Abstract
Purpose We sought to determine factors influencing survival and resource utilization in patients undergoing surgical resection of congenital lung malformations (CLM). Additionally, we used propensity score-matched analysis (PSMA) to compare these outcomes for thoracoscopic versus open surgical approaches. Methods Kids’ Inpatient Database (1997–2009) was used to identify congenital pulmonary airway malformation (CPAM) and pulmonary sequestration (PS) patients undergoing resection. Open and thoracoscopic CPAM resections were compared using PSMA. Results 1547 cases comprised the cohort. In-hospital survival was 97%. Mortality was higher in small vs. large hospitals, p < 0.005. Survival, pneumothorax (PTX), and thoracoscopic procedure rates were higher, while transfusion rates and length of stay (LOS) were lower, in children ≥ 3 vs. < 3 months (p < 0.001). Multivariate analysis demonstrated longer LOS for older patients and Medicaid patients (all p < 0.005). Total charges (TC) were higher for Western U.S., older children, and Medicaid patients (p < 0.02). PSMA for thoracoscopy vs. thoracotomy in CPAM patients showed no difference in outcomes. Conclusion CLM resections have high associated survival. Children < 3 months of age had higher rates of thoracotomy, transfusion, and mortality. Socioeconomic status, age, and region were independent indicators for resource utilization. Extent of resection was an independent prognostic indicator for in-hospital survival. On PSMA, thoracoscopic resection does not affect outcomes.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 1414-1420 |
| Number of pages | 7 |
| Journal | Journal of pediatric surgery |
| Volume | 51 |
| Issue number | 9 |
| DOIs | |
| State | Published - Sep 1 2016 |
Funding
The Kids' Inpatient Database (KID) is a national database available from the Healthcare Cost and Utilization Project (HCUP), sponsored by the Agency for Healthcare Research and Quality (AHRQ). The dataset contains information on approximately 2–3 million pediatric inpatient discharges per triennial release. Clinical diagnoses and procedures are coded using the International Classification of Diseases , Ninth Revision , Clinical Modification (ICD-9-CM). This study included cases from the 1997, 2000, 2003, 2006, and 2009 releases. Data was not collected from intervening years. Children and adolescents (< 20 years old) diagnosed with CLM were identified using codes 748.4 (CPAM) and 748.5 (PS). We limited the cohort to those undergoing surgical intervention for CLM in the form of segmentectomy (32.3, 32.30, 32.39), lobectomy (32.4, 32.41, 32.49), pneumonectomy (32.5, 32.50, 32.59), thoracotomy (34.01, 34.02, 34.99), or other pulmonary resection (32.20, 32.6, 32.9, 33.1), using their respective procedure codes. The procedures were further classified as either thoracoscopic or open surgery. Those with disposition coded as transfer to short term hospital and other transfers, including skilled nursing facility, intermediate care, and other type of facility were excluded from survival analyses to avoid the potential for duplicate case reporting from the receiving hospital. All analyses were limited to data available for each category and, as such, all analyzed outcomes were limited to the hospital stay associated with the operation. Categorical measures were compared using chi-square and Fisher's exact tests as appropriate. Continuous measures were compared using Student's U.S. dollar (USD) values according to rates determined by the United States (U.S.) Department of Labor t- and Mann–Whitney U tests as appropriate. Ordinal logistic regression models were constructed to predict demographic, hospital, and clinical characteristics associated with higher resource utilization in the forms of LOS and TC. Binary logistic regression modeling, using a backward-stepwise method, was performed with the same covariates to analyze in-hospital survival characteristics. All analyses were two-sided, and significance was defined at alpha level 0.05. Owing to the nature of the KID, all survival/mortality analyses in our study refer to in-hospital survival/mortality. TC analyses were adjusted to 2009 [40] . All cases were weighted to project nationally representative estimates. Propensity score matching to analyze outcomes of thoracoscopic vs. open resection for CPAM was performed using the nearest neighbor 1:1 ratio method. The resulting groups were weighted after matching, thus resulting in unequal numbers in the comparison groups. Briefly, a multivariate logistic regression model was constructed to assign propensity scores for each case included in this analysis. Covariates used for modeling included demographic (age, gender, race), socioeconomic (primary payer, median income quartile), and hospital characteristics (size, location/teaching status, region, type). Risk-adjustment was performed using the comorbidity codes as described in the Elixhauser method, which has been validated in multiple previous studies [32,33,41] . Case matching was performed using MatchIt version 2.4–20 (Cambridge, MA), a supplemental module available for R commander version 2.14.2 (R foundation for Statistical Computing; Vienna, Austria) [42] . All other statistical analyses were performed using SPSS Statistics version 21 (IBM Corporation, Armonk, NY). The Institutional Review Board at the University of Miami Miller School of Medicine (Miami, FL) deemed this retrospective study to be exempt from full review. 2
Keywords
- Bronchopulmonary sequestration
- Congenital pulmonary airway malformation
- Health resources
- Outcome assessment
- Thoracoscopy
- Thoracotomy
ASJC Scopus subject areas
- Surgery
- Pediatrics, Perinatology, and Child Health
