End-to-End Operative Repair of Achilles Tendon Rupture*

Transcription

1 End-to-End Operative Repair of Achilles Tendon Rupture* Jeffery J. Soldatis, MAJ, MC, USA, Donald B. Goodfellow, MD, and John H. Wilber, MD From the Orthopaedic Surgery Service, Brooke Army Medical Center, Ft. Sam Houston, Texas, and the Department of Orthopaedic Surgery, University Hospitals of Cleveland, Case Western Reserve University, Cleveland, Ohio ABSTRACT We present the long-term results of operative repair in 23 consecutive patients with Achilles tendon ruptures, treated between 1984 and 1991, to evaluate our treatment method and determine the clinical causes of rupture. Fifty-four percent of ruptures occurred in people in their 30s; 90% occurred during participation in acceleration-deceleration sports. All but three patients were treated within 1 week of injury with open, operative, end-to-end repair of the Achilles tendon. The remaining three patients were treated more than 3 weeks after injury. All patients followed a standard postoperative regimen. Followup averaged 3.6 years (range, 1 to 7.5). Seventeen patients were available for Cybex analysis, and the remaining patients were interviewed personally or by telephone. Subjectively, patients were very satisfied with the results of treatment. Objectively. physical examination and Cybex testing to measure strength and endurance revealed results somewhat better than those previously reported with operative repair. No patient experienced a rerupture, although one attenuated repair was noted 9 months postoperatively. Only two minor wound problems were recorded. Long-term results revealed near-normal function when comparing the injured side with the uninjured side. Ninety-two percent of patients returned fully to their preinjury levels of activity. Rupture of the Achilles tendon is a relatively infrequent injury, although recent occurrences in several professional * Presented at the 21st annual meeting of the AOSSM, Toronto, Ontano, Canada, July Preliminary results presented at the 43rd Annual 38th Parallel Medical Society Meeting, Seoul, Korea, November t Address correspondence and repnnt requests to Jeffery J Soldatis, MD, 11 Grants Lake Dnve, The Hollow at Inwood, San Antonio, TX The views expressed in this paper are those of the authors and do not reflect the official policy or position of the Department of the Army, Department of Defense, or the United States Government No author or related institution has received any financial benefit from research in this study. athletes have brought its existence to the attention of the national press. These ruptures usually occur in men between 30 and 50 years of age, most often without prodromal symptoms. 1, 11 There is an apparent predilection for the weekend athlete engaging in acceleration-deceleration sports to experience this injury. However, the exact causes of injury in many cases remain unclear.4,8,11 Controversy also exists regarding treatment. Early publications supported operative repair,5,14,16 but a few, more recent, studies suggest that operative and nonoperative treatments produce similar results, with a lower complication rate associated with the latter. 7,11 Nistor 15 reported a prospective study in 1981 that concluded operative repair provided no significant advantage over nonoperative care. Operative treatment can result in complications, ranging from infection, wound breakdown, and sensory loss, to rerupture, 2,4,13,15 but nonoperative treatment is repeatedly cited as resulting in a u much ~.~~v~. higher ~~~bmu rerupture rate. i vm.alrvum mw. Nonoperative treatment also appears to result in diminished functional levels compared with operative repair. This has been especially noted in the high-demand athlete.9, Inglis et a1.8 reported less strength and endurance in their patients after nonoperative treatment compared with after operative repair. Additionally, Cetti et a1.3 concluded that operative repair results in a greater resumption of athletic performance at preinjury levels than does nonoperative treatment. We report common clinical causes of Achilles tendon rupture and preoperative findings on physical examination. We also report the long-term results of a simple, open, end-to-end operative repair and standardized postoperative regimen. This report is in contrast to most previous studies on operative repair in which several types of repairs were grouped together in reporting outcomes.l ~ 9 14> 17 MATERIALS AND METHODS Between 1984 and 1991, 30 patients with 30 ruptures of the Achilles tendon were treated with operative repair by 90

2 91 one of the two senior authors (JHW or DBG) at the University Hospitals of Cleveland. Most repairs were performed within 1 week of injury, leaving five patients treated more than 1 week after injury, but none more than 4 weeks after injury. The same operative technique, described later, was used in all cases. For purposes of demographic analysis only, all patients on whom charts and operative notes were available were considered (N 30). = The 30 patients for demographic analysis included 28 men and 2 women. The histories and physical examination results from admission, as well as the operative notes, were thoroughly reviewed to determine demographic patterns, mechanisms of injury, physical examination findings, and findings at the time of surgery. Charts were reviewed to determine 1) symptoms at the time of injury; 2) presence or absence of prodromal symptoms; 3) activity at the time of injury (acceleration or deceleration); 4) whether the patient could continue the activity; and 5) frequency of the activity. Diagnosis of Achilles tendon rupture was based on the history, especially the inability to continue the activity because of pain or weakness or both. The results of the Thompson test,18 the presence or absence of a palpable defect, and the ability to actively plantar flex the ankle were recorded from the chart for all patients. Partial rupture was determined by a negative Thompson test and absence ot a palpable defect but less than 4/5 manual plantar flexion strength in the presence of diminished function, as noted previously, and palpable tenderness and swelling over the Achilles tendon itself. The finding of some intact Achilles tendon fibers (but less than 50%), as noted in the operative note, confirmed the &dquo;partial&dquo; nature of the rupture. These patients were selected for operative treatment based on their preoperative symptoms and plantar flexion weakness, as described. Additionally, histories of gout, rheumatoid arthritis, previous steroid injection, or previous rupture were noted. Twenty-three of the 30 patients were available for longterm follow-up evaluation 1 year or more after surgery. Five patients were interviewed either personally or by telephone using a standard patient information form, and 18 patients underwent personal interview and physical examination of specific parameters. Seventeen of these 18 patients underwent specific Cybex dynamometer testing (Lumex Inc., Ronkonkoma, New York) by either of two athletic trainers. Patients for long-term followup were interviewed regarding the following: 1) their perceived percentage return to preinjury activity level; 2) whether all preinjury activities had been resumed and, if not, why; 3) whether they experienced any pain or weakness; and 4) whether they had suffered any long-term complications (e.g., rerupture, infection). Physical examination consisted of measuring calf circumference, palpable tenderness at the repair site, tendon width 5 cm proximal to the Achilles tendon insertion, dorsiflexion and plantar flexion range of motion and objective strength, and the ability to perform 20 single-foot toe raises on the involved and uninvolved sides. This is a standard evaluation technique used by other investigators.l 9-11 Cybex testing, using a Cybex II dynamometer, was used to evaluate plantar flexion strength and endurance objectively. The uninvolved side was used as a control. The patient was positioned sitting with his or her leg extended fully at the knee, with the foot, in a shoe, secured to the footplate. Strength was measured using a preset speed of 30 deg/sec to determine the maximal torque of five repetitions of maximal effort. Endurance was measured by the performance of 25 cycles at 180 deg/sec, noting the parameters of peak torque, total output (in watts), total work (in foot-pounds), and 50% failure point. These measurements were then compared with those of the uninvolved side as well as with measurements found in previous studies.17 Operative Technique All patients were treated by one of the two senior authors (DBG or JHW). A simple operative technique (described later) was used in all cases. All patients received perioperative prophylactic antibiotics (a 1-g dose of a first generation cephalosporin preoperatively and an additional two to three 1-g doses during the first 24 hours postoperatively). For the 23 patients available for longterm followup, a medial approach was used in 19 cases and a lateral approach was used in 4. All of the lateral approaches were performed in the early years of the period covered. A change to the medial approach evolved because of its ease of use and the avoidance of the sural nerve, which is encountered in the lateral approach. A direct end-to-end repair using either a Bunnell s or modified Kessler suture of No. 0 nonabsorbable material was used. Four patients had severe shredding of the tendon ends. In these cases, a whip-type suture was used to create three bundles that were then sutured together with the free ends, with either Bunnell s or modified Kessler sutures. A single bundle was fashioned distally with two opposing bundles fashioned proximally or vice versa, depending on which end of the rupture appeared more frayed. The surgeon fashioned two bundles from the end that appeared more frayed (Fig. 1). The plantaris tendon was harvested when available (15 repairs) and used as a simple suture onlay to bridge the repair site. No additional fascial grafts were used or required. The tendon end surfaces were then further coapted with interrupted absorbable No. 0 sutures. The tendon sheath or inter- was repaired in all cases with either running rupted No. 0 braided absorbable suture. A subcutaneous closure with absorbable synthetic suture was used in all cases. The repair was performed to allow a relaxed equinus positioning of the foot when complete. A sterile dressing and below-knee plaster splint in a relaxed equinus position were then applied. All procedures were performed using a tourniquet, with an average time of 61 minutes (range, 44 to 100). The postoperative regimen was standardized within the confines of scheduling constraints and individual patients schedules. All patients legs were immobilized in a belowknee nonweightbearing short leg cast in relaxed equinus

3 92 within 4 to 6 months postoperatively. Preinjury level was determined by measuring the uninvolved side s ankle range of motion (dorsiflexion and plantar flexion), manual muscle testing, and proprioceptive ability with a BAPS Board (Biomechanic Ankle Platform System, Camp-Spectrum Therapy Products, Jackson, Michigan). Return to sports was allowed once this level was achieved. RESULTS Figure 1. Whip suture technique of running &dquo;baseball-type&dquo; nonabsorbable suture to create three separate repair bundles. before they were discharged from the hospital (3 to 4 days after surgery). The average use of the cast was 4.2 weeks (range, 4 to 8), after which the patients used a short leg walking cast in neutral position for an additional 4 weeks (average, 4 weeks; range, 2 to 6). Thus, the total mean casting time was 8.2 weeks (range, 6 to 12). A heel lift was used only in those patients lacking at least 10 of dorsiflexion after removal of the second cast (N 5). Three = patients, the two professional athletes and a physician, were given CAM walkers 4 weeks after surgery to allow greater mobility. The device was locked at neutral ankle flexion for the first 4 weeks. Ankle plantar and dorsiflexion were gradually increased thereafter. A physical therapy program, under the direction of a licensed therapist, was initiated in all patients once the final cast was removed. The program initially consisted of range of motion and stretching exercises, with gradual addition of progressive resistance exercises once ankle motion was nearly the same as the uninvolved side. The therapist set goals for each patient for improving dorsiflexion, plantar flexion, and ankle eversion and inversion motion and strength to within 80% of the preinjury level The preoperative examination of the original 30 patients revealed the following. The average age for the men was 38 years (range, 25 to 72); the women were 32 and 42 years old. The highest incidence of rupture occurred in patients in their 30s (54%) (Fig. 2). Subjectively, only 6 (20%) patients experienced any Achilles tendon symptoms before rupture. For the remaining patients, the tendon ruptures were not preceded by any heel cord symptoms. Twenty-four patients described feeling a painful &dquo;snap&dquo; or &dquo;explosion&dquo; in the heel region during athletic activities. Twelve patients described the occurrence with forceful plantar flexion from a dorsiflexed ankle position (acceleration), and 12 with dorsiflexion with eccentric contraction of the calf muscles (deceleration). None of our patients were able to continue play after injury, except for one patient who was later diagnosed as having a partial rupture. Acceleration-deceleration sports predominated the list of activities associated with rupture (Table 1). Basketball was the most common sport (30%), followed by racquet sports (tennis and racquetball; 16%) and volleyball and soccer (15% each). Fourteen patients (47%) were frequent participants (2 to 3 times/week) in their respective sports, 9 (28%) were occasional participants (2 to 3 times/month), and 4 patients participated infrequently (less than 1 time/ month). Two patients were professional athletes injured...1. ;1... ~ ~.+;.,:... +;... ;.., +1.,..; z..,..,..+: ~...,.+,. -KT *--4- i while participating in their respective sports. 1V V patient had a history of gout, rheumatoid arthritis, previous steroid injection, or previous rupture. On physical examination, the Thompson test was posi- Figure 2. Number of patients in each age group.

4 93 TABLE 1 Activities Leading to Injury tive in all patients except for three, those who had experienced preinjury pain; in two of these patients partial ruptures were diagnosed. A palpable defect was noted in all but five patients, which again included the two with partial ruptures. Only eight patients had any discernible active plantar flexion. At surgery, rupture of the tendon was noted in the middle three-fifths of the tendon body in 25 of 30 patients. Twenty-three patients were available for long-term followup at an average of 3.5 years (range, 1 to 7.5). These patients were an average age of 35.5 years (range, 25 to 70). These patients were grouped together for analysis because all but three patients underwent early repairs (within 1 week of injury). The subgroup operated on later than 1 week after injury was too small to allow statistically significant comparison with those patients with early repairs, and the results among these patients were not dissimilar from the others. The results of the subjective questionnaire were as follows. Patients believed they had attained an average of 93% return to preinjury levels of function. Fourteen of the 23 patients felt that they were at 100% functioning. All but four patients had returned to all preinjury activities. None of these four patients had returned to the activity associated with the injury. This group included a patient who was considered to have an attenuated repair. Ten patients had no symptoms of pain on the injured side; 11 seldom experienced minor symptoms, which were not activity limiting; and 2 patients had occasional discomfort, again including the patient with the attenuated repair. Seven patients perceived some degree of weakness on the injured side and the rest perceived none. No patients complained of rerupture, serious infection, or severe pain. Physical examination of 18 patients revealed an average calf circumference difference of -1 cm (range, + 1 to -2.9 cm) for the involved side compared with the uninvolved side (measurement made 15 cm distal to the medial joint line). Three patients had mild palpable tenderness at the repair site (including the patient with the attenuated repair). All patients had palpable thickening at the repair site, which measured an average of 6.1 mm wider (range, 2 to 10 mm) than the control side (measurements were made using calipers at 5 cm proximal to the calcaneal insertion site). All but 2 patients were easily able to perform 20 single-foot toe raises on the involved side; the 2 patients who could not included the patient with the attenuated repair, who performed none, and a patient who performed only 12 raises on the uninvolved side. Active dorsiflexion averaged 2 less (range, -7 to +5 ) and plantar flexion averaged 2.6 less (range, -10 to +5 ) than the uninvolved side. Objectively, all patients demonstrated 5/5 dorsiflexion strength and averaged 5/5 plantar flexion strength for the involved side; all patients had 5/5 for both parameters on the uninvolved side. Cybex testing was performed on 17 of the 18 patients examined. Thirteen of these patients had the repair augmented with the plantaris as described in the operative technique. Comparing the involved side with the uninvolved side as a control, this group averaged 92% return of strength (range, 73% to 115%) as achievable peak torque in five repetitions at 30 deg/sec. Endurance testing of 25 repetitions at 180 deg/sec revealed average peak torque of 92% (range, 73% to 112%), average total output of 91% (range, 54% to 125%), and average total work of 81% (range, 57% to 120%). Additionally, the 50% failure point (number of repetitions at which peak torque was only 50% of the highest previously measured peak torque) was calculated at an average of 79% (range, 56% to 100%), again compared with the uninvolved side. No patient experienced difficulty or discomfort during the testing. Of the 23 patients, only two developed delayed healing of a superficial wound, which resolved within 3 months postoperatively. There were no deep or superficial infections. One patient had an attenuated repair noted at 9 months after surgery. This patient, a 68-year-old man, could not tolerate his cast postoperatively and required four cast changes during the initial nonweightbearing period, for an overall casting period of only 6 weeks. He had a decrease in plantar flexion strength from 4+/5 at 6 months postoperatively to 4-/5 at 9 months, a mild defect distal to the repair site, and 10 less active plantar flexion compared with the uninvolved side. No additional studies were performed. As noted above, he was unable to perform any toe raises on the repaired side, had occasional pain that limited activity, and did not undergo Cybex dynamometer testing. DISCUSSION Spontaneous rupture of the Achilles tendon remains an incompletely understood phenomenon. The exact cause of Achilles tendon rupture remains controversial, although the theory proposed by Inglis et a1.,8 which implicates a malfunction of the normal inhibitory mechanism of the musculotendinous unit, has the most merit. They hypothesized that this mechanism may malfunction in the fatigued or unconditioned athlete. They noted no degenerative changes of the tendon in their operatively treated patients but rather acute hemorrhage and inflammation, suggesting muscular override of the normal tendon. In contrast, the findings of J6zsa et al.11 on 292 ruptures of the Achilles tendon revealed histologic degenerative processes in more than half of those patients operated on within 48 hours of injury. Most of these patients were white-collar workers participating in recreational sports activities. These findings suggest a &dquo;silent&dquo; degenerative

5 94 process, possibly from recurrent microtrauma, in the recreational athlete that may weaken the tendon. The low number of reported ruptures in well-conditioned, professional athletes lends support to both of these theories, which in fact may be interrelated. Additionally, the tendon usually fails in midsubstance rather than at the musculotendinous junction or the insertion onto the calcaneus. More than 75% of tendon ruptures in the patient group reported by Jozsa et al. were noted in the middle three-fifths of the tendon, a finding we also noted at the time of surgery in our patient group. Unfortunately, none of the tendons in our patient group were examined histologically to note the presence or absence of degenerative changes. Several predisposing factors for rupture of the Achilles tendon have been identified. The injury occurs predominantly in middle-aged men. In our study, 54% of the ruptures occurred in people in their 30s, and 28 of our 30 patients were men. These findings are similar to those described previously.1,6,15 When prodromal symptoms were present before rupture, Kannus and Jozsa 12 attributed rupture to progressive degenerative changes within the tendon substance itself. However, only 20% of our initial group experienced such symptoms, leaving the majority of patients without warning, as others have noted. 1,6,8, 11 This finding further implicates a mechanism other than a degenerative process in the cause of tendon failure in those cases without prodromal symptoms. Acceleration-deceleration sports have also been implicated in this injury, and more than 90% of our patients experienced ruptures during participation in such sports: 30% during basketball and 16% during racquet sports. Forty-seven percent and 28% were frequent or occasional participants, respectively, in the activity responsible for their injury. Almost all of our patients described a &dquo;snap&dquo; in their heel, with the inability to continue play. The number of our patients describing their symptoms after an acceleration or a deceleration maneuver were the same. This is in contrast with the study by Beskin et al. ; their patients all reported acceleration events. It appears that either acceleration or deceleration can result in an injury if the conditions are appropriately severe. None of our patients demonstrated any other predisposing factors, although gout and steroid injections have been implicated by previous authors. 1,2,8,12 We found the presence of a positive Thompson test and, slightly less so, the presence of a palpable defect to be highly reliable indicators for the presence of complete Achilles tendon rupture. Three patients were treated with late repairs (more than 1 week after injury), and two of these, who themselves came to their primary care physicians late, were considered to have partial tears and had negative Thompson tests. Because of the small size of this group, we cannot accurately comment on the effect of a delayed or missed diagnosis. Inglis et al.,8 however, estimated that as many as 20% to 25% of Achilles tendon ruptures may go undiagnosed by the primary treating physician, and both Beskin et al/ and they did report a trend for less consistent, although still acceptable, results in delayed operative treatment groups. Many different techniques for operative repair of Achilles tendon ruptures have been presented in the literature. We present our technique for comparison with these various operative treatments and the reported operative complications, sometimes as high as 25%.13 A direct comparison with nonoperative treatment is beyond the scope of this report. The hypothesis for our study was that a simple operative repair with a standardized postoperative regimen gave consistently good results with a low complication rate. No standard grading scale exists to directly compare subjective and objective results of previous studies. Poor or unsatisfactory results have not been explicitly delineated by previous authors, other than they noted the occurrence of rerupture, wound complications, and to what functional level the patient returned. Results of Cybex dynamometer analysis appear to be the most objective results by which to compare one study with another. The debate regarding operative versus nonoperative treatment remains unsettled. However, operative treatment generally appears to be indicated in the active older patient, the young athlete, and the elite athlete. 1-3, 8, 13 Although Nistor 15 noted only minor differences between operatively and nonoperatively treated patients, he did report 5 reruptures in his nonoperatively treated group of 60 patients (rerupture rate of 8.3%) versus only 2 in the operatively treated group of 44 patients (rate of 4.5%). The reviews by Kellam et a1.13 and Wills et al.,19 as well as the comparative work by Inglis et a1.,8 show a significantly reduced rate of rerupture with operative treatment when compared with nonoperative treatment. The two reviews, which compile the results of all studies on Achilles tendon rupture through the mid 1980s, reported an average rerupture rate among patients with nonoperative treatment of 17.7%, compared with a 1.7% rate among patients with various operative repairs. Inglis et al. reported rerupture rates of 30% in their nonoperative group (9 of 31 patients) and 0% in their operative group (0 of 48 patients). This higher reported rate of the complication of rerupture in the nonoperatively treated groups is of concern. Additionally, operative repair appears to result in better functional performance. Gillies and Chalmers and Jacobs et al.l noted statistically significant greater strength between operatively and nonoperatively treated ruptures on isometric testing. Using Cybex dynamometer testing, Inglis et al. reported average strength of 101% and average endurance of 91% in their operatively treated patients as a group, compared with 73% and 64%, respectively, in their nonoperatively treated patient group. This is consistent with the similar findings by Shields et al. 17 and is contrasted only by Nistor,15 who did not report a statistically significant difference between his operative and nonoperative groups on Cybex testing. However, Nistor s results are not reported as a percentage value compared with the uninjured control side, and this makes direct comparison to other reports difficult. No other studies that directly compare operative and nonoperative treatment using a Cybex dynamometer have been performed. The recent prospective study by Cetti et a1.3 reports patients treated operatively had a statistically significant higher resumption of preinjury function than nonopera-

6 95 tively treated patients. In a review of the literature, these authors also noted a higher percentage of athletes returning to the same athletic level after operative repair (62%) than after nonoperative treatment (52%). Optimal operative outcome demands good operative technique, respect for contraindications, appropriate patient selection, and early repair, as noted earlier. Our method is simple and direct, without complicated fascial grafts. Operative time averaged 1 hour. Repair can proceed either medially or laterally, although care must be taken to identify and avoid the sural nerve and lesser saphenous vein in the lateral approach. As Beskin et all also noted, avoidance of disrupting the anterior tendon sheath, the access for the tendon s blood supply, is crucial. If poor local skin conditions are present, open repair may be delayed, or percutaneous or nonoperative care may be considered, even though there is a higher reported rerupture incidence with these methods. Bradley and Tibone 2 reported two reruptures in 12 percutaneous repairs, and they advocate extended protection with the use of a dorsiflexion stop brace as well as a heel lift. Additionally, reports on other operative repairs have shown higher incidences of wound problems. The repairs in the series of Cetti et al.4 were performed through a straight posterior approach, and they reported an 18% rate of major or minor wound problems. Beskin et al/ reported 3 of 42 patients required oral antibiotics for superficial wound infections. fix7n Only two nf of nf our nati,-nfq patients ],ad _h_ had a minnr minor wniind nroblem; problem; there were no infections or major wound complications. Our patients were subjectively very satisfied with the results of their treatments. As a group, they returned to an average of 93% of preinjury level of activity, and all but four patients returned to all preinjury activities. Differences in measured physical parameters, including plantar flexion and dorsiflexion range of motion, between the operative and uninvolved sides were small and consistent with other reports on operative repair. 1-3 There were no reruptures. Only the patient with the attenuated repair had symptoms that limited his level of activity. This patient was considered to have achieved an unsatisfactory result based on his functional limitations. This patient did not tolerate the postoperative casting and may not have fared well with nonoperative treatment, cast immobilization, either. Cybex dynamometer testing results were very consistent with those reported after other operative treatment methods. Our average 92% return of strength compares well with the 80% to 82% reported by Bradley and Tibone,2 and the 82% average reported by Beskin et al. Endurance testing also revealed excellent results, on average 92% peak torque, again comparing well with the previously mentioned studies (90% and 74%, respectively). We consider the 50% failure point values as less accurate because there was a wide range of results (56% to 100% of the control side) and the values were obtained manually from the oscillator readout, which had notable variability. CONCLUSIONS Participation by the middle-aged, male athlete in acceleration-deceleration sports, regardless of the frequency of participation, predisposes one to Achilles tendon rupture. We advocate early, open, end-to-end operative repair of Achilles tendon rupture for all active patients and patients who cannot afford the higher risk of rerupture associated with conservative methods (nonoperative treatment as well as percutaneous repair). Our method produces predictable and consistent results with very low morbidity, near normal or normal functional outcome, and somewhat superior objective results compared with previous studies. We recommend nonoperative treatment only for the sedentary patient or patients with significant risk factors (e.g., diabetes, peripheral vascular disease, skin problems), in whom the risk of operative complication would be unacceptably high. ACKNOWLEDGMENTS We thank Gary Sutton, LPT, ATC, and Sharon Kelly, ATC, University Suburban Heath Center, South Euclid, Ohio, for their invaluable assistance and expertise in making this study possible. REFERENCES 1 Beskm J, Sanders RA, Hunter SC, et al Surgical repair of Achilles tendon ruptures Am J Sports Med , BradIQy JP, Tibnna JF- Percutaneous and open surgical repairs nf Achilles tendon ruptures A comparative study. Am J Sports Med , Cetti R, Christensen SE, Ejsted R, et al: Operative versus nonoperative treatment of Achilles tendon rupture. A prospective randomized study and a review of the literature. Am J Sports Med , Cetti R, Christensen SE, Reuther K Ruptured Achilles tendons treated surgically under local anesthesia Acta Orthop Scand , Christensen I Rupture of the Achilles tendon Analysis of 57 cases Acta Chir Scand , Fitzgibbons RE, Hefferon J, Hill J Percutaneous Achilles tendon repair Am J Sports Med 21: , Gillies H, Chalmers JJ. The management of fresh ruptures of the tendo Achilles. J Bone Joint Surg 52A , Inglis AE, Scott WN, Sculco TR, et al Ruptures of the tendo Achillis: An objective assessment of surgical and non-surgical treatment J Bone Joint Surg 58A , Inglis AE, Sculco TR- Surgical repair of ruptures of the tendo Achilles. Clin Orthop 156: , Jacobs D, Martens M, Van Audekerke R, et al: Companson of conservative and operative treatment of Achilles tendon rupture Am J Sports Med 6: , Jozsa L, Kvist M, Bálint BJ, et al. The role of recreational sports in Achilles tendon rupture. A clinical, pathoanatomical, and sociological study of 292 cases Am J Sports Med 17: , Kannus P, Józsa L Histopathological changes preceding spontaneous rupture of a tendon A controlled study of 891 patients J Bone Joint Surg 73A , Kellam JF, Hunter GA, McElwam JP, et al Review of the operative treatment of Achilles tendon rupture. Clin Orthop 201: 80-83, Lawrence GH, Cave EF, O Conner H Irijury to the Achilles tendon Experience at the Massachusetts General Hospital, Am J Surg 89: , Nistor L: Surgical and non-surgical treatment of Achilles tendon rupture A prospective randomized study. J Bone Jomt Surg 63A: , Quenu J, Stoianomtch Les ruptures du tendon d Achilles Rev Chir , Shields CL Jr, Kerlan RK, Jobe FW, et al The Cybex II evaluation of surgically repaired Achilles tendon ruptures. Am J Sports Med 6: , Thompson TC, Doherty DH: Spontaneous rupture of tendon of Achilles. A new clinical diagnostic test J Trauma 2: , Wills CA, Washburn S, Caiozzo V, et al. Achilles tendon rupture A review of the literature companng surgical versus nonsurgical treatment Clm Orthop , 1986