Improved patient safety with a simplified operating room to pediatric intensive care unit handover tool (PATHQS).

Introduction: Patient handover is a crucial transition requiring a high level of coordination and communication. In the BC Children's Hospital (BCCH) pediatric intensive care unit (PICU), 10 adverse events stemming from issues that should have been addressed at the operating room (OR) to PICU handover were reported into the patient safety learning system (PSLS) within 1 year. We aimed to undertake a quality improvement project to increase adherence to a standardized OR to PICU handover process to 100% within a 6-month time frame. In doing so, the secondary aim was to reduce adverse events by 50% within the same 6-month period. Methods: The model for improvement and a Plan, Do, Study, Act method of quality improvement was used in this project. The adverse events were reviewed to identify root causes. The findings were reviewed by a multidisciplinary inter-departmental group comprised of members from surgery, anesthesia, and intensive care. Issues were batched into themes to address the most problematic parts of handover that were contributing to risk. Intervention: A bedside education campaign was initiated to familiarize the team with an existing handover standard. The project team then formulated a new simplified visual handover tool with the mnemonic "PATHQS" where each letter denoted a step addressing a theme that had been noted in the pre-intervention work as contributing to adverse events. Results: Adherence to standardized handover at 6 months improved from 69% to 92%. This improvement was sustained at 12 months and 3 years after the introduction of PATHQS. In addition, there were zero PSLS events relating to handover at 6 and 12 months, with only one filed by 36 months. Notably, staff self-reporting of safety concerns during handover reduced from 69% to 13% at 6 months and 0% at 3 years. The PATHQS tool created in this work also spread to six other units within the hospital as well as to one adult teaching hospital. Conclusion: A simplified handover tool built collaboratively between departments can improve the quality and adherence of OR to PICU handover and improve patient safety. Simplification makes it adaptable and applicable in many different healthcare settings.

Patch repair is an independent predictor of morbidity and mortality in congenital diaphragmatic hernia.

BACKGROUND: Infants with congenital diaphragmatic hernia (CDH) have variable outcomes. There is a considerable potential benefit in being able to predict perinatally, which infants have severe hypoplasia and are thus more likely to die or survive with significant morbidity. We examine the relationship between a need for patch repair of CDH (PR) and outcome, using a national database. METHODS: Baseline characteristics of patients undergoing PR or non-patch repair (NPR) were compared. Multivariate analysis was performed to determine the association of PR with mortality and morbidity independent of other known predictors. RESULTS: Baseline characteristics of PR and NPR infants were similar although those infants with PR had higher SNAP-II scores. PR was an independent predictor of mortality with an odds ratio of 17.1 (95%CI 2.0-149.2) and was independently associated with secondary outcome measures of morbidity, including the need for oxygen at discharge and the duration of ventilation. CONCLUSIONS: Infants requiring PR have significantly higher mortality and suffer greater morbidity than those undergoing NPR. This association is independent of other known predictors of mortality. Identifying prenatal features associated with this high risk group would be of great clinical value.

Does pre-biopsy contrast enema delay the diagnosis of long segment Hirschsprung's disease?

INTRODUCTION: The diagnosis of long segment Hirschsprung's disease (LSHD) is frequently delayed. Our purpose was to: 1) summarize contrast enema (CE) findings in patients with LSHD, and 2) evaluate the utility of CE by comparing LSHD patients managed with/without pre-biopsy CE. METHODS: All LSHD cases (transition zone [TZ] proximal to the splenic flexure) treated between 1984 and 2009 were stratified according to whether a pre-biopsy CE was done (Group 1) or not (Group 2). CE were reviewed by a single pediatric radiologist, and the original reports were categorized as "helpful", "inconclusive" or "misleading". Group comparisons included elapsed days from admission to diagnostic rectal biopsy/first operation and initial hospitalization length of stay (LOS). RESULTS: 29 patients (16 in Group 1; 13 in Group 2) were identified. CE review revealed TZ in 7/16 (44%); and of these, 6 (86%) underestimated the actual aganglionic segment length. 6/16 (38%) original CE reports were "misleading". Overall, Group 1 patients experienced a significant delay in time to biopsy (p=0.047), first operation (p=0.005), and showed a trend towards prolonged LOS. CONCLUSIONS: Pre-biopsy CE offers little to the diagnosis of LSHD and may contribute to diagnosis/treatment delays. Even if a TZ is recognized in biopsy proven HD, the predicted aganglionic segment length should not guide the operative planning.

Impact of target blood gases on outcome in congenital diaphragmatic hernia (CDH).

INTRODUCTION: Neonatal intensive care unit (NICU) stabilization strategies which normalize physiology according to predetermined blood gas targets may contribute to observed improved survival rates of patients with CDH. The purpose of our study was to compare risk-adjusted outcomes of CDH patients managed with or without blood gas targets established at NICU admission. METHODS: Cases were collected from a national CDH network between May 2005 and November 2007. On NICU admission, the responsible neonatologist was asked to establish target ranges for pH, pCO (2), pO (2), and pre/post-ductal O (2) saturation. The outcomes analyzed were mortality, need for ECMO, days of mechanical ventilation/supplemental oxygen, and length of stay. RESULTS: Of 147 CDH infants, 63 had admission blood gas targets. Severity of illness and gestational age in both groups were comparable (SNAP-II score). Infants with blood gas targets had a significantly lower mortality than those without (Hazard ratio 0.27, p=0.006). CONCLUSIONS: Blood gas targets for the management of infants with CDH are associated with improved survival. Although the willingness to create and use stabilization targets to guide early NICU care may be a surrogate for other factors (experience, staffing, lack of interest), it is clearly associated with improved survival in CDH.

Perinatal predictors of outcome in gastroschisis.

OBJECTIVE: To identify perinatal risk variables predictive of outcome in gastroschisis. STUDY DESIGN: Gastroschisis cases were collected over a 3-year period from a national database. Risk variables evaluated included gestational age (GA), birth weight, time of birth, admission illness severity (score for neonatal acute physiology-II, SNAP-II) score, and abdominal closure type. Mortality and survival outcomes were analyzed. Multivariate analyses were performed. RESULT: In all, 239 infants were survived (96%). SNAP-II score predicted mortality (relative risk (RR)=1.07, 95% confidence interval (CI)=1.0 to 1.1). Length of hospital stay (LOS) and ventilation days were predicted by GA and by SNAP-II score. SNAP-II score predicted total parenteral nutrition (TPN) days (P=0.006). Severe cholestasis (conjugated bilirubin of >10 mg per 100 ml) was inversely related to GA (RR=0.77, 95% CI=0.61 to 0.97) and directly to categorical SNAP-II score (RR=3.4, 95% CI=1.2 to 10.1). Urgent closure predicted fewer TPN days (P=0.003) and shorter LOS (P=0.0002). CONCLUSION: SNAP-II scores significantly predict mortality and survival outcomes. Urgent closure favors fewer TPN days and shorter LOS. Our data refute routine preterm delivery in gastroschisis.

Abdominal compartment syndrome complicating paediatric extracorporeal life support: diagnostic and therapeutic challenges.

We report three paediatric cases, and summarise the reported experience in two others, with cardiorespiratory failure requiring extracorporeal life support for which supportive pump flows could not be maintained due to abdominal compartment syndrome. In two of our patients, the mechanism of abdominal compartment syndrome was massive intra-abdominal fluid extravasation secondary to sepsis, while in the third, the mechanism was post-traumatic intra-abdominal haemorrhage. Although all three children eventually died, decompressive laparotomy and arrest of haemorrhage in the trauma patient restored venous return and enabled technically adequate extracorporeal life support. In two previously reported cases of sepsis with massive fluid resuscitation resulting in abdominal compartment syndrome, one patient died without attempted decompression, while the other patient survived after peritoneal catheter placement restored venous return. Once correctable causes of inadequate venous cannula drainage have been excluded, abdominal compartment syndrome should be considered in any patient on extracorporeal life support with a taut abdomen and reduced venous return. If abdominal compartment syndrome can be proven or is strongly suspected, there may be a role for selective decompressive laparotomy.

Correlation of transcutaneous hepatic near-infrared spectroscopy readings with liver surface readings and perfusion parameters in a piglet endotoxemic shock model.

PURPOSE: To determine whether transcutaneous liver near-infrared spectrophotometry (NIRS) measurements correlate with NIRS measurements taken directly from the liver surface, and invasive blood flow measurements. PROCEDURE: Laparotomy was performed in 12 Yorkshire piglets, and ultrasound blood flow probes were placed on the hepatic artery and portal vein. Intravascular catheters were inserted into the hepatic and portal veins for intermittent blood sampling, and a pulmonary artery catheter was inserted via the jugular vein for cardiac output measurements. NIRS optodes were placed on skin overlying the liver and directly across the right hepatic lobe. Endotoxemic shock was induced by continuous infusion of Escherichia coli lipopolysaccharide O55:B5. Pearson's correlations were calculated between the NIRS readings and the perfusion parameters. FINDINGS: After endotoxemic shock induction, liver blood flow, and oxygen delivery decreased significantly. There were statistically significant correlations between the transcutaneous and liver-surface NIRS readings for oxyhemoglobin, deoxyhemoglobin, and cytochrome c oxidase concentrations. There were similar significant correlations of the transcutaneous oxyhemoglobin with both the mixed venous and hepatic vein saturation, and mixed venous and hepatic vein lactate. CONCLUSIONS: Transcutaneous NIRS readings of the liver, in an endotoxemic shock model, correlate with NIRS readings taking directly from the liver surface, as well as with global and specific organ-perfusion parameters.

Primary reversed gastric tube reconstruction in long gap esophageal atresia.

BACKGROUND/PURPOSE: Esophageal reconstruction in long-gap esophageal atresia (EA) poses a technical challenge with several surgical options. The purpose of this study was to review the authors' experience with the reversed gastric tube (RGT) in esophageal reconstruction. METHODS: This series describes 7 babies with pure EA treated at 2 centers between 1989 and 2001. Data, gathered by retrospective chart review, included clinical details of the esophageal and associated malformations, technique and timing of repair, early and late complications, and long-term follow-up. Institutional review board (IRB) approval of this study has been obtained. RESULTS: Seven babies were included. Associated malformations were present in 4: trisomy 21 in 2 and imperforate anus in 2. After gastrostomy tube placement, patients were treated with gastrostomy tube feedings and continuous upper pouch suction. Median gap length was 5.5 vertebral segments (range, 3 to 9). RGT with a posterior mediastinal esophagogastric anastomosis was performed at median age of 62 days (range, 38 to 131). There were no anastomotic leaks. Three patients had strictures, one required resection. Exclusive oral nourishment was achieved in 5 patients by 6 months of age. At last follow-up (mean, 4.5 years), 6 patients were receiving oral nutrition exclusively, and all were maintaining growth curves. CONCLUSIONS: In long gap EA, early esophageal reconstruction using an RGT can be performed with minimal morbidity and promising short-term results.

The role of prospective randomized clinical trials in pediatric surgery: state of the art?

PURPOSE: This study sought to determine the role of randomized controlled trials (RCT) in the evolution of pediatric surgical practice. METHODS: The authors used a computer-assisted literature search to identify all clinical trials related to pediatric surgery published in the English-language literature from 1966 through 1999. Each article was reviewed in detail for purpose, content, conduct, and quality of the trial. The authors assessed quality with a previously validated instrument (Chalmers Qualitative Assessment). RESULTS: The authors identified 134 RCTs related to pediatric surgery over the past 33 years. This accounts for 0.17% of 80,377 articles published in the field. The areas of surgery studied were analgesia 65 (49%), antibiotics 17 (13%), extracorporeal membrane oxygenation (ECMO) 9 (7%), gastrointestinal, burns, oncology, minimally invasive surgery, vascular access, congenital anomalies, and trauma (each <5%). Only 16 (12%) trials compared 2 surgical therapies, 9 (7%) compared a medical versus a surgical therapy, and 109 (81%) compared 2 medical therapies in surgical patients. Fourteen (10%) RCTs were funded by peer-reviewed agencies. Only 17 (13%) RCTs included a biostatistician as an author or a consultant. Trial design included calculation of sample size and statistical power in 21 (16%) RCTs. Method of randomization was reported in only 51 (38%). The test statistic and observed probability value was reported in 15 (11%). CONCLUSIONS: Clinical trials are used infrequently to answer questions related to pediatric surgery. When RCTs are utilized, they often suffer from poor trial design, inadequate statistical analysis, and incomplete reporting. Pediatric surgery could benefit from increased expertise, funding, and participation in clinical trials.

Fetal therapy with rhIGF-1 in a rabbit model of intrauterine growth retardation.

BACKGROUND: Intrauterine growth retardation (IUGR) may, in part, be due to a deficiency of insulin-like growth factor-1 (IGF-1). The objectives of this study were to determine the relationship between fetal serum IGF-1 levels and fetal and placental size in a rabbit model of IUGR and to compare two techniques of selective, exogenous IGF-1 administration (transamniotic and branch uterine arterial catheter infusion) to growth-retarded fetuses in utero. MATERIALS AND METHODS: Pregnant rabbits (n = 6) had their fetuses harvested near term (31 days) for fetal and placental weighing and serum collection. Growth-retarded fetuses were selectively infused for 7 days with recombinant human IGF-1 (rhIGF-1; 1,440 ng/day) either through a transamniotic catheter (n = 8) or via an adjacent uterine arterial branch catheter (n = 6). Opposite horn runts were sham catheterized, but not infused. At term, the fetal runt pairs and their placentas were harvested and weighed, and their serum was collected. The correlation between fetal and placental weight and endogenous serum IGF-1 was calculated (Pearson coefficient, r), while paired t-tests were used to compare the means between the IGF-1-infused and control groups. RESULTS: There was a significant correlation between fetal (r = 0.4230; P = 0.022) and placental weight (r = 0.4166; P = 0.025) and endogenous serum levels of IGF-1. Transamniotic infusion of rhIGF-1 was associated with an increase in serum IGF-1 level (254 +/- 79 vs 351 +/- 101 ng/ml, P = 0.04) and placental weight (5.4 +/- 2.3 vs 7.1 +/- 3.2 g, P = 0.005), and with a trend toward increased fetal weight between matched fetal runt pairs. Fetal mortality in the uterine arterial catheterized group was 76%, and there was no significant difference in fetal or placental weight or IGF-1 levels between infused and noninfused survivors. CONCLUSIONS: Endogenous fetal serum levels correlate with fetal and placental size in the rabbit IUGR model. Transamniotic administration of rhIGF-1 significantly increases serum IGF-1 levels and placental weight of fetal runts, while uterine vessel catheterization results in prohibitive fetal mortality and does not increase fetal or placental growth or IGF-1 levels.