Does obesity affect early outcome of rotator cuff repair?

J Shoulder Elbow Surg (2010) 19, 1250-1255 www.elsevier.com/locate/ymse Does obesity affect early outcome of rotator cuff repair? Surena Namdari, MD, MSca, Keith Baldwin, MD, MPH, MSPTa, David Glaser, MDa, Andrew Green, MDb,* a Department of Orthopaedic Surgery, University of Pennsylvania, Philadelphia, PA, USA Department of Orthopaedic Surgery, Warren Alpert Medical School Brown University/Rhode Island Hospital, Providence, RI, USA b Background: Obesity is linked to major health conditions and poor surgical outcomes. The impact of obesity on self-perceived outcome after rotator cuff repair (RCR) is unclear. Materials and methods: We studied 154 patients who underwent RCR. Obesity was considered a body mass index (BMI) greater than 30. Preoperative and postoperative evaluations included the Disabilities of the Arm, Shoulder and Hand (DASH), Simple Shoulder Test, and visual analog scales for pain, function, and quality of life. Obese and control patients were compared for baseline demographics, surgical findings, and postoperative outcomes. Results: Our overall population had a mean BMI of 28.4 (95% confidence interval, 27.7-29.1). There were 57 obese patients (BMI >30) and 97 nonobese patients with 1- or 2-tendon rotator cuff tears. Mean followup was 54.8 weeks (range, 52.0-88.7 weeks). Preoperative DASH score was 45.2 for obese patients and 43.4 for control patients (P ¼ .524). The mean improvement in DASH score was 30.7 for obese patients and 26.1 for nonobese patients (P ¼ .152). There were no significant differences in the Simple Shoulder Test and visual analog scale scores. Worse follow-up DASH scores in both groups were associated with worker’s compensation status (P ¼ .003) and total comorbidities (P < .001). Multiple linear regression analysis showed that BMI (continuous) and obesity (dichotomous) were not significantly related to outcome after we controlled for confounding variables. Conclusions: Although obesity is considered a risk factor for poor postoperative outcomes after some surgical procedures, in our experience, obesity does not have an independent, significant effect on selfreported early outcomes after RCR. Level of evidence: Level III, Retrospective Case Control, Treatment Study. Ó 2010 Journal of Shoulder and Elbow Surgery Board of Trustees. Keywords: Rotator cuff tear; rotator cuff repair; obesity; outcomes; comorbidity; body mass index The World Health Organization defines obesity as a body mass index (BMI) greater than 30 and morbid obesity as a BMI greater than 40.27 Unfortunately, the prevalence of obesity is increasing in the developed world. It is estimated that 32% of non-institutionalized adults in *Reprint requests: Andrew Green, MD, University Orthopedics, 2 Dudley St, Ste 200, Providence, RI 02905. E-mail address: [email protected] (A. Green). the United States are obese.18 Although obesity has been linked to a number of major health conditions including ischemic heart disease,28 diabetes mellitus,8 stroke,8 osteoarthritis,8 and chronic renal failure,6 among others, it is also associated with surgical complications including thromboembolic disease5 and difficulty with general anesthesia.1,8 Despite this, obesity as a risk factor for orthopaedic surgery is less commonly evaluated, and the findings have 1058-2746/$ - see front matter Ó 2010 Journal of Shoulder and Elbow Surgery Board of Trustees. doi:10.1016/j.jse.2010.03.003 Obesity and rotator cuff repair been variable. Strauss et al22 found that obesity had no effect on the outcome of ankle fractures. Batsis et al3 found that there were no greater cardiac complications in obese hip fracture patients compared with patients who were not obese. Obesity has been evaluated in total hip and knee arthroplasties, with studies showing a greater risk of infection15 and pulmonary embolism16 in obese patients and no increased hospital resource utilization for elective total hip arthroplasty.4 For these reasons, some arthroplasty surgeons consider high BMI a relative contraindication for hip or knee arthroplasty. In terms of shoulder surgery and, more specifically, rotator cuff repair (RCR), it has been postulated that obesity may contribute to decreased vascularity through its association with risk factors such as elevated cholesterol level, atherosclerosis, chronic diseases such as diabetes and hypertension, and decreased physical activity.10,12,13,26 Wendelboe et al25 showed that individuals who are obese are at increased risk for rotator cuff tendinitis and rotator cufferelated surgery and noted that the apparent risk increased with the degree of obesity. The purpose of this study was to determine the effect of obesity and BMI on the preoperative and postoperative outcomes of patients undergoing RCR. We hypothesized that obesity and BMI would be a negative predictor of patient self-perceived outcome after RCR. Materials and methods The study was approved by the Rhode Island Hospital Institutional Review Board (study No. CMT-0097-02 and registration No. 00000396). Patients were identified from a prospectively collected database. The records of 240 consecutive RCRs were retrospectively re-evaluated for this study. Inclusion criteria included patients who underwent surgical repair of a full-thickness rotator cuff tear (RCT) between January 1, 1998, and November 1, 2001. Exclusion criteria included revision RCR (8 patients), contralateral shoulder symptoms or surgery (16 patients), missing baseline or follow-up data (15 patients), follow-up of less than 1 year (17 patients), or subscapularis tear (30 patients). In total, 86 patients were eliminated based on these criteria, yielding a sample size of 154 patients. A power analysis was performed with operating characteristics of a threshold of type I error of .05 and a power of 0.80. We wished to be able to detect a standardized medium-sized difference on a continuous primary outcome measure (Disabilities of the Arm, Shoulder and Hand [DASH] and change in DASH). To detect this change using a 2-sided t test, we needed a harmonic mean of 64 patients per group for an unbalanced design. With 97 cases and 57 controls, a harmonic mean of 72 patients, which exceeds this minimum number, was achieved.21 Baseline data evaluation The preoperative data included patient demographics, a comprehensive medical history, and an outcomes questionnaire package. 1251 The outcomes questionnaire was administered to the patients before surgery as previously described by Tashjian et al.23 The questionnaire package was completed in the senior author’s office at the time of the preoperative visit 1 to 2 weeks before surgery. The first section of the self-assessment questionnaire addressed general medical problems via a series of binary (yes or no) questions. Comorbidities were defined as heart disease, high blood pressure, lung disease, diabetes, ulcer or stomach disease, kidney disease, liver disease, anemia or other blood disease, cancer, depression, osteoarthritis, back pain, rheumatoid arthritis, or other medical problems (specified). Each yes answer was considered to be 1 medical comorbidity. In addition, the questionnaire package included the DASH questionnaire2; visual analog scales (VASs) for pain, shoulder function, and quality of life; and the Simple Shoulder Test (SST).14 Additional data regarding worker’s compensation status, smoking history, and dominant extremity involvement were also obtained. The indications for surgery were dependent on a number of factors, including the patient’s age, general health, activity level, severity of pain and disability, and goals and expectations. The condition of the rotator cuff muscles as shown on preoperative imaging was also strongly considered. Surgical data included surgical technique (open, mini-open, arthroscopic), size of tear and tendon involvement, and additional shoulder pathology. Adjunctive procedures to the RCR, including subacromial decompression, distal clavicle excision, synovial debridement, and biceps tenodesis or tenotomy, were also considered. The postoperative rehabilitation protocol was dependent on the surgical repair technique.23 All of the patients were immobilized with a sling for 5 weeks after the repair. Patients with an open or mini-open repair began self-assisted passive pendulum circumduction, supine forward elevation, supine external rotation, supine cross-chest adduction, and standing internal rotation within a few days of surgery. Patients with an arthroscopic repair began passive self-assisted pendulum circumduction and supine external rotation within a few days of surgery. During the fifth week after surgery, these patients added the other ranges of motion. After sling use was discontinued, all of the patients began light active use. Stretching exercises were continued until the patients achieved maximal shoulder motion. Strengthening was initiated with isometric exercises during the seventh week after surgery and progressed to resistance-band exercises after 12 weeks. Postoperative evaluation Patients were evaluated at a mean of 54.8 weeks (range, 52.0-88.7 weeks). The same questionnaire package that was administered preoperatively was again administered postoperatively. Data from these patient visits were retrospectively reviewed. Statistical analysis Paired 2-tailed Student t tests were used to determine significant differences between obese and control patients at baseline and follow-up. Each variable was assessed to determine whether it was significantly related to the variable of interest (obesity) or the outcome of interest (DASH or change in DASH). In cases where the variable was dichotomous, mean DASH and change in DASH were compared by use of the Student t test for independent samples to determine whether the two were significantly related. 1252 In cases where the two variables were continuous, the Pearson correlation coefficient was used to determine whether the two variables were significantly related. In cases where the obese group was compared with the nonobese group to determine whether some characteristic of interest was present or absent, a Yates c2 test or Fisher exact test was used. P < .05 was considered statistically significant. Multiple linear regression analysis was used to analyze differences in postoperative outcomes (postoperative DASH and change in DASH). BMI was analyzed as a dichotomous variable to maintain our initial case/control structure. All medical, surgical, and demographic variables were considered potential confounders. For each model, variables were considered individually. If they were found to be significantly related to the outcome of interest (postoperative DASH in the first model or change in DASH in the second model), they were placed in the regression model. After this, regression analysis was carried out by use of all variables found to be significantly related to either the outcome of interest or the variable (obesity) of interest with a threshold P value of .10. Multiple linear regression was carried out with a threshold of .05 for statistical significance. Post hoc analysis with BMI as a continuous variable was also carried out. All statistics were calculated with SPSS software, version 15.0 (SPSS, Chicago, IL). Results The mean BMI for the entire sample of patients was 28.4 (95% confidence interval, 27.7-29.1). There were 57 obese patients (mean BMI, 33.3) and 97 nonobese patients (mean BMI, 25.5). The mean age was 55.2 years (range, 36-80 years) for obese patients and 58.6 years (range, 18-84 years) for control patients (P ¼ .055). Table I outlines preoperative patient demographics for both groups. There were no significant differences between obese patients and control patients in terms of gender, total comorbidities, smoking history, injury to dominant side, worker’s compensation status, or length of follow-up. When the presence of individual comorbidities was evaluated, only diabetes mellitus was significantly more common in the obese group (9 of 57) compared with control patients (5 of 92) (P ¼ .05). Table II compares the surgical data for obese patients and control patients. There were no significant differences between obese patients and control patients in terms of surgical technique (open vs mini-open vs arthroscopic), retraction of tear, size of tear, and prevalence of SLAP tears (superior labrum anterior-posterior) and tears of the long head of the biceps. There was no difference in the adjunctive use of distal clavicle resections, subacromial decompressions, synovectomies, or other procedures between the groups. Table III lists the changes in outcomes score from preoperative to follow-up assessment for obese patients and control patients. The mean preoperative DASH score was 45.2 for obese patients and 43.4 for control patients S. Namdari et al. (P ¼ .524). The mean postoperative DASH score was 17.3 in obese patients and 14.5 in control patients (P ¼ .336). There was a statistically significant improvement in the DASH scores in each group (P < .001 for both groups). There was no statistically significant difference in the change in DASH scores between the groups (P ¼ .152). Equivalent results were observed for the SST and VASs for pain, function, and quality of life (Table III), and there were no statistically significant differences observed between preoperative scores, postoperative scores, or changes in scores for each instrument. BMI was not significantly correlated with final DASH score but was correlated with change in DASH score (P ¼ .497 and P ¼ .034, respectively). BMI was not significantly correlated with final SST (P ¼ .411) or change in number of positive responses (P ¼ .906). When analyzed by size of tear, there was no difference in DASH score or change in DASH score for obese patients compared with control patients. When stratified by surgical approach (open, mini-open, arthroscopic), there was no difference in DASH score or change in DASH score for obese patients compared with control patients. There were 2 deep wound infections in the obese group (mini-open and open approaches) and 1 in the control group (open). During the follow-up period, 2 patients in each group underwent revision RCRs to treat a clinical outcome failure. The presence of increasing medical comorbidities (r ¼ 0.356) and presence of worker’s compensation claims were significantly associated with lower postoperative DASH score (P < .001 and P ¼ .003, respectively). The final linear regression model for postoperative DASH score found that worker’s compensation status (P < .001) and total comorbidities (P < .001) were significantly related to worse postoperative DASH scores. Change in DASH score was not related to either of these variables in our analysis. Obesity was not found to be significantly related to postoperative DASH score (P ¼ .172) or change in DASH score (P ¼ .153) when we controlled for diabetes mellitus, total comorbidities, worker’s compensation status, or other demographic variables. Discussion Obesity is a major problem in the developed world and is associated with a number of medical conditions and complications.1,5-8,28 Previously published studies in the orthopaedic literature are rare, and those that exist report somewhat conflicting results. Whereas some studies suggest that there is little effect of obesity on outcome,4,22 others show that obesity is associated with greater surgical difficulty, increased incidence of wound complications, and general medical complications.11,19,20,24 The primary purpose of this study was to determine the effect of obesity on the outcome of RCR. We found that Obesity and rotator cuff repair Table I 1253 Patient characteristics Variable Obese (n ¼ 57) Control (n ¼ 97) P value Age (y) Gender (male/female) BMI Total comorbidities Smoking Worker’s compensation Injury to dominant side Follow-up (wk) 55.8 26 (46%)/31 (54%) 33.3 2.3 10 (18%) 18 (32%) 40 (67%) 54.5 58.5 57 (59%)/40 (41%) 25.5 1.9 15 (15%) 25 (26%) 66 (68%) 54.6 .139 .158 <.001 .147 .911 .556 .924 .904 Table II Operative characteristics Variable Open surgery Mini-open Arthroscopic Isolated SS tear SS/IS tear Tear width (cm) Tear retraction (cm) SLAP tear Biceps tear Acromioplasty Distal clavicle resection Biceps tenodesis Biceps tenotomy Synovectomy Obese (n ¼ 57) Control (n ¼ 97) P value 10 (17%) 20 (38%) 27 (47%) 42 (72%) 14 (24%) 2.25 1.75 7 (26%) 4 (7%) 54 (95%) 14 (25%) 0 (0%) 4 (7%) 5 (9%) 10 (10%) 35 (36%) 52 (54%) 87 (83%) 18 (17%) 1.97 1.63 7 (15%) 10 (10%) 95 (98%) 24 (25%) 8 (8%) 2 (2%) 16 (17%) .322 .999 .561 .171 .384 .172 .402 .416 .548 .360 .999 .531 .195 .261 SS, Supraspinatus; IS, infraspinatus; SLAP, superior labrum anteriorposterior. obese patients with full-thickness RCTs requiring surgery are more likely to be diabetic. Our analysis of the data showed that obesity is not an independent risk factor for poor outcomes, because the functional results in both obese and control patients were similar when assessed by 3 commonly used instruments of assessment (DASH, SST, and VAS). This is the first study that specifically evaluates the effect of obesity on the outcome of RCR. Miranda et al17 conducted a prospective study to evaluate individual factors that influence the incidence and persistence of shoulder pain among forestry workers. Although they did not evaluate surgical procedures or outcomes, they found that the risk of incident shoulder pain was positively correlated with BMI. Their cohort comprised heavy laborers, however, and is not necessarily representative of the sample of patients that we studied. Wendelboe et al25 showed an association between BMI and shoulder repair surgery and noted that increasing BMI is a risk factor for rotator cuff tendinitis and related conditions. However, they did not evaluate outcomes of operative treatment in this patient population. Table III Changes in DASH, SST, and VAS for pain, function, and quality of life preoperatively and postoperatively 6 DASH 6 SST score (%) 6 SST (No. of yes responses) 6 VAS for pain 6 VAS for function 6 VAS for quality of life Obese (n ¼ 57) Control (n ¼ 97) P value 30.7  18.5 39.0  26.4 4.4  3.1 26.1  20.3 44.0  28.3 5.0  3.1 .152 .289 .308 4.8  2.8 4.8  3.2 4.0  2.8 4.4  2.6 4.3  2.6 3.4  2.8 .443 .304 .216 All values given are absolute values of the differences, presented as mean  SD. All values were improved. P values given are based on independent-samples t tests; equal variances are not assumed. In our study, regardless of the presence of obesity, patients with greater total comorbidities had worse followup DASH scores. However, when we evaluated the change in DASH score, the total number of comorbidities did not correlate with the outcome. This finding is consistent with previous work from our group reported by Tashjian et al,23 showing that patients with greater numbers of medical comorbidities have a worse general health status after RCR and greater improvement in pain, function, and DASH scores. Worker’s compensation was likewise associated with lower postoperative DASH scores but not with a worse change in DASH. This is also in keeping with a previous report by the senior author.9 Henn et al9 reported that patients with worker’s compensation claims had worse outcomes, even after controlling for confounding variables. Dominant extremity surgery was associated with a greater improvement in DASH score in our study. It is reasonable to expect that restoration of function to the dominant extremity would result in greater self-perceived improvement. Although there were 2 cases of infection in the obese group compared with 1 in the control group, we lacked sufficient power to analyze this variable. However, it is important to note that there were no cases of infection in either the obese or control patients after arthroscopic RCR. In this study, the nonobese control patients were more likely to undergo biceps tenodesis than obese patients for 1254 treatment of partial-thickness biceps tears. The senior author’s preferred treatment for pathology of the long head of the biceps tendon is to perform biceps tenotomy in patients with a greater BMI because of the reduced likelihood of cosmetic deformity in a patient with a larger extremity. In thinner, nonobese patients, a biceps tenodesis is preferred to avoid a cosmetic deformity of the biceps muscle. We recognize that this study has a number of limitations and weaknesses. Although the data were collected prospectively, the study design was retrospective, and there may be important factors that we did not consider that could affect the outcome. For example, we do not know whether obesity leads to a higher rate of RCTs or a higher likelihood for an irreparable tear because the only patients included were those with a symptomatic cuff rupture that was repaired. Furthermore, we lacked data regarding operative time and do not know whether obesity led to greater intraoperative difficulties. Although BMI is a quantitative determinant of obesity, it does not account for the distribution of body weight and cannot specifically determine the influence of upper extremity size on rotator cuff disease and patient self-reported functional outcome. We did not evaluate or consider whether obesity might be associated with patient compliance with postoperative management protocols. We evaluated subjective outcome measures and did not consider objective variables such as strength or range of motion. It is possible that obese patients have a less active lifestyle such that they could be satisfied with less motion and strength than the nonobese patients. In addition, we did not review preoperative and postoperative magnetic resonance images and thus did not consider the effects of rotator cuff muscle quality and postoperative integrity of the repair. Despite this, both obese and control patients had similar self-perceived outcomes at early follow-up and the intraoperatively determined size of the tear did not correlate with the outcomes assessed in this study. Nevertheless, we believe that our study also has significant strengths. We identified an appropriate control group to highlight differences in demographics, surgical variables, and outcome between obese and control patients. Obese patients comprised 35.6% of our study sample, which is similar to that reported, 32%, in the general population.18 We designed this investigation with a clearly identified number of variables to analyze and thereby minimized the influence of multiplicity in our data analysis. Finally, we used multiple instruments to assess patient-derived outcomes including the DASH, VAS, and SST and were able to control for potential confounders in our data analysis. Even when evaluating BMI as a continuous variable or when controlling for potential confounders such as total comorbidities, worker’s compensation status, smoking, or dominant-sided injury, we found that the outcomes did not appear to be affected by the presence of obesity. S. Namdari et al. Conclusion Although obesity is considered a risk factor for poor postoperative outcomes for certain surgical procedures, we found that obesity does not have an independent, significant effect on self-reported early outcomes after RCR. Thus, we conclude that obesity should not be a deterrent to surgical treatment for RCTs in otherwise appropriate clinical settings. Although there was no significant difference between obese patients and control patients, inferior outcomes in both groups were associated with worker’s compensation status and greater total comorbidities. This is consistent with the senior author’s previously reported experience. Lastly, the findings of this study only relate to the duration of follow-up that we assessed. Longer-term follow-up might show differences that relate to the effects of obesity on rotator cuff disease and functional outcome. Disclaimer The authors, their immediate families, and any research foundations with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article. References 1. Adams JP, Murphy PG. Obesity in anaesthesia and intensive care. Br J Anaesth 2000;85:91-108. 2. American Academy of Orthopaedic Surgeons. Disabilities of the arm, shoulder and hand. Rosemont (IL): American Academy of Orthopaedic Surgeons; 1995. 3. Batsis JA, Huddleston JM, Melton LJ IV, Huddleston PM, LopezJimenez F, Larson DR, et al. Body mass index and risk of adverse cardiac events in elderly patients with hip fracture: a population based study. J Am Geriatr Soc 2009;57:419-26. 4. Batsis JA, Naessens JM, Keegan MT, Wagie AE, Huddleston PM, Huddleston JM. Impact of body mass on hospital resource use in total hip arthroplasty. Public Health Nutr 2009;12:1122-32. 5. Darvall KA, Sam RC, Silverman SH, Bradbury AW, Adam DJ. Obesity and thrombosis. Eur J Vasc Endovasc Surg 2007;33:223-33. 6. Ejerblad E, Fored CM, Lindblad P, Fryzek J, McLaughlin JK, Nyrén O. Obesity and risk for chronic renal failure. J Am Soc Nephrol 2006;17:1695-702. 7. Finkelstein EA, Fiebelkorn IA, Wang G. National medical spending attributable to overweight and obesity: how much, and who’s paying. Health Affairs: Online (May) 2003. Available from: URL: http://content. healthaffairs.org/cgi/content/full/hlthaff.w3.219v1/DC1. Accessed Nov 15, 2009. 8. Haslam DW, James WP. Obesity. Lancet 2005;366:1197-209. 9. Henn RF III, Kang L, Tashjian RZ, Green A. Patients with workers’ compensation claims have worse outcomes after rotator cuff repair. J Bone Joint Surg Am 2008;90:2105-13. 10. Hubert HB, Feinleib M, McNamara PM, Castelli WP. Obesity as an independent risk factor for cardiovascular disease: a 26-year follow-up of participants in the Framingham Heart Study. Circulation 1983;67:968-77. Obesity and rotator cuff repair 11. Karunakar MA, Shah SN, Jerabek S. Body mass index as a predictor of complications after operative treatment of acetabular fractures. J Bone Joint Surg Am 2005;87:1498-502. 12. Kelsey JL. Epidemiology of musculoskeletal disorders. New York: Oxford University Press; 1982. 13. Kim KS, Owen WL, Williams D, Adams-Campbell LL. A comparison between BMI and Conicity index on predicting coronary heart disease: the Framingham Heart Study. Ann Epidemiol 2000;10:424-31. 14. Lippit SB, Harryman DT, Matsen FA III. A practical tool for evaluating function: the simple shoulder test. In: Matsen FA III, Fu FH, Hawkins RJ, editors. The shoulder: a balance of mobility and stability. Rosemont (IL): American Academy of Orthopaedic Surgeons; 1993. p. 501-18. 15. Malinzak RA, Ritter MA, Berend ME, Meding JB, Olberding EM, Davis KE. Morbidly obese, diabetic, younger, and unilateral joint arthroplasty patients have elevated total joint arthroplasty infection rates. J Arthroplasty 2009;24(Suppl):84-8. 16. Memtsoudis SG, Besculides MC, Gaber L, Liu S, González Della Valle A. Risk factors for pulmonary embolism after hip and knee arthroplasty: a population-based study. Int Orthop 2009;33: 1739-45. 17. Miranda H, Viikari-Juntura E, Martikainen R, Takala EP, Riihimäki H. A prospective study of work related factors and physical exercise as predictors of shoulder pain. Occup Environ Med 2001;58:528-34. 18. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Prevalence of overweight and obesity in the United States, 19992004. JAMA 295:1549-1555. 19. Porter SE, Graves ML, Qin Z, Russell GV. Operative experience of pelvic fractures in the obese. Obes Surg 2008;18:702-8. 1255 20. Porter SE, Russell GV, Dews RC, Qin Z, Woodall J, Graves M. Complications of acetabular fracture surgery in morbidly obese patients. J Orthop Trauma 2008;22:589-94. 21. Portney LG, Watkins MP. Appendix C: power analysis and determination of sample size. In: Foundations of clinical research: applications to practice. Stamford (CT): Appleton & Lange; 1993. 651e667. 22. Strauss EJ, Frank JB, Walsh M, Koval KJ, Egol KA. Does obesity influence the outcome after the operative treatment of ankle fractures? J Bone Joint Surg Br 2007;89:794-8. 23. Tashjian RZ, Henn RF, Kang L, Green A. The effect of comorbidity on self-assessed function in patients with a chronic rotator cuff tear. J Bone Joint Surg Am 2004;86:355-62. 24. Tucker MC, Schwappach JR, Leighton RK, Coupe K, Ricci WM. Results of femoral intramedullary nailing in patients who are obese versus those who are not obese: a prospective multicenter comparison study. J Orthop Trauma 2007;21:523-9. 25. Wendelboe AM, Hegmann KT, Gren LH, Alder SC, White GL Jr, Lyon JL. Associations between body-mass index and surgery for rotator cuff tendinitis. J Bone Joint Surg Am 2004;86:743-7. 26. Wilson PW. Established risk factors and coronary artery disease: the Framingham Study. Am J Hypertens 1994;7(Pt 2):7S-12. 27. World Health Organization. Technical report series 894. Obesity: preventing and Managing the Global Epidemic. Geneva: World Health Organization. Available from: URL: http://whqlibdoc.who.int/trs/ WHO_TRS_894_(part1).pdf. Accessed Nov 15, 2009. 28. Yusuf S, Hawken S, Ounpuu S, Dans T, Avezum A, Lanas F, et al. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study). Lancet 2004;364:937-52.