Treating deep sternal wound infection with pectoralis major flap transposition: a systemic factor analysis of efficacy and safety

Abstract

Background: Deep sternal wound infection (DSWI) is a life-threatening postoperative complication of cardiac surgery. Currently, there are many therapies used to treat patients with DSWI. However, none of these therapies have been shown to be the optimal choice for patients with DSWI. Additionally, these methods may require additional facilities which limit their widespread use. Therefore, we aimed to investigate the effectiveness, safety, and risk factors related to the prognosis of DSWI patients treated with pectoralis major flap transposition (PMFT), a more concise method.

Methods: A retrospective, systemic analysis was conducted of DSWI patients at the Beijing Anzhen Hospital from January 2010 until December 2020. All the patients were diagnosed with DSWI according to the relevant guidelines, and treated with PMFT. The patients were divided into the following two groups based on their prognosis after treatment with PMFT: (I) the wound healing (WH) group; and (II) the delayed wound healing or death (DWHD) group. All the participants were followed up for 1 year.

Results: In total, 9.7% (76/785) of the DSWI patients experienced DWHD in the present study. The all-cause mortality rate was 3.7% (29/785). While 90.3% (709/785) of the patients achieved WH after undergoing PMFT. The multivariate logistic regression model indicated that patients with diabetes mellitus, obesity, a history of smoking, abnormal liver function, anemia, chronic infection, immune disease, hypothermia, a longer gap time, and patients requiring extracorporeal membrane oxygenation (ECMO) assistance, salvage surgery, and secondary cardiac surgery were more likely to experience DWDH.

Conclusions: This study showed PMFT was a safe and effective method for treating DSWI after cardiac surgery. Patients with risk factors, such as those mentioned above, require more attention. Prospective studies should be conducted to explore the relationships among the novel risk factors and DSWI.

Keywords: Deep sternal wound infection (DSWI); cardiac surgery; efficacy; pectoralis major flap; safety.

Figure 1

The process of pectoralis major flap transposition. (A) Deep sternal wound infection; (B) the wound was irrigated with hydrogen peroxide or diluted iodine; (C) the necrotic skin, peripheral hyperplastic tissue, and muscle were cleaned out, and the sternal wires or sternal fixators were removed entirely; (D) the wound was irrigated with hydrogen peroxide or diluted iodine, and then repeatedly rinsed with sterile saline; (E) the pectoralis major muscle was separated from the midline of the sternum; (F) the pedicled muscle was drawn and curled to the wound surface to cover the sternum defect; (G,H) interrupted sutures and tension-reducing treatment of the wound.

Figure 2

The flow of participants through the study. DSWI, deep sternal wound infection; PMFT, pectoralis major flap transposition; WH, wound healing; DWHD, delayed wound healing or death.

Figure 3

The univariate logistic regression model revealed that many factors were related to delayed wound healing or death in DSWI patients. The categorical data are reported as the n (%), and the continuous data are reported as the median [IQR]. WH, wound healing; DWHD, delayed wound healing or death; CI, confidence interval; COPD, chronic obstructive pulmonary disease; NYHA, New York Heart Association; BIMA, bilateral internal mammary arteries; SIMA, single internal mammary arteries; IABP, intra-aortic balloon pump; ECMO, extracorporeal membrane oxygenation; DSWI, deep sternal wound infection; IQR, interquartile range.

Figure 4

The preoperative multivariate logistic regression model revealed that many factors were related to delayed wound healing or death in DSWI patients. The categorical data are reported as n (%), and the continuous data are reported as median [IQR]. WH, wound healing; DWHD, delayed wound healing or death; CI, confidence interval; DSWI, deep sternal wound infection; IQR, interquartile range.

Figure 5

The operation-related multivariate logistic regression model revealed that many factors were related to delayed wound healing or death in DSWI patients. The categorical data are reported as n (%), and the continuous data are reported as median [IQR]. WH, wound healing; DWHD, delayed wound healing or death; CI, confidence interval; IABP, intra-aortic balloon pump; ECMO, extracorporeal membrane oxygenation; ICU, intensive care unit; DSWI, deep sternal wound infection; IQR, interquartile range.

Figure 6

The discrimination power of both models was assessed using the ROC curve, and the area under the curve of the ROC for the preoperative predictive model was 0.811. Youden index was 0.143; sensitivity was 0.632, and specificity was 0.846. ROC, receiver operating characteristic; AUC, area under the curve.

Figure 7

The discrimination power of both models was assessed with the ROC curve, and the area under the curve of the ROC for the operation-related predictive model was 0.885. Youden index was 0.111; sensitivity was 0.882, and specificity was 0.843. ROC, receiver operating characteristic; AUC, area under the curve.