Do Antibiotic-impregnated Calcium Sulfate Beads Improve the Healing of Neuropathic Foot Ulcers With Osteomyelitis Undergoing Surgical Debridement?

Background. Neuropathic foot ulcers are common and difficult to treat. Calcium sulfate (CAS) has been used for antibiotic delivery in the treatment of osteomyelitis with success. Recent case series suggest success in treating osteomyelitis of the foot with CAS in a mean time to healing of 4 months; however, few studies with a control group for comparison exist. Objective. This study aims to determine if antibiotic-impregnated CAS beads improved the healing of neuropathic foot ulcers with proven osteomyelitis undergoing surgical debridement. Materials and Methods. A consecutive retrospective cohort study of 50 patients undergoing surgical debridement of neuropathic foot ulcers for osteomyelitis from December 2015 to May 2016 was performed. Exclusion criteria consisted of amputations and microbiology findings inconsistent with osteomyelitis. Patients were divided into 2 groups: the surgical debridement (SD) group was treated with SD alone and the other (CAS) was treated with debridement and implantation of vancomycin- and gentamicin-impregnated CAS beads. Results. After exclusion criteria, 42 patients were included: 13 in the SD group and 29 in the CAS group. In the SD group, the mean time to healing was 5.8 months (range, 2–9 months), and in the CAS group, it was 5.5 months (range, 2–13 months). There was no significant difference in ulcer healing (P = .81), time to healing (P = .79), reoperation rate (P = .51), length of stay (P = .74), or mortality (P = .13) between the 2 groups. Conclusions. Ulcer healing in patients treated with antibiotic-impregnated CAS beads did not show statistical significance. Healing rates in both groups were similar to those in recent literature. Surgical debridement alone may be as effective as supplementation with local antibiotics in a bioabsorbable carrier.

RELATED CONTENT

Diagnosing Diabetic Foot Osteomyelitis

Fully Synthetic Bioengineered Nanomedical Scaffold in Chronic Neuropathic Foot Ulcers

Comparison Between Tc-99m WBC SPECT/CT and MRI for the Diagnosis of Biopsy-proven Diabetic Foot Osteomyelitis

As of November 2017, the prevalence of diabetes in the United Kingdom was 3.7 million¹; diabetic foot ulcers (DFUs) are a common problem with a global prevalence of 6.3%.² This is partly due to about 50% of patients with diabetes being diagnosed with peripheral neuropathy with loss of protective sensation due to microvascular damage.³ The lifetime risk of a foot ulcer is estimated at 10% in this patient population, with a 5-year risk of amputation estimated at 70% and a 5-year risk of death at 50%.⁴ Foot ulcers are difficult to heal and carry a high recurrence rate.⁵ 

The use of antibiotics within a bioabsorbable carrier has demonstrated a reduction of dead space, improvement of local antibiotic delivery to tissues delivering higher concentrations to the wound,⁶ and eradication and prevention of bacterial biofilms.^6,7 Historically, polymethyl methacrylate (PMMA) cement beads have been used, but they carry the disadvantage of needing a second procedure to remove them.⁸

STIMULAN (Biocomposites Ltd, Keele, UK) is a bioabsorbable calcium sulfate (CAS) product prepared at the time of surgery and can be mixed with chosen antibiotics. In vitro studies^9,10 have demonstrated antibiotic elution is effective for more than 40 days. A recent study⁸ found local concentrations above the minimum inhibitory concentration for staphylococci (4 mg/L) 4 weeks after treatment for hip joint infection. Researchers have reported^9-11 success in treating tibial and femoral osteomyelitis.

There are few published studies on the outcomes of the use of bioabsorbable antibiotic-impregnated CAS beads in the neuropathic foot. A case report¹² of incision and drainage for forefoot osteomyelitis with packing of the wound with vancomycin- and gentamicin-impregnated CAS beads found the ulcer healed in 4 months. A 12-patient case series¹³ concluded CAS beads to be effective in the treatment of calcaneal osteomyelitis when combined with hydroxyapatite, with a mean time to healing of 4 months. In addition, a large case series¹⁴ of 337 patients with lower extremity osteomyelitis treated over a 5-year period found success with CAS beads in treating osteomyelitis in which 86% of patients ultimately healed; however, the time to healing was not reported. A study¹⁵ of 20 patients with forefoot ulcers undergoing treatment with vancomycin- and gentamicin-impregnated CAS beads reported a median healing time of 5 weeks. 

In a comparative case series¹⁶ of 60 patients with transmetatarsal amputation for diabetic forefoot ulcers that failed nonoperative management, the use of tobramycin-impregnated CAS beads showed no difference in length of stay, time to a dry wound, or time to healing when compared with amputation alone. Further, a study¹⁷ of 46 patients illustrated an improved average healing time for CAS beads (4 months) when compared with gentamicin-impregnated PMMA beads (6 months).

A literature search revealed no study directly comparing the outcomes of the use of CAS in osteomyelitis of the foot in neuropathic ulcers when compared with surgical debridement (SD; excluding amputation); antibiotic-impregnated CAS beads have not been proven more effective than surgical debridement. A local audit within the Trauma and Orthopaedic Surgery Division of Nottingham University Hospitals NHS Trust (Nottingham, UK) demonstrated that patients treated with amputation for foot ulcers all healed within 2 weeks and had amputations through healthy tissue far from the ulcer that grew no organism on microbiological culture. Surgical wounds unable to close surgically by definition take longer to heal as they must do so by granulation; thus, they were the subject of the present study.

This study aims to determine if antibiotic-impregnated CAS beads improved the healing of neuropathic foot ulcers with proven osteomyelitis undergoing concurrent SD when compared with SD alone.

Study design
A retrospective cohort study was performed on a consecutive group of 50 patients undergoing foot ulcer debridement. Inclusion criteria were both medical inpatients and outpatients attending the authors’ unit with deep ulcers penetrating to bone. Patients were managed by a multidisciplinary team, including an orthopedic surgeon (foot and ankle consultant), specialist nurse, occupational therapist, and consultant physician (consultant in diabetes and endocrinology). A diagnosis of neuropathy was made by the inability to perceive the force of a 5.07/10-g Semmes-Weinstein monofilament.¹⁸ Both patients with and without diabetes were included. Exclusion criteria consisted of amputations and patients that did not culture microbiological organisms from samples obtained at surgical debridement.

Patient selection
A consecutive series of 50 patients were identified from the operating room records. A retrospective analysis of medical and operative records between December 2015 to May 2016 was performed.

Outcome measures
Data were extracted for whether the ulcer ultimately healed, whether the ulcer was active at 6 months, if further surgery was performed at any point within the study period, time to healing, length of stay, and if the patient was alive at 6 months. 

Ulcers were graded using the SINBAD (Site, Ischemia, Neuropathy, Bacterial infection, Area, and Depth) score,¹⁹ which is a standardized system recommended by the National Institute of Clinical Excellence (NICE).⁴ The score range is 0 to 6, with a higher score indicating a more severe foot ulcer. Midfoot and hindfoot ulcers score higher than forefoot ulcers; ischemia is defined as the absence of palpable pedal pulses; neuropathy is defined as the loss of protective sensation; ulcer area is divided into > or < 1 cm²; and depth is divided into ulcers superficial to subcutaneous tissues and those deep reaching to muscle, tendon, or bone.

Surgical procedure
Patients were positioned on the operating table without the use of a tourniquet in most cases. The skin was prepared using povidone iodine and extremity draping. In all cases, SD was performed of all infected and nonviable tissue with multiple bony and soft tissue samples taken using separate sterile instruments to avoid cross contamination. The ulcer was then thoroughly lavaged using 0.9% saline. Those receiving CAS beads received these based on availability, not on patient or ulcer characteristics. There was no randomization. In the cases in which CAS beads were used, following the surgical debridement, 5 cc of CAS was mixed with 1 g of vancomycin and 160 mg of gentamicin. Then, 3-mm beads were prepared and packed into the clean ulcer bed. The wound was left open and dressed with either nonadhesive silicone dressings or gauze swabs and bandages. Systemic antibiotics were used following discussion with a consultant microbiologist based on organism growth and antibiotic sensitivities from operative samples and clinical features. Patients who underwent repeat debridement had the same procedure repeated. Any existing CAS beads from a previous procedure were irrigated and debrided from wounds in which sufficient time had not elapsed for the beads to dissolve.

Ethical approval
This research was registered locally under the Trauma and Orthopaedic Surgery Division of Nottingham University Hospitals NHS Trust as a service evaluation (evaluation no. 116-019c). 

Statistical analysis
Statistical analysis was performed using SPSS²⁰ (version 23; IBM Corp, Armonk, NY). Patients were analyzed in 2 groups: the first underwent SD alone and second received debridement followed by packing of CAS beads into the wound. Categorical outcomes (ulcer healed by the end of the study, active ulcer at 6 months, patient alive at 6 months, further surgery performed) were compared using a chi-squared test. Time to healing was compared using a t test, as the data were normally distributed. Length of stay and duration of antibiotics were compared using the Mann-Whitney U test, as the data were not normally distributed. A correlation analysis was performed to determine if there was a relationship between length of ulcer duration and time to healing.

Of the 50 patients identified, 8 were excluded (5 simple toe amputations, 3 had no osteomyelitis found and no growth on culture of specimens). All excluded cases healed within 2 weeks. The remaining 42 eligible patients were separated into the SD group (13) and CAS group (29) (Figure 1). Both groups had similar features with no significant differences, including age, ulcer duration at presentation, and ulcer severity as per SINBAD Classification and Score (Table 1). Of the 42 patients, 37 had diabetes (SD: 12/13, 92.3%; CAS: 25/29, 86.2%). The cause of neuropathic foot was renal failure and transplant in 1 patient (CAS group), and 4 patients had Charcot-Marie-Tooth hereditary sensory and motor neuropathy (2 in CAS group, 2 in SD group). The median duration of antibiotic treatment was 6 weeks in both groups (interquartile range, 6–12 weeks for the CAS group and 2–6 weeks in the SD group; P = .07). All patients received systemic antibiotics.

There were no significant differences in any of the outcome measures (Table 2, Figure 2 and Figure 3). The proportion of patients healed by 12 months in the SD and CAS groups was 76.9% and 72.4%, respectively (P = .81). The proportion of active ulcers at 6 months in the SD and CAS groups was 53.8% and 48.3%, respectively (P = .74). The proportion of patients alive at 6 months in the SD and CAS groups was 92.3% and 100%, respectively (P = .13). Further surgery was performed in 30.8% of the SD group and 58.6% of the SD group (P = .51). Mean time to healing in the SD and CAS groups was 5.8 months and 5.5 months, respectively (P = .79). Median length of stay in the SD and CAS groups was 16 days and 10 days, respectively (P = .74).

The most common organisms grown were gram-stain positive bacteria, usually Staphylococcus aureus. Similar organisms were grown from both patient groups (Figure 4).

A subgroup analysis revealed no differences in outcome measures comparing surgeon grade (consultant vs. registrar), use of pulsed lavage against washout using syringe, and intravenous giving set with whether the organisms were sensitive to vancomycin. Duration of ulcer at presentation did not correlate with time to healing (P = .980, R = -0.005).

There was 1 death in the SD group unrelated to the foot ulcer and not due to sepsis. There were no cases of malignant transformation.

This study aimed to determine if antibiotic-impregnated CAS beads improved the healing of neuropathic foot ulcers with proven osteomyelitis undergoing SD when compared with SD alone.

The mean time to healing of 5.7 months overall was not significantly different between the groups (CAS: mean, 5.5 months, range, 2–13, standard deviation = 2.9; SD: mean, 5.8 months, range, 2–9, standard deviation = 2.4). This is comparable to the literature with other authors reporting healing in about 4 months.^13,16,17 The patient population, however, differed from other studies as patients with vascular deficiency and sepsis were included and cases of simple forefoot amputation with wounds suitable for primary closure were excluded. Wounds were left open to granulate and CAS beads were used to pack the dead space within the soft tissue as well as the bony defects. This may explain the slightly longer healing time observed. The inclusion of patients with vascular insufficiency is significant because 31% of the patients in this study did not have palpable peripheral pulses (Table 1), reflecting the high incidence of this comorbidity that adversely effects outcomes.

Patients received antibiotic treatment individually guided by bacterial growth and antibiotic sensitivity from tissue samples taken at the time of debridement. There was no difference in the duration of antibiotics between groups. Likewise, there was no difference in the number of patients who had growth of an organism from biopsy samples taken following debridement and irrigation in the groups.

Although other authors^11-14 have found that treatment with antibiotic-impregnated CAS beads is effective in treating infection in the diabetic foot and elsewhere in the body, this study represents the first comparative study of the outcomes of treatment of infected foot ulcers that included all regions of the foot. The only other study¹⁶ comparing outcomes found in the literature search was a case series of 60 patients with transmetatarsal amputations of forefoot DFUs with failed nonoperative management. The authors compared tobramycin-impregnated CAS beads to amputation without CAS beads and found no difference in length of stay, time to a dry wound, or time to healing.¹⁶

Other studies did not have a comparison group. In a 12-patient case series by Drampalos et al,¹³ the authors treated calcaneal osteomyelitis with combined CAS and hydroxyapatite using a novel technique, achieving a mean time to healing of 16 weeks. Gauland¹⁴ evaluated a large case series of 337 patients with osteomyelitis of the lower extremity over a 5-year period and found success in the treatment of osteomyelitis with 86% ultimately healing, however, the time to healing was not reported. This study¹⁴ excluded patients suffering from peripheral vascular disease, Charcot arthropathy, sepsis, and inability to offload. In a series of 20 patients with forefoot ulcers treated with gentamicin- and vancomycin-impregnated CAS beads, Jogia et al¹⁵ found a median healing time of 5 weeks; however, they¹⁵ only included forefoot ulcers that had all underwent primary closure. 

Within the present authors’ unit, they used CAS with vancomycin and gentamicin supplementation for more than 2 years with perceived success. Their results suggest SD was as effective. Interestingly, the only other study¹⁶ comparing a group treated with and without CAS also found similar outcomes.

The cost implication of the use of CAS beads is significant. In the authors’ hospital, 5 cc of CAS cost £260 ($336), 1 g of vancomycin cost £1.97 ($2.55), and 160 mg of gentamicin cost £1.06 ($1.37). In total, the price for treatment in materials alone was £263.03 ($340) in addition to the operative time required to prepare the beads.

This study has some limitations, one of which being a small series of retrospectively analyzed patients. In the authors’ unit, gentamicin- and vancomycin-impregnated CAS beads were being introduced, thus were often used as an adjunct in the treatment of foot ulcers following surgical debridement when available. Without randomization, there is a possibility of selection bias. The use of a consecutive series reduced this bias but resulted in unequal groups with more patients in the CAS group. Since SINBAD scoring for vascular assessment was used, palpation of pedal pulses was used, which may have been inaccurate and prone to bias.

In conclusion, the use of antibiotic-impregnated CAS beads in conjunction with surgical debridement of neuropathic foot ulcers with osteomyelitis did not result in significantly faster healing, reduced length of stay, or need for further debridement undergoing surgical debridement. Further prospective investigation with randomized controlled trials is warranted to provide higher-level evidence.

Authors: Andrew Peter Dekker, BMBS, BMedSci(Hons), FRCS; Chukwudi Uzoho, MRCS; and Brigitte Scammell, DM, FRCS(Orth)

Affiliation: Nottingham University Hospitals NHS Trust, Nottingham, United Kingdom

Correspondence: Andrew Peter Dekker, BMBS, BMedSci(Hons), FRCS, Nottingham University Hospitals NHS Trust, Trauma and Orthopaedic Surgery, Derby Road, Nottingham, Nottinghamshire NG7 2UH, United Kingdom; apdekker@doctors.org.uk

Disclosure: The authors disclose no financial or other conflicts of interest.