Successful Treatment of a Metacarpal Trophic Ulcer Utilizing a Neurovascular Island Flap

Introduction

Trophic ulceration of the metacarpal, metatarsal, and digital pads is an uncommon sequela of peripheral nerve damage.^1–4 Medical management of cutaneous trophic ulceration is intensive, long term, and frequently unrewarding in both human and veterinary patients.^1,3–6 In the veterinary literature, two case series describe the use of neurovascular island flaps to provide sensitized skin coverage to trophic ulcers of the metacarpal pad. Outcomes of these five cases varied from excellent to poor.^1,4 This case report represents the modification to a previously described technique with a successful outcome for a dog.⁴

Case Report

A4-year-old, neutered male, soft coated wheaten terrier mixed-breed dog was presented for evaluation and treatment of a nonhealing ulcer on the metacarpal pad of the left thoracic limb. The ulceration had been present for 10 to 12 weeks. Approximately 3 years prior, the dog was hit by a car and suffered neurological dysfunction of the left thoracic limb, consistent with brachial plexus injury. The dog gradually recovered neurological function of the limb but had a persistent, left thoracic limb lameness. The referring veterinarian (RDVM) initially instructed the owner to manage the wound by soaking the foot in a chlorhexidine solution on a daily basis. Four weeks prior to presentation, the RDVM reevaluated the lesion on the metacarpal pad and performed a punch biopsy at the margin of the lesion. Initial histopathological evaluation of the lesion was consistent with a diagnosis of squamous cell carcinoma.

On presentation, a 4-mm, circular ulcer was seen on the proximolateral aspect of the metacarpal pad of the left thoracic limb [Figure 1]. Abnormalities in the neurological examination were limited to the left thoracic limb. The dog exhibited a normal gait; however, mild hyperextension of the left carpus resulted in a slight palmigrade stance during weightbearing. The withdrawal and panniculus reflexes and muscle tone were normal in the left thoracic limb. Atrophy of the muscles of the brachium and antebrachium was noted when compared to the corresponding musculature of the right thoracic limb. Cutaneous sensation was absent in the area of the ulcer and the lateral aspect of the metacarpal pad. Cutaneous sensation was normal proximal to the metacarpal pad and normal in the autonomous zone for the radial nerve [Figure 2].

Neuroanatomical diagnosis was consistent with a peripheral neuropathy affecting the cutaneous innervation in the distribution of the ulnar, median, and musculocutaneous nerves. The palmigrade stance suggested an abnormality in the palmar carpal ligament. The underlying cause of the palmar carpal ligament abnormality may be related to a neurological deficit or traumatic injury. It was not possible to determine if this carpal laxity was a direct result of the dog being hit by a car or was a later development, because stressed view radiographs had not been made at the time of injury; however, the injury was not believed to be the cause of the ulcer. Complete blood count and serum biochemical profile were within reference ranges, except for a mild thrombocytopenia (144,000 cells/μL; reference range 235,000 to 694,000 cells/μL) that was considered to be clinically insignificant. Based on the initial histopathological evaluation of the ulceration, fine-needle aspiration was performed on the left prescapular lymph node, and cytological evaluation was normal.

The dog was anesthetized for biopsies of the ulcer and for electrophysiological testing of the nerves supplying the ulcerated skin. Motor nerve conduction studies were performed on the ulnar and radial nerves as previously described.⁷ The radial nerve had normal motor nerve conduction velocity (mNCV). The ulnar nerve had a reduced compound muscle action potential. The mNCV of the ulnar nerve proximal to the elbow was normal but was prolonged distal to the elbow. These results were consistent with a distal neuropathy affecting the ulnar nerve. Because of the neurological deficits, a trophic ulcer was suspected. To confirm the diagnosis of squamous cell carcinoma, a wedge incisional biopsy was taken from the junction of normal and abnormal tissue in the area of ulceration on the metacarpal pad. Histopathological evaluation of the specimen was consistent with chronic, focally extensive, ulcerative, hyper-plastic, suppurative pododermatitis. The original biopsy was reviewed by two separate pathologists (neither of which was the original pathologist), who both read the sample out as consistent with chronic, ulcerative pododermatitis rather than squamous cell carcinoma.

Based on the history, neurological examination findings, electrophysiological testing, and histological evaluation of both biopsy specimens, the cutaneous ulcer was believed to be the result of chronic denervation, and a diagnosis of a trophic ulcer was made. In order to treat the ulcer, a neurovascular island skin flap was planned, which would preserve the neurovascular supply to the overlying skin. A nonadhesive, soft padded bandage was placed on the paw to protect the sutures at the biopsy site. The bandage was changed every 72 hours and maintained for 15 days until surgery was performed. The biopsy site had healed, but the ulcer remained and appeared unchanged.

Carpal arthrodesis was not considered as an initial therapeutic option, because the palmigrade stance was considered to be mild and not the cause of the ulcer. Additionally, the complication rate reported with carpal arthrodesis is high.^8,9 Instead, a neurovascular island flap was performed, and carpal arthrodesis was reserved as a salvage procedure if the flap failed to resolve the ulcer.

Neurological testing of the donor site was performed immediately prior to surgery. Stimulation of the skin on the dorsal aspect of the base of the fourth digit with forceps resulted in a withdrawal reflex and a conscious response; this indicated normal sensory innervation supplied by the lateral branch of the superficial branch of the radial nerve.¹⁰

The left thoracic limb was clipped and prepared for aseptic surgery. An Esmarch tourniquet was applied by tightly wrapping the forelimb with sterile elastic tape, ^a beginning at the toes and wrapping in a proximal direction to the midante-brachium. The elastic tape was then cut, beginning at the toes and extending proximally. A 3-cm band of tape was left at the proximal-most aspect of the bandage. This acted as a tourniquet to prevent venous congestion and reduce the arterial blood supply, thereby providing a relatively bloodless field that allowed accurate visualization of neurovascular structures. An incision was made through the skin over the dorsal aspect of the fourth metacarpal bone.

The trifurcation of the neurovascular bundle was identified and isolated. This bundle contained the lateral branch of the superficial radial nerve, the cranial superficial antebrachial artery, and the accessory cephalic vein into the dorsal common digital nerves, arteries, and veins (II, III, and IV). The neurovascular bundle containing the dorsal common digital nerve artery and vein IV was dissected from the underlying connective tissue, at the point of the trifurcation to just proximal to the base of the fourth digit. A premade paper template, patterned after the defect in the metatarsal pad, was used to outline the island of skin at the base of the fourth digit that would be used as the island flap. The outlined area of skin was incised and dissected free of the underlying fascia from distal to proximal, with care taken to elevate the neurovascular bundle with the flap. The island and its pedicle were wrapped in a saline-soaked gauze sponge while the recipient site was prepared.

The ulcer was debrided to grossly normal tissue on all margins. The path of the flap was chosen based on the shortest distance to the recipient site. An incision was made through the skin one-third of the way from the recipient site to the base of the pedicle. A subcutaneous tunnel was made under the remaining skin between the pedicle and the recipient site. The flap and pedicle were passed through this tunnel [Figure 3], the island was laid into the recipient site, and the pedicle was laid into the channel created by the skin incision.

Excess tension on the flap was alleviated by additional undermining of the trifurcation of the lateral branch of the superficial radial nerve, the cranial superficial antebrachial artery, the accessory cephalic vein, and the dorsal common digital nerves, arteries, and veins (II, III). The flap was secured in place only with simple interrupted skin sutures (4-0 nylon). The channel for the pedicle was closed using only simple interrupted skin sutures (4-0 nylon). The defect from the donor site was closed with simple interrupted sub-cutaneous sutures (3-0 polydioxanone), and the skin was apposed with simple interrupted skin sutures (4-0 nylon). A soft padded bandage and fiberglass clamshell splint¹¹ were applied in such a manner that the dog could walk on the end of the splint, reducing weightbearing on the metacarpal pad.

Hydromorphone (0.05 mg/kg intravenously q 4 hours) was administered for analgesia for 18 hours, followed by tramadol (2.5 mg/kg per os q 12 hours) for 3 days. The dog was managed in the hospital for 23 days at the owner’s request to enforce strict exercise restrictions. The dog was bearing weight on the splinted left thoracic limb during short leash walks. The bandage was changed every 24 to 48 hours. At each evaluation, the flap appeared healthy and viable; no swelling, discoloration, or discharge was noted. Five days after surgery, a 3- to 4-mm area on the proximo-medial surface of the flap appeared to be separating from the recipient site [Figure 4], but it healed without intervention. Ten days postoperatively, the flap appeared to be healing well with no evidence of necrosis or additional incisional dehiscence; all skin sutures were removed. Twelve days postoperatively, the bandage was changed from a clamshell splint to a soft padded bandage. The bandage was changed every 24 to 36 hours until it was removed permanently 23 days after surgery.

Mild pododermatitis was noted during bandage changes. This was managed successfully with cleansing, using a 2% chlorhexidine solution diluted 1:10 with tap water^b and talcum powder-coated cotton balls placed between the toes. On day 12 after surgery, a 1 × 1-cm portion of the superficial layer of the heavily keratinized epithelium of the metacarpal pad had sloughed. Also, a 0.5 × 0.5-cm portion of the superficial layer of the heavily keratinized epithelium of the first digital pad had sloughed. These areas redeveloped their tough, heavily keratinized layer sometime between bandage removal 23 days after surgery and recheck examination 55 days after surgery. The superficial sloughing was attributed to disuse of the pads. The dog was discharged 24 days postoperatively, with owner instructions to leave the dog’s paw unbandaged and to use a protective foot cover when walking the dog on rough surfaces.

The dog was reevaluated 55 days after surgery, and the owner reported the dog was using the leg better than prior to surgery. The dog did not lick the donor site excessively, and the flap site appeared normal. On examination, the donor site had healed completely. The surface of the flap appeared to have developed a heavily keratinized layer grossly similar to the normal, heavily keratinized metacarpal pad in the region [Figure 5]. For practical reasons, we were unable to confirm this change histopathologically. Sensation of the flap was not assessed. Mild hyperextension of the left carpus persisted. The dog was discharged with owner instructions to allow the dog to walk with the paw exposed and to monitor the metacarpal pad for signs of ulceration. Telephone contact was made 95 days postoperatively, and the owner reported the dog was doing well with no signs of ulceration of the metacarpal pad.

Discussion

Amodification of a previously described technique creating a neurovascular island flap was used to treat a suspected trophic ulcer of the metacarpal pad in this dog.^1,4 The modifications included passing the island flap and its pedicle through a subcutaneous tunnel and using a clamshell splint to protect the healing flap.⁴

The pathogenesis of trophic ulceration secondary to denervation of the overlying skin is not completely understood. It appears, at least in part, that loss of cutaneous sensation inhibits normal somatosensory reflexes and voluntary changes in posture, allowing for prolonged ischemia and subsequent ulceration of the desensitized tissues.^5,6,12 In the veterinary literature, reports of trophic ulcers have been limited to the metacarpal, metatarsal, and digital pads.^1–4 Medical management of trophic ulcers, both in human and veterinary patients, requires intensive, long-term nursing care and is frequently unrewarding.^1,3–6 Reported surgical therapies attempt to provide healthy donor tissue with normal sensory innervation and blood supply. These neurovascular island flaps are based on known neurovascular bundles, but success has been variable in humans and dogs.^1,4,6,13

Digital pad transposition has been used successfully to treat defects of the metacarpal and metatarsal pads, but it requires the effective amputation of at least one digit.^14,15 The use of free segmental grafts from the digital paw pads has been shown to be successful in experimentally induced metatarsal pad defects.^16,17 However, the application of such grafts to trophic ulceration of the metacarpal/tarsal pad may be difficult due to the apparent difficulties in trying to achieve a healthy bed of granulation tissue at the recipient site.^1,4 Treatment of trophic ulceration by using the fourth dorsal common digital neurovascular bundle as a pedicle for an island flap has been described.^1,4 Reported complications include subsequent self-mutilation of the fifth digit, necrosis of the flap, and superficial abrasions of the flap from heavy use. The outcomes in five cases varied from excellent to poor.^1,4 In the two successful cases of island flaps reported by Gourley et al, the dogs were reported to develop abrasions with excessive activity, indicating that the flap had not become heavily keratinized.¹

The dog of this report had a successful outcome in the short term, with few complications experienced. Sloughing occurred in the superficial layers of the heavily keratinized epithelium of the metacarpal pad and the first digital pad. These areas redeveloped their heavily keratinized layer without complication. This was considered to be a minor complication secondary to bandaging and not bearing weight on the pads. The mild pododermatitis observed during bandage changes also resolved with conservative management and was considered to be a minor complication.

Wounds of the paw pads can be protected using different types of external coaptation. Soft padded bandages or splints have been used extensively. Recent research indicates a clamshell-type splint significantly reduces the pressure applied to the metacarpal pad during ambulation, and, for this reason, a clamshell splint was used for this dog to protect the flap.¹¹ Clamshell splints were not used in the previous case series reported, and only one island flap failed;⁴ however, the authors feel the clamshell splint protected the flap during healing. The development of what grossly appeared to be a heavily keratinized surface on the island flap, similar to the surface of the normal pad, is noteworthy.

Conclusion

The surgical procedure described requires careful dissection of delicate neurovascular structures, but it is relatively straightforward. This technique should also be adaptable for treating cutaneous defects involving the hind paw. Although an uncommon sequela of peripheral neuropathy, neurovascular island flaps appear to be a useful option for the treatment of chronic trophic ulcers of the metacarpal pad when they are unresponsive to conservative therapy. Owners should be made aware of the risks of flap failure, the potential for future ulcers to result from excessive exercise, and the prolonged and involved care needed during healing.