Purpose:  Tracking surgical complications in plastic surgery is crucial to ensuring high quality care, patient safety and informed patient choice.  Unfortunately, surgical outcomes data are most often generated from a single surgeon or a single center, limiting the generalizability to other healthcare providers and settings.  Large databases present another option and allow us to evaluate outcomes in a large cohort of patient.  We tested this approach using two national databases that capture both procedural and outcomes data.  Two common procedures, breast augmentation and abdominoplasty, were selected to cross validate the data and assess the face validity of the results using this approach.

Methods: Our study population included all cases that underwent a single or combined procedure abdominoplasty or breast augmentation in the TOPS and CosmetAssure databases from July, 2003 to August, 2007.  The independent variables included patient age, patient gender, diabetes, nicotine use and facility type.  The primary outcomes of interest were the proportion of patients with hematoma, infection or DVT/PE in the two surgical groups.

Results:  The TOPS and CosmetAssure databases included 7,310 and 3,350 patients with single procedure abdominoplasty; 13,193 and 7,777 combined procedure abdominoplasty; 30,831 and 14,227 patients with single procedure breast augmentation; and 17,894 and 7,797 combined procedure breast augmentation, respectively.  Many of the reported complications were statistically higher in the TOPS compared to the CosmetAssure populations, although many of the differences were not clinically significant.  However, both reported complication rates below the published literature.  Specifically, in the TOPS and CosmetAssure populations, the complication rates for single procedure abdominoplasty were: 0.9% and 0.5% with hematoma (p = 0.29); 3.5% and 0.7% with infection (p < 0.001); 0.3% and 0.1% with DVT/PE (p = 0.05), respectively.  In the TOPS and CosmetAssure populations, the complication rates for primary breast augmentation were: 0.6% and 0.7% with hematoma (p = 0.21); 0.3% and 0.1% with infection (p < 0.001); 0.02% and <0.01% with DVT/PE (p = 0.31), respectively.  In addition, combining procedures appeared to increase the risk of postoperative complications, especially with breast augmentation.  In particular, combined compared to single procedure breast augmentation in TOPS had significantly higher rates of hematoma, infection and DVT/PE (see Table 1).   

Table 1. Breast Augmentation Complications in the TOPS Population, 2003 – 2007
Patient Characteristics	Single Procedure   (N=30,831)	Combined Procedure     (N=17,894)	P value
Age (mean, years)	33.3	39.7	<0.001
Female Gender	99.7%	99.8%	0.038
Hematoma	0.6%	0.8%	0.009
Infection	0.3%	1.2%	<0.001
DVT/PE	0.02%	0.1%	<0.001
Hospital    Ambulatory Care      Office-Based Suite    office/other	16.9%       37.7%       44.2%         1.2%	23.2%          33.8%          41.2%            1.8%	<0.001   <0.001   <0.001   <0.001

Conclusions:  Complication rates for abdominoplasty and breast augmentation were similar in both databases providing a measure of cross validation.  The low complication rates reported for abdominoplasty and breast augmentation in the TOPS and CosmetAssure populations support the safety of these procedures when performed by board-certified plastic surgeons.  However, surgeons should be aware of the higher complication rates associated with combining breast augmentation with other procedures.  These data should be used by individual practitioners for outcomes benchmarking.