BACKGROUND: Post-bariatric body contouring represents a rapidly growing field. With long operative times, evidence-based guidelines for operative management are vital. This study examined perioperative factors and correlated these parameters with outcomes.

METHODS: Patients who lost ≥ 50 pounds and underwent body contouring were enrolled in an IRB-approved prospective registry over three years. Weight loss was achieved either by bariatric surgery or diet and exercise. All cases were performed by the senior author at two community hospitals and two academic hospitals. Hypothermia was defined as a minimum temperature of 35°C or lower.
The t-test and Mann-Whitney U test were used for comparison of dichotomous groups with linear and Spearman regression for continuous variables. Univariate logistic regression was used to analyze complications with multivariate regression for possible confounders.

RESULTS: Three hundred and eight patients (272 female, 36 male) were analyzed. The average operative time was 4.7 hours and 71.4% of cases were performed in an academic hospital. Urine output did not differ by gender, BMI or age and failed to correlate with complications whether analyzed per hour or by total output. The average estimated blood loss was 269 ± 330mL (range minimal-1500). EBL was not predictive of complications, but in multivariate analysis a lower minimum temperature was associated with an increased blood loss (p = 0.014).
The average minimum operative temperature was 35.6 ± 0.63°C (range 34.0-38.0) while the average maximum operative temperature was 36.5 ± 0.75°C (range 34.8-39.0). Gender was not correlated to low or high temperature continuously but men did have an increased risk of becoming hypothermic (p = 0.034; OR 2.3). Fifty nine cases (19%) had a minimum temperature recording of 35°C or less.
Multivariate analysis provided a seroma model including intraoperative time, gender and minimum temperature. The transfusion model included BMI at time of operation, intraoperative time and minimum temperature. Hypothermia predicted both seroma development (p = 0.003; OR 3.1 per 1° decrease; Figure 1) and the need for transfusion (p = 0.005; OR 2.4 per 1° decrease).

CONCLUSIONS: Operative hypothermia below 35°C was present in one fifth of cases. After controlling for other factors, lower intraoperative temperature was still associated with an increased risk of seroma formation, blood loss and the need for transfusion. This is the first study to report an increased incidence of seroma in hypothermic patients; our findings of increased rates of blood loss and transfusion are consistent with the literature in other types of surgery. Maintaining normothermia is a critical component of perioperative management. In response to our findings we have adopted a hypothermia protocol involving patient prewarming, a higher operating room temperature and the routine use of warmed fluids.

Figure 1. Seroma probability versus minimum operative temperature (°C); (colored) actual seroma rates in zones of risk; (curve) predicted probability of seroma development from multivariate model.