The impact of smoking on surgical complications after head and neck reconstructive surgery with a free vascularised tissue flap: a systematic review and meta-analysis.

Reconstructive surgery with a free vascularised tissue flap is indicated in large defects in the head and neck region, which arise mostly because of head and neck cancer. Tobacco smoking is a major risk factor for head and neck cancer, and many patients undergoing reconstructive surgery in the head and neck have a history of smoking. The objective of this meta-analysis was to determine the impact of smoking on surgical complications after head and neck reconstructive surgery with a free vascularised tissue flap. A systematic review was undertaken for articles reporting and comparing the incidence of overall surgical complications after reconstructive surgery with a free vascularised tissue flap between smokers and nonsmokers. Relevant articles were searched using PubMed, Cochrane, and Embase databases, and screened for eligibility according to the PRISMA guidelines. The risk of bias analysis was conducted using the Newcastle-Ottawa quality assessment scale. A meta-analysis was performed to quantitatively compare the incidence rate of overall surgical complications, flap failure, surgical site infection, fistula, and haematoma between smokers and nonsmokers using OpenMetaAnalyst (open source) software. Only qualitative analysis was performed for wound dehiscence, bleeding, nerve injury, and impaired wound healing. Forty-six articles were screened for eligibility; 30 full texts were reviewed, and 19 studies were included in the quantitative meta-analysis. From the 19 studies, 18 were retrospective and 1 was a prospective study. In total, 2155 smokers and 3124 nonsmokers were included in the meta-analysis. Smoking was associated with a significantly increased risk of 19.12% for haematoma (95% Confidence Interval (CI): 4.75-33.49; p<0.01), and of 4.57% for overall surgical complications (95% CI: 1.97-7.15; p<0.01). No significant difference in risk was found for flap failure (95% CI: -4.33-9.90; p=0.44), surgical site infection (95% CI: -0.88-2.60; p=0.33) and fistula formation (95% CI: -3.81-3.71; p=0.98) between smokers and nonsmokers. Only for flap failure was a significant heterogeneity found (I2=63.02%; p=0.03). Smoking tobacco was significantly associated with an increased risk of overall surgical complications and haematoma, but did not seem to affect other postoperative complications. Encouraging smoking cessation in patients who need reconstructive head and neck surgery remains important, but delaying surgery to create a non-smoking interval is not needed to prevent the investigated complications. More high-quality retrospective or prospective studies with a standardised protocol are needed to allow for definitive conclusions.