Plantar Versus Dorsal Incision in the Treatment of Primary Intermetatarsal Morton s Neuroma

Transcription

1 FOOT &ANKLE INTERNATIONAL Copyright 2008 by the American Orthopaedic Foot & Ankle Society DOI: /FAI Plantar Versus Dorsal Incision in the Treatment of Primary Intermetatarsal Morton s Neuroma Christian Åkermark, M.D. 1 ; Hans Crone, M.D. 2 ;Tönu Saartok, M.D., Ph.D. 3 ; Zbigniew Zuber, M.D. 4 Stockholm, Sweden ABSTRACT INTRODUCTION Background: Only few studies have compared plantar and dorsal incisions in the treatment of primary intermetatarsal Morton s neuroma (PIMN). The results and guidelines are, however, still controversial, mainly due to confounding factors and study design. The present study is an attempt to systematically compare the two approaches. Materials and Methods: With a 2- to 5-year followup, we retrospectively compared the results of 125 patients (132 feet) with PIMN. All specimens had histology assessments. Longitudinal plantar incisions were performed by one experienced surgeon (n = 69) and dorsal incisions by another (n = 56). Records were reviewed, questionnaires evaluated, and physical examinations performed by one of two independent orthopaedic surgeons. Results: Histology verified nerve resections in all specimens except in three cases of missed nerves in the dorsal group. There were significant differences, in favor of the plantar group, regarding long-term sensory loss, postoperative sick-leave weeks and complications. The clinical outcome regarding postoperative pain at followup and overall satisfaction rating were similar. Conclusion: We conclude that the two surgical approaches were comparable for clinical outcome and patient satisfaction at followup, whereas significant differences, in favor of plantar incisions, were present regarding residual sensory loss and number of complications. The more serious complication with the dorsal approach, missed neuroma, may result in an increased risk of failure with the dorsal incision. 1 Department of Orthopaedics, Ortopediska Huset, Stockholm, Sweden 2 Department of Orthopaedics, Läkarhuset Odenplan, Stockholm, Sweden 3 Department of Orthopaedics, Visby Hospital, Visby & Stockholm Sports Trauma Research Center, Karolinska Institutet, Stockholm, Sweden 4 Department of Orthopaedics, St Görans Hospital, Stockholm, Sweden Corresponding Author: Christian Åkermark, M.D. Birger Jarlsgatan 106 A SE Stockholm Sweden christian.akermark@telia.com For information on prices and availability of reprints, call x When painful symptoms persist after initial conservative treatment, surgery is the method of choice in the treatment of primary intermetatarsal Morton s neuroma (PIMN). 3 The placement of the incision has, however, been under debate ever since the introduction of the different surgical approaches. Betts 4 reported a longitudinal plantar approach and Kaplan 13 suggested a transverse, plantar incision; whereas McElvenny 16 advocated a dorsal websplitting incision and McKeever 17 another dorsal incision. Not only does the location of the incision differ, but various surgical treatments for PIMN s have been reported: nerve resection, 6,15 neurolysis, 10 transposition, 5 and decompression via different methods. 10 Reports of postoperative results with plantar 1,14,18,20 as well as with dorsal incisions 6,7,9,11,15,20,24 of PIMNs have documented similar overall satisfactory ratings (excellent or good), ranging from 65% to 93% (plantar) and 57% to 85% (dorsal), respectively. Failure-rates (poor results) of 7% are reported with plantar incisions 1,14 and of 3% to 24% with dorsal incisions. 6,7,9,11,15,24 However, most of these studies are, as previously reported by Coughlin and Pinsonneault, 6 characterized by confounding factors. This applies also to the only four studies found in the literature of PIMNs where reasonable adequate comparisons of the two approaches have been performed. 12,20,22,25 The purpose of this study was to, retrospectively, analyze and compare the results of patients with PIMNs, treated with either a longitudinal plantar incision or a dorsal incision. MATERIALS AND METHODS With the approval of the review board in two orthopaedic clinics, informed consent was obtained from all patients in this study. Study Method All patients (n = 145) surgically treated for PIMN between January 1995 and December 1998 were selected from the

2 Foot & Ankle International/Vol. 29, No. 2/February 2008 MORTON S NEUROMA, PLANTAR VERSUS DORSAL 137 Table 1: Demographic and clinical data Plantar Group n= 69 (73 feet) Dorsal Group n= 56 (59 feet) p-value Male/female 19/54 8/51 Mean age (range) 52 (24-77) 49 (24-80) Left/right foot 41/32 24/35 Bilateral excision 4 3 Web-spaces: 2 nd rd nd and 3 rd 5 7 Preoperative duration of pain, 37 (48) 60 (72) 0.03 mean in months (SD) Followup time, months (range) 29 (24-46) 37 (24-60) Previous surgery 7/73 8/ Preoperative steroid injection 4/73 (5%) 15/59 (25%) surgical diary and their records reviewed. All surgery was performed on an ambulatory basis by two equally experienced surgeons in two orthopedic clinics. One surgeon had treated all his patients with a longitudinal plantar incision, and the other with a dorsal incision. Inclusion in the study required PIMNs only, a thorough preoperative clinical diagnosis based on the patients history and clinical examination of pain. A Mulder s test (click with medial / lateral compression of the forefoot) 27 and sensory examination of the affected lesser toes was performed. Histological examination of the specimens, as well as followup of at least 2 years, was also required. Patients were excluded if they had been operated on for other foot conditions, either simultaneously or during the followup period. None of the other procedures were for failure of the neuroma surgery. Of the 145 patients, 20 (14%) were excluded or lost at followup: five refused to participate due to acceptable circumstances (night work and other work-related conditions), four had simultaneously been operated for other forefoot conditions, and three were unreachable. Of the remaining eight, two had moved abroad, three could not attend due to poor general health unrelated to surgery of the foot, two were excluded due to violation of data (i.e., changing the wording of questions and scales, as well as marking two alternatives on questions and scales), and one due to resections of PIMNs of the same foot in different web-spaces on 2 different occasions (not a case of failed surgery). Thus, the resulting followup rate was 86% (125 of 145 patients). Demographic data and clinical characteristics of the subjects at baseline are provided in Table 1. Operative Technique Surgery was indicated when conservative treatment had failed and pain, in either the second or third or both webspaces, persisted. Each surgeon had at least 20 years of experience. Either a longitudinal plantar incision, or a dorsal incision was used (Table 1). The operations in both groups were performed on an outpatient basis, using a tourniquet. In the plantar group, all operations were performed under intravenous regional anesthesia, whereas, in the dorsal group, a posterior tibial nerve block together with local anesthesia was used in 55 of the 59 feet, and local anesthesia only in 4 of the 59 feet. Both the longitudinal plantar incision and the dorsal incision were performed with use of a standard technique as earlier described by Åkermark et al. 1 and Coughlin and Pinsonneault, 6 respectively. The two different surgical approaches are shown in Figure 1. A division of the deep transverse ligament was performed in conjunction with the dorsal incision, but not with the plantar incision. Patients were encouraged to elevate their feet, limit activities, and were allowed weightbearing at 2 to 3 days postoperatively, or as tolerated. In both groups, sutures were removed 2 to 3 weeks after surgery. Followup Investigations Two independent orthopedic surgeons performed similar numbers of followup evaluations in the two groups. The followups included a review of history data, surgical charts, histological examination data, and other outcome variables. A patient-administered questionnaire was used to evaluate the primary outcome variable, pain at followup [evaluated by a 100-mm visual analog scale (VAS), with the endpoints no pain and worst possible pain], The secondary outcome

3 138 ÅKERMARK ET AL. Foot & Ankle International/Vol. 29, No. 2/February 2008 Fig. 1: A cross-section through the metatarsal heads with some joint structures indicated. A, The plantar approach. Patient is in a prone position. B, The dorsal approach. Patient is in a supine position. Table 2: Patients evaluation of pain at daily activities at followup measured by a 100-mm VAS scale and by a 4-step scale Pain Plantar Group Dorsal Group p-value VAS (SD) 8 (17) mm 11 (19) mm 0.41 Never 48/72 (67 %) 39/55 (71 %) 0.70 Monthly-weekly-daily 24/72 (33 %) 16/55 (29 %) variables were subjective residual pain in the operated foot (estimated by a 4-step Likert scale, with the categories of never, monthly, weekly, or daily ), and the overall satisfaction rating of the result of surgery (excellent, good, fair, or poor). The following adverse events were also estimated: restriction of daily activities, scar tenderness and patient s subjective estimation of sensory loss of the involved toes, all evaluated by a 4-step Likert scale with the categories of none, slight, moderate, or severe. Expectations with operative results were evaluated by a VAS estimation with the endpoints, not at all and totally. Patients were also asked if they would have the surgery performed again on the other foot under similar circumstances. Finally, the patients were asked for how long they had been on sick leave and their records were retrospectively reviewed. Sensory loss of the involved area including alongside the toes and the tips of the distal phalanges was tested with a light pin-prick touch. Any tenderness of the scar was also noted. All these findings were evaluated by a 4-step Likert scale, with the categories none, slight, moderate, and severe. Statistical Analysis The descriptive results are presented as mean (±SD), or median (range). The groups are compared using nonparametric tests, most often Fischer s exact test. The risk of complications was described as an odds ratio with 95% confidence interval (95% CI). The level of significance was set at p < The intraobserver, as well as the intrapatient, reliability of the main and some of the secondary outcome variables in the survey were tested in a separate analysis. The patientoriented subjective estimations of the 5 ordinal variables and 2 continuous (VAS) variables, as well as the investigatorestimations of 4 ordinal and 2 continuous (VAS) variables, were tested for reliability by applying the survey after a oneweek interval for the patients, and 1- to 2-hour intervals for the investigator. RESULTS With the numbers available, there were no significant differences between the two groups regarding baseline data, except preoperative mean pain duration, mean followup time and number of preoperative corticosteroid injections (see Table 1). Histology revealed confirmation of neuromas in 72 of 73 feet (99%) in the plantar group (one normal nerve resected), and in 56 of 59 feet (95%) in the dorsal group (one artery and two granulation tissues were resected). At the followup evaluation, there were no significant differences between the groups in pain during daily activities, neither by VAS nor by Likert scale (Table 2). Similar outcome findings were found between the groups regarding both activity restrictions, and the scar tenderness (Table 3). There were significantly fewer patients complaining of subjective sensory loss of the involved lesser toes in the

4 Foot & Ankle International/Vol. 29, No. 2/February 2008 MORTON S NEUROMA, PLANTAR VERSUS DORSAL 139 Table 3: Patients assessment of restrictions in daily activities, incision tenderness, as well as sensory loss of the involved lesser toes at followup None Slightmoderate-severe p-value Activity restrictions Plantar Group 54/73 (74 %) 19/73 (26 %) 0.69 Dorsal Group 39/56 (70 %) 17/56 (30 %) Scar tenderness Plantar Group 51/73 (70 %) 22/73 (30 %) 0.10 Dorsal Group 47/56 (84 %) 9/56 (16 %) Sensory loss Plantar Group 34/72 (47 %) 38/72 (53 %) 0.03 Dorsal Group 15/56 (27 %) 41/56 (73 %) plantar group (53%) than in the dorsal group (73%), (p = 0.03) (Table 3). Of the patients stating sensory loss, numbness was considered bothersome in comparable numbers of patients in the two groups: 22 of 38 (58%) in the plantar group and 24 of 40 (60%) in the dorsal group. Significantly more postoperative treatments, such as corticosteroid injections and NSAID medications for various complaints were given to patients in the dorsal group (16 of 59) compared to the plantar group (2 of 70) (p < 0.001). At followup, the physical evaluation of the involved lesser toes (digits II-IV) revealed significantly lower prevalence of slight to severe sensory losses in the plantar group, 49 of 207 (24%) compared to the dorsal group, 65 of 165 (39%, p = 0.002). Overall satisfaction was, in the plantar group, rated as excellent by 53 patients (73%), good by 11 (15%), fair by 4 (5%) and poor by 5 (7%). The corresponding numbers in the dorsal group were 34 (61%), 13 (23%), 5 (9%) and 4 (7%). With the numbers available, there were no significant differences between the groups regarding the overall clinical outcome (excellent or good) as subjectively assessed by the patients. The bilaterally operated patients, as well as those who had two web-spaces operated simultaneously on the same foot, showed similar results. All bilateral operations, except one in the dorsal group (excellent result), were performed on different occasions, whereas all operations in two web-spaces (2nd and 3rd) were performed simultaneously. The values regarding expectations of surgical outcome were also comparable: mean 88 (±25) mm (n = 71) in the plantar group versus 81 (±29) mm (n = 56) in the dorsal group. Of the unilaterally operated patients, 60 of the 65 patients (92%) in the plantar group and 45 of the 50 patients (90%) in the dorsal group would have the other foot operated in a similar way, if necessary NS, not significant. Patients in the dorsal group were on sick leave, postoperatively, significantly longer [mean 3.7 (±2.8) weeks, n = 45], than patients in the plantar group [mean 2.2 (±1.7) weeks, n = 49; p = 0.003]. Patients in the dorsal group had significantly more complications (n = 10, 17%) than patients in the plantar group (n = 4, 5%);[RR= 1.13 (95% CI: ), p = 0.047]. The rate of re-operation was the same, 5% in each group (NS) All complications in the plantar group, three minor (2 3 mm) hypertrophic scars, and one minor epidermal inclusion cyst that developed postoperatively, were surgically excised before the followup. These four patients were, however, included in the followup evaluation as the operations were of a minor nature and performed with the same type of approach. The following complications in the dorsal group were re-operated: One of three missed nerves, one recurrent pain case (missed nerve-branch) and one painful stump neuroma. There were five additional complications in the dorsal group: three neuropathy cases and two postoperative infections. The histological examinations of specimens from the three patients with missed nerves showed an artery in one case and granulation tissue in the two other cases. The three re-operations in the dorsal group were performed with a plantar incision during the followup period, and these patients, regarded as failures of the dorsal approach, were thus excluded from the followup examinations. Reliability of the Survey The intrapatient as well as intra-investigator, agreement for both ordinal and continuous variables showed excellent agreement (a kappa value of greater than 0.9) in all 4 cases. DISCUSSION This retrospective study was performed as an attempt to compare the outcome of plantar and dorsal incisions in the treatment of PIMN. Even though the primary outcome variable of pain and overall satisfaction were similar in

5 140 ÅKERMARK ET AL. Foot & Ankle International/Vol. 29, No. 2/February 2008 both groups, some strikingly significant differences in the secondary outcome variables, in favor of the plantar group, were found. These included rate of complications, both subjective and objective sensory loss, and the number of sick leave weeks. Also of note, there were no differences between the groups regarding scar tenderness. The three patients in the dorsal group who had to be re-operated due to poor results with a plantar incision during the followup period were considered as failures and thus excluded from the overall satisfaction rating, pain evaluation and other outcome analyses at the 2-year followup. The negative influence the three failures would have had on the outcome would markedly have strengthened the differences between the groups, in favor of the plantar group. There are, to our knowledge, only four other studies that compare the results of plantar versus dorsal incisions of PIMNs. 12,20,22,25 These studies suffer, however, from considerable confounding factors such as simultaneous surgery for other conditions, 12 insufficient information on followup, unequal or limited number of patients 20,22,25 as well as lack of pathology reports. 12,25 It is difficult to draw confident conclusions from such studies. We are aware that our own retrospective study has its limitations, but tried to minimize the confounding factors such as including only primary neuromas. Furthermore, we managed to collect larger groups than the 4 studies cited above, had at least a 2-year followup, performed histological assessment of all specimens, and had two independent orthopaedic surgeons unrelated to the surgery perform the evaluations. We believe that a histo-pathological evaluation of a specimen, to verify removal of the nerve, enhances the value of our study. It certainly facilitated understanding the cause of failure in the three cases (with dorsal incisions), where non-neural tissue was removed. The validity of studies without histological examinations of the excised tissue may therefore confound the results. Pre- and postoperative radiographic examinations were not performed on a regular basis in our patients, which limit our possibilities to rule out possible osseous pathologies. The different models of anesthesia used in the two groups may, in part, have affected the difference in sensory loss of the lesser toes between the groups. In at least one patient, where a sudden nerve-pain appeared during the ankle nerve block anesthesia, minor damage to the tibial posterior nerve may have resulted in a postoperative neuropathy at followup in the dorsal group. The comparison of the two incisions in the present report may suggest selection bias, due to the two surgeons. However, the patients of both surgeons were mainly referrals from various general practitioners, without any apparent selection preferences. The two operating surgeons had equally long (more than 20 years) experience with the plantar and the dorsal approach, respectively, and both had exclusively been faithful to their choice of technique over the years. Thus, any learning curve as a confounding factor can be ruled out. A significantly longer followup time was noted in the dorsal group (p = ). In order to elucidate whether length of followup had any influence on outcome variables, we compared the results in patients with close to 2-years followup versus those with longer followup times. We found, however, with the numbers available, no statistical differences in the outcome variables at two yeasr of followup between the dorsal group and the plantar group. The significant differences between the two groups in our study, with more frequent and more serious complications in the dorsal group, is noticeable. The 3 missed nerves in the dorsal group in our study, verified by histological assessment (an artery in one case and granulation tissue in 2 cases), may be due to poor exposure and an enhanced risk of misinterpretation of plantar structures with the dorsal incision. We find support in the literature for the opinion that the plantar incision is easier and gives a better exposure, 3,8,14,18 but we are aware that there are pro and cons with the two incisions. Nøkleby 22 as well as Dereymaker 7 reported one missed nerve each with a dorsal approach (arteries resected in both cases). Also Dick, 8 using a dorsal approach, reported one missed nerve. Reported complications in other studies with a plantar incision are 4%, 18 8%, 20 31%, 21 and with a dorsal incision 2%, 7 8%, 9 and 19%. 20 In the plantar group in our study, 2 minor hyperkeratotic scars and one inclusion cyst, occurred (5%). Other studies, where a plantar incision had been performed and where information on complications were described, presented from zero to 27% painful incisional scars. 3,12 However, a similar occurrence of incisional problems (7% to 21%) has been reported with dorsal incisions. 3,20 In contrast to our expectations of more pronounced trouble with the plantar incisions, patients activity restrictions and subjective assessment of scar tenderness showed no significant differences between the approaches (Table 3). We agree with Thomson et al. 23 that there are, at best, very limited indications to suggest that dorsal incisions for resection of the plantar digital nerve may result in less symptomatic postoperative scars when compared to plantar excision of the nerve. Sensory testing can be extremely variable and of uncertain value. 2 The simple tests generally performed for the objective sensory differences between the groups in this and other studies may be of doubtful value and significance. With this in mind, both patients subjective assessment of sensory loss (see Table 3), as well as the objective values at physical examination in this study, lead to poorer results with the dorsal incision. Only a few studies, using dorsal incisions, have findings of postoperative sensory loss. Coughlin and Pinsonneault 6 noted 51% subjective numbness and 72% objective numbness between the digits and 41% numbness in the terminal aspects of the digits. Mann and Reynold 15

6 Foot & Ankle International/Vol. 29, No. 2/February 2008 MORTON S NEUROMA, PLANTAR VERSUS DORSAL 141 reported 68% objective numbness between the involved toes at followup. Another possible explanation for the differences in sensory-loss at followup may be the number of corticosteroid injections given. Patients in our dorsal group received significantly more injections, both pre- and postoperatively, than patients in the plantar group possibly leading to more nerve damage (Table 1 and Results above). The patients subjective satisfaction ratings of the surgery (excellent or good results) were comparable between the groups in our study, 88% and 84%, respectively. The vast majority of patients who had bilateral operations, as well as those who had their web-spaces 2 and 3 operated simultaneously in the same foot, showed surprisingly high values of excellent or good results. CONCLUSION We found that the main outcome of PIMN surgery was similar using plantar or dorsal incisions, but noticed significantly more problems associated with the dorsal approach regarding the number of complications and sensory loss, both subjectively and objectively. Also, the patients in the dorsal group were on sick-leave significantly longer. The occurrence of mostly small, occasional hypertrophic scars could be troublesome for the patients but should be compared with the more serious complications that occurred with the dorsal approach in this and other studies. These results have encouraged us to perform a randomized, prospective study. Our preliminary results in that study support our findings and conclusions presented here. REFERENCES 1. Åkermark, C; Saartok, T; Zuber, Z: A prospective 2-year followup study of plantar incisions in the treatment of primary intermetatarsal neuromas (Morton s Neuroma). Accepted for publication in Foot and Ankle Surg Amis, JA: Primary interdigital neuroma resection. In: Johnson, KA (ed): Master techniques in orthopedic surgery. The Foot and Ankle. New York, Raven Press, , Beskin, JL; Baxter, DE: Recurrent pain following interdigital neurectomy a plantar approach. Foot Ankle. 9:34 39, Betts, L: Morton s metatarsalgia: neuritis of the fourth digital nerve. Med. J. Australia. 1: , Colgrove, RC; Huang, EY; Barth, AH; Greene, MA: Interdigital neuroma: intermuscular neuroma transposition compared with resection. Foot Ankle Int. 21: , Coughlin, MJ; Pinsonneault, T: Operative treatment of interdigital neuroma. J. Bone Joint Surg. Am. 83: , Dereymaeker, G; Schroven, I; Steenwerckx, A; Stuer, P: Results of excision of the interdigital nerve in the treatment of Morton s metatarsalgia. Acta Orthop. Belg. 62:22 25, Dick, W: Morton s metatarsalgia. Orthopäde 11: , Friscia, DA; Strom, DE; Parr, JW; Saltzman, CL; Johnson, KA: Surgical treatment for primary interdigital neuroma. Orthopedics 14: , Gauthier, G: Thomas Morton s disease: a nerve entrapment syndrome. A new surgical technique. Clin. Orthop. 142:90 92, Giannini, S; Bacchini, P; Ceccarelli, F; Vannini, F: Interdigital neuroma: clinical examination and histopathologic results in 63 cases treated with excision. Foot Ankle Int. 25:79 84, Jarde, O; Trinquier, JL; Pleyber, A; Meire, P; Vives, P: Treatment of Morton s neuroma by neurectomy. A review of 43 cases. Revue de Chirurgie Orthopédique. 81: , (In French). 13. Kaplan, EB: Surgical approach to the plantar digital nerves. Bull. Hosp. Joint Dis. 11:96 97, Karges, DE: Plantar excision of primary interdigital neuromas. Foot Ankle. 9: , Mann, RA; Reynolds, JC: Interdigital neuroma a critical clinical analysis. Foot Ankle. 3: , McElvenny, RT: The etiology and surgical treatment of intractable pain about the fourth metatarsophalangeal joint (Morton s toe). J. Bone Joint Surg. Am. 25: , McKeever, DC: Surgical approach for neuroma of plantar nerve (Morton s metatarsalgia). J. Bone Joint Surg. Am. 34:490, Morris, MA: Morton s metatarsalgia. Clin. Orthop. Relat. Res. 127: , Mulder, JD: The causative mechanism in Morton s metatarsalgia. J. Bone Joint Surg. Br. 33: 94 95, Nashi, M; Venkatachalam, AK; Muddu, BN: Surgery of Morton s neuroma: dorsal or plantar approach? J. R. Coll. Surg. Edinb. 42:36 37, Nissen, KI: Plantar digital neuritis, Morton s metatarsalgia. J. Bone Joint Surg. 30B:84 94, Nøkleby, K: Morton s metatarsalgia. Acta Orthop. Scand. 56: , Thomson, CE; Gibson, JNA; Martin, D: Interventions for the treatment of Morton s neuroma. The Cochrane Database of Systematic Reviews 2004, Issue Younger, AS; Claridge, RJ: The role of diagnostic block in the management of Morton s neuroma. Can. J. Surg. 41: , Wilson, S; Kuwada, GT: Retrospective study of the use of a plantar transverse incision versus a dorsal incision for excision of neuroma. J. Foot Ankle Surg. 34: , 1995.