Luther Sigurdsson

The spectrum of pediatric eosinophilic esophagitis beyond infancy: a clinical series of 30 children

OBJECTIVES:Eosinophilic esophagitis, previously confused with esophageal inflammation due to gastroesophageal reflux, has recently begun to be distinguished from it. We undertook this analysis of our large series of children with the condition to clarify its spectrum: its presenting symptoms; its relation to allergy, respiratory disease, and reflux; its endoscopic and histological findings; and its diagnosis and therapy.METHODS:We analyzed the details of our clinical series of 30 children with eosinophilic esophagitis, defining it as ≥5 eosinophils per high power field in the distal esophageal epithelium. Retrospective chart review was supplemented by prospective, blinded, duplicate quantitative evaluation of histology specimens, and by telephone contact with some families to clarify subsequent course. Presentation and analysis of the series as a whole is preceded by a case illustrating a typical presentation with dysphagia and recurrent esophageal food impactions.RESULTS:Presenting symptoms encompass vomiting, pain, and dysphagia (some with impactions or strictures). Allergy, particularly food allergy, is an associated finding in most patients, and many have concomitant asthma or other chronic respiratory disease. A subtle granularity with furrows or rings is newly identified as the endoscopic herald of histological eosinophilic esophagitis. Histological characteristics include peripapillary or juxtaluminal eosinophil clustering in certain cases. Association with eosinophilic gastroenteritis occurs, but is not common. Differentiation from gastroesophageal reflux disease is approached by analyzing eosinophil density and response to therapeutic trials. Therapy encompasses dietary elimination and anti-inflammatory pharmacotherapy.CONCLUSION:Awareness of the spectrum of eosinophilic esophagitis should promote optimal diagnosis and treatment of this elusive entity, both in children and in adults.

Eosinophilic esophagitis: strictures, impactions, dysphagia.

Eosinophilic esophagitis, long known to be a feature of acid reflux, has recently been described in patients with food allergies and macroscopically furrowed esophagus. The pathophysiology and optimal management of patients with eosinophilic esophagitis is unclear. We describe our clinical experience related to eosinophilic esophagitis and obstructive symptoms in children and propose etiopathogenesis and management guidelines. Twelve children with obstructive esophageal symptoms (11 male), median age 5 years, and identified to have eosinophilic esophagitis with >5 eosinophils per high-power field (eos/hpf) are reported. Of these, four had strictures, six had impactions, and two had only dysphagia. A diagnostic evaluation included esophagogastroduodenoscopy with biopsies in all and upper gastrointestinal series, IgE, radioallergosorbent tests, and skin tests for food allergies in some cases. Esophageal histology specimens were independently analyzed for eosinophil density by two authors. Four of five children with >20 eos/hpf responded to elimination diets/steroids. The fifth child responded to a fundoplication. Seven children had 5–20 eos/hpf and three of them with no known food allergies responded to antireflux therapy alone. Three others in this group with positive food allergies responded to treatment with elimination diets and/or steroids. The seventh patient in this group was lost to follow-up. In conclusion, on the basis of response to therapy, eosinophilic esophagitis can be subdivided into two groups: those with likely gastroesophageal reflux disease if <20 eos/hpf and no food allergies, and others with allergic eosinophilic esophagitis associated with food allergies and often with >20 eos/hpf.

Intestinal transplantation in children with chronic intestinal pseudo-obstruction

BACKGROUND Children with chronic intestinal pseudo-obstruction (CIPO) often require total parenteral nutrition (TPN) which puts them at risk of liver failure and recurrent line infections. Intestinal transplantation has become a therapeutic option for TPN dependent children with intestinal failure who are failing management with TPN. AIMS To investigate the outcome of children with CIPO referred for intestinal transplantation. METHODS A retrospective review was carried out of records and diagnostic studies from 27 patients with CIPO referred for intestinal transplantation. RESULTS Five children were not listed for transplantation: two because of parental decision, two because of suspicion of Munchausen syndrome by proxy, and one because he tolerated enteral nutrition. Six are still TPN dependent and awaiting transplantation. Eight children died awaiting transplantation. Eight children underwent transplantation. Three died (two months, seven months, and four years after transplant). Five children are alive with a median follow up of 2.6 years (range two months to six years). All transplanted children were able to tolerate full enteral feedings. The postoperative course was complicated by dumping syndrome, Munchausen syndrome by proxy, narcotic withdrawal, and uncovering of achalasia.Conclusion—Intestinal transplantation may be a life saving procedure in children with CIPO. Early referral and thorough pretransplant evaluation are keys to successful transplantation.

Bacteremia after intestinal transplantation in children correlates temporally with rejection or gastrointestinal lymphoproliferative disease.

BACKGROUND Bacteremia occurs frequently after intestinal transplantation (ITx) in children. During our initial experience with this procedure, we noted that bacteremic episodes tended to occur simultaneously with the presence of rejection and/or gastrointestinal (GI) posttransplant lymphoproliferative disease (PTLD). AIM To document the association of bacteremia with rejection and GI PTLD in pediatric ITx recipients. METHODS Retrospective analysis of all medical records from 62 children who underwent ITx between July 1990 and January 1998 at Childrens Hospital of Pittsburgh. A bacteremic episode was defined as two positive blood cultures from different sites at the same time or from the same site at different times. Rejection and PTLD were defined using previously published criteria. RESULTS A total of 39/62 ITx recipients had 133 blood stream infections (2.1 episodes/patient) including 121 episodes of bacteremia and 12 of fungemia. Enteric organisms were the most frequently recovered pathogens (Gram negative rods, n=76; enterococci, n=36). Enteric organisms were recovered as a single organism (n=57), with another enteric bacteria (n=23), or with coagulase negative staphylococci (CONS) (n=24). CONS were recovered as a single organism on 21 occasions. An obvious source of bacteremia was not found for 115/121 episodes. Endoscopy was performed for 107 of the 115 bacteremia episodes; an abnormal histology was identified in 74 revealing rejection (n=36), GI PTLD (n=21), or both (n=17). When endoscopy showed GI pathology, enteric organisms alone or in combination with CONS were recovered on 63/107 occasions, although CONS were recovered alone only 11 times. CONCLUSIONS Bacteremia accompanies GI rejection and intestinal PTLD in ITx recipients. Endoscopy should be performed to inspect the allograft when bacteremia occurs without an obvious source in these patients. This is especially true for patients with bacteremia due to enteric organisms.

Anatomic variability of Rejection in intestinal allografts after pediatric intestinal transplantation

BACKGROUND Rejection of the allograft is a major contributor to morbidity and mortality in children who undergo a small intestinal transplant. Operational definitions for histologic rejection have been established, but little is known about the anatomic variability of the histologic abnormalities. STUDY DESIGN Biopsy reports were reviewed retrospectively from more than 1200 endoscopies performed on the 41 children who received intestinal transplantation between 1990 and 1995. RESULTS Biopsies were performed in the proximal jejunum and distal ileum allograft simultaneously on 248 occasions. In 168 biopsies, neither site was histologically abnormal; in 80, rejection was found. In 42, both regions were abnormal; however, in 17 only the jejunum was involved and in 21 the rejection exclusively involved the ileum. Among children whose allograft included colon, rejection was absent in colon and ileum in 34 biopsies, involved both in 6, involved ileum but not colon in another 6 and involved colon but not ileum in only one. When the allograft included stomach, rejection was absent in the stomach and jejunum 39 times, involved both sites 8 times, involved jejunum and not the stomach 10 times, but involved the stomach and not jejunum only once. Endoscopic appearance correctly predicted histologic rejection 63% of the time. CONCLUSION Anatomic variability may exist in the rejection process. Sampling the jejunum and ileum seems to have similar sensitivity in detecting rejection, whereas sampling stomach and the colon is less sensitive. When allograft rejection is suspected on clinical grounds, sampling more than one area of the graft may be necessary for histologic confirmation.