Images processed using ImageJ, (National Institutes of Health), an open-source image processing software program. Areas of interest are outlined, and corresponding pixel areas are tabulated.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
David AP, Heaton C, Park A, Seth R, Knott PD, Markey JD. Association of Bolster Duration With Uptake Rates of Fibula Donor Site Skin Grafts. JAMA Otolaryngol Head Neck Surg. 2020;146(6):537–542. doi:10.1001/jamaoto.2020.0160
Is longer bolster duration associated with improved split-thickness skin graft (STSG) uptake in fibula free flap donor sites?
In this cohort study including 42 adults receiving fibula free flaps, based on photographs taken 1 month postoperatively, 14-day bolster duration was associated with a better STSG uptake rate than 5-day bolster duration (77.5% vs 59.9%) with a moderate effect size. Patients with the highest Adult Comorbidity Evalution-27 scores were also found to have poorer rates of STSG uptake with a moderate effect size.
Larger studies are needed, but the findings of this cohort study suggest that longer bolster duration may be associated with improved outcomes for patients undergoing STSGs.
The fibula free flap donor site is associated with both short-term and long-term morbidity. Split-thickness skin graft (STSG) loss can lead to long delays in donor site healing and is associated with significant adverse sequelae. Patients may experience initial good STSG uptake after bolster removal but may have subsequent partial or total loss related to contact pressure or shearing.
To determine if increased duration of bolster use is associated with increased STSG uptake rates following fibula free flap reconstruction.
Design, Setting, and Participants
This retrospective cohort study included patients 18 years and older undergoing fibula free flap reconstruction following head and neck extirpative surgery at a tertiary care academic medical center from May 2013 to March 2019. The donor sites were photographed 4 weeks postoperatively, and areas of graft uptake were measured using image processing software. The baseline demographic, comorbidity, and operative characteristics were also collected.
A fine mesh gauze with 3% bismuth tribromophenate and petrolatum blend bolster was sutured over leg STSGs placed on fibula free flap donor sites intraoperatively, and the ankle and lower leg were immobilized for 5 days in a plaster splint. Bolsters were either removed on postoperative day 5 or 14. Thereafter, the STSGs were covered with a petroleum and bismuth gauze and a cotton dressing.
Main Outcomes and Measures
Rates of donor site infection and STSG percentage uptake at 4 weeks.
Of the 42 included patients, 31 (74%) were male, and the mean (SD) age was 62.1 (13.1) years. A total of 20 patients were included in the 5-day group, and 22 were included in the 14-day group. The 14-day bolster group had a higher mean percentage skin graft uptake rate compared with the 5-day bolster duration group (77.5% vs 59.9%), with an effect size of −0.632 (95% CI, −1.260 to −0.004). Patients with Adult Comorbidity Evalution-27 scores of 3 had poorer rates of STSG uptake compared with patients with Adult Comorbidity Evalution-27 scores of 0 to 2 (65.9% vs 82.9%), with an effect size of 0.599 (95% CI, −0.191 to 1.389). No donor site infections were noted in either group.
Conclusions and Relevance
Fourteen-day bolster application to the fibula free flap donor site was associated with better STSG uptake rates than 5-day bolster application.
From its early use in 1989 by Hidalgo1 to successfully reconstruct a series of mandibular defects, the fibula free flap has become an important reconstructive option for osseous defects of the mandible and maxilla.2 It has been used for defects caused by malignant tumors, benign tumors, osteoradionecrosis, bisphosphonate-related osteonecrosis of the jaw, and osteomyelitis.3 Advantages of the fibula free flap include its long pedicle length, ability to use a 2-team approach for simultaneous resection and flap harvest, length of bone stock (up to 25 cm), a bicorticocancellous structure that can accept dental implants, the ability to harvest a skin paddle, and acceptable donor site morbidity.4
The most common donor site complications include delayed wound healing and skin graft loss.5-7 In a 2012 systematic review,8 early donor site wound complications, defined as delayed wound healing or wound infections within 30 days of surgery, occurred in 19% of sites closed with skin grafts. Patients undergoing neoadjuvant chemotherapy experience an even greater risk of delayed wound healing.6 Other donor site complications include compartment syndrome, chronic pain, gait abnormality, ankle instability, limited ankle range of motion, reduced muscle strength, claw toe, weakness in great toe dorsiflexion, and sensory deficits.8 The incidence of compartment syndrome is rare (a rate of 0% to 1% is reported in the literature) and is avoided in many cases by using a skin graft for closure.9
Split-thickness skin grafts (STSGs) are frequently needed at the donor site following fibula harvest. However, their use is associated with prolonged healing time—34 days on average10—and the creation of a secondary donor site if the STSGs are not harvested from the fibula skin paddle.11 It has been our practice to secure a bolster over the skin graft and immobilize the ankle in a plaster lower leg splint intraoperatively, with both remaining in place until 5 days after surgery. These measures minimize movement and prevent fluid accumulation beneath the skin graft.12 Our center has not used negative pressure dressings, which are associated with mixed results and therefore limited adoption.7,13 In this study, we sought to determine the association of bolster duration with donor site wound healing. We hypothesized that maintaining the bolster for 14 days compared with 5 days would lead to decreased shearing and result in improved skin graft survival.
This retrospective cohort study included 42 adult patients undergoing fibula free flap reconstruction for head and neck defects at a tertiary academic medical center between May 2013 and March 2019. Duration of bolster application changed during the study period as the participating surgeon altered his technique based on disappointing success rates. Inclusion criteria included the need for STSG for fibula donor site closure. Exclusion criteria included antecedent fibula trauma or surgery, history of organ transplant, medical condition requiring immunosuppressive medication use, or preoperative albumin level less than 2.0 g/dL (to convert to grams per liter, multiply by 10). Baseline demographic and clinical characteristics were collected from the electronic medical record and a proprietary microvascular free flap head and neck reconstruction REDCap database (Vanderbilt University). These factors included age, sex, preoperative American Society of Anesthesiologists (ASA) score, Adult Comorbidity Evalution-27 (ACE-27) score, surgical indication, STSG source, primary ischemia time, secondary ischemia time, smoking status (current, former, or never), current alcohol use, body mass index, and severe obesity. No patients undergoing fibula free flap reconstruction had vascular disease noted on their preoperative lower extremity computed tomographic angiography or magnetic resonance angiography. These factors were chosen because they have been associated with increased donor site complications.5-7,14 Patients were excluded if donor site photographs were not taken at the 4-week postoperative visit. Study approval was obtained from the University of California, San Francisco, Institutional Review Board, and informed consent was waived because deidentified data were used.
The STSGs were harvested either from the fibula skin paddle (to avoid a secondary skin graft donor site) or from the patient’s ipsilateral thigh. The skin grafts were harvested with a dermatome set to a depth of 0.020 inches or 0.51 mm and were pie-crusted prior to inset. They were trimmed and secured to the size of the fibula donor site defect with 4-0 chromic sutures (Ethicon). The bolster was then made using sterile mineral oil–soaked cotton wrapped with Xeroform gauze (Covidien), a fine mesh gauze impregnated with 3% bismuth tribromophenate and petrolatum blend. This material helped to provide a bacteriostatic barrier that retained moisture and heat. The bolster was secured over the STSG using tie-over 2-0 Perma-Hand silk sutures (Ethicon) (Figure 1), and a posterior plaster slab splint was then fashioned and secured to the lower extremity with an elastic bandage to immobilize the ankle. On postoperative day 5, the splint and in-dwelling suction drain were removed for all patients. The bolster was either removed at that time or at the first postoperative visit, approximately 14 days after surgery, depending on the senior surgeon’s practice pattern. Once the bolster was removed, the STSG was then covered with Xeroform gauze and wrapped securely with a cotton gauze dressing and replaced daily.
Recipient site percentage STSG uptake rate at approximately 4 weeks after surgery was the main outcome measure. This was determined using photographs taken of the donor site typically at the second postoperative clinic visit. Percentage graft uptake was measured using ImageJ (National Institutes of Health), an open-source image processing software program. Areas of graft uptake and graft loss were outlined by one of us (A.P.D.) and agreed on by 3 of us (R.S., P.D.K., and J.D.M.) who were all blinded to the treatment group of each patient. The pixel area was measured and then a percentage for the graft uptake was calculated (Figure 2). Additional outcomes included the rate of complete STSG loss and number of donor site infections treated with oral antimicrobials.
Patient demographic characteristics and comorbidities, operative characteristics, and postoperative outcomes were compared between the 5-day and 14-day bolster groups using R version 3.6.2 (The R Foundation). Descriptive statistics including mean and SD as well as median and interquartile range were reported for the quantitative variables, and counts and frequencies were reported for categorical variables. The association of key variables with percentage STSG uptake at 4 weeks was determined using Hedges method because the 2 groups had different sample sizes. A negligible effect size was defined as 0.10 or greater; small effect size, 0.2 or greater; moderate effect size, 0.5 or greater; and large effect size, 0.8 or greater. For this analysis, certain variables had to be dichotomized, including patient age of 65 years or older vs younger than 65 years, ASA score of 1 or 2 vs 3 or 4, ACE-27 scores of 3 vs 0 to 2, and current/former smokers vs never smokers. Effect sizes were calculated using the effsize package in R. A sample size calculation was not performed prior to initiating the study.
Of the 42 included patients, 31 (74%) were male, and the mean (SD) age was 62.1 (13.1) years. The 5-day bolster duration group included 20 patients and the 14-day bolster duration group included 22 patients. Squamous cell carcinoma extirpation was the most common indication for fibula free flap reconstruction. The baseline patient demographic characteristics, comorbidities, surgical indications, and perioperative characteristics are summarized in Table 1.
Longer bolster duration and higher ACE-27 scores had moderate effect sizes on percentage STSG uptake measured at 1 month. The 14-day bolster group had a higher mean percentage skin graft uptake rate compared with the 5-day bolster duration group (77.5% vs 59.9%; mean difference, 17.5%), with an effect size of −0.632 (95% CI, −1.260 to −0.004). Patients with an ACE-27 score of 3 had poorer rates of STSG uptake compared with patients with ACE-27 scores of 0 to 2 (65.9% vs 82.9%; mean difference, 17.0%), with an effect size 0.599 (95% CI, −0.191 to 1.389). Patient sex, older age (age of 65 years or older ), ASA score, current or former smoking status, current alcohol use, and STSG donor site had small to negligible effect sizes (Table 2). There were 2 cases (10%) of total graft loss in the 5-day group and 1 case (5%) of total graft loss in the 14-day group. There were no postoperative wound infections. In a subgroup analysis of patients with ACE-27 scores of 3 (18 patients in the 5-day bolster group and 16 patients in the 14-day bolster group), those in the 14-day bolster group had a higher mean STSG uptake rate (75.7% vs 57.2%; mean difference, 18.5%), with an effect size of −0.641 (95% CI, −1.341 to 0.060).
Delayed wound healing with or without skin graft loss is the most common complication affecting fibula donor sites.14 In a systematic review including 10 studies,8 STSG recipients had a mean incidence of partial skin graft loss of 8.1% and total graft loss of 4.7%. Participants in our series had a greater percentage skin graft uptake when the bolster was in place for 14 days rather than 5 days.
Prior study has illustrated that tie-over bolsters are beneficial for STSG uptake, but the underlying mechanism of action appears to be different than originally thought. In 1929, Blair and Brown15 suggested that pressure dressings applied an even pressure over the skin graft and provided fixation and thus maximized skin graft take. It has been proposed that the tie-over bolsters apply a pressure in the range of the capillary closing pressure, approximately 25 mm Hg, to reduce hematoma and seroma formation. However, in a 2003 study by Seymour and Giele,16 calibrated pressure transducers were used to measure the pressure at the graft wound–bed interface, and they found that no significant pressure was exerted. They hypothesized that tie-over bolster dressings primarily provided fixation of the skin graft instead.
In their 1929 study, Blair and Brown15 also recommended that skin graft pressure dressings remain in place for 4 to 5 days and upwards of 10 days when necessary. In our review of the literature regarding skin grafted fibula donor sites, we found that many studies do not explicitly mention the duration of skin graft bolster placement.1,9,11,17-20 Regarding the minority that do mention the bolster duration, 4 studies kept bolsters in place for 5 days,6,7,21,22 1 study for 7 days,23 1 study for 10 days,13 and another study for 14 days.10
It is our clinical observation that although skin grafts appeared to have excellent uptake rates when the bolsters were removed after 5 days, the ultimate uptake rate at the 2-week and 4-week follow-up visits were consistently lower than initially anticipated. It is our supposition that when bolsters are removed on postoperative day 5, the grafts are exposed to repetitive shearing, even though they are covered with Xeroform dressings and wrapped snugly with cotton dressings. We believe that by maintaining the bolster for a longer duration, the skin grafts remain fixed, and shear stresses are reduced. Moreover, the bolster protects the STSGs from the daily dressing changes, which may also cause injury. Repetitive contact pressure is therefore reduced because the skin grafts have consolidated their wound healing and neovascularization, leading to an increased likelihood of successful graft uptake.
Negative pressure dressings, also known as vacuum-assisted closure devices, have also been used on skin-grafted fibula free flap donor sites. These devices promote wound healing by increasing local nutrient blood flow, promoting granulation tissue formation, and decreasing bacterial loads.24 The effects of conventional bolsters on skin graft healing of fibula free flap donor sites have been compared with negative pressure dressings in 2 studies.7,13 Ho et al13 found no statistically significant difference in partial or total graft loss, wound infections, healing problems, or days to complete healing when comparing a negative pressure dressing placed for 5 days with a conventional tie-over bolster dressing placed for 10 days. Bach et al7 showed that 5-day use of a negative pressure dressing compared with a traditional bolster for 5 days had lower mean graft loss rates (19% in the negative pressure dressing group vs 37% in the traditional dressing group). The negative pressure dressing group reached the complete healing threshold more quickly when defined as no longer needing any additional dressings or wound care (67 days in the negative pressure dressing group vs 163 days in the traditional bolster group). Further study comparing negative pressure dressings to standard bolsters will be valuable to corroborate these findings.14
In this study, patients with higher ACE-27 scores were found to have lower rates of STSG uptake. This finding is consistent with prior studies that validate the ACE-27 score as a predictor of postoperative complications in patients undergoing surgical treatment of head and neck squamous cell carcinoma25 and those undergoing head and neck free flap reconstruction.26-29 In our subgroup analysis of patients with ACE-27 scores of 3, the 14-day bolster duration group was still found to have a greater rate of STSG uptake compared with the 5-day group with a moderate effect size. One may infer that a 14-day bolster should be considered, especially in patients with greater comorbidities and at greater risk of poor STSG uptake.
The strength of this study includes the use of image processing software to measure the percentage graft uptake. This could be taken a step further if a linear measurement reference scale was included in the photographs, as the software could be capable of calculating areas of skin graft uptake and of delayed healing. Our study has limitations. This study is limited by its retrospective nature, as it relied on the documentation in the electronic health record. A potential source of bias was that not all patients in the 14-day bolster group had their bolsters removed on exactly postoperative day 14; this ranged from 12 to 18 days but was a mean of 14.0 days. Additionally, the 4-week postoperative photography from which we calculated percentage graft uptake was not always taken on exactly the same day; this ranged from 22 to 37 days but was a mean of 30 days (Table 1). For future studies, a randomized prospective study with a larger sample size will provide a higher level of evidence to answer whether duration of bolster is related to STSG uptake as well as a sample size calculation prior to initiating the study. In the analysis, accounting for laboratory test results that have been correlated with postoperative wound healing, like thyroid-stimulating hormone levels, prealbumin levels, or albumin levels, may also be important.30,31 Furthermore, patients’ evaluation of short-term and long-term skin graft outcomes and the burdens of delayed wound healing, such as continued STSG wound care, would also be worth investigating in a future study.
In this study, delayed wound healing was the most common complication reported in fibula free flap donor sites. Bolsters that remained in place for 14 days were associated with greater percentage STSG uptake at the 1-month postoperative time point than bolsters in place for 5 days, even in patients with higher ACE-27 scores. These findings promote the importance of prolonged skin graft immobilization on wound healing at the fibula free flap donor site, especially in those affected by a greater degree of medical comorbidity.
Accepted for Publication: February 18, 2020.
Corresponding Author: P. Daniel Knott, MD, Division of Facial Plastic and Reconstructive Surgery, Department of Otolaryngology–Head and Neck Surgery, University of California, San Francisco, 2233 Post St, 3rd Floor, San Francisco, CA 94115 (email@example.com).
Published Online: April 16, 2020. doi:10.1001/jamaoto.2020.0160
Author Contributions: Drs David and Knott had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: David, Heaton, Knott.
Acquisition, analysis, or interpretation of data: David, Park, Seth, Knott, Markey.
Drafting of the manuscript: David, Knott, Markey.
Critical revision of the manuscript for important intellectual content: David, Heaton, Park, Seth, Knott.
Statistical analysis: David.
Administrative, technical, or material support: Park, Knott, Markey.
Study supervision: Heaton, Seth, Knott, Markey.
Conflict of Interest Disclosures: None reported.
Create a personal account or sign in to: