Abstract
We report the case of a 63-year-old female patient with liver cirrhosis who presented with symptoms of severe hypoalbuminaemia and diarrhoea. After ruling out other causes of hypoalbuminaemia and confirmation of an elevated faecal α-1 antitrypsin clearance, the diagnosis of protein-losing enteropathy (PLE) could be established. Since PLE is a syndrome caused by various diseases, classified into erosive and non-erosive gastrointestinal diseases or lymphatic obstruction, an extensive work-up was necessary, establishing the final diagnosis of Crohn’s disease.
Keywords: endoscopy, Crohn's disease, malabsorption, general practice / family medicine, inflammatory bowel disease
Background
Although the burden of Crohn’s disease (CD) is rising globally with a prevalence of around 300 per 100 000 in Europe and North America,1 there still is a rather long median diagnostic delay of around 6–9 months, and even >2 years in a quarter of patients.2 3 One potential explanation for this long diagnostic delay could be that in addition to well-established symptoms, such as diarrhoea and abdominal pain,4 5 a number of patients present with atypical first manifestations of CD. One of these rare atypical presentations can be peripheral oedema, ascites and pleural effusion caused by severe hypoalbuminaemia due to protein-losing enteropathy (PLE).
Our case illustrates the challenging pathway to the final diagnosis of CD in a comorbid patient with PLE as the main clinical symptom.
Case presentation
A 63-year-old woman was referred for further assessment of presumed portal hypertensive enteropathy (PHE) causing severe PLE.
The patient reported on pitting oedema in her legs, chronic watery diarrhoea and occasional episodes of sharp abdominal pain over a period of 2 years. Her medical history was notable for Child-Pugh score A liver cirrhosis as a result of chronic hepatitis C infection, as well as a schizoaffective disorder. She had no pertinent family medical history. Case-relevant medication consisted of spironolactone, torasemide, analgetics, such as metamizole and morphine, as well as esomeprazole and antipsychotic (escitalopram, quetiapine, pipamperone) therapy. The patient denied the use of non-steroidal anti-inflammatory drugs. She continued to smoke 1.5 packs of cigarettes per day and had done that for 50 years; intravenous drug abuse was stopped years ago.
Prior to referral, infectious and parasitic causes of diarrhoea were ruled out. Repetitively performed ileocolonoscopies demonstrated neither significant inflamed regions nor aphthous ulceration in the terminal ileum or in the colon. Accordingly, biopsies of the terminal ileum and the colon were unremarkable. Specifically, granulomas, focal crypt architectural abnormalities with chronic inflammation or basal plasmacytosis were absent, so that a diagnosis of inflammatory bowel disease (IBD) could not have been established. An upper endoscopy was performed to exclude, among others, giant hypertrophic gastritis and coeliac disease.
Due to progressive clinical symptoms, elevated faecal calprotectin and radiological evidence (CT scan) of terminal ileitis of around 25 cm, a therapy with prednisone and later on with infliximab was initiated, despite the lack of respective endoscopic/histological findings. However, the further course was complicated by progressive oedema of both arms and legs with unchanged abdominal symptoms. Laboratory findings showed severe hypoalbuminaemia and CT revealed persisting thickening of the terminal ileum as well as a dilated portal vein along with dilated mesenteric veins. Transthoracic echocardiography ruled out right-sided heart failure and urinalysis revealed no significant protein loss excluding nephrotic syndrome. A repeated colonoscopy, including ileal biopsies, again did not show signs for IBD. With known liver cirrhosis, a PHE causing PLE was suspected and an elevated faecal α-1 antitrypsin (α1AT) clearance was measured. Diuretics and beta blockage were initiated. Due to persisting abdominal symptoms and PLE with clinically significant pitting oedema and ascites, the patient was referred to our tertiary clinic for further assessment.
On admission, the abdominal examination showed tenderness to palpation in the right lower quadrant without signs of peritonitis. Additionally, there was pitting oedema in the legs bilaterally. The remainder of the physical examination was normal. Laboratory studies revealed severe hypoalbuminaemia with an albumin of 14 g/L, macrocytic anaemia and slightly elevated inflammatory parameters. PLE, as the potential cause of the hypoalbuminaemia, was confirmed by a significantly elevated faecal α1AT clearance (60 mL/24 hours; normal values ≤27 mL/24 hours).
To further clarify the suspected PHE, a transjugular measurement of the hepatic venous pressure gradient (HVPG) was performed, which, however, revealed an only slightly elevated HVPG of 7 mm Hg, ruling out clinical significant PHE. In addition, significant hepatocellular dysfunction was ruled out by missing coagulopathy (international normalised ratio 1.1) and normal levels of bilirubin.
Due to persisting signs for ileitis of around 30 cm in the bowel ultrasound (figure 1) and in MRI, we repeated ileocolonoscopy, which, for the first time since the beginning of endoscopic diagnostic management, revealed multiple ulcerations in the last 10 cm of the terminal ileum. However, due to an impassable ileal narrowing, the whole extent of the inflammation could not be visualised and an initially planned balloon-assisted enteroscopy was cancelled. Histology showed no signs of chronicity, leading the pathologist to suspect an infectious or drug-induced aetiology. An extensive evaluation for infectious diseases in the biopsy specimen, including actinomyces, PCR detection of mycobacteria tuberculosis and immunohistochemistry of intestinal spirochetosis, were all negative. Furthermore, a multiplex PCR-based stool assays for gastrointestinal (GI) pathogens were repetitively negative. Neither in the MRI nor in the biopsies signs of a lymphoma could be detected. A CT angiography excluded non-occlusive mesenteric ischaemia and a vasculitis as a rare cause of ileitis.
Figure 1.
Thickened bowel wall as a sign for ileitis in bowel ultrasound.
To rule out more proximal lesions explaining the severe PLE, we additionally performed a capsule endoscopy (figure 2), which confirmed the terminal ileitis with multiple ulcerations, but no other small bowel involvement.
Figure 2.
Ulcerations in the terminal ileum during capsule endoscopy.
With respect to the extensive diagnostic work-up without evidence of other causes of ileitis and the suggestive clinical presentation, we ultimately interpreted severe ileitis most probably based on CD as the cause of PLE, irrespectively of inconclusive histopathological results.
Due to persisting opioid-demanding abdominal symptoms, severe hypoalbuminaemia and in the past steroid-refractory disease course as well as primary non-response to infliximab, we considered laparoscopic ileocaecal resection as a favourable diagnostic and therapeutic option. After optimising metabolic needs with parenteral nutrition and oral nutritional supplements, an ileocaecal resection with a Kono-S anastomosis was performed. The surgical sample (around 50 cm of small bowel) demonstrated chronic transmural inflammation with crypt architectural distortion, pyloric metaplasia and transmural lymphoid aggregates as well as acute ulcerative inflammation with aphtoid lesions (figure 3), ultimately establishing the diagnosis of CD.
Figure 3.
Surgical sample. (A) Chronic transmural inflammation with crypt architectural distortion and transmural lymphoid aggregates; (B) same image in higher resolution showing acute ulcerative inflammation with aphthoid lesion.
Outcome and follow-up
Following the recovery after surgery, the patient gained 5 kg of weight and her albumin levels improved significantly to normal values. However, 3 months after the ileocaecal resection abdominal pain recurred with endoscopic evidence of inflammation of the neoterminal ileum (Rutgeerts score i3) and colon, resulting in the initiation of treatment with vedolizumab. After an initial clinical response, abdominal pain with signs of active inflammation (small bowel wall thickening in the ultrasound and elevated calprotectin) returned. Despite of dose intensification, response to vedolizumab could not be restored and ustekinumab was initiated.
Discussion
We present a case of severe PLE with ileocaecal resection establishing the final diagnosis of previously unknown CD. Despite PLE being recognised as a clinical feature of CD, reports on PLE antecedent to CD diagnosis are scarce and its prevalence remains unknown. One reason could be the relatively low awareness of PLE among physicians resulting in late or missed diagnoses. Since albumin is a negative acute-phase protein, hypoalbuminaemia is often observed in active IBD. However, the most common symptoms are fatigue, diarrhoea and abdominal pain,4 5 and clinical manifestation of hypoalbuminaemia is rarely observed as the leading symptom of undiagnosed CD.
In PLE, an abnormal amount of serum proteins is lost into the GI tract, ultimately leading to hypoalbuminaemia, potentially causing pitting oedema, ascites, pericardial/pleural effusions and malabsorption of fat and fat-soluble vitamins6; diarrhoea, however, is often absent.7 Albumin homeostasis is depending on hepatic synthesis and urinary and intestinal clearance, as well as catabolic conditions, leading to a broad spectrum of differential diagnoses with various clinical presentations. Table 1 gives an overview of differential diagnoses of hypoalbuminaemia, including basic diagnostic approaches. Focusing on intestinal loss (table 2), three major underlying conditions have been proposed: (1) increased lymphatic pressure (eg, lymphangiectasia), (2) non-erosive GI disorders (eg, coeliac disease) and (3) erosive GI disorders.6 The latter is hallmarked by mucosal injury (wider tight junctions), with the degree of mucosal damage correlating with increased permeability for proteins.8 More than 20 diseases causing mucosal erosions have been described,6 including CD, which even in early stages may cause inflammation and ulcerations in various bowel areas, potentially with endoscopically visible oozing (of protein rich fluids) into the GI lumen.9
Table 1.
Overview of differential diagnoses of hypoalbuminaemia
| Pathomechanism | Differential diagnosis | Diagnostic approach |
| Decreased albumin synthesis | Hepatic dysfunction, congestive heart failure, Kwashiorkor* | Liver function tests, brain natriuretic peptide |
| Increased renal albumin loss | Nephrotic syndrome, chronic kidney disease |
Serum creatinine, proteinuria |
| Increased gut albumin loss | Protein-losing enteropathy | Gastroenterological work-up (see table 2), α-1 antitrypsin clearance† |
| Increased catabolism and extravascular loss (3rd space loss) | Inflammation, burn, sepsis | C reactive protein, erythrocyte sedimentation rate |
*Despite Kwashiorkor is associated with hypoalbuminaemia, albumin is not a marker of malnutrition.
†Normal ≤27 mL/24 hours (pitfalls: in patients with diarrhoea up to ≤56 mL/24 hours is considered normal; if pH is <3.5 in the stomach (such as in Menetrier’s disease) initiation of acid suppressive therapy prior to measurement is necessary).
Table 2.
Differential diagnosis of protein-losing enteropathy
| Pathomechanism | Differential diagnosis |
| Increased lymphatic pressure |
|
| Non-erosive GI disorders |
|
| Erosive GI disorders |
|
GI, gastrointestinal.
Currently, the diagnostic method of choice to determine GI protein loss is the α1AT clearance, a quotient of the measured amount of α1AT in a 24 hours’ faecal collection and the plasma α1AT level.7 10
The concept of α1AT clearance to quantify GI protein loss was first described in 1978.11 The α1AT, a hepatic protein, resistant to intestinal proteolysis and neither absorbed nor secreted into the GI tract, has a similar molecular size to albumin; faecal loss as a consequence to mucosal injury is therefore comparable with that of albumin.7 10 Timing of serum α1AT measurement is irrelevant due to relatively constant serum levels.7 However, α1AT is sensitive to pH <3 and therefore unreliable if gastric protein loss (eg, Menetrier’s disease) is suspected.10 12 In cases with highly suspected GI protein loss but normal α1AT clearance, measurement should be repeated on acid suppressive medication. It needs to be mentioned, however, that in patients with severe portal hypertensive gastropathy, significant gastric protein loss may be found as well.13 Therefore, serum α1AT measurements should be interpreted cautiously in patients with liver cirrhosis. Normal α1AT clearance is ≤27 mL/24 hours, but may increase in any cause of diarrhoea with a proposed normal value of ≤56 mL/24 hours.7 8
Theoretically, α1AT (and therefore albumin) clearance should be proportional to the amount of involved mucosa10 and has been proposed to reflect clinical activity of CD.14 15 However, PLE was also described in inactive disease, leading to the conclusion that CD, independent of clinical activity, may lead to mucosal changes with increased intestinal protein exudation.16 17
We conducted a literature research using the PubMed database and only found seven individual cases,18–23 in which the investigation of clinically symptomatic PLE led to the final diagnosis of previously unknown CD (table 3). Despite all of these presenting with clinically relevant oedema, three of these cases (patients 1, 5 and 7) primarily did not present with abdominal pain and three (patients 1–3) without a history of diarrhoea. In addition, in no case an elevation of C reactive protein (CRP) was reported (in patients 2–5 normal CRP values were specifically stated). Although the symptoms in our case were consistent with a diagnosis of CD, there were several factors initially rendering doubts about the suspected diagnosis. Previous ileocolonoscopies, biopsies and the non-response to empirically given glucocorticoids and infliximab were not indicative for CD. Further, the established diagnosis of liver cirrhosis was misleading. Furthermore, after having diagnosed ulcerative ileitis, confidence was lacking whether the cause of the ulceration was CD and whether the severe hypoalbuminaemia was really the result of 30 cm of intestinal inflammation.
Table 3.
Summary of published case reports: protein-losing enteropathy leading to diagnosis of Crohn’s disease
| Patient | Sex; age (years) |
Symptoms | Laboratory work-up | Endoscopic work-up; biopsy | Biopsy | Radiological work-up | Diagnostic modality that finally leads to the diagnosis of CD | Therapy | Follow-up | |
| Moynagh18 | 1 | Female, 48 | Ankle oedema, 4 years later abdominal pain with palpable mass in lower abdomen | Hypoalbuminaemia, iron deficiency anaemia, high faecal 131I polyvinylpyrrolidone | None | None | Barium follow-through (areas of narrowing and of abnormal pattern in jejunum) | Histopathology of resected specimen (patchy mucosal ulceration, chronic inflammation, giant cell granulomata) | Resection of diseased jejunum | Excelled recovery |
| Barkay et al19 | 2 | Male, 13 | Abdominal pain (2 years’ duration), no fever, no diarrhoea, no oedema | Hypoproteinaemia, no proteinuria, normal CRP, iron and vitamin B12 deficiency anaemia, elevated α1AT clearance | Colonoscopy with ileoscopy and push enteroscopy (100 cm of jejunum) | Inconclusive (mild neutrophilic infiltration and lymphoid hyperplasia) | Abdominal CT (abdominal lymphadenopathy) | WCE (multiple aphthous and linear ulcers with surrounding oedema and erythema in mid-jejunum) | 5-ASA | Abdominal pain disappeared, albumin and haemoglobin normalised |
| 3 | Male, 14 | Abdominal pain (1 year duration), facial and leg oedema | Hypoproteinaemia, no proteinuria, normal CRP, elevated α1AT clearance | Multiple gastroscopies and colonoscopies with ileoscopy and 2 enteroscopies (100 cm of jejunum) normal | Normal | 99mTC leucocyte scintigraphy (increased uptake in terminal ileum) | WCE (multiple linear ulcers in deep jejunum and ileum) | 5-ASA and budesonide followed by 6-mercaptopurine | Good clinical response, normalised albumin and haemoglobin levels | |
| Kia et al20 | 4 | Male, 61 | Rapid onset of abdominal pain, pedal and scrotal oedema, mild intermittent diarrhoea | Hypoproteinaemia, no proteinuria, normal CRP | Gastroscopy and colonoscopy with ileoscopy | Normal in duodenum and colon | Abdominal/thoracic CT (pleural effusion and ascites, no features of portal hypertension); small bowel barium series (strictures and ulcerations, entero-sigmoid fistula) | Terminal ileal inflammatory stricture during repeated colonoscopy; biopsies from stricture overall consistent with typical CD | 5-ASA | Clinical condition and serum albumin improved; no recurrence 12 months after initial presentation |
| Rashid et al21 | 5 | Male, 20 | Intermittent diarrhoea, anorexia, weight loss, facial and pedal oedema, no abdominal pain | Hypoalbuminaemia, no proteinuria, iron deficiency anaemia, no α1AT clearance measured | Gastroscopy and colonoscopy with ileoscopy | N/A | N/A | WCE (multiple small bowel ulcerations, strictures and possible fistula) | Remicade, azathioprine and mesalamine | Clinical remission |
| Cakir et al22 | 6 | Female, 15 | Abdominal pain (2 months’ duration), ascites, lower extremity oedema, watery diarrhoea | Hypoalbuminaemia, no proteinuria, normal CRP, no α1AT clearance measured | Gastroscopy and colonoscopy (diffuse aphthous and mucosal ulcers with ‘punched out holes’, mucosal fissures and fragility) | Severe active colitis and ulceration and CMV inclusions | Abdominal CT (massive ascites and diffuse thickening in intestinal wall including terminal ileum and whole colonic segments) | Histopathology | Oral prednisone | Patient died 22 days after initial evaluation due to disseminated CMV infection |
| Ito et al23 | 7 | Male, 44 | Systemic oedema (+6 kg in 14 days), regularly soft stools or diarrhoea with 5 bowel movements per day for 20 years, no abdominal pain | Hypoalbuminaemia | Colonoscopy with ileoscopy (long segmental stenosis of ileum with longitudinal ulcer) | Inconclusive | Abdominal CT and small bowel follow-through (ileal stenosis with dilation of oral side) | Histopathology of resected specimen (transmural inflammation, granulomas, atrophic villus, pyloric gland metaplasia) | Mesalazine followed by surgical resection and adalimumab | Good clinical response |
5-ASA, 5-aminosalicylic acid; α1AT, α-1 antitrypsin; CD, Crohn’s disease; CMV, cytomegalovirus; CRP, C reactive protein; 99mTC, technetium-99m; N/A, not available; WCE, wireless capsule endoscopy.
Similar to our case, only in one case (patient 6) the diagnosis of CD was established by the initial histopathology of ileal biopsy. In all other cases, an extensive work-up with multiple diagnostic modalities was necessary to establish the final diagnosis. As in our case, nearly all patients (except patient 6 who died due to a disseminated cytomegalovirus infection) showed a benign course with normalisation of albumin values after the diagnosis of CD and initiation of therapy.
Having in mind that CD represents a disease with potentially atypical initial presentation, we believe that clinical manifestation of PLE, unrelated to cardiac, nephrological or cirrhotic conditions, should prompt an extensive gastroenterological evaluation. The possibility of underlying IBD, especially CD with patchy appearance, has to be included in the differential diagnosis. In patients with a history of abdominal pain and/or diarrhoea, repeated endoscopies and small bowel imaging using wireless capsule endoscopy or magnetic resonance enterography have to be considered. Studies evaluating the prevalence of PLE at the time of IBD diagnosis and during course of disease are lacking.
In summary, our case highlights the importance of considering CD in a patient with clinical manifestations of severe hypoalbuminaemia and PLE.
Learning points.
Crohn’s disease represents a disease with potentially atypical initial presentation.
Consider Crohn’s disease in patients with clinical manifestation of severe hypoalbuminaemia and protein-losing enteropathy.
Currently, the diagnostic method of choice to determine gastrointestinal protein loss is the α1-antitrypsin clearance.
Acknowledgments
The authors thank Dr Antonia Töpfer from the Department of Pathology of the University Hospital Zurich for her critical review and providing the images of the histology sample taken during the surgical intervention.
Footnotes
Contributors: FRM: substantial contributions to the conception or design of the work, drafting the work and approved the final version. BM: interpretation of data for the work, revising it critically for important intellectual content and approved the final version. LB: interpretation of data for the work, revising it critically for important intellectual content and approved the final version. PS: substantial contributions to the conception or design of the work, drafting the work and approved the final version.
Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests: None declared.
Patient consent for publication: Obtained.
Provenance and peer review: Not commissioned; externally peer reviewed.
References
- 1.Ng SC, Shi HY, Hamidi N, et al. Worldwide incidence and prevalence of inflammatory bowel disease in the 21st century: a systematic review of population-based studies. Lancet 2018;390:2769–78. 10.1016/S0140-6736(17)32448-0 [DOI] [PubMed] [Google Scholar]
- 2.Schoepfer A, Santos J, Fournier N, et al. Systematic Analysis of the Impact of Diagnostic Delay on Bowel Damage in Paediatric Versus Adult Onset Crohn’s Disease. Journal of Crohn’s colitis 2019;13:1334–42. 10.1093/ecco-jcc/jjz065 [DOI] [PubMed] [Google Scholar]
- 3.Vavricka SR, Spigaglia SM, Rogler G, et al. Systematic evaluation of risk factors for diagnostic delay in inflammatory bowel disease. Inflamm Bowel Dis 2012;18:496–505. 10.1002/ibd.21719 [DOI] [PubMed] [Google Scholar]
- 4.Perler BK, Ungaro R, Baird G, et al. Presenting symptoms in inflammatory bowel disease: descriptive analysis of a community-based inception cohort. BMC Gastroenterol 2019;19:47. 10.1186/s12876-019-0963-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Sands BE. From symptom to diagnosis: clinical distinctions among various forms of intestinal inflammation. Gastroenterology 2004;126:1518–32. 10.1053/j.gastro.2004.02.072 [DOI] [PubMed] [Google Scholar]
- 6.Greenwald DA. Protein-losing gastroenteropathy : Feldman M, Friedman LS, Brandt LJ, Sleisenger and Fordtran’s Gastrointestinal and Liver Disease. Vol 1 10th ed Philadelphia, PA: Saunders, 2016: 44–70. [Google Scholar]
- 7.Umar SB, DiBaise JK. Protein-Losing enteropathy: case illustrations and clinical review. Am J Gastroenterol 2010;105:43–9. quiz 50. 10.1038/ajg.2009.561 [DOI] [PubMed] [Google Scholar]
- 8.Ferrante M, Penninckx F, De Hertogh G, et al. Protein-Losing enteropathy in Crohn's disease. Acta Gastroenterol Belg 2006;69:384–9. [PubMed] [Google Scholar]
- 9.Dubcenco E, Jeejeebhoy KN, Petroniene R, et al. Capsule endoscopy findings in patients with established and suspected small-bowel Crohn's disease: correlation with radiologic, endoscopic, and histologic findings. Gastrointest Endosc 2005;62:538–44. 10.1016/j.gie.2005.06.026 [DOI] [PubMed] [Google Scholar]
- 10.Levitt DG, Levitt MD. Protein losing enteropathy: comprehensive review of the mechanistic association with clinical and subclinical disease states. Clin Exp Gastroenterol 2017;10:147–68. 10.2147/CEG.S136803 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Bernier JJ, Florent C, Desmazures C, et al. Diagnosis of protein-losing enteropathy by gastrointestinal clearance of alpha1-antitrypsin. Lancet 1978;2:763–4. 10.1016/S0140-6736(78)92650-8 [DOI] [PubMed] [Google Scholar]
- 12.Florent C, L'Hirondel C, Desmazures C, et al. Intestinal clearance of alpha 1-antitrypsin. A sensitive method for the detection of protein-losing enteropathy. Gastroenterology 1981;81:777–80. [PubMed] [Google Scholar]
- 13.Becheur H, Pauwels A, Mostefa-Kara N, et al. [Gastric protein loss and alcoholic cirrhosis. Study by measurement of gastric clearance of alpha 1-antitrypsin]. Gastroenterol Clin Biol 1996;20:669–73. [PubMed] [Google Scholar]
- 14.Becker K, Berger M, Niederau C, et al. Individual fecal alpha 1-antitrypsin excretion reflects clinical activity in Crohn's disease but not in ulcerative colitis. Hepatogastroenterology 1999;46:2309–14. [PubMed] [Google Scholar]
- 15.Biancone L, Fantini M, Tosti C, et al. Fecal alpha 1-antitrypsin clearance as a marker of clinical relapse in patients with Crohn's disease of the distal ileum. Eur J Gastroenterol Hepatol 2003;15:261–6. 10.1097/00042737-200303000-00009 [DOI] [PubMed] [Google Scholar]
- 16.Karbach U, Ewe K, Bodenstein H. Alpha 1-antitrypsin, a reliable endogenous marker for intestinal protein loss and its application in patients with Crohn's disease. Gut 1983;24:718–23. 10.1136/gut.24.8.718 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Fischbach W, Becker W, Mössner J, et al. Faecal alpha-1-antitrypsin and excretion of 111indium granulocytes in assessment of disease activity in chronic inflammatory bowel diseases. Gut 1987;28:386–93. 10.1136/gut.28.4.386 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Moynagh PD. Crohn's disease of the jejunum with protein-losing enteropathy. Proc R Soc Med 1968;61:443–4. 10.1177/003591576806100510 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Barkay O, Moshkowitz M, Reif S. Crohn's disease diagnosed by wireless capsule endoscopy in adolescents with abdominal pain, protein-losing enteropathy, anemia and negative endoscopic and radiologic findings. Isr Med Assoc J 2005;7:216–8. [PubMed] [Google Scholar]
- 20.Kia R, White D, Sarkar S. An unusual presentation of fistulating Crohn's disease: ascites. World J Gastrointest Endosc 2010;2:41–3. 10.4253/wjge.v2.i1.41 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Rashid S, Dahl K, Subramani K. Small Bowel Crohn’s Disease Presenting as Protein-Losing Enteropathy and Iron-Deficiency Anemia Diagnosed by Wireless Capsule Endoscopy:. 2011;106:S366–7. 980. [Google Scholar]
- 22.Cakir M, Ersoz S, Akbulut UE. Disseminated cytomegalovirus infection and protein losing enteropathy as presenting feature of pediatric patient with Crohn's disease. Pediatr Gastroenterol Hepatol Nutr 2015;18:60–5. 10.5223/pghn.2015.18.1.60 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Ito S, Higashiyama M, Horiuchi K, et al. Atypical clinical presentation of Crohn's disease with superior mesenteric vein obstruction and protein-losing enteropathy. Intern Med 2019;58:369–74. 10.2169/internalmedicine.1192-18 [DOI] [PMC free article] [PubMed] [Google Scholar]



