Geoffrey L. Bird

Risk factors for interstage death after stage 1 reconstruction of hypoplastic left heart syndrome and variants

OBJECTIVE The risk of death during the interstage period remains high after stage 1 reconstruction for single ventricle lesions, despite improved surgical results. The purpose of this study is to identify risk factors for interstage death and to describe the events leading to interstage death. METHODS A nested case-control study was conducted of 368 patients who underwent stage 1 reconstruction at a single center between January 1998 and April 2005. RESULTS Among the 313 (85%) hospital survivors, there were 33 (10.5%) interstage deaths. Cases more frequently presented with intact or restrictive atrial septum (9 [27%] vs 4 [4%]; P < .001), were older at the time of surgery (5 [2-40] vs 3 [1-42] days; P = .005), had more postoperative arrhythmias (12 [36%] vs 15 [15%]; P = .01), and a higher incidence of airway or respiratory complications (12 [36%] vs 19 [19%]; P = .04). By multivariate analysis, only intact atrial septum (odds ratio 7.6; 95% confidence intervals 1.9-29.6; P = .003) and age at operation greater than 7 days (odds ratio 3.8; 95% confidence intervals 1.3-11.2; P = .017) were predictors of interstage death. CONCLUSIONS The presence of intact atrial septum and older age at the time of surgery are associated with a higher risk of interstage death. In addition, postoperative arrhythmia and airway complications are associated with a higher risk of interstage death in univariate analysis. The results of this study provide a focus for interstage monitoring and risk stratification of these high-risk infants, which may improve overall survival.

Weight change in infants with a functionally univentricular heart: from surgical intervention to hospital discharge.

OBJECTIVE The purpose of this study was to assess the pattern of weight change from surgical intervention to home discharge and to determine predictors of poor growth in this population of infants with congenital cardiac disease. METHODS Neonates with functionally univentricular physiology enrolled in a prospective cohort study examining growth between March, 2003 and May, 2007 were included. Weights were collected at birth, before surgical intervention, and at hospital discharge. In addition, retrospective echocardiographic data and data about post-operative complications were reviewed. Primary outcome variables were weight-for-age z-score at discharge and change in weight-for-age z-score between surgery and discharge. RESULTS A total of 61 infants met the inclusion criteria. The mean change in weight-for-age z-score between surgery and hospital discharge was minus 1.5 plus or minus 0.8. Bivariate analysis revealed a significant difference in weight-for-age z-score between infants who were discharged on oral feeds, minus 1.1 plus or minus 0.8 compared to infants with feeding device support minus 1.7 plus or minus 0.7, p-value equal to 0.01. Lower weight-for-age z-score at birth, presence of moderate or greater atrioventricular valve regurgitation, post-operative ventilation time, and placement of an additional central venous line were associated with 60% of the variance in weight-for-age z-score change. CONCLUSION Neonates undergoing staged surgical repair for univentricular physiology are at significant risk for growth failure between surgery and hospital discharge. Haemodynamically significant atrioventricular valve regurgitation and a complex post-operative course were risk factors for poor post-operative weight gain. Feeding device support appears to be insufficient to ensure adequate weight gain during post-operative hospitalisation.

Clinical Epidemiology of Extubation Failure in the Pediatric Cardiac ICU: A Report From the Pediatric Cardiac Critical Care Consortium

Objective: To describe the clinical epidemiology of extubation failure in a multicenter cohort of patients treated in pediatric cardiac ICUs. Design: Retrospective cohort study using prospectively collected clinical registry data. Setting: Pediatric Cardiac Critical Care Consortium registry. Patients: All patients admitted to the CICU at Pediatric Cardiac Critical Care Consortium hospitals. Interventions: None. Measurements and Main Results: Analysis of all mechanical ventilation episodes in the registry from October 1, 2013, to July 31, 2014. The primary outcome of extubation failure was reintubation less than 48 hours after planned extubation. Repeated-measures analysis using generalized estimating equations to account for within patient and center correlation was performed to identify risk factors for extubation failure. Adjusted extubation failure rates for each hospital were calculated using logistic regression controlling for patient factors. Of 1,734 mechanical ventilation episodes (1,478 patients at eight hospitals) ending in a planned extubation, there were 100 extubation failures (5.8%). In multivariable analysis, only longer duration of mechanical ventilation was significantly associated with extubation failure (p = 0.01); the failure rate was 4% when ventilated less than 24 hours, 9% after 24 hours, and 13% after 7 days. For 503 patients intubated and extubated in the cardiac operating room, 15 patients (3%) failed extubation within 48 hours (12 within 24 hr). Case-mix-adjusted extubation failure rates ranged from 1.1% to 9.8% across hospitals. Patients failing extubation had greater median cardiac ICU length of stay (15 vs 3 d; p < 0.001) and in-hospital mortality (7.9 vs 1.2%; p < 0.001). Conclusions: Though extubation failure is uncommon overall, there may be opportunities to improve extubation readiness assessment in patients ventilated more than 24 hours. These data suggest that extubation in the operating room after cardiac surgery can be done with a low failure rate. We observed variation in extubation failure rates across hospitals, and future investigation must elucidate the optimal strategies of high-performing centers to reduce ventilation time while limiting extubation failures.

Prediction of periventricular leukomalacia. Part I: Selection of hemodynamic features using logistic regression and decision tree algorithms

OBJECTIVE Periventricular leukomalacia (PVL) is part of a spectrum of cerebral white matter injury which is associated with adverse neurodevelopmental outcome in preterm infants. While PVL is common in neonates with cardiac disease, both before and after surgery, it is less common in older infants with cardiac disease. Pre-, intra-, and postoperative risk factors for the occurrence of PVL are poorly understood. The main objective of the present work is to identify potential hemodynamic risk factors for PVL occurrence in neonates with complex heart disease using logistic regression analysis and decision tree algorithms. METHODS The postoperative hemodynamic and arterial blood gas data (monitoring variables) collected in the cardiac intensive care unit of Childrens Hospital of Philadelphia were used for predicting the occurrence of PVL. Three categories of datasets for 103 infants and neonates were used-(1) original data without any preprocessing, (2) partial data keeping the admission, the maximum and the minimum values of the monitoring variables, and (3) extracted dataset of statistical features. The datasets were used as inputs for forward stepwise logistic regression to select the most significant variables as predictors. The selected features were then used as inputs to the decision tree induction algorithm for generating easily interpretable rules for prediction of PVL. RESULTS Three sets of data were analyzed in SPSS for identifying statistically significant predictors (p<0.05) of PVL through stepwise logistic regression and their correlations. The classification success of the Case 3 dataset of extracted statistical features was best with sensitivity (SN), specificity (SP) and accuracy (AC) of 87, 88 and 87%, respectively. The identified features, when used with decision tree algorithms, gave SN, SP and AC of 90, 97 and 94% in training and 73, 58 and 65% in test. The identified variables in Case 3 dataset mainly included blood pressure, both systolic and diastolic, partial pressures pO(2) and pCO(2), and their statistical features like average, variance, skewness (a measure of asymmetry) and kurtosis (a measure of abrupt changes). Rules for prediction of PVL were generated automatically through the decision tree algorithms. CONCLUSIONS The proposed approach combines the advantages of statistical approach (regression analysis) and data mining techniques (decision tree) for generation of easily interpretable rules for PVL prediction. The present work extends an earlier research [Galli KK, Zimmerman RA, Jarvik GP, Wernovsky G, Kuijpers M, Clancy RR, et al. Periventricular leukomalacia is common after cardiac surgery. J Thorac Cardiovasc Surg 2004;127:692-704] in the form of expanding the feature set, identifying additional prognostic factors (namely pCO(2)) emphasizing the temporal variations in addition to upper or lower values, and generating decision rules. The Case 3 dataset was further investigated in Part II for feature selection through computational intelligence.

Prediction of periventricular leukomalacia. Part II: Selection of hemodynamic features using computational intelligence

OBJECTIVE The objective of Part II is to analyze the dataset of extracted hemodynamic features (Case 3 of Part I) through computational intelligence (CI) techniques for identification of potential prognostic factors for periventricular leukomalacia (PVL) occurrence in neonates with congenital heart disease. METHODS The extracted features (Case 3 dataset of Part I) were used as inputs to CI based classifiers, namely, multi-layer perceptron (MLP) and probabilistic neural network (PNN) in combination with genetic algorithms (GA) for selection of the most suitable features predicting the occurrence of PVL. The selected features were next used as inputs to a decision tree (DT) algorithm for generating easily interpretable rules of PVL prediction. RESULTS Prediction performance for two CI based classifiers, MLP and PNN coupled with GA are presented for different number of selected features. The best prediction performances were achieved with 6 and 7 selected features. The prediction success was 100% in training and the best ranges of sensitivity (SN), specificity (SP) and accuracy (AC) in test were 60-73%, 74-84% and 71-74%, respectively. The identified features when used with the DT algorithm gave best SN, SP and AC in the ranges of 87-90% in training and 80-87%, 74-79% and 79-82% in test. Among the variables selected in CI, systolic and diastolic blood pressures, and pCO(2) figured prominently similar to Part I. Decision tree based rules for prediction of PVL occurrence were obtained using the CI selected features. CONCLUSIONS The proposed approach combines the generalization capability of CI based feature selection approach and generation of easily interpretable classification rules of the decision tree. The combination of CI techniques with DT gave substantially better test prediction performance than using CI and DT separately.

Change in regional (somatic) near-infrared spectroscopy is not a useful indictor of clinically detectable low cardiac output in children after surgery for congenital heart defects

Objective: Near-infrared spectroscopy correlation with low cardiac output has not been validated. Our objective was to determine role of splanchnic and/or renal oxygenation monitoring using near-infrared spectroscopy for detection of low cardiac output in children after surgery for congenital heart defects. Design: Prospective observational study. Setting: Pediatric intensive care unit of a tertiary care teaching hospital. Patients: Children admitted to the pediatric intensive care unit after surgery for congenital heart defects. Interventions: None. Measurements and Main Results: We hypothesized that splanchnic and/or renal hypoxemia detected by near-infrared spectroscopy is a marker of low cardiac output after pediatric cardiac surgery. Patients admitted after cardiac surgery to the pediatric intensive care unit over a 10-month period underwent serial splanchnic and renal near-infrared spectroscopy measurements until extubation. Baseline near-infrared spectroscopy values were recorded in the first postoperative hour. A near-infrared spectroscopy event was a priori defined as ≥20% drop in splanchnic and/or renal oxygen saturation from baseline during any hour of the study. Low cardiac output was defined as metabolic acidosis (pH <7.25, lactate >2 mmol/L, or base excess ⩽−5), oliguria (urine output <1 mL/kg/hr), or escalation of inotropic support. Receiver operating characteristic analysis was performed using near-infrared spectroscopy event as a diagnostic test for low cardiac output. Twenty children were enrolled: median age was 5 months; median Risk Adjustment for Congenital Heart Surgery category was 3 (1–6); median bypass and cross-clamp times were 120 mins (45–300 mins) and 88 mins (17–157 mins), respectively. Thirty-one episodes of low cardiac output and 273 near-infrared spectroscopy events were observed in 17 patients. The sensitivity and specificity of a near-infrared spectroscopy event as an indicator of low cardiac output were 48% (30%–66%) and 67% (64%–70%), respectively. On receiver operating characteristic analysis, neither splanchnic nor renal near-infrared spectroscopy event had a significant area under the curve for prediction of low cardiac output (area under the curve: splanchnic 0.45 [95% confidence interval 0.30–0.60], renal 0.51 [95% confidence interval 0.37–0.65]). Conclusions: Splanchnic and/or renal hypoxemia as detected by near-infrared spectroscopy may not be an accurate indicator of low cardiac output after surgery for congenital heart defects.

Results of elective repair at 6 months or younger in 277 patients with tetralogy of Fallot: A 14-year experience at a single center

OBJECTIVE To report practice and outcomes in infants undergoing elective repair of tetralogy of Fallot. METHODS A review of a retrospective cohort of elective complete repair of infants age 6 months or younger from 1995 to 2009 was performed. Patients were excluded because of previous interventions, hypercyanotic episodes, intensive care admissions, additional major cardiac defects, or if they were not discharged after birth. Length of stay, mortality, and complications were recorded. Association was determined using logistic or linear regression models and univariate testing determined the multivariate model. RESULTS There were 277 patients included. The hospital mortality rate was zero. A total of 87.4% of patients were discharged home within 7 days of repair, and 21.6% of patients were discharged on or before the third postoperative day. The postoperative course was uncomplicated in 245 patients (88.4%). Longer support time was associated independently with increased odds of complications (P < .001). Longer support time, younger age, chromosomal abnormality, and presence of a complication were associated independently with a longer hospital stay (all P < .001). Patients younger than 3 months (n = 110) had a longer median hospital stay (4 vs 3 days; P < .001) and longer support times (77.3 ± 35.1 min vs 66.4 ± 34 min; P < .01). CONCLUSIONS Elective tetralogy of Fallot repair was performed at 6 months or younger with low morbidity, no hospital mortality, and an 11.6% complication rate. Longer support times, lower weight, chromosomal abnormalities, and complications were associated with a significantly increased duration of hospital stay.

A tool to measure shared clinical understanding following handoffs to help evaluate handoff quality.

BACKGROUND Information exchanged during handoffs contributes importantly to a teams shared mental model. There is no established instrument to measure shared clinical understanding as a marker of handoff quality. OBJECTIVE To study the reliability, validity, and feasibility of the pediatric cardiology Patient Knowledge Assessment Tool (PKAT), a novel instrument designed to measure shared clinical understanding for pediatric cardiac intensive care unit patients. DESIGN To estimate reliability, 10 providers watched 9 videotaped simulated handoffs and then completed a PKAT for each scenario. To estimate construct validity, we studied 90 handoffs in situ by having 4 providers caring for an individual patient each complete a PKAT following handoff. Construct validity was assessed by testing the effects of provider preparation and patient complexity on agreement levels. SETTING A 24-bed pediatric cardiac intensive care unit in a freestanding childrens hospital. RESULTS Video simulation results demonstrated score reliability. Average inter-rater agreement by item ranged from 0.71 to 1.00. During in situ testing, agreement by item ranged from 0.41 to 0.87 (median 0.77). Construct validity for some items was supported by lower agreement rates for patients with increased length of stay and increased complexity. DISCUSSION Results suggest that the PKAT has high inter-rater reliability and can detect differences in understanding between handoff senders and receivers for routine and complex patients. Additionally, the PKAT is feasible for use in a real-time clinical environment. The PKAT or similar instruments could be used to study effects of handoff improvement efforts in inpatient settings.

Successful support and separation from veno–venous extracorporeal membrane oxygenation support in a three-month-old patient following bidirectional Glenn procedure

While extracorporeal membrane oxygenation (ECMO) is a useful mechanism of providing support in pediatric patients with cardiopulmonary dysfunction following surgery for congenital heart disease, outcomes have varied dramatically between distinct cardiac diagnoses. Reported outcomes of ECMO support following a bidirectional Glenn procedure in patients with single ventricle physiology are uniformly poor due in part to physiological and anatomical challenges inherent to cannulation in this population. We describe a unique veno-venous cannulation that can be applied to this patient population and has allowed for successful decannulation in our practice.

Evaluation of QT interval correction methods in normal pediatric resting ECGs

Different methods for heart rate correction of QT intervals have been proposed, but the classic Bazett method remains widely used in current clinical practice, despite being often criticized for limitations and disadvantages. To investigate the value of commonly employed methods in pediatric patients, we evaluated four QT correction methods (Bazett, Fridericia, Framingham and Hodges formulae) in a set of 2,170 normal pediatric resting ECGs. The dataset (age 0 to 20 years) is typical of a pediatric population for age, gender and heart rate. Scatter plotting of uncorrected QT versus heart rate reveals curvilinearity not typically identified in previously reported adult normal ECG datasets. Among the four algorithms tested, corrected QT (QTc) values calculated from the Bazett formula yielded the most consistent results across different ranges of heart rate and age. Further statistical regression modeling demonstrated that the Bazett method better fits the overall curvilinear trend in QT-heart rate distribution than the other formulae. The Bazett method also has the least residual heart rate dependence after correction. This study provides support for the use of the Bazett QT correction method over others in normal pediatric resting ECGs. The Bazett method may represent a balance point of accuracy, simplicity and generalizability not yet surpassed by other commonly applied alternatives. Our analysis suggests that its use in general pediatric patients is appropriate as a current best option.