Change of Outcomes in Pediatric Intestinal Failure: Use of Time-Series Analysis to Assess the Evolution of an Intestinal Rehabilitation Program

• Published in: Journal of the American College of Surgeons Vol. 222; no. 6; pp. 1180 - 1188.e3
• Main Authors: Oliveira, Carol, MD, PhD, de Silva, Nicole T., MSc, Stanojevic, Sanja, PhD, Avitzur, Yaron, MD, Bayoumi, Ahmed M., MD, MSc, Ungar, Wendy J., PhD, Hoch, Jeffrey S., PhD, Wales, Paul W., MD, MSc, FACS
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.06.2016
• Subjects: Combined Modality Therapy; Fat Emulsions, Intravenous - therapeutic use; Fatty Acids, Omega-3 - therapeutic use; Female; Follow-Up Studies; Humans; Infant; Infant, Newborn; Linear Models; Male; Multivariate Analysis; Retrospective Studies; Short Bowel Syndrome - mortality; Short Bowel Syndrome - rehabilitation; Surgery; Treatment Outcome; intestinal failure; autoregressive integrative moving average models; IRP; catheter-related bloodstream infections; central venous catheter; IQR; interquartile range; CVC; intestinal rehabilitation program; parenteral nutrition; ARIMA; IF; PN; CRBSI
• ISSN: 1072-7515; 1879-1190
• DOI: 10.1016/j.jamcollsurg.2016.03.007

Background The clinical picture of pediatric intestinal failure has changed over the past 15 years, while effectiveness evolving treatment options remains unclear. This study explored evolution in care and quantified independent effects of new treatment options. Study Design Consecutive patients (n = 196) with neonatal or infantile intestinal failure, born between July 1996 and December 2011, were derived from an intestinal rehabilitation program (IRP) patient registry. Change over time was analyzed using multivariable Box-Jenkins method-based autoregressive integrative moving average models (ARIMA), robust linear regression, and nonparametric trend analysis. Four systematically introduced treatment options (IRP, serial transverse enteroplasty, omega-3 lipid emulsions, and ethanol locks) were evaluated. Analyses were adjusted for patient characteristics and disease severity. The primary outcome was disease-specific mortality from liver failure and sepsis. Secondary outcomes included parenteral nutrition weaning, transplantations, catheter complications, and liver disease. Results Patient characteristics remained unchanged over time, except for decreasing small bowel length (−0.5%/quarter; 95% CI −0.85, −0.16) and ICU admission time (−0.6 days/quarter; 95% CI −1.03, −0.18). Disease-specific mortality diminished significantly over time (−0.02 deaths/quarter; 95% CI −0.03, −0.01) by IRP and omega-3 lipids introduction (−0.6 deaths/quarter each, 95% CI −1.23, −0.02 and −0.77, −0.45, respectively). Serial transverse enteroplasty and ethanol locks had no significant impact. Parenteral nutrition weaning and transplantations remained unchanged, while catheter sepsis and complication rates decreased by 0.3 episodes/1,000 catheter-days each (95% CI −0.43, −0.2 and −0.45, −0.24, respectively). Conclusions Introduction of IRP and omega-3 lipids independently decreased disease-specific mortality. For the first time, time series analysis was applied to evaluate effectiveness of treatment options in intestinal rehabilitation.