Postoperative Seromas after Abdominoplasty: A Retrospective Analysis of 494 Patients and Possible Risk Factors

Abstract

Sir: Abdominoplasty is a procedure that manifests postoperative seromas in 5 to 22 percent of cases.1,2 Different risk factors have been investigated. Although age and body mass index have been associated with their occurrence,3 progressive tension sutures gave contrasting results, and the use of drains or concomitant liposuction was not related to seroma occurrence.4 In this retrospective analysis, we tried to correlate the risk of seromas with the amount of flap resected and of fat aspirated with liposuction. We also analyzed the influence of associated liposuction, progressive tension sutures, fibrin tissue adhesives (Tissucol), patient smoking status, and two techniques for flap raising (diathermocoagulation versus cold knife). Data were retrospectively collected from the personal archive of two surgeons (A.A. and V.C.) working at the Dolan Park Hospital, in Bromsgrove, United Kingdom, and at the Plastic Surgery Department of the University “Tor Vergata,” in Rome, Italy. We excluded from the analysis morbidly obese and postbariatric patients who had undergone panniculectomy following massive weight loss. A total of 494 patients who underwent full abdominoplasty were analyzed (A.A.: September of 2004 to December of 2007; V.C.: January of 2001 to December of 2007). Descriptive statistics and clinical characteristics are summarized in Table 1.Table 1: Descriptive Statistics and Clinical CharacteristicsWe recorded 23 seromas (4.7 percent) and 34 hematomas (6.9 percent). Wound infections were present in 60 patients (12.1 percent). They occurred after a mean period of 8 ± 3 days. The most common organism isolated was Staphylococcus epidermidis. No cases of deep vein thrombosis or pulmonary embolism were observed. The chi-square and Mann-Whitney tests confirmed that groups (seromas versus nonseromas) were homogeneous for all variables analyzed, except for the amount of flap resected during the abdominoplasty (Table 1). The analysis of smoking status in both groups produced no significant differences, nor did the amount of flap aspirated with liposuction, the use of tissue adhesives, or the use of diathermocoagulation versus the cold knife approach. The comparison between groups produced a prognostic cut-off value for the amount of flap removed (Figs. 1 and 2). This value (700 g) was determined with receiver operating characteristic curves to find the greatest areas of sensitivity and specificity (area = 0.665). According to this cut-off, the relative risk for the incidence of postoperative seromas in patients who removed more than 700 g of fat was 3.8 compared with the others [(15/167)/(8/327)] (chi-square test; p < 0.001).Fig. 1.: Scatter plot with cut-off line for the amount of flap resected of patients who experienced seromas (red circles) versus those who did not (blue circles).Fig. 2.: Area of receiver operating characteristic (ROC) curves refers to cut-off of tissue removed. Diagonal segments are produced by ties.Results of our study suggest that the amount of fat removed during surgery could be an important factor for the occurrence of postoperative seromas, increasing their risk of occurrence almost four times when the quantity removed exceeded 700 g. We previously demonstrated that the quantity of fat removed influences the risk of pulmonary embolism when it exceeds 1500 g,5 and our personal observations also suggest a relationship with the occurrence of postoperative surgical-site infections. No other factor was associated with the occurrence of seromas, especially concomitant liposuction or the use of progressive tension sutures. However, although based on a large series, our analysis is retrospective in nature and the results obtained need to be confirmed in future prospective trials. A. Araco Dolan Park Hospital Bromsgrove, United Kingdom G. Gravante Department of Surgery Whipps Cross University Hospital London, United Kingdom F. Araco Dolan Park Hospital Bromsgrove, United Kingdom R. Sorge Department of Human Physiology Laboratory of Biometry University Tor Vergata Rome, Italy V. Cervelli Department of Plastic Surgery University Tor Vergata Rome, Italy