Cristina Camarero

Human small-intestinal epithelium contains functional natural killer lymphocytes

BACKGROUND & AIMS CD3(-) non-T lymphocytes constitute the second most abundant lymphoid subset in the human small-bowel epithelium, and these CD3(-) intraepithelial lymphocytes are virtually absent in active celiac disease. Phenotypically, they resemble natural killer cells and have been termed natural killer-like intraepithelial lymphocytes. Because of the limited availability of appropriate human samples, functional studies have not yet been reported, and it is not yet clear whether these are true natural killer cells. METHODS We used magnetic bead-based purification and flow cytometry to study several aspects of normal human small-bowel natural killer-like intraepithelial lymphocytes: intracellular cytokine content (basally and after activation); ability to lyse natural killer-sensitive K562 target cells; and expression of perforins, Fas ligand, and other functional markers. RESULTS CD3(-) intraepithelial lymphocytes cultured in interleukin-2 showed a higher lymphokine-activated killer activity than CD3(+) intraepithelial lymphocytes (48%-83% lysis exerted by CD3(-) intraepithelial lymphocytes at an effector-target cell ratio of 2:1 vs. 8%-18% by CD3(+) intraepithelial lymphocytes). Perforin content correlated with this lytic potential (75% +/- 4% in CD3(-) vs. 5% +/- 4% in CD3(+) intraepithelial lymphocytes). Both CD3(-) and CD3(+) cells displayed a type I cytokine profile (interferon-gamma > tumor necrosis factor-alpha > interleukin-2; undetectable interleukin-4 and interleukin-10). In addition to their activated phenotype, subsets of natural killer-like intraepithelial lymphocytes expressed CD8alphaalpha and intracellular CD3epsilon chain, showing the existence of heterogeneity within this cell lineage. CONCLUSIONS This is the first demonstration of functional natural killer cells within the human gut epithelium. These cells might play an important role in innate mucosal immunity (host defense and tumor surveillance) and tolerance.

Hepatitis E Virus: Relevance in Blood Donors and Risk Groups

Background and Objectives: Hepatitis E virus (HEV) infection usually causes acute self–limited disease. HEV is associated with faecal–contaminated drinking water, but other vectors, such as blood, are possible. The aim of this study was to investigate the prevalence of HEV in blood donors and in two groups at high risk for parenteral infections, namely, haemodialysis patients, and children infected with HCV via blood transfusion. Materials and Methods: We investigated the prevalence of anti–HEV in 863 blood donors, 63 haemodialysis patients, and 42 children infected post transfusion with HCV. Results: The prevalence rates were 2.8, 6.3%, and zero, respectively. Conclusions: (1) The incidence of HEV in Spain is similar to that in other Western European countries, and (2) HEV is probably not transmitted parenterally to children.

Intraepithelial lymphocytes and coeliac disease: permanent changes in CD3−/CD7+ and T cell receptor γβ subsets studied by flow cytometry

Permanent changes in intestinal intraepithelial lymphocytes have been observed in coeliac patients. The aim of this investigation was to study small intestinal intraepithelial lymphocytes by using flow cytometry and to evaluate its diagnostic value in coeliac disease. Three‐colour flow cytometry analyses were performed on isolated epithelial cells of 117 intestinal biopsies obtained from 113 children (54 coeliac disease, 4 other enteropathies, 18 Helicobacter pylori associated gastritis and 37 normal controls). A multiple logistic regression model was developed to select the best intraepithelial lymphocytes subset predictor of coeliac disease. Coeliac patients had significant higher levels of T cell receptor yS intraepithelial lymphocytes than control patients (p < 0.01), H. pylori patients (p < 0.01) and other enteropathies (p < 0.05). The density of CD3∼/CD7+ intraepithelial lymphocytes, a intraepithelial lymphocyte subset poorly characterized by immuno‐histochemical methods, was significantly lower in coeliac patients than in the control group (p < 0.01), H. pylori group (p < 0.01) and other enteropathies (p < 0.01). Both changes remained altered independent of the coeliac patients diet. The data were used on a logistic regression analysis in order to calculate sensitivity [94.4%; 95% confidence interval (CI) 83.7‐98.6%], specificity (94.9%; 95% CI 84.9‐98.7%) and likelihood ratio for a positive test 18.5 (95% CI 6.1‐55.8) in the diagnosis of coeliac disease.

Cytokine Production by Intestinal Intraepithelial Lymphocyte Subsets in Celiac Disease

One of the earliest signs of mucosal immune activation in celiac disease (CD) is an increase in the intraepithelial lymphocyte (IEL) count in the small intestinal epithelium. Though most of those IELs express T cell receptor (TcR)-αβ chains, CD is characterized by an increase in TcR-γ δ+ IELs and by the loss of CD3− IELs. There is currently little evidence that these changes in IEL subset distribution are of relevance in the pathogenesis of CD. We aimed to determine the pattern of cytokine production by IEL subsets isolated from duodenal biopsy specimens from control subjects and CD patients at different stages of the disease. We quantified the capacity of IEL subsets to produce IFN-γ, TNF-α, IL-2, IL-4, and IL-10 by intracellular staining by flow cytometry. All IEL subsets studied displayed a type I cytokine profile in both CD and control subjects, with TcR-αβ+ IELs being the main IFN-γ producers. Untreated CD exhibited a trend toward a superior accumulation of IFN-γ per cell but a reduced proportion of INF-γ+ cells in vitro in association with a significantly increased apoptotic rate of IELs. IL-4 was almost undetectable in all cases and IL-10 showed a tendency to increase in treated and “silent” celiac patients. IEL subsets have a similar Th1 profile in controls and CD patients, and the superior in vitro apoptosis of IELs from CD patients may reflect their superior in vivo activation. The induction of IL-10-dependent regulatory Tr1 responses may be of potential clinical significance in this disease and merits further investigation.

Horizontal transmission of hepatitis C virus in households of infected children

Anti-hepatitis C antibodies were measured in 80 household contacts of 27 children infected with hepatitis C virus. Antibodies were demonstrated in only one brother of an infected patient. The results suggest that intrafamilial transmission from infected children may occur, but at a low rate.

Age-related variation of intraepithelial lymphocytes subsets in normal human duodenal mucosa.

The enumeration of intestinal intraepithelial lymphocytes (IELs), and the phenotyping of CD3+CD103+ (TcRαβ, TcRγδ) and CD3−CD103+ IEL subsets constitute useful diagnostic tools for the correct interpretation of the mucosal histology of duodenal/jejunal biopsies in many pathological conditions of the small intestine, particularly celiac disease (CD). This work evaluates the ranges of duodenal IEL counts by flow cytometry in healthy mucosa from pediatric and adult controls, establishing normal reference values for CD3+ TcRγδ and CD3− subsets and their variation with age. Seventy-four pediatric controls and 36 adult controls were identified on the basis of their normal histology from more than 1,000 duodenal diagnostic biopsies performed in Caucasian subjects. Total IEL counts and IEL subsets (“IEL lymphogram”) were analyzed by four-color flow cytometry (FCM). IEL represent 7.7%±0.4 (mean±SE) and 8.5%±0.5 of the cells isolated from the epithelium in the pediatric and adult series, respectively. The upper normal range, considered as the 97 percentile, is 14% in pediatrics and 15% in adults. No significant difference was observed between TcRγδIEL percentages in children (6.9%±0.5 of the total IELs) and adults (6.6%±0.8). However, the density of CD3− IELs is significantly higher (p < 0.001) in the mucosa from controls under 3 years (50.2%±2.6) than in adults (25.5%±2.1). IEL lymphogram by flow cytometry is an easy, quick and reliable analysis performed in one of the biopsy specimens obtained during a diagnostic endoscopy, and confers specificity to the histopathological findings. IEL counts below 14% in children and 15% in adults should be considered within a normal range in the evaluation of duodenal mucosa by FCM. No differences with age were observed with respect to TcRγδIEL, while the CD3− IEL fraction was significantly higher on children under 3 years, with a trend to increase again in the elderly.

Intramural haematoma of the duodenum following endoscopic biopsy: an unusual complication of non-therapeutic endoscopy in children

This case report points out the rarity of intramural duodenal haematomas after intestinal biopsy in children and serves as a reminder to paediatric endoscopists to be aware of this complication. A 4-year-old girl with short stature and increased anti-gliadin antibodies and no history of bleeding disorders underwent upper gastrointestinal endoscopy to obtain a small bowel biopsy. Results of laboratory tests performed prior to the procedure showed a normal platelet count, prothrombin time, and activated partial thromboplastin time. Five duodenal biopsy specimens were obtained using endoscopic grasp forceps with no excessive bleeding being observed. Six hours later the patient presented with abdominal pain and bilious vomiting. Physical examination disclosed normal vital signs and diffuse abdominal tenderness. Levels of haemoglobin, haematocrit, amylase and serum electrolytes were normal. Abdominal ultrasound showed a solid and cystic mass in the second and third duodenal portions. An abdominal CT scan was performed and confirmed the presence of an asymmetrical 4·5 cm mass located within the second and third duodenal portions consistent with an intramural haematoma. The patient followed conservative treatment with total parenteral nutrition. Complete resolution of the haematoma was observed on ultrasound examination on day 19. Two days later the patient was discharged from the hospital. The histological investigation of the intestinal biopsy was normal. Three years later the girl remains asymptomatic and without sequelae. In most cases, intramural duodenal haematoma is an entity caused by abdominal trauma and a complication of therapeutic upper gastrointestinal endoscopy. The development of an intramural duodenal haematoma after endoscopic small bowel biopsy has been reported in nine children, some of them being leukaemic patients or bone marrow transplant recipients [1] and in seven children when using capsule biopsy. It is difficult to state the frequency of intramural duodenal haematoma after biopsy; complications of this procedure develop in less than 2% of cases and are usually mild [4]. In our institution we have performed 2,640 intestinal capsule biopsies and 2,797 upper gastrointestinal endoscopies in children. The only one patient who presented with clinical manifestations of an intramural haematoma is the one described in this report. No other serious complications were observed in our population. The cause of the duodenal haematoma in patients with no underlying disease or therapy-altered coagulation is not clear. Some special features of duodenal anatomy have been involved [3]. Two other facts that could increase the likelihood of haematoma occurrence when biopsies are obtained by endoscopy are the greater number of biopsy specimens obtained and the shear injury associated with obtaining deeper portions of biopsy specimens [5]. In order to prevent this complication, Zinelis et al. [6] have suggested obtaining biopsies from the duodenum extending the forceps no more than 2–3 cm from the endoscope. The clinical presentation of intramural duodenal haematoma is similar in almost all cases and includes severe abdominal pain and vomiting, frequently associated with pancreatitis. Diagnosis is confirmed using imaging techniques with ultrasound, CT scan and upper intestinal series being the most frequently used [2]. Once diagnosis is confirmed and intestinal perforation excluded, conservative treatment with nasogastric C. Camarero (&) AE D. Herrera AE F. Olivares AE B. Roldan Servicio de Pediatria, Hospital Ramon y Cajal, Universidad de Alcala, Madrid, Spain E-mail: ccamareros@yahoo.es Tel.: +34-91-3165443 Fax: +34-91-3368417

Distal duodenum versus duodenal bulb: intraepithelial lymphocytes have something to say in celiac disease diagnosis.

Background and AimAfter clinical screening and the serological test, many patients still require a duodenal biopsy for celiac disease diagnosis. Mild histological lesions, unspecific findings and patchiness are frequent outcomes of this mandatory diagnostic tool, thus complicating clinical decisions.MethodsWe analyzed the lymphoid components [number of total intraepithelial lymphocytes (IELs), TcR-γδ and CD3−IELs] of the duodenal epithelium by flow cytometry in samples obtained from bulb and distal duodenum during upper gastrointestinal endoscopies performed for diagnostic purposes.ResultsIEL counts and IEL subset distribution (IEL lymphogram) remain invariant along duodenal mucosa revealing a specific profile (immunophenotype) that characterizes either a healthy mucosa or a celiac mucosa. The celiac immunophenotype persists regardless of the biopsy’s anatomical location or the corresponding histological findings.ConclusionsWe propose the IEL lymphogram by flow cytometry as an immunological parameter to discern celiac condition from healthy mucosa. This obviates not only misinterpretation of minor histological changes, but also patchiness and the concerns about the location and number of biopsies.

Specificity of IEL profiling in the diagnosis of celiac disease.

TO THE EDITOR: We read with interest the recent article by Jarvinen et al. (1). This article describes the changes in intraepithelial lymphocytes (IEL) subsets observed in a vast series of patients. This excellent article provides data that apparently contradict our previous reports in a couple of aspects that we would like to comment on in order to make clear that our message is the same: IEL phenotyping is useful in the diagnosis of celiac disease (CeD). Jarvinen et al. report a specificity of 88% for the increase of γ δ+ IEL in the diagnosis of CeD (1) and quote our proposal of a diagnostic algorithm for CeD (2) when they claim that “even γ δ+ IEL are not as specific for the disease as recently suggested.” We want to clarify that the figure of 94% sensitivity and 95% specificity that we have reported (2, 3) was for the combined consideration of two different parameters (and not only γ δ+ IEL), namely, the percentage of γ δ+ IEL and of natural killer (NK) IEL, both with respect to total IEL. NK IEL (4) are virtually absent in CeD (5, 6), and their determination increases the efficacy of the test (3). Secondly, Jarvinen et al. report a decrease in the absolute number of γ δ+ IEL in patients under gluten-free diet (GFD) (1), while we reported that γ δ+ IEL are permanently elevated in relation to total IEL, even under GFD (3). This apparent discrepancy can probably be explained by the different methodologies of our respective works. Jarvinen et al. use immunohistochemistry, which provides absolute values, while we use flow cytometry, which provides relative values (6). Despite the permanently elevated relative value of γ δ+ IEL by flow cytometry, compliance with the GFD can also be confirmed by this method if total IEL and αβ+ IEL decrease and NK IEL reappear (3). We believe that flow cytometry is somewhat superior to immunohistochemistry for IEL phenotyping, since it allows for a more accurate quantification and for an easier determination of NK IEL, which are difficult to identify by in situ double staining (CD3− CD7+). But, regardless of the method used, we could not agree more with the authors of this solid report when they state the usefulness of IEL phenotyping in the diagnosis of CeD (1). Francisco León, M.D., Ph.D. Cristina Camarero, M.D., Ph.D. Pablo Eiras, M.D., Ph.D. Garbiñe Roy, M.D., Ph.D.

Anal endosonography in children with chronic constipation

Background. Hypertrophy of the internal anal sphincter may be apparent in some children, but its significance has not yet been determined. Objective. To assess anal endosonographic findings in children with chronic constipation. Materials and methods. We performed anal endosonography in 46 children with chronic constipation and compared the results with values considered normal. Results. We did not find a significant relationship between age and thickness of the internal or external anal sphincters. The clinical response to medical management did not differ between patients with or without sphincter hypertrophy. Conclusions. Although we did not find a significant correlation between sphincter hypertrophy and constipation or age, further studies may clarify its place amongst other techniques which are used in the investigation of anorectal pathology.