Acute Tears of the Rotator Cuff

Size: px

Start display at page:

Download "Acute Tears of the Rotator Cuff"

Jocelyn Christine Quinn
5 years ago
Views:

1 Acute Tears of the Rotator Cuff The Timing of Surgical Repair RICK W. BASSETT, M.D., AND ROBERT H. COFIELD, M.D. Thirty-seven patients had surgical repair within three months after significant ruptures of the shoulder rotator cuff. Twelve were repaired within three weeks (Group l), six within three to six weeks (Group 2), and 19 within six to 12 weeks after injury (Group 3). Follow-up periods averaged 7.0 years (range, OO years). Pain relief was generally satisfactory; however, active postoperative abduction averaged 168" for Group 1, 126" for Group 2, and only 129" for Group 3. At open operation, the tear size was graded as small, medium, or large. The correlation of average values of postoperative abduction (148", 152", and 133") with tear size was not statistically significant. Roentgenograms showed that rotator cuff disease is associated with glenohumeral arthritis. With evidence of an acute and complete disruption of the rotator cuff if one must consider surgery, early surgical repair (with 3 weeks of injury) affords the best opportunity for maximal recovery of shoulder function. Rupture of the rotator cuff mechanism in the shoulder is a common injury. The disruption occurs through an area of weakened tendon," the result of degenerative changes' that may have been augmented by subacromial impingement." Although most rotator cuff tears (perhaps 90%) are viewed as chronic disabilities, some patients have acute injuries that produce a significant change in active shoulder movement and strength. For recent injuries, many authors have advocated a substantial trial of conservative therapy, empha- From the Department of Orthopedics, Mayo Clinic and Mayo Foundation, Rochester, Minnesota. Reprint requests to R. W. Bassett, M.D., 70 Section of Publications, Mayo Clinic, Rochester, MN Received: May 3, I X/S3/0500/0 I8 $0 1. I5 0 J. B. Lippincott Co. 18 sizing that surgery should be a secondary option.5,9-l I McLaughlin9."' specifically reported equally satisfactory results in early and late cuff repair. However, with time, tears may enlarge, and shoulder instability and disuse may cause attritional changes of the tendons, making repair more difficult.' In view of these changes, others have stressed the importance of early surgical repair. In this study the authors sought to evaluate a series of patients with the uncommon acute tear of the rotator cuff and severely compromised shoulder function whose condition at surgery was known, and then to determine whether there is an optimal time for surgical repair of a rotator cuff tear after a significant acute injury. The study does not include conservatively treated patients with uncertain pathologic shoulder changes, nor does it analyze the role of nonsurgical treatment methods for this problem. MATERIALS From 1954 to 1978, 510 rotator cuff repairs were done at the Mayo Clinic. Only those patients with a significunt acute injury and a full-thickness rotator cuff tear who had surgical repair of their torn tendon(s) within three months of injury were included in the study. A three-month limit was selected because with a longer period to surgery details surrounding the injury often became vague. Also, by three months the lesion is approaching a chronic state, for which the results of treatment are well known. Only 43 patients (43 shoulder injuries) fulfilled these criteria, reflecting in part traditional conservative treatment. Four of these, however, died of causes unrelated to their shoulder

2 Number 175 May Acute Rotator Cuff Tears 19 disease before follow-up data were complete, and two patients could not be located for follow-up examination, leaving 37 patients (37 rotator cuff tears) to constitute the study group. All were followed up by examination years after surgery (average, 7 years): in only two patients was the follow-up period less than two years. Twentyfive patients were examined personally by the authors. The remaining 12 patients were examined by other orthopedic surgeons and interviewed by the authors. All patients completed a detailed questionnaire. Of the 37 patients, 34 were men and three were women. The average patient age was 56 years (range, years). Three patients were younger than 40 years of age, and four were older than 70 years. The ages of the remaining patients were distributed rather evenly among the fifth, sixth, and seventh decades of life. The right shoulder was injured in 19 (5 1%) patients. A fall on the shoulder or outstretched hand was the mechanism of injury in 22 (59%) of the patients. Other causes of injury included stressful lifting of heavy objects (6 patients), grabbing a rail to prevent falling (3 patients), and motor vehicle accident, pulling on a wrench, football, hockey, golf, and tennis (1 patient each). With pain categorized as none, slight, moderate, or severe, all patients had moderate or severe pain before operation. All had moderate or pronounced weakness in abduction. Active abduction averaged 33" (range. 0"-180"). Thirty-five of the 37 patients could not actively abduct more than 90". In 25 cases, arthrography was used to confirm the diagnosis. To evaluate the effects of early as compared with delayed surgery, the repairs were placed into one of three groups on the basis of time from injury to surgery. Twelve shoulders were operated on within three weeks of injury (Group I), 6 between three and six weeks (Group 2), and 19 between six and 12 weeks (Group 3). At operation, acute injury only or a significant acute tear superimposed on chronic tendon disease was noted in 36 shoulders, and in one shoulder, chronic disease alone was present. Rotator cuff tears were graded as small, moderate, or large. Small tears were confined to a supraspinatus tendon and were less than 3 cm in greatest diameter. Medium and large tears extended into the infraspinatus or subscapularis tendon, or both. Medium tears were from 3 to 5 crn in greatest diameter, and large tears were greater than 5 cm. The distribution of tear size for each of the three groups is documented in Table 1. The distribution of tear size is comparable among the three groups, TABLE 1. Sizes of Rotator Cuff Tears Size of Teur Small* Moderutr Large Group 1 (<3 wks to repair) Group 2 (repair at 3-6 wks) Group 3 (repair at 6-12 wks) * Small tears involved only the supraspinatus tendon and were <3 cm in longest diameter. Moderate and large tears extended into adjacent tendons. Moderate tears were 3-5 cm in diameter. and larger tears were >5 cm. with the exception of Group 2, which had slightly more of the larger tears. The rotator cuff tears were repaired by primary tendon suture in three of 12 patients in Group I, three of six in Group 2, and three of 19 in Group 3. The remaining patients required side-to-side tendon suture and then suturing of the remaining healthy tendon to exposed cancellous bone of the humeral tuberosities. In addition to rotator cuff repair, anterior acromioplasty" was done in 25 shoulders. and lateral acromioplasty was done in two. The distal clavicle was excised for associated acromioclavicular arthritis in six shoulders, and the biceps tendon was tenodesed to the upper humerus in two shoulders. operative support included the use of a Velpeau dressing, humeral abduction splint, or shoulder spica cast in seven of 12 shoulders in Group I, three of six shoulders in Group 2, and 11 of 19 shoulders in Group 3. The remainder were treated either with a sling or commercial shoulder immobilizer. The average duration of postoperative support was 3.3 weeks in Group I, 3.5 weeks in Group 2, and 3.2 weeks in Group 3. A uniform physical rehabilitation program was employed, beginning with pendulum and passive range of motion exercises; progressing to assisted range of motion exercises and isometrics for the nonsurgically violated muscles; and finally, active range of motion exercises, stretching, and muscle strengthening for all groups. Motion exercises were begun during the first two weeks for shoulders with slings, shoulder immobilizers, or abduction splints and after removal of the external support for shoulders with Velpeau dressings or casts. Physical rehabilitation was monitored by a physiotherapist for an average of 3. I, 3.0, and 2.9 weeks, respectively, for the three groups.

3 20 Bassett and Cofield RESULTS At follow-up examination, 13 patients stated that they had no pain and 24, that they had slight pain. No patient complained of moderate or severe pain. Patient satisfaction was rated as follows: much better = 4; better = 3; the same = 2; worse = 1. In Group 1, ten patients reported that they were much better and two that they were better. In Group 2, three were much better, one was better, and two were the same. In Group 3, eight were much better, nine were better, and two were the same. The average patient satisfaction value was 3.83 for Group 1, 3.17 for Group 2, and 3.32 for Group 3. Active abduction was measured with the patient sitting, with the arm in 45" of horizontal flexion (approximating the plane of the body of the scapula) rather than in the frontal plane. This position was chosen because the am is aligned with the long axis of the supraspinatus-the most commonly involved structure in rotator cuff tears-and movement most directly tests its competence. Flexion in the sagittal plane may overestimate the capabilities of upward movement of the arm. Additionally, true abduction in the frontal plane places the glenohumeral joint in extension due to the oblique position of the scapula on the chest wall. In this position, substantial rotational Clinical Onhopaedlcs and Related Research movement of the arm is necessary to affect upward, overhead movement.**" The relationships of postoperative active abduction to time of surgical repair and to tear size were analyzed. Change in preoperative to postoperative abduction for the three groups is illustrated in Figure 1. On preoperative examination, Group 1 patients averaged 28" of active shoulder abduction; Group 2,42"; and Group 3,34". At the most recent examination, the values for average active abduction of the shoulder were 168", 126", and 129", respectively (Table 2, Fig. 1). Improvement in active abduction was 140" for Group 1,84" for Group 2, and 95" for Group 3. Improvement was significantly better in Group 1 than in Group 2 or 3 (p < 0.05). In Figure 2, the change in active abduction as measured prior to and after operation is graphed according to rotator cuff tear size. Active postoperative abduction was 148" for those with small rotator cuff tears, 152" for those with medium rotator cuff tears, and 133" for those with large rotator cuff tears. There was no significant difference in final active abduction between the small, medium, and large tears. Documentation of the examiner's impression of the postoperative strength of abduction and external rotation was recorded at follow-up examination for 30 of 37 patients. Active abduction 0-3wk Pre 3-6wk Time to repair Pre 6-12wk FIG. 1. Changes in preoperative and postoperative active abduction for each shoulder within the three groups, based on the time from injury to surgical repair. Dashed lines indicate mean values for each group. Patients who had surgical repair within 3 weeks of injury had consistently better postoperative active abduction (p = 0.05).

4 Number 175 May, 1983 Acute Rotator Cuff Tears 21 TABLE 2. Active Abduction in Patients Undergoing Rotator Cuff Repair Preoperative operative Improvement (average) (average) (average) Group 1 (<3 wks to repair) 28" 168"* 140 * Group 2 (repair at 3-6 wks) 42" 126" 84" Group 3 (repair at 6-12 wks) 34" 129" 95" * Significantly different from respective values for Groups 2 and 3 (p < 0.05). The muscle strength of abduction and external rotation was graded as poor when the muscle could move the shoulder through its range of motion but not against gravity; fair when the muscle could move the arm through its range of motion against gravity but not against any resistance; good when the muscle could lift the arm against gravity and added resistance; and normal if equal to the uninjured side. All shoulders except one had a full passive range of motion on testing. Complete information on muscle strength was available for 1 1 of 12 shoulders in Group I, five of six shoulders in Group 2, and 14 of 19 shoulders in Group 3, as well as for seven of seven shoulders with small rotator cuff tears, nine of ten shoulders with medium rotator cuff tears, and 14 of 20 shoulders with large rotator cuff tears. To simplify analysis, shoulders with normal or good muscle function were grouped together, as were those with fair or poor muscle function in abduc- tion or external rotation (Table 3). Patients whose muscle tears were repaired within six weeks of injury had consistently better muscle strength in abduction and external rotation than those who had repair six weeks to three months after injury; however, the difference was not statistically significant. Similarly, shoulders with small and medium rotator cuff tears had consistently better abduction and external rotation strength after surgical repair than those with large rotator cuff tears. However, again, the difference was not statistically significant. An acromioplasty was done along with rotator cuff repair in 27 patients: 8 of 12 in Group 1 ; four of six in Group 2, and 15 of 19 in Group 3. Preoperative active abduction averaged 37" for these 27 patients and 22" for those who did not have partial acromial resection. operative abduction averaged 138" for those with the combined procedure and 150" for those with rotator cuff repair FIG. 2. Changes in preoperative and postoperative active abduction for each shoulder group according to the size of rotator cuff tear found at surgery. Dashed lines indicate mean values for each group. No significant difference was noted in final active abduction among patients with small, medium, or large tears when surgical repair was done within 3 months. Active abduction Pre Small Pre Medium Po91 Pre Large Tear size

5 22 Bassett and Cofield Clinical Orthopaedics and Related Research TABLE 3. Muscle Strength in Patients after Rotator Cuff Repair Patients with: Normal or Good Strength Fair or Poor Strength Time from injury to repair <3 wks (Group 1) Abduction 9 2 External rotation wks (Group 2) Abduction 4 I External rotation wks (Group 3) Abduction 8 6 External rotation 10 4 Tear size Small (<3 cm) Abduction 5 5 External rotation 6 1 Medium (3-5 cm) Abduction 7 2 External rotation 7 2 Large (>5 cm) Abduction 9 5 External rotation 10 4 alone. (This small difference may be explained by the relatively fewer combined procedures done in Group 1.) Preoperative roentgenograms were available for review in 24 cases and postoperative roentgenograms, in 27. Nonspecific changes associated with rotator cuff tendinitis included tuberosity sclerosis or rarefaction (9), partial tuberosity resorption (7), sclerosis of the acromion (8), reversal of the inferior acromial convexity (8), and upward subluxation of the humeral head, of varying amount but usually mild (14). Acromioclavicular arthritis was present in 11 shoulders, and mild glenohumeral arthritis was present in six. operative roentgenograms of 27 shoulders were available for review at follow-up examination. Both preoperative and postoperative roentgenograms were available in 21 shoulders. Of these, 13 had upward hu- meral subluxation at surgery. After operation, upward subluxation remained the same in six and increased slightly in seven; it never decreased. Mild glenohumeral arthritic changes were present before operation in four shoulders. At follow-up examination, mild glenohumeral arthritis was present in nine shoulders, moderate arthritis in two, and severe arthritic changes in one. DISCUSSION Experimental studies indicate that healthy tendons do not tear through their substance and that all tendon tears occur along with chronic changes of the tendinous tissue. With one exception, all shoulders in the present series had acute changes that definitely could be identified and that represented a significant part of the pathologic changes associated with the cuff injury. Often, however, associated changes occurred within the tendon adjacent to the tear, suggesting an acute process in addition to the chronic underlying disease. Indeed, the roentgenograms showing changes attributable to nonspecific tendinitis of the rotator cuff support the view that chronic disease existed within the tendon and that an acute injury was superimposed upon this. The present study indicates that pain relief was acceptable regardless of the time from injury to surgical repair, within the threemonth limit. However, return of function, as measured by active abduction and muscle strength, was better if the repair was done earlier. In particular, the return to active abduction was consistently greater for patients who were operated on within three weeks of injury. Soon after injury, retraction of the musculotendinous tissues and frictional attrition of the torn edges are minimal, and scarring about the shoulder is not present, rendering early repair easier and more secure. The good but not uniformly excellent results in patients of the present series who had their repairs done later than three to six weeks

Number 175 May. 1983 Acute Rotator Cuff Tears 23 after injury, and in those series in the literature that report results of chronic rotator cuff tears6,15.j6.

6 Number 175 May Acute Rotator Cuff Tears 23 after injury, and in those series in the literature that report results of chronic rotator cuff tears6,15.j6.18 may be due to the development of tissue scarring with inelasticity. Alternatively, the repair may be less secure, delaying the initiation of physiotherapy, or the rotator cuff may retear after surgical repair. Although there has been speculation that many patients who have acute rotator cuff tears will recover with nonsurgical treatment, unfortunately, no patient series in which the pathologic changes are known is available. Several series have reported the results of nonsurgical treatment of patients with rotator cuff tears, but generally, the groups are mixed, consisting of patients who have had either acute or chronic onset of shoulder symptoms. In the series of Samilson and Binder, 59% of the patients managed nonsurgically had good or excellent results. Only 33% of Wolfgang s patients (3 of 9) treated nonsurgically improved, compared with 74% treated surgically. Brown,3 reporting on a series of patients with clinically diagnosed tears of the supraspinatus tendon, classified symptoms as mild, moderate, or severe. Twentytwo patients presented with moderate symptoms. Thirteen (59%) recovered fully while seven (32%) recovered incompletely. Two patients evidenced no recovery. Of 34 patients with severe clinical symptoms, 18 (53%) recovered fully and 12 (35%) recovered incompletely; four patients evidenced no recovery. In neither of these series was arthrography always done to confirm the clinical diagnosis. While arthrography would have allowed the absolute diagnosis of a rotator cuff tear, it would not, with any degree of security, have defined the size of the tear or the quality of the tendon tissue. In the present study, all patients presented with severe clinical symptoms. The nonsurgical results are not nearly as satisfactory as those of surgical repair performed soon after injury (within 3 weeks). In the presence of an acute, significant rotator cuff tear with moderate and certainly severe compromise of function, a randomized study with surgical and nonsurgical groups may not be ju_stified. CONCLUSIONS AND RECOMMENDATIONS Surgical repair of rotator cuff tears done within three weeks of injury afforded the best surgical results. Tear size was not a significant variable in patients with acute injury repaired within three months. Rotator cuff disease is associated with glenohumeral arthritis. This study was not designed to reveal whether conservative treatment or surgical repair of acute rotator cuff tears produced better results. Nor was it intended to define the treatment for chronic rotator cuff disease. On the basis of this information, the authors make the following recommendations. In patients who place high demands on their shoulders and who have an acute injury with severe compromise of shoulder function (which may be a rotator cuff tear), the diagnosis should be made using an arthrogram, if necessary, and surgery should be done as soon as possible. In patients who place moderate demands on their shoulders, a ten-day to two-week trial of conservative measures may be indicated, but if improvement is not dramatic, a secure diagnosis should be obtained and surgery considered. In patients who place very low demands on their shoulders or who have severe general medical illness, the benefits of surgery may not match the risks. Hence, the above mentioned plan would be inappropriate unless pain relief could not be achieved otherwise. For an acute rotator cuff tear the shoulder may or may not improve with conservative treatment. Hence, for significant disability after an acute, complete disruption of the cuff, surgical repair must be considered within three weeks of injury to obtain maximal return of shoulder function.

7 24 Bassett and Cofield Clinical Orthopaedics and Related Research REFERENCES I. Bateman, J. E.: The diagnosis and treatment of ruptures of the rotator cuff. Surg. Clin. North Am. 43:1523, Brewer, B. J.: Aging ofthe rotator cuff. Am. J. Sports Med. 7:102, Brown, J. T.: Early assessment of supraspinatus tears: Procaine infiltration as a guide to treatment. J. Bone Joint Surg. 31B:423, Codman, E. A,: Rupture of the supraspinatus. Am. J. Surg. 42:603, DePalma, A. F.: Surgery ofthe Shoulder, ed. 2. Philadelphia, J. B. Lippincott, 1973, pp Godsil, R. D. Jr., and Linscheid, R. L.: Intratendinous defects of the rotator cuff. Clin. Orthop. 69:181, Heikel, H. V. A.: Rupture of the rotator cuff of the shoulder: Experiences of surgical treatment. Acta Orthop. Scand. 39:477, Johnston, T. B.: The movements of the shoulderjoint: A plea for the use of the plane of the scapula as the plane of reference for movements occumng at the humero-scapular joint. Br. J. Surg. 25:252, McLaughlin, H. L.: Rupture of the rotator cuff. J. Bone Joint Surg. 44A:979, McLaughlin, H. L.: Repair of major cuff ruptures. Surg. Clin. North Am. 43:1535, I. Moseley, H. F.: Shoulder Lesions, ed. 3. Baltimore, Williams & Wilkins, 1969, pp , Neer, C. S. 11: Anterior acromioplasty for the chronic impingement syndrome in the shoulder: A preliminary report. J. Bone Joint Surg. 54A:41, Poppen, N. K., and Walker, P. S.: Normal and abnormal motion of the shoulder. J. Bone Joint Surg. 58A:195, Rowe, C. R.: Ruptures of the rotator CUE Selection of cases for conservative treatment. Surg. Clin. North Am. 43:1531, Samilson, R. L., and Binder, W. F.: Symptomatic full thickness tears of the rotator CUR An analysis of 292 shoulders in 276 patients. Orthop. Clin. North Am. 6:449, Weiner, D. S., and Macnab, 1.: Ruptures of the rotator cuff: Follow-up evaluation of operative repairs. Can. J. Surg. 13:219, Wilson, C. L.: Lesions of the supraspinatus tendon: Degeneration, rupture and calcification. Arch. Surg. 46:307, Wolfgang, G. L.: Surgical repair of tears of the rotator cuff of the shoulder: Factors influencing the result. J. Bone Joint Surg. 56A:14, 1974.

Rehabilitation Guidelines for Large Rotator Cuff Repair

Rehabilitation Guidelines for Large Rotator Cuff Repair The true shoulder joint is called the glenohumeral joint and consists humeral head and the glenoid. It is a ball and socket joint. Anatomy of the