Long-Term Results of the Modified Hoffman Procedure in the Rheumatoid Forefoot

Thomas, S. MBChB, BSc, MRCS; Kinninmonth, A.W.G. MB, FRCS(Ed)Orth; Kumar, C. Senthil FRCS(Tr & Orth)

Journal of Bone & Joint Surgery - American Volume:
doi: 10.2106/JBJS.C.01696
Scientific Articles

Background: Rheumatoid arthritis commonly affects the forefoot, causing metatarsalgia, hallux valgus, and deformities of the lesser toes. Various types of surgical correction have been described, including resection of the lesser-toe metatarsal heads coupled with arthrodesis of the great toe, resection arthroplasty of the proximal phalanx or metatarsal head, and metatarsal osteotomy. We report the results at an average of five and a half years following thirty-seven consecutive forefoot arthroplasties performed in twenty patients by one surgeon using a technique involving resection of all five metatarsal heads.

Methods: All patients were treated with the same technique of resection of all five metatarsal heads through three dorsal incisions. All surviving patients were asked to return for follow-up, which included subjective assessment (with use of visual analogue pain scores, AOFAS [American Orthopaedic Foot and Ankle Society] foot scores, and SF-12 [Short Form-12] mental and physical disability scores), physical examination, and radiographic evaluation.

Results: All results were satisfactory to excellent in the short term (six weeks postoperatively), and no patient sought additional surgical treatment for the feet. A superficial infection subsequently developed in two feet, and two feet had delayed wound-healing. At an average of 64.9 months postoperatively, the average AOFAS forefoot score was 64.5 points and the average hallux valgus angle was 22.3°. There were no reoperations.

Conclusions: Resection of all five metatarsal heads in patients with metatarsalgia and hallux valgus associated with rheumatoid arthritis can be a safe procedure that provides reasonable, if rarely complete, relief of symptoms.

Level of Evidence: Therapeutic Level IV. See Instructions to Authors for a complete description of levels of evidence.

Author Information

1 Department of Trauma and Orthopaedics, Ninewells Hospital, Dundee DD1 9SY, United Kingdom. E-mail address for S. Thomas: simon_thomas97@hotmail.com

2 Department of Orthopaedics, Golden Jubilee National Hospital, Beardmore Street, Clydebank G81 4HX, United Kingdom

3 Department of Orthopaedic and Trauma Surgery, Glasgow Royal Infirmary, Glasgow G4 0SF, United Kingdom

Article Outline

Nearly 90% of patients with rheumatoid arthritis have involvement of the feet, with soft-tissue instability and joint destruction resulting in hallux valgus, subluxation or dislocation of the lesser toes with valgus sweeping, and displacement of the plantar fat pad with associated metatarsalgia1 (Fig. 1-A). Often, the patient needs to wear custom-made shoes and has difficulty walking because of pain. Efforts to treat these problems have resulted in a number of surgical procedures with countless modifications2. Most attention has been given to the treatment of the first ray, with some surgeons fusing the metatarsophalangeal joint3-5 and others resecting part of it (the proximal half of the proximal phalanx, or the metatarsal head)2,6-11.

This study addressed the long-term results of one such procedure, involving resection of all five metatarsal heads through three dorsal incisions, performed by one surgeon over a nine-year period.

Back to Top | Article Outline

Materials and Methods

This retrospective cohort study was approved by the hospital review board, and all patients gave informed consent before participating. Patients were identified by a search of an operating theater logbook, and retrieval of case notes confirmed that all patients were treated with the same operation by the same surgeon. The main indication for surgery in all patients was metatarsalgia associated with hallux valgus and rheumatoid arthritis. Conservative management, including the use of domed insoles and orthotic shoes with a wide toe-box, had failed for all patients. The clinical course, including any complications, repeat referrals, or subsequent foot surgery, was noted.

All patients were examined by one of us (S.T.) in the clinic. Complications were recorded, and a 10-cm visual analogue scale was used to record pain at rest and pain while the patient walked in his or her usual shoes. The end points of this scale were labeled as “no pain at all” (0) and “worst pain imaginable” (10). The patients then stated whether they could wear any shoes that they liked, had to be careful choosing shoes, or had to wear specially made shoes. They also stated how well they could walk and what aids they needed and whether they had any problems balancing. Each patient then filled out a standard SF-12 (Short Form-12) questionnaire (four-week recall version)12, which was scored with use of the manual13, to derive a physical component summary score and a mental component summary score.

On examination, any obvious residual deformity when the patient was weight-bearing, the presence of calluses, and the stability and mobility of the first metatarsophalangeal and the interphalangeal joints were noted.

Standing anteroposterior radiographs were then made, and the hallux valgus angle was measured on each. The metatarsophalangeal angle of the second toe was also measured in the sagittal plane (between the long axes of the metatarsal and the proximal phalanx), and the axial alignment of the lesser toes was assessed according to the method described by Coughlin4.

AOFAS (American Orthopaedic Foot and Ankle Society) forefoot scores were calculated for all feet14. The results were also classified with use of a slightly modified version of the criteria described by Mann and Thompson3 (Table I). A grade of excellent was given if the patient had minimal pain on walking (<1 on the visual analogue scale of 0 to 10) and was able to wear any shoes that he or she liked. The result was considered to be good when there was minimal pain but the patient was restricted with regard to the types of shoes that could be worn. The result was considered to be fair when the patient had moderate pain (1 to 8 on the visual analogue scale) but had improvement compared with the preoperative state, regardless of which shoes could be worn. The result was considered to be poor when the patient had severe pain (>8), regardless of shoe wear (Table I).

The Spearman rank correlation coefficient was used to test for correlation between the age at the operation, postoperative time, and hallux valgus angle and three subjective clinical outcome measures: the score for pain with walking, the score for resting pain, and the AOFAS score.

Back to Top | Article Outline
Surgical Procedure

No perioperative antibiotics were used. Three dorsal incisions were made: the first was over the hallux metatarsophalangeal joint; the second, between the second and third metatarsophalangeal joints; and the last, between the fourth and fifth metatarsophalangeal joints. Through these incisions, all five metatarsal heads were excised, with the surgeon working from medial to lateral to produce a smooth crescentic curve of the remaining metatarsal necks (Fig. 1-B). The wounds were sutured, and no Kirschner wires were used to stabilize the metatarsophalangeal joints, although a carefully applied padded bandage was used to maintain the positions of the toes. All patients were then restricted to bed rest for a week in the hospital ward and were allowed to walk only after the wounds were noted to be healing with no signs of infection.

Back to Top | Article Outline


From 1994 to 2001, forty-three feet in twenty-three patients were operated on with this procedure. There were nineteen women and four men, with an average age of 60.9 years (range, 36.1 to 79.8 years) at the time of the operation. Twenty-two right feet and twenty-one left feet were operated on. At the time of this review, three patients had died of causes unrelated to the surgery. The remainder were evaluated for this study. Thirty-seven feet in twenty patients were assessed at an average of 64.9 months (range, twenty-two to 108 months) postoperatively.

Two wounds had delayed healing, and another two had a superficial infection and were treated with oral antibiotics. All wounds had healed well by the time of discharge from the clinic, six weeks after the surgery. No patient had any more operations on the feet.

At the time of final follow-up, the average score for pain at rest, on the visual analogue scale, was 1.8 (range, 0 to 6.1). The average score for pain when walking was 4.1 (range, 0 to 8.9). All patients stated that the pain had decreased from the preoperative level, although this was a retrospective judgment. Twelve patients (60%) reported problems with balance. Any shoes that the patient liked could be worn on five feet, the shoes had to be chosen carefully for twenty-four feet, and eight feet required specially supplied shoes (orthotic shoes with a wide toe-box).

On examination, eighteen feet had no residual deformity, fourteen had mild deformity, and five had severe deformity. All were hallux valgus deformities. Eleven feet had callosities: six were on the dorsal aspect of the second proximal interphalangeal joint, three were on the lateral aspect of the fifth toe, and two were underneath the first metatarsophalangeal joint.

Weight-bearing dorsoplantar radiographs demonstrated an average hallux valgus angle of 22.3° (range, 0° to 52°) and an average second-toe metatarsophalangeal angle of 23.1° (range, 0° to 52°). All of the lesser-toe proximal phalanges were aligned in thirty feet. One toe had subluxated in five feet, two toes had subluxated in one foot, and three toes had subluxated in another foot.

The AOFAS scores ranged from 37 to 90 points (of a possible 100 points for a normal asymptomatic foot), with an average of 64.5 points. An association was found only between the age at the operation and the AOFAS score (p = 0.036) and between the hallux valgus angle and the AOFAS score (p = 0.0032). Both of these relationships were inverse. The average SF-12 physical component summary score was 33.1 points, and the average mental component summary score was 43.9 points. According to the classification outlined in Table I, there were three excellent, eight good, seventeen fair, and nine poor results (Table II).

Back to Top | Article Outline


Many procedures have been described for the treatment of hallux valgus and the related forefoot deformities associated with rheumatoid arthritis. In 1912, Hoffman6 resected all five metatarsal heads through a single transverse plantar incision. Since then, surgeons have reported varied results with dorsal and plantar incisions, with repositioning the plantar fat pad, for excision of the metatarsal heads and/or the proximal halves of the proximal phalanges2. Perhaps the most controversial issue has been the treatment of the first ray, with various approaches being advocated, including excision of the first metatarsal head, excision of the proximal half of the proximal phalanx, or arthrodesis of the metatarsophalangeal joint.

Barton described the results of sixty-five operations performed with a variety of techniques, all of which involved resection of part of the first metatarsophalangeal joint without arthrodesis2. Although most patients were satisfied because they had received some relief from pain, only twenty-nine (45%) were pain-free. Stockley et al.7 reported that forty-two (70%) of sixty feet were pain-free at three years following a Kates procedure (a plantar incision with resection of all metatarsal heads and replacement of the fat pad)8. Patsalis et al. followed thirty-six feet in twenty-three patients for an average of more than ten years after resection of all five metatarsal heads through a plantar incision and found a 56% rate of dissatisfaction due to pain9. Other authors have found less than satisfactory results with resection arthroplasty of the first metatarsophalangeal joint10,11.

Fusion of the first metatarsophalangeal joint may provide better results, as advocated by Mann and Thompson3, Coughlin4, and others5. However, Beauchamp et al. claimed that there was no difference in pain between thirty-four feet treated with arthrodesis and thirty treated with excision, although they noted better results in terms of appearance and function following the arthrodeses5.

We described the results of what we consider to be a modified Hoffman procedure, which differs from the original technique only in terms of the surgical approach. Our results, after more than five years of follow-up, present a mixed comparative picture. When the results were assessed with use of the criteria of Mann and Thompson3, <30% were classified as excellent or good and nearly a quarter were classified as poor, findings that were notably inferior to those reported by Mann and Thompson (Table II). However, these criteria are strict, allowing an excellent or good grading only if there is no pain at all. Patients with rheumatoid arthritis frequently have multiple foot deformities, with metatarsalgia accounting for only part of their discomfort. If the metatarsal heads alone are treated, there may be some residual pain that does not originate from the forefoot but is difficult for the patient to distinguish from forefoot pain. It is noteworthy that only two of the thirty-seven feet in our series had calluses under the first metatarsal head and none had a callus under the second metatarsal head, suggesting that the residual pain may indeed not have been metatarsalgia. All patients were pleased with the result; thus, subjectively, the direct comparison with the results reported by Mann and Thompson may not accurately reflect the success of this procedure in this patient group.

To provide a more standardized measure of forefoot pain and function, AOFAS forefoot scores have been used since 199414. However, there are fewer published series with which to compare these scores. In 2000, Coughlin4 assessed the results of a technique involving arthrodesis of the first metatarsophalangeal joint in forty-seven feet in thirty-two patients. At an average of seventy-four months, the average AOFAS forefoot score was 69 points, the average postoperative hallux valgus angle was 20°, and the average second-toe metatarsophalangeal angle was 14°. All patients could wear ordinary shoes, but fourteen feet (30%) required a second operation (e.g., removal of the hardware or additional procedures on the interphalangeal joint or on one or more of the lesser toes). In comparison, our average AOFAS score was 64.5 points, none of our patients required another operation, the average hallux valgus angle was 22.3°, and the average secondtoe metatarsophalangeal angle was 23.1° (Table III). This is a more favorable comparison, which highlights the importance of clearly defining the aims of the operation.

The SF-12 survey was used to assess the physical and mental disability levels of our patients12. We used only the physical component summary and mental component summary scores. The survey is designed to give an average (and standard deviation) of 50.0 ± 10.0 for each score in a normal population. The physical component summary score of 33.1 points in our study indicates a substantial level of physical disability compared with a normal population, as would be expected. The mental component summary score of 43.9 points suggests that, despite the physical disability of this group of patients, they did not have a substantial mental disability due to their illness. Although these results may not be surprising in themselves, they allow better comparisons in future studies.

This technique is safe. There were no complications apart from two cases of delayed wound-healing and two cases of superficial wound infection, all of which resolved without difficulty. In addition, there was no need for additional operations.

It is difficult to draw definite conclusions about how this technique compares with others. There is a lack of studies in which validated outcome scales were used, so comparison has limited value. The polyarthritic nature of rheumatoid arthritis as well as its variable systemic and psychological effects also makes valid comparison difficult.

We tried to allow for these factors by including a visual analogue scale for pain, the AOFAS score for foot pain and function, and the SF-12 score, which provides a measure of physical and psychological impairment.

This study's strengths are that we had a 100% rate of follow-up of survivors at an average of more than five years, all procedures were performed by the same surgeon in the same fashion, and standardized scoring systems were used to aid comparison with other studies. Its weaknesses are that we were not able to include pedobarographic data or detailed preoperative assessments.

In conclusion, the results of thirty-seven forefoot arthroplasties in twenty patients were reviewed at an average of 5.5 years. The procedure involved excision of all five metatarsal heads through three dorsal incisions. All of the patients in the study were satisfied with the result of the operation. This procedure provides a safe method of treating metatarsalgia associated with hallux valgus and rheumatoid arthritis. Comparisons with studies of other techniques are variable and inconclusive. ▪

The authors did not receive grants or outside funding in support of their research or preparation of this manuscript. They did not receive payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the authors are affiliated or associated.

Investigation performed at the Department of Orthopaedic and Trauma Surgery, Glasgow Royal Infirmary, Glasgow, United Kingdom

1. , Kile TA. The foot in rheumatoid arthritis. In: Bulstrode CJK, Buckwalter J, Carr A, Fairbank J, Marsh L, editors. Oxford textbook of orthopedics and trauma. 1st ed. New York: Oxford University Press; 2002. p 1279-90.
2. . Arthroplasty of the forefoot in rheumatoid arthritis. J Bone Joint Surg Br. 1973;55: 126-33.
3. , Thompson FM. Arthrodesis of the first metatarsophalangeal joint for hallux valgus in rheumatoid arthritis. J Bone Joint Surg Am. 1984;66: 687-92.
4. . Rheumatoid forefoot reconstruction. A long-term follow-up study. J Bone Joint Surg Am. 2000;82: 322-41.
5. , Kirby T, Rudge SR, Worthington BS, Nelson J. Fusion of the first metatarsophalangeal joint in forefoot arthroplasty. Clin Orthop. 1984;190: 249-53.
6. . An operation for severe grades of contracted or clawed toes. Am J Orthop Surg. 1912;9: 441-9.
7. , Betts RP, Getty CJ, Rowley DI, Duckworth T. A prospective study of forefoot arthroplasty. Clin Orthop. 1989;248: 213-8.
8. , Kessel L, Kay A. Arthroplasty of the forefoot. J Bone Joint Surg Br. 1967;49: 552-7.
9. , Georgousis H, Göpfert S. Long-term results of forefoot arthroplasty in patients with rheumatoid arthritis. Orthopedics. 1996;19: 439-47.
10. , Johnson KA. Keller arthroplasty in combination with resection arthroplasty of the lesser metatarsophalangeal joints in rheumatoid arthritis. Foot Ankle. 1988;9: 75-80.
11. , Piirainen H, Kettunen P. Resection arthroplasty of the metatarsophalangeal joints in rheumatoid arthritis. A follow-up study of 100 patients. Scand J Rheumatol. 1980;9: 257-65.
12. , Kosinski M, Keller SD. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care. 1996;34: 220-33.
13. , Kosinski M, Keller SD. SF-12: how to score the SF-12 Physical and Mental Health Summary Scales. 2nd ed. Boston: The Health Institute, New England Medical Center; 1995.
14. , Alexander IJ, Adelaar RS, Nunley JA, Myerson MS, Sanders M. Clinical rating systems for the ankle-hindfoot, midfoot, hallux, and lesser toes. Foot Ankle Int. 1994;15: 349-53.
Copyright 2005 by The Journal of Bone and Joint Surgery, Incorporated