 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1999 The Journal of Bone and Joint Surgery, Inc. Triple Arthrodesis: Twenty-five and Forty-four-Year Average Follow-up of the Same Patients*   , IOWA CITY, IOWAInvestigation performed at the Department of Orthopaedic Surgery, University of Iowa, Iowa City
Background: Triple arthrodesis is used to treat major deformities of the hindfoot and is often performed in young patients. The purpose of this study was to assess the long-term outcomes of triple arthrodesis in young patients. Methods: Sixty-seven feet of fifty-seven patients were evaluated at an average of twenty-five and forty-four years after triple arthrodesis. The most common indication for the operation was neuromuscular imbalance of the hindfoot, which was secondary to poliomyelitis in thirty-seven feet (55 percent), Charcot-Marie-Tooth disease in six (9 percent), spinal cord abnormalities in four (6 percent), cerebral palsy in three (4 percent), and Guillain-Barré syndrome in one (1 percent). Results: Fifty-two feet (78 percent) had some residual deformity after the arthrodesis. However, these deformities appeared to be nonprogressive between 1973 and 1994. Pseudarthrosis occurred in thirteen feet. Thirty feet or ankles (45 percent) were painful at the first follow-up evaluation, and thirty-seven feet or ankles (55 percent) were painful at the second follow-up evaluation. Of the thirty feet or ankles that were painful at the first follow-up evaluation, twenty-three were painful at the second follow-up evaluation. Of the thirty-seven feet or ankles that were not painful at the first follow-up evaluation, fourteen were painful at the second follow-up evaluation. Eighteen patients (32 percent) needed walking support at the time of the first follow-up, and thirty-nine patients (68 percent) needed it at the time of the second follow-up. Two of the patients who needed support at the first follow-up evaluation did not need it at the second follow-up evaluation. At the first follow-up evaluation, twenty-one ankles (31 percent) had no radiographic evidence of degenerative changes. However, by the second follow-up evaluation, all of the ankles had some degenerative changes. Similar progressive arthritic findings were noted at the naviculocuneiform and tarsometatarsal joints. According to the system of Angus and Cowell, the overall result at the time of the first follow-up was rated as good in fifty feet (75 percent) and as fair in seventeen feet (25 percent). At the time of the second follow-up, nineteen feet (28 percent) were rated as good, forty-six (69 percent) were rated as fair, and two (3 percent) were rated as poor. Conclusions: Despite progressive symptoms and radiographic degeneration in the joints of the ankle and midfoot, fifty-four patients (95 percent) were satisfied with the result of the operation. The triple arthrodesis was a satisfactory solution for imbalance of the hindfoot in this group of patients.
The triple arthrodesis consists of arthrodeses of the talocalcaneal, calcaneocuboid, and talonavicular joints, as described in 1923 by Ryerson6. The procedure was designed to stabilize the foot, correct deformity, relieve pain, and achieve a plantigrade foot. The operation, which is still in use today, has been the preferred procedure for stabilization of the hindfoot for most of this century. However, questions have been raised regarding the long-term functional results of this procedure and whether a painless, plantigrade foot can be maintained over the long term1-7. Several investigators have described the results of triple arthrodesis after ten to twenty-five years of follow-up. Angus and Cowell2 reported the results of eighty triple arthrodeses after an average follow-up period of thirteen years. In their series, pseudarthrosis was noted in eighteen feet (23 percent), degenerative changes were noted in thirty-one ankles (39 percent), and degenerative changes in the midfoot were noted in forty-three feet (54 percent). Poor results were seen in twenty-five feet (31 percent). Drew4 studied twenty-seven patients who had had thirty-one triple arthrodeses; the follow-up period was eleven to twenty years. In his series, eight patients (30 percent) had mild or moderate pain and ten ankles (32 percent) had radiographic evidence of osteoarthritis. He found that the position of the foot and arthritis of the ankle joint did not seem to have an important effect on the long-term result. Adelaar et al.1 reported on a series of sixty-two operations with an average duration of follow-up of twenty years. Five talonavicular pseudarthroses (8 percent) were noted, and radiographic evidence of osteoarthritis was reported in nine ankle joints (15 percent), eighteen naviculocuneiform joints (29 percent), and ten tarsometatarsal joints (16 percent), but none of the arthritic changes could be determined to be associated with symptoms. The best results were in patients who had good muscle balance. Wetmore and Drennan7 reported the long-term results of triple arthrodesis in patients who had Charcot-Marie-Tooth disease. They followed thirty feet for an average of twenty-one years. Only seven feet (23 percent) were rated as having a good or excellent result, and radiographic evidence of osteoarthritis was noted in twenty-three feet (77 percent). To the best of our knowledge, the study of triple arthrodesis with the longest duration of follow-up was recently reported by Haritidis et al.5. Forty-two patients were evaluated at an average of twenty-five years postoperatively. Thirteen patients (31 percent) had a good result, twenty-six (62 percent) had a fair result, and three (7 percent) had a poor result. Twelve ankles (29 percent) showed osteoarthritic changes. We are not aware of any study that has detailed the results of triple arthrodesis after an average duration of follow-up of twenty-five years. This procedure continues to be used to manage young patients who live a fully active life and have a normal life span. Without long-term data, we cannot advise patients as to whether the well documented changes in symptoms, function, and arthritis will continue or will stabilize after the first few postoperative decades. The purpose of the study was to assess the outcome of this operative procedure over a longer period than has been previously reported. We performed the same two examinations, more than two decades apart, on a single group of patients. In each examination, we assessed the level of symptoms, function, arthritis, and satisfaction with the result of the operation.
Study Group The average age at the time of the triple arthrodesis in the study group was sixteen years (range, seven to fifty-seven years) (Table I). The average duration of follow-up in 1973 was twenty-five years (range, ten to forty-nine years), and the average duration in 1994 was forty-four years (range, twenty-nine to sixty-eight years). The average age at the second follow-up evaluation was sixty years (range, forty-one to eighty-eight years). A triple arthrodesis was performed unilaterally in forty-seven patients and bilaterally in ten patients. In four of the patients who had a bilateral condition, both feet were operated on at the same time. In three patients, there was a one-year interval between operations; in one, a two-year interval; in one, a nine-year interval; and in one, a fourteen-year interval.
The etiology of the problems with the foot included poliomyelitis in thirty-four patients (thirty-seven feet; 55 percent), clubfoot in five (seven feet; 10 percent), Charcot-Marie-Tooth disease in five (six feet; 9 percent), tarsal coalition in four (six feet; 9 percent), spinal cord abnormalities in three (four feet; 6 percent), trauma in three (three feet; 4 percent), cerebral palsy in two (three feet; 4 percent), and Guillain-Barré syndrome in one (one foot; 1 percent). The preoperative deformities included equinocavovarus in twenty-eight (42 percent) of the sixty-seven feet, planovalgus in twelve (18 percent), equinovalgus in seven (10 percent), flail foot in seven (10 percent), calcaneovalgus in five (7 percent), hindfoot varus in four (6 percent), cavovarus in two (3 percent), and hindfoot valgus in two (3 percent). Twenty-one of the sixty-seven feet had had procedures performed before the triple arthrodesis. These procedures included tendon transfers in thirteen feet, Achilles-tendon lengthening in seven, a Steindler operation (plantar stripping) in three, midtarsal wedge resection in two, and medial release in one. In thirty-five feet (52 percent), forty-seven other procedures were performed within eight weeks after the triple arthrodesis. These operations included twenty-eight tendon transfers, six Achilles-tendon lengthenings, six midfoot arthrodeses, four arthrodeses of the interphalangeal joint of the hallux, and three releases of the plantar fascia. In general, the tendon transfers were performed six weeks after the triple arthrodeses. At our institution, patients with neuromuscular disorders are not typically managed with a brace to maintain correction after removal of the cast.
Clinical Evaluation At both follow-up intervals, all of the patients completed a detailed questionnaire. Data included the presence or absence of pain, use of analgesic medication, need for and type of walking supports, maximum walking distance, and additional operative procedures on the lower extremity. The physical examination consisted of goniometric measurement of ankle and foot motion in the sagittal plane, assessment of residual deformity while the patient was standing, and identification of plantar callosities.
Radiographic Evaluation
Overall Assessment
Statistical Analysis
Postoperative Complications
Results of the Questionnaire
Pain in the other major joints of the lower extremity was similarly progressive. At the time of the first follow-up, three patients reported pain in the contralateral knee; two, pain in the ipsilateral hip; one, pain in the ipsilateral knee; and one, pain in the contralateral hip. At the time of the second follow-up, ten patients reported pain in the ipsilateral hip; ten, pain in the contralateral knee; nine, pain in the ipsilateral knee; and four, pain in the contralateral hip. At the time of the first follow-up seven patients (12 percent) used analgesic pain medications, and at the time of the second follow-up sixteen patients (28 percent) used such medications. With the numbers available, we detected no association between the underlying diagnosis, whether neurological (in forty-five patients) or structural (in twelve patients), and the level of pain reported at either the first or the second follow-up evaluation. Thirteen feet had a pseudarthrosis, and fifty-four did not. Of the feet with a pseudarthrosis, ten were painful at the first follow-up evaluation and twelve were painful at the second follow-up evaluation. Of the fifty-four feet without a pseudarthrosis, twenty-five (46 percent) were painful or were associated with pain at the first follow-up evaluation and forty-four (81 percent) were painful or were associated with pain at the second follow-up evaluation. In the group of feet without a pseudarthrosis, there was a significantly greater prevalence of pain at the second follow-up evaluation than at the first follow-up evaluation (p = 0.0002). Eighteen patients (32 percent) needed walking support at the time of the first follow-up, and thirty-nine patients (68 percent) needed support at the time of the second follow-up. At the second follow-up evaluation, fourteen patients (25 percent) used a brace, fifteen (26 percent) used an in-shoe orthosis, thirteen (23 percent) used a crutch or a cane, eight (14 percent) used an external shoe modification, and three (5 percent) used a walker. At the first follow-up evaluation fifty-two patients (91 percent) stated that they could walk more than one mile (1.6 kilometers), whereas at the second follow-up evaluation only twenty-two patients (39 percent) reported that they could walk more than a mile. At the time of the first follow-up, no patients reported a maximum walking distance of less than three blocks. At the time of the second follow-up, thirteen patients (23 percent) reported a maximum walking distance of less than three blocks. We did not specifically diagnose so-called post-poliomyelitis syndrome in these patients, although gradual weakness was clearly the trend. The factor that limited walking by the thirteen patients who could not walk three blocks was generally pain, not weakness. Other, subsequent operations on the ipsilateral lower extremity were performed in eighteen patients (32 percent). These operations included a toe procedure in seven patients, tendon transfers in four, a knee procedure in four, repeat triple arthrodesis in three (for the treatment of pseudarthroses and uncontrollable deformity), dorsal wedge resection in one, a total hip replacement in one, and removal of a heel spur in one.
Findings on Physical Examination At the second follow-up evaluation, fifteen (22 percent) of the feet had no residual deformity. The remaining fifty-two feet (78 percent) had some residual deformity: nineteen (28 percent) had a mean varus deformity of 7 degrees, thirteen (19 percent) had a mean valgus deformity of 8 degrees, eleven (16 percent) had residual supination, eleven (16 percent) had a cavus deformity, seven (10 percent) had an equinus deformity, four (6 percent) had forefoot adduction, two (3 percent) had residual pronation, and one (1 percent) had a calcaneal deformity. These deformities appeared to be nonprogressive between 1973 and 1994. Twenty-eight feet had a tendon transfer within eight weeks after the triple arthrodesis, and thirty-nine feet did not. Of the twenty-eight feet that had a tendon transfer within eight weeks, twenty (71 percent) had a residual deformity. Of the thirty-nine feet that did not have a tendon transfer within eight weeks, thirty-two (82 percent) had a residual deformity (p = 0.378, Fisher's exact test). Of the fifty-two feet that had a residual deformity, four (8 percent) had a tendon transfer more than eight weeks after the triple arthrodesis. Of the fifteen feet that had no residual deformity, one had a tendon transfer more than eight weeks after the triple arthrodesis (p = 1.0, Fisher's exact test). A callosity or callosities were seen on sixteen feet (24 percent) at the first follow-up evaluation and twenty-eight feet (42 percent) at the second follow-up evaluation. At the time of the first follow-up, twelve feet had one callosity and four had more than one. At the time of the second follow-up, sixteen feet (57 percent) had one callosity and twelve (43 percent) had more than one.
Radiographic Findings
A complete radiographic evaluation of the foot and ankle was performed at both follow-up intervals for all patients. At the time of the first follow-up, there were no degenerative changes in twenty-one ankles (31 percent), mild changes in forty-three (64 percent), moderate changes in two (3 percent), and severe changes in one (1 percent). At the time of the second follow-up, all of the ankles had degenerative changes. Thirty-three ankles (49 percent) had mild changes, twenty-eight (42 percent) had moderate changes, and six (9 percent) had severe changes. At the first follow-up evaluation, there were no degenerative changes in the naviculocuneiform joint in twelve feet (18 percent), mild changes in forty-seven (70 percent), moderate changes in three (4 percent), and severe changes in five (7 percent). At the second follow-up evaluation, degenerative changes in the naviculocuneiform joint were rated as none in one foot (1 percent), mild in thirty-nine (58 percent), moderate in twenty-one (31 percent), and severe in six (9 percent). At the first follow-up evaluation, no degenerative changes were found in the tarsometatarsal joint in nineteen feet (28 percent); mild changes, in forty-two (63 percent); and moderate changes, in six (9 percent). At the second follow-up evaluation, one foot (1 percent) had no degenerative changes in the tarsometatarsal joint; thirty-five (52 percent), mild changes; and thirty-one (46 percent), moderate changes (Figs. 1-A, 1-B, 1-C and 1-D).
We evaluated the possible association between the age at the time of the operation (thirteen years or younger [thirty-five feet] or fourteen years or older [thirty-two feet]) and the eventual development of arthritis. With the numbers available, no significant difference could be detected between the two age-groups with regard to radiographically apparent degenerative changes on either the first or the second follow-up radiographs.
Overall Results Surprisingly, the need for walking aids, the neurological status, and the range of motion of the ankle or foot did not seem to be associated with the overall outcome at either follow-up interval. Likewise, the amount of narrowing radiographically at the ankle, naviculocuneiform, or tarsometatarsal joint did not seem to be associated either with the patient's perception of pain or the degree of deformity. At the first follow-up evaluation, the presence of a nonunion was associated with pain (risk ratio, 1.66; 95 percent confidence interval, 1.10 to 2.51). However, by the second follow-up evaluation, pain was equally prevalent in patients who had a nonunion and those who did not. At the second follow-up evaluation, the patients were asked whether they were satisfied with the result of the operation and whether they would recommend the operation to others who had a similar foot problem. Fifty-four patients (95 percent) were satisfied with the result. Fifty-two patients (91 percent) stated that they would recommend triple arthrodesis to patients who had the same foot problem, whereas five patients (9 percent) stated that they would not.
The strength of the present study lies in the long-term nature of the follow-up. We serially evaluated a single group of patients who had had a standardized operative procedure. The condition of these patients was carefully observed and recorded at each of the follow-up periods, which allowed us to determine the change over time. The relatively long average follow-up intervals of twenty-five and forty-four years impart additional weight to the findings of this study. The weaknesses of the study are due to the design. Most of these patients had a neuromuscular problem that resulted in the need for an operation on the hindfoot. These various problems (poliomyelitis, hereditary sensory motor neuropathies, cerebral palsy, abnormalities of the spinal cord, and Guillain-Barré syndrome6) have a profound and continuing effect on the gait, joint-loading history, and walking ability of these patients and can contribute to the development of arthritis. Many of these patients also had other operative procedures on the ipsilateral lower extremity, and that may have further confounded the analysis. We were unable to detect any differences in the presence of residual deformity between the thirty-three feet that had a subsequent tendon transfer and the thirty-four feet that did not. There are many potential causes of residual deformity, and probably more than one factor caused the residual deformity in most patients. In this retrospective review, we were unable to delineate the specific influence of untreated motor imbalance on the residual position of the foot from other factors such as progressive weakness, development of degenerative joint-space disease, and intraoperative or perioperative malpositioning. Moreover, the twenty-eight feet that had a tendon transfer within eight weeks after the triple arthrodesis were, in general, more deformed before the triple arthrodesis; thus, they were more likely to have a residual deformity because of their preoperative condition. Another substantial concern with the present study is the lack of a control group. Without a nonoperatively treated group, we do not know the extent to which our findings are independent of the operative procedure. In particular, we do not know how much of the change in symptoms, mobility, or arthritis is due to the underlying disease process, the index procedure, or the passage of time and the effects of aging. Despite these clear limitations, this follow-up study does provide useful information on the long-term results of a commonly performed orthopaedic procedure. To our knowledge, it is the longest follow-up study of triple arthrodesis. It shows that the results, in terms of pain, function, and adjacent joint arthritis, deteriorate between an average of twenty-five and an average of forty-four years postoperatively. The general decline appears to be independent of the age at the time of the operation, use of walking aids, neurological status, or range of motion of the ankle or foot. Because of changes in disease prevalence, the findings of the present study have a somewhat diminished relevance in the present clinical environment. When patients in this study had the operation, the most prevalent indication was uncontrollable deformity secondary to poliomyelitis. With the advent of an oral vaccine, poliomyelitis has nearly disappeared in developed nations. In the United States today, the disease processes necessitating triple arthrodeses are vastly different from those in the 1940s. At our institution, the current major indications for a triple arthrodesis are degenerative arthritis of the hindfoot, uncontrolled deformity due to rheumatoid arthritis or severe acquired adult flatfoot deformity, and poorly treated or untreated clubfoot. Although the operative goals are the same, the recommended technique has changed over the last half-century. Many textbooks now advocate the use of two incisions to gain better access to the talonavicular joint, especially in feet with major, rigid deformities. Currently, the joint surfaces are often held together rigidly with internal fixation devices to promote healing and to maintain correction of the deformity, particularly in adults. The findings of the present study must be considered in light of the changes in disease prevalence and operative technique. For example, in this study we found that a major factor related to pain in the first decades after the operation was the occurrence of pseudarthrosis. Changes in the standard operative technique have been directed at reducing the occurrence of this problem. With these modifications in operative technique, the dramatic changes in disease prevalence, and the associated increasing average age of patients needing triple arthrodesis, the long-term results of triple arthrodeses performed in North America today likely will be different from those reported in the present study. Despite these clear changes, the results of the present study suggest that the outcomes of triple arthrodesis deteriorate over a long period of time. When counseling young patients and their families regarding the potential long-term effects of triple arthrodesis, the physician should explain that, if the patient lives long enough, he or she should expect to have pain with walking, to require the use of walking aids, and to have limited functional mobility. However, they should also be told that almost all of the patients who have had this procedure have been satisfied with the improved function that it gave them during the major portion of their lives and that most would recommend it to others.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |