- Email: [email protected]
Arthroscopically assisted rotator cuff repair: Correlation of functional results with integrity of the cuff
Arthroscopy: The Journal of Arthroscopic and Related Surgery 10(1):54-60
Published by Raven Press, Ltd. © 1994ArthroscopyAssociationof North America
Arthroscopically Assisted Rotator Cuff Repair: Correlation of Functional Results with Integrity of the Cuff Stephen H. Liu, M.D., and Champ L. Baker, M.D.
Summary: Thirty-three patients (35 full-thickness rotator cuff tears) who underwent arthroscopically assisted mini open repair between June 1987 and January 1990 were evaluated for shoulder function and cuff integrity. The study population was composed of 19 women and 14 men with an average age of 63 years (range 35-76) and an average follow-up of 3.7 years (range 2.5-5.1). Functional results were obtained using the UCLA Shoulder Rating Scale. Integrity of the rotator cuff was established by shoulder arthrography at a minimum 2 years postoperatively. UCLA Shoulder Rating Scale showed 86% good/excellent results with 92% patient satisfaction. The shoulder arthrography showed 12 (34%) full-thickness tears, seven (20%) partial tears, and 16 (46%) no tears. The size of the arthrographic defect correlated well with the size of the intraoperative tear. Seventy percent of the large tears had follow-up full-thickness defects and 80% of the small tears had no defect. Eighty percent and 88% good/excellent functional results were achieved in pateints with fullthickness defects and no defects, respectively, without significant differences. The size of the tear at the time of the repair is a major determinant of the integrity of the cuff after repair. The integrity of the cuff at follow-up does not determine the functional outcome of the operated shoulder. Key Words: Rotator cuff Arthroscopic repair Rotator cuff Integrity--Functional result.
to better appreciate the s p e c t r u m of rotator cuff injury and to treat this lesion with a less invasive technique. Preliminary results of arthroscopically assisted mini open rotator cuff repair h a v e b e e n promising (1%20). In our recent study comparing open and arthroscopically assisted repairs, the results of arthroscopically assisted repair were found to be similar to those of the o p e n procedure (20). The importance of the integrity of the rotator cuff after open repairs has been studied by several investigators (12,14-16). H o w e v e r , the integrity o f the cuff after arthroscopically assisted repair has not been evaluated. Because m o r e arthroscopic rotator cuff surgeries are being p e r f o r m e d , it would be equally important to evaluate the integrity of the rotator cuff after such a p r o c e d u r e . To our knowledge, there has not been a published report on the integrity of the rotator cuff after arthroscopically assisted mini open repair. The p u r p o s e of this study
The diagnosis and m a n a g e m e n t of rotator cuff disease has b e e n well described in the literature (18). M a n y h a v e obtained good results with rotator cuff repairs (1-6). H o w e v e r , others h a v e achieved similar results with acromioplasty, without repairing the tear (7-11). Results of the studies on the integrity of the rotator cuff after open repair have been controversial and have left unsettled the correlation b e t w e e n the integrity of the cuff and the shoulder function (12-16). Recent i m p r o v e m e n t and refinement of shoulder a r t h r o s c o p y have enabled us From the Department of Orthopaedic Surgery, Sports Medicine Section, University of California at Los Angeles School of Medicine (S.H.L.), and Hughston Orthopaedic Clinic (C.L.B.), Columbus, Georgia, U.S.A. Address correspondence and reprint requests to Dr. Stephen H. Liu, Department of Orthopaedic Surgery, UCLA School of Medicine, CHS 76-119, 10833 LeConte Avenue, Los Angeles, CA 90024-6902, U.S.A. 54
ROTATOR CUFF REPAIR
was to correlate the functional outcome of arthroscopically assisted rotator cuff repair with the integrity of the repaired cuff. Investigations into the integrity of the rotator cuff have included the use of ultrasonography, arthrography, and magnetic resonance imaging (MRI). The effectiveness of ultrasound in evaluating rotator cuff lesions has been reported (21-25). Ultrasound study is noninvasive and inexpensive; however, its usefulness depends strongly on the training, experience, and expertise of the person performing the test. Results of MRI studies of a postoperative rotator cuff have been inconsistent (26,27). Furthermore, it adds cost and complexity in comparison with arthrography. Shoulder arthrography is highly accurate in diagnosing full-thickness rotator cuff tears. We selected arthrography because it is quick and reliable, with minimum morbidity and low cost. In addition, we have had extensive experience and success with this procedure. MATERIALS AND METHOD Forty-eight arthroscopically assisted mini open rotator cuff repairs for chronic rotator cuff tears were performed on 45 patients between June 1987 and January 1990. We were able to evaluate 33 patients having 35 arthroscopically assisted rotator cuff repairs with a minimum 2-year follow-up. Indications for surgery were persistent pain and/or functional disability without improvement after conservative treatment for >6 months. All patients had full-thickness rotator cuff tears documented by preoperative arthrography. Thirteen patients who had come from out of state did not return for followup. All operations were performed by, or under the direct supervision of, one of the authors. Our surgical technique for arthroscopically assisted repair has been described previously (20). Intraoperative rotator cuff tear size was recorded as small (< 1 cm), moderate (1-3 cm), large (3-5 cm), or massive (>5 cm). All patients were followed at least 2 years after surgery. The average follow-up was 3.7 years (range 2.5-5.1). The study population was composed of 19 women and 14 men. The dominant extremity was involved in 70% of the patients. The average age was 63 (range 35-76). Postoperative management and rehabilitation were similar for all patients. After surgery, the operated shoulder was immobilized in a 45 ° abduction splint for 3 weeks. Modalities such as pendulum exercises, isometrics,
55
and protected passive range of motion (ROM) were initiated during that first 3-week period. Sling immobilization was continued for the next 3 weeks. Strength and active ROM was initiated at 6 weeks. At follow-up, single-contrast shoulder arthrography was performed on all patients. Functional evaluations using the UCLA Shoulder Rating Scale and additional active ROM and strength (forward flexion, abduction, external rotation) were performed by the same physical therapist both pre- and postsurgery. Time for the patient's return to full participation of his or her previous activity was recorded in months. Results of the follow-up shoulder arthrograms, return to previous activity, and the UCLA shoulder functional ratings were compared. The UCLA shoulder rating scale The UCLA Shoulder Rating Scale rating for pain and function of the shoulder was used to grade each patient before operation and at follow-up. According to this schema, pain and function are each rated on a scale of 1-100 points, with 1 point being the worst and 10 points the best score; the range of active forward flexion and strength in forward flexion are each scored from 0 to 5 points. The maximum score that can be achieved is 35 points. Excellent is 34 to 35 points, good is 29 to 33, and poor is <29.
Surgical technique A standard shoulder arthroscopy is performed with the patient in the lateral decubitus position. Initial evaluation included inspection of the glenohumeral joint, treating any pathology in addition to the rotator cuff tear. The extent of the tear is noted and graded as <1 cm, 1-3 cm, 3-5 cm, or >5 cm. After identification of the tear, the cuff edges are debrided. Next, bursoscopy is performed for evaluation of the subacromial space, as well as subacromial decompression with release of the coracoacromial ligament. A burr is used to create a bony trough on the greater tuberosity. Multiple 0 PDS sutures (polydioxanone; Ethicon, Somerville, NJ) are placed in the leading edges of the torn tendon facilitated by the suture passer via the anterolateral portal, Traction is removed, and the arm is placed horizontally on the patient's side. Rotator cuff repair is now performed through a small transverse incision incorporating the anterolateral portal incision. The deltoid is split bluntly in the longitudinal direction following the previous defect created by the anterolateral portal. The deltoid should only be Arthroscopy, Vol. t0, No. 1, 1994
56
S. H. L I U A N D C. L. BAKER
F I G . 1. A r t h r o s c o p i c rotator cuff repair. U C L A s c o r e s pre- a n d p o s t r e p a i r .
I
I POST-OP
PRE-OP
separated from the acromion by no more than 5 cm to prevent injury to the axillary nerve. At this point, the leading edge of the torn cuff can be visualized easily and pulled laterally by the previously placed PDS sutures. Once the cuff can be brought to the bone trough without much tension, three adjacent bone holes are made with a towel clip. The PDS sutures on the cuff are passed through the bone hole facilitated by the suture passer. The rotator cuff is then pulled with mild tension onto the bone trough. All sutures are tied. Since 1990, we have used Mitek (Mitek Surgical Products, Norwood, MA, U.S.A.) anchoring sutures to facilitate our repair, without transosseous attachment. After the cuff repair, 40 ml saline fluid is injected into the shoulder to assess the intraoperative integrity of the repair.
rity. Patients' overall scores on the UCLA Scale, as well as their scores on the subcategories of pain, function, ROM, strength, and satisfaction were compared for pre- and postsurgery, as well as with the arthrogram defect (normal, partial, full). Additional ROM and strength in active forward flexion, abduction, and external rotation also were compared, as was the months to return to previous activity. A nonparametric (Kruskal-WaUis H) test was used for these comparisons. Tests were also run to determine if a relationship existed between the defect as observed on an arthrogram and the size of the tear as observed at surgery. These were conducted for the overall group. A ×2 test with 4 dfwas performed to test for these possible relationships.
DATA ANALYSIS
Postoperative arthrography was performed at >/2 years follow-up. All arthrograms were performed by the senior radiologist at our institution. Under fluoroscopic control, a 22-gauge spinal needle was used to inject 5 ml of contrast material into the an-
Multiple comparisons were made to determine if any differences could be observed between the shoulder function scores and the rotator cuff integ-
SHOULDER ARTHROGRAPHY
F I G . 2. Pre- a n d p o s t o p e r a t i v e r a t i n g s for pain and function.
[] PRE-OP [] POST-OP
u [
I PAIN
Arthroscopy, Vol. 10, No. 1, 1994
[
[ FUNCTION
ROTATOR CUFF REPAIR
[] PRE-OP [] POST-OP
R.O.M.
57
FIG. 3. Pre- and postoperative ratings for range of motion and strength.
STRENGTH
terior shoulder joint. The arthrograms were made as single-contrast studies, and radiographs were made both before and after exercise of the shoulder. Extravasation of contrast medium into the subacromial-subdeltoid space was considered diagnostic of a full-thickness tear, whereas contrast medium extending into but not through the rotator cuff was considered diagnostic of a partial-thickness tear.
duction 45 (20-54). On the average, all patients improved at least one grade on manual strength testing after surgery. The average strength grading ranged from 3.0 of external rotation, 3.2 of forward flexion, and 2.95 of abduction preoperatively to 4.3 of external rotation, 4.5 of forward flexion, and 4.75 of abduction at follow-up. FOLLOW-UP ARTHROGRAPHY
RESULTS All patients improved after surgery. The overall scores and the subcategory scores are shown on Figs. 1-3. The mean patient postoperative scores were improved for all subcategories of pain, ROM, function, and strength. Eighty-five percent of the patients achieved excellent/good results and 93% of the patients were satisfied (Fig. 4). Additional ROM and strength were measured in forward flexion, abduction, and external rotation (Figs. 5 and 6). The average gain in external rotation was 13° (range 738), in forward flexion 47 (range 15-55), and in ab-
Of the 35 rotator cuffs studied, 19 (54%) had abnormal arthrograms. Twelve (34%) had fullthickness tears, seven (20%) had partial tears, and 16 (46%) had no tears. Results of the tear size of the rotator cuff repaired and the follow-up arthrographic defects are shown in Table 1. A large number of recurrent defects were found in large (3- to 5-cm) tears that were repaired, and the majority of the small tears had normal study results. Eighty percent of the large tears had abnormal arthrogram results (70% full thickness and 10% partial), and 80% of the small tears had no defect.
FIG. 4. A r t h r o s c o p i c a l l y a s s i s t e d rotator tendon cuff repair according to the U C L A Rating Scale. 15"W
EXCELLENT/ GOOD
FAIR / POOR
SATISFACTORY
Arthroscopy, Vol. 10, No. 1, 1994
58
S. H. L1U A N D C. L. BAKER
3 PRE-OP
FIG. 5. Pre- and p o s t o p e r a t i v e ratings for strength in flexion, abduction, and external
rotation.
2
POST-OP
FLEXION
ABDUCTION
Correlation of arthrography defect with the functional results Results of integrity versus function are shown in Fig. 7. Eighty percent of the full-thickness defect group achieved good/excellent results, and the remaining 20% had fair results. Seventy-five percent of the partial defect group achieved good results, and 25% had fair results. Eighty-eight percent of the no defect group achieved good/excellent results, and 12% had fair results. None of the groups had poor results. There were no statistically significant differences among these groups. DISCUSSION The goal of this study was to evaluate shoulder function after arthroscopically assisted repairs and to correlate these findings with rotator cuff integrity. To our knowledge, there has not been a documentation of the integrity of the rotator cuff after 180-
EXTERNAL ROTATION
arthroscopically assisted mini open rotator cuff repair. Results of the integrity of the rotator cuff after standard open repair have been controversial. Lundberg (16) used single-contrast arthrography to study 21 shoulders in which the cuff had been repaired, and he found that contrast medium leaked from the cuff in seven (33%). Shoulders without leakage functioned better than those that had leakage. Harryman (15) used ultrasound to study 105 rotator cuff repairs at an average of 5 years followup. Overall, he found 35% abnormality in the follow-up repaired rotator cuff. Tears involving just the supraspinatus tendon had 20% defect, and those involving more than the supraspinatus tendon had >50% defect. In addition, he noted that integrity of the rotator cuff was directly related to the shoulder functions. In contrast, Calvert (12), using doublecontrast arthrography, found that 18 of the 20 follow-up rotator cuffs repaired had persistent defect
t57
160140120[] PRE-OP
100-
[] POST-OP
805O
6040--200
I FLEXION
Arthroscopy, Vol. 10, No. 1, 1994
L ABDUCTION
EXTERNAL ROTATION
FIG. 6. Pre- and postoperative ratings for ROM in flexion, abduction, and external rotation.
R 0 TA TOR C U F F R E P A I R T A B L E 1. Distribution o f defects in relation to tear size o f rotator cuffs studied
Full Partial Normal Total
Small
Medium
Large
Total
2 1 9 12
2 4 4 10
8 2 3 13
12 7 16 35
in the cuff; however, 17 patients were asymptomatic and 15 had full ROM. Pettersson (13), using arthrography, found full or partial rotator cuff defects in 13 of 27 asymptomatic patients. Packer (21) noted that defects in the repair are not problematic and that the goal of repair is to restore the continuity of the muscle-tendon unit to provide a mechanically effective rotator cuff. This study showed that 54% of follow-up arthrogram results were abnormal, with 34% full thickness defect. These findings correlated well with those of Lundberg and Harryman's open repaired results. The full-thickness defect, partial-thickness defect, and no defect groups each achieved similar excellent/good results, 80% versus 75% versus 88%, respectively, with no significant differences between these groups. Thus, no correlation was established between the integrity of the rotator cuff after arthroscopically assisted repair and the followup functional outcome. These results correlated well with the open repair study of Calvert that good shoulder function can occur with a full-thickness defect in the rotator cuff. Additional results of this study showed that the size of the tear correlated well with the follow-up arthrogram defect. Seventy percent of the large tears had follow-up full-thickness defects, and 80% of the small tears had no defect. These data corre-
80% -
59
lated well with Harryman's study that there is a direct relationship between the size of the tear repaired and the follow-up rotator cuff defect. Previous reports on arthroscopically assisted rotator cuff repairs have shown results similar to those of the standard open procedure (17-20). Recent studies have shown that both procedures are equally effective; however, the arthroscopically assisted group had less postoperative morbidity and had return to full work earlier than the open repaired group (20,28). The results of this study demonstrated that the functional outcome and integrity of the rotator cuff after arthroscopically assisted repair are similar to those of open repair. Furthermore, the integrity of the rotator cuff at follow-up does not determine the functional outcome of the repair. These findings may indicate that water-tight closure repair may not be essential in arthroscopically assisted rotator cuff repair. The residual defect in the repair may not be problematic as long as the continuity of the muscle-tendon unit is restored to provide a mechanically effective rotator cuff. In contrast, results from this study may also imply that rotator cuff tears need not be repaired. However, the reader must be cautioned that just because the results of the integrity and function of the cuff do not correlate, it does not indicate that an unrepaired rotator cuff will result in good shoulder function. Further investigations on shoulder function after unrepaired or irreparable massive rotator cuff tears should provide further information on the treatment of these tears. CONCLUSIONS The results of arthroscopically assisted rotator cuff repair are favorable and comparable with those
75% 63%
60% FIG. 7. Ratings for integrity versus function in the full-thickness, partial, and no defect groups as seen by arthrography.
~,~,~
~
~
~!
[] GOOD
*'~ 40%-
~~ ~
ion,
[] EXCELLENT i~NN
~
0%NORMAL
• FAIR
20¶
-oo, l~;~;~ ~ PARTIAL
1 FULL Arthroscopy, Vol. 10, No. 1, 1994
60
S. H. LIU AND C. L. BAKER
of the standard open repair technique, especially with small and moderate sized tears. The size of the tear at the time of repair is a major determinant of the integrity of the rotator cuff after repair. The integrity of the rotator cuff at follow-up does not determine the functional outcome of the operated shoulder. REFERENCES 1. Hawkins RJ, Misamore GW, Hobeika PE. Surgery for fullthickness rotator cuff tears. J Bone Joint Surg [Am] 1985 ;6: 134%55. 2. Ellman H, Hanker G, Bayer M. Repair of the rotator cuff: end results study of factors influencing reconstruction. J Bone Joint Surg [Am] 1986;68:1136--44. 3. Cofield RH. Current concepts review: rotator cuffdisease of the shoulder. J Bone Joint Surg [Am] 1985;67:974-9. 4. Cofield RH, Hoffmeyer P, Lanzer WH. Surgical repair of chronic rotator cuff tears. Orthop Trans 1990;14:251-2. 5. Neviaser JS, Neviaser RJ, Neviaser TJ. The repair of chronic massive ruptures of the rotator cuff of the shoulder by use of a freeze dried rotator cuff. J Bone Joint Surg [Am] 1978;60:681-4. 6. Bigliani LU, Mcllveen S J, Cordasco FA, et al. Operative repair of massive rotator cuff tears: long term results. Orthop Trans 1990;14:251. 7. Burkhart SS. Arthroscopic treatment of massive rotator cuff tears: clinical results and biomechanical rationale. Orthop Trans 1990;14:173. 8. Rockwood CA, Bnrkhead WZ. Management of patient with massive rotator cuff defect acromioplasty and rotator cuff debridement. Orthop Trans 1988; 12:190-1. 9. Ellman H. Arthroscopic subacromial decompression: analysis of 1-3 Year results. Arthroscopy 1987;3:73. 10. Ogilvie-Harris DJ, Demaziere A. Arthroscopic debfidement versus open repair for rotator cuff tears 1 to 4 cm in size: a prospective cohort study. J Bone Joint Surgery [Brl 1993; 75:416--20. 11. Burkhart SS. Consideration for debridement in full thickness rotator cuff tears. Presented at the Arthroscopy Association Meeting, Specialty Day, San Francisco, California, February 21, 1993. 12. Calvert PT, Packer MP, Stoker DJ, e t a l . Arthrography of the shoulder after operative repair of torn rotator cuff. J Bone Joint Surg [Br] 1986;68:147-50.
Arthroscopy, Vol. 10, No. 1, 1994
13. Pettersson G. Rupture of the tendon aponeurosis of the shoulder joint in anterior/inferior dislocation. Acta Chir Scand 1942;77(suppl): 1-184. 14. Bark DL, Karasick D, Kurt A, e t a l . Rotator cuff tears: a prospective comparison of MRI, arthrography, sonography, and surgery. A JR 1989;153:87-92. 15. Harryman DT, Mack LA, Wayne KA, et al. Repairs of the rotator cuff: correlation of functional results with integrity of the cuff. J Bone Joint Surg [Am] 1991;73:92-9. 16. Lundberg BJ. The correlation of clinical varialion of operative findings and prognosis in rotator cuff rupture. In: Bayley I, Kessel L, eds. Shoulder surgery. Berlin: Springer Verlag, 1982:35-8. 17. Levy HJ, Uribe JW, Delaney LG. Arthroscopic assisted rotator cuff repair: preliminary results. Arthroscopy 1990;6: 55-60, 18. Warner JP, Altchek DW, Warren RF. Arthroscopic management of rotator cuff tears with emphasis on the throwing athlete. Oper Tech Orthop 1991;1:235-9. 19. Seltzer DG, Uribe JW, Posada A, Gaines R. Arthroscopic assisted rotator cuff repair: two year follow-up. Presented at the AOSS meeting, San Francisco, February 1992. 20. Liu SH, Baker CL. Comparison of open and artbroscopically assisted rotator cuff repair. Presented at the AAOS meeting, San Francisco, February 1993. 21. Packer MP, Calvert PT, Bailey JIL, et al. Operative treatment of chronic ruptures of the rotator cuff of the shoulder. J Bone Joint Surg [Br] 1983;65:171-5. 22. Bretzke CA, Crass JR, Craig EV, etal. Ultrasonography of rotator cuff: normal and pathologic anatomy. Invest Radiol 1985;20:311-5. 23. Hodler J, Fritz CJ, Terrier F, etal. Rotator cuff tears: correlation of the sonographic and surgical findings. Radiology 1988;I69:791-4. 24. Middleton DW, Reinus WR, Totty WG, et al. Ultrasonographic evaluation of rotator cuff and biceps tendon. J Bone Joint Surg [Am] 68:440-50. 25. Mack LA, Nyberg D, Matsen FA, III. Sonography of the postoperative shoulder. Am J Radiol 1988;150:108%93. 26. Crass JR, CraigEV. Noninvasiveimagingoftherotator cuff. Orthopedics 1988;11:57-64. 27. Kieft G J, Bloem JL, Rozing PM, Obermann WR. Rotator cuff impingement syndrome: MR imaging. Radiology 1988; 166:2ll-4. 28. Weber SC, Schaefer R. Mini open vs traditional open technique in the management of tears of the rotator cuff. Presented at the AOSM and JOSSM Trans-Pacific Meeting, Maui, Hawaii, March 1993.
Published by Raven Press, Ltd. © 1994ArthroscopyAssociationof North America
Arthroscopically Assisted Rotator Cuff Repair: Correlation of Functional Results with Integrity of the Cuff Stephen H. Liu, M.D., and Champ L. Baker, M.D.
Summary: Thirty-three patients (35 full-thickness rotator cuff tears) who underwent arthroscopically assisted mini open repair between June 1987 and January 1990 were evaluated for shoulder function and cuff integrity. The study population was composed of 19 women and 14 men with an average age of 63 years (range 35-76) and an average follow-up of 3.7 years (range 2.5-5.1). Functional results were obtained using the UCLA Shoulder Rating Scale. Integrity of the rotator cuff was established by shoulder arthrography at a minimum 2 years postoperatively. UCLA Shoulder Rating Scale showed 86% good/excellent results with 92% patient satisfaction. The shoulder arthrography showed 12 (34%) full-thickness tears, seven (20%) partial tears, and 16 (46%) no tears. The size of the arthrographic defect correlated well with the size of the intraoperative tear. Seventy percent of the large tears had follow-up full-thickness defects and 80% of the small tears had no defect. Eighty percent and 88% good/excellent functional results were achieved in pateints with fullthickness defects and no defects, respectively, without significant differences. The size of the tear at the time of the repair is a major determinant of the integrity of the cuff after repair. The integrity of the cuff at follow-up does not determine the functional outcome of the operated shoulder. Key Words: Rotator cuff Arthroscopic repair Rotator cuff Integrity--Functional result.
to better appreciate the s p e c t r u m of rotator cuff injury and to treat this lesion with a less invasive technique. Preliminary results of arthroscopically assisted mini open rotator cuff repair h a v e b e e n promising (1%20). In our recent study comparing open and arthroscopically assisted repairs, the results of arthroscopically assisted repair were found to be similar to those of the o p e n procedure (20). The importance of the integrity of the rotator cuff after open repairs has been studied by several investigators (12,14-16). H o w e v e r , the integrity o f the cuff after arthroscopically assisted repair has not been evaluated. Because m o r e arthroscopic rotator cuff surgeries are being p e r f o r m e d , it would be equally important to evaluate the integrity of the rotator cuff after such a p r o c e d u r e . To our knowledge, there has not been a published report on the integrity of the rotator cuff after arthroscopically assisted mini open repair. The p u r p o s e of this study
The diagnosis and m a n a g e m e n t of rotator cuff disease has b e e n well described in the literature (18). M a n y h a v e obtained good results with rotator cuff repairs (1-6). H o w e v e r , others h a v e achieved similar results with acromioplasty, without repairing the tear (7-11). Results of the studies on the integrity of the rotator cuff after open repair have been controversial and have left unsettled the correlation b e t w e e n the integrity of the cuff and the shoulder function (12-16). Recent i m p r o v e m e n t and refinement of shoulder a r t h r o s c o p y have enabled us From the Department of Orthopaedic Surgery, Sports Medicine Section, University of California at Los Angeles School of Medicine (S.H.L.), and Hughston Orthopaedic Clinic (C.L.B.), Columbus, Georgia, U.S.A. Address correspondence and reprint requests to Dr. Stephen H. Liu, Department of Orthopaedic Surgery, UCLA School of Medicine, CHS 76-119, 10833 LeConte Avenue, Los Angeles, CA 90024-6902, U.S.A. 54
ROTATOR CUFF REPAIR
was to correlate the functional outcome of arthroscopically assisted rotator cuff repair with the integrity of the repaired cuff. Investigations into the integrity of the rotator cuff have included the use of ultrasonography, arthrography, and magnetic resonance imaging (MRI). The effectiveness of ultrasound in evaluating rotator cuff lesions has been reported (21-25). Ultrasound study is noninvasive and inexpensive; however, its usefulness depends strongly on the training, experience, and expertise of the person performing the test. Results of MRI studies of a postoperative rotator cuff have been inconsistent (26,27). Furthermore, it adds cost and complexity in comparison with arthrography. Shoulder arthrography is highly accurate in diagnosing full-thickness rotator cuff tears. We selected arthrography because it is quick and reliable, with minimum morbidity and low cost. In addition, we have had extensive experience and success with this procedure. MATERIALS AND METHOD Forty-eight arthroscopically assisted mini open rotator cuff repairs for chronic rotator cuff tears were performed on 45 patients between June 1987 and January 1990. We were able to evaluate 33 patients having 35 arthroscopically assisted rotator cuff repairs with a minimum 2-year follow-up. Indications for surgery were persistent pain and/or functional disability without improvement after conservative treatment for >6 months. All patients had full-thickness rotator cuff tears documented by preoperative arthrography. Thirteen patients who had come from out of state did not return for followup. All operations were performed by, or under the direct supervision of, one of the authors. Our surgical technique for arthroscopically assisted repair has been described previously (20). Intraoperative rotator cuff tear size was recorded as small (< 1 cm), moderate (1-3 cm), large (3-5 cm), or massive (>5 cm). All patients were followed at least 2 years after surgery. The average follow-up was 3.7 years (range 2.5-5.1). The study population was composed of 19 women and 14 men. The dominant extremity was involved in 70% of the patients. The average age was 63 (range 35-76). Postoperative management and rehabilitation were similar for all patients. After surgery, the operated shoulder was immobilized in a 45 ° abduction splint for 3 weeks. Modalities such as pendulum exercises, isometrics,
55
and protected passive range of motion (ROM) were initiated during that first 3-week period. Sling immobilization was continued for the next 3 weeks. Strength and active ROM was initiated at 6 weeks. At follow-up, single-contrast shoulder arthrography was performed on all patients. Functional evaluations using the UCLA Shoulder Rating Scale and additional active ROM and strength (forward flexion, abduction, external rotation) were performed by the same physical therapist both pre- and postsurgery. Time for the patient's return to full participation of his or her previous activity was recorded in months. Results of the follow-up shoulder arthrograms, return to previous activity, and the UCLA shoulder functional ratings were compared. The UCLA shoulder rating scale The UCLA Shoulder Rating Scale rating for pain and function of the shoulder was used to grade each patient before operation and at follow-up. According to this schema, pain and function are each rated on a scale of 1-100 points, with 1 point being the worst and 10 points the best score; the range of active forward flexion and strength in forward flexion are each scored from 0 to 5 points. The maximum score that can be achieved is 35 points. Excellent is 34 to 35 points, good is 29 to 33, and poor is <29.
Surgical technique A standard shoulder arthroscopy is performed with the patient in the lateral decubitus position. Initial evaluation included inspection of the glenohumeral joint, treating any pathology in addition to the rotator cuff tear. The extent of the tear is noted and graded as <1 cm, 1-3 cm, 3-5 cm, or >5 cm. After identification of the tear, the cuff edges are debrided. Next, bursoscopy is performed for evaluation of the subacromial space, as well as subacromial decompression with release of the coracoacromial ligament. A burr is used to create a bony trough on the greater tuberosity. Multiple 0 PDS sutures (polydioxanone; Ethicon, Somerville, NJ) are placed in the leading edges of the torn tendon facilitated by the suture passer via the anterolateral portal, Traction is removed, and the arm is placed horizontally on the patient's side. Rotator cuff repair is now performed through a small transverse incision incorporating the anterolateral portal incision. The deltoid is split bluntly in the longitudinal direction following the previous defect created by the anterolateral portal. The deltoid should only be Arthroscopy, Vol. t0, No. 1, 1994
56
S. H. L I U A N D C. L. BAKER
F I G . 1. A r t h r o s c o p i c rotator cuff repair. U C L A s c o r e s pre- a n d p o s t r e p a i r .
I
I POST-OP
PRE-OP
separated from the acromion by no more than 5 cm to prevent injury to the axillary nerve. At this point, the leading edge of the torn cuff can be visualized easily and pulled laterally by the previously placed PDS sutures. Once the cuff can be brought to the bone trough without much tension, three adjacent bone holes are made with a towel clip. The PDS sutures on the cuff are passed through the bone hole facilitated by the suture passer. The rotator cuff is then pulled with mild tension onto the bone trough. All sutures are tied. Since 1990, we have used Mitek (Mitek Surgical Products, Norwood, MA, U.S.A.) anchoring sutures to facilitate our repair, without transosseous attachment. After the cuff repair, 40 ml saline fluid is injected into the shoulder to assess the intraoperative integrity of the repair.
rity. Patients' overall scores on the UCLA Scale, as well as their scores on the subcategories of pain, function, ROM, strength, and satisfaction were compared for pre- and postsurgery, as well as with the arthrogram defect (normal, partial, full). Additional ROM and strength in active forward flexion, abduction, and external rotation also were compared, as was the months to return to previous activity. A nonparametric (Kruskal-WaUis H) test was used for these comparisons. Tests were also run to determine if a relationship existed between the defect as observed on an arthrogram and the size of the tear as observed at surgery. These were conducted for the overall group. A ×2 test with 4 dfwas performed to test for these possible relationships.
DATA ANALYSIS
Postoperative arthrography was performed at >/2 years follow-up. All arthrograms were performed by the senior radiologist at our institution. Under fluoroscopic control, a 22-gauge spinal needle was used to inject 5 ml of contrast material into the an-
Multiple comparisons were made to determine if any differences could be observed between the shoulder function scores and the rotator cuff integ-
SHOULDER ARTHROGRAPHY
F I G . 2. Pre- a n d p o s t o p e r a t i v e r a t i n g s for pain and function.
[] PRE-OP [] POST-OP
u [
I PAIN
Arthroscopy, Vol. 10, No. 1, 1994
[
[ FUNCTION
ROTATOR CUFF REPAIR
[] PRE-OP [] POST-OP
R.O.M.
57
FIG. 3. Pre- and postoperative ratings for range of motion and strength.
STRENGTH
terior shoulder joint. The arthrograms were made as single-contrast studies, and radiographs were made both before and after exercise of the shoulder. Extravasation of contrast medium into the subacromial-subdeltoid space was considered diagnostic of a full-thickness tear, whereas contrast medium extending into but not through the rotator cuff was considered diagnostic of a partial-thickness tear.
duction 45 (20-54). On the average, all patients improved at least one grade on manual strength testing after surgery. The average strength grading ranged from 3.0 of external rotation, 3.2 of forward flexion, and 2.95 of abduction preoperatively to 4.3 of external rotation, 4.5 of forward flexion, and 4.75 of abduction at follow-up. FOLLOW-UP ARTHROGRAPHY
RESULTS All patients improved after surgery. The overall scores and the subcategory scores are shown on Figs. 1-3. The mean patient postoperative scores were improved for all subcategories of pain, ROM, function, and strength. Eighty-five percent of the patients achieved excellent/good results and 93% of the patients were satisfied (Fig. 4). Additional ROM and strength were measured in forward flexion, abduction, and external rotation (Figs. 5 and 6). The average gain in external rotation was 13° (range 738), in forward flexion 47 (range 15-55), and in ab-
Of the 35 rotator cuffs studied, 19 (54%) had abnormal arthrograms. Twelve (34%) had fullthickness tears, seven (20%) had partial tears, and 16 (46%) had no tears. Results of the tear size of the rotator cuff repaired and the follow-up arthrographic defects are shown in Table 1. A large number of recurrent defects were found in large (3- to 5-cm) tears that were repaired, and the majority of the small tears had normal study results. Eighty percent of the large tears had abnormal arthrogram results (70% full thickness and 10% partial), and 80% of the small tears had no defect.
FIG. 4. A r t h r o s c o p i c a l l y a s s i s t e d rotator tendon cuff repair according to the U C L A Rating Scale. 15"W
EXCELLENT/ GOOD
FAIR / POOR
SATISFACTORY
Arthroscopy, Vol. 10, No. 1, 1994
58
S. H. L1U A N D C. L. BAKER
3 PRE-OP
FIG. 5. Pre- and p o s t o p e r a t i v e ratings for strength in flexion, abduction, and external
rotation.
2
POST-OP
FLEXION
ABDUCTION
Correlation of arthrography defect with the functional results Results of integrity versus function are shown in Fig. 7. Eighty percent of the full-thickness defect group achieved good/excellent results, and the remaining 20% had fair results. Seventy-five percent of the partial defect group achieved good results, and 25% had fair results. Eighty-eight percent of the no defect group achieved good/excellent results, and 12% had fair results. None of the groups had poor results. There were no statistically significant differences among these groups. DISCUSSION The goal of this study was to evaluate shoulder function after arthroscopically assisted repairs and to correlate these findings with rotator cuff integrity. To our knowledge, there has not been a documentation of the integrity of the rotator cuff after 180-
EXTERNAL ROTATION
arthroscopically assisted mini open rotator cuff repair. Results of the integrity of the rotator cuff after standard open repair have been controversial. Lundberg (16) used single-contrast arthrography to study 21 shoulders in which the cuff had been repaired, and he found that contrast medium leaked from the cuff in seven (33%). Shoulders without leakage functioned better than those that had leakage. Harryman (15) used ultrasound to study 105 rotator cuff repairs at an average of 5 years followup. Overall, he found 35% abnormality in the follow-up repaired rotator cuff. Tears involving just the supraspinatus tendon had 20% defect, and those involving more than the supraspinatus tendon had >50% defect. In addition, he noted that integrity of the rotator cuff was directly related to the shoulder functions. In contrast, Calvert (12), using doublecontrast arthrography, found that 18 of the 20 follow-up rotator cuffs repaired had persistent defect
t57
160140120[] PRE-OP
100-
[] POST-OP
805O
6040--200
I FLEXION
Arthroscopy, Vol. 10, No. 1, 1994
L ABDUCTION
EXTERNAL ROTATION
FIG. 6. Pre- and postoperative ratings for ROM in flexion, abduction, and external rotation.
R 0 TA TOR C U F F R E P A I R T A B L E 1. Distribution o f defects in relation to tear size o f rotator cuffs studied
Full Partial Normal Total
Small
Medium
Large
Total
2 1 9 12
2 4 4 10
8 2 3 13
12 7 16 35
in the cuff; however, 17 patients were asymptomatic and 15 had full ROM. Pettersson (13), using arthrography, found full or partial rotator cuff defects in 13 of 27 asymptomatic patients. Packer (21) noted that defects in the repair are not problematic and that the goal of repair is to restore the continuity of the muscle-tendon unit to provide a mechanically effective rotator cuff. This study showed that 54% of follow-up arthrogram results were abnormal, with 34% full thickness defect. These findings correlated well with those of Lundberg and Harryman's open repaired results. The full-thickness defect, partial-thickness defect, and no defect groups each achieved similar excellent/good results, 80% versus 75% versus 88%, respectively, with no significant differences between these groups. Thus, no correlation was established between the integrity of the rotator cuff after arthroscopically assisted repair and the followup functional outcome. These results correlated well with the open repair study of Calvert that good shoulder function can occur with a full-thickness defect in the rotator cuff. Additional results of this study showed that the size of the tear correlated well with the follow-up arthrogram defect. Seventy percent of the large tears had follow-up full-thickness defects, and 80% of the small tears had no defect. These data corre-
80% -
59
lated well with Harryman's study that there is a direct relationship between the size of the tear repaired and the follow-up rotator cuff defect. Previous reports on arthroscopically assisted rotator cuff repairs have shown results similar to those of the standard open procedure (17-20). Recent studies have shown that both procedures are equally effective; however, the arthroscopically assisted group had less postoperative morbidity and had return to full work earlier than the open repaired group (20,28). The results of this study demonstrated that the functional outcome and integrity of the rotator cuff after arthroscopically assisted repair are similar to those of open repair. Furthermore, the integrity of the rotator cuff at follow-up does not determine the functional outcome of the repair. These findings may indicate that water-tight closure repair may not be essential in arthroscopically assisted rotator cuff repair. The residual defect in the repair may not be problematic as long as the continuity of the muscle-tendon unit is restored to provide a mechanically effective rotator cuff. In contrast, results from this study may also imply that rotator cuff tears need not be repaired. However, the reader must be cautioned that just because the results of the integrity and function of the cuff do not correlate, it does not indicate that an unrepaired rotator cuff will result in good shoulder function. Further investigations on shoulder function after unrepaired or irreparable massive rotator cuff tears should provide further information on the treatment of these tears. CONCLUSIONS The results of arthroscopically assisted rotator cuff repair are favorable and comparable with those
75% 63%
60% FIG. 7. Ratings for integrity versus function in the full-thickness, partial, and no defect groups as seen by arthrography.
~,~,~
~
~
~!
[] GOOD
*'~ 40%-
~~ ~
ion,
[] EXCELLENT i~NN
~
0%NORMAL
• FAIR
20¶
-oo, l~;~;~ ~ PARTIAL
1 FULL Arthroscopy, Vol. 10, No. 1, 1994
60
S. H. LIU AND C. L. BAKER
of the standard open repair technique, especially with small and moderate sized tears. The size of the tear at the time of repair is a major determinant of the integrity of the rotator cuff after repair. The integrity of the rotator cuff at follow-up does not determine the functional outcome of the operated shoulder. REFERENCES 1. Hawkins RJ, Misamore GW, Hobeika PE. Surgery for fullthickness rotator cuff tears. J Bone Joint Surg [Am] 1985 ;6: 134%55. 2. Ellman H, Hanker G, Bayer M. Repair of the rotator cuff: end results study of factors influencing reconstruction. J Bone Joint Surg [Am] 1986;68:1136--44. 3. Cofield RH. Current concepts review: rotator cuffdisease of the shoulder. J Bone Joint Surg [Am] 1985;67:974-9. 4. Cofield RH, Hoffmeyer P, Lanzer WH. Surgical repair of chronic rotator cuff tears. Orthop Trans 1990;14:251-2. 5. Neviaser JS, Neviaser RJ, Neviaser TJ. The repair of chronic massive ruptures of the rotator cuff of the shoulder by use of a freeze dried rotator cuff. J Bone Joint Surg [Am] 1978;60:681-4. 6. Bigliani LU, Mcllveen S J, Cordasco FA, et al. Operative repair of massive rotator cuff tears: long term results. Orthop Trans 1990;14:251. 7. Burkhart SS. Arthroscopic treatment of massive rotator cuff tears: clinical results and biomechanical rationale. Orthop Trans 1990;14:173. 8. Rockwood CA, Bnrkhead WZ. Management of patient with massive rotator cuff defect acromioplasty and rotator cuff debridement. Orthop Trans 1988; 12:190-1. 9. Ellman H. Arthroscopic subacromial decompression: analysis of 1-3 Year results. Arthroscopy 1987;3:73. 10. Ogilvie-Harris DJ, Demaziere A. Arthroscopic debfidement versus open repair for rotator cuff tears 1 to 4 cm in size: a prospective cohort study. J Bone Joint Surgery [Brl 1993; 75:416--20. 11. Burkhart SS. Consideration for debridement in full thickness rotator cuff tears. Presented at the Arthroscopy Association Meeting, Specialty Day, San Francisco, California, February 21, 1993. 12. Calvert PT, Packer MP, Stoker DJ, e t a l . Arthrography of the shoulder after operative repair of torn rotator cuff. J Bone Joint Surg [Br] 1986;68:147-50.
Arthroscopy, Vol. 10, No. 1, 1994
13. Pettersson G. Rupture of the tendon aponeurosis of the shoulder joint in anterior/inferior dislocation. Acta Chir Scand 1942;77(suppl): 1-184. 14. Bark DL, Karasick D, Kurt A, e t a l . Rotator cuff tears: a prospective comparison of MRI, arthrography, sonography, and surgery. A JR 1989;153:87-92. 15. Harryman DT, Mack LA, Wayne KA, et al. Repairs of the rotator cuff: correlation of functional results with integrity of the cuff. J Bone Joint Surg [Am] 1991;73:92-9. 16. Lundberg BJ. The correlation of clinical varialion of operative findings and prognosis in rotator cuff rupture. In: Bayley I, Kessel L, eds. Shoulder surgery. Berlin: Springer Verlag, 1982:35-8. 17. Levy HJ, Uribe JW, Delaney LG. Arthroscopic assisted rotator cuff repair: preliminary results. Arthroscopy 1990;6: 55-60, 18. Warner JP, Altchek DW, Warren RF. Arthroscopic management of rotator cuff tears with emphasis on the throwing athlete. Oper Tech Orthop 1991;1:235-9. 19. Seltzer DG, Uribe JW, Posada A, Gaines R. Arthroscopic assisted rotator cuff repair: two year follow-up. Presented at the AOSS meeting, San Francisco, February 1992. 20. Liu SH, Baker CL. Comparison of open and artbroscopically assisted rotator cuff repair. Presented at the AAOS meeting, San Francisco, February 1993. 21. Packer MP, Calvert PT, Bailey JIL, et al. Operative treatment of chronic ruptures of the rotator cuff of the shoulder. J Bone Joint Surg [Br] 1983;65:171-5. 22. Bretzke CA, Crass JR, Craig EV, etal. Ultrasonography of rotator cuff: normal and pathologic anatomy. Invest Radiol 1985;20:311-5. 23. Hodler J, Fritz CJ, Terrier F, etal. Rotator cuff tears: correlation of the sonographic and surgical findings. Radiology 1988;I69:791-4. 24. Middleton DW, Reinus WR, Totty WG, et al. Ultrasonographic evaluation of rotator cuff and biceps tendon. J Bone Joint Surg [Am] 68:440-50. 25. Mack LA, Nyberg D, Matsen FA, III. Sonography of the postoperative shoulder. Am J Radiol 1988;150:108%93. 26. Crass JR, CraigEV. Noninvasiveimagingoftherotator cuff. Orthopedics 1988;11:57-64. 27. Kieft G J, Bloem JL, Rozing PM, Obermann WR. Rotator cuff impingement syndrome: MR imaging. Radiology 1988; 166:2ll-4. 28. Weber SC, Schaefer R. Mini open vs traditional open technique in the management of tears of the rotator cuff. Presented at the AOSM and JOSSM Trans-Pacific Meeting, Maui, Hawaii, March 1993.