Split-Thickness Skin Grafts to the Foot and Ankle of Diabetic Patients

Abstract

Background: There is an increased prevalence of foot ulceration in patients with diabetes, leading to hospitalization. Early wound closure is necessary to prevent further infections and, ultimately, lower-limb amputations. There is no current evidence stating that an elevated preoperative hemoglobin A_1c (HbA_1c) level is a contraindication to skin grafting. The purpose of this review was to determine whether elevated HbA_1c levels are a contraindication to the application of skin grafts in diabetic patients. Methods: A retrospective review was performed of 53 consecutive patientswho underwent splitthickness skin graft application to the lower extremity between January 1, 2012, and December 31, 2015. A uniformsurgical techniquewas used across all of the patients. A comparison of HbA1c levels between failed and healed skin grafts was reviewed. Results: Of 43 surgical sites (41 patients) that met the inclusion criteria, 27 healed with greater than 90% graft take and 16 had a skin graft that failed. There was no statistically significant difference in HbA_1c levels in the group that healed a skin graft compared with the group in which skin graft failed to adhere. Conclusions: Preliminary data suggest that an elevated HbA_1c level is not a contraindication to application of a skin graft. The benefits of early wound closure outweigh the risks of skin graft application in patients with diabetes.

There is an alarming increase in the prevalence of diabetes mellitus worldwide, with foot ulceration as the leading cause of hospitalizations [1]. The prolonged hospitalization course and outpatient wound care not only expose patients to a higher risk of infections and complications but also incur a higher burden in terms of the cost of providing these services in the population of patients with diabetes. In addition, the higher risk of infections in these chronic wounds places the patient at higher risk for needing some form of an amputation. Thus, early wound closure is necessary to prevent further infections and, ultimately, lower-limb amputations.

Currently, there is no evidence stating that an elevated preoperative hemoglobin A_1c (HbA_1c) level is a contraindication to the application of split-thickness skin graft (STSG). Although the literature is replete in the discouragement of elective surgery in patients with poorly controlled diabetes, reporting an increase in postoperative complications in people with diabetes [1], little is known regarding STSG. Although STSG application risk is low, there is an increased hesitancy in using it to obtain quick wound coverage in these high-risk patients with diabetes. However, newer data have suggested that there is actually a low complication rate when treating the patient with a diabetic foot wound with STSG [2]. The purpose of this study was to determine whether an elevated preoperative HbA_1c level is a contraindication to the application of STSG in people with diabetes.

Material and Methods

Approval was obtained from the University of Texas Southwestern Medical Center, Dallas. A retrospective review was performed on 53 consecutive patients who underwent STSG application to the lower extremity, distal to the ankle, between January 1, 2012, and December 31, 2015, using Current Procedural Terminology code 15120. Patients were excluded if they did not have diabetes, were lost to follow-up, or had incomplete documentation. To determine outcomes, a Student t test was used to compare the mean HbA_1c level of the failed group compared with that of the healed group.

The STSG technique was standardized between the two senior authors (L.A.L., J.L.). The donor site was the ipsilateral thigh, which was prepared with mineral oil to allow for smooth harvest with the pneumatic dermatome. The dermatome setting ranged from 0.010 to 0.018 inches, with the most common thickness set at 0.014 inches.

After graft harvest, the donor site was dressed with a Tegaderm (3M, St. Paul, Minnesota). The STSG was secured with either staples or absorbable sutures and either meshed 1:1.5 or “pie-crusted” using stab incision with a scalpel to allow for fluid egress. A bolster dressing was applied that consisted of saline-moistened cotton balls, bacitracin/xeroform or a negative pressure wound therapy (NPWT) device.

Results

A total of 53 patients underwent STSG to the lower extremity between January 1, 2012, and December 31, 2015. Three patients were lost to follow-up, two had an incomplete medical record, and seven did not have diabetes, leaving 41 patients (43 surgical sites) who that met the inclusion criteria.

Based on review, 16 sites were deemed as failures of graft take based on charting that stated “graft failed.” There were 27 sites that had healed based on charting that stated “healed” or “greater than 90% take of graft.” In the patients with diabetes who failed STSG, the mean ± SD HbA_1c level was 8.4% ± 3.1%. In the diabetic patients who healed, the mean ± SD HbA_1c level was 8.7% ± 2.8%. The difference in HbA_1c levels between the two groups was not significant (P = .72). Wound size also was not significantly different in each group (P = .29) (Table 1). In addition, the mean time to heal STSG in patients with diabetes was 45.2 days, or 6.5 weeks (range, 13–111 days).

Table 1. Patients Who Healed and Failed to Heal a Skin Graft with Corresponding HbA_1c Levels

Table 1. Patients Who Healed and Failed to Heal a Skin Graft with Corresponding HbA_1c Levels

Discussion

Few studies have reported on the use of STSG to the lower extremity in the poorly controlled diabetic patient. This patient population and wound location offer a unique challenge due to the weightbearing surface of the foot, dependency, and microvascular and macrovascular insults.

Recently, Rose et al [3] indicated that there was a perceived higher incidence of failure for the application of STSG in chronic lower-extremity wounds. Those included studies from Rose et al [3], Ramanujam et al [4], Mahmoud et al [5], Puttirutvong [6], and Younes et al [7], reporting on a total of 257 diabetic patients with foot ulcers, but did not demonstrate a correlation of success or failure to HbA_1c level.

The objective of this study was to retrospectively evaluate the rate of success and failure of STSG in diabetic foot ulcers as it relates to HbA_1c level. We chose this as an indicator because 1) it was the most reported laboratory result without having to add to the cost of patient care and 2) it is the average of 3 months of glucose control, it demonstrates the patient’s level of compliance to their overall health, and we posited that it will continue to trend this way with their postoperative care.

Glucose measurements were also recorded preoperatively but were not analyzed critically because we believed that this would have much more fluctuation due to recent infection or dietary changes from hospital admission.

Preliminary data suggest that an elevated HbA_1c level is not necessarily a contraindication to the application of STSG, which is in agreement with a recent study [4]. Early wound closure could still be obtained in patients with poorly controlled diabetes once an appropriate wound base has been established, cleared of infection, and there is adequate vascular status. The mean healing time for patients with diabetes who have lower-extremity wounds in the present population was 45.9 days, or 6.5 weeks. This is comparable with other studies of wound healing in patients with diabetes. In comparison, most studies of patients without diabetes demonstrate healed STSG in 2 to 4 weeks [2].

The foot wound in patients with diabetes is usually multifactorial in nature, encompassing some form of vascular insult, sensory nerve insult, and weakened immune response to infection [8]. When these are combined, not only can the patient with diabetes incur a wound, but also this wound is more likely to become infected and lead to deeper structures (bone, soft tissue) being infected as well. This often leads to the only treatment option being limb amputation.

Given the propensity for an infection to occur in the patient who has diabetes and a foot ulcer, the need to obtain a skin barrier to cover these open wounds is paramount. One of the faster and more efficient ways of doing this is with a skin graft. It provides pliable soft tissue, with minimal donor site morbidity, and accelerates the wound-healing time [8,9].

Unique to this study was the use of NPWT as a postoperative bolster dressing for high-risk patients with diabetes. Based on our clinical experience, it seems that it did not harm or improve graft take. This study did not allow us to conclude a benefit versus risk of using NPWT, and we recommend that this should be based on the operating surgeon’s experience and comfort level at this time.

A complete review of skin graft physiology is outside the scope of this article; however, an excellent review of skin graft take was performed by Thornton and Gosman [10]. Skin grafts fail in several ways, most commonly by shear forces, infection, and seroma/hematoma formation [11,12]. We believe that the use of NPWT provides an optimal dressing for the STSG recipient site. The use of NPWT helps minimize the problems that commonly cause grafts to fail by removing excess fluid and preventing it from accumulating under the skin graft and by providing uniform compression of the graft to the wound bed [13,14]. This is especially beneficial for patients with diabetes with grafts to the foot and ankle because they are in a dependent position and more prone to fluid egress.

The use of skin grafts to close foot wounds in diabetic patients is an effective and often optimal method of obtaining wound closure in a timely manner. Although there is minimal research regarding foot wound skin grafting in patients with diabetes, a recent literature review showed that skin grafts were 78% successful at closing 90% of the wound by 8 weeks [15]. Skin grafting is often the best option in obtaining viable wound coverage because diabetes has historically been considered a relative contraindication for a free flap to the lower extremity for wound coverage [16]. Although recently this has been shown to be a faulty paradigm [17], it still remains a contraindication for many reconstructive surgeons.

We further evaluated the present data and grouped them based on varying HbA_1c levels. There were no statistically significant differences in healed versus failed skin grafts even for very high HbA_1c levels greater than 9%. We believe that this further validates our point regarding the misconception about HbA_1c levels being a contraindication for a skin graft.

There are several limitations to this study. First is the retrospective nature of the study; however, the inclusion of consecutive patients was used to limit the selection bias. Second, the follow-up and evaluation time was variable due to the nature of a teaching institution with multiple providers, but the intraoperative and initial postoperative management was standardized by standardizing the senior authors’ (L.A.L., J.L.) surgical technique for graft harvest and placement.

Conclusions

We believe that the benefits to early wound closure outweigh the risks of STSG application. Consideration should be made to quickly obtain wound coverage even in high-risk patients with elevated HbA_1c levels because these data have shown that there was no difference in the failed versus healed groups.