Back to Annual Meeting Program
Vasculogenesis and the Incidence of Surgical Site Infection in Patients Undergoing Ventral Hernia Repair by Components Separation
Benjamin Levi, MD, Jeffrey H. Kozlow, BS, Jeffrey Lisiecki, BS, Jacob Rinkinen, BS, Michael N. Terjimanian, BS, Alexis Donneys, MD, Peng Zhang, PhD, Cameron Brownley, BA, Sven A. Holcombe, PhD, Stewart C. Wang, MD, PhD, William M. Kuzon, MD, PhD.
University of Michigan, Ann Arbor, MI, USA.
PURPOSE:The component separation technique (CST) is an important option for ventral hernia repair (VHR) in complicated ventral hernias. Obesity is known to play a role in inadequate wound healing and the development of surgical site infections (SSI), but we currently lack an objective method to assess this increased risk. Our hypothesis is that morphometric measures of fat and body area obtained from routine pre-operative Computed Tomography (CT) scans are more precise quantitative measures and better associated with the risk of SSI than heuristic proxies for obesity, such as BMI, in patients undergoing VHR using a CST. Additionally, we hypothesize that adipose derived stem cells (ASCs) from patients who require CST demonstrate decreased vasculogenesis compared to age matched abdominoplasty patients.
METHODS: 111 patients underwent component release VHR (2002 to 2011). Using semiautomated analytic morphomics, the subcutaneous fat area, total body area, vertebra to fascia distance and vertebra to skin distance along the linea alba was measured between T12 and L4 of the routine pre-operative CT scans. Logistic regression analyses were used to identify factors associated with the incidence of SSI, controlling for patient age, gender, and BMI. ASCs obtained from the subcutaneous fat of CST and abdominoplasty (control) patients, were grown in standard growth medium and subjected to hypoxic conditions for 12 and 24 hours (n=3 per group). Protein was collected for Western blotting with VEGF-A antibodies. Both groups of cells were grown in Matrigel under hypoxic conditions and assayed for tubule formation after 24 hours.
RESULTS: SSIs were observed in 35% (n =39) of the population. Pearson’s correlation coefficients showed significant associations between several morphomic indices and the occurrence of SSI. Multivariate logistic regression revealed that those patients with larger vertebra to skin distance, larger vertebra to fascia distance, and larger subcutaneous fat area also had higher incidences of SSI (p=0.0369 (Table. 1), p=0.0338, and p=0.0366 respectively), controlling for patient age, gender, and BMI. Hernia patient ASCs showed decreases in VEGF-A protein expression after 12 and 24 hours of growth in hypoxic conditions compared to control ASCs (Fig. 1). Finally, the hernia patient ASCs formed fewer tubules per high-powered field than the control patient ASCs after 24 hours in hypoxia (4.33±2.27 tubules/HPF vs. 8.51±2.77 tubules/HPF (Fig. 2).
CONCLUSION:We confirm that obesity is closely related to post-operative infection. However, proxy measurements for obesity such as BMI are not as good as more specific morphomic factors in predicting the risk of SSI. At a cellular level, ASCs from hernia patients have less vasculogenic activity in hypoxic conditions, than those derived from patients without previous abdominal surgeries.
Logistic regression of abd morphomics vs surgical site infection (controlling for age, gender, BMI)
|Estimate of logistic odds ratio||p value|
|Vertebra to skin distance||0.1371||0.0369|
|Vertebra to fascia distance||0.1629||0.0338|
|Subcutaneous fat area||0.0040||0.0366|
Back to Annual Meeting Program