Risk Factors for Development of Pneumatosis Intestinalis after Pediatric Hematopoietic Stem Cell Transplantation: A...

The significance of pneumatosis intestinalis (PI) in pediatric patients following hematopoietic stem cell transplantation (HSCT) is poorly understood. A knowledge gap remains with respect to the etiology, risk factors, and evidence-based treatment of these patients. As a result, management is frequently based on each center's clinical practice, without standardization across treatment centers. In this single-center trial, we aimed to validate both previously proposed and additional risk factors for the development of PI and to examine our management and outcomes for these patients. We performed a retrospective case-control study examining risk factors for the development of PI in pediatric HSCT patients at a single tertiary referral children's hospital. We used univariate and multivariable conditional logistic regression analysis to explore differences in pharmacologic and other transplantation-specific risk factors. Between 2012 and 2019, PI was diagnosed in 212 patients at our pediatric hospital, of whom 42 were HSCT recipients. The majority of patients (88%; n = 37 of 42) with PI were diagnosed by X-ray. Eighteen patients (43%) were asymptomatic and diagnosed incidentally after imaging was obtained for standard post-transplantation surveillance or other nonrelated indications. All patients with PI were hospitalized and placed on strict bowel rest while receiving parenteral nutrition and antibiotics. Recurrence of PI occurred in 4 patients (10%) following their initial diagnosis. Increased doses of steroid exposure within 30 days of PI diagnosis (odds ratio [OR], 5.7; 95% confidence interval [CI], 2.1 to 15.3; P = .0006), presence of grade II-IV gastrointestinal acute graft-versus-host disease (GVHD) (OR, 5.3; 95% CI, 1.0 to 28.1; P = .05), and receipt of >50% of total daily nutrition by nasogastric (NG) tube feeds (OR, 22.0; 95% CI, 1.3 to 370.2; P = .03) were identified as independent risk factors for the development of PI. Intensity of the conditioning regimen, exposure to total body irradiation, stem cell source, donor type, HLA matching, use of mycophenolate mofetil, and presence of bacterial or viral infection at the time of PI diagnosis were not demonstrably associated with the development of PI in our study. We conclude that development of asymptomatic PI is a benign condition following HSCT, and that the risk for PI is increased in patients with gastrointestinal GVHD, patients receiving steroid therapy, and patients relying on supplemental NG tube feeds for at least one-half of their total daily nutrition.