Customize your JAMA Network experience by selecting one or more topics from the list below.
Kuriakose MA, Loree TR, Spies A, Meyers S, Hicks, Jr WL. Sensate Radial Forearm Free Flaps in Tongue Reconstruction. Arch Otolaryngol Head Neck Surg. 2001;127(12):1463–1466. doi:10.1001/archotol.127.12.1463
Successful rehabilitation after ablative surgery requires not only the reconstruction of 3-dimensional form but also the restoration of physiologic function.
To assess sensory recovery of reinnervated radial forearm flaps used for tongue reconstruction.
Patients and Methods
Seventeen patients, who underwent reconstruction of glossectomy defects with reinnervated radial forearm free flaps, formed the study group. Recovery of sensation was measured by both subjective and detailed objective tests 8 months after surgery. Sensory function of the flap was compared with that of the normal residual tongue or the adjacent oral mucosa and the contralateral forearm donor site.
All patients involved in this study had tongue defects of hemiglossectomy or greater and adjacent floor of the mouth. Sensory recovery was observed in all of the 17 patients within 8 months. Detailed sensory testing showed that median static 2-point discrimination, moving 2-point discrimination, and pressure sensitivity (1.2 cm, 0.8 cm, and 3.7 psi, respectively) were subjectively greater in the innervated forearm flaps than in the contralateral forearm donor site (2.3 cm, 1.7 cm, and 4.6 psi, respectively) (P=.064) and similar to those of the normal tongue (0.9 cm, 0.5 cm, and 3.6 psi).
In all modalities examined, sensate free flaps proved superior in sensory fidelity to the native forearm donor site and closely approached that of the normal tongue. Microsurgical reinnervation of flaps should be considered in tongue reconstruction.
ADVANCES AND refinements in head and neck reconstructive procedures have shifted the aim of reconstructive surgery from closure of resection defects to the more complex goal of restoration of form and function of the resected parts. With the introduction of microsurgical free tissue transfer, the range of tissues that can be transferred for the reconstruction of oral cavity defects has dramatically increased.
Innervated free flaps in tongue reconstruction have been proposed to improve deglutition, speech, and airway protection. The objective of this study was to assess the level of sensory recovery that may be achieved by the use of sensate radial forearm free flaps in tongue reconstruction.
Twenty-two consecutive patients underwent reconstruction of the tongue and adjacent floor of the mouth with reinnervated neurofasciocutaneous radial forearm free flaps. Five patients were excluded from the study, 4 because of recurrent disease and 1 because of unavailability for follow-up. All patients included in the study had undergone ablative surgery for squamous cell carcinoma except for 1 patient with mucoepidermoid carcinoma. All glossectomy defects were larger than hemiglossectomy according to the classification system proposed by Urken et al.1 Eight of these patients had received adjuvant postoperative radiotherapy, and 1 patient received preoperative radiotherapy. The demographic details of the patients are listed in Table 1.
The radial forearm free flap is harvested from the nondominant arm by previously well-described techniques2 with a few modifications to incorporate the sensory element of the flap. A rectangular skin paddle is designed to be centered over the radial artery and the antebrachial cutaneous septum. The skin paddle should be confined within the overlapping zones of the lateral antebrachial cutaneous nerve somatosome3,4 and radial artery angiosome.5,6 This extends from the flexor crease of the wrist to antecubital fossa. The medial limit is the midline of ventral forearm and laterally over the radial surface of the forearm to the lateral third of the dorsal surface. For sensate forearm free flap, the skin paddle is more lateral than that of the conventional radial forearm free flap.
Care should be taken to identify and preserve the superficial branch of the radial nerve during the subfascial dissection on the radial side. The lateral antebrachial cutaneous nerve is a continuation of the musculocutaneous nerve (C5-6). It pierces the coracobrachialis muscle and runs downward and laterally between the tendon of biceps and brachialis to reach the lateral side of the arm. At the interepicondylar line it pierces the deep fascia on the lateral side of the tendon of biceps and is continued into the forearm as the lateral antebrachial cutaneous nerve of the forearm.7 The nerve can be identified medial to the cephalic vein as it courses along with the vein distal to the antecubital fossa. Both the nerve and the vein are traced toward the skin paddle. The nerve divides into multiple smaller branches as it traverses distally, making it difficult to identify at this location. After the vascular anastomosis, the antebrachial cutaneous nerve is approximated to the cut end of the lingual nerve by interrupted epineurial sutures.
During routine postoperative follow-up, sensory recovery was evaluated in all patients. Ten of the 17 patients included in the study were available for detailed objective sensory evaluation, which was carried out by an experienced occupational therapist (S.M.). This sensory evaluation was carried out at 8 to 38 months (median, 14 months) after the surgical procedure. The following tests were performed on the central portion of the flap: (1) subjective ability to sense food, (2) sharp-dull discrimination, (3) pain sensation, (4) Semmes-Weinstein monofilament (pressure) sensation, (5) static and moving 2-point discrimination, (6) hot-cold perception, and (7) sensory recovery of the flap compared with normal residual tongue or adjacent oral mucosa and the contralateral forearm donor site.
The subjective assessment was carried out by asking whether the patient could sense food on the reconstructed tongue after mastication. To determine the sharp-dull discrimination, a 27-gauge hypodermic needle and a mirror handle were used to probe the area to be tested. Pain sensation was tested by pinprick with a 27-gauge needle. Serial stimulation with Semmes-Weinstein monofilament (North Coast Medicals, Inc, Morgan Hill, Calif) was used to determine the tactile pressure threshold. To estimate moving and static 2-point discrimination, a disk-criminator was used. Hot-cold perception was determined with cotton wool soaked in ethyl chloride and a warmed mirror handle.
All patients included in the study (N = 17) had recovery of subjective ability to perceive food in the reinnervated free flaps. All of the patients included in this study were seen monthly after surgery. From retrospective analysis of the patient records, the time course to subjective recovery of sensation was determined. In no patient was sensation present at 1 month postoperatively. In 6 patients, sensation was present at 2 months after surgery. All patients had sensation at 8 months. Overall, the mean and median time to sensory recovery was 4 and 3 months, respectively.
Results of the detailed objective sensory evaluations of the study patients are tabulated in Table 2. Restoration of sharp-dull discrimination and perception of cold and hot sensation of the new tongue were observed in all patients who were subjected to detailed sensory testing (n = 10). Comparison of the thresholds for pressure sensation estimated by Semmes-Weinstein monofilaments showed the level of recovery of the reinnervated free flap (3.78 g/mm2) to be closer to that of the normal residual tongue (3.61 g/mm2) and better than that of the contralateral forearm donor site (4.56 g/mm2). A similar observation of sensory upgrading of sensate free flaps was observed with respect to both static and moving 2-point discrimination; none of these observations reached statistical significance. The average static 2-point discrimination of the flap, normal tongue, and contralateral donor site were 12, 9, and 22.5 mm, and moving 2-point discrimination, 8, 5, and 17 mm, respectively. None of the patients included in the study reported dysesthesia.
Oral mucosal sensory feedback plays a critical role in many stomatognathic functions, such as mastication, oral hygiene, phonation, and swallowing, and can influence patient quality of life. Kapur et al8 clearly demonstrated the detrimental effects of oral mucosal anesthesia. Their study showed a proportionally worsening functional impact with an increasing area of anesthesia. Restoration of sensibility should be, therefore, one of the important components of the functional rehabilitation of glossectomy defects. This would require transfer of a composite functional unit with its own vascular supply (angiosome5,6) and innervated by a sensory nerve (neurosome3,4).
Taylor et al,3 in detailed cadaveric studies, demonstrated various neurovascular territories of the body. They observed that cutaneous nerves often run along with blood vessels in an overlapping distribution of angiosomes and neurosomes. This work suggests that, clinically, many of the currently used axial or fasciocutaneous flaps can be potentially modified as neurovascular flaps.
The successful transfer of neurosomes in extremity reconstruction is well established.9-13 David14 first reported attempts to restore sensation in intraoral reconstruction by means of a reinnervated deltopectoral flap. In this procedure, he anastomosed supraclavicular nerves supplying the flap to recipient nerves in the head and neck. He did not report results of restoration of sensation with this technique. In 1979, Franklin et al15 reported an unsuccessful attempt to restore intraoral sensation with the use of reinnervated dorsalis pedis flaps, where the superficial peroneal nerve was anastomosed to the lingual nerve. Ten years later, Matloub et al,16 with the use of reinnervated lateral arm flaps, reported the restoration of sensation in 2 of the 4 flaps used for intraoral reconstruction. Urken et al17 reported the successful transfer of a sensate radial forearm free flap for pharyngeal reconstruction through the anastomosis of the antebrachial cutaneous nerves to the great auricular nerve. With the use of the same flap, Dubner and Heller18 reported high-fidelity restoration of intraoral sensation. The most conclusive evidence of recovery of intraoral sensation was reported by Boyd et al.19 In this prospective study of 8 radial forearm free flaps, they observed the return of sensation in the reinnervated flaps approaching the quality of sensation of the adjacent normal tissue. Many of the potentially useful sensate flaps for head and neck reconstruction are listed in Table 3; some, such as the dorsalis pedis and tensor fascia lata, are included for their historical interest only.
Neurofasciocutaneous radial forearm free flaps are now considered the ideal choice of all available flaps for subtotal tongue reconstruction. The thin skin paddle with minimum subcutaneous tissue is well suited to contour the glossectomy defect while, at the same time, providing adequate bulk for reconstruction. These flaps have a very reliable vascular pedicle with an excellent vessel caliber match to donor vessels in the neck. The radial artery has a consistent angiosome, which is well documented in cadaveric studies.5 Rhee et al4 studied the extent of lateral antebracheal cutaneous nerve somatosome by intraoperative mapping. The mean ± SD surface area was 187 ± 30 cm2, distributed in an elliptical shape. There was consistency in the 15 neurosomes tested.
The results from this study have shown that, with reinnervated free flaps, sensory recovery can be expected in patients as early as 8 months postoperatively. We have documented that, by microsurgical anastomosis of the lingual nerve to the lateral antebrachial cutaneous nerve, the sensory function of the radial forearm free flap approximates that of the normal tongue. The phenomenon of sensory upgrading has been previously reported in extremity reconstruction24 and in tongue reconstruction by Boyd et al.19 The quality of sensory recovery of the reinnervated free flaps came close to that of the normal tongue as observed by Semmes-Weinstein monofilament testing and static and moving 2-point discrimination.
The improvement in tactile pressure sensation observed in the sensate flaps as compared with the contralateral forearm donor site can be explained by the higher level of amplification of signals at neuronal synapses and the larger cortical representation of the tongue in the sensory homunculus as compared with the forearm. We believe that the improvement of 2-point discrimination of forearm skin from 22.5 mm to 12 mm is due to the relative sophistication of the lingual nerve as compared with the lateral antebrachial cutaneous nerve. The capacity of the sensorium to distinguish between 2 points of stimulation is effected through a mechanism called lateral inhibition. Virtually all sensory pathways give rise to lateral inhibitory signals through interneurons. These limit the lateral spread of excitatory signals and increase the degree of contrast of the sensory pattern perceived by the sensory cortex. These inhibitory signals occur at each synaptic level: the gasserian ganglion, thalamus, and sensory cortex. Sensory fidelity is further augmented by inhibitory descending fibers from the cerebral cortex to these synaptic levels.25 We believe that the lingual nerve, with its more refined neuronal pathways, can amplify and increase the fidelity of impulses received from the forearm skin to obtain near-normal sensory function.
The return of sensation in noninnervated flaps has been well documented in head and neck reconstruction.26-28 Although spontaneous reinnervation does occur in noninnervated flaps, it takes a longer period to develop and it does not restore adequate functional sensation, nor does it provide useful tactile sensation or 2-point discrimination.29 These sensory modalities are important in a patient's ability to handle oral secretions and food boluses. A potential problem of nerve anastomosis is dysesthesia. We did not observe this phenomenon in any of our patients. Further study is necessary to determine whether better sensation translates into improved functional outcome. The problem of replication of the discrete motor functions of the tongue, of course, remains an enigma and a significant challenge to future reconstructive efforts.
In all modalities examined, sensate neurofasciocutaneous free flaps proved subjectively superior in sensory fidelity to the native forearm donor site and closely approached that of the normal residual tongue. Microsurgical reinnervation of flaps should be considered, whenever feasible, in tongue reconstruction.
Accepted for publication July 25, 2001.
Presented at the Joint Meeting of the Society of Head and Neck Surgeons and American Society of Head and Neck Surgery, Palm Beach, Fla, May 16, 1998.
A special thanks to Mark DeLacure, MD, for his interest in this project and to Nestor Rigual, MD, for his editorial expertise.
Corresponding author and reprints: Thom R. Loree, MD, Department of Head and Neck Surgery, Roswell Park Cancer Institute, Buffalo, NY 14263.
Create a personal account or sign in to: