Preoperative risk assessment in children undergoing major urologic surgery

Predicting postoperative complications in pediatric surgery: A novel pediatric comorbidity index

INTRODUCTION/BACKGROUND

Comorbidity-driven surgical risk assessment is essential for informed patient counseling, risk-stratification, and outcomes-based health-services research. Existing mortality-focused comorbidity indices have had mixed success at risk-adjustment in children.

OBJECTIVE

To develop a new comorbidity-driven multispecialty surgical risk index predicting 30-day postoperative complications in children.

STUDY DESIGN

This retrospective cohort study investigated children undergoing surgical procedures across seven specialties in 2014-2015 using the MarketScan® Research databases. The risk index was derived separately for ambulatory and inpatient surgery patients using logistic regression with backward selection. The performance of the novel index in discriminating postoperative complications vis-à-vis three existing comorbidity indices was compared using bootstrapping and area under the receiver operating characteristics curves (AUC).

RESULTS

We identified 190,629 ambulatory and 22,633 inpatient patients. The novel index had the best performance for discriminating postoperative complications for inpatients (AUC 0.76, 95% confidence interval [CI] 0.75-0.77) relative to the Charlson Comorbidity Index (CCI, 0.56, 95% CI 0.56-0.57), Van Walraven Index (VWI, 0.60, 95% CI 0.60-0.61), and Rhee Score (RS, 0.69, 95% CI 0.68-0.70). In the ambulatory cohort, the novel index outperformed all three existing indices, though none demonstrated excellent discriminatory ability for complications (novel score 0.68, 95% CI 0.67-0.68; CCI 0.53, 95% CI 0.52-0.53; VWI 0.53, 95% CI 0.52-0.53; RS 0.50, 95% CI 0.49-0.50).

DISCUSSION

In both inpatient and ambulatory pediatric settings, our novel comorbidity index demonstrated better performance at predicting postoperative complications than three widely used alternatives. This index will be useful for research and may be adaptable to clinical settings to identify high-risk patients and facilitate perioperative planning.

CONCLUSION

We developed a novel pediatric comorbidity index with better performance at predicting postoperative complications than three widely used alternatives.

Combining adult with pediatric patient data to develop a clinical decision support tool intended for children: leveraging machine learning to model heterogeneity

Background

Clinical decision support (CDS) tools built using adult data do not typically perform well for children. We explored how best to leverage adult data to improve the performance of such tools. This study assesses whether it is better to build CDS tools for children using data from children alone or to use combined data from both adults and children.

Methods

Retrospective cohort using data from 2017 to 2020. Participants include all individuals (adults and children) receiving an elective surgery at a large academic medical center that provides adult and pediatric services. We predicted need for mechanical ventilation or admission to the intensive care unit (ICU). Predictor variables included demographic, clinical, and service utilization factors known prior to surgery. We compared predictive models built using machine learning to regression-based methods that used a pediatric or combined adult-pediatric cohort. We compared model performance based on Area Under the Receiver Operator Characteristic.

Results

While we found that adults and children have different risk factors, machine learning methods are able to appropriately model the underlying heterogeneity of each population and produce equally accurate predictive models whether using data only from pediatric patients or combined data from both children and adults. Results from regression-based methods were improved by the use of pediatric-specific data.

Conclusions

CDS tools for children can successfully use combined data from adults and children if the model accounts for underlying heterogeneity, as in machine learning models.

Postoperative complications in pediatric patients with cerebral palsy

BACKGROUND/PURPOSE

To assess surgical outcomes of patients with cerebral palsy (CP) and if they differ from patients without CP.

METHODS

The NSQIP-Pediatric database from 2012 to 2019 was used to compare differences in presenting characteristics and outcomes between patients with and without CP. Chi-square tests and multivariable logistic regression analysis were used to determine significance.

RESULTS

119,712 patients, 433 (0.4%) with CP, 119,279 (99.6%) without, were identified. Patients with CP had more postoperative complications (19.4% vs. 6.9%, p < 0.001) with an OR of 3.2, (95%CI 2.5-4.1, p < 0.001) on univariable analysis. They underwent fewer laparoscopic procedures (79.1% vs. 90.8%, p < 0.001), had more readmissions (10.2% vs. 3.8%, p < 0.001), reoperations (5.1% vs. 1.2%, p < 0.001), and longer length of stays (LOS) (median 3 versus 1 day, p < 0.001). On multivariable analysis, having CP did not increase the odds of postoperative morbidity (OR 0.99, 95% CI 0.7-1.3), but higher ASA class, congenital lung malformation, gastrointestinal disease, coagulopathy, preoperative inotropic support, oxygen use, nutritional support, and steroid use significantly increase the odds of morbidity, all of which were more common in patients with CP.

CONCLUSION

Patients with CP have more postoperative complications, open procedures, and longer LOS. Patient complexity may account for these differences and risk-directed perioperative planning may improve outcomes.

LEVEL OF EVIDENCE

Level IV.

Racial Disparities in Failure to Rescue Following Unplanned Reoperation in Pediatric Surgery

BACKGROUND

Failure to rescue (FTR) and unplanned reoperation following an index surgical procedure are key indicators of the quality of surgical care. Given that differences in unplanned reoperation and FTR rates among racial groups may contribute to persistent disparities in postsurgical outcomes, we sought to determine whether racial differences exist in the risk of FTR among children who required unplanned reoperation following inpatient surgical procedures.

METHODS

We used the National Surgical Quality Improvement database (2012-2017) to assemble a cohort of children (<18 years), who underwent inpatient surgery and subsequently returned to the operating room within 30 days of the index surgery. We used logistic regression models to estimate the odds ratio (OR) and 95% confidence interval (CI) of FTR, comparing African American (AA) to White children. We estimated the risk-adjusted odds ratio (aOR) for FTR by controlling the analyses for demographic characteristics, surgical profile, and preoperative comorbidities. We further evaluated the racial differences in FTR by stratifying the analyses by the timing of unplanned reoperation.

RESULTS

Of 276,917 children who underwent various inpatient surgical procedures, 10,425 (3.8%) required an unplanned reoperation, of whom 2016 (19.3%) were AA and 8409 (80.7%) were White. Being AA relative to being White was associated with a 2-fold increase in the odds of FTR (aOR: 2.03; 95% CI, 1.5-2.74; P < .001). Among children requiring early unplanned reoperation, AAs were 2.38 times more likely to die compared to their White peers (8.9% vs 3.4%; aOR: 2.38; 95% CI, 1.54-3.66; P < .001). In children with intermediate timing of return to the operating room, the risk of FTR was 80% greater for AA children compared to their White peers (2.2% vs 1.1%; aOR: 1.80; 95% CI, 1.07-3.02; P = .026). Typically, AA children die within 5 days (interquartile range [IQR]: 1-16) of reoperation while their White counterparts die within 9 days following reoperation (IQR: 2-26).

CONCLUSIONS

Among children requiring unplanned reoperation, AA patients were more likely to die than their White peers. This racial difference in FTR rate was most noticeable among children requiring early unplanned reoperation. Time to mortality following unplanned reoperation was shorter for AA than for White children. Race appears to be an important determinant of FTR following unplanned reoperation in children and it should be considered when designing interventions to optimize unplanned reoperation outcomes.

Postoperative course following complex major pediatric urologic surgery: A single surgeon experience

BACKGROUND/PURPOSE

Payers have established unplanned returns as quality measures tied to reimbursement. We sought to identify patient characteristics and surgical factors associated with unplanned returns and postoperative complications following intraabdominal pediatric urologic procedures. Only one study has emergency room visits in pediatric urology, but included only limited case complexity.

METHODS

We retrospectively reviewed electronic medical records of all pediatric patients who underwent urological surgery from 2007 to 2016 by a single surgeon. Univariate and multivariate regressions were used to determine factors associated with length of stay, unplanned ER visits and readmissions.

RESULTS

A total of 372 cases were included, of which 251 were robotic and 121 were open. Overall, the rate of ER visits was 19%, rate of readmissions was 8.7%, and rate of reoperations was 4.3%. We found that the odds of a urologic ER visit and Clavien grade II (infectious) complications were comparatively lower for robotic surgery, but not readmission. Undergoing a lower tract procedure was significantly associated with a related ER visit; however, patient insurance status was not related to unplanned returns.

CONCLUSIONS

Procedure characteristics, not patient demographics were associated with unplanned returns. Robotic approach was associated with fewer ER visits and infectious complications, indicating potential cost savings.

LEVELS OF EVIDENCE

Level 3 Treatment Study.

Return to emergency department after pediatric urology procedures

BACKGROUND

Unplanned postoperative return visits to the emergency department (ED) and readmission represent a quality bench outcome and pose a considerable cost burden to health-care systems.

OBJECTIVE

The aim of this study is to evaluate ED return visits after pediatric urology procedures in a tertiary care children's hospital, trying to identify potential causes. This may constitute a platform for future improvement areas.

MATERIALS AND METHODS

A Quality Board-approved retrospective study was performed identifying all urologic cases completed between October 2012 and September 2015. Baseline demographics, American Society of Anesthesia class, operating surgeon, type of admission, type and duration of surgical procedure, and type of anesthesia given were evaluated. Patients who returned to the ED within 30 days of the surgery date were identified. The ED records were reviewed for time of return, etiology for visit, and management received. Univariate and subsequent multivariate logistic regression statistical analyses were performed to identify variables associated with ED return. Odds ratio (OR) and 95% confidence intervals (95% CIs) were generated to determine the significance of relationships.

RESULTS

Total of 4125 cases was identified. Median age was 32.9 months, with 85.1% of them being male. 349 (8.5%) cases returned to the ED within 30 days of the surgery. The majority of the returned patients, 295 (84.5%), managed conservatively with medications or reassurance, whereas 54 (15.5%) required readmission, and of those readmitted, 15 (4.3%) cases needed further surgical interventions, mainly urinary tract drainage procedures. Multivariate logistic regression analysis identified that the age, residence, admission type, inguinoscrotal surgery, and duration of surgical procedure were significantly associated with ED return (Table). The most common reason for the ED visit was UTI in 17.2%, followed by stent and catheter issues in 14.3%, wound-related in 14.3%, and bleeding in 11.7%.

DISCUSSION

Pediatric literature show varying rates of ED return ranging from 2.4% to 2.6% after urologic procedures. Our return to ED rate exceeds that found in US studies, which can perhaps be attributed to the differences between the Canadian and US health-care systems. As found with other studies, age, inpatient admission, procedure type, and increased operative time were related to ED returns, possibly because of the difficulty of young children expressing themselves and the presumed complex nature of longer operations that mostly need inpatient admission. The most common reason for ED return in this study as in others was presumptive UTI. A known limitation of this study is its retrospective nature, along with the possible missed visits of patients who presented to outside hospitals.

CONCLUSION

We present an account of the status of ED return visits after pediatric urology procedures in our institute. The majority of ED returns can be managed conservatively and are probably preventable.

Impact of positive preoperative urine cultures before pediatric lower urinary tract reconstructive surgery

PURPOSE

Children who undergo lower urinary tract reconstruction (LUTR) often have asymptomatic bacteriuria or recurrent urinary tract infections (UTI). We aimed to determine the prevalence of positive preoperative urine cultures (PPUC) before LUTR and to analyze any impact on postoperative outcomes.

METHODS

This retrospective review included all pediatric LUTR procedures utilizing bowel segments performed by one surgeon over 2 years. Preoperative cultures were obtained 1-2 days before surgery. Baseline characteristics and 90-day infection/readmission rates between patients with and without PPUC were compared using descriptive statistics, Fisher's exact, and Mann-Whitney tests with significance p < 0.05.

RESULTS

54 patients with mean age 10.1 ± 5.6 years underwent LUTR procedures using bowel including continent catheterizable channel (85%), enterocystoplasty (81%), and/or urinary diversion (9%). PPUC was present in 28 patients (52%). Postoperatively, 20% had inpatient infections, including eight UTI, four surgical site infections, and two bloodstream infections with no difference between those with or without PPUC. Within 90 days of discharge, 28% of patients were readmitted to the hospital, and there was no difference between groups. Postoperative urine cultures were positive in 83% of patients within 90 days.

CONCLUSIONS

Half of the patients undergoing LUTR have PPUC, but it does not increase the risk of postoperative infections or hospital readmissions. We believe complex LUTR can be safely performed in patients with PPUC.

The evaluation of three comorbidity indices in predicting postoperative complications and readmissions in pediatric urology

INTRODUCTION

The surgical comorbidity assessment is important for patient risk stratification, counseling, and research. In adults, risk assessment indices, such as the Charlson Co-morbidity Score (CCS) or Van Walraven Index (VWI), are well established. In pediatrics, however, risk assessment indices are scarce. Recently, a pediatric-specific risk assessment index, the Rhee index, was developed to discriminate mortality for pediatric general surgery patients. Currently, there is no validated risk assessment tool in pediatric urology.

OBJECTIVE

We compared the performance of the CCS, VWI, and Rhee Index in discriminating postoperative complications and readmissions to the emergency room/inpatient unit after pediatric urological procedures.

METHODS

We analyzed the Nationwide Readmissions Database (NRD), State Inpatient Databases (SID), and State Emergency Department Databases (SEDD). We included patients (<18 years) who underwent the following urological procedures: ureteroneocystostomy, ureteroureterostomy, radical/partial nephrectomy, pyeloplasty, appendicovesicostomy, enterocystoplasty, vesicostomy, and bladder neck sling. Complications were identified based on definitions in the National Surgical Quality Improvement Program (NSQIP). Thirty-day emergency room admission and inpatient readmissions were extracted. Comorbidity scores were calculated using each of the three indices. We compared the performance of each index in discriminate primarily postoperative complications in the NRD and both admission types in the SID/SEDD by constructing a receiver operating characteristics (ROC). AUCs were compared using the Delong method. This protocol was reviewed by our Institutional Review Board and deemed to be exempt.

RESULTS

We identified a total of 8006 patients in NRD and 6236 patients in SID/SEDD. The Rhee index had the best performance for discriminating postoperative complications (AUC = 0.67, 95% CI 0.64-0.70) compared to CCS (AUC = 0.62, 95% CI 0.60-0.65) and VWI (AUC = 0.62, 95% CI 0.59-0.65); p < 0.01. The CCS had the best performance for discriminating 30-day inpatient readmissions (AUC = 0.63, 95% CI 0.61-0.66) than VWI (AUC = 0.54, 95% CI 0.52-0.57), and Rhee Index (AUC = 0.56, 95% CI 0.54-0.59); p < 0.0001. All three indices had similarly poor discrimination for 30-day ER admissions: CCS (AUC = 0.52), VWI (AUC = 0.51), and Rhee Index (AUC = 0.50); p = 0.5 (see Table).

DISCUSSION

The Rhee Index had the best performance for discriminating postoperative complications, while the CCS was superior for discriminating inpatient readmissions among the three indices. Limitations to our study include inpatient-only procedures, inability to identify complications managed in clinics, omission of secondary operations, accounting for parental anxiety, and the generalizability of SID.

CONCLUSIONS

The three comorbidity indices evaluated are poor discriminators for postoperative complications, 30-day inpatient readmissions or 30-day ER admissions. A new index is needed for pediatric urology patients.

Relative to open surgery, minimally-invasive renal and ureteral pediatric surgery offers no improvement in 30-day complications, yet requires longer operative time: Data from the National Surgical Quality Improvement Program Pediatrics

Purpose

Performance of minimally-invasive surgery (MIS) is increasing relative to open surgery. We sought to compare the contemporary rates of short-term complications of open versus laparoscopic renal and ureteral surgery in pediatric patients.

Materials and Methods

A retrospective cross-sectional analysis of the National Surgical Quality Improvement Program Pediatrics database was performed of all cases in 2014 identified using CPT procedure codes for nephrectomy, partial nephrectomy (PN), ureteroneocystostomy (UNC), and pyeloplasty, and reviewed for postoperative complications. Univariate analysis was performed to determine 30-day complications, with comparison between open and MIS approaches. Receiver operator curve (ROC) analysis was performed to determine differences in body surface area (BSA) and age for open versus MIS.

Results

Review identified 207 nephrectomies, 72 PN, 920 UNC, and 625 pyeloplasties. MIS was associated with older age and larger BSA except for cases of UNC. Apart from PN, operative durations were longer with MIS. However, only PN was associated with significantly longer length of hospital stay (LOS). There was no difference in incidence of all other 30-day complications. When evaluating BSA via ROC, the area under the curve (AUC) was found to be 0.730 and was significant. Children with a BSA greater than 0.408 m² were more likely to have MIS (sensitivity, 66.9%; specificity, 69.3%). Regarding age, the AUC was 0.732. Children older than 637.5 days were more likely to have MIS (sensitivity, 72.8%; specificity, 63.3%).

Conclusions

Pediatric MIS is associated with longer operative time for nephrectomy, but shorter LOS following PN. Surgical approach was not associated with difference in short-term complications.

Role of body mass index in school-aged children with lower urinary tract dysfunction: Does weight classification predict treatment outcome?

INTRODUCTION

Lower urinary tract (LUT) dysfunction comprises a large percentage of pediatric urology referrals. Childhood obesity is a major health concern, and has been associated with voiding symptoms. We assessed the impact of body mass index (BMI) on treatment outcomes of children presenting with LUT or bladder-bowel dysfunction (BBD).

STUDY DESIGN

Children aged 5-17 years diagnosed with non-neurogenic LUT dysfunction and no prior urologic diagnoses were identified. Patient demographics including BMI, lower urinary tract symptoms, constipation, medical and psychologic comorbidities, imaging, and treatment outcomes were evaluated. BMI was normalized by age and gender according to percentiles: underweight < 5th, healthy 5th to <85th, overweight 85th to <95th, and obese > 95th percentile. Uni- and multivariate analyses were performed to identify predictors of treatment response.

RESULTS

During an 18-month period, 100 children (54 girls, 46 boys) met the inclusion criteria. The mean age at diagnosis was 7.7 ± 2.4 years, and mean length of follow-up 15.3 ± 13.1 months. Sixty-nine patients were a normal weight, 22 were overweight, and nine were obese. Fifteen percent of the children had complete treatment response, 63% partial response, and 22% non-response. On univariate analysis, children with elevated BMI (p = 0.04) or history of urinary tract infection (p = 0.01) were statistically more likely to not respond to treatment. Controlling for all other variables, children with BMI > 85th percentile had 3.1 times (95% CI 1.11-8.64; p = 0.03) increased odds of treatment failure (Table).

DISCUSSION

BBD management includes implementation of a bowel program and timed voiding regimen, with additional treatment modalities tailored on the basis of the prevailing symptoms. We observed that school-aged children with a BMI ≥ 85th percentile were over three times more likely to experience treatment failure when controlling for all other patient characteristics including constipation and a history of urinary tract infection. Limitations of the study include the relatively small sample size, lack of uroflow with electromyography to confirm the presence or the absence of detrusor sphincter dyssynergia, and inconsistent anticholinergic dosing.

CONCLUSIONS

Nearly one-third of school-aged children presenting to our institution with LUT or BBD were overweight or obese when normalized for age and gender. Children with LUT dysfunction and elevated BMI are significantly less likely to experience treatment response.