Back to Annual Meeting Program
Increased Anesthesia Duration Increases Venous Thromboembolism Risk in Plastic Surgery: A Six-Year Analysis of Over 19,000 Cases
Alexei S. Mlodinow, BA1, Nima Khavanin, BS1, Jon P. Ver Halen, MD2, Aksharananda Rambachan, BS1, Karol A. Gutowski, MD3, John YS Kim, MD1.
1Northwestern Feinberg School of Medicine, Chicago, IL, USA, 2Ingram Cancer Center, Memphis, TN, USA, 3Private Practice, Northbrook, IL, USA.
Purpose: Venous thromboembolism (VTE) is a well-documented cause of morbidity, mortality and excess healthcare cost. Various risk stratification schema such as the Caprini score and its derivatives exist in the plastic surgery literature, but do not take into account variations in procedure length. The putative risk of VTE conferred by increased length of time under anesthesia has never been rigorously explored. The goals of this study are to quantitatively assess this relationship, as well as provide a benchmark VTE rates among plastic surgery patients.
Methods: The National Surgical Quality Improvement (ACS-NSQIP) database was queried for plastic and reconstructive surgery procedures performed under general anesthesia between 2005 and 2011. Z-scores were calculated based on procedure-specific mean surgical durations, to assess each case’s length in comparison to other cases with the same primary procedure. Patients with and without post-operative VTE were compared with respect to a variety of demographics, comorbidities, and intraoperative characteristics. Potential confounders for VTE were included in a regression model, along with the Z-score for each case. This yielded an independent odds ratio for VTE, with respect to each unit increment in Z-score.
Results: A total of 19,276 cases met inclusion criteria. VTE occurred in a variety of procedures, both cosmetic and reconstructive, at a total rate of 0.36% (70 incidents). There were incidents in each Z-Score interval analyzed. Increased relative surgical duration (as measured by the Z-score) was associated with increased VTE rates (Figure 1). Further, regression analysis showed a higher Z-score to be a statistically significant (p<0.001) independent risk factor for post-operative VTE, with an odds ratio of 1.772 per additional unit (Table 1).
Conclusions: This validates the long-held view that increased surgical time confers risk of VTE, as well as benchmarks VTE rates in plastic surgery procedures. While this in itself does not suggest an intervention, surgical time under general anesthesia would be a useful addition to risk models in plastic surgery.
Figure 1. Incidence of VTE across Z-Score Intervals. As general anesthesia increases relative to the mean for a given procedure, the incidence of venous thromboembolism increases.
Table 1. Odds Ratios for Z-Score on the Incidence of Venous Thromboembolism
|Odds Ratio||95% Confidence Interval||p Value||H-L Statistic||c-statistic|
|*Variables for adjustment were selected through a bivariate screen. Variables with a p-value ≤ 0.20 and n ≥ 10 were included in the model|
Back to Annual Meeting Program