L J Blair, MD, C R Huntington, MD, T C Cox, MD, T Prasad, MA, A E Lincourt, PhD, MBA, V A Augenstein, MD, FACS, B Todd Heniford, MD, FACS. Carolinas Medical Center
Introduction: Sepsis is a devastating postoperative consequence and carries a high mortality. The objective of this study was to examine a common general surgery procedure, laparoscopic cholecystectomy, and identify risk factors which could potentially predict risk of sepsis preoperatively.
Methods: The American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database was queried from 2005-2012 for patients who developed sepsis postoperatively following laparoscopic cholecystectomy for acute cholecystitis. Patients with evidence of systemic inflammatory response syndrome or sepsis present preoperatively were excluded. Data was analyzed using standard statistical methods including the chi square test for categorical variables and Wilcoxon two-sample test for continuous variables.
Results: Over an eight year time period, 12,387 patients underwent laparoscopic cholecystectomy for acute cholecystitis. The incidence of postoperative sepsis in this population was 0.60%(n=75). Patients who developed sepsis postoperatively were older with a decreased body mass index(BMI) and were more likely to have associated comorbidities including diabetes, and an elevated Charlson Comorbidity Index(CCI)(p<0.01);see Table 1. Patients who developed sepsis had lower preoperative albumin and increased operative time(p<0.0001). A multiple logistic regression model with age, race, gender, BMI, preoperative albumin, CCI score, and operative time as covariates supported the conclusions from univariate analyses. Overall LOS was elevated in the sepsis group, 7.9±12.9 days versus 1.3±4.6 days(p<0.0001). There was a trend toward increased in-hospital interval between admission and operation in patients who developed postoperative sepsis(3.2±13.2days versus 1.1±3.9days, p=0.05).
Conclusions: Predictors of postoperative sepsis in laparoscopic cholecystectomy include older age, associated comorbidities, decreased preoperative albumin, increased operative time, and a trend toward increased in-hospital interval prior to surgery.
| Sepsis (n=75) | No Sepsis (n=12,312) | P value | |
| Age (years) | 65.2±16 | 49.5±17.5 | 0.0001 |
| BMI (kg/m2) | 29.2±7.7 | 31.1±7.6 | 0.0137 |
| Diabetes (% patients) | 36% | 11% | 0.0001 |
| Hypertension (% patients) | 63% | 35% | 0.0001 |
| COPD (% patients) | 8% | 2% | 0.002 |
| CHF (% patients) | 4% | 0.4% | 0.001 |
| Renal failure (% patients) | 1% | 0.09% | 0.002 |
| Chronic steroid use (% patients) | 11% | 2% | 0.0001 |
| Charlson Comorbidity Index | 1.3±2.3 | 0.3±0.8 | 0.0001 |
| PreOp Albumin | 3.3±0.6 | 3.8±0.6 | 0.0001 |
| Length of Stay (days) | 7.6±8.6 | 2.5±4.7 | 0.0001 |
| Operative time (minutes) | 96±49 | 74±40 | 0.0001 |
| In-Hospital time to OR (days) | 3.2±13.2 | 1.1±3.9 | 0.0536 |