Clinical significance of duodenal lymphangiectasia incidentally found during routine upper gastrointestinal endoscopy

• Introduction
• Materials and methods
• Retrospective study
• Prospective study
• Endoscopic classification and histologic grade of duodenal lymphangiectasia
• Statistical analysis
• Results
• Retrospective study
• Prospective study
• Pathologic findings
• Discussion
• References

Background and study aim: Although duodenal lymphangiectasia in individuals without clinical evidence of malabsorption has been reported, the prevalence and clinical significance in this situation are not yet known. The aim of this study was to evaluate the prevalence and clinical significance of incidentally found duodenal lymphangiectasia.

Patients and methods: A retrospective review of medical records was undertaken for consecutive patients who had undergone diagnostic upper endoscopy between January 2005 and June 2006. A prospective study was then performed in consecutive individuals undergoing routine upper endoscopy for health examination between July 2006 to October 2006. Endoscopic features of duodenal lymphangiectasia were classified into three types: (1) multiple scattered pinpoint white spots; (2) diffuse prominent villi with whitish-discolored tips; and (3) focal small whitish macule or nodule. The histologic grade of duodenal lymphangiectasia was classified according to the depth and severity of lymphatic duct dilatations. Prevalence and clinical data of incidentally found duodenal lymphangiectasia were evaluated in the retrospective and prospective studies.

Results: Among 1866 retrospective cases, duodenal lymphangiectasia was endoscopically suspected in 59 cases (3.2 %), and histologically confirmed in 35 cases (1.9 %). No clinical evidence of malabsorption was noted in the duodenal lymphangiectasia cases. The ”scattered pinpoint white spots” type was the most frequently found endoscopic feature (40.0 %). Duodenal lymphangiectasia was persistent in seven of 10 individuals who underwent repeat endoscopy after a median of 12 months. Among 134 prospective cases, duodenal lymphangiectasia was histologically confirmed in 12 cases (8.9 %). There was no significant clinical difference between groups with and without duodenal lymphangiectasia. Lymphatic duct dilatation was histologically more severe in the ”focal small whitish macule or nodule” type than in the other types.

Conclusion: Duodenal lymphangiectasia without clinical evidence of malabsorption is not extremely rare among cases undergoing routine upper gastrointestinal endoscopy.

Introduction

Intestinal lymphangiectasia is a disease characterized by focal or generalized dilatation of intestinal lymphatic ducts, and may induce protein-losing enteropathy, steatorrhea, and lymphocytopenia [1]. Primary intestinal lymphangiectasia is caused by congenital malformation of the generalized lymphatic system and primarily affects children and young adults [2] [3]. A secondary form is caused by various diseases that induce lymphatic duct obstruction or central venous pressure elevation [4] [5] [6] [7].

Since the reporting of transient duodenal lymphangiectasia in healthy volunteers after nasogastric olive oil infusion [8], there have been several reports describing endoscopic evidence of duodenal lymphangiectasia in patients without clinical evidence of malabsorption [9] [10] [11].

Recently, a report was pubslished on the submucosal yellow plaques (an endoscopic finding of lymphangiectasia), observed in the small intestine during double-balloon enteroscopy [12]. The detection rate was 13.3 % of all study subjects. Most of the lesions were found in the jejunum. However, the prevalence and clinical significance of the incidentally found lymphangiectasia in the duodenum during routine upper gastrointestinal endoscopy are not yet known.

The aim of this study was to evaluate the prevalence and clinical significance of duodenal lymphangiectasia incidentally found during routine upper gastrointestinal endoscopy.

Materials and methods

Retrospective study

Endoscopic data were retrospectively analyzed from 1866 consecutive patients (mean age 52.0±14.0 years; men/women 920/946) who had undergone routine diagnostic upper gastrointestinal endoscopy for various reasons. All endoscopies had been performed by one endoscopist (Y.T.B.) at the Korea University Guro Hospital, Seoul, Korea between January 2005 and June 2006. All endoscopies were carried out after at least 12 hours of fasting. A standard-resolution endoscope (GIF-H260; Olympus, Tokyo, Japan) was used, and the duodenum was scrutinized up to the mid-portion of the descending part, as is standard for upper gastrointestinal endoscopy. If necessary, two or three pieces of mucosal specimen were taken using standard biopsy forceps (FB-24U-1; Olympus, Tokyo, Japan) for histologic examination. All endoscopic pictures taken during the study were stored in the main hospital computer system for picture archiving and communication, and endoscopic reports were typed and saved by the endoscopist in the hospital electronic medical record system. The reports were in free text forms and included separate descriptions of the endoscopic findings such as number, size, shape, color, and location of any lesions in the pharynx, esophagus, stomach, and duodenum.

The prevalence of duodenal lymphangiectasia was calculated, and endoscopic features and clinical characteristics including age, sex, symptoms, and laboratory data were reviewed from the patients who showed duodenal lymphangiectasia endoscopically and histologically.

Prospective study

Endoscopic and clinical data from all 134 consecutive asymptomatic individuals (mean age 47.3 ± 12.3 years; men/women 75/59) undergoing upper gastrointestinal endoscopy, which was performed by another endoscopist (J.H.K.), for routine health examination at Korea University Guro Hospital, Seoul, Korea between July 2006 and October 2006 were collected prospectively without a single dropout case. The endoscopic procedure was the same as that followed in the retrospective study. Endoscopic biopsy and histologic examination were performed to confirm duodenal lymphangiectasia in all endoscopically suspected cases.

Before endoscopy, each individual was asked for frequency of defecation and stool consistency during the past week. Stool consistency was recorded using the Bristol stool form scale [13], and stools of Bristol scales 5 – 7 were defined as “loose stool”.

Serum levels of calcium, phosphate, and immunoglobulins G, A, and M, stool fat test using glacial acetic acid and Sudan III staining, and abdominopelvic ultrasonography were checked in individuals with histologically confirmed duodenal lymphangiectasia in order to detect the presence of a subclinical form of malabsorption.

The primary outcome of this prospective study was to measure the prevalence of incidentally found duodenal lymphangiectasia. The secondary outcomes included evaluating clinical differences between the individuals with and without duodenal lymphangiectasia and evaluating the correlation between endoscopic classification and pathologic grade.

Endoscopic classification and histologic grade of duodenal lymphangiectasia

Endoscopic features of duodenal lymphangiectasia were classified into three cardinal findings: (1) multiple scattered pinpoint white spots; (2) diffuse prominent villi with whitish-discolored tips; and (3) focal small whitish macule or nodule ([Fig. 1]) [14] [15].

Histologic interpretation was performed by a pathologist (B.K.S.), and histologic grade of duodenal lymphangiectasia was classified according to the depth and severity of lymphatic duct dilatations. The depth of duodenal lymphangiectasia was classified as either mucosal involvement or submucosal involvement [15]. The severity of lymphatic duct dilatation was classified into three groups: (1) mild, if the diameter of the dilated lymphatic duct was less than half of the maximal diameter of the villi; (2) moderate, if the diameter of the dilated lymphatic duct was between half and the maximal diameter of the villi; (3) severe, if the diameter of the dilated lymphatic duct was greater than the maximal diameter of the villi with consequent shortening or widening of the villi ([Fig. 2]). Electron microscopic examination was performed in cases with severely dilated lymphatic ducts.

Correlation between endoscopic classifications and histologic grades was analyzed.

Statistical analysis

Statistical analysis was done with the SAS program (v9.1.3, SAS Institute Inc., Cary, North Carolina, USA). Data were expressed as mean ± SD. Age, body mass index, and laboratory data were compared between groups with presence and absence of duodenal lymphangiectasia in the prospective study using the independent t-test. Differences in sex and proportions of individuals with loose stool were assessed using the chi-square test and Fisher’s exact test, respectively. The correlation between endoscopic classifications and histologic grades was assessed using the generalized Cochran–Mantel–Haenszel test. P-values of less than 0.05 were considered statistically significant.

This study was performed in accordance with the humane and ethical principles of research set forth in the Helsinki Declaration. Before each endoscopy, written informed consent was obtained from each individual after thorough explanation about endoscopic procedures, including the indications for taking biopsy specimens for histologic examinations if necessary.

Results

Retrospective study

Among 1866 patients, endoscopic duodenal lymphangiectasia was observed in 59 (3.2 %) (mean age 56.9 ± 14.0 years; men/women 29/30). All lesions suggesting lymphangiectasia were found on the descending part of the duodenum. Histologic confirmation of duodenal lymphangiectasia was made in 35 (1.9 % of total endoscopies) (mean age 57.1 ± 14.0 years; men/women 15/20) of the 37 patients in whom endoscopic biopsy was performed. No clinical evidence of protein-losing enteropathy, steatorrhea, and leukocytopenia was found among the 59 patients with endoscopic duodenal lymphangiectasia. Associated medical conditions in the histologically confirmed cases are shown in [Table 1]. Epigastric pain was the most common symptom, and gastroesophageal reflux symptoms such as heartburn and acid regurgitation followed in frequency ([Table 2]).

The type of ”scattered pinpoint white spots” was the most common endoscopic finding in both the endoscopically suspected group and the histologically confirmed group ([Table 3]). The ”focal small whitish macule or nodule” lesion type mostly occurred singly and the lesions of the two other types were usually too numerous to count.

Among histologically confirmed duodenal lymphangiectasia cases, follow-up endoscopy was performed for other reasons in 10 cases after a median period of 12 months (range 2 – 20 months), and duodenal lymphangiectasia was persistent in seven of them.

Prospective study

A total of 134 asymptomatic individuals underwent upper gastrointestinal endoscopy for routine health examination. Of these 15 were endoscopically suspected of having duodenal lymphangiectasia (11.2 % of total cases), and this was histologically confirmed in 12 (8.9 % of total cases) (mean age 50.0 ± 16.7; men/women 5/7). All lymphangiectasia lesions were found on the descending part of the duodenum.

There was no statistical difference in demographic characteristics such as age, sex and body mass index, laboratory data such as white blood cell count and serum protein, albumin, total cholesterol and triglyceride levels, and proportion of individuals with loose stools between the groups with and without histologically confirmed duodenal lymphangiectasia ([Table 4]). Serum levels of calcium, phosphate, and immunoglobulins were within normal range, and stool fat test was negative in all histologically confirmed cases. Abdominopelvic ultrasonography was negative in all 12 cases with histologically confirmed duodenal lymphangiectasia, whereas four cases of fatty liver, one case of gallbladder stone, and one case of hepatic hemangioma were diagnosed in the remaining 122 cases without duodenal lymphangiectasia. The ”scattered pinpoint white spots” type was the most common endoscopic finding (seven cases).

Pathologic findings

In 29 of the 47 histologically confirmed duodenal lymphangiectasia cases, endoscopic biopsy specimens contained submucosal layers; correlation between endoscopic classification and depth of lymphatic duct dilatation could be analyzed in these cases. The correlation between endoscopic classification and severity of lymphatic duct dilatation could be analyzed in all 47 histologically confirmed duodenal lymphangiectasia cases. There was no correlation between endoscopic classification and histologic depth of lymphatic duct dilatation, but the grade of lymphatic duct dilatation was more severe in the group of ”focal small whitish macule or nodule” than the other two groups ([Table 5]).

Electron microscopy was performed in four cases of severe lymphatic duct dilatation on light microscopy. Dilated lymphatic vessels and perivascular edema were observed but definite structural anomaly of lymphatic ducts was not found ([Fig. 3]).

Discussion

In our retrospective study, the prevalence of incidentally found duodenal lymphangiectasia confirmed by histology was 1.9 % among cases undergoing routine diagnostic upper gastrointestinal endoscopies for various reasons. The prevalence of duodenal lymphangiectasia could be 3.2 %, if all individuals in whom duodenal lymphangiectasia was suspected endoscopically were included. The prevalence of histologically proven duodenal lymphangiectasia from our prospective study was 9.0 % among asymptomatic individuals undergoing upper gastrointestinal endoscopies for routine health check-ups. The difference of the study population between the retrospective and prospective studies might have contributed to the difference in the detection rates of the lesions. However, the result might more likely be due to the increased awareness of the endoscopist who performed the prospective study and the resultant endeavor to look for the duodenal lymphangiectasia lesions.

According to a recent study, the submucosal yellow plaques that suggest lymphangiectasia were endoscopically found in 13.3 % and the presence of lymphangiectasias was pathologically proven in 10.0 % of total patients undergoing double-balloon enteroscopy, which was performed for various indications [12]. The endoscopic and pathologic detection rates of lymphangiectasia were quite similar to those of the present report, although their lymphangiectasia lesions were mostly found in the jejunum and all belonged to the type of ”focal small whitish macule or nodule.”

In the present report, among 52 individuals in whom duodenal lymphangiectasia was endoscopically suspected and from whom endoscopic biopsies were taken, 47 (90.4 %) showed characteristic histologic findings of dilated lymphatic vessels in the mucosa and submucosa. Therefore, a careful endoscopic observation is usually sufficient for the diagnosis of lymphangiectasia without pathologic examination with endoscopic biopsy specimen.

The ”scattered pinpoint white spots” type was the most common endoscopic finding of incidentally found duodenal lymphangiectasia in both retrospective and prospective studies. Lymphatic duct dilatation was more severe in cases with ”focal small whitish macule or nodule” type than the two other types. Electron microscopy showed perivascular edema and dilatation of lymphatic vessels, but structural anomalies of lymphatic vessels were not observed.

In the retrospective study, 14 of 35 cases (40 %) with duodenal lymphangiectasia had no associated disease. Two cases had medical conditions that might induce secondary intestinal lymphangiectasia (i. e. systemic lupus erythematosus [n = 1] and chronic pancreatitis [n = 1]) but no clinical evidence of malabsorption was found in these two cases.

The exact etiology of asymptomatic duodenal lymphangiectasia has not been elucidated so far. Transient disturbance of fat transport had been suggested as a possible etiology [8]. However, all cases in our study had been fasting for 12 hours or longer prior to the endoscopy and, therefore, the appearance of duodenal lymphangiectasia cannot be explained by this mechanism.

Among 10 cases of duodenal lymphangiectasia patients in whom follow-up endoscopies were done, duodenal lymphangiectasia was persistent in seven cases after a median period of 12 months after initial diagnosis. This result is contradictory to a previous study reporting that functional duodenal lymphangiectasia was a transient and reversible finding [8].

In our study, quantitative evaluations such as α₁-antitrypsin fecal clearance test or fat absorption test to detect inapparent malabsorption were not performed. However, given that the symptoms or signs suggestive of malabsorption did not develop during the follow-up period in seven cases in whom persistent duodenal lymphangiectasia was found on the follow-up endoscopy in the retrospective study and that the laboratory data such as serum levels of protein, immunoglobulin and cholesterol were within normal range in duodenal lymphangiectasia cases in the prospective study, it is likely that there was little possibility of presence of inapparent or subclinical malabsorption.

In conclusion, duodenal lymphangiectasia without clinical evidence of malabsorption is not extremely rare among cases undergoing routine upper gastrointestinal endoscopy, and repeat endoscopy or biopsy is not warranted in the absence of clinical evidence of malabsorption.

Competing interests: None

References

• 1 Mistilis S P, Skyring A P, Stephen D D. Intestinal lymphangiectasia: mechanism of enteric loss of plasma protein and fat.  Lancet. 1965;  1 77-79
  PubMedSearch in Google Scholar
• 2 Vallet H L, Holtzapple P G, Eberlein W R. et al . Noonan syndrome with intestinal lymphangiectasia. A metabolic and anatomic study.  J Pediatr. 1972;  80 269-274
  PubMedSearch in Google Scholar
• 3 Pomerantz M, Waldmann T A. Systemic lymphatic abnormalities associated with gastrointestinal protein loss secondary to intestinal lymphangiectasia.  Gastroenterology. 1963;  45 703-711
  PubMedSearch in Google Scholar
• 4 Edworthy S M, Fritzler M J, Kelly J K. et al . Protein-losing enteropathy in systemic lupus erythematosus associated with intestinal lymphangiectasia.  Am J Gastronenterol. 1990;  85 1398-1402
  PubMedSearch in Google Scholar
• 5 van Tilburg A J, van Blankenstein M, Verschoor L. Intestinal lymphangiectasia in systemic sclerosis.  Am J Gastroenterol. 1988;  83 1418-1419
  PubMedSearch in Google Scholar
• 6 Berkowitz I, Segal I. Protein-losing enteropathy in congestive cardiac failure: an entity of minor clinical significance.  Am J Gastroenterol. 1990;  85 154-156
  PubMedSearch in Google Scholar
• 7 Broder S, Callihan T R, Jaffe E S. et al . Resolution of longstanding protein-losing enteropathy in a patient with intestinal lymphangiectasia after treatment for malignant lymphoma.  Gastroenterology. 1981;  80 166-168
  PubMedSearch in Google Scholar
• 8 Fempell J, Lux G, Kaduk B. et al . Functional lymphangiectasia of the duodenal mucosa.  Endoscopy. 1978;  10 44-46
  Thieme ConnectPubMedSearch in Google Scholar
• 9 Van der Meer S B, Forget P P, Willebrand D. Intestinal lymphangiectasia without protein loss in a child with abdominal pain.  J Pediatr Gastroenterol Nutr. 1990;  10 246-248
  PubMedSearch in Google Scholar
• 10 Patel A S, DeRidder P H. Endoscopic appearance and significance of functional lymphangiectasia of the duodenal mucosa.  Gastrointest Endosc. 1990;  36 376-378
  CrossrefPubMedSearch in Google Scholar
• 11 Barnes R E, DeRidder P H. Fat absorption in patients with functional intestinal lymphangiectasia and lymphangiectic cysts.  Am J Gastroenterol. 1993;  88 887-890
  PubMedSearch in Google Scholar
• 12 Bellutti M, Mönkemüller K, Fry L C. et al . Characterization of yellow plaques found in the small bowel during double-balloon enteroscopy.  Endoscopy. 2007;  39 1059-1063
  Thieme ConnectPubMedSearch in Google Scholar
• 13 Heaton K W, Ghosh S, Braddon F E. How bad are the symptoms and bowel dysfunction of patients with the irritable bowel syndrome? A prospective, controlled study with emphasis on stool form.  Gut. 1991;  32 73-79
  CrossrefPubMedSearch in Google Scholar
• 14 Asakura H, Miura S, Morishita T. et al . Endoscopic and histopathological study on primary and secondary intestinal lymphangiectasia.  Dig Dis Sci. 1981;  26 312-320
  CrossrefPubMedSearch in Google Scholar
• 15 Aoyagi K, Iida M, Yao T. et al . Characteristic endoscopic features of intestinal lymphangiectasia: correlation with histological findings.  Hepatogastroenterology. 1997;  44 133-138
  PubMedSearch in Google Scholar

Y. T. Bak, MD

Department of Gastroenterology
Korea University Guro Hospital

97 Gurodong-gil
Guro-gu
Seoul 152-703
Korea

Fax: +82-505-1151778

Email: drbakyt@korea.ac.kr

References

• 1 Mistilis S P, Skyring A P, Stephen D D. Intestinal lymphangiectasia: mechanism of enteric loss of plasma protein and fat.  Lancet. 1965;  1 77-79
  PubMedSearch in Google Scholar
• 2 Vallet H L, Holtzapple P G, Eberlein W R. et al . Noonan syndrome with intestinal lymphangiectasia. A metabolic and anatomic study.  J Pediatr. 1972;  80 269-274
  PubMedSearch in Google Scholar
• 3 Pomerantz M, Waldmann T A. Systemic lymphatic abnormalities associated with gastrointestinal protein loss secondary to intestinal lymphangiectasia.  Gastroenterology. 1963;  45 703-711
  PubMedSearch in Google Scholar
• 4 Edworthy S M, Fritzler M J, Kelly J K. et al . Protein-losing enteropathy in systemic lupus erythematosus associated with intestinal lymphangiectasia.  Am J Gastronenterol. 1990;  85 1398-1402
  PubMedSearch in Google Scholar
• 5 van Tilburg A J, van Blankenstein M, Verschoor L. Intestinal lymphangiectasia in systemic sclerosis.  Am J Gastroenterol. 1988;  83 1418-1419
  PubMedSearch in Google Scholar
• 6 Berkowitz I, Segal I. Protein-losing enteropathy in congestive cardiac failure: an entity of minor clinical significance.  Am J Gastroenterol. 1990;  85 154-156
  PubMedSearch in Google Scholar
• 7 Broder S, Callihan T R, Jaffe E S. et al . Resolution of longstanding protein-losing enteropathy in a patient with intestinal lymphangiectasia after treatment for malignant lymphoma.  Gastroenterology. 1981;  80 166-168
  PubMedSearch in Google Scholar
• 8 Fempell J, Lux G, Kaduk B. et al . Functional lymphangiectasia of the duodenal mucosa.  Endoscopy. 1978;  10 44-46
  Thieme ConnectPubMedSearch in Google Scholar
• 9 Van der Meer S B, Forget P P, Willebrand D. Intestinal lymphangiectasia without protein loss in a child with abdominal pain.  J Pediatr Gastroenterol Nutr. 1990;  10 246-248
  PubMedSearch in Google Scholar
• 10 Patel A S, DeRidder P H. Endoscopic appearance and significance of functional lymphangiectasia of the duodenal mucosa.  Gastrointest Endosc. 1990;  36 376-378
  CrossrefPubMedSearch in Google Scholar
• 11 Barnes R E, DeRidder P H. Fat absorption in patients with functional intestinal lymphangiectasia and lymphangiectic cysts.  Am J Gastroenterol. 1993;  88 887-890
  PubMedSearch in Google Scholar
• 12 Bellutti M, Mönkemüller K, Fry L C. et al . Characterization of yellow plaques found in the small bowel during double-balloon enteroscopy.  Endoscopy. 2007;  39 1059-1063
  Thieme ConnectPubMedSearch in Google Scholar
• 13 Heaton K W, Ghosh S, Braddon F E. How bad are the symptoms and bowel dysfunction of patients with the irritable bowel syndrome? A prospective, controlled study with emphasis on stool form.  Gut. 1991;  32 73-79
  CrossrefPubMedSearch in Google Scholar
• 14 Asakura H, Miura S, Morishita T. et al . Endoscopic and histopathological study on primary and secondary intestinal lymphangiectasia.  Dig Dis Sci. 1981;  26 312-320
  CrossrefPubMedSearch in Google Scholar
• 15 Aoyagi K, Iida M, Yao T. et al . Characteristic endoscopic features of intestinal lymphangiectasia: correlation with histological findings.  Hepatogastroenterology. 1997;  44 133-138
  PubMedSearch in Google Scholar

Y. T. Bak, MD

Department of Gastroenterology
Korea University Guro Hospital

97 Gurodong-gil
Guro-gu
Seoul 152-703
Korea

Fax: +82-505-1151778

Email: drbakyt@korea.ac.kr