Back to Annual Meeting Posters
Analysis Of 1,102 Complex Hernia Repairs From The 2005-2010 NSQIP Datasets - Assessment Of Factors Associated With Post-operative Respiratory Failure (PRF)
John P. Fischer, MD, Jonas A. Nelson, MD, Eric Shang, MD, Joseph M. Serletti, MD, Stephen J. Kovach, MD.
Hospital of the Univesity of Pennsylvania, Philadelphia, PA, USA.
Complex abdominal wall reconstruction often requires component separation to achieve fascial approximation. These reconstructions can be associated significant rates of surgical complications and respiratory events. We aim to characterize and derive a model to predict post-operative respiratory failure (PRF) after complex abdominal wall reconstruction using the ACS-NSQIP database.
We reviewed the 2005-2010 ACS-NSQIP databases identifying encounters for CPT codes for both hernia repair and component separation (complex abdominal wall reconstruction). We defined PRF as either an unplanned intubation or prolonged ventilation (>48 hours). Exploratory univariate analyses and regressions were used to identify predictors. A model was created and validated using a bootstrap technique. A simplified risk score with greater clinical accessibility was subsequently constructed using weighted logistic regression coefficients.
1,102 complex abdominal repairs were performed during the study period. The following factors were associated with PRF: age (P=0.007), resident involvement (P=0.02), obesity (P=0.002), diabetes (P=0.01), smoking (P=0.02), dyspnea (P<0.001), functional status (P<0.001), COPD (P<0.001), CHF (P=0.03), weight loss (P=0.04), albumin (P<0.001), malnutrition (P<0.001), wound class (P=0.003), ASA (P<0.001), and operative time (P<0.001) (Table 1). Patients experiencing a respiratory complication stayed on average 10.8 days longer. Regression analysis revealed that morbid obesity (P=0.01), functional status (P=0.03), malnutrition (P=0.01), ASA (P=0.03), and operative time (P=0.02) were independently associated with higher rates of PRF events (Table 2). Regression modeling was performed using identifiable preoperative risks associated with PRF. The variables and weights are summarized in Table 3. The logistic regression’s c-statistic (measure of discrimination) was 0.75 in the derivation dataset, 0.73 by internal validation (optimism-corrected), and with good calibration (goodness of fit χ2=8.179, P=0.416). A simplified respiratory risk score was subsequently derived and validated with nearly equivalent discrimination (c=0.73). Patients with relatively low risk (RRS 0-1, n=478), intermediate (RRS 2-4, n= 588) and high risk (RRS 5-8, n= 36) had respiratory complication rates of 2.7%, 10.0%, and 44.4%, respectively.
This study characterizes PRF events after complex abdominal wall reconstruction and describes a validated predictive model and clinical risk assessment tool. Respiratory complications on average added 11 hospital days and were associated with morbid obesity, lower functional and nutritional status, and higher ASA scores. When high risk patients undergo longer operations the risk of major respiratory morbidity is clinically significant. Data derived from this large cohort study can be used to risk-stratify patients and to enhance peri-operative decision-making and patient counseling.
Back to Annual Meeting Posters