- Email: [email protected]
Rotator cuff repair in patients with type I diabetes mellitus
Rotator cuff repair in patients with type I diabetes mellitus Andrew L. Chen, MD, MS,a Joel A. Shapiro, MD,b Anthony K. Ahn, MD,b Joseph D. Zuckerman, MD,b and Frances Cuomo, MD,b Vail, CO, and New York, NY
Insulin-dependent diabetes mellitus is associated with shoulder stiffness and a propensity toward postoperative wound complications and infection. We compared our results of open repair of full-thickness rotator cuff tears in 30 diabetic patients with those of a matched, nondiabetic population. No differences were observed in preoperative range of motion, although at a mean of 34 months, significant differences in shoulder active range of motion and passive range of motion were found postoperatively at 6 weeks, 6 months, and final follow-up (P ⬍ .05). On the basis of American Shoulder and Elbow Surgeons shoulder scoring, there were 27 (90%) and 28 (93%) good or excellent results in the diabetic and comparison groups, respectively. Complications occurred in 5 diabetic patients (17%), with 2 failures (7%) and 3 infections (10%), as compared with 1 failure (3%) and no infections in the comparison group. Repair of the diabetic rotator cuff may be performed with the expectation of improved motion and function, although less than nondiabetic counterparts. The surgeon should remain cognizant that a higher rate of complications, infection in particular, may occur after rotator cuff repair in the diabetic population. (J Shoulder Elbow Surg 2003;12:416-21.)
T
ears of the rotator cuff are a common cause of shoulder pain and dysfunction.8 Since Codman’s classic description of open surgical repair of fullthickness tears of the supraspinatus tendon in 1911,7 the goals of rotator cuff repair have been pain relief and functional improvement. Results of open repair of full-thickness lesions of the rotator cuff have generally been favorable, with relief of pain, inNo 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. From Steadman Hawkins Sports Medicine Foundation, Vail, CO,a and the Hospital for Joint Diseasesb, New York, NY. Reprint requests: Andrew L. Chen, MD, MS, Steadman Hawkins Foundation, Suite 1000, 181 West Meadow Drive, Vail, CO 81657 (E-mail: [email protected]). Copyright © 2003 by Journal of Shoulder and Elbow Surgery Board of Trustees. 1058-2746/2003/$35.00 ⫹ 0 doi:10.1016/S1058-2746(03)00172-1
416
creased range of motion, increased functional capacity, and high rates of patient satisfaction in most studies.1,3,8,9,12-14,21,25,30 Complications have been reported to occur in up to 10.5% of cases, with an overall revision rate of 3% for failure of repair.18 The presence of insulin-dependent diabetes mellitus may confer increased surgical risks, including a higher risk of infection, compromised tissue quality for repair, and a propensity for problems with wound healing. A higher incidence of shoulder stiffness has been reported in insulin-dependent diabetic patients, with shoulder stiffness developing in as many as 36% of insulin-dependent diabetic patients, as compared with 3% of the general population.5 In 77% of these cases, bilateral shoulder stiffness is present.27 Chronic fluctuating serum glucose levels and longterm administration of exogenous insulin have been shown to increase the risk of shoulder stiffness developing.20 This has been shown to be not merely a consequence of the general disability caused by diabetic sequelae but a result of pathologic changes at the biochemical level that result in irreversible crosslinking of collagen molecules.21,23,24,26 Whether this increased propensity toward shoulder stiffness manifests postoperatively after shoulder surgery has not previously been investigated. To our knowledge, no study has specifically examined the results of rotator cuff repair in the diabetic population. The purpose of this investigation was to compare our experience with open repair of fullthickness defects of the rotator cuff in the insulindependent diabetic population with those of a matched, nondiabetic comparison population. MATERIALS AND METHODS We reviewed our results of open repair of full-thickness rotator cuff defects in 30 consecutive patients with insulindependent diabetes mellitus (mean age, 61.0 years; range, 34-77 years). All patients in the diabetic group required insulin for control for serum glucose levels for at least 5 years before referral to our institution. A group of 30 patients (mean age, 58.6 years; range, 32-83 years) matched for age, comorbidities, and length of follow-up but without history of diabetes mellitus (diet-, oral medication–, or insulin-controlled) were selected for comparison. Excluded were patients with a history of prior shoulder dislocation or fracture, acromioclavicular pain, degenerative or inflammatory arthrosis, infection, neuropathic changes,
Chen et al
J Shoulder Elbow Surg Volume 12, Number 5
prior surgical procedures to the shoulder, active workers’ compensation claims, or fewer than 24 months of postoperative follow-up. A total of 33 men and 27 women were evaluated. The right side was involved in 40 patients and the left in 20. Twenty-four patients (forty percent) could recall a specific, traumatic event. Initial evaluation included a standardized questionnaire and physical examination with determination of functional status as per the American Shoulder and Elbow Surgeons (ASES) standardized shoulder assessment form.25 This system is based on a 100-point scoring system including the patient’s report of pain on a visual analog scale (50 points), in addition to the cumulative score of 10 activities of daily living on a 4-point ordinal scale. With this system, results were rated as excellent (90-100 points), good (80-89 points), fair (70-79 points), and poor (⬍70 points). Greater than 80 points was considered a satisfactory result. Active and passive ranges of motion were measured with a goniometer. Forward elevation of the shoulder was measured in the scapular plane. External rotation was assessed with the elbow flexed at the patient’s side. Internal rotation behind the back was determined by the superiormost vertebral level reached by the hand. In cases in which the patient was unable to rotate internally sufficiently to the vertebrae, the rotation was categorized as being to the posterior-superior iliac spine, greater trochanter of the hip, or anterior-superior iliac spine. All patients had standard radiographic projections of the affected shoulder, including anteroposterior, scapular lateral, and axillary views. Two patients underwent arthrography, and fifty patients underwent magnetic resonance imaging for documentation of rotator cuff pathology. Two patients in the nondiabetic group received corticosteroid injections to the affected shoulder within 3 months of surgery. No patient in the diabetic group received a corticosteroid injection to the affected shoulder within 6 months of surgery. All patients were initially managed with nonoperative treatment, including nonsteroidal anti-inflammatory medications and formal physical therapy for strengthening and optimization of shoulder range of motion. In all cases the decision to proceed with rotator cuff repair was based on the history, physical examination, degree of functional debilitation, radiographic documentation of a tear, and failure of nonoperative measures. The mean duration between onset of symptoms and the surgical repair was 38 weeks (range, 2-56 weeks) in the diabetic group and 35 weeks (range, 4-48 weeks) in the nondiabetic group. A standardized treatment protocol was used. All patients received preoperative prophylactic antibiotics consisting of a first-generation cephalosporin (eg, 1 g cefazolin) or clindamycin (600 mg) if allergic to penicillin. All underwent open repair of full-thickness rotator cuff tears through an anterosuperior approach to the shoulder, subperiosteal elevation of the anterior deltoid, acromioplasty, and resection of the coracoacromial ligament and any osteophytes on the inferior aspect of the acromioclavicular joint. Any patient requiring distal clavicle resection was excluded from this investigation. Intraoperative measurement of tear size was performed according to the system of De Orio and Cofield.10 Patients with massive (⬎5 cm in length) tears of the rotator cuff were excluded from the study. Rotator cuff
417
repair was performed by a combination of tendon-to-tendon and tendon-to-bone techniques. As per our protocol during the study period, all patients were admitted for observation, during which time they received 24 hours of postoperative parenteral antibiotics and formal physical therapy twice per day. Diabetic patients were re-started on their preoperative insulin regimens with serum glucose measurement by fingersticks every 4 hours. Supplemental regular human insulin was given according to a sliding scale based on serum glucose determination. Passive range-of-motion exercises were initiated on the first postoperative day, with active range-of-motion and resistive exercises instituted when rotator cuff healing was believed to be secure— generally by 6 to 8 weeks postoperatively. Isometric and isotonic strengthening exercises were added based on tear size and healing, generally 10 to 12 weeks postoperatively. Determination of range of motion (forward elevation, external rotation, and internal rotation) and functional assessment with the ASES rating scale25 were performed at 6 weeks, 6 months, and final follow-up visits. Active and passive range-of-motion and functional scores were then plotted as a function of time postoperatively. Statistical analysis was performed with a matched Student t test. Statistical significance was defined as a P value less than .05. Ninety-five percent confidence intervals were calculated.
RESULTS The mean length of stay after surgery was 1.3 days (range, 1-4 days) in the diabetic group, as compared with 1.2 days (range, 1-3 days) in the comparison group (P ⬎ .10). The range-of-motion data are summarized in Tables I and II. No significant differences were observed between the diabetic and comparison populations in mean preoperative range of motion. Preoperative shoulder scores as per the ASES rating scale25 averaged 51.8 points (range, 28-74 points) in the diabetic group and 54.3 points (range, 33-74 points) in the comparison group (P ⬎ .7). Intraoperative measurement of tear size averaged 3.1 cm (range, 1.04.5 cm) in the diabetic group versus 3.2 cm (range, 0.5-4.5 cm) in the comparison group (P ⬎ .9). Length of follow-up averaged 34 months in both the diabetic group (range, 24-70 months) and comparison group (range, 24-64 months). Active motion and passive motion were significantly decreased in all planes of shoulder motion postoperatively in the diabetic group at 6 weeks, 6 months, and final followup. These results are summarized in Tables I and II and Figure 1. No significant differences were noted in ASES scores at final follow-up between the diabetic population (mean, 90.3 points; range, 53-100 points) and the comparison group (mean, 92.6 points; range, 64-100 points). In the diabetic population, the results were excellent in 12, good in 15, fair in 2, and poor in 1.
418
Chen et al
J Shoulder Elbow Surg September/October 2003
Table I Mean passive shoulder range of motion (in degrees) in diabetic and control populations Preoperative
Forward elevation Diabetic Control External rotation Diabetic Control Internal rotation Diabetic Control
Postoperative 6 wk
6 mo
Final follow-up
142 (50-170) 146 (100-175)
138 (80-170) 153 (85-175)†
153 (110-175) 164 (105-175)†
155 (120-175)* 170 (130-180)‡
46 (20-70) 48 (20-70)*
35 (10-55) 45 (20-55)§
L2 (T9-ASIS) L1 (T9-GT)*
L4 (L2-GT) L2 (T10-PSIS)#
44 (15-65) 58 (20-65)㛳 L2 (T10-PSIS) T11 (T7-L4)¶
48 (25-70) 63 (25-85)¶ T12 (T8-L5) T9 (T6-L4)
T, thoracic vertebrae; ASIS anterosuperior iliac spine; L, Lumbar vertebra; GT, greater trochanter; PSIS, posterosuperior iliac spine. *P ⬎ .30 (not significant); †P ⬍ .006; ‡P ⬍ .005; §P ⬍ .004; 㛳P ⬍ .002; ¶P ⬍ .001; #P ⬍ .009.
Table II Mean active shoulder range of motion (in degrees) in diabetic and control populations Preoperative
Forward elevation Diabetic Control External rotation Diabetic Control Internal rotation Diabetic Control
Postoperative 6 wk
6 mo
Final follow-up
128 (65-170) 144 (65-165)†
147 (100-170) 160 (90-175)†
150 (115-175) 166 (125-180)‡
31 (5-55) 32 (5-70)*
26 (–5-45) 39 (10-45)§
42 (10-65) 56 (20-60)㛳
L3 (T10-ASIS) L3 (T9-ASIS)*
L5 (L3-GT) L2 (T11-GT)#
113 (40-165) 117 (85-160)*
L3 (T11-GT) T12 (T9-L5)¶
47 (20-70) 61 (25-75)¶ T12 (T9-L5) T9 (T7-L4)¶
T, thoracic vertebrae; ASIS anterosuperior iliac spine; L, Lumbar vertebra; GT, greater trochanter; PSIS, posterosuperior iliac spine. *P ⬎ .30 (not significant); †P ⬍ .006; ‡P ⬍ .005; §P ⬍ .004; 㛳P ⬍ .002; ¶P ⬍ .001; #P ⬍ .009.
Twenty-seven patients (ninety percent) had good or excellent results. Complications in the diabetic group included two cases of deep infection, one case of superficial infection, and two failures of repair, one of which required revision. The single poor result (ASES score, 53) occurred in a patient in whom there was a failure to regain adequate strength and range of motion after rotator cuff repair. Magnetic resonance imaging confirmed failure of healing at 6 months, necessitating revision at 9 months with a good result. The two fair results represented patients who had deep postoperative wound infections at a mean of 3 weeks postoperatively (range, 2-4 weeks) necessitating multiple debridements (mean, 2.5 debridements; range, 2-3 debridements). In one case, deltoid healing was incomplete as a result of soft-tissue compromise. Incomplete healing of the rotator cuff repair was noted in each case; at the time of final debridement in each case, the rotator cuff was re-repaired. There was no recognized osteomyelitis in either case. Intraoperative cultures revealed that Staphylococcus aureus was the offending organism in one case, and Staphylococcus epidermidis was responsible in the second. Postoperative intravenous antibiotics (cefazolin in one
case and vancomycin in one case) were used for 4 weeks in each case on the basis of culture sensitivity, followed by 2 weeks of oral antibiotics (cephalexin). Resolution of infection resulted in a fair result in one case (ASES score, 71) and a good result in the other (ASES score, 83). In one patient with a good result, local wound erythema and tenderness were found 1 week postoperatively; this patient was placed on a 10-day course of oral cephalexin with complete resolution. In another patient with a good result, a minor fall at 57 months prompted an outside physician to obtain imaging studies of the postoperative shoulder, documenting a full-thickness tear of the rotator cuff (1 cm) at the site of repair. After resolution of the acute incident, the patient returned to full activities without complaint or functional debilitation. In the comparison group, the results were excellent in 11, good in 17, fair in 1, and poor in 1. Twentyeight patients (ninety-three percent) had good or excellent results. The single failure occurred in a patient with a fair result because of chronic pain issues. The patient felt a pop in her shoulder with trivial lifting 10 months after rotator cuff repair and was successfully revised to an excellent result. The single poor result
J Shoulder Elbow Surg Volume 12, Number 5
Chen et al
419
Figure 1 Shoulder range of motion as a function of time period. Ninety-five percent confidence intervals (error bars) are indicated at each time period for forward elevation (A), external rotation (B), and internal rotation (C). L, Lumbar vertebra; T, thoracic vertebra; GT, greater trochanter.
occurred in a patient with chronic pain issues, in whom subjective complaints of pain limited her ultimate result, despite satisfactory strength and range of motion. No infectious complications were observed in the comparison group. DISCUSSION The results of open rotator cuff repair reported in this study compare favorably with those reported in the literature, with predictable abatement of pain, increased range of motion and strength, and a general restoration of functional capacity as per the ASES rating system,25 with 93% of patients having good or excellent results. Multiple studies have documented satisfactory results and functional improvement in 85% to 95% of patients who undergo primary open repair of full-thickness defects of the rotator cuff,4,13-15,22,25 with a high degree of correlation between pain relief and patient satisfaction.16 To our knowledge, no previous study has specifically examined the results of rotator cuff repair in an insulindependent, diabetic population. In this investigation, despite statistically significant decreases in range of motion achieved at each post-
operative time point studied (6 weeks, 6 months, and final follow-up), 90% of our diabetic study population achieved good or excellent results at final follow-up, as compared with 93% good or excellent results in the comparison population. Joint stiffness is a common musculoskeletal manifestation of diabetes mellitus and is more frequent in those who have been chronically insulin-dependent.28 It has been attributed to long-standing high-circulating serum glucose levels that result in irreversible, non-enzymatic glycosylation and chemical cross-linking of collagen chains.11 Multiple reports have observed that the diabetic stiff shoulder, even in well-controlled diabetic patients, has the potential to be more resistant to all treatment modalities.2,23,24 This has been attributed to the elaboration of chemotactic inflammatory mediators as well as excessive generation of potent mitogenic polypeptides, such as platelet-derived growth factor, capable of eliciting an exuberant fibrogenic response in response to injury.28 The dense collagen production results in a thickened capsule and scar tissue histopathologically similar to the palmar fascia found in Dupuytren’s contracture. Despite the observed reductions in range of motion in the diabetic group,
420
Chen et al
90% of patients had good or excellent results. This is likely because of an improvement in pain, functional capacity, and preservation of shoulder mobility within a functional range, despite a loss of terminal motion. Complications after open repair of the rotator cuff have been well described. Mansat et al18 suggested that complications after rotator cuff repair are underreported and may occur in up to 33% of cases, instead of the 10% complication rate classically cited. In their series 16% of these represented major surgical complications affecting the final result, including failure of repair, shoulder stiffness, infection, and dislocation, in decreasing order of frequency, and no cases of fracture (acromion or greater tuberosity) or deltoid dehiscence.18 They postulated that stiffness after rotator cuff repair is related to an inflammatory response that is present preceding surgery and typically occurs in association with smaller rotator cuff tears in which a larger, intact rotator cuff surface area may participate in the inflammatory response.18 In this study we observed an overall complication rate of 16.7% (5 patients) in the diabetic group, as compared with 3.3% (1 patient) in the comparison group. A failure of repair necessitating revision occurred in 1 patient (3.3%) in each group (diabetic and comparison). One patient in the diabetic group had a full-thickness rotator cuff tear after a fall nearly 6 years after repair; this was confirmed by magnetic resonance imaging, although the patient remained pain-free and functionally unchanged after the initial abrasion and contusion resolved. Diminished pain associated with increased strength and functional capacity despite magnetic resonance documentation of structural repair failure has previously been reported.17 Importantly, in 3 of 30 diabetic patients (10%) an infectious complication occurred, compared with none in the comparison population. Infection after rotator cuff repair has been previously reported to be an uncommon complication with potentially disastrous results.8,19,29 Settecerri et al29 reported their institutional incidence of infection after rotator cuff repair to be 0.27%, whereas Mansat et al,18 in a review of 2948 rotator cuff repairs reported in the literature between 1982 and 1995, found the incidence to be 1.1%. This is notably lower than the observed incidence of infection of 10% within our diabetic study population, although it is comparable to our institutional incidence of 0.5%. In a retrospective review of 13 patients with deep infections after rotator cuff repair, Mirzayan et al19 found that whereas only one patient was a type A host (no risk factors for infection or compromised wound healing) in the classification by Cierny and Mader,6 the majority of their patients were type B hosts with evidence of systemic compromise. In keeping with their results,
J Shoulder Elbow Surg September/October 2003
all of our patients in the diabetic group were type B hosts and were, therefore, theoretically at increased risk for infection after rotator cuff repair. Moreover, the responsible organisms in this study (S aureus and S epidermidis) were the same as those previously identified as being most commonly involved in infections after rotator cuff repair.19 In a review of 16 cases of deep wound infection after rotator cuff repair in nondiabetic patients, Settecerri et al29 recognized that although the results after treatment of an infected rotator cuff repair are generally inferior to those of an uncomplicated repair, they are not uniformly poor. In that series, of the 11 patients with sufficient information to calculate UCLA shoulder scores, 5 had satisfactory results.29 Our results are comparable, with one of two patients achieving a satisfactory result with reoperation for deep infection. In our series all patients received a single preoperative dose of antibiotics, followed by 24 hours of postoperative parenteral antibiotics. As the antibiotic regimens were identical in both the diabetic and matched groups, we believe that the observed differences in the complication rate are indicative of the relative state of immunocompromise of the diabetic patient. In diabetic patients undergoing rotator cuff repair, we have since become more aggressive in treating wound-healing abnormalities, such as incisional erythema or prolonged wound drainage. As this investigation evaluated the outcome of open rotator cuff repairs through an anterosuperior approach, it remains unknown whether our findings are applicable to rotator cuff repairs through more limited approaches, such as mini-open repairs or those performed arthroscopically. We conclude that open rotator cuff repair may be performed in the diabetic population with the expectation of increased motion and functional capacity. Postoperative shoulder stiffness in the diabetic patient is reflected in all planes of shoulder motion, although in our series this did not significantly affect functional outcome. The surgeon must remain cognizant that a higher rate of complications, infection in particular, may occur. Prompt recognition, multiple surgical debridements, and intravenous antibiotics are necessary for resolution of infection and maximization of functional recovery. REFERENCES
1. Adamson G, Tibone J. Ten-year assessment of primary rotator cuff repairs. J Shoulder Elbow Surg 1993;2:57-63. 2. Arkkila PE, Kantola IM, Viikari JS, Ronnemaa T. Shoulder capsulitis in type I and type II diabetic patients: association with diabetic complications and related diseases. Ann Rheum Dis 1996;55:907-14. 3. Bassett RW, Cofield RH. Acute tears of the rotator cuff. The timing of surgical repair. Clin Orthop 1983;175:18-24. 4. Bjorkenheim JM, Paavolainen P, Ahovuo J, Slatis P. Surgical
J Shoulder Elbow Surg Volume 12, Number 5
5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.
16. 17.
repair of the rotator cuff and surrounding tissues. Factors influencing the results. Clin Orthop 1988;236:148-53. Bunker TD, Anthony PP. The pathology of frozen shoulder: a Dupuytren-like disease. J Bone Joint Surg Br 1995;77:677-83. Calhoun JH, Cierny G III, Holtom P, Mader J, Nelson CL. Symposium: current concepts in the management of osteomyelitis. Contemp Orthop 1994;28:157-85. Codman EA. Complete rupture of the supraspinatus tendon. Operative treatment with report of two successful cases. Boston Med Surg J 1911;164:708-10. Cofield RH. Rotator cuff disease of the shoulder. J Bone Joint Surg Am 1985;67:974-9. Cofield RH, Hoffmeyer P, Lanzer W. Surgical repair of chronic rotator cuff tears. Orthop Trans 1990;14:251-2. De Orio JK, Cofield RH. Results of a second attempt at surgical repair of a failed initial rotator cuff repair. J Bone Joint Surg Am 1984;66:563-7. Fisher L, Kurtz A, Shipley M. Association between cheirarthropathy and frozen shoulder in patients with insulin-dependent diabetes mellitus. Br J Rheumatol 1986;25:141-6. Gazielly D, Gleyze P, Montagnon C. Functional and anatomic results after rotator cuff repair. Clin Orthop 1994;304:43-53. Harryman DT II, Mack LA, Wang KY, et al. Repairs of the rotator cuff. Correlation of functional results with integrity of the cuff. J Bone Joint Surg Am 1991;73:982-9. Hawkins RJ, Misamore GW, Hobeika PE. Surgery for full thickness rotator cuff tears. J Bone Joint Surg Am 1985;67:1349-55. Iannotti JP, Bernot MP, Kuhlman JR, Kelley MJ, Williams GR. Postoperative assessment of shoulder function: a prospective study of full-thickness rotator cuff tears. J Shoulder Elbow Surg 1996;5:449-57. Iannotti JP. Full-thickness rotator cuff tears: factors affecting surgical outcome. J Am Acad Orthop Surg 1994;2:87-95. Jost B, Pfirrmann CW, Gerber C. Clinical outcome after structural failure of rotator cuff repair. J Bone Joint Surg Am 2000;82:30414.
Chen et al
421
18. Mansat P, Cofield RH, Kersten TE, Rowland CM. Complications of rotator cuff repair. Orthop Clin North Am 1997;28:205-13. 19. Mirzayan R, Itamura JM, Vangsness CT Jr, et al. Management of chronic deep infection following rotator cuff repair. J Bone Joint Surg Am 2000;82:1115-21. 20. Moren-Hybbinette I, Moritz U, Schersten B. The clinical picture of the painful diabetic shoulder: natural history, social consequences and analysis of concomitant hand syndrome. Acta Med Scand 1987;221:73-82. 21. Neer C, Flatow E, Lech O. Tears of the rotator cuff: long-term results of anterior acromioplasty and repair. Orthop Trans 1988; 12:673-4. 22. Neer CS II. Anterior acromioplasty for the chronic impingement syndrome in the shoulder: a preliminary report. J Bone Joint Surg Am 1972;54:41-50. 23. Ogilvie-Harris DJ, Myerthall S. The diabetic frozen shoulder: arthroscopic release. Arthroscopy 1997;13:1-8. 24. Pal B, Anderson J, Dick WC, Griffiths ID. Limitation of joint mobility and shoulder capsulitis in insulin- and non-insulin dependent diabetes mellitus. Br J Rheumatol 1986;25:147-51. 25. Richards R, An K, Bigliani L, et al. A standardized method for the assessment of shoulder function. J Shoulder Elbow Surg 1994;3: 347-52. 26. Rosenbloom AL, Silverstein JH. Connective tissue and joint disease in diabetes mellitus. Endocrinol Metab Clin North Am 1996;25:473-83. 27. Sattar MA, Lugman WA. Periarthritis: another duration-related complication of diabetes mellitus. Diabetes Care 1985;8:50710. 28. Scarlat MM, Harryman DT. Management of the diabetic stiff shoulder. Instr Course Lect 2000;49:293-4. 29. Settecerri JJ, Pitner MA, Rock MG, Hanssen AD, Cofield RH. Infection after rotator cuff repair. J Shoulder Elbow Surg 1999; 8:1-5. 30. Wolfgang GL. Surgical repair of tears of the rotator cuff of the shoulder. J Bone Joint Surg Am 1974;56:14-26.
Insulin-dependent diabetes mellitus is associated with shoulder stiffness and a propensity toward postoperative wound complications and infection. We compared our results of open repair of full-thickness rotator cuff tears in 30 diabetic patients with those of a matched, nondiabetic population. No differences were observed in preoperative range of motion, although at a mean of 34 months, significant differences in shoulder active range of motion and passive range of motion were found postoperatively at 6 weeks, 6 months, and final follow-up (P ⬍ .05). On the basis of American Shoulder and Elbow Surgeons shoulder scoring, there were 27 (90%) and 28 (93%) good or excellent results in the diabetic and comparison groups, respectively. Complications occurred in 5 diabetic patients (17%), with 2 failures (7%) and 3 infections (10%), as compared with 1 failure (3%) and no infections in the comparison group. Repair of the diabetic rotator cuff may be performed with the expectation of improved motion and function, although less than nondiabetic counterparts. The surgeon should remain cognizant that a higher rate of complications, infection in particular, may occur after rotator cuff repair in the diabetic population. (J Shoulder Elbow Surg 2003;12:416-21.)
T
ears of the rotator cuff are a common cause of shoulder pain and dysfunction.8 Since Codman’s classic description of open surgical repair of fullthickness tears of the supraspinatus tendon in 1911,7 the goals of rotator cuff repair have been pain relief and functional improvement. Results of open repair of full-thickness lesions of the rotator cuff have generally been favorable, with relief of pain, inNo 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. From Steadman Hawkins Sports Medicine Foundation, Vail, CO,a and the Hospital for Joint Diseasesb, New York, NY. Reprint requests: Andrew L. Chen, MD, MS, Steadman Hawkins Foundation, Suite 1000, 181 West Meadow Drive, Vail, CO 81657 (E-mail: [email protected]). Copyright © 2003 by Journal of Shoulder and Elbow Surgery Board of Trustees. 1058-2746/2003/$35.00 ⫹ 0 doi:10.1016/S1058-2746(03)00172-1
416
creased range of motion, increased functional capacity, and high rates of patient satisfaction in most studies.1,3,8,9,12-14,21,25,30 Complications have been reported to occur in up to 10.5% of cases, with an overall revision rate of 3% for failure of repair.18 The presence of insulin-dependent diabetes mellitus may confer increased surgical risks, including a higher risk of infection, compromised tissue quality for repair, and a propensity for problems with wound healing. A higher incidence of shoulder stiffness has been reported in insulin-dependent diabetic patients, with shoulder stiffness developing in as many as 36% of insulin-dependent diabetic patients, as compared with 3% of the general population.5 In 77% of these cases, bilateral shoulder stiffness is present.27 Chronic fluctuating serum glucose levels and longterm administration of exogenous insulin have been shown to increase the risk of shoulder stiffness developing.20 This has been shown to be not merely a consequence of the general disability caused by diabetic sequelae but a result of pathologic changes at the biochemical level that result in irreversible crosslinking of collagen molecules.21,23,24,26 Whether this increased propensity toward shoulder stiffness manifests postoperatively after shoulder surgery has not previously been investigated. To our knowledge, no study has specifically examined the results of rotator cuff repair in the diabetic population. The purpose of this investigation was to compare our experience with open repair of fullthickness defects of the rotator cuff in the insulindependent diabetic population with those of a matched, nondiabetic comparison population. MATERIALS AND METHODS We reviewed our results of open repair of full-thickness rotator cuff defects in 30 consecutive patients with insulindependent diabetes mellitus (mean age, 61.0 years; range, 34-77 years). All patients in the diabetic group required insulin for control for serum glucose levels for at least 5 years before referral to our institution. A group of 30 patients (mean age, 58.6 years; range, 32-83 years) matched for age, comorbidities, and length of follow-up but without history of diabetes mellitus (diet-, oral medication–, or insulin-controlled) were selected for comparison. Excluded were patients with a history of prior shoulder dislocation or fracture, acromioclavicular pain, degenerative or inflammatory arthrosis, infection, neuropathic changes,
Chen et al
J Shoulder Elbow Surg Volume 12, Number 5
prior surgical procedures to the shoulder, active workers’ compensation claims, or fewer than 24 months of postoperative follow-up. A total of 33 men and 27 women were evaluated. The right side was involved in 40 patients and the left in 20. Twenty-four patients (forty percent) could recall a specific, traumatic event. Initial evaluation included a standardized questionnaire and physical examination with determination of functional status as per the American Shoulder and Elbow Surgeons (ASES) standardized shoulder assessment form.25 This system is based on a 100-point scoring system including the patient’s report of pain on a visual analog scale (50 points), in addition to the cumulative score of 10 activities of daily living on a 4-point ordinal scale. With this system, results were rated as excellent (90-100 points), good (80-89 points), fair (70-79 points), and poor (⬍70 points). Greater than 80 points was considered a satisfactory result. Active and passive ranges of motion were measured with a goniometer. Forward elevation of the shoulder was measured in the scapular plane. External rotation was assessed with the elbow flexed at the patient’s side. Internal rotation behind the back was determined by the superiormost vertebral level reached by the hand. In cases in which the patient was unable to rotate internally sufficiently to the vertebrae, the rotation was categorized as being to the posterior-superior iliac spine, greater trochanter of the hip, or anterior-superior iliac spine. All patients had standard radiographic projections of the affected shoulder, including anteroposterior, scapular lateral, and axillary views. Two patients underwent arthrography, and fifty patients underwent magnetic resonance imaging for documentation of rotator cuff pathology. Two patients in the nondiabetic group received corticosteroid injections to the affected shoulder within 3 months of surgery. No patient in the diabetic group received a corticosteroid injection to the affected shoulder within 6 months of surgery. All patients were initially managed with nonoperative treatment, including nonsteroidal anti-inflammatory medications and formal physical therapy for strengthening and optimization of shoulder range of motion. In all cases the decision to proceed with rotator cuff repair was based on the history, physical examination, degree of functional debilitation, radiographic documentation of a tear, and failure of nonoperative measures. The mean duration between onset of symptoms and the surgical repair was 38 weeks (range, 2-56 weeks) in the diabetic group and 35 weeks (range, 4-48 weeks) in the nondiabetic group. A standardized treatment protocol was used. All patients received preoperative prophylactic antibiotics consisting of a first-generation cephalosporin (eg, 1 g cefazolin) or clindamycin (600 mg) if allergic to penicillin. All underwent open repair of full-thickness rotator cuff tears through an anterosuperior approach to the shoulder, subperiosteal elevation of the anterior deltoid, acromioplasty, and resection of the coracoacromial ligament and any osteophytes on the inferior aspect of the acromioclavicular joint. Any patient requiring distal clavicle resection was excluded from this investigation. Intraoperative measurement of tear size was performed according to the system of De Orio and Cofield.10 Patients with massive (⬎5 cm in length) tears of the rotator cuff were excluded from the study. Rotator cuff
417
repair was performed by a combination of tendon-to-tendon and tendon-to-bone techniques. As per our protocol during the study period, all patients were admitted for observation, during which time they received 24 hours of postoperative parenteral antibiotics and formal physical therapy twice per day. Diabetic patients were re-started on their preoperative insulin regimens with serum glucose measurement by fingersticks every 4 hours. Supplemental regular human insulin was given according to a sliding scale based on serum glucose determination. Passive range-of-motion exercises were initiated on the first postoperative day, with active range-of-motion and resistive exercises instituted when rotator cuff healing was believed to be secure— generally by 6 to 8 weeks postoperatively. Isometric and isotonic strengthening exercises were added based on tear size and healing, generally 10 to 12 weeks postoperatively. Determination of range of motion (forward elevation, external rotation, and internal rotation) and functional assessment with the ASES rating scale25 were performed at 6 weeks, 6 months, and final follow-up visits. Active and passive range-of-motion and functional scores were then plotted as a function of time postoperatively. Statistical analysis was performed with a matched Student t test. Statistical significance was defined as a P value less than .05. Ninety-five percent confidence intervals were calculated.
RESULTS The mean length of stay after surgery was 1.3 days (range, 1-4 days) in the diabetic group, as compared with 1.2 days (range, 1-3 days) in the comparison group (P ⬎ .10). The range-of-motion data are summarized in Tables I and II. No significant differences were observed between the diabetic and comparison populations in mean preoperative range of motion. Preoperative shoulder scores as per the ASES rating scale25 averaged 51.8 points (range, 28-74 points) in the diabetic group and 54.3 points (range, 33-74 points) in the comparison group (P ⬎ .7). Intraoperative measurement of tear size averaged 3.1 cm (range, 1.04.5 cm) in the diabetic group versus 3.2 cm (range, 0.5-4.5 cm) in the comparison group (P ⬎ .9). Length of follow-up averaged 34 months in both the diabetic group (range, 24-70 months) and comparison group (range, 24-64 months). Active motion and passive motion were significantly decreased in all planes of shoulder motion postoperatively in the diabetic group at 6 weeks, 6 months, and final followup. These results are summarized in Tables I and II and Figure 1. No significant differences were noted in ASES scores at final follow-up between the diabetic population (mean, 90.3 points; range, 53-100 points) and the comparison group (mean, 92.6 points; range, 64-100 points). In the diabetic population, the results were excellent in 12, good in 15, fair in 2, and poor in 1.
418
Chen et al
J Shoulder Elbow Surg September/October 2003
Table I Mean passive shoulder range of motion (in degrees) in diabetic and control populations Preoperative
Forward elevation Diabetic Control External rotation Diabetic Control Internal rotation Diabetic Control
Postoperative 6 wk
6 mo
Final follow-up
142 (50-170) 146 (100-175)
138 (80-170) 153 (85-175)†
153 (110-175) 164 (105-175)†
155 (120-175)* 170 (130-180)‡
46 (20-70) 48 (20-70)*
35 (10-55) 45 (20-55)§
L2 (T9-ASIS) L1 (T9-GT)*
L4 (L2-GT) L2 (T10-PSIS)#
44 (15-65) 58 (20-65)㛳 L2 (T10-PSIS) T11 (T7-L4)¶
48 (25-70) 63 (25-85)¶ T12 (T8-L5) T9 (T6-L4)
T, thoracic vertebrae; ASIS anterosuperior iliac spine; L, Lumbar vertebra; GT, greater trochanter; PSIS, posterosuperior iliac spine. *P ⬎ .30 (not significant); †P ⬍ .006; ‡P ⬍ .005; §P ⬍ .004; 㛳P ⬍ .002; ¶P ⬍ .001; #P ⬍ .009.
Table II Mean active shoulder range of motion (in degrees) in diabetic and control populations Preoperative
Forward elevation Diabetic Control External rotation Diabetic Control Internal rotation Diabetic Control
Postoperative 6 wk
6 mo
Final follow-up
128 (65-170) 144 (65-165)†
147 (100-170) 160 (90-175)†
150 (115-175) 166 (125-180)‡
31 (5-55) 32 (5-70)*
26 (–5-45) 39 (10-45)§
42 (10-65) 56 (20-60)㛳
L3 (T10-ASIS) L3 (T9-ASIS)*
L5 (L3-GT) L2 (T11-GT)#
113 (40-165) 117 (85-160)*
L3 (T11-GT) T12 (T9-L5)¶
47 (20-70) 61 (25-75)¶ T12 (T9-L5) T9 (T7-L4)¶
T, thoracic vertebrae; ASIS anterosuperior iliac spine; L, Lumbar vertebra; GT, greater trochanter; PSIS, posterosuperior iliac spine. *P ⬎ .30 (not significant); †P ⬍ .006; ‡P ⬍ .005; §P ⬍ .004; 㛳P ⬍ .002; ¶P ⬍ .001; #P ⬍ .009.
Twenty-seven patients (ninety percent) had good or excellent results. Complications in the diabetic group included two cases of deep infection, one case of superficial infection, and two failures of repair, one of which required revision. The single poor result (ASES score, 53) occurred in a patient in whom there was a failure to regain adequate strength and range of motion after rotator cuff repair. Magnetic resonance imaging confirmed failure of healing at 6 months, necessitating revision at 9 months with a good result. The two fair results represented patients who had deep postoperative wound infections at a mean of 3 weeks postoperatively (range, 2-4 weeks) necessitating multiple debridements (mean, 2.5 debridements; range, 2-3 debridements). In one case, deltoid healing was incomplete as a result of soft-tissue compromise. Incomplete healing of the rotator cuff repair was noted in each case; at the time of final debridement in each case, the rotator cuff was re-repaired. There was no recognized osteomyelitis in either case. Intraoperative cultures revealed that Staphylococcus aureus was the offending organism in one case, and Staphylococcus epidermidis was responsible in the second. Postoperative intravenous antibiotics (cefazolin in one
case and vancomycin in one case) were used for 4 weeks in each case on the basis of culture sensitivity, followed by 2 weeks of oral antibiotics (cephalexin). Resolution of infection resulted in a fair result in one case (ASES score, 71) and a good result in the other (ASES score, 83). In one patient with a good result, local wound erythema and tenderness were found 1 week postoperatively; this patient was placed on a 10-day course of oral cephalexin with complete resolution. In another patient with a good result, a minor fall at 57 months prompted an outside physician to obtain imaging studies of the postoperative shoulder, documenting a full-thickness tear of the rotator cuff (1 cm) at the site of repair. After resolution of the acute incident, the patient returned to full activities without complaint or functional debilitation. In the comparison group, the results were excellent in 11, good in 17, fair in 1, and poor in 1. Twentyeight patients (ninety-three percent) had good or excellent results. The single failure occurred in a patient with a fair result because of chronic pain issues. The patient felt a pop in her shoulder with trivial lifting 10 months after rotator cuff repair and was successfully revised to an excellent result. The single poor result
J Shoulder Elbow Surg Volume 12, Number 5
Chen et al
419
Figure 1 Shoulder range of motion as a function of time period. Ninety-five percent confidence intervals (error bars) are indicated at each time period for forward elevation (A), external rotation (B), and internal rotation (C). L, Lumbar vertebra; T, thoracic vertebra; GT, greater trochanter.
occurred in a patient with chronic pain issues, in whom subjective complaints of pain limited her ultimate result, despite satisfactory strength and range of motion. No infectious complications were observed in the comparison group. DISCUSSION The results of open rotator cuff repair reported in this study compare favorably with those reported in the literature, with predictable abatement of pain, increased range of motion and strength, and a general restoration of functional capacity as per the ASES rating system,25 with 93% of patients having good or excellent results. Multiple studies have documented satisfactory results and functional improvement in 85% to 95% of patients who undergo primary open repair of full-thickness defects of the rotator cuff,4,13-15,22,25 with a high degree of correlation between pain relief and patient satisfaction.16 To our knowledge, no previous study has specifically examined the results of rotator cuff repair in an insulindependent, diabetic population. In this investigation, despite statistically significant decreases in range of motion achieved at each post-
operative time point studied (6 weeks, 6 months, and final follow-up), 90% of our diabetic study population achieved good or excellent results at final follow-up, as compared with 93% good or excellent results in the comparison population. Joint stiffness is a common musculoskeletal manifestation of diabetes mellitus and is more frequent in those who have been chronically insulin-dependent.28 It has been attributed to long-standing high-circulating serum glucose levels that result in irreversible, non-enzymatic glycosylation and chemical cross-linking of collagen chains.11 Multiple reports have observed that the diabetic stiff shoulder, even in well-controlled diabetic patients, has the potential to be more resistant to all treatment modalities.2,23,24 This has been attributed to the elaboration of chemotactic inflammatory mediators as well as excessive generation of potent mitogenic polypeptides, such as platelet-derived growth factor, capable of eliciting an exuberant fibrogenic response in response to injury.28 The dense collagen production results in a thickened capsule and scar tissue histopathologically similar to the palmar fascia found in Dupuytren’s contracture. Despite the observed reductions in range of motion in the diabetic group,
420
Chen et al
90% of patients had good or excellent results. This is likely because of an improvement in pain, functional capacity, and preservation of shoulder mobility within a functional range, despite a loss of terminal motion. Complications after open repair of the rotator cuff have been well described. Mansat et al18 suggested that complications after rotator cuff repair are underreported and may occur in up to 33% of cases, instead of the 10% complication rate classically cited. In their series 16% of these represented major surgical complications affecting the final result, including failure of repair, shoulder stiffness, infection, and dislocation, in decreasing order of frequency, and no cases of fracture (acromion or greater tuberosity) or deltoid dehiscence.18 They postulated that stiffness after rotator cuff repair is related to an inflammatory response that is present preceding surgery and typically occurs in association with smaller rotator cuff tears in which a larger, intact rotator cuff surface area may participate in the inflammatory response.18 In this study we observed an overall complication rate of 16.7% (5 patients) in the diabetic group, as compared with 3.3% (1 patient) in the comparison group. A failure of repair necessitating revision occurred in 1 patient (3.3%) in each group (diabetic and comparison). One patient in the diabetic group had a full-thickness rotator cuff tear after a fall nearly 6 years after repair; this was confirmed by magnetic resonance imaging, although the patient remained pain-free and functionally unchanged after the initial abrasion and contusion resolved. Diminished pain associated with increased strength and functional capacity despite magnetic resonance documentation of structural repair failure has previously been reported.17 Importantly, in 3 of 30 diabetic patients (10%) an infectious complication occurred, compared with none in the comparison population. Infection after rotator cuff repair has been previously reported to be an uncommon complication with potentially disastrous results.8,19,29 Settecerri et al29 reported their institutional incidence of infection after rotator cuff repair to be 0.27%, whereas Mansat et al,18 in a review of 2948 rotator cuff repairs reported in the literature between 1982 and 1995, found the incidence to be 1.1%. This is notably lower than the observed incidence of infection of 10% within our diabetic study population, although it is comparable to our institutional incidence of 0.5%. In a retrospective review of 13 patients with deep infections after rotator cuff repair, Mirzayan et al19 found that whereas only one patient was a type A host (no risk factors for infection or compromised wound healing) in the classification by Cierny and Mader,6 the majority of their patients were type B hosts with evidence of systemic compromise. In keeping with their results,
J Shoulder Elbow Surg September/October 2003
all of our patients in the diabetic group were type B hosts and were, therefore, theoretically at increased risk for infection after rotator cuff repair. Moreover, the responsible organisms in this study (S aureus and S epidermidis) were the same as those previously identified as being most commonly involved in infections after rotator cuff repair.19 In a review of 16 cases of deep wound infection after rotator cuff repair in nondiabetic patients, Settecerri et al29 recognized that although the results after treatment of an infected rotator cuff repair are generally inferior to those of an uncomplicated repair, they are not uniformly poor. In that series, of the 11 patients with sufficient information to calculate UCLA shoulder scores, 5 had satisfactory results.29 Our results are comparable, with one of two patients achieving a satisfactory result with reoperation for deep infection. In our series all patients received a single preoperative dose of antibiotics, followed by 24 hours of postoperative parenteral antibiotics. As the antibiotic regimens were identical in both the diabetic and matched groups, we believe that the observed differences in the complication rate are indicative of the relative state of immunocompromise of the diabetic patient. In diabetic patients undergoing rotator cuff repair, we have since become more aggressive in treating wound-healing abnormalities, such as incisional erythema or prolonged wound drainage. As this investigation evaluated the outcome of open rotator cuff repairs through an anterosuperior approach, it remains unknown whether our findings are applicable to rotator cuff repairs through more limited approaches, such as mini-open repairs or those performed arthroscopically. We conclude that open rotator cuff repair may be performed in the diabetic population with the expectation of increased motion and functional capacity. Postoperative shoulder stiffness in the diabetic patient is reflected in all planes of shoulder motion, although in our series this did not significantly affect functional outcome. The surgeon must remain cognizant that a higher rate of complications, infection in particular, may occur. Prompt recognition, multiple surgical debridements, and intravenous antibiotics are necessary for resolution of infection and maximization of functional recovery. REFERENCES
1. Adamson G, Tibone J. Ten-year assessment of primary rotator cuff repairs. J Shoulder Elbow Surg 1993;2:57-63. 2. Arkkila PE, Kantola IM, Viikari JS, Ronnemaa T. Shoulder capsulitis in type I and type II diabetic patients: association with diabetic complications and related diseases. Ann Rheum Dis 1996;55:907-14. 3. Bassett RW, Cofield RH. Acute tears of the rotator cuff. The timing of surgical repair. Clin Orthop 1983;175:18-24. 4. Bjorkenheim JM, Paavolainen P, Ahovuo J, Slatis P. Surgical
J Shoulder Elbow Surg Volume 12, Number 5
5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.
16. 17.
repair of the rotator cuff and surrounding tissues. Factors influencing the results. Clin Orthop 1988;236:148-53. Bunker TD, Anthony PP. The pathology of frozen shoulder: a Dupuytren-like disease. J Bone Joint Surg Br 1995;77:677-83. Calhoun JH, Cierny G III, Holtom P, Mader J, Nelson CL. Symposium: current concepts in the management of osteomyelitis. Contemp Orthop 1994;28:157-85. Codman EA. Complete rupture of the supraspinatus tendon. Operative treatment with report of two successful cases. Boston Med Surg J 1911;164:708-10. Cofield RH. Rotator cuff disease of the shoulder. J Bone Joint Surg Am 1985;67:974-9. Cofield RH, Hoffmeyer P, Lanzer W. Surgical repair of chronic rotator cuff tears. Orthop Trans 1990;14:251-2. De Orio JK, Cofield RH. Results of a second attempt at surgical repair of a failed initial rotator cuff repair. J Bone Joint Surg Am 1984;66:563-7. Fisher L, Kurtz A, Shipley M. Association between cheirarthropathy and frozen shoulder in patients with insulin-dependent diabetes mellitus. Br J Rheumatol 1986;25:141-6. Gazielly D, Gleyze P, Montagnon C. Functional and anatomic results after rotator cuff repair. Clin Orthop 1994;304:43-53. Harryman DT II, Mack LA, Wang KY, et al. Repairs of the rotator cuff. Correlation of functional results with integrity of the cuff. J Bone Joint Surg Am 1991;73:982-9. Hawkins RJ, Misamore GW, Hobeika PE. Surgery for full thickness rotator cuff tears. J Bone Joint Surg Am 1985;67:1349-55. Iannotti JP, Bernot MP, Kuhlman JR, Kelley MJ, Williams GR. Postoperative assessment of shoulder function: a prospective study of full-thickness rotator cuff tears. J Shoulder Elbow Surg 1996;5:449-57. Iannotti JP. Full-thickness rotator cuff tears: factors affecting surgical outcome. J Am Acad Orthop Surg 1994;2:87-95. Jost B, Pfirrmann CW, Gerber C. Clinical outcome after structural failure of rotator cuff repair. J Bone Joint Surg Am 2000;82:30414.
Chen et al
421
18. Mansat P, Cofield RH, Kersten TE, Rowland CM. Complications of rotator cuff repair. Orthop Clin North Am 1997;28:205-13. 19. Mirzayan R, Itamura JM, Vangsness CT Jr, et al. Management of chronic deep infection following rotator cuff repair. J Bone Joint Surg Am 2000;82:1115-21. 20. Moren-Hybbinette I, Moritz U, Schersten B. The clinical picture of the painful diabetic shoulder: natural history, social consequences and analysis of concomitant hand syndrome. Acta Med Scand 1987;221:73-82. 21. Neer C, Flatow E, Lech O. Tears of the rotator cuff: long-term results of anterior acromioplasty and repair. Orthop Trans 1988; 12:673-4. 22. Neer CS II. Anterior acromioplasty for the chronic impingement syndrome in the shoulder: a preliminary report. J Bone Joint Surg Am 1972;54:41-50. 23. Ogilvie-Harris DJ, Myerthall S. The diabetic frozen shoulder: arthroscopic release. Arthroscopy 1997;13:1-8. 24. Pal B, Anderson J, Dick WC, Griffiths ID. Limitation of joint mobility and shoulder capsulitis in insulin- and non-insulin dependent diabetes mellitus. Br J Rheumatol 1986;25:147-51. 25. Richards R, An K, Bigliani L, et al. A standardized method for the assessment of shoulder function. J Shoulder Elbow Surg 1994;3: 347-52. 26. Rosenbloom AL, Silverstein JH. Connective tissue and joint disease in diabetes mellitus. Endocrinol Metab Clin North Am 1996;25:473-83. 27. Sattar MA, Lugman WA. Periarthritis: another duration-related complication of diabetes mellitus. Diabetes Care 1985;8:50710. 28. Scarlat MM, Harryman DT. Management of the diabetic stiff shoulder. Instr Course Lect 2000;49:293-4. 29. Settecerri JJ, Pitner MA, Rock MG, Hanssen AD, Cofield RH. Infection after rotator cuff repair. J Shoulder Elbow Surg 1999; 8:1-5. 30. Wolfgang GL. Surgical repair of tears of the rotator cuff of the shoulder. J Bone Joint Surg Am 1974;56:14-26.