Indocyanine green angiography for preventing postoperative mastectomy skin flap necrosis in immediate breast reconstruction

BACKGROUND: Breast cancer will affect one in eight women during their lifetime. The opportunity to restore the removed tissue and cosmetic appearance is provided by reconstructive breast surgery following skin-sparing mastectomy (SSM). Mastectomy skin flap necrosis (MSFN) is a common complication following SSM breast reconstruction. This postoperative complication can be prevented by intraoperative assessment of mastectomy skin flap viability and intervention when tissue perfusion is compromised. Indocyanine green fluorescence angiography is presumed to be a better predictor of MSFN compared to clinical evaluation alone. OBJECTIVES: To assess the effects of indocyanine green fluorescence angiography (ICGA) for preventing mastectomy skin flap necrosis in women undergoing immediate breast reconstruction following skin-sparing mastectomy. To summarise the different ICGA protocols available for assessment of mastectomy skin flap perfusion in women undergoing immediate breast reconstructions following skin-sparing mastectomy. SEARCH METHODS: We searched the Cochrane Breast Cancer Specialised Register, Cochrane Central Register of Controlled Trials (CENTRAL; Issue 3, 2019), MEDLINE, Embase, the World Health Organization's International Clinical Trials Registry Platform (ICTRP) and Clinicaltrials.gov in April 2019. In addition, we searched reference lists of published studies. SELECTION CRITERIA: We included studies that compared the use of ICGA to clinical evaluation to assess mastectomy skin vascularisation and recruited women undergoing immediate autologous or prosthetic reconstructive surgery following SSM for confirmed breast malignancy or high risk of developing breast cancer. DATA COLLECTION AND ANALYSIS: Two review authors independently assessed the risk of bias of the included nonrandomised studies and extracted data on postoperative outcomes, including postoperative MSFN, reoperation, autologous flap necrosis, dehiscence, infection, haematoma and seroma, and patient-related outcomes. The quality of the evidence was assessed using the GRADE approach and we constructed two 'Summary of finding's tables: one for the comparison of ICGA to clinical evaluation on a per patient basis and one on a per breast basis. MAIN RESULTS: Nine nonrandomised cohort studies met the inclusion criteria and involved a total of 1589 women with 2199 breast reconstructions. We included seven retrospective and two prospective cohort studies. Six studies reported the number of MSFN on a per breast basis for a total of 1435 breasts and three studies reported the number of MSFN on a per patient basis for a total of 573 women. Five studies reported the number of other complications on a per breast basis for a total of 1370 breasts and four studies reported the number on a per patient basis for a total of 613 patients. Therefore, we decided to pool data separately. Risk of bias for each included nonrandomised study was assessed using the Newcastle-Ottawa Scale for cohort studies. There was serious concern with risk of bias due to the nonrandomised study design of all included studies and the low comparability of cohorts in most studies. The quality of the evidence was found to be very low, after downgrading the quality of evidence twice for imprecision based on the small sample sizes and low number of events in the included studies. Postoperative complications on a per patient basis We are uncertain about the effect of ICGA on MSFN (RR 0.79, 95% CI 0.40 to 1.56; three studies, 573 participants: very low quality of evidence), infection rates (RR 0.91, 95% CI 0.60 to 1.40; four studies, 613 participants: very low quality of evidence), haematoma rates (RR 0.87, 95% CI 0.30 to 2.53; two studies, 459 participants: very low quality of evidence) and seroma rates (RR 1.68, 95% CI 0.41 to 6.80; two studies, 408 participants: very low quality of evidence) compared to the clinical group. We found evidence that ICGA may reduce reoperation rates (RR 0.50, 95% CI 0.35 to 0.72; four studies, 613 participants: very low quality of evidence). One study considered dehiscence as an outcome. In this single study, dehiscence was observed in 2.2% of participants (4/184) in the ICGA group compared to 0.5% of participants (1/184) in the clinical group (P = 0.372). The RR was 4.00 (95% CI 0.45 to 35.45; one study; 368 participants; very low quality of evidence). Postoperative complications on a per breast basis We found evidence that ICGA may reduce MSFN (RR 0.62, 95% CI 0.48 to 0.82; six studies, 1435 breasts: very low quality of evidence), may reduce reoperation rates (RR 0.65, 95% CI 0.47 to 0.92; five studies, 1370 breasts: very low quality of evidence) and may reduce infection rates (RR 0.65, 95% CI 0.44 to 0.97; five studies, 1370 breasts: very low quality of evidence) compared to the clinical group. We are uncertain about the effect of ICGA on haematoma rates (RR 1.53, CI 95% 0.47 to 4.95; four studies, 1042 breasts: very low quality of evidence) and seroma rates (RR 0.71, 95% CI 0.37 to 1.35; two studies, 528 breasts: very low quality of evidence). None of the studies reported patient-related outcomes. ICGA protocols: eight studies used the SPY System and one study used the Photodynamic Eye imaging system (PDE) to assess MSFN. ICGA protocols in the included studies were not extensively described in most studies. AUTHORS' CONCLUSIONS: Although mastectomy skin flap perfusion is performed more frequently using ICGA as a helpful tool, there is a lack of high-quality evidence in the context of randomised controlled trials. The quality of evidence in this review is very low, since only nonrandomised cohort studies have been included. With the results from this review, no conclusions can be drawn about what method of assessment is best to use during breast reconstructive surgery. High-quality randomised controlled studies that compare the use of ICGA to assess MSFN compared to clinical evaluation are needed.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.