Smoking as a Risk Factor for Forehead Flap Wound Outcomes: An Analysis of 1030 Patients

The forehead flap is recognized as the ideal donor for reconstruction of large nasal defects due to its significant color and texture match, vascularity, and capacity for subtotal and total nasal resurfacing. Supplied by vessels from the supratrochlear, supraorbital, superficial temporal, postauricular, and occipital arteries, the forehead flap can be transposed on a wide range of pedicles; the most common of these is the vertical paramedian forehead flap, an axial flap based on a single supratrochlear artery and the design of choice in patients with major nasal defects.¹

Often, the patients undergoing these procedures are smokers, and the problematic effects of smoking have been reported in a number of scientific and medical data.^1-4 High-level animal studies show that cigarette smoke decreases flap survival through impaired oxygenation, introduction of free radicals, and arterial vasoconstriction, leading to lower arterial oxygen tensions in tobacco-smoking animals,^5,6 yet comparable human studies have not been well defined. Furthermore, despite the general agreement on the detrimental effects of smoking, the limited prospective studies in flap procedures have shown no significant difference in outcomes between smokers and nonsmokers.^7,10-14

To our knowledge, the American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) is the most abundant database of multiinstitutional, high-powered retrospective studies. The ACS NSQIP plastic surgery dataset contains over 250 patient demographic variables, including 30-day postoperative outcomes, and may provide detailed knowledge about the burden of smoking on forehead flap procedures.⁸

The primary objective of our study is to use the ACS NSQIP database to examine smoking as a preoperative risk factor for complications following reconstructive surgery employing forehead flaps.

Study Sample

After obtaining approval from our local institutional review board, we retrospectively analyzed data collected from the ACS NSQIP from 2005 to 2019. Each record in the ACS NSQIP data set contains Health Insurance Portability and Accountability Act–compliant preoperative, intraoperative, and 30-day postoperative data. The detailed process of data collection, including the complete list of variables collected, are publicly available by the ACS NSQIP.⁸

A search for Current Procedural Terminology code 15731 identified 1128 forehead flap cases performed by the plastic surgery departments of ACS NSQIP–participating institutions. Patients were excluded from our study if they had undergone other surgical procedures in the preceding 30 days or had preoperative wound infection or sepsis; patients with missing information, such as age, sex, wound classification, or American Society of Anesthesiology (ASA) physical status classification, were also excluded.

Baseline Characteristics

Unidentifiable patient demographics analyzed included age and sex. Preoperative health variables included such conditions as diabetes, hypertension, bleeding disorders, and chronic steroid use. Operative variables analyzed included classifications of wound type (clean, clean/contaminated, contaminated, dirty/nfected) and ASA classification (1-2 or 3-4). A review of demographic information is summarized in Table 1.

Dependent Variables

Dependent variables were postoperative wound complications, including surgical-site infection (superficial incisional, deep incisional, and organ/space surgical-site infection), wound dehiscence, and flap failure.

Independent Variables

The primary independent variable for this study was current smoking status, defined by the ACS NSQIP database as patients who reported smoking cigarettes within 12 months before admission for surgery. Definitions of all variables used in this study can be found in the 2020 ACS NSQIP Participant Use Data File.⁹

Statistical Analysis

Forehead flap cases were identified and separated into 2 cohorts based on current smoking status. Baseline characteristics for cohorts were summarized, and unadjusted univariate analyses were performed using χ², Fisher exact test, or Wilcoxon rank sum test. Demographics and operation variables were analyzed for current nonsmoker and current smoker categories as displayed in Table 1. All demographic, preoperative health status, and operative variables with a P ≤ .05 on univariate analysis were included in the multivariate logistic regression analysis. Multivariate logistic regression models, which were adjusted for sex, age, diabetes, hypertension, bleeding disorders, chronic steroid use, wound class, and ASA classification, were fitted to evaluate the association between smoking and development of wound complications as seen in Table 2. The significance level was set at α = .05, and odds ratios (ORs), 95% confidence intervals (CIs), and P values were calculated. Additionally, cohorts were 1:1 nearest-neighbor propensity score–matched without replacement, using a 0.15 caliper width and controlling for age, sex, diabetes, hypertension, bleeding disorders, chronic steroid use, wound class, and ASA classification.  Wound complication rates were evaluated among the matched cohorts. Data analysis was performed using SAS version 9.4 M3 for Linux.

Sample Characteristics

A total of 1030 forehead flaps cases met the inclusion criteria of the study period from 2005 to 2019. Of these, 789 procedures were performed on patients who were smokers, and 241 were performed on nonsmokers. Differences in demographics and preoperative conditions among cohorts were reported as follows. The nonsmoker cohort had a significantly older mean age (68.7 vs 56.3 years; P < .0001), higher percentage of patients with hypertension (56.4% vs 41.9%; P < .0001), and more patients on chronic steroid therapy (4.7% vs 1.25%; P = .0126) compared with the current smoker cohort. After propensity score matching, 236 cases were identified per cohort. Once cohorts were matched, nonsmokers were used as controls to compare the cohorts to ensure that the matching process was successful. This analysis was performed by using the t test to compare each cohort’s mean age, and chi-square test and Fisher exact test to compare categorical variables, verifying matching with no significant differences between the groups. Demographics for matched and unmatched groups are depicted in Table 1.

Primary Endpoint Data

The incidence of wound complications occurred in 31 (3.01%) of all forehead flap procedures in our sample; of these, 21 (2.7%) were in nonsmokers, and 10 (4.1%) were in current smokers, which was not statistically significant (P = .0807).

Multivariate logistic regression analysis for complication development among cohorts adjusted for sex, age, ASA class, chronic steroid use, diabetes, hypertension, bleeding disorders, and wound class revealed no significant difference in risk for development of wound complications (OR 1.297, 95% CI 0.55-2.9; P = .5174). Additionally, there were no significant differences in frequencies of wound complications when comparing matched cohorts. Detailed results are summarized in Table 2.

Interpretation

In our study, 23.4% of patients who underwent forehead flap procedures identified as current smokers based on the ACS NSQIP criteria. This value is considerably greater than the estimated national average (14.5%) of adults aged 45 to 64 years who identify as smokers, based on a study conducted by the Centers for Disease Control in 2020.¹⁵ This overrepresentation of smokers in patients requiring facial reconstruction stresses the damaging health effects associated with smoking; yet, the findings of this study demonstrate that smoking status does not significantly increase postoperative complication rates in patients who undergo reconstruction with forehead flaps when controlled for preexisting conditions.

Comparison With the Literature

In the retrospective study by Little et al,¹⁶ the investigators performed a chart review on patients who underwent forehead flap surgery for nasal reconstruction from 1995 to 2008. This study showed that smokers had significantly increased odds of developing flap necrosis (OR 4.34, 95% CI 1.26-14.93; P = .02). Due to the nature of chart review, specific details of patient smoking habits such as the number of cigarettes smoked per day or smoking frequency could not be analyzed. In a study measuring 854 consecutive transpositional or free flap procedures, Kroll et al⁷ showed no significant differences in flap survival among smokers, ex-smokers, and nonsmokers (P = .084); however, these cases were limited to abdominal flap procedures.Additionally, like the study by Little et al, the number of cigarettes, or smoking habits were not measured in the study.^7,16 A well-cited study on smokers by Golminz et al¹⁰ compared the risk of necrosis in head and neck region flaps and full-thickness grafts in 4 patient populations divided by cigarette use. The study compared necrosis rates in low-level smokers (1 pack or less per day), high-level smokers (1 pack or more per day), former smokers, and never smokers. In this study, 916 flaps and full-thickness grafts in 220 patients were analyzed. The results showed former smokers (relative risk 1.4, 95% CI 0.6-3.2) and low-level smokers (relative risk 1.1, 95% CI 0.2-6.1) were at a negligible increased risk for necrosis that was not significantly different from never smokers (P > .1); however, smoking cessation for 2 days before surgery and 7 days after surgery was associated with lower rates of flap necrosis. Rohrich⁴ recommends not pursuing flap procedures on current smokers and advises a minimum of 4 weeks preoperative smoking cessation.However, based on the scarcity of studies in the literature, the period of preoperative smoking cessation and or the number of pack years prior to cessation required to impair wound healing remains unclear.⁶

A possible explanation to the unclear effect smoking has on facial wound healing may be explained by the robustness of the facial circulation. Anatomic studies by McCarthy et al¹⁷ mapped out the vigorous arcade of vessels supplied by the supratrochlear, infratrochlear, supraorbital, dorsonasal, and angular branches of the facial artery. The center of this axis of collateral circulation is positioned on the medial canthal region, the anatomic center of the face, thus forming a network of stout arterial coverage capable of providing strong perfusion to all facial sections.Additionally, the concept of vascular delay, also referred to as the “delay phenomenon,” may be a possible explanation to outcomes in forehead flaps being minimally affected by the harms of smoking. Vascular delay describes the process of rendering a tissue ischemic to increase vascularity before tissue transfer.¹⁸ This concept is applied to the forehead flap via purposeful pedicle division at least 1 to 3 weeks following initial flap transfer. After cutting the donor site blood supply, the distal portion is rendered ischemic and theoretically pushed into a state of amplified vascular ingrowth.¹⁹

Extending this thinking, Kent and Defazio²⁰ described a successful case of nasal reconstruction in a patient who smoked 4 packs of cigarettes per day.In this case, a paramedian forehead flap was used with an additional staged vascular delay procedure. The authors described creating lateral incisions along the axial border of the flap, thus elevating the flap from the loose areolar tissue for 1 week before subsequent flap elevation, transfer, and traditional pedicle division to maximize the benefits of vascular delay.²⁰

Among other factors, the large variability in smoking habits and the subjective nature of patient-reported smoking history make objective measurements of smoking difficult to assess. This may be a reason for the mixed results on smoking risk when pertaining to flap procedures.

In our study that reviewed the outcomes of 241 smokers, we did not find a significantly higher likelihood of complications than that reported in nonsmokers 30 days after undergoing a forehead flap. On the basis of our findings, cigarette smoking may not pose a significant risk of deleterious outcomes when concerning forehead flaps.

This study has several limitations worth noting. The ACS NSQIP database contains only 30-day postoperative data, which may lead to a potential underestimation of complication development. Additionally, the ACS NSQIP–participating institutions are predominantly hospitals; thus, our results may be less generalizable to other practice settings. Furthermore, for this specific analysis, the ACS NSQIP database lacks nicotine biomarkers, thus failing to quantify individual smoking habits or smoking cessation rates in this period, which may potentially underestimate the outcome data of this study.

Smoking has previously been demonstrated to be a risk factor for plastic surgery procedures. In forehead flap facial reconstruction, our study showed that smokers did not have worse outcomes compared with nonsmokers. Though it is still advised to counsel patients on smoking cessation given the multiple health benefits, a history of past or current smoking status should not preclude a patient in need of undergoing a forehead flap based on the theoretical increased risk of complications alone.

Affiliations: ¹Division of Plastic Surgery, Department of Surgery, Tulane University School of Medicine, New Orleans, LA; ²Department of Dermatology, University of Illinois at Chicago, Chicago, IL; ³Department of Plastic and Reconstructive Surgery, Ochsner Clinic Foundation, New Orleans, LA

Correspondence: Salomon Puyana, MD, MS; salomonpuyana@gmail.com

Disclosures: The authors disclose no financial or proprietary interest in the subject matter of this article.