+
MODEL
Journal of Plastic, Reconstructive & Aesthetic Surgery (2015) xx, 1e11
REVIEW
A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits? Martijn Baas a, Liron S. Duraku a, Eveline M.L. Corten, Marc A.M. Mureau* Department of Plastic and Reconstructive Surgery, Erasmus MC Cancer Institute, University Medical Centre Rotterdam, Rotterdam, The Netherlands Received 29 December 2014; accepted 27 April 2015
KEYWORDS Tongue reconstruction; Free flaps; Nerve reinnervation; Sensory recovery
Summary Background: Tongue reconstruction after (hemi)glossectomy including sensory recovery is challenging. Although sensory recovery could improve functional outcome, no consensus on the need for reinnervation of the neo-tongue exists. Therefore, a systematic review was performed to determine if sensory reinnervation of free flaps in tongue reconstruction is better than no sensory reinnervation. The secondary study aim was to assess the effect of sensory reinnervation on overall functional outcome, such as speech and deglutition. Methods: Seven databases (Embase, Medline, Web of Science, Scopus, PubMed publisher, Cochrane, and Google Scholar) were searched. Studies that reported the effect of sensory reinnervation on overall functional outcome were identified. Results: Fourteen articles were included in the systematic review, concerning a total of 271 tongue reconstructions. Free flaps that were used were the radial forearm (RF) flap (n Z 137), the anterolateral thigh (ALT) flap (n Z 65), the rectus abdominis (RA) flap (n Z 20), and the tensor fascia latae (TFL) flap (n Z 5). Seven out of seven articles directly comparing sensory reinnervation with no sensory reinnervation revealed superior sensibility in the reinnervated group. Moreover, the innervated RF and ALT flaps showed superior recovery of sensibility compared to other flaps used for the reconstruction of hemiglossectomy as well as total glossectomy defects. There are indications that sensory reinnervation may have a beneficial effect on overall tongue function. Age, smoking, and sex did not affect sensory recovery. Four out of five articles showed that postoperative radiotherapy does not have a longterm adverse effect on sensory recovery.
* Corresponding author. Department of Plastic and Reconstructive Surgery, Erasmus MC Cancer Institute, University Medical Centre Rotterdam, PO Box 2040, 3000 CA Rotterdam, The Netherlands. E-mail address: [email protected] (M.A.M. Mureau). a Both authors contributed equally to this article. http://dx.doi.org/10.1016/j.bjps.2015.04.020 1748-6815/ª 2015 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
2
M. Baas et al. Conclusions: Sensory reinnervation of free flaps in the reconstruction of (hemi)glossectomy defects improves sensory recovery; however, evidence for beneficial effects on function is poor. ª 2015 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
Introduction The tongue is an important muscle, with numerous functions, such as articulation, deglutition, and taste. Impairment of tongue function can severely affect quality of life.1 Therefore, it is paramount to restore the tongue’s main function after oncologic (hemi)glossectomy. Because of its complex anatomy, it is challenging to restore bulk, mobility, and sensibility in order for the reconstructed tongue to be functional. Because of advances in microvascular tissue transfer, various free flaps for tongue reconstruction are available.2 Small defects can usually be closed primarily with good function.3,4 After hemiglossectomy, the appropriate reconstruction of choice includes a thin fasciocutaneous flap such as the radial forearm (RF) flap or a thin anterolateral thigh (ALT) flap. However, larger defects need more bulk to restore deglutition and articulation. In these situations, thicker fasciocutaneous or musculocutaneous flaps such as the ALT and rectus abdominis (RA) flap may be indicated.3,4 Interestingly, sensory recovery is a factor that is often neglected in the reconstructive algorithm,5e7 whereas it is supposed to be an essential factor for the proper function of the tongue. Previous articles showed that sensory reinnervation of free flaps in tongue reconstruction may have an advantage in the recovery of articulation, deglutition, sensibility, and even quality of life.8e10 However, still no consensus exists on the need for reinnervation of the neotongue. Therefore, a systematic review was performed to determine the advantages and reasons for sensory reinnervation of free flaps in tongue reconstruction. The primary research question of this review was whether sensory reinnervation of free flaps in tongue reconstruction leads to better sensibility of the neo-tongue compared with no sensory reinnervation. In addition, the effect of recipient nerves and factors such as age, tobacco use, sex, and radiation history on the return of sensibility was investigated. The secondary study question was whether the improved sensibility of free flaps also implies improved functional outcome of the reconstruction, such as speech and deglutition.
Methods The search strings that were used to search in seven different databases (Embase, Medline (OvidSP), Web of Science, Scopus, PubMed publisher, Cochrane, and Google Scholar) are listed in Appendix 3. A combination of search criteria was used to identify all articles concerning
oncological tongue as well as intraoral reconstructions with either innervated or non-innervated free flap. The initial search was performed on April 2014, and it was monitored during the review progress. Subsequently, two reviewers (M.B. and L.S.D.) performed a manual secondary selection based on the following inclusion criteria for our primary and secondary outcome measures. Eligibility criteria were formulated to select articles with comparable, preferably standardized, measures of reinnervation. The criteria for eligibility were as follows: Only original articles studying patients (no reviews) and written in English were included At least one of the study groups consisted of tongue reconstructions Tongue reconstructions were performed with free flaps only Articles had to contain objective sensory testing modalities Only articles with a design classification of levels IeV were included, as classified by Jovell and Navarro-Rubio (Table 1). Bias was assessed using the Cochrane tool for assessing the risk of bias, addressing sequence generation, allocation, blinding, incomplete outcome data, selective outcome reporting as well as other sources of bias.11
Table 1 Classification of strength of evidence by Jovell and Narvarro-Rubio. Level
Strength of evidence
Type of study design
I
Good
Meta-analysis of randomized controlled trials Large-sample randomized controlled trials (N > 25 for each group) Small-sample randomized controlled trials (N < 25 for each group) Non-randomized controlled prospective trials Non-randomized controlled retrospective trials Cohort studies Caseecontrol studies Noncontrolled clinical series; descriptive studies Anecdotes or case reports
II
III
Good to fair
IV V VI VII VIII IX
Fair Poor
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
Review on reinnervation of tongue reconstruction free flaps Finally, all differences between reviewers were resolved by consensus. The high heterogeneity in the population and outcome measures of the included articles precluded statistical analyses of pooled data. Therefore, conclusions were made on the basis of key findings in the included articles. Our protocol was not registered in PROSPERO. The Preferred Reporting Items for Systematic Review and MetaAnalyses (PRISMA) guidelines were followed, and the checklist is available in the online supplements of this article (Appendices 1e5).
Results Patient and reconstruction characteristics The systematic search identified 952 articles. After exclusion of 525 duplicate articles, 14 articles were selected using the inclusion criteria (Figure 1).3,4,8,9,12e21 There was a high risk of bias in the individual articles mainly due to poor study design (Appendix 5); no articles were excluded. A total of 271 tongue reconstructions were performed using 137 RF flaps, 65 ALT flaps, 20 RA flaps, and five tensor fascia latae (TFL) flaps. Other (pedicled) flaps that were also used for tongue reconstruction served as controls, but they are not discussed in great detail. The sex of the patient was reported in 197 out of 271 patients; 146 patients were male and 51 patients were female. Patient age in the included articles ranged from 29 to 78 years. All resections were performed to treat malignant tumors. Although most
Figure 1
3 authors reported resecting at least half of the mobile tongue (hemiglossectomy), the defect size was highly variable and ranged from 20 to 140 cm2. The use of (neo) adjuvant radiotherapy was reported in 10 articles, and it ranged from 15.4% to 100% of the included subjects.3,4,8,13,14,17e21 The prevalence of smoking in the studied populations was only reported twice, and it was 39.3% and 47.5%.20,21
Sensory recovery after tongue reconstruction with different free flaps RF flap Nine articles described the sensory recovery of the neotongue after reconstruction using an RF flap. A total of 136 patients were examined, of which 72 flaps were sensory reinnervated and 64 flaps were used as non-sensate flaps. Three articles compared the use of sensate with nonsensate RF flaps in tongue reconstruction.5,6,8 All authors reported superior sensibility in sensate compared with nonsensate RF flaps, assessed with multiple objective sensory modalities (Table 2). In addition, Katou et al. showed that sensate RF flap tongue reconstructions also resulted in the earlier recovery of sensibility (mean, 13 months) than nonsensate RF flaps (mean, 25.4 months).6 When a sensate RF flap was compared to the healthy contralateral tongue, temperature discrimination and two-point discrimination showed no statistically significant differences, indicating complete recovery of sensibility for those modalities in the sensate RF flap.5
Flowchart regarding the selection of included articles by PRISMA standards.
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
4
M. Baas et al. Table 2 Details on reconstruction, tested modalities, and outcome of all included articles on RF free-flap reconstruction of the tongue. Author
Population
Non-sensate versus sensate RF free flap Boyd et al.8 Eight sensate RF flaps 10 non-sensate RF flaps 10 historical controls (PM flaps)
Tested modalities
Outcome
Hot/cold perception, Semmes eWeinstein tests, sharp/dull discrimination, two-point discrimination, and perception of food presence.
Sharp/dull discrimination was present in 75% versus 10% (sensate vs. non-sensate). Hot ecold perception: 100% (both) versus 50% (hot) and 40% (cold). Two-point discrimination: 14.25 g/mm2 (3.25e31.5) versus 27 g/mm2 (3.25e31.5). Perception of food presence: 7/10 versus 3/10.b Sharp/dull discrimination was present in 4/4 versus 4/9 patients (sensate vs. nonsensate). Hot/cold discrimination 4/4 versus 2/9. Innervated flaps were tested after a shorter follow-up period (13 vs. 25.4 months). Using immunohistochemistry, more fibers with better structured arrangement were detected in sensate reconstructions. Perception of tactile stimuli was slightly better in the sensate group (89% and 79% for central and peripheral areas compared to 71% for both in the non-sensate group). Sharp/ dull discrimination and cold perceptions were superior (78% vs. 43% and 100% vs. 57%, respectively) in sensate reconstructions. Subjective speech was superior (78% vs. 43%) in the sensate group. For swallowing, a minimal difference was reported (61% vs. 57%).b
Katou et al.9
Four sensate RF flaps Nine non-sensate RF flaps
Hot/cold perception, touch perception, sharp/dull discrimination, and immunohistochemistry.
Biglioli et al.12
Nine non-sensate RF flaps Seven sensate RF flaps
Hot/cold perception, touch perception, sharp/dull discrimination, perception and localization of tactile stimuli, speech and swallow function, and deglutition pressure.
Non-sensate RF Free Flap Close et al.17 Four fasciocutaneous RF flaps Four fasciocutaneous ALT flaps Three pedicled musculocutaneous PM flaps One pedicled musculocutaneous LD flaps
Hot/cold perception, Semmes eWeinstein tests, sharp/dull discrimination, two-point discrimination, and speech and swallow function.
10 out of 12 patients had restoration of sensibility after at least 6 months (4e24 months). Touch sensation recovered in 8/8 FCFF and 2/4 MCF (p Z 0.09), sharp/dull discrimination 6/8 FCFF and 2/ 4 MFC. Two-point discrimination returned in 5/8 FCFF and 1/4 MCF; 7/8 FCFF versus 2/4 MCF patients had satisfactory articulation and adequate oral intake. Articulation and swallowing correlated statistically to return of sensibility.
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
Review on reinnervation of tongue reconstruction free flaps
5
Table 2 (continued ) Author Shindo et al.
19
Sabesan et al.20
Population
Tested modalities
Outcome
Nine fasciocutaneous RF flaps Nine osteocutaneous fibula flaps
Hot/cold perception, touch perception, sharp/dull discrimination, two-point discrimination.
23 fasciocutaneous RF flaps Six gastro-omental free flaps 10 jejunal flaps
Hot/cold perception, touch perception, sharp/dull discrimination, and two-point discrimination.
No return of sensibility was reported in the two reconstructed tongues after total glossectomy. In partial lesions, sensibility was reported in all five patients after 6 (1), 7 (1), and 12 months.3 The area of sensibility was largest in hot temperature testinga. The mean scores for RF, gastroomental, and jejunal flaps for the different modalities were (p-values represent test of RF vs. gastro-omental and jejunal flaps); Light touch: 6.04 versus 4.1 versus 3.52 (p Z 0.084). Sharp prick 6.84 versus 4.83 versus 3.72 (p < 0.05). Hot temp sensation 6.82 versus 4.93 versus 4.14 (p < 0.05). Cold temp sensation 4.52 versus 2.81 versus 2.1 (p < 0.05). Twopoint discrimination. 20.31 mm versus 23.86 mm versus 23.28 mm (p < 0.05).
Sensate RF Free Flap 16 sensate (lingual) RF flaps Santamaria et al.21 Six sensate (inferior alveolar) RF flaps, three sensate (posterior auricular) RF flaps, two sensate (cervical plexus) RF flaps one sensate (hypoglossus) RF flaps
Kuriakose et al.13
17 sensate RF flaps
Hot/cold perception, touch perception, Semmes eWeinstein tests, sharp/dull discrimination, two-point discrimination.
Hot/cold perception, Semmes eWeinstein tests, sharp/dull discrimination, two-point discrimination, and subjective perception of food.
Two-point discrimination in sensate reconstruction was comparable to the contralateral tongue (largest difference was observed in the dorsal aspect of the tongue 4.4 1.6 mm vs. 2.6 0.6 mm). Light-touch sensation was statistically significantly diminished in the ventral (reconstructed) tongue indicating loss of protective sensibility (4.8 0.8 mm vs. 2.6 0.2 mm). Pain sensation was significantly decreased (p < 0.05) on the floor of the mouth as compared to other tongue surfaces. Hot and cold perception was 15%b less than control values. Sensory recovery was observed in all 17 patients within 8 months. Two-point discrimination (moving and static) and pressure sensitivity were greater than the contralateral donor site (1.2 vs. 2.3 cm, 0.8 vs. 1.7 cm and 3.7 psi vs. 4.6 psi), however, not (continued on next page)
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
6
M. Baas et al. Table 2 (continued ) Author
Loewen et al.14
Population
Eight sensate RF flaps Eight matched healthy controls
Tested modalities
Hot/cold discrimination, touch perception, Semmes eWeinstein test, two-point discrimination, form recognition, texture recognition, taste, and swallow and speech function.
Outcome significantly different (p Z 0.064). Values were similar to normal tongue (0.9 cm, 0.5 cm, and 3.6 psi) For nearly all modalities, there was a significant difference (p < 0.05) with healthy controls (two-pot reconstruction 20 vs. 90% correct response, light touch 50 vs. 90% correct response, temperature sensation 50 vs. 90% correct response, form recognition 40 vs. 60% correct response). There was no significant difference only in texture recognition and taste. Significant differences (p < 0.05) in intelligible speech were found for both words and sentences in favor of the healthy control group. Masticatory efficiency was greater in healthy controls
RF: radial forearm, PM: pectoralis major, ALT: anterolateral thigh, LD: latissimus dorsi. FCFF: Fasciocutaneous Free Flap. MCF: Musculocutaneous Flap. a Specific values of test were not published. b Original article only provides percentages for these parameters.
In three articles, the sensibility of non-sensate RF flaps was compared to other non-sensate flaps in tongue reconstruction.13,15,16 The other flaps described were the ALT, fibula, and gastro-omental and jejunal flaps. In contrast to other non-sensate flaps, there was recovery of sensibility in all non-sensate RF flap tongue reconstructions. The last three articles exclusively examined sensate RF flaps in tongue reconstruction.9,10,17 Loewen et al. showed that the sensibility of sensate RF flaps in tongue reconstruction never matches the sensibility of healthy controls.10 By contrast, Santamaria et al. showed a comparable two-point discrimination for the sensate RF flap as compared with the contralateral tongue; however, lighttouch perception, pain perception, and hot and cold discrimination remained inferior.17 Kuriakose et al. reported full restoration of sensibility in the reconstructed tongue with sensate RF flaps as compared to the contralateral donor site.9 ALT flap Five articles described the sensory recovery of the tongue after reconstruction using an ALT flap.3,4,11,13,14 A total of 65 patients were examined, of which 40 flaps were sensate and 25 flaps were used as non-sensate flaps (Table 3). Three articles compared sensate ALT flaps with nonsensate ALT flaps in tongue reconstruction.3,4,14 All articles reported superior sensibility of innervated ALT flaps in light touch, pain sensation, temperature, SemmeseWeinstein monofilament tests, and two-point discrimination as
compared with non-sensate ALT flaps. In addition, in these articles, the return of sensibility in non-sensate ALT flaps did not occur in the majority of patients. One study compared the sensibility of sensate ALT flaps to the donor site of the flap and to the tip of the preoperatively native tongue.15 This report revealed that sensate ALT flaps show inferior sensibility compared to the preoperative tip of the tongue. However, it was also found that the sensibility of sensate ALT flaps did recover to the level of the preoperative donor site.15 The last study16 compared non-sensate ALT flaps to other non-sensate flaps for the reconstruction of the tongue. The other flaps used were the RF, the lattisimus dorsi (LD), and the pectoralis major (PM) flap. This report revealed that in non-sensate tongue reconstruction, all non-sensate ALT flaps had gained sensibility of light touch, and over 7/12 patients obtained sharp/dull and twopoint discrimination.17 In non-sensate PM flaps, no recovery of sensibility was reported, and in non-sensate LD flaps, all but temperature discrimination recovered. RA flap Only one article used the RA flap as a sensate flap in tongue reconstruction.14 This article compared sensate (five patients) with non-sensate RA flaps (10 patients). Additionally, both sensate (eight patients) and non-sensate ALT flaps (five patients) were examined. Sensate RA flaps showed a statistically significant better sensibility compared to non-sensate RA flaps in the majority of the tested sensory modalities (moving and static two-point
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
Review on reinnervation of tongue reconstruction free flaps
7
Table 3 Details on reconstruction, tested modalities, and outcome of all included articles on ALT free-flap reconstruction of the tongue. Author
Population
Non-sensate vs. sensate ALT flap Kimata et al.18 Eight sensate ALT flaps, Six non-sensate ALT flaps, Five sensate RA flaps 10 non-sensate RA flaps
Tested modalities
Outcome
Hot/cold perception, touch perception, Semmes eWeinstein tests, sharp/dull discrimination, and two-point discrimination.
Up to 87.5% sensibility of light touch (vs. 30% non-sensate) and hot/cold discrimination (vs. 0% non-sensate) in sensate ALT flapsa; 7/8 patients had intact two-point discrimination (vs. 1/6 non-innervated) in sensate flaps. Sensate ALT flaps were able to detect lower pressures (6.62 g vs. 184.7 g, p < 0.05) Innervated flaps were superior in two-point discrimination (0/ 5 vs. 5/8 points) (mean 2.6 0.9 mm), Semmes eWeinstein (Tip: log force 3.8 vs. 6.8, p < 0.05), pain perception (0/5 vs. 8/8), and temperature differentiation (0/5 vs. 5/6). No significant difference in speech scores was observed (3.7 0.8 vs. 3.3 0.5). Swallow scores were significantly better in the innervated group (5 2.5 vs. 2.6 0.9 points). Innervated flaps were superior in two-point discrimination 0/ 10 versus 9/11 (mean 2.6 0.9 mm), Semmes eWeinstein (Tip: log force 3.8 vs. 6.8, p Z 0.005) pain perception (0/10 vs. 11/11), and temperature differentiation (1/10 vs. 11/ 11). No significant difference in speech scores was observed (3.7 0.8 vs. 3.3 0.5). Swallow scores were significantly better in the innervated group (5 2.5 vs. 2.6 0.9 points).
Yu et al.3
Eight sensate ALT flaps Five non-sensate ALT flaps
Hot/cold perception, Semmes eWeinstein tests, two-point discrimination, pain sensation, and speech and swallow function.
Yu et al.4
11 sensate ALT flaps 10 non-sensate ALT flaps
Hot/cold perception, Semmes eWeinstein tests, two-point discrimination, pain sensation, and speech and swallow function.
Non-sensate ALT Flap Four fasciocutaneous RF flaps Close et al.17 Four fasciocutaneous ALT flaps Three pedicled musculocutaneous PM flaps One pedicled musculocutaneous LD flaps
Hot/cold perception, Semmes eWeinstein tests, sharp/dull discrimination, two-point discrimination, and speech and swallow function.
10 out of 12 patients had restoration of sensibility after at least 6 months (4e24 months). Touch sensation recovered in 8/8 FCFF and 2/4 MCF (p Z 0.09) Sharp/dull discrimination 6/8 FCFF and 2/ 4 MFC. Two-point discrimination returned in 5/8 FCFF and 1/4 MCF; 7/8 FCFF vs. 2/4 MCF patients had satisfactory articulation and (continued on next page)
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
8
M. Baas et al. Table 3 (continued ) Author
Population
Tested modalities
Outcome adequate oral intake. Articulation and swallowing correlated statistically to return of sensibility.
Sensate ALT Flap Longo et al.15
Seven sensate ALT total tongue reconstruction Six sensate ALT subtotal tongue reconstruction
Hot/cold perception, touch perception, Semmes eWeinstein tests, two-point discrimination, and speech and swallow function
Full recovery of hot/cold perception, pain perception, and two-point discrimination was observed in sensate reconstruction to the level of the donor site discrimination (Tip vs. donor site: 4.08 3.09 vs. 4.47 2.15). Normal diet was reached in 46.2%, mild impairment in diet in 53.8%. Normal intelligible speech in 53.8%, acceptable speech in 46.2%a.
RF: radial forearm, RA: rectus abdominis, PM: pectoralis major, ALT: anterolateral thigh, LD: latissimus dorsi. a Original article only provides percentages for these parameters.
discriminations and touch and pain perception).14 Only for hot and cold discrimination, there was no difference between the sensate and non-sensate RA groups. TFL flap There is only one article that examined the TFL flap as an option for the reconstruction of the tongue.16 Primary closure and pedicled pectoralis musculocutaneous flaps were the comparison groups. Clinical evaluation was performed by means of speech, oral pressure, cineradiography, and electromyography at 10 and 24 months post operatively. The free TFL flap showed superior clinical outcomes: deglutition pressure had recovered to 60% in the TFL group compared to 20% and 505 in the PM flap and primary closure group, respectively, with faster swallowing time and lesser residual barium at cineradiography. On speech evaluation, excellent or good intelligibility scores were achieved by all five patients in the TFL group, whereas in the other groups, a total of three patients were rated as poor (2/5 in the PM flap group and 1/5 in the primary closure group).12,16
Effect of recipient nerve and patient characteristics on sensory outcome Recipient nerves Different recipient nerves and neurorrhaphy techniques (end-to-end or end-to-side) may be used for flap reinnervation in tongue reconstruction. The majority of articles used the lingual nerve as the primary recipient nerve. Santamaria et al. were the only ones who studied different recipient nerves for flap reinnervation in tongue reconstruction (lingual, inferior alveolar, posterior auricular, cervical plexus, or hypoglossal nerve).17 They also investigated whether there was a difference between end-to-end and end-to-side repairs. They reported that two-point
discrimination, light touch, pain, and temperature perception of the RF flap were statistically better after coaptation to the lingual and inferior alveolar nerves compared to the other nerves. Interestingly, there existed no statistically significant difference in sensory recovery between the end-to-end and end-to-side techniques when the lingual or inferior alveolar nerves were used as recipient nerve. Patient characteristics Five articles examined the effect of various patient characteristics on the sensory recovery of free flaps in tongue reconstruction.3,4,13,16,17 No statistically significant relationships between sensory recovery, age, sex, or tobacco use were reported. Sabesan et al.16 as well as Santamaria et al.17 could not find statistically significant differences between males versus females, patients older versus younger than 60 years, and smokers versus nonsmokers. Close et al. did not address tobacco use or sex; however, they also could not find a relationship between age and return of sensibility.13 In four articles, postoperative radiotherapy did not seem to affect sensory recovery of free flaps in tongue reconstruction in the long term (after at least 12 months).3,4,10,16 However, two of those articles did show inferior results in the short term (at 6 months).3,4 One other article found that free flaps showed inferior light-touch sensation and pain and temperature perception after postoperative radiotherapy at a mean follow-up of 18.2 months (range, 6e32 months).17
Effect of sensory recovery on overall functional outcome Functional outcomes were reported in eight of 14 articles.3,4,12e17 Most functional results were assessed using
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
Review on reinnervation of tongue reconstruction free flaps non-standardized questionnaires on swallowing and dietary limitations and intelligibility of speech. Although superior results were obtained following sensate flap reconstruction, most articles did not correlate this to functional outcomes. Close et al. did prove a statistically significant correlation between return of sensibility and articulation as well as swallowing in non-sensate RF flaps.13 Yu et al. reported better swallowing scores after innervated ALT flap tongue reconstruction; however, these differences were not statistically significant.3,4 Loewen et al. noted that, although sensory outcomes after free-flap tongue reconstruction remained suboptimal compared to healthy tongues, no statistically significant differences were found in taste and texture recognition compared to healthy tongues.10
Discussion The comprehensive search in seven databases together with the individual manual selection by two authors makes it unlikely that articles concerning our research question(s) may have been missed. However, the quality of the results of our review is highly dependent on the limited number of articles and the low sample sizes examined in the selected articles. We found no randomized controlled trials or other articles in which there was a blinding of outcomes that we reviewed. We chose not to exclude any articles based on their risk of bias, because the risk of bias was judged to be equal for articles on sensate and non-sensate reconstructions as well as comparative studies. The RF and ALT flaps were the most frequently used flaps in the articles we reviewed. The RF flap is very suitable because of its thinness and pliability. The ALT flap, just like the RA and TFL flaps, is larger and bulkier, and therefore it is most commonly used to reconstruct larger tongue defects. The articles reviewed did not clearly describe the rationale for free-flap choice. However, based on the defect sizes derived from the included articles, it did not appear that flap choices were primarily based on defect size. Other reasons for flap choice could include patient comorbidities, desire for flap reinnervation, concern about donor site complications, and the desire to perform the oncological resection and harvest of the free flap simultaneously. When these parameters are considered, the RF and ALT flaps could be more favorable for tongue reconstruction. Direct comparison of sensate with non-sensate free-flap tongue reconstruction reveals superior sensory recovery after flap reinnervation in all included articles.3,4,8,9,12,18 In addition, the superiority in sensibility of sensate (RF) flaps has also been confirmed by immunohistochemistry findings: similar innervation architecture compared to normal skin was found in contrast to non-reinnervated RF flaps, which showed signs of Wallerian degeneration.9 However, reasonable recovery of sensibility was also reported by the articles just addressing non-sensate free-flap tongue reconstructions. Furthermore, when outcomes are compared to the contralateral tongue or healthy control subjects, the recovery of sensibility remains inferior. Therefore, several considerations need to be made when comparing the outcomes from these articles, and these will be discussed in the following paragraphs.
9 Kimata et al. reported hardly any recovery of pain sensation, two-point discrimination, and hot/cold differentiation in non-sensate ALT flaps in their comparison to sensate ALT flaps.18 Other articles on non-sensate ALT flap reconstructions did report the recovery of sensibility in the reconstructed tongue eventually.3,4,17 This difference may be due to insufficient follow-up. Kimata et al. reported an average follow-up of only 12 months after non-sensate flap tongue reconstructions compared to 18 months after sensate flap tongue reconstructions.14 Therefore, results and comparison with non-sensate ALT flap reconstruction could have been subject to lengthetime bias. Kimata et al. also reported that sensate RA flaps were superior to nonsensate RA flaps; however, recovery was inferior to sensate ALT flaps.18 This discrepancy may have been caused by the use of intercostal nerves in RA flap reconstruction, which are mixed nerves, consisting of 70% motor and only 30% sensory nerves. The lateral cutaneous femoral nerve (LCFN) of the ALT flap, by contrast, is a 100% sensory nerve. Second, it has been shown that fasciocutaneous flaps have better sensory recovery compared to musculocutaneous flaps, possibly because of the difference in flap thickness.17 When tongue sensibility after sensate RF flap tongue reconstruction is compared to healthy controls, instead of the contralateral tongue or non-sensate flaps, RF flaps showed inferior sensibility.13,14 These results question the use of healthy subjects as controls in sensory testing, as Boyd et al. and Katou et al. described no significant difference between sensate RF flap tongue reconstruction and the contralateral tongue.8,9 A reason for this discrepancy may be that the innervation of the contralateral tongue suffers from radiotherapy or the resection of the affected tongue tissue. Unfortunately, no articles in this review examined the collateral damage to sensibility of the adjacent healthy tongue after radiotherapy or surgery. The use of the TFL flap for tongue reconstruction was reported by Cheng et al. who showed superior clinical outcome measures as opposed to primary closure and the PM flap.16 It cannot be excluded that motor reinnervation, which was additionally performed in their study, may also have improved the functional outcome of the TFL tongue reconstruction. Because a PM flap tongue reconstruction does not include an adequate sensory or motor reinnervation, comparison with this method may not be valid. A better comparison would be to compare reinnervated and non-reinnervated TFL muscles for tongue reconstruction to obtain more conclusive results. One of the considerations that have to be made during tongue reconstruction is which recipient nerve and what neurorrhaphy technique (end to end or end to side) should be used for coaptation. All articles in the present review used the lingual or inferior alveolar nerve as the recipient nerve. Only one study examined whether there was a different sensory outcome after utilizing different recipient nerves for the RF flap in tongue reconstruction.21 This study showed that the lingual or alveolar nerve had superior sensory recovery when used as a recipient nerve compared to the posterior auricular, cervical plexus, or hypoglossal nerves. An explanation for this difference may be that the lingual and alveolar nerves have a bigger representation in the sensory cerebral cortex as opposed to the other
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
10 investigated nerves. Interestingly, several articles have shown that if a free flap is used for tongue reconstruction that is anastomosed to the lingual or alveolar nerve, it has better sensibility at the recipient site than at the original donor site.8,13e15 Possibly, this could also be explained by the better cortical representation of the lingual and inferior alveolar nerves as compared to the nerves from the donor site such as, for example, the RF (lateral antebrachial cutaneous nerve (LABCN)) and the ALT flap (LCFN). There is no difference in the coaptation technique (end to end vs. end to side) when the lingual or the inferior alveolar nerve was used. Therefore, on the basis of this review, both the lingual and the alveolar nerves could be used, in an end-to-end or end-to-side fashion, as a recipient nerve in free-flap tongue reconstruction.21 Although it has been suggested frequently that patient characteristics such as age, smoking, sex, and postoperative radiation therapy affect the flap sensory recovery, this was not confirmed by any of the articles that we reviewed: no statistically significant differences in the recovery of sensibility were found between different age groups, smokers and nonsmokers, or different sexes.3,4,13,16,17 However, only few articles addressed these clinical factors in a limited number of patients. Therefore, these analyses may not have reached statistical significance due to lack of power. For postoperative radiotherapy, there is still no consensus whether or not it has a negative effect on sensory reinnervation. Delayed sensory recovery was reported twice,3,4 and one article18 reported permanent inferior sensibility after radiotherapy. However, the timing of testing was unclear from the latter article,18 and the follow-up time ranged from 6 to 32 months. Therefore, in at least a part of the studied cohort, it might be possible that there was a delayed good sensory recovery as seen in the study by Yu et al.,3,4 which was not observed due to insufficient follow-up. Proper long-term follow-up is needed in future research to study the effect of radiotherapy on the sensory recovery after free-flap tongue reconstruction more closely. It has been assumed that the recovery of flap sensibility in tongue reconstruction is needed for good functional outcomes; however, we found limited evidence to support this. Three articles actually described a statistically significant relationship between return of sensibility and improved functional outcomes.3,4,17 Other articles that reported superior outcomes, however, did not support these findings with statistical analyses.8,12 This lack of evidence is caused by various methodological shortcomings. First, the assessment of functional outcomes was often not standardized and subjective. It is known that questionnaires regarding subjective results require extensive psychometric testing to measure valid, reliable, and reproducible results. Second, as Loewen et al. reported,14 functions such as texture recognition, taste, and mastication are largely dependent upon the resection’s extent and the amount of sensibility of the remainder of the oral cavity. As defect sizes varied, the contribution of healthy surrounding tissues differed, and their impact on these functions was unclear. As oral stereognosis has been reported to depend upon sensory feedback from the tip of the tongue,22 the tests applied may not have had sufficient sensory thresholds to detect minute differences.
M. Baas et al.
Conclusion Sensory nerve coaptation of free flaps in tongue reconstruction improves sensory recovery, and it may improve clinical functional outcome (i.e., deglutition and articulation). The two free flaps with the most promising results are the RF and ALT flaps, using the LABC and LCFN as sensory nerves for coaptation, respectively. As recipient nerves, the lingual and inferior alveolar nerves appear to be the best option for neurorrhaphy, regardless of the coaptation method (end to end vs. end to side). Clinical factors such as age, tobacco use, and sex did not seem to have an effect on sensory recovery. The long-term effect of postoperative radiotherapy, however, is still unknown, although it appears to reduce the speed of sensory recovery. Our conclusions are limited by the small number and heterogeneity of the patient groups and different methods of reconstruction and assessment of the outcomes. Prospective long-term follow-up studies with a large number of patients and with objective measurements are needed for better conclusions.
Financial disclosure None of the authors has a financial interest in any of the products, devices, or drugs mentioned in this manuscript.
Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.bjps.2015.04.020.
References 1. Elfring T, Boliek CA, Winget M, et al. The relationship between lingual and hypoglossal nerve function and quality of life in head and neck cancer. J Oral Rehabil 2014;41:133e40. 2. Bokhari WA, Wang SJ. Tongue reconstruction: recent advances. Curr Opin Otolaryngol Head Neck Surg 2007;15:202e7. 3. Yu P. Reinnervated anterolateral thigh flap for tongue reconstruction. Head Neck 2004;26:1038e44. 4. Yu P, Robb GL. Reconstruction for total and near-total glossectomy defects. Clin Plast Surg 2005;32:411e9. vii. 5. Broyles JM, Abt NB, Shridharani SM, et al. The fusion of craniofacial reconstruction and microsurgery: a functional and aesthetic approach. Plast Reconstr Surg 2014;134:760e9. 6. Hanasono MM, Matros E, Disa JJ. Important aspects of head and neck reconstruction. Plast Reconstr Surg 2014;134:968ee80e. 7. Welkoborsky HJ, Deichmuller C, Bauer L, Hinni ML. Reconstruction of large pharyngeal defects with microvascular free flaps and myocutaneous pedicled flaps. Curr Opin Otolaryngol Head Neck Surg 2013;21:318e27. 8. Boyd B, Mulholland S, Gullane P, et al. Reinnervated lateral antebrachial cutaneous neurosome flaps in oral reconstruction: are we making sense? Plast Reconstr Surg 1994;93:1350e9. discussion 60e2. 9. Katou F, Shirai N, Kamakura S, et al. Intraoral reconstruction with innervated forearm flap: a comparison of sensibility and reinnervation in innervated versus noninnervated forearm flap. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1995;80: 638e44.
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
Review on reinnervation of tongue reconstruction free flaps 10. Nordgren M, Hammerlid E, Bjordal K, et al. Quality of life in oral carcinoma: a 5-year prospective study. Head Neck 2008; 30:461e70. 11. Higgins JP, Altman DG, Gotzsche PC, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ 2011;343:d5928. 12. Biglioli F, Liviero F, Frigerio A, Rezzonico A, Brusati R. Function of the sensate free forearm flap after partial glossectomy. J Craniomaxillofac Surg 2006;34:332e9. 13. 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;127:1463e6. 14. Loewen IJ, Boliek CA, Harris J, Seikaly H, Rieger JM. Oral sensation and function: a comparison of patients with innervated radial forearm free flap reconstruction to healthy matched controls. Head Neck 2010;32:85e95. 15. Longo B, Pagnoni M, Ferri G, Morello R, Santanelli F. The mushroom-shaped anterolateral thigh perforator flap for subtotal tongue reconstruction. Plast Reconstr Surg 2013;132: 656e65. 16. Cheng N, Shou B, Zheng M, Huang A. Microneurovascular transfer of the tensor fascia lata musculocutaneous flap for reconstruction of the tongue. Ann Plast Surg 1994;33:136e41.
11 17. Close LG, Truelson JM, Milledge RA, Schweitzer C. Sensory recovery in noninnervated flaps used for oral cavity and oropharyngeal reconstruction. Arch Otolaryngol Head Neck Surg 1995;121:967e72. 18. Kimata Y, Uchiyama K, Ebihara S, et al. Comparison of innervated and noninnervated free flaps in oral reconstruction. Plast Reconstr Surg 1999;104:1307e13. 19. Shindo ML, Sinha UK, Rice DH. Sensory recovery in noninnervated free flaps for head and neck reconstruction. Laryngoscope 1995;105:1290e3. 20. Sabesan T, Ramchandani PL, Ilankovan V. Sensory recovery of noninnervated free flap in oral and oropharyngeal reconstruction. Int J Oral Maxillofac Surg 2008;37:819e23. 21. Santamaria E, Wei FC, Chen IH, Chuang DC. Sensation recovery on innervated radial forearm flap for hemiglossectomy reconstruction by using different recipient nerves. Plast Reconstr Surg 1999;103:450e7. 22. Colletti EA, Geffner D, Schlanger P. Oral stereognostic ability among tongue thrusters with interdental lisp, tongue thrusters without interdental lisp and normal children. Percept Mot Skills 1976;42:259e68.
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
MODEL
Journal of Plastic, Reconstructive & Aesthetic Surgery (2015) xx, 1e11
REVIEW
A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits? Martijn Baas a, Liron S. Duraku a, Eveline M.L. Corten, Marc A.M. Mureau* Department of Plastic and Reconstructive Surgery, Erasmus MC Cancer Institute, University Medical Centre Rotterdam, Rotterdam, The Netherlands Received 29 December 2014; accepted 27 April 2015
KEYWORDS Tongue reconstruction; Free flaps; Nerve reinnervation; Sensory recovery
Summary Background: Tongue reconstruction after (hemi)glossectomy including sensory recovery is challenging. Although sensory recovery could improve functional outcome, no consensus on the need for reinnervation of the neo-tongue exists. Therefore, a systematic review was performed to determine if sensory reinnervation of free flaps in tongue reconstruction is better than no sensory reinnervation. The secondary study aim was to assess the effect of sensory reinnervation on overall functional outcome, such as speech and deglutition. Methods: Seven databases (Embase, Medline, Web of Science, Scopus, PubMed publisher, Cochrane, and Google Scholar) were searched. Studies that reported the effect of sensory reinnervation on overall functional outcome were identified. Results: Fourteen articles were included in the systematic review, concerning a total of 271 tongue reconstructions. Free flaps that were used were the radial forearm (RF) flap (n Z 137), the anterolateral thigh (ALT) flap (n Z 65), the rectus abdominis (RA) flap (n Z 20), and the tensor fascia latae (TFL) flap (n Z 5). Seven out of seven articles directly comparing sensory reinnervation with no sensory reinnervation revealed superior sensibility in the reinnervated group. Moreover, the innervated RF and ALT flaps showed superior recovery of sensibility compared to other flaps used for the reconstruction of hemiglossectomy as well as total glossectomy defects. There are indications that sensory reinnervation may have a beneficial effect on overall tongue function. Age, smoking, and sex did not affect sensory recovery. Four out of five articles showed that postoperative radiotherapy does not have a longterm adverse effect on sensory recovery.
* Corresponding author. Department of Plastic and Reconstructive Surgery, Erasmus MC Cancer Institute, University Medical Centre Rotterdam, PO Box 2040, 3000 CA Rotterdam, The Netherlands. E-mail address: [email protected] (M.A.M. Mureau). a Both authors contributed equally to this article. http://dx.doi.org/10.1016/j.bjps.2015.04.020 1748-6815/ª 2015 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
2
M. Baas et al. Conclusions: Sensory reinnervation of free flaps in the reconstruction of (hemi)glossectomy defects improves sensory recovery; however, evidence for beneficial effects on function is poor. ª 2015 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
Introduction The tongue is an important muscle, with numerous functions, such as articulation, deglutition, and taste. Impairment of tongue function can severely affect quality of life.1 Therefore, it is paramount to restore the tongue’s main function after oncologic (hemi)glossectomy. Because of its complex anatomy, it is challenging to restore bulk, mobility, and sensibility in order for the reconstructed tongue to be functional. Because of advances in microvascular tissue transfer, various free flaps for tongue reconstruction are available.2 Small defects can usually be closed primarily with good function.3,4 After hemiglossectomy, the appropriate reconstruction of choice includes a thin fasciocutaneous flap such as the radial forearm (RF) flap or a thin anterolateral thigh (ALT) flap. However, larger defects need more bulk to restore deglutition and articulation. In these situations, thicker fasciocutaneous or musculocutaneous flaps such as the ALT and rectus abdominis (RA) flap may be indicated.3,4 Interestingly, sensory recovery is a factor that is often neglected in the reconstructive algorithm,5e7 whereas it is supposed to be an essential factor for the proper function of the tongue. Previous articles showed that sensory reinnervation of free flaps in tongue reconstruction may have an advantage in the recovery of articulation, deglutition, sensibility, and even quality of life.8e10 However, still no consensus exists on the need for reinnervation of the neotongue. Therefore, a systematic review was performed to determine the advantages and reasons for sensory reinnervation of free flaps in tongue reconstruction. The primary research question of this review was whether sensory reinnervation of free flaps in tongue reconstruction leads to better sensibility of the neo-tongue compared with no sensory reinnervation. In addition, the effect of recipient nerves and factors such as age, tobacco use, sex, and radiation history on the return of sensibility was investigated. The secondary study question was whether the improved sensibility of free flaps also implies improved functional outcome of the reconstruction, such as speech and deglutition.
Methods The search strings that were used to search in seven different databases (Embase, Medline (OvidSP), Web of Science, Scopus, PubMed publisher, Cochrane, and Google Scholar) are listed in Appendix 3. A combination of search criteria was used to identify all articles concerning
oncological tongue as well as intraoral reconstructions with either innervated or non-innervated free flap. The initial search was performed on April 2014, and it was monitored during the review progress. Subsequently, two reviewers (M.B. and L.S.D.) performed a manual secondary selection based on the following inclusion criteria for our primary and secondary outcome measures. Eligibility criteria were formulated to select articles with comparable, preferably standardized, measures of reinnervation. The criteria for eligibility were as follows: Only original articles studying patients (no reviews) and written in English were included At least one of the study groups consisted of tongue reconstructions Tongue reconstructions were performed with free flaps only Articles had to contain objective sensory testing modalities Only articles with a design classification of levels IeV were included, as classified by Jovell and Navarro-Rubio (Table 1). Bias was assessed using the Cochrane tool for assessing the risk of bias, addressing sequence generation, allocation, blinding, incomplete outcome data, selective outcome reporting as well as other sources of bias.11
Table 1 Classification of strength of evidence by Jovell and Narvarro-Rubio. Level
Strength of evidence
Type of study design
I
Good
Meta-analysis of randomized controlled trials Large-sample randomized controlled trials (N > 25 for each group) Small-sample randomized controlled trials (N < 25 for each group) Non-randomized controlled prospective trials Non-randomized controlled retrospective trials Cohort studies Caseecontrol studies Noncontrolled clinical series; descriptive studies Anecdotes or case reports
II
III
Good to fair
IV V VI VII VIII IX
Fair Poor
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
Review on reinnervation of tongue reconstruction free flaps Finally, all differences between reviewers were resolved by consensus. The high heterogeneity in the population and outcome measures of the included articles precluded statistical analyses of pooled data. Therefore, conclusions were made on the basis of key findings in the included articles. Our protocol was not registered in PROSPERO. The Preferred Reporting Items for Systematic Review and MetaAnalyses (PRISMA) guidelines were followed, and the checklist is available in the online supplements of this article (Appendices 1e5).
Results Patient and reconstruction characteristics The systematic search identified 952 articles. After exclusion of 525 duplicate articles, 14 articles were selected using the inclusion criteria (Figure 1).3,4,8,9,12e21 There was a high risk of bias in the individual articles mainly due to poor study design (Appendix 5); no articles were excluded. A total of 271 tongue reconstructions were performed using 137 RF flaps, 65 ALT flaps, 20 RA flaps, and five tensor fascia latae (TFL) flaps. Other (pedicled) flaps that were also used for tongue reconstruction served as controls, but they are not discussed in great detail. The sex of the patient was reported in 197 out of 271 patients; 146 patients were male and 51 patients were female. Patient age in the included articles ranged from 29 to 78 years. All resections were performed to treat malignant tumors. Although most
Figure 1
3 authors reported resecting at least half of the mobile tongue (hemiglossectomy), the defect size was highly variable and ranged from 20 to 140 cm2. The use of (neo) adjuvant radiotherapy was reported in 10 articles, and it ranged from 15.4% to 100% of the included subjects.3,4,8,13,14,17e21 The prevalence of smoking in the studied populations was only reported twice, and it was 39.3% and 47.5%.20,21
Sensory recovery after tongue reconstruction with different free flaps RF flap Nine articles described the sensory recovery of the neotongue after reconstruction using an RF flap. A total of 136 patients were examined, of which 72 flaps were sensory reinnervated and 64 flaps were used as non-sensate flaps. Three articles compared the use of sensate with nonsensate RF flaps in tongue reconstruction.5,6,8 All authors reported superior sensibility in sensate compared with nonsensate RF flaps, assessed with multiple objective sensory modalities (Table 2). In addition, Katou et al. showed that sensate RF flap tongue reconstructions also resulted in the earlier recovery of sensibility (mean, 13 months) than nonsensate RF flaps (mean, 25.4 months).6 When a sensate RF flap was compared to the healthy contralateral tongue, temperature discrimination and two-point discrimination showed no statistically significant differences, indicating complete recovery of sensibility for those modalities in the sensate RF flap.5
Flowchart regarding the selection of included articles by PRISMA standards.
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
4
M. Baas et al. Table 2 Details on reconstruction, tested modalities, and outcome of all included articles on RF free-flap reconstruction of the tongue. Author
Population
Non-sensate versus sensate RF free flap Boyd et al.8 Eight sensate RF flaps 10 non-sensate RF flaps 10 historical controls (PM flaps)
Tested modalities
Outcome
Hot/cold perception, Semmes eWeinstein tests, sharp/dull discrimination, two-point discrimination, and perception of food presence.
Sharp/dull discrimination was present in 75% versus 10% (sensate vs. non-sensate). Hot ecold perception: 100% (both) versus 50% (hot) and 40% (cold). Two-point discrimination: 14.25 g/mm2 (3.25e31.5) versus 27 g/mm2 (3.25e31.5). Perception of food presence: 7/10 versus 3/10.b Sharp/dull discrimination was present in 4/4 versus 4/9 patients (sensate vs. nonsensate). Hot/cold discrimination 4/4 versus 2/9. Innervated flaps were tested after a shorter follow-up period (13 vs. 25.4 months). Using immunohistochemistry, more fibers with better structured arrangement were detected in sensate reconstructions. Perception of tactile stimuli was slightly better in the sensate group (89% and 79% for central and peripheral areas compared to 71% for both in the non-sensate group). Sharp/ dull discrimination and cold perceptions were superior (78% vs. 43% and 100% vs. 57%, respectively) in sensate reconstructions. Subjective speech was superior (78% vs. 43%) in the sensate group. For swallowing, a minimal difference was reported (61% vs. 57%).b
Katou et al.9
Four sensate RF flaps Nine non-sensate RF flaps
Hot/cold perception, touch perception, sharp/dull discrimination, and immunohistochemistry.
Biglioli et al.12
Nine non-sensate RF flaps Seven sensate RF flaps
Hot/cold perception, touch perception, sharp/dull discrimination, perception and localization of tactile stimuli, speech and swallow function, and deglutition pressure.
Non-sensate RF Free Flap Close et al.17 Four fasciocutaneous RF flaps Four fasciocutaneous ALT flaps Three pedicled musculocutaneous PM flaps One pedicled musculocutaneous LD flaps
Hot/cold perception, Semmes eWeinstein tests, sharp/dull discrimination, two-point discrimination, and speech and swallow function.
10 out of 12 patients had restoration of sensibility after at least 6 months (4e24 months). Touch sensation recovered in 8/8 FCFF and 2/4 MCF (p Z 0.09), sharp/dull discrimination 6/8 FCFF and 2/ 4 MFC. Two-point discrimination returned in 5/8 FCFF and 1/4 MCF; 7/8 FCFF versus 2/4 MCF patients had satisfactory articulation and adequate oral intake. Articulation and swallowing correlated statistically to return of sensibility.
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
Review on reinnervation of tongue reconstruction free flaps
5
Table 2 (continued ) Author Shindo et al.
19
Sabesan et al.20
Population
Tested modalities
Outcome
Nine fasciocutaneous RF flaps Nine osteocutaneous fibula flaps
Hot/cold perception, touch perception, sharp/dull discrimination, two-point discrimination.
23 fasciocutaneous RF flaps Six gastro-omental free flaps 10 jejunal flaps
Hot/cold perception, touch perception, sharp/dull discrimination, and two-point discrimination.
No return of sensibility was reported in the two reconstructed tongues after total glossectomy. In partial lesions, sensibility was reported in all five patients after 6 (1), 7 (1), and 12 months.3 The area of sensibility was largest in hot temperature testinga. The mean scores for RF, gastroomental, and jejunal flaps for the different modalities were (p-values represent test of RF vs. gastro-omental and jejunal flaps); Light touch: 6.04 versus 4.1 versus 3.52 (p Z 0.084). Sharp prick 6.84 versus 4.83 versus 3.72 (p < 0.05). Hot temp sensation 6.82 versus 4.93 versus 4.14 (p < 0.05). Cold temp sensation 4.52 versus 2.81 versus 2.1 (p < 0.05). Twopoint discrimination. 20.31 mm versus 23.86 mm versus 23.28 mm (p < 0.05).
Sensate RF Free Flap 16 sensate (lingual) RF flaps Santamaria et al.21 Six sensate (inferior alveolar) RF flaps, three sensate (posterior auricular) RF flaps, two sensate (cervical plexus) RF flaps one sensate (hypoglossus) RF flaps
Kuriakose et al.13
17 sensate RF flaps
Hot/cold perception, touch perception, Semmes eWeinstein tests, sharp/dull discrimination, two-point discrimination.
Hot/cold perception, Semmes eWeinstein tests, sharp/dull discrimination, two-point discrimination, and subjective perception of food.
Two-point discrimination in sensate reconstruction was comparable to the contralateral tongue (largest difference was observed in the dorsal aspect of the tongue 4.4 1.6 mm vs. 2.6 0.6 mm). Light-touch sensation was statistically significantly diminished in the ventral (reconstructed) tongue indicating loss of protective sensibility (4.8 0.8 mm vs. 2.6 0.2 mm). Pain sensation was significantly decreased (p < 0.05) on the floor of the mouth as compared to other tongue surfaces. Hot and cold perception was 15%b less than control values. Sensory recovery was observed in all 17 patients within 8 months. Two-point discrimination (moving and static) and pressure sensitivity were greater than the contralateral donor site (1.2 vs. 2.3 cm, 0.8 vs. 1.7 cm and 3.7 psi vs. 4.6 psi), however, not (continued on next page)
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
6
M. Baas et al. Table 2 (continued ) Author
Loewen et al.14
Population
Eight sensate RF flaps Eight matched healthy controls
Tested modalities
Hot/cold discrimination, touch perception, Semmes eWeinstein test, two-point discrimination, form recognition, texture recognition, taste, and swallow and speech function.
Outcome significantly different (p Z 0.064). Values were similar to normal tongue (0.9 cm, 0.5 cm, and 3.6 psi) For nearly all modalities, there was a significant difference (p < 0.05) with healthy controls (two-pot reconstruction 20 vs. 90% correct response, light touch 50 vs. 90% correct response, temperature sensation 50 vs. 90% correct response, form recognition 40 vs. 60% correct response). There was no significant difference only in texture recognition and taste. Significant differences (p < 0.05) in intelligible speech were found for both words and sentences in favor of the healthy control group. Masticatory efficiency was greater in healthy controls
RF: radial forearm, PM: pectoralis major, ALT: anterolateral thigh, LD: latissimus dorsi. FCFF: Fasciocutaneous Free Flap. MCF: Musculocutaneous Flap. a Specific values of test were not published. b Original article only provides percentages for these parameters.
In three articles, the sensibility of non-sensate RF flaps was compared to other non-sensate flaps in tongue reconstruction.13,15,16 The other flaps described were the ALT, fibula, and gastro-omental and jejunal flaps. In contrast to other non-sensate flaps, there was recovery of sensibility in all non-sensate RF flap tongue reconstructions. The last three articles exclusively examined sensate RF flaps in tongue reconstruction.9,10,17 Loewen et al. showed that the sensibility of sensate RF flaps in tongue reconstruction never matches the sensibility of healthy controls.10 By contrast, Santamaria et al. showed a comparable two-point discrimination for the sensate RF flap as compared with the contralateral tongue; however, lighttouch perception, pain perception, and hot and cold discrimination remained inferior.17 Kuriakose et al. reported full restoration of sensibility in the reconstructed tongue with sensate RF flaps as compared to the contralateral donor site.9 ALT flap Five articles described the sensory recovery of the tongue after reconstruction using an ALT flap.3,4,11,13,14 A total of 65 patients were examined, of which 40 flaps were sensate and 25 flaps were used as non-sensate flaps (Table 3). Three articles compared sensate ALT flaps with nonsensate ALT flaps in tongue reconstruction.3,4,14 All articles reported superior sensibility of innervated ALT flaps in light touch, pain sensation, temperature, SemmeseWeinstein monofilament tests, and two-point discrimination as
compared with non-sensate ALT flaps. In addition, in these articles, the return of sensibility in non-sensate ALT flaps did not occur in the majority of patients. One study compared the sensibility of sensate ALT flaps to the donor site of the flap and to the tip of the preoperatively native tongue.15 This report revealed that sensate ALT flaps show inferior sensibility compared to the preoperative tip of the tongue. However, it was also found that the sensibility of sensate ALT flaps did recover to the level of the preoperative donor site.15 The last study16 compared non-sensate ALT flaps to other non-sensate flaps for the reconstruction of the tongue. The other flaps used were the RF, the lattisimus dorsi (LD), and the pectoralis major (PM) flap. This report revealed that in non-sensate tongue reconstruction, all non-sensate ALT flaps had gained sensibility of light touch, and over 7/12 patients obtained sharp/dull and twopoint discrimination.17 In non-sensate PM flaps, no recovery of sensibility was reported, and in non-sensate LD flaps, all but temperature discrimination recovered. RA flap Only one article used the RA flap as a sensate flap in tongue reconstruction.14 This article compared sensate (five patients) with non-sensate RA flaps (10 patients). Additionally, both sensate (eight patients) and non-sensate ALT flaps (five patients) were examined. Sensate RA flaps showed a statistically significant better sensibility compared to non-sensate RA flaps in the majority of the tested sensory modalities (moving and static two-point
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
Review on reinnervation of tongue reconstruction free flaps
7
Table 3 Details on reconstruction, tested modalities, and outcome of all included articles on ALT free-flap reconstruction of the tongue. Author
Population
Non-sensate vs. sensate ALT flap Kimata et al.18 Eight sensate ALT flaps, Six non-sensate ALT flaps, Five sensate RA flaps 10 non-sensate RA flaps
Tested modalities
Outcome
Hot/cold perception, touch perception, Semmes eWeinstein tests, sharp/dull discrimination, and two-point discrimination.
Up to 87.5% sensibility of light touch (vs. 30% non-sensate) and hot/cold discrimination (vs. 0% non-sensate) in sensate ALT flapsa; 7/8 patients had intact two-point discrimination (vs. 1/6 non-innervated) in sensate flaps. Sensate ALT flaps were able to detect lower pressures (6.62 g vs. 184.7 g, p < 0.05) Innervated flaps were superior in two-point discrimination (0/ 5 vs. 5/8 points) (mean 2.6 0.9 mm), Semmes eWeinstein (Tip: log force 3.8 vs. 6.8, p < 0.05), pain perception (0/5 vs. 8/8), and temperature differentiation (0/5 vs. 5/6). No significant difference in speech scores was observed (3.7 0.8 vs. 3.3 0.5). Swallow scores were significantly better in the innervated group (5 2.5 vs. 2.6 0.9 points). Innervated flaps were superior in two-point discrimination 0/ 10 versus 9/11 (mean 2.6 0.9 mm), Semmes eWeinstein (Tip: log force 3.8 vs. 6.8, p Z 0.005) pain perception (0/10 vs. 11/11), and temperature differentiation (1/10 vs. 11/ 11). No significant difference in speech scores was observed (3.7 0.8 vs. 3.3 0.5). Swallow scores were significantly better in the innervated group (5 2.5 vs. 2.6 0.9 points).
Yu et al.3
Eight sensate ALT flaps Five non-sensate ALT flaps
Hot/cold perception, Semmes eWeinstein tests, two-point discrimination, pain sensation, and speech and swallow function.
Yu et al.4
11 sensate ALT flaps 10 non-sensate ALT flaps
Hot/cold perception, Semmes eWeinstein tests, two-point discrimination, pain sensation, and speech and swallow function.
Non-sensate ALT Flap Four fasciocutaneous RF flaps Close et al.17 Four fasciocutaneous ALT flaps Three pedicled musculocutaneous PM flaps One pedicled musculocutaneous LD flaps
Hot/cold perception, Semmes eWeinstein tests, sharp/dull discrimination, two-point discrimination, and speech and swallow function.
10 out of 12 patients had restoration of sensibility after at least 6 months (4e24 months). Touch sensation recovered in 8/8 FCFF and 2/4 MCF (p Z 0.09) Sharp/dull discrimination 6/8 FCFF and 2/ 4 MFC. Two-point discrimination returned in 5/8 FCFF and 1/4 MCF; 7/8 FCFF vs. 2/4 MCF patients had satisfactory articulation and (continued on next page)
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
8
M. Baas et al. Table 3 (continued ) Author
Population
Tested modalities
Outcome adequate oral intake. Articulation and swallowing correlated statistically to return of sensibility.
Sensate ALT Flap Longo et al.15
Seven sensate ALT total tongue reconstruction Six sensate ALT subtotal tongue reconstruction
Hot/cold perception, touch perception, Semmes eWeinstein tests, two-point discrimination, and speech and swallow function
Full recovery of hot/cold perception, pain perception, and two-point discrimination was observed in sensate reconstruction to the level of the donor site discrimination (Tip vs. donor site: 4.08 3.09 vs. 4.47 2.15). Normal diet was reached in 46.2%, mild impairment in diet in 53.8%. Normal intelligible speech in 53.8%, acceptable speech in 46.2%a.
RF: radial forearm, RA: rectus abdominis, PM: pectoralis major, ALT: anterolateral thigh, LD: latissimus dorsi. a Original article only provides percentages for these parameters.
discriminations and touch and pain perception).14 Only for hot and cold discrimination, there was no difference between the sensate and non-sensate RA groups. TFL flap There is only one article that examined the TFL flap as an option for the reconstruction of the tongue.16 Primary closure and pedicled pectoralis musculocutaneous flaps were the comparison groups. Clinical evaluation was performed by means of speech, oral pressure, cineradiography, and electromyography at 10 and 24 months post operatively. The free TFL flap showed superior clinical outcomes: deglutition pressure had recovered to 60% in the TFL group compared to 20% and 505 in the PM flap and primary closure group, respectively, with faster swallowing time and lesser residual barium at cineradiography. On speech evaluation, excellent or good intelligibility scores were achieved by all five patients in the TFL group, whereas in the other groups, a total of three patients were rated as poor (2/5 in the PM flap group and 1/5 in the primary closure group).12,16
Effect of recipient nerve and patient characteristics on sensory outcome Recipient nerves Different recipient nerves and neurorrhaphy techniques (end-to-end or end-to-side) may be used for flap reinnervation in tongue reconstruction. The majority of articles used the lingual nerve as the primary recipient nerve. Santamaria et al. were the only ones who studied different recipient nerves for flap reinnervation in tongue reconstruction (lingual, inferior alveolar, posterior auricular, cervical plexus, or hypoglossal nerve).17 They also investigated whether there was a difference between end-to-end and end-to-side repairs. They reported that two-point
discrimination, light touch, pain, and temperature perception of the RF flap were statistically better after coaptation to the lingual and inferior alveolar nerves compared to the other nerves. Interestingly, there existed no statistically significant difference in sensory recovery between the end-to-end and end-to-side techniques when the lingual or inferior alveolar nerves were used as recipient nerve. Patient characteristics Five articles examined the effect of various patient characteristics on the sensory recovery of free flaps in tongue reconstruction.3,4,13,16,17 No statistically significant relationships between sensory recovery, age, sex, or tobacco use were reported. Sabesan et al.16 as well as Santamaria et al.17 could not find statistically significant differences between males versus females, patients older versus younger than 60 years, and smokers versus nonsmokers. Close et al. did not address tobacco use or sex; however, they also could not find a relationship between age and return of sensibility.13 In four articles, postoperative radiotherapy did not seem to affect sensory recovery of free flaps in tongue reconstruction in the long term (after at least 12 months).3,4,10,16 However, two of those articles did show inferior results in the short term (at 6 months).3,4 One other article found that free flaps showed inferior light-touch sensation and pain and temperature perception after postoperative radiotherapy at a mean follow-up of 18.2 months (range, 6e32 months).17
Effect of sensory recovery on overall functional outcome Functional outcomes were reported in eight of 14 articles.3,4,12e17 Most functional results were assessed using
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
Review on reinnervation of tongue reconstruction free flaps non-standardized questionnaires on swallowing and dietary limitations and intelligibility of speech. Although superior results were obtained following sensate flap reconstruction, most articles did not correlate this to functional outcomes. Close et al. did prove a statistically significant correlation between return of sensibility and articulation as well as swallowing in non-sensate RF flaps.13 Yu et al. reported better swallowing scores after innervated ALT flap tongue reconstruction; however, these differences were not statistically significant.3,4 Loewen et al. noted that, although sensory outcomes after free-flap tongue reconstruction remained suboptimal compared to healthy tongues, no statistically significant differences were found in taste and texture recognition compared to healthy tongues.10
Discussion The comprehensive search in seven databases together with the individual manual selection by two authors makes it unlikely that articles concerning our research question(s) may have been missed. However, the quality of the results of our review is highly dependent on the limited number of articles and the low sample sizes examined in the selected articles. We found no randomized controlled trials or other articles in which there was a blinding of outcomes that we reviewed. We chose not to exclude any articles based on their risk of bias, because the risk of bias was judged to be equal for articles on sensate and non-sensate reconstructions as well as comparative studies. The RF and ALT flaps were the most frequently used flaps in the articles we reviewed. The RF flap is very suitable because of its thinness and pliability. The ALT flap, just like the RA and TFL flaps, is larger and bulkier, and therefore it is most commonly used to reconstruct larger tongue defects. The articles reviewed did not clearly describe the rationale for free-flap choice. However, based on the defect sizes derived from the included articles, it did not appear that flap choices were primarily based on defect size. Other reasons for flap choice could include patient comorbidities, desire for flap reinnervation, concern about donor site complications, and the desire to perform the oncological resection and harvest of the free flap simultaneously. When these parameters are considered, the RF and ALT flaps could be more favorable for tongue reconstruction. Direct comparison of sensate with non-sensate free-flap tongue reconstruction reveals superior sensory recovery after flap reinnervation in all included articles.3,4,8,9,12,18 In addition, the superiority in sensibility of sensate (RF) flaps has also been confirmed by immunohistochemistry findings: similar innervation architecture compared to normal skin was found in contrast to non-reinnervated RF flaps, which showed signs of Wallerian degeneration.9 However, reasonable recovery of sensibility was also reported by the articles just addressing non-sensate free-flap tongue reconstructions. Furthermore, when outcomes are compared to the contralateral tongue or healthy control subjects, the recovery of sensibility remains inferior. Therefore, several considerations need to be made when comparing the outcomes from these articles, and these will be discussed in the following paragraphs.
9 Kimata et al. reported hardly any recovery of pain sensation, two-point discrimination, and hot/cold differentiation in non-sensate ALT flaps in their comparison to sensate ALT flaps.18 Other articles on non-sensate ALT flap reconstructions did report the recovery of sensibility in the reconstructed tongue eventually.3,4,17 This difference may be due to insufficient follow-up. Kimata et al. reported an average follow-up of only 12 months after non-sensate flap tongue reconstructions compared to 18 months after sensate flap tongue reconstructions.14 Therefore, results and comparison with non-sensate ALT flap reconstruction could have been subject to lengthetime bias. Kimata et al. also reported that sensate RA flaps were superior to nonsensate RA flaps; however, recovery was inferior to sensate ALT flaps.18 This discrepancy may have been caused by the use of intercostal nerves in RA flap reconstruction, which are mixed nerves, consisting of 70% motor and only 30% sensory nerves. The lateral cutaneous femoral nerve (LCFN) of the ALT flap, by contrast, is a 100% sensory nerve. Second, it has been shown that fasciocutaneous flaps have better sensory recovery compared to musculocutaneous flaps, possibly because of the difference in flap thickness.17 When tongue sensibility after sensate RF flap tongue reconstruction is compared to healthy controls, instead of the contralateral tongue or non-sensate flaps, RF flaps showed inferior sensibility.13,14 These results question the use of healthy subjects as controls in sensory testing, as Boyd et al. and Katou et al. described no significant difference between sensate RF flap tongue reconstruction and the contralateral tongue.8,9 A reason for this discrepancy may be that the innervation of the contralateral tongue suffers from radiotherapy or the resection of the affected tongue tissue. Unfortunately, no articles in this review examined the collateral damage to sensibility of the adjacent healthy tongue after radiotherapy or surgery. The use of the TFL flap for tongue reconstruction was reported by Cheng et al. who showed superior clinical outcome measures as opposed to primary closure and the PM flap.16 It cannot be excluded that motor reinnervation, which was additionally performed in their study, may also have improved the functional outcome of the TFL tongue reconstruction. Because a PM flap tongue reconstruction does not include an adequate sensory or motor reinnervation, comparison with this method may not be valid. A better comparison would be to compare reinnervated and non-reinnervated TFL muscles for tongue reconstruction to obtain more conclusive results. One of the considerations that have to be made during tongue reconstruction is which recipient nerve and what neurorrhaphy technique (end to end or end to side) should be used for coaptation. All articles in the present review used the lingual or inferior alveolar nerve as the recipient nerve. Only one study examined whether there was a different sensory outcome after utilizing different recipient nerves for the RF flap in tongue reconstruction.21 This study showed that the lingual or alveolar nerve had superior sensory recovery when used as a recipient nerve compared to the posterior auricular, cervical plexus, or hypoglossal nerves. An explanation for this difference may be that the lingual and alveolar nerves have a bigger representation in the sensory cerebral cortex as opposed to the other
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
10 investigated nerves. Interestingly, several articles have shown that if a free flap is used for tongue reconstruction that is anastomosed to the lingual or alveolar nerve, it has better sensibility at the recipient site than at the original donor site.8,13e15 Possibly, this could also be explained by the better cortical representation of the lingual and inferior alveolar nerves as compared to the nerves from the donor site such as, for example, the RF (lateral antebrachial cutaneous nerve (LABCN)) and the ALT flap (LCFN). There is no difference in the coaptation technique (end to end vs. end to side) when the lingual or the inferior alveolar nerve was used. Therefore, on the basis of this review, both the lingual and the alveolar nerves could be used, in an end-to-end or end-to-side fashion, as a recipient nerve in free-flap tongue reconstruction.21 Although it has been suggested frequently that patient characteristics such as age, smoking, sex, and postoperative radiation therapy affect the flap sensory recovery, this was not confirmed by any of the articles that we reviewed: no statistically significant differences in the recovery of sensibility were found between different age groups, smokers and nonsmokers, or different sexes.3,4,13,16,17 However, only few articles addressed these clinical factors in a limited number of patients. Therefore, these analyses may not have reached statistical significance due to lack of power. For postoperative radiotherapy, there is still no consensus whether or not it has a negative effect on sensory reinnervation. Delayed sensory recovery was reported twice,3,4 and one article18 reported permanent inferior sensibility after radiotherapy. However, the timing of testing was unclear from the latter article,18 and the follow-up time ranged from 6 to 32 months. Therefore, in at least a part of the studied cohort, it might be possible that there was a delayed good sensory recovery as seen in the study by Yu et al.,3,4 which was not observed due to insufficient follow-up. Proper long-term follow-up is needed in future research to study the effect of radiotherapy on the sensory recovery after free-flap tongue reconstruction more closely. It has been assumed that the recovery of flap sensibility in tongue reconstruction is needed for good functional outcomes; however, we found limited evidence to support this. Three articles actually described a statistically significant relationship between return of sensibility and improved functional outcomes.3,4,17 Other articles that reported superior outcomes, however, did not support these findings with statistical analyses.8,12 This lack of evidence is caused by various methodological shortcomings. First, the assessment of functional outcomes was often not standardized and subjective. It is known that questionnaires regarding subjective results require extensive psychometric testing to measure valid, reliable, and reproducible results. Second, as Loewen et al. reported,14 functions such as texture recognition, taste, and mastication are largely dependent upon the resection’s extent and the amount of sensibility of the remainder of the oral cavity. As defect sizes varied, the contribution of healthy surrounding tissues differed, and their impact on these functions was unclear. As oral stereognosis has been reported to depend upon sensory feedback from the tip of the tongue,22 the tests applied may not have had sufficient sensory thresholds to detect minute differences.
M. Baas et al.
Conclusion Sensory nerve coaptation of free flaps in tongue reconstruction improves sensory recovery, and it may improve clinical functional outcome (i.e., deglutition and articulation). The two free flaps with the most promising results are the RF and ALT flaps, using the LABC and LCFN as sensory nerves for coaptation, respectively. As recipient nerves, the lingual and inferior alveolar nerves appear to be the best option for neurorrhaphy, regardless of the coaptation method (end to end vs. end to side). Clinical factors such as age, tobacco use, and sex did not seem to have an effect on sensory recovery. The long-term effect of postoperative radiotherapy, however, is still unknown, although it appears to reduce the speed of sensory recovery. Our conclusions are limited by the small number and heterogeneity of the patient groups and different methods of reconstruction and assessment of the outcomes. Prospective long-term follow-up studies with a large number of patients and with objective measurements are needed for better conclusions.
Financial disclosure None of the authors has a financial interest in any of the products, devices, or drugs mentioned in this manuscript.
Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.bjps.2015.04.020.
References 1. Elfring T, Boliek CA, Winget M, et al. The relationship between lingual and hypoglossal nerve function and quality of life in head and neck cancer. J Oral Rehabil 2014;41:133e40. 2. Bokhari WA, Wang SJ. Tongue reconstruction: recent advances. Curr Opin Otolaryngol Head Neck Surg 2007;15:202e7. 3. Yu P. Reinnervated anterolateral thigh flap for tongue reconstruction. Head Neck 2004;26:1038e44. 4. Yu P, Robb GL. Reconstruction for total and near-total glossectomy defects. Clin Plast Surg 2005;32:411e9. vii. 5. Broyles JM, Abt NB, Shridharani SM, et al. The fusion of craniofacial reconstruction and microsurgery: a functional and aesthetic approach. Plast Reconstr Surg 2014;134:760e9. 6. Hanasono MM, Matros E, Disa JJ. Important aspects of head and neck reconstruction. Plast Reconstr Surg 2014;134:968ee80e. 7. Welkoborsky HJ, Deichmuller C, Bauer L, Hinni ML. Reconstruction of large pharyngeal defects with microvascular free flaps and myocutaneous pedicled flaps. Curr Opin Otolaryngol Head Neck Surg 2013;21:318e27. 8. Boyd B, Mulholland S, Gullane P, et al. Reinnervated lateral antebrachial cutaneous neurosome flaps in oral reconstruction: are we making sense? Plast Reconstr Surg 1994;93:1350e9. discussion 60e2. 9. Katou F, Shirai N, Kamakura S, et al. Intraoral reconstruction with innervated forearm flap: a comparison of sensibility and reinnervation in innervated versus noninnervated forearm flap. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1995;80: 638e44.
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
+
MODEL
Review on reinnervation of tongue reconstruction free flaps 10. Nordgren M, Hammerlid E, Bjordal K, et al. Quality of life in oral carcinoma: a 5-year prospective study. Head Neck 2008; 30:461e70. 11. Higgins JP, Altman DG, Gotzsche PC, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ 2011;343:d5928. 12. Biglioli F, Liviero F, Frigerio A, Rezzonico A, Brusati R. Function of the sensate free forearm flap after partial glossectomy. J Craniomaxillofac Surg 2006;34:332e9. 13. 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;127:1463e6. 14. Loewen IJ, Boliek CA, Harris J, Seikaly H, Rieger JM. Oral sensation and function: a comparison of patients with innervated radial forearm free flap reconstruction to healthy matched controls. Head Neck 2010;32:85e95. 15. Longo B, Pagnoni M, Ferri G, Morello R, Santanelli F. The mushroom-shaped anterolateral thigh perforator flap for subtotal tongue reconstruction. Plast Reconstr Surg 2013;132: 656e65. 16. Cheng N, Shou B, Zheng M, Huang A. Microneurovascular transfer of the tensor fascia lata musculocutaneous flap for reconstruction of the tongue. Ann Plast Surg 1994;33:136e41.
11 17. Close LG, Truelson JM, Milledge RA, Schweitzer C. Sensory recovery in noninnervated flaps used for oral cavity and oropharyngeal reconstruction. Arch Otolaryngol Head Neck Surg 1995;121:967e72. 18. Kimata Y, Uchiyama K, Ebihara S, et al. Comparison of innervated and noninnervated free flaps in oral reconstruction. Plast Reconstr Surg 1999;104:1307e13. 19. Shindo ML, Sinha UK, Rice DH. Sensory recovery in noninnervated free flaps for head and neck reconstruction. Laryngoscope 1995;105:1290e3. 20. Sabesan T, Ramchandani PL, Ilankovan V. Sensory recovery of noninnervated free flap in oral and oropharyngeal reconstruction. Int J Oral Maxillofac Surg 2008;37:819e23. 21. Santamaria E, Wei FC, Chen IH, Chuang DC. Sensation recovery on innervated radial forearm flap for hemiglossectomy reconstruction by using different recipient nerves. Plast Reconstr Surg 1999;103:450e7. 22. Colletti EA, Geffner D, Schlanger P. Oral stereognostic ability among tongue thrusters with interdental lisp, tongue thrusters without interdental lisp and normal children. Percept Mot Skills 1976;42:259e68.
Please cite this article in press as: Baas M, et al., A systematic review on the sensory reinnervation of free flaps for tongue reconstruction: Does improved sensibility imply functional benefits?, Journal of Plastic, Reconstructive & Aesthetic Surgery (2015), http://dx.doi.org/ 10.1016/j.bjps.2015.04.020
