times higher than an SLE patient who was anti-SSA-negative. If the SLE patient was anti-SSB-positive, the odds of developing PLE were as high as 7.225, but this value was not statistically significant (P = 0.08).
Diagnostic test evaluation
As with the mean values for the laboratory parameters described above, the frequency of laboratory abnormalities was much greater for patients in the PLE group compared with the control group. Exceptions included comparable frequencies of higher plasma triglyceride levels and lower serum C4 levels (S4 Table in S1 File).
Sensitivity, specificity, and the area under the ROC curves for each laboratory characteristic, assessed using the standard threshold or the best threshold obtained through ROC curve analysis are shown in Table 4.
Plasma total cholesterol and triglyceride levels yielded similar sensitivity and specificity using the standard or the best threshold. Platelet count; serum albumin, C3, and C4; plasma calcium, and 24-hour urine protein levels yielded different sensitivity and specificity depending on whether the standard or best threshold was used. Overall, hypoalbuminemia had the best sensitivity and specificity, followed by decreased plasma calcium levels and 24-hour urine protein levels.
We sought the combination of laboratory variables in series that had the greatest diagnostic accuracy. The laboratory variables included were serum albumin, plasma calcium, and 24-hour urine protein levels. For each combination of variables, sensitivity, specificity, PPV, NPV, PLR, and NLR were calculated (Table 5). The best results were obtained using the simultaneous detection of serum albumin (<22 g/l) and 24-hour urine protein levels (<0.8 g/24 h), which yielded sensitivity 0.818, specificity 0.989, PPV 0.973, NPV 0.916, PLR 74.36, and NLR 0.184.
Endoscopy and imaging
Twenty-four patients (54.5%) in the PLE group underwent esophagogastroduodenoscopy (OGD). Of these, 23 (95.8%) had chronic superficial gastritis, generalized mucosal edema, and diffuse non-erosive erythematous GI mucosa. Approximately one-third of the PLE patients (13/44) underwent abdominal contrast-enhanced computed tomography (CT) scans. Circumferential bowel wall thickening due to submucosal edema was documented in 10 (76.9%) of 13 patients. A 99 m Tc-labeled HSA scan documented protein leakage in 18 (100%) of patients (S5 Table in S1 File).
Treatment and prognosis
All patients in the PLE group received intravenously administered corticosteroid therapy. The initial steroid dosage was 1 mg/kg/d of intravenously administered prednisone or the equivalent dosage of another corticosteroid. Fourteen patients (31.8%) had 1000 mg methylprednisolone (MP) pulse therapy for three days and then changed to intravenous prednisone (1 mg/kg/d). Forty (90.9%) of 44 patients received immunosuppressive therapy. Cyclophosphamide (CTX) was the most commonly administered immunosuppressive medication (37 cases), followed by mycophenolate mofetil (MMF, 1 case), and methotrexate (MTX, 1 case).
Eighteen cases were followed-up for longer than six months. After two months of treatment, eight (44.4%) of these patients had CR, four (22.2%) had PR and six (33.3%) had NR prognoses. Sixteen (88.9%) patients had CR after six months of treatment and the total response rate was 88.9% (S6 Table in S1 File).
We documented the changes in serum albumin and C3 levels in the PLE patients during treatment (Fig. 1). A significant elevation in serum albumin was noted in the first two months (P<0.001), which was accompanied by a significant elevation in serum C3 levels (P = 0.028). Serum albumin levels gradually increased, reaching maximal levels during the third month of treatment (P<0.001).
We documented the clinical presentation, diagnoses, treatment, and prognoses of 44 SLE-related PLE patients that were admitted into our care over a 12 year period. To our knowledge, this is the first case-control study of lupus-related PLE reported in the literature.
Although no specific epidemiological data exist for SLE-related PLE, it is estimated that the point prevalence of PLE is 3.2–7.5% of Chinese individuals with SLE in Hong Kong , . We previously reported that the prevalence of SLE-related PLE in a smaller case series was 1.9% . In this study, we extended the study period to 12 years and surveyed 4692 SLE patients that attended the lupus clinics at our hospital, thus decreasing the prevalence to 0.94%. This figure may be an underestimation as screening for protein leakage from the GI tract is not routinely performed unless there is a clinical indication for unexplained hypoalbuminemia .
Aoki et al. have previously concluded that lupus patients might initially present with PLE symptoms, and Gornisiewicz et al showed that PLE often occurs in patients with clinically severe SLE , . In the current study, close to 30% of the PLE patients had single GI involvement as the initial manifestation of SLE. This is a reminder that gastroenterologists should be aware of the possibility of SLE whenever this syndrome is encountered. Over 70% of the PLE patients in our case series had other organ involvement; LN was the most common, followed by hematologic involvement, and NPSLE. Therefore, PLE can occur in SLE patients with multiple system involvement or it can occur alone. Our data are consistent with other the reported cases in which common presenting features of SLE-related PLE are edema, while GI symptoms were infrequent , , .
We also found that SLE-related PLE patients had hypoalbuminemia, hypocalcemia, hypercholesterolemia, and C3 and C4 hypocomplementemia, and 23% of patients had anti-dsDNA seropositivity. The frequency of anti-dsDNA seropositivity reported in previous studies ranges from “negative in most cases”  to 52% (25/48) , 57% (4/7) , and 75% (12/16) . Future studies are required to further elucidate the frequency of anti-dsDNA seropositivity in SLE-related PLE patients.
We note that the positive rate of anti-SSA/SSB was much higher in patients with SLE-related PLE. We confirmed this phenomenon through case-control analysis, and these data are consistent with those of Zheng et al . Interestingly, anti-SSA/SSB seropositivity is associated with columnar epithelium cells of labial and salivary glands . Whether anti-SSA/SSB acts on intestinal columnar epithelium cells and causes damage is unknown, and this possibility requires further examination.
PLE is clinically indistinguishable from other complications such as nephrotic syndrome . Furthermore, PLE and LN can successively or simultaneously appear in a SLE patient. In our study, over two thirds of PLE patients had concomitant organ manifestation of SLE upon admission. This further increases the difficulty of diagnosing PLE. With this in mind, we performed a risk factor analysis and diagnostic test evaluation. In the multivariate logistic regression analysis, anti-SSA seropositivity was an independent risk factor, and it was positively associated with SLE-related PLE.
Most laboratory tests provided high sensitivity, but low specificity. Some showed excellent discriminating properties with use of the best threshold. The laboratory characteristics that had the strongest ability to discriminate SLE-related PLE from active SLE were hypoalbuminemia and hypocalcemia, whose sensitivity and specificity were approximately 85%, and AUC were approximately 0.9. To decrease the misdiagnosis rate, we explored the accuracy of laboratory characteristics that were serially combined. The best results were obtained using the simultaneous presence of serum albumin (<22 g/l) and 24-hour urine protein (<0.8 g/24 h). This combination had higher specificity, PPV, NPV, and PLR, lower NLR, and no significant reduction in sensitivity.
99 m Tc-HSA scintigraphy is regarded as the diagnostic tool of choice for patients with PLE , , , as it has very high sensitivity for diagnosis and localization of PLE –. However, its application is limited because many hospitals are not able to perform the procedure. Furthermore, some cases show negative results during disease duration. Law et al. reported that two SLE patients in their cohort had negative scans, but PLE was later confirmed by fecal alpha-1-antitrypsin clearance (AATC) tests . We compared the typical clinical and laboratory characteristics of PLE in patients with definite and probable PLE. Almost all characteristics were comparable, including anti-SSA seropositivity, albumin, and total cholesterol level. This finding suggests that SLE-related PLE can be diagnosed in the absence of evidence from 99 m Tc-HSA scintigraphy, and a negative scintigraphy result does not negate a PLE diagnosis.
The exact pathogenesis of SLE-related PLE is unclear. The postulated mechanisms include mucosal ulceration, increased mucosal capillary permeability as a result of intravascular activation and conversion of complement, complement- or cytokine-mediated (e. g. tumor necrosis factor-α and interleukin-6) vascular or mucosal damage, and ruptured mucosal lacteals resulting in lymphangiectasia –, .
Corticosteroids and immunosuppressants were the mainstay treatment for SLE-related PLE patients in our study. Because of low cost efficacy, CTX is one of the most commonly used immunosuppressants for SLE in China. However, there are no studies demonstrating CTX efficacy for the treatment of SLE-related PLE. Most SLE-related PLE patients respond well to combined prednisolone and azathioprine (AZA) , . In our study, almost 90% of our patients responded well to corticosteroids and immunosuppressive therapy within six months. Serum albumin and C3 levels significantly increased after two months of treatment, and by the third month, serum albumin had achieved normal levels and remained stable until six months after treatment. However, it remains unclear whether steroid monotherapy or combination therapy with cyclophosphamide is appropriate for SLE-related PLE patients.
Nonetheless, our study should be interpreted along with several limitations. First, all patients in this study came from a single healthcare center, and this raises the possibility of a referral bias. Second, data collection was conducted retrospectively. Third, the number of cases was too small to allow global conclusions to be drawn from the results of some of the statistical tests. Fourth, a relatively short follow-up period is unfavorable for comprehensive evaluation of this treatment. The main strengths of our study lie in the case-control population, and in the number of patients with SLE-related PLE.
In summary, PLE is an uncommon complication of SLE. It can present as an initial manifestation of SLE, or accompany other organ damage that is often observed in SLE patients. Anti-SSA seropositivity, hypoalbuminemia and hypercholesterolemia were independent risk factors for SLE-related PLE patients. When a patient with SLE presents with generalized edema, severe hypoalbuminemia, and low levels of 24-hour urine protein, SLE-related PLE should be considered. Results from 99 m Tc-HSA scintigraphy may be negative even in the presence of typical PLE clinical and laboratory characteristics. Aggressive treatment for lupus had a high response rate.
This file contains S1–S6 Tables. S1 Table. Frequency of clinical characteristics of patients with definite and probable PLE. S2 Table. Demographic characteristics and clinical data according to SLICC criteria in patients with definite and probable PLE. S3 Table. Laboratory findings in patients with definite and probable PLE. S4 Table. Frequency of laboratory characteristics in the PLE and control groups. S5 Table. Endoscopic and imaging findings from patients in the PLE group. S6 Table. Treatment and prognosis of PLE patients.
This work was earlier presented in abstract form as Z CHEN, M LI, DONG XU, HONG YANG, JING LI, J ZHAO, H ZHANG, S HAN, TAO XU, LINA ZHANG, F ZHANG, YAN ZHAO, X ZENG: Protein-losing enteropathy in systemic lupus erythematosus. International Journal of Rheumatic Diseases, Special Issue: Abstracts of the 16th Asia Pacific League of Associations for Rheumatology Congress, 31 March–4 April 2014, Radisson Blu Hotel, Cebu City, Philippines. Volume 17, Issue Supplement s1, pages 1–133, April 2014 APLAR-0175.
Conceived and designed the experiments: ZC MTL XFZ. Performed the experiments: ZC MTL DX JL JLZ. Analyzed the data: ZC SMH TX. Contributed reagents/materials/analysis tools: ZC MTL HY HHZ. Wrote the paper: ZC MTL.
- 1. Umar SB, DiBaise JK (2009) Protein-Losing Enteropathy: Case Illustrations and Clinical Review. The American Journal of Gastroenterology 105(1):43–49.
- 2. Al-Mogairen SM (2011) Lupus protein-losing enteropathy (LUPLE): a systematic review. Rheumatol Int 31(8):995–1001.
- 3. Tian XP, Zhang X (2010) Gastrointestinal involvement in systemic lupus erythematosus: Insight into pathogenesis, diagnosis and treatment. World J Gastroenterol 16(24):2971–7.
- 4. Law ST, Ma KM, Li KK (2012) The clinical characteristics of lupus related protein-losing enteropathy in Hong Kong Chinese population: 10 years of experience from a regional hospital. Lupus 21(8):840–7.
- 5. Rajendiran A, Viswanathan S, Remalayam B, Muthu V, Alexander T (2012) Chronic diarrhea as the presenting complaint of systemic lupus erythematosus in a man. Intern Med 51(22):3131–4.
- 6. Ratnayake EC, Riyaaz AA, Wijesiriwardena BC (2012) Systemic lupus erythematosus presenting with protein losing enteropathy in a resource limited centre: a case report. Int Arch Med 5(1):1.
- 7. Awazawa R, Yamamoto Y, Mine Y, Nakamura I, Kishimoto K, et al. (2012) Systemic lupus erythematosus complicated with protein-losing enteropathy: a case report and review of the published works. J Dermatol 39(5):454–61.
- 8. Carneiro FO, Sampaio LR, Brandão LA, Braga LL, Rocha FA (2012) Protein-losing enteropathy as initial manifestation of systemic lupus erythematosus. Lupus 21(4):445–8.
- 9. Tsao YT, Wu CC, Hsu YJ, Chu P (2010) Acute right ventricular failure in lupus-associated protein-losing enteropathy. Am J Emerg Med 28(7) 845:e1–3.
- 10. Kim YG, Lee CK, Byeon JS, Myung SJ, Oh JS, et al. (2008) Serum cholesterol in idiopathic and lupus-related protein-losing enteropathy. Lupus 17:575–579.
- 11. Wang YF, Tseng KC, Chiu JS Chuang MH, Chung MI, et al. (2008) Outcome of surgical resection for protein-losing enteropathy in systemic lupus erythematosus. Clin Rheumatol 27(10):1325–8.
- 12. Zheng WJ, Tian XP, Li L, Jing HL, Li F, et al. (2007) Protein-Losing Enteropathy in Systemic Lupus Erythematosus Analysis of the Clinical Features of Fifteen Patients. J Clin Rheumatol 13(6):313–6.
- 13. Mok CC, Ying KY, Mak A, To CH, Szeto ML (2006) Outcome of protein-losing gastroenteropathy in systemic lupus erythematosus treated with prednisolone and azathioprine. Rheumatology (Oxford) 45:425–429.
- 14. Türkçapar N, Ozyüncü N, Cinar K, Ensari A, Küçük O, et al. (2006) A case of systemic lupus erythematosus presenting with protein-losing enteropathy. Turk J Gastroenterol 17(3):226–30.
- 15. Oh DC, Ng TM, Ho J Leong KP (2006) Systemic lupus erythematosus with concurrent protein-losing enteropathy and primary sclerosing cholangitis: a unique association. Lupus 15(2):102–4.
- 16. Hung J, Wood CA, Woronik V, Vieira JM Jr, Barros RT (2006) Protein-losing gastroenteropathy in a patient with systemic lupus erythematosus and antiphospholipid antibody syndrome simulating nephrotic syndrome. Nephrol Dial Transplant 21(7):2027–8.
- 17. Werner de Castro GR, Appenzeller S, Bértolo MB, Costallat LT (2005) Protein-losing enteropathy associated with systemic lupus erythematosus: response to cyclophosphamide. Rheumatol Int 25(2):135–8.
- 18. Park JM, Ahn SY, Shin JI, Yun MJ, Lee JS (2004) A systemic lupus erythematosus patient with protein losing enteropathy. Yonsei Med J 45(5):923–6.
- 19. Wu CC, Lin SH, Chu P, Lai JH, Chang DM, et al. (2004) An unrecognized cause of oedema in a patient with lupus nephritis: protein losing enteropathy. Nephrol Dial Transplant 19(8):2149–50.
- 20. Ossandon A, Bombardieri M, Coari G, Graziani G, Valesini G (2002) Protein losing enteropathy in systemic lupus erythematosus: role of diet and octreotide. Lupus 11(7):465–6.
- 21. Lee CK, Han JM, Lee KN, Lee EY, Shin JH, et al. (2002) Concurrent occurrence of chylothorax, chylous ascites, and protein-losing enteropathy in systemic lupus erythematosus. J Rheumatol 29(6):1330–3.
- 22. Chang JM, Hwang SJ, Chen HC, Lai YH (2002) Edema due to protein-losing enteropathy-a disorder rarely considered by nephrologists. Clin Nephrol 57(5):392–7.
- 23. Yazici Y, Erkan D, Levine DM, Parker TS, Lockshin MD (2002) Protein-losing enteropathy in systemic lupus erythematosus: report of a severe, persistent case and review of pathophysiology. Lupus 11(2):119–23.
- 24. Gattorno M, Buoncompagni A, Barabino A, Barbano G, Loy A, et al. (2002) Severe hypoalbuminaemia in a systemic lupus erythematosus-like patient. Eur J Pediatr 161(2):84–6.
- 25. Aoki T, Noma N, Takajo I, Yamaga J, Otsuka M, et al. (2002) Protein-losing gastropathy associated with autoimmune disease: successful treatment with prednisolone. J Gastroenterol 37(3):204–9.
- 26. Gornisiewicz M, Rodriguez M, Smith JK Saag K, Alarcón GS (2001) Protein-losing enteropathy in a young African-American woman with abdominal pain, diarrhea and hydronephrosis. Lupus 10:835–40.
- 27. Xu D, Yang H, Lai CC, Li P, Zhang X, et al. (2010) Clinical analysis of systemic lupus erythematosus with gastrointestinal manifestations. Lupus 19(7):866–9.
- 28. Takenaka H, Ohmiya N, Hirooka Y, Nakamura M, Ohno E, et al. (2012) Endoscopic and imaging findings in protein-losing enteropathy. J Clin Gastroenterol 46:575–580.
- 29. Hochberg MC (1997) Updating the American College of Rheumatology revised criteria for the classification of systemic lupus erythematosus [letter]. Arthritis Rheum 40:1725.
- 30. Vitali C, Bombardieri S, Jonsson R, Moutsopoulos HM, Alexander EL, et al. (2002) Classification criteria for Sjögren’s syndrome: a revised version of the European criteria proposed by the American-European Consensus Group. Ann Rheum Dis 61(6):554–8.
- 31. Petri M, Orbai AM, Alarcón GS, Gordon C, Merrill JT, et al. (2012) Derivation and validation of the Systemic Lupus International Collaborating Clinics Classification Criteria for systemic lupus erythematosus. Arthritis Rheum 64(8):2677–86.
- 32. Gladman DD, Ibañez D, Urowitz MB (2002) Systemic lupus erythematosus disease activity index 2000. J Rheumatol 29(2):288–91.
- 33. Parodi A, Davì S, Pringe AB, Pistorio A, Ruperto N, et al. (2009) Macrophage activation syndrome in juvenile systemic lupus erythematosus:a multinational multicenter study ofthirty-eight patients. Arthritis Rheum 60(11):3388–99.
- 34. Peng SL, Craft J (2009) Antinuclear antibodies. In Firestein GS, Budd RC, Harris ED, et al: Kelley’s Textbook of Rheumatology pp748.
- 35. Chau TN, Mok MY, Chan EYT, Luk WH, Lai KB, et al. (2011) Evaluation of performance of measurement of faecal α1-antitrypsin clearance and Technetium-99m human serum albumin scintigraphy in protein-losing enteropathy. Digestion 84(3):199–206.
- 36. Al-Mogairen SM (2011) Lupus protein-losing enteropathy (LUPLE): a systematic review. Rheumatol Int 31(8):995–1001.
- 37. Chiu NT, Lee BF, Hwang SJ, Chang JM, Liu GC, et al. (2001) Protein-losing enteropathy: diagnosis with (99 m)Tc-labeled human serum albumin scintigraphy. Radiology 219(1):86–90.
- 38. Khalesi M, Nakhaei AA, Seyed AJ, Treglia G, Zakavi SR, et al. (2013) Diagnostic accuracy of nuclear medicine imaging in protein losing enteropathy: systematic review and meta-analysis of the literature. Acta Gastroenterol Belg 76(4):413–22.