Fouad Youssef

Gastroschisis outcomes in North America: a comparison of Canada and the United States

BACKGROUND Care of infants with gastroschisis is centralized in Canada and noncentralized in the United States. We conducted an outcomes comparison between the two countries and analyzed the determinants of such outcomes. METHODS Inpatient mortality and hospital stay of gastroschisis patients from the Canadian Pediatric Surgery Network prospective clinical database for the period 2005-2013 were compared with those from the US Kids Inpatient Database for the period 2003-2012. Potential outcome determinants were analyzed using univariate and multivariate analyses. RESULTS A comparison was made between 695 Canadian patients and 5216 American patients. Complex gastroschisis was found in 16.0% and 13.7% of patients in Canada and the US, respectively; P=0.11. Canada had less premature births, more normal birth weight (BW) infants, less cesarean section deliveries, and more inborn patients compared to the US. For simple gastroschisis, Canadian mortality was lower (1.4% vs. 3.4%; P=.008) and hospital stay was longer (45±38 vs. 41±32days; P=.04). US mortality correlated strongly with low BW (P=.002) and marginally with cesarean section delivery (P=.08). A longer Canadian hospital stay was associated with lower gestational age (P=0.01) and western region (P=0.04), while a longer American hospital stay was associated with medium neonatal intensive care unit gastroschisis volume (P=.03), low socioeconomic status (P=.06), low BW (P=0.06), and public insurance (P=0.07). Outcomes for complex gastroschisis did not differ between Canada and the US. CONCLUSIONS Mortality for simple gastroschisis is higher in the US than in Canada, whereas no outcome differences exist for complex gastroschisis. Outcome determinants are different between the 2 countries.

A risk-stratified comparison of fascial versus flap closure techniques on the early outcomes of infants with gastroschisis

BACKGROUND While fascial closure is traditionally used in gastroschisis (GS), flap closure (skin or umbilical cord) has gained popularity. We evaluated early outcomes and complications of the two techniques. METHODS A national, population-based gastroschisis data registry was analyzed from 2005 to 2011. We compared fascial to flap closures and stratified patients into low or high-risk groups using the Gastroschisis Prognostic Score (GPS), a validated marker of post-natal bowel injury. Demographic and outcome data, including length of stay, complications, and markers of resource utilization were analyzed using Fishers exact and Students t-tests for categorical and continuous variables, respectively (p<0.05 significant). RESULTS The analyzed dataset included 436 fascial closures (344 [78.8%] low-risk, 92 high-risk) and 129 flap closures (112 [86.7%] low-risk, 17 high-risk; p=0.06). Demographics and birth weight did not differ between groups. In patients with low GPS, flap closure demonstrated significant decreases in resource utilization and failure of closure, without differences in complication rates. Analysis of high-risk patients revealed no statistically significant differences in outcome. CONCLUSION Flap closure was not associated with an increase in patient morbidity and seemed suitable as a definitive closure method for gastroschisis patients irrespective of disease severity. Furthermore, flap closure reduced several markers of resource utilization in patients with low-risk disease.

The correlation between the time spent in utero and the severity of bowel matting in newborns with gastroschisis

BACKGROUND Optimal timing of delivery in fetuses with gastroschisis (GS) is unknown. Some favor early induced delivery to prevent bowel injury. This study evaluates the correlation between bowel injury and the gestational age at birth using the Gastroschisis Prognostic Score (GPS). METHODS A national database was analyzed from 2005 to 2013. Patients were pooled based on their gestational age at birth. The mean GPS and % of patients with severe bowel matting were tabulated for each week in utero. Regression modeling was used to evaluate the relationship between the dependent (severe matting and GPS) and independent (gestational age) variables and the R(2) coefficient of determination was derived to evaluate model strength. Additional factors influencing the timing of delivery were evaluated. RESULTS Of 780 cases, 88 were excluded because of missing data. A linear relationship is seen between increasing gestational age and decreasing bowel matting (R(2)=0.66) and GPS (R(2)=0.72). For every week in utero, the % of patients with severe matting decreases by 3.6%. CONCLUSION Early induced delivery simply to protect the bowel from ongoing in utero damage appears unfounded and should be reserved for evidence of closing gastroschisis or traditional obstetrical/fetal indications.

Flap versus fascial closure for gastroschisis: a systematic review and meta-analysis

BACKGROUND Flap closure represents an alternative to fascial closure for gastroschisis. We performed a systematic review and meta-analysis of outcomes comparing these techniques. METHODS A registered systematic review ( PROSPERO CRD42015016745) of comparative studies was performed, querying multiple databases without language or date restrictions. Gray literature was sought. Outcomes analyzed included: mortality, ventilation days, feeding parameters, length of stay (LOS), wound infection, resource utilization, and umbilical hernia incidence. Multiple reviewers independently assessed study eligibility and literature quality. Meta-analysis of outcomes was performed where appropriate (Revman 5.2). RESULTS Twelve studies met inclusion criteria, of which three were multi-institutional. Quality assessment revealed unbiased patient selection and exposure, but group comparability was suboptimal in four studies. Overall, 1124 patients were evaluated, of which 350 underwent flap closure (210 immediately; 140 post-silo). Meta-analysis revealed no significant differences in mortality, LOS, or feeding parameters between groups. Flap patients had less wound infections (OR 0.40 [95%CI 0.22-0.74], P=0.003). While flap patients had an increased risk of umbilical hernia, they were less likely to undergo repair (19% vs. 41%; P=0.01). CONCLUSIONS Flap closure has equivalent or superior outcomes to fascial closure for patients with gastroschisis. Given potential advantages of bedside closure and reduced sedation requirements, flap closure may represent the preferred closure strategy.

Determinants of outcomes in patients with simple gastroschisis

PURPOSE We analyzed the determinants of outcomes in simple gastroschisis (GS) not complicated by intestinal atresia, perforation, or necrosis. METHODS All simple GS patients enrolled in a national prospective registry from 2005 to 2013 were studied. Patients below the median for total parenteral nutrition (TPN) duration (26days) and hospital stay (34days) were compared to those above. Univariate and multivariate logistic and linear regression analyses were employed using maternal, patient, postnatal, and treatment variables. RESULTS Of 700 patients with simple GS, representing 76.8% of all GS patients, 690 (98.6%) survived. TPN was used in 352 (51.6%) and 330 (48.4%) patients for ≤26 and >26days, respectively. Hospital stay for 356 (51.9%) and 330 (48.1%) infants was ≤34 and >34days, respectively. Univariate analysis revealed significant differences in several patient, treatment, and postnatal factors. On multivariate analysis, prenatal sonographic bowel dilation, older age at closure, necrotizing enterocolitis, longer mechanical ventilation, and central-line associated blood stream infection (CLABSI) were independently associated with longer TPN duration and hospital stay, with CLABSI being the strongest predictor. CONCLUSIONS Prenatal bowel dilation is associated with increased morbidity in simple GS. CLABSI is the strongest predictor of outcomes. Bowel matting is not an independent risk factor. LEVEL OF EVIDENCE 2c.

Standardization of care for pediatric perforated appendicitis improves outcomes

BACKGROUND The treatment of perforated appendicitis in children is characterized by significant variability in care, morbidity, resource utilization, and outcomes. We prospectively studied how minimization of care variability affects outcomes. METHODS A clinical pathway for perforated appendicitis, in use for three decades, was further standardized in May 2015 by initiation of a disease severity classification, refinement of discharge criteria, standardization of the operation, and establishment of criteria for use of postoperative total parenteral nutrition, imaging, and invasive procedures. Prospective evaluation of all children treated for 20months on the new fully standardized protocol was conducted and compared to a retrospective cohort treated over 58months prior to standardization. Differences between outcomes before and after standardization were analyzed using regression analysis techniques to adjust for disease severity. RESULTS Median follow-up time post discharge was 25 and 14days in the post- and prestandardization groups, respectively. Standardization significantly reduced postoperative abscess (9.8% vs. 17.4%, p=0.001) and hospital stay (p=0.002). Standardization reduced the odds of developing a postoperative abscess by four fold. CONCLUSION Minimizing variability of care at all points in the treatment of perforated appendicitis significantly improves outcomes. TYPE OF STUDY Prospective Cohort Study. LEVEL OF EVIDENCE Level II.

Risk stratification in pediatric perforated appendicitis: Prospective correlation with outcomes and resource utilization

PURPOSE Despite a wide spectrum of severity, perforated appendicitis in children is typically considered a single entity in outcomes studies. We performed a prospective cohort study to define a risk stratification system that correlates with outcomes and resource utilization. METHODS A prospective study was conducted of all children operated for perforated appendicitis between May 2015 and December 2016 at a tertiary free-standing university childrens hospital. Surgical findings were classified into one of four grades of perforation: I. localized or contained perforation, II. Contained abscess with no generalized peritonitis, III. Generalized peritonitis with no dominant abscess, IV. Generalized peritonitis with one or more dominant abscesses. All patients were treated on a clinical pathway that involved all points of care from admission to final follow-up. Outcomes and resource utilization measures were analyzed using Fishers exact test, Kruskal-Wallis test, One-way ANOVA, and logistic regression. RESULTS During the study period, 122 patients completed treatment, and 100% had documented follow-up at a median of 25days after operation. Grades of perforation were: I, 20.5%; II, 37.7%; III, 10.7%; IV, 31.1%. Postoperative abscesses occurred in 12 (9.8%) of patients, almost exclusively in Grade IV perforations. Hospital stay, duration of antibiotics, TPN utilization, and the incidence of postoperative imaging significantly increased with increasing grade of perforation. CONCLUSION Outcomes and resource utilization strongly correlate with increasing grade of perforated appendicitis. Postoperative abscesses, additional imaging, and additional invasive procedures occur disproportionately in patients who present with diffuse peritonitis and abscess formation. The current stratification allows risk-adjusted outcome reporting and appropriate assignment of resource burden. LEVEL OF EVIDENCE I (Prognosis Study).

Loop versus divided colostomy for the management of anorectal malformations: a systematic review and meta-analysis

BACKGROUND The ideal colostomy type for patients with anorectal malformations (ARM) is undetermined. We performed a systematic review and meta-analysis of short-term complications comparing loop and divided colostomies. METHODS After review registration (PROSPERO: CRD42016036481), multiple databases were searched for comparative studies without language or date restrictions. Gray literature was sought. Complications investigated included stomal prolapse/hernia/retraction, wound infections, and urinary tract infections (UTIs). Two reviewers independently assessed study eligibility and the quality of included studies. Meta-analysis of selected complications was performed using Revman 5.3, with p<0.05 considered significant. RESULTS Twenty-six studies were included, and four were multi-institutional. Reporting standards were highly variable. Studies scored between 6 and 9 of possible nine stars on the NOS. Overall, 3866 neonates with ARM were incorporated, in which 2241 loop colostomies and 1994 divided colostomies were reported. Of 10 studies reporting short-term complications, the overall rate was 27%. Meta-analysis demonstrated no significant difference in the incidence of UTIs, (OR: 2.55 [0.76, 8.58], p=0.12), while loop colostomies had a significantly higher prolapse rate (See figure). No publication bias was noted. CONCLUSIONS A colostomy for patients with an ARM is a source of considerable morbidity. Divided colostomies reduce the risk of subsequent prolapse and may represent the preferred approach. LEVEL OF EVIDENCE 3A.

Appropriate use of total parenteral nutrition in children with perforated appendicitis

BACKGROUND Total parenteral nutrition (TPN) is often used in children with perforated appendicitis, despite the absence of clear indications. We assessed the validity of specific clinical indications for initiation of TPN in this patient cohort. METHODS Data were gathered prospectively on duration of nil per os (NPO) status and TPN use in a cohort of children treated under a perforated appendicitis protocol during a 19-month period. TPN was started in the immediate postoperative period in patients who had generalized peritonitis and severe intestinal dilatation at operation, or later per the discretion of the attending surgeon. At discharge, TPN was considered to have been used appropriately, according to consensus guidelines, if the patient was NPO≥7days or received TPN≥5days. RESULTS During the study period, TPN was initiated in 31 (25.4%) of 122 patients operated for perforated appendicitis. Sixteen (51.6%) received TPN per operative finding indications and 15 (48.4%) for prolonged ileus. The operative indications demonstrated 47% sensitivity, 86% specificity, a positive predictive value (PPV) of 35%, and a negative predictive value (NPV) of 91%, when adherence to TPN consensus guidelines was considered the gold standard. CONCLUSION Patients without severe intestinal dilatation and generalized peritonitis at operation should not be placed on TPN in the immediate postoperative period. Refinement of selection criteria is necessary to further decrease inappropriate TPN use in children with perforated appendicitis. TYPE OF STUDY Diagnostic Test. LEVEL OF STUDY II.

The utility of magnetic resonance imaging in the diagnosis and management of pediatric benign ovarian lesions

BACKGROUND The utility of magnetic resonance imaging (MRI) in the diagnosis and management of pediatric ovarian lesions has not been well defined. METHODS A retrospective review of all girls who underwent MRI evaluation of ovarian masses during the period 2009-2015 was performed. The accuracy of MRI was evaluated by comparing results with surgical findings, pathology reports, and subsequent imaging. The influence of the MRI on the treatment plan was specifically explored. RESULTS Eighteen girls, 12-17years of age, underwent 27 MRIs, subsequent to ultrasound identification of ovarian lesions. Of 9 neoplastic lesions diagnosed on MRI, 8 (89%) were confirmed by surgical and pathological findings. Of 18 functional lesions, 17 (94.4%) were confirmed pathologically or by resolution on subsequent imaging. Twenty MRI exams (74%) directly influenced the treatment plan, by leading to appropriate operative intervention in 9 and appropriate observation in 11. The extent of ovarian resection was guided by MRI findings in 8 of 9 (89%) neoplastic lesions. For characterizing lesions as neoplastic, the sensitivity, specificity, negative predictive value, positive predictive value, and accuracy of MRI were 89%, 94%, 94%, 89%, and 93% respectively. CONCLUSIONS MRI can differentiate functional from neoplastic pediatric ovarian masses, and guide ovarian resection in appropriate cases. LEVEL OF STUDY II.