- Email: [email protected]
Collagen Production at the Edge of Ruptured Rotator Cuff Tendon is Correlated With Postoperative Cuff Integrity
Original Articles
Collagen Production at the Edge of Ruptured Rotator Cuff Tendon is Correlated With Postoperative Cuff Integrity Isao Shirachi, M.D., Masafumi Gotoh, M.D., Yasuhiro Mitsui, M.D., Tetsu Yamada, M.D., Kenjiro Nakama, M.D., Kazuyuki Kojima, M.D., Takahiro Okawa, M.D., Fujio Higuchi, M.D., and Kensei Nagata, M.D.
Purpose: The purpose was to evaluate the correlation between messenger RNA (mRNA) expression of collagen at the edge of the ruptured rotator cuff tendon and postoperative cuff integrity. Methods: The edge of the ruptured tendon was sampled during open rotator cuff surgery in 12 patients with full-thickness rotator cuff tears (mean age, 58.2 years). The mean period from symptom onset was 9.3 months (range, 1 to 36 months), and the mean tear size was 4.1 cm. As controls, rotator cuff tendons with no gross rupture were taken from 5 fresh cadavers. Production of type I and type III collagen was examined by real-time reverse transcription polymerase chain reaction. By use of magnetic resonance imaging, postoperative cuff integrity was evaluated based on the classification of Sugaya et al. and then scored, ranging from 5 points for type I to 1 point for type V. Results: Looking at the mRNA of type I and type III collagen in tendons, we found that the expression of mRNA for both collagen types in ruptured tendons was significantly greater than in control tendons (P ⫽ .0462 for type I collagen and P ⫽ .0306 for type III collagen). Correlating the mRNA of type I and type III collagen with repaired cuff integrity on postoperative magnetic resonance imaging, we found a close relation between expression of mRNA for both collagen types and postoperative rotator cuff integrity (r ⫽ 0.63 [P ⫽ .038] for type I collagen and r ⫽ 0.626 [P ⫽ .03] for type III collagen). Furthermore, expression of type I collagen mRNA showed a significant inverse correlation with the period from symptom onset (r ⫽ ⫺0.845, P ⬍ .0005). Conclusions: This study showed that expression of mRNA for type I and type III collagen at the edge of the ruptured rotator cuff tendon was significantly correlated with postoperative cuff integrity and that mRNA expression for type I collagen was significantly associated with the period from symptom onset. These results may suggest that conservative treatment should not be prolonged if patients do not respond within a certain period. Level of Evidence: Level III, prognostic case-control study.
R
upture of the rotator cuff tendon occurs frequently in middle-aged and elderly individuals. The clinical results of both open and arthroscopic cuff repair are favorable, although postoperative imaging
From the Department of Orthopedic Surgery, Kurume University (I.S., T.Y., K. Nakama, K. Nagata); and Departments of Orthopedic Surgery (M.G., Y.M., T.O., F.H.) and Radiology (K.K.), Kurume University Medical Center, Kurume, Japan. The authors report no conflict of interest. Received June 15, 2010; accepted March 16, 2011. Address correspondence to Masafumi Gotoh, M.D., Department of Orthopedic Surgery, Kurume University Medical Center, 155 Kokubu-machi, Kurume, Fukuoka 839-0863, Japan. E-mail: [email protected] © 2011 by the Arthroscopy Association of North America 0749-8063/10356/$36.00 doi:10.1016/j.arthro.2011.03.078
studies have shown a high failure rate of biological healing, thus raising questions about the healing ability of the ruptured rotator cuff tendon.1-6 A normal tendon consists mainly of type I collagen, organized into fibrils grouped parallel to form organized bundles.7 Type III collagen is a molecule that occurs normally, but it is also synthesized as a repair response to tendon healing.8 This collagen is gradually replaced by type I collagen as the scar tissue matures.9,10 Because these are the primary fibrillar collagens of both normal and ruptured tendons, type I collagen and type III collagen have been assessed in previous studies and have been shown to play important roles in the healing of ruptured tendons.11-14 Specifically, messenger RNA (mRNA) for type I and type III collagen is used as an indicator of de novo production of these collagens.10,15
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 27, No 9 (September), 2011: pp 1173-1179
1173
1174
I. SHIRACHI ET AL.
There has been controversy regarding the healing ability of the ruptured tendon edge.16,17 One laser Doppler study found that tissue less than 2.5 cm from the edge of the tear appeared to be histologically viable in terms of microvasculature.15 The edge of the ruptured tendon appears to be active, with increased cellular synthesis of type I and type III collagen.8,16 In contrast, Yuan et al.4 reported that ruptured rotator cuff tendons have a reduced ability to heal, showing an increased degree of apoptosis. Thus determination of whether the edge of the ruptured rotator cuff tendon has healing potential is of great importance for both arthroscopic and open rotator cuff repair. Proper examination of the molecular species present in the ruptured rotator cuff tendon may help to shed light on this issue. The purpose of this study was to compare the expression of mRNA for collagen at the edge of the ruptured rotator cuff tendon with that in the normal tendon, as well to correlate the mRNA of type I and type III collagen at the edge of the ruptured tendon with preoperative clinical parameters and postoperative cuff integrity. We hypothesized that collagen mRNA expression at the edges of ruptured rotator cuff tendons would be correlated with postoperative cuff integrity. METHODS Specimens were obtained during surgery from 12 patients with full-thickness rotator cuff tears. All lesions were diagnosed by preoperative magnetic resonance imaging (MRI) and verified by operative evaluation. A margin 5 to 7 mm wide was removed from each ruptured rotator cuff tendon by a sharp scalpel during surgery. The tissue specimens were snap frozen and stored at – 80°C for isolation of total cellular RNA and protein. Five cadaveric rotator cuff tendons without apparent rupture were obtained as controls within 6 hours after death at our institution.18 Shoulder pain had not been reported before death by any of these individuals: we examined the patient’s record and/or asked the physician in charge. The rotator cuff tendon was obtained as close as possible to the “critical portion,” that is, the site where the tear would mostly occur in the supraspinatus tendon (1 cm posteriorly and 1 cm proximally).19 Therefore an area measuring approximately 2 ⫻ 2 cm was harvested, with the critical portion in the center. Inclusion criteria included patients with full-thickness rotator cuff tears who had a traumatic injury from a variety of mechanisms that resulted in their shoulder symptoms. Exclusion criteria included the presence of unequivocally diagnosed concomitant disorders of the
shoulder (e.g., glenohumeral arthritis, fracture, osteonecrosis, infection, or tumor) in both the clinical and cadaveric cases. All specimens were obtained with informed consent and with the permission of the ethics committee of our hospital. Assessment of Clinical Parameters Preoperative evaluation included a patient questionnaire and physical examination by an independent examiner who was blinded to the study. The examination consisted of functional assessment based on the Japanese Orthopaedic Association (JOA) scoring system, range-of-motion testing, and strength testing.20 A visual analog scale (VAS) was used to rate the patient’s perceived level of pain (with a range of 0 to 10).21 The largest dimension of the tear was measured during surgery. Surgical Technique of Open Rotator Cuff Repair All procedures were performed with the patient in the beach-chair position. An 8-cm incision was made along the Langer lines just lateral to the coracoid process. The deltoid was split longitudinally from the anterior edge of the acromion to a distance of 5 cm, and the insertion of the anterior deltoid was detached from the acromion. A sharp scalpel was used to skeletonize the acromion; a chisel was then used to remove the anterolateral undersurface of the acromion. After the subacromial bursa overlying the cuff had been excised, the largest dimension of the rupture was measured (in centimeters) with the arm positioned at the patient’s side. Multiple single interrupted sutures were placed at the lateral edge of the torn cuff to mobilize the retracted cuff. If tension still existed, the capsule on the superior aspect of the glenoid was transected, and anterior and posterior releases were performed. After the margin of the ruptured cuff had been excised, the ruptured cuff was fixed with multiple single interrupted sutures to the bone tunnel prepared in the greater tuberosity. “Watertight repair” was completed in all patients without excessive tension at the repair site, that is, the torn cuff was firmly sutured to the greater tuberosity with the arm at the patient’s side and complete coverage of the humeral head by the repaired cuff was achieved. Once repaired, the deltoid was reattached with nonabsorbable braided sutures to its previous, anatomic position. In all patients the affected arm was kept in a sling for 5 weeks, but self-assisted passive motion was allowed after postoperative day 4. Full active motion was
COLLAGEN PRODUCTION AND CUFF INTEGRITY started 6 weeks after surgery, and resistive muscle strengthening began 8 weeks after surgery. Measurement of Collagen mRNA by Reverse Transcriptase Polymerase Chain Reaction Total RNA was isolated from samples by use of an ISOGEN-LS Poly (A)⫹ RNA Isolation Pack (Wako Junyaku, Osaka, Japan). Total RNA was reverse transcribed to complementary DNA and then amplified by polymerase chain reaction (PCR). A 5700 Sequence Detection System (Applied Biosystems, Foster City, CA) was used with SYBR Green PCR Master Mix (Applied Biosystems). After reverse transcription, realtime PCR was performed by use of primers synthesized according to published human sequences.12 The conditions for real-time PCR were initial treatment at 50°C for 2 minutes, followed by incubation at 95°C for 15 seconds and then annealing at 62°C for 1 minute without extension. This process (denaturing and annealing) was repeated, with calculation of the values in the linear phase. The amount of mRNA expression for each type of collagen were measured and normalized against -actin as an internal standard by the delta-delta-CT methods.22,23 The relative ratios of the PCR products for the collagens in each sample (ruptured and control tendons) compared with those in MG cells (an osteosarcomaderived cell line used as a positive control for collagens) were calculated. Western Blot Analysis of Protein Production Tissue specimens were solubilized in buffer (0.25mol/L sucrose, 3-mmol/L Tris-hydrochloride [pH 7.5], and 0.1-mmol/L ethylenediaminetetraacetic acid) after the insoluble materials had been removed by centrifugation for 60 minutes at 40,000g. Electrophoresis of specimens was performed on 15% sodium dodecyl sulfate–polyacrylamide gel electrophoresis gels, and the specimens were then transferred onto polyvinyl difluoride membranes (Millipore, Natick, MA). After inhibition of nonspecific binding, the protein blots were incubated with antibodies (type I anti-collagen [Santa Cruz Biotechnology, Santa Cruz, CA]; type III anti-collagen [Santa Cruz Biotechnology]; and tubulin [Sigma, St. Louis, MO]) for 1 hour at 37°C and then reacted with anti-mouse immunoglobulin G antibody labeled with horseradish peroxidase (Amersham, Buckinghamshire, United Kingdom). Immune complex was detected by chemiluminescence (ECL Western Blotting Detection System and Hyperfilm; Amersham International, Arlington Heights, IL).
1175
Evaluation by MRI The patients underwent MRI (1.5-T magnetic resonance unit [Excelart; Toshiba Medical Systems, Tokyo, Japan]) at our institution preoperatively and postoperatively. Our MRI protocol consisted of oblique coronal imaging, oblique sagittal imaging, and axial imaging in a T1-weighted spin echo sequence, a fat-suppressed T2weighted fast spin echo sequence, and a T2-weighted gradient echo sequence. The slice thickness was 3.5 mm, with an interslice gap of 0.5 mm. The field of view was 20 cm, and the image matrix was 256 ⫻ 512. All scans were judged adequate for the assessment of rotator cuff tendon integrity. Any tendon defect filled with fluid was considered a tear. Postoperative cuff integrity was examined at least 1 year after surgery (mean, 14.1 months; range, 12 to 15 months) by use of the classification of Sugaya et al.24 Cuff integrity was classified into 5 categories from 3 T2-weighted image views: type I, repaired cuff with sufficient thickness and homogenously low intensity on each image; type II, sufficient thickness with partial high-intensity area; type III, insufficient thickness but without discontinuity, suggesting a partial-thickness delaminated tear; type IV, presence of minor discontinuity in only 1 or 2 slices on both coronal and sagittal images, suggesting a small full-thickness tear; and type V, presence of a major discontinuity in more than 2 slices on both coronal and sagittal images, suggesting a medium or large full-thickness tear.24 The magnetic resonance reading was performed in a blind manner by orthopaedic specialists and a radiologist trained in musculoskeletal medicine. A consensus discussion determined the final reading judgment. Therefore intraobserver statistics were not used. Statistical Analysis Our data were derived from a varying distribution, and thus nonparametric analysis was considered better for the analysis. Accordingly, we used the MannWhitney U test and Spearman rank correlation test. The Mann-Whitney U test was used for comparison among the different parameters recorded. The Spearman rank correlation test was used to analyze possible relations between the procollagen mRNAs and clinical parameters. Results with P ⬍ .05 were considered significant. We used SAS software (version 9.3; SAS, Cary, NC) for statistical analysis. A power analysis was not performed.
1176
I. SHIRACHI ET AL. TABLE 1.
Comparison of Collagen mRNA Between Tendon at Edge of Ruptured Rotator Cuff and Control Tendon
Type I collagen Type III collagen
Rotator Cuff Tear Patients
Cadaveric Controls
P Value
2.22 (0.414-3.88) 0.903 (0.442-2.86)
0.252 (0.164-0.403) 0.291 (0.11-0.295)
.0462 .0306
NOTE. The relative ratios of the PCR products for the collagens in each sample (ruptured and control tendons) compared with those in MG cells were calculated. Data are given as median (25th to 75th percentile).
RESULTS Comparison of Collagen mRNAs at Edges of Ruptured Rotator Cuff Tendons With Those in Control Tendons A total of 17 rotator cuff tendons (12 ruptured and 5 control) were subjected to real-time reverse transcriptase (RT)–PCR analysis. The mean patient age was 58.2 years (range, 47 to 68 years), and the mean age of the controls was 66.2 years (range, 57 to 76 years). Expression of mRNA for type I and type III collagen at the edges of the ruptured rotator cuff tendons was compared with that in the control tendons. In the case of type I collagen mRNA expression, the relative ratio of the PCR products in each sample (ruptured and control tendons) compared with that in the control MG cells was 2.22 (25th to 75th percentile, 0.414 to 3.88) in the ruptured rotator cuff tendon and 0.252 (25th to 75th percentile, 0.164 to 0.403) in the control tendon. In the case of type III collagen mRNA expression, the relative ratio of the PCR products in each sample compared with that in the MG cells was 0.903 (25th to 75th percentile, 0.442 to 2.86) in the ruptured rotator cuff tendon and 0.291 (25th to 75th percen-
tile, 0.11 to 0.295) in the control tendon. Expression of mRNA for both collagen types at the edges of the ruptured tendons was significantly greater than that in the controls (P ⫽ .0462 for type I collagen and P ⫽ .0306 for type III collagen) (Table 1). Collagen Protein Production at Edges of Ruptured Rotator Cuff Tendons and in Control Tendons Next, collagen protein production was examined by Western blot analysis. Because the amount of harvested tissue was limited, a total of 7 rotator cuff tendons (5 ruptured and 2 control) were available for this analysis. As with the results of real-time RT-PCR analysis, there was increased production of both collagen proteins at the edges of the ruptured rotator cuff tendons, as compared with the controls. Representative data are given in Fig 1. Correlation Between Expression of Collagen mRNA at Edges of Ruptured Rotator Cuff Tendons and Preoperative Clinical Parameters Correlations between preoperative clinical parameters in the 12 patients and the level of expression of
FIGURE 1. Detection of collagen protein production by Western blot analysis.
COLLAGEN PRODUCTION AND CUFF INTEGRITY TABLE 2.
1177
Correlation Between Collagen mRNA and Clinical Parameters Type I Collagen mRNA
Type III Collagen mRNA
Clinical Parameter
Correlation Coefficient
P Value
Correlation Coefficient
P Value
Age Period from symptom onset Tear size JOA score VAS score
⫺0.18 ⫺0.84 0.46 ⫺0.21 0.25
.57 .0005 .13 .50 .43
⫺0.14 ⫺0.53 0.11 ⫺0.26 0.11
.66 .07 .73 .42 .73
NOTE. The relative ratios of the PCR products for the collagens in samples (ruptured tendons) compared with those in MG cells were calculated.
collagen mRNA at the edges of the ruptured tendons were examined. The preoperative clinical parameters included were age, JOA score, rupture size, period from symptom onset, and VAS score. The mean JOA score was 55.7 points (range, 33 to 79 points), and the mean rupture size was 4.1 cm, comprising 2 small, 3 medium, 3 large, and 4 massive rotator cuff tears.17 The mean period from symptom onset was 9.3 months (range, 1 to 36 months), and the mean VAS score was 5.92 (range, 4 to 8). The results showed a significant inverse correlation between the period from symptom onset and the level of expression of mRNA for type I collagen (r ⫽ – 0.845, P ⬍ .0005) but not for type III collagen (r ⫽ – 0.53, P ⫽ .074). There were no significant correlations between the expression of mRNA for either collagen type and the other preoperative clinical parameters, including the size of the rotator cuff tendon rupture. Details are given in Table 2. Correlation Between Expression of Collagen mRNA at Edges of Ruptured Rotator Cuff Tendons and Postoperative Cuff Integrity Postoperative cuff integrity was evaluated with the classification system of Sugaya et al.24 There were 6 type I cases, 3 type II cases, 2 type III cases, and 1 type IV case. When graded from 5 points for type I to 1 point for type 5, postoperative cuff integrity was significantly correlated with the expression of mRNA for both collagen types at the edges of the ruptured tendons (r ⫽ 0.63 [P ⫽ .038] for type I collagen and r ⫽ 0.626 [P ⫽ .03] for type III collagen) (Table 3). DISCUSSION Collagen production at the edge of the rotator cuff tendon has been examined before.16,17 However, no
studies have related it to postoperative structural outcome. Our study showed that expression of mRNA for both type I collagen and type III collagen at the edge of the ruptured rotator cuff tendon was closely correlated with postoperative cuff integrity. Lo et al.,17 using RT-PCR, found that expression of mRNA for type I and type III collagen at the edges of ruptured rotator cuff tendons was significantly greater than that in control rotator cuff tendons obtained from fresh cadavers. They concluded that the tendon at the edge of a ruptured rotator cuff may potentially contribute to the healing process after repair. The results of our RT-PCR analysis of type I and type III collagen were therefore in accord with those of Lo et al. Furthermore, Western blot analysis showed increased production of both collagen proteins at the edge of the ruptured rotator cuff tendon. These results therefore suggested that the tissue at the edge of the ruptured rotator cuff tendon in this series showed active protein production. Previous studies have examined the correlation between the period from symptom onset and the expression of type I and type III collagen.16,17 Lo et al.17 reported an inverse correlation between the period
TABLE 3.
Correlation Between Collagen mRNA and Postoperative Cuff Integrity Type I Collagen mRNA
Postoperative cuff integrity
Correlation Coefficient
P Value
0.63
P ⬍ .038
Type III Collagen mRNA Correlation Coefficient P Value ⫺0.14
P ⬍ .03
NOTE. The relative ratios of the PCR products for the collagens in samples (ruptured tendons) compared with those in MG cells were calculated.
1178
I. SHIRACHI ET AL.
from symptom onset and the expression of type I collagen mRNA. An in situ hybridization study has shown that, at the edge of a ruptured rotator cuff tendon, expression of type I collagen mRNA in tenocytes is correlated with the period from symptom onset whereas the expression of type III collagen mRNA in these cells is not.16 Our study confirmed a significant and inverse correlation between the period from symptom onset and expression of type I collagen mRNA at the edge of the ruptured rotator cuff tendon (r ⫽ – 0.845, P ⫽ .0005). There has been controversy as to the quality of tissue at the edge of the ruptured rotator cuff tendon.13-15 Goodmurphy et al.14 showed that, in rotator cuff tendons that had been ruptured for 3 to 5 years, the edge of the ruptured tendon was less vascular. Matthews et al.25 found an overall deterioration in tissue quality at the edge of the ruptured rotator cuff tendon, with a mean period from symptom onset of 32.6 months, and recommended excision of the tendon edge. In contrast, laser Doppler flowmetry studies have shown blood flow throughout the rotator cuff tendon and a hyperemic response at the edge of the rupture.15 After a mean period of 9.3 months from symptom onset, this study showed significantly greater expression of mRNA for type I collagen (P ⫽ .0462) and type III collagen (P ⫽ .0306) at the tendon edge than in controls. In addition, expression of mRNA for both collagen types at the edge of the ruptured tendon was correlated with postoperative cuff integrity, and expression of type I collagen mRNA was correlated with the period from symptom onset. Although this study found no significant correlation between the expression of type III collagen mRNA and period from symptom onset, the data indicated that as the period from onset became prolonged, type I collagen production at the edge of the ruptured tendon was further decreased, eventually leading to deterioration of tissue quality at this edge. Larger ruptures seem to be associated with a greater degree of tissue deterioration.26-28 Matthews et al.27 stated that smaller ruptures retained a greater potential to heal and that the longer duration of symptoms associated with larger ruptures could account for their greater degree of degeneration. In contrast, Gazielly et al.29 suggested that the tissue quality of a ruptured tendon was a more important factor than rupture size in determining the outcome of surgical repair. In accordance with their study, our study showed the importance of expression of mRNA for type I and type III collagen at the edge of the ruptured tendon, being positively correlated
with postoperative cuff integrity and not with rupture size. Taken together with previous reports, our findings suggest that the physiologic capacity of a ruptured rotator cuff tendon to heal is not always associated with rupture size. Japanese studies have shown that postoperative signal intensity on MRI alters within 1 year, suggesting that postoperative evaluation by MRI should be performed at least 1 year after repair.30,31 Therefore we examined cuff integrity at least 1 year postoperatively. This study had several limitations. First, posttranscriptional/post-translational regulation may affect the eventual expression of functional protein. It is difficult to distinguish newly synthesized collagen from older collagen, because collagen has an extremely long half-life (300 to 500 days).32 Thus measurement of the expression of collagen mRNA by RT-PCR is a rational method and can be regarded as a real-time indicator of newly synthesized collagen.6,17 Second, the amount of control tendon tissue was limited, because we harvested the area of the control tendon as close to the critical portion as possible. Furthermore, we preferentially used a larger amount of tissue for PCR, in comparison with Western blot analysis. Consequently, we were unable to analyze all the control tendons at both the mRNA and protein levels; only 2 tendons were subjected to both examinations. Although the number of control tendons analyzed was less, we believe that the Western blot data partially strengthened the validity of the mRNA data. The third limitation was that the use of fresh postmortem controls might have affected the data. However, a previous report has confirmed the postmortem stability of total RNA isolated from dense connective tissue within 96 hours (within 6 hours in our study), thus supporting the validity of our data.33 Fourth, the range of postoperative cuff integrity was narrow, with a low incidence of retear (8.3% [1 of 12 cases]). Fifth, we did not examine the relation between onset and the quality of the tissue at the time of surgery/1-year clinical result. However, we confirmed that the “1-month sample” did not bias the data set in our study. Finally, because our sample size was small and a power analysis was not performed, there is a possibility that the absence of a significant correlation for some items might not have been observed if the study had been a larger-scale study. Despite these limitations, however, we were able to show a correlation between collagen production at the edge of the ruptured rotator cuff tendon and postoperative cuff integrity.
COLLAGEN PRODUCTION AND CUFF INTEGRITY CONCLUSIONS This study showed that expression of mRNA for type I and type III collagen at the edge of the ruptured rotator cuff tendon was significantly correlated with postoperative cuff integrity and that mRNA expression for type I collagen was significantly associated with the period from symptom onset. These results may suggest that conservative treatment should not be prolonged if patients do not respond within a certain period. Acknowledgment: The authors thank H. Wakita and K. Yoshida for their technical assistance.
REFERENCES 1. Nirschl RP. Rotator cuff tendinitis: Basic concepts of pathoetiology. Instr Course Lect 2007;38:439-445. 2. Harryman DT II, Mack LA, Wang KY, Jackins SE, Richardson ML, Matsen FA III. Repairs of the rotator cuff. Correlation of functional results with integrity of the cuff. J Bone Joint Surg Am 1991;73:982-989. 3. Liu SJ, Baker CL. Arthroscopically assisted rotator cuff repair. Correlation of functional result with integrity of the cuff. Arthroscopy 1994;10:54-60. 4. Yuan J, Murrell GA, Wei AQ, Wang MX. Apoptosis in rotator cuff tendonopathy. J Orthop Res 2002;20:1372-1379. 5. Riley G. The pathogenesis of tendinopathy. A molecular perspective. Rheumatology (Oxford) 2004;43:131-142. 6. Hamada K, Tomonaga A, Gotoh M, Yamakawa H, Fukuda H. Intrinsic healing capacity and tearing process of torn supraspinatus tendons: In situ hybridization study of alpha 1 (I) procollagen mRNA. J Orthop Res 1997;15:24-32. 7. Blevins FT, Djurasovic M, Flatow EL, Vogel KG. Biology of the rotator cuff tendon. Orthop Clin North Am 1997;28:1-15. 8. Kumagai J, Uhthoff K, Sarkar K, Murnaghan JP. Collagen type III in rotator cuff tears: An immunohistochemical study. J Shoulder Elbow Surg 1992:187-192. 9. Boykiw R, Sciore P, Reno C, et al. Altered levels of extracellular matrix molecule mRNA in healing rabbit ligaments. Matrix Biol 1998;17:371-378. 10. Fan L, Sarkar K, Franks DJ, Uhthoff HK. Estimation of total collagen and type I and III collagen in canine rotator cuff tendons. Calcif Tissue Int 1997;61:223-229. 11. Riley GP, Harrall RL, Constant CR, et al. Tendon degeneration and chronic shoulder pain: Changes in the collagen composition of the human rotator cuff tendon in rotator cuff tendinitis. Ann Rheum Dis 1994;53:359-366. 12. Yamada T, Gotoh M, Nakama K, et al. Effects of hyaluronan on cell proliferation and mRNA expression of procollagens ␣1 (I) and ␣1 (III) in tendon-derived fibroblasts from patients with rotator cuff disease. Am J Sports Med 2007;35:1870-1876. 13. Fukuda H, Hamada K, Yamanaka K. Pathology and pathogenesis of bursal-side rotator cuff tears from en bloc histologic sections. Clin Orthop Relat Res 1990:75-80. 14. Goodmurphy CW, Osborn J, Akesson EJ, et al. An immunocytochemical analysis of torn rotator cuff tendon taken at the time of repair. J Shoulder Elbow Surg 2003;12:368-374.
1179
15. Swiontkowski MF, Lannotti JP, Boulas HJ, Esterhai JL. Intraoperative assessment of rotator cuff vascularity using laser Doppler flowmetry. In: Post M, Morrey BF, eds. Surgery of the shoulder. St. Louis: Mosby-Year Book, 1990. 16. Hamada K, Okawara Y, Fryer JN, Tomonaga A, Fukuda H. Localization of mRNA of procollagen alpha 1 type I in torn supraspinatus tendons: In situ hybridization using digoxigeninlabeled oligonucleotide probe. Clin Orthop Relat Res 1994: 18-21. 17. Lo IKY, Boorman R, Marchuk L, et al. Matrix molecule mRNA levels in the bursa and rotator cuff of patients with full-thickness rotator cuff tears. Arthroscopy 2005;21:645-651. 18. Seear PJ, Sweeney GE. Stability of RNA isolated from postmortem tissues of Atlantic salmon (Salmo salar L.). Fish Physiol Biochem 2008;34:19-24. 19. Rathbun JB, Macnab I. The microvascular pattern of the rotator cuff. J Bone Joint Surg Br 1970;52:540-553. 20. Ide J, Tokiyoshi A, Hirose J, Mizuta H. Arthroscopic repair of traumatic combined rotator cuff tears involving the subscapularis tendon. J Bone Joint Surg Am 2007;89:2378-2388. 21. Gotoh M, Hamada K, Yamakawa H, et al. Interleukin-1induced subacromial synovitis and shoulder pain in rotator cuff diseases. Rheumatology 2001;40:995-1001. 22. Nago M, Mitsui Y, Gotoh M, et al. Hyaluronan modulates cell proliferation and mRNA expression of adhesion-related procollagens and cytokines in glenohumeral synovial/capsular fibroblasts in adhesive capsulitis. J Orthop Res 2010;28:726-731. 23. Ben-Shahar Y, Robichon A, Sokolowski MB, Robinson GE. Influence of gene action across different time scales on behavior. Science 2002;296:741-744. 24. Sugaya H, Maeda K, Matsuki K, Moriishi J. Functional and structural outcome after arthroscopic full-thickness rotator cuff repair: Single-row versus dual-row fixation. Arthroscopy 2005; 21:1307-1316. 25. Matthews TJ, Smith SR, Peach CA, Ress JL, Urban JP, Carr AJ. In vivo measurement of tissue metabolism in tendons of the rotator cuff. J Bone Joint Surg Br 2007;89:633-638. 26. Hashimoto T, Nobuhara K, Hamada T. Pathologic evidence of degeneration as a primary cause of rotator cuff tear. Clin Orthop Relat Res 2003:111-120. 27. Matthews TJ, Hand GC, Ress JL, Athanasou NA, Carr AJ. Pathology of the torn rotator cuff tendon: Reduction in potential for repair as tear size increases. J Bone Joint Surg Br 2006;88:489-495. 28. Schmutz S, Fuchs T, Regenfelder F, et al. Expression of atrophy mRNA relates to tendon tear size in supraspinatus muscle. Clin Orthop Relat Res 2009;467:457-464. 29. Gazielly DF, Gleyze P, Montagnon C. Functional and anatomical results after rotator cuff repair. Clin Orthop Relat Res 1994:43-53. 30. Oizumi N, Suenaga N, Hisada Y, Minami A, Fukuda K. Postoperative changes of MRI findings over time after rotator cuff repair. Shoulder Joint 2007;31:261-264. 31. Ohnada Y, Matsumoto D, Fukuda K, Oizumi N, Sugimoto H. Postoperative changes of MRI findings after rotator cuff repair within 1 year and over 1 year. Shoulder Joint 2007;34:427-430. 32. Neuberger A, Slack HG. The metabolism of collagen from liver, bone, skin and tendon in the normal rat. Biochem J 1953;53:47-52. 33. Marchuk L, Sciore P, Reno C, Frank CB, Hart DA. Postmortem stability of total RNA isolated from rabbit ligament, tendon and cartilage. Biochim Biophys Acta 1998;1379:171177.
Collagen Production at the Edge of Ruptured Rotator Cuff Tendon is Correlated With Postoperative Cuff Integrity Isao Shirachi, M.D., Masafumi Gotoh, M.D., Yasuhiro Mitsui, M.D., Tetsu Yamada, M.D., Kenjiro Nakama, M.D., Kazuyuki Kojima, M.D., Takahiro Okawa, M.D., Fujio Higuchi, M.D., and Kensei Nagata, M.D.
Purpose: The purpose was to evaluate the correlation between messenger RNA (mRNA) expression of collagen at the edge of the ruptured rotator cuff tendon and postoperative cuff integrity. Methods: The edge of the ruptured tendon was sampled during open rotator cuff surgery in 12 patients with full-thickness rotator cuff tears (mean age, 58.2 years). The mean period from symptom onset was 9.3 months (range, 1 to 36 months), and the mean tear size was 4.1 cm. As controls, rotator cuff tendons with no gross rupture were taken from 5 fresh cadavers. Production of type I and type III collagen was examined by real-time reverse transcription polymerase chain reaction. By use of magnetic resonance imaging, postoperative cuff integrity was evaluated based on the classification of Sugaya et al. and then scored, ranging from 5 points for type I to 1 point for type V. Results: Looking at the mRNA of type I and type III collagen in tendons, we found that the expression of mRNA for both collagen types in ruptured tendons was significantly greater than in control tendons (P ⫽ .0462 for type I collagen and P ⫽ .0306 for type III collagen). Correlating the mRNA of type I and type III collagen with repaired cuff integrity on postoperative magnetic resonance imaging, we found a close relation between expression of mRNA for both collagen types and postoperative rotator cuff integrity (r ⫽ 0.63 [P ⫽ .038] for type I collagen and r ⫽ 0.626 [P ⫽ .03] for type III collagen). Furthermore, expression of type I collagen mRNA showed a significant inverse correlation with the period from symptom onset (r ⫽ ⫺0.845, P ⬍ .0005). Conclusions: This study showed that expression of mRNA for type I and type III collagen at the edge of the ruptured rotator cuff tendon was significantly correlated with postoperative cuff integrity and that mRNA expression for type I collagen was significantly associated with the period from symptom onset. These results may suggest that conservative treatment should not be prolonged if patients do not respond within a certain period. Level of Evidence: Level III, prognostic case-control study.
R
upture of the rotator cuff tendon occurs frequently in middle-aged and elderly individuals. The clinical results of both open and arthroscopic cuff repair are favorable, although postoperative imaging
From the Department of Orthopedic Surgery, Kurume University (I.S., T.Y., K. Nakama, K. Nagata); and Departments of Orthopedic Surgery (M.G., Y.M., T.O., F.H.) and Radiology (K.K.), Kurume University Medical Center, Kurume, Japan. The authors report no conflict of interest. Received June 15, 2010; accepted March 16, 2011. Address correspondence to Masafumi Gotoh, M.D., Department of Orthopedic Surgery, Kurume University Medical Center, 155 Kokubu-machi, Kurume, Fukuoka 839-0863, Japan. E-mail: [email protected] © 2011 by the Arthroscopy Association of North America 0749-8063/10356/$36.00 doi:10.1016/j.arthro.2011.03.078
studies have shown a high failure rate of biological healing, thus raising questions about the healing ability of the ruptured rotator cuff tendon.1-6 A normal tendon consists mainly of type I collagen, organized into fibrils grouped parallel to form organized bundles.7 Type III collagen is a molecule that occurs normally, but it is also synthesized as a repair response to tendon healing.8 This collagen is gradually replaced by type I collagen as the scar tissue matures.9,10 Because these are the primary fibrillar collagens of both normal and ruptured tendons, type I collagen and type III collagen have been assessed in previous studies and have been shown to play important roles in the healing of ruptured tendons.11-14 Specifically, messenger RNA (mRNA) for type I and type III collagen is used as an indicator of de novo production of these collagens.10,15
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 27, No 9 (September), 2011: pp 1173-1179
1173
1174
I. SHIRACHI ET AL.
There has been controversy regarding the healing ability of the ruptured tendon edge.16,17 One laser Doppler study found that tissue less than 2.5 cm from the edge of the tear appeared to be histologically viable in terms of microvasculature.15 The edge of the ruptured tendon appears to be active, with increased cellular synthesis of type I and type III collagen.8,16 In contrast, Yuan et al.4 reported that ruptured rotator cuff tendons have a reduced ability to heal, showing an increased degree of apoptosis. Thus determination of whether the edge of the ruptured rotator cuff tendon has healing potential is of great importance for both arthroscopic and open rotator cuff repair. Proper examination of the molecular species present in the ruptured rotator cuff tendon may help to shed light on this issue. The purpose of this study was to compare the expression of mRNA for collagen at the edge of the ruptured rotator cuff tendon with that in the normal tendon, as well to correlate the mRNA of type I and type III collagen at the edge of the ruptured tendon with preoperative clinical parameters and postoperative cuff integrity. We hypothesized that collagen mRNA expression at the edges of ruptured rotator cuff tendons would be correlated with postoperative cuff integrity. METHODS Specimens were obtained during surgery from 12 patients with full-thickness rotator cuff tears. All lesions were diagnosed by preoperative magnetic resonance imaging (MRI) and verified by operative evaluation. A margin 5 to 7 mm wide was removed from each ruptured rotator cuff tendon by a sharp scalpel during surgery. The tissue specimens were snap frozen and stored at – 80°C for isolation of total cellular RNA and protein. Five cadaveric rotator cuff tendons without apparent rupture were obtained as controls within 6 hours after death at our institution.18 Shoulder pain had not been reported before death by any of these individuals: we examined the patient’s record and/or asked the physician in charge. The rotator cuff tendon was obtained as close as possible to the “critical portion,” that is, the site where the tear would mostly occur in the supraspinatus tendon (1 cm posteriorly and 1 cm proximally).19 Therefore an area measuring approximately 2 ⫻ 2 cm was harvested, with the critical portion in the center. Inclusion criteria included patients with full-thickness rotator cuff tears who had a traumatic injury from a variety of mechanisms that resulted in their shoulder symptoms. Exclusion criteria included the presence of unequivocally diagnosed concomitant disorders of the
shoulder (e.g., glenohumeral arthritis, fracture, osteonecrosis, infection, or tumor) in both the clinical and cadaveric cases. All specimens were obtained with informed consent and with the permission of the ethics committee of our hospital. Assessment of Clinical Parameters Preoperative evaluation included a patient questionnaire and physical examination by an independent examiner who was blinded to the study. The examination consisted of functional assessment based on the Japanese Orthopaedic Association (JOA) scoring system, range-of-motion testing, and strength testing.20 A visual analog scale (VAS) was used to rate the patient’s perceived level of pain (with a range of 0 to 10).21 The largest dimension of the tear was measured during surgery. Surgical Technique of Open Rotator Cuff Repair All procedures were performed with the patient in the beach-chair position. An 8-cm incision was made along the Langer lines just lateral to the coracoid process. The deltoid was split longitudinally from the anterior edge of the acromion to a distance of 5 cm, and the insertion of the anterior deltoid was detached from the acromion. A sharp scalpel was used to skeletonize the acromion; a chisel was then used to remove the anterolateral undersurface of the acromion. After the subacromial bursa overlying the cuff had been excised, the largest dimension of the rupture was measured (in centimeters) with the arm positioned at the patient’s side. Multiple single interrupted sutures were placed at the lateral edge of the torn cuff to mobilize the retracted cuff. If tension still existed, the capsule on the superior aspect of the glenoid was transected, and anterior and posterior releases were performed. After the margin of the ruptured cuff had been excised, the ruptured cuff was fixed with multiple single interrupted sutures to the bone tunnel prepared in the greater tuberosity. “Watertight repair” was completed in all patients without excessive tension at the repair site, that is, the torn cuff was firmly sutured to the greater tuberosity with the arm at the patient’s side and complete coverage of the humeral head by the repaired cuff was achieved. Once repaired, the deltoid was reattached with nonabsorbable braided sutures to its previous, anatomic position. In all patients the affected arm was kept in a sling for 5 weeks, but self-assisted passive motion was allowed after postoperative day 4. Full active motion was
COLLAGEN PRODUCTION AND CUFF INTEGRITY started 6 weeks after surgery, and resistive muscle strengthening began 8 weeks after surgery. Measurement of Collagen mRNA by Reverse Transcriptase Polymerase Chain Reaction Total RNA was isolated from samples by use of an ISOGEN-LS Poly (A)⫹ RNA Isolation Pack (Wako Junyaku, Osaka, Japan). Total RNA was reverse transcribed to complementary DNA and then amplified by polymerase chain reaction (PCR). A 5700 Sequence Detection System (Applied Biosystems, Foster City, CA) was used with SYBR Green PCR Master Mix (Applied Biosystems). After reverse transcription, realtime PCR was performed by use of primers synthesized according to published human sequences.12 The conditions for real-time PCR were initial treatment at 50°C for 2 minutes, followed by incubation at 95°C for 15 seconds and then annealing at 62°C for 1 minute without extension. This process (denaturing and annealing) was repeated, with calculation of the values in the linear phase. The amount of mRNA expression for each type of collagen were measured and normalized against -actin as an internal standard by the delta-delta-CT methods.22,23 The relative ratios of the PCR products for the collagens in each sample (ruptured and control tendons) compared with those in MG cells (an osteosarcomaderived cell line used as a positive control for collagens) were calculated. Western Blot Analysis of Protein Production Tissue specimens were solubilized in buffer (0.25mol/L sucrose, 3-mmol/L Tris-hydrochloride [pH 7.5], and 0.1-mmol/L ethylenediaminetetraacetic acid) after the insoluble materials had been removed by centrifugation for 60 minutes at 40,000g. Electrophoresis of specimens was performed on 15% sodium dodecyl sulfate–polyacrylamide gel electrophoresis gels, and the specimens were then transferred onto polyvinyl difluoride membranes (Millipore, Natick, MA). After inhibition of nonspecific binding, the protein blots were incubated with antibodies (type I anti-collagen [Santa Cruz Biotechnology, Santa Cruz, CA]; type III anti-collagen [Santa Cruz Biotechnology]; and tubulin [Sigma, St. Louis, MO]) for 1 hour at 37°C and then reacted with anti-mouse immunoglobulin G antibody labeled with horseradish peroxidase (Amersham, Buckinghamshire, United Kingdom). Immune complex was detected by chemiluminescence (ECL Western Blotting Detection System and Hyperfilm; Amersham International, Arlington Heights, IL).
1175
Evaluation by MRI The patients underwent MRI (1.5-T magnetic resonance unit [Excelart; Toshiba Medical Systems, Tokyo, Japan]) at our institution preoperatively and postoperatively. Our MRI protocol consisted of oblique coronal imaging, oblique sagittal imaging, and axial imaging in a T1-weighted spin echo sequence, a fat-suppressed T2weighted fast spin echo sequence, and a T2-weighted gradient echo sequence. The slice thickness was 3.5 mm, with an interslice gap of 0.5 mm. The field of view was 20 cm, and the image matrix was 256 ⫻ 512. All scans were judged adequate for the assessment of rotator cuff tendon integrity. Any tendon defect filled with fluid was considered a tear. Postoperative cuff integrity was examined at least 1 year after surgery (mean, 14.1 months; range, 12 to 15 months) by use of the classification of Sugaya et al.24 Cuff integrity was classified into 5 categories from 3 T2-weighted image views: type I, repaired cuff with sufficient thickness and homogenously low intensity on each image; type II, sufficient thickness with partial high-intensity area; type III, insufficient thickness but without discontinuity, suggesting a partial-thickness delaminated tear; type IV, presence of minor discontinuity in only 1 or 2 slices on both coronal and sagittal images, suggesting a small full-thickness tear; and type V, presence of a major discontinuity in more than 2 slices on both coronal and sagittal images, suggesting a medium or large full-thickness tear.24 The magnetic resonance reading was performed in a blind manner by orthopaedic specialists and a radiologist trained in musculoskeletal medicine. A consensus discussion determined the final reading judgment. Therefore intraobserver statistics were not used. Statistical Analysis Our data were derived from a varying distribution, and thus nonparametric analysis was considered better for the analysis. Accordingly, we used the MannWhitney U test and Spearman rank correlation test. The Mann-Whitney U test was used for comparison among the different parameters recorded. The Spearman rank correlation test was used to analyze possible relations between the procollagen mRNAs and clinical parameters. Results with P ⬍ .05 were considered significant. We used SAS software (version 9.3; SAS, Cary, NC) for statistical analysis. A power analysis was not performed.
1176
I. SHIRACHI ET AL. TABLE 1.
Comparison of Collagen mRNA Between Tendon at Edge of Ruptured Rotator Cuff and Control Tendon
Type I collagen Type III collagen
Rotator Cuff Tear Patients
Cadaveric Controls
P Value
2.22 (0.414-3.88) 0.903 (0.442-2.86)
0.252 (0.164-0.403) 0.291 (0.11-0.295)
.0462 .0306
NOTE. The relative ratios of the PCR products for the collagens in each sample (ruptured and control tendons) compared with those in MG cells were calculated. Data are given as median (25th to 75th percentile).
RESULTS Comparison of Collagen mRNAs at Edges of Ruptured Rotator Cuff Tendons With Those in Control Tendons A total of 17 rotator cuff tendons (12 ruptured and 5 control) were subjected to real-time reverse transcriptase (RT)–PCR analysis. The mean patient age was 58.2 years (range, 47 to 68 years), and the mean age of the controls was 66.2 years (range, 57 to 76 years). Expression of mRNA for type I and type III collagen at the edges of the ruptured rotator cuff tendons was compared with that in the control tendons. In the case of type I collagen mRNA expression, the relative ratio of the PCR products in each sample (ruptured and control tendons) compared with that in the control MG cells was 2.22 (25th to 75th percentile, 0.414 to 3.88) in the ruptured rotator cuff tendon and 0.252 (25th to 75th percentile, 0.164 to 0.403) in the control tendon. In the case of type III collagen mRNA expression, the relative ratio of the PCR products in each sample compared with that in the MG cells was 0.903 (25th to 75th percentile, 0.442 to 2.86) in the ruptured rotator cuff tendon and 0.291 (25th to 75th percen-
tile, 0.11 to 0.295) in the control tendon. Expression of mRNA for both collagen types at the edges of the ruptured tendons was significantly greater than that in the controls (P ⫽ .0462 for type I collagen and P ⫽ .0306 for type III collagen) (Table 1). Collagen Protein Production at Edges of Ruptured Rotator Cuff Tendons and in Control Tendons Next, collagen protein production was examined by Western blot analysis. Because the amount of harvested tissue was limited, a total of 7 rotator cuff tendons (5 ruptured and 2 control) were available for this analysis. As with the results of real-time RT-PCR analysis, there was increased production of both collagen proteins at the edges of the ruptured rotator cuff tendons, as compared with the controls. Representative data are given in Fig 1. Correlation Between Expression of Collagen mRNA at Edges of Ruptured Rotator Cuff Tendons and Preoperative Clinical Parameters Correlations between preoperative clinical parameters in the 12 patients and the level of expression of
FIGURE 1. Detection of collagen protein production by Western blot analysis.
COLLAGEN PRODUCTION AND CUFF INTEGRITY TABLE 2.
1177
Correlation Between Collagen mRNA and Clinical Parameters Type I Collagen mRNA
Type III Collagen mRNA
Clinical Parameter
Correlation Coefficient
P Value
Correlation Coefficient
P Value
Age Period from symptom onset Tear size JOA score VAS score
⫺0.18 ⫺0.84 0.46 ⫺0.21 0.25
.57 .0005 .13 .50 .43
⫺0.14 ⫺0.53 0.11 ⫺0.26 0.11
.66 .07 .73 .42 .73
NOTE. The relative ratios of the PCR products for the collagens in samples (ruptured tendons) compared with those in MG cells were calculated.
collagen mRNA at the edges of the ruptured tendons were examined. The preoperative clinical parameters included were age, JOA score, rupture size, period from symptom onset, and VAS score. The mean JOA score was 55.7 points (range, 33 to 79 points), and the mean rupture size was 4.1 cm, comprising 2 small, 3 medium, 3 large, and 4 massive rotator cuff tears.17 The mean period from symptom onset was 9.3 months (range, 1 to 36 months), and the mean VAS score was 5.92 (range, 4 to 8). The results showed a significant inverse correlation between the period from symptom onset and the level of expression of mRNA for type I collagen (r ⫽ – 0.845, P ⬍ .0005) but not for type III collagen (r ⫽ – 0.53, P ⫽ .074). There were no significant correlations between the expression of mRNA for either collagen type and the other preoperative clinical parameters, including the size of the rotator cuff tendon rupture. Details are given in Table 2. Correlation Between Expression of Collagen mRNA at Edges of Ruptured Rotator Cuff Tendons and Postoperative Cuff Integrity Postoperative cuff integrity was evaluated with the classification system of Sugaya et al.24 There were 6 type I cases, 3 type II cases, 2 type III cases, and 1 type IV case. When graded from 5 points for type I to 1 point for type 5, postoperative cuff integrity was significantly correlated with the expression of mRNA for both collagen types at the edges of the ruptured tendons (r ⫽ 0.63 [P ⫽ .038] for type I collagen and r ⫽ 0.626 [P ⫽ .03] for type III collagen) (Table 3). DISCUSSION Collagen production at the edge of the rotator cuff tendon has been examined before.16,17 However, no
studies have related it to postoperative structural outcome. Our study showed that expression of mRNA for both type I collagen and type III collagen at the edge of the ruptured rotator cuff tendon was closely correlated with postoperative cuff integrity. Lo et al.,17 using RT-PCR, found that expression of mRNA for type I and type III collagen at the edges of ruptured rotator cuff tendons was significantly greater than that in control rotator cuff tendons obtained from fresh cadavers. They concluded that the tendon at the edge of a ruptured rotator cuff may potentially contribute to the healing process after repair. The results of our RT-PCR analysis of type I and type III collagen were therefore in accord with those of Lo et al. Furthermore, Western blot analysis showed increased production of both collagen proteins at the edge of the ruptured rotator cuff tendon. These results therefore suggested that the tissue at the edge of the ruptured rotator cuff tendon in this series showed active protein production. Previous studies have examined the correlation between the period from symptom onset and the expression of type I and type III collagen.16,17 Lo et al.17 reported an inverse correlation between the period
TABLE 3.
Correlation Between Collagen mRNA and Postoperative Cuff Integrity Type I Collagen mRNA
Postoperative cuff integrity
Correlation Coefficient
P Value
0.63
P ⬍ .038
Type III Collagen mRNA Correlation Coefficient P Value ⫺0.14
P ⬍ .03
NOTE. The relative ratios of the PCR products for the collagens in samples (ruptured tendons) compared with those in MG cells were calculated.
1178
I. SHIRACHI ET AL.
from symptom onset and the expression of type I collagen mRNA. An in situ hybridization study has shown that, at the edge of a ruptured rotator cuff tendon, expression of type I collagen mRNA in tenocytes is correlated with the period from symptom onset whereas the expression of type III collagen mRNA in these cells is not.16 Our study confirmed a significant and inverse correlation between the period from symptom onset and expression of type I collagen mRNA at the edge of the ruptured rotator cuff tendon (r ⫽ – 0.845, P ⫽ .0005). There has been controversy as to the quality of tissue at the edge of the ruptured rotator cuff tendon.13-15 Goodmurphy et al.14 showed that, in rotator cuff tendons that had been ruptured for 3 to 5 years, the edge of the ruptured tendon was less vascular. Matthews et al.25 found an overall deterioration in tissue quality at the edge of the ruptured rotator cuff tendon, with a mean period from symptom onset of 32.6 months, and recommended excision of the tendon edge. In contrast, laser Doppler flowmetry studies have shown blood flow throughout the rotator cuff tendon and a hyperemic response at the edge of the rupture.15 After a mean period of 9.3 months from symptom onset, this study showed significantly greater expression of mRNA for type I collagen (P ⫽ .0462) and type III collagen (P ⫽ .0306) at the tendon edge than in controls. In addition, expression of mRNA for both collagen types at the edge of the ruptured tendon was correlated with postoperative cuff integrity, and expression of type I collagen mRNA was correlated with the period from symptom onset. Although this study found no significant correlation between the expression of type III collagen mRNA and period from symptom onset, the data indicated that as the period from onset became prolonged, type I collagen production at the edge of the ruptured tendon was further decreased, eventually leading to deterioration of tissue quality at this edge. Larger ruptures seem to be associated with a greater degree of tissue deterioration.26-28 Matthews et al.27 stated that smaller ruptures retained a greater potential to heal and that the longer duration of symptoms associated with larger ruptures could account for their greater degree of degeneration. In contrast, Gazielly et al.29 suggested that the tissue quality of a ruptured tendon was a more important factor than rupture size in determining the outcome of surgical repair. In accordance with their study, our study showed the importance of expression of mRNA for type I and type III collagen at the edge of the ruptured tendon, being positively correlated
with postoperative cuff integrity and not with rupture size. Taken together with previous reports, our findings suggest that the physiologic capacity of a ruptured rotator cuff tendon to heal is not always associated with rupture size. Japanese studies have shown that postoperative signal intensity on MRI alters within 1 year, suggesting that postoperative evaluation by MRI should be performed at least 1 year after repair.30,31 Therefore we examined cuff integrity at least 1 year postoperatively. This study had several limitations. First, posttranscriptional/post-translational regulation may affect the eventual expression of functional protein. It is difficult to distinguish newly synthesized collagen from older collagen, because collagen has an extremely long half-life (300 to 500 days).32 Thus measurement of the expression of collagen mRNA by RT-PCR is a rational method and can be regarded as a real-time indicator of newly synthesized collagen.6,17 Second, the amount of control tendon tissue was limited, because we harvested the area of the control tendon as close to the critical portion as possible. Furthermore, we preferentially used a larger amount of tissue for PCR, in comparison with Western blot analysis. Consequently, we were unable to analyze all the control tendons at both the mRNA and protein levels; only 2 tendons were subjected to both examinations. Although the number of control tendons analyzed was less, we believe that the Western blot data partially strengthened the validity of the mRNA data. The third limitation was that the use of fresh postmortem controls might have affected the data. However, a previous report has confirmed the postmortem stability of total RNA isolated from dense connective tissue within 96 hours (within 6 hours in our study), thus supporting the validity of our data.33 Fourth, the range of postoperative cuff integrity was narrow, with a low incidence of retear (8.3% [1 of 12 cases]). Fifth, we did not examine the relation between onset and the quality of the tissue at the time of surgery/1-year clinical result. However, we confirmed that the “1-month sample” did not bias the data set in our study. Finally, because our sample size was small and a power analysis was not performed, there is a possibility that the absence of a significant correlation for some items might not have been observed if the study had been a larger-scale study. Despite these limitations, however, we were able to show a correlation between collagen production at the edge of the ruptured rotator cuff tendon and postoperative cuff integrity.
COLLAGEN PRODUCTION AND CUFF INTEGRITY CONCLUSIONS This study showed that expression of mRNA for type I and type III collagen at the edge of the ruptured rotator cuff tendon was significantly correlated with postoperative cuff integrity and that mRNA expression for type I collagen was significantly associated with the period from symptom onset. These results may suggest that conservative treatment should not be prolonged if patients do not respond within a certain period. Acknowledgment: The authors thank H. Wakita and K. Yoshida for their technical assistance.
REFERENCES 1. Nirschl RP. Rotator cuff tendinitis: Basic concepts of pathoetiology. Instr Course Lect 2007;38:439-445. 2. Harryman DT II, Mack LA, Wang KY, Jackins SE, Richardson ML, Matsen FA III. Repairs of the rotator cuff. Correlation of functional results with integrity of the cuff. J Bone Joint Surg Am 1991;73:982-989. 3. Liu SJ, Baker CL. Arthroscopically assisted rotator cuff repair. Correlation of functional result with integrity of the cuff. Arthroscopy 1994;10:54-60. 4. Yuan J, Murrell GA, Wei AQ, Wang MX. Apoptosis in rotator cuff tendonopathy. J Orthop Res 2002;20:1372-1379. 5. Riley G. The pathogenesis of tendinopathy. A molecular perspective. Rheumatology (Oxford) 2004;43:131-142. 6. Hamada K, Tomonaga A, Gotoh M, Yamakawa H, Fukuda H. Intrinsic healing capacity and tearing process of torn supraspinatus tendons: In situ hybridization study of alpha 1 (I) procollagen mRNA. J Orthop Res 1997;15:24-32. 7. Blevins FT, Djurasovic M, Flatow EL, Vogel KG. Biology of the rotator cuff tendon. Orthop Clin North Am 1997;28:1-15. 8. Kumagai J, Uhthoff K, Sarkar K, Murnaghan JP. Collagen type III in rotator cuff tears: An immunohistochemical study. J Shoulder Elbow Surg 1992:187-192. 9. Boykiw R, Sciore P, Reno C, et al. Altered levels of extracellular matrix molecule mRNA in healing rabbit ligaments. Matrix Biol 1998;17:371-378. 10. Fan L, Sarkar K, Franks DJ, Uhthoff HK. Estimation of total collagen and type I and III collagen in canine rotator cuff tendons. Calcif Tissue Int 1997;61:223-229. 11. Riley GP, Harrall RL, Constant CR, et al. Tendon degeneration and chronic shoulder pain: Changes in the collagen composition of the human rotator cuff tendon in rotator cuff tendinitis. Ann Rheum Dis 1994;53:359-366. 12. Yamada T, Gotoh M, Nakama K, et al. Effects of hyaluronan on cell proliferation and mRNA expression of procollagens ␣1 (I) and ␣1 (III) in tendon-derived fibroblasts from patients with rotator cuff disease. Am J Sports Med 2007;35:1870-1876. 13. Fukuda H, Hamada K, Yamanaka K. Pathology and pathogenesis of bursal-side rotator cuff tears from en bloc histologic sections. Clin Orthop Relat Res 1990:75-80. 14. Goodmurphy CW, Osborn J, Akesson EJ, et al. An immunocytochemical analysis of torn rotator cuff tendon taken at the time of repair. J Shoulder Elbow Surg 2003;12:368-374.
1179
15. Swiontkowski MF, Lannotti JP, Boulas HJ, Esterhai JL. Intraoperative assessment of rotator cuff vascularity using laser Doppler flowmetry. In: Post M, Morrey BF, eds. Surgery of the shoulder. St. Louis: Mosby-Year Book, 1990. 16. Hamada K, Okawara Y, Fryer JN, Tomonaga A, Fukuda H. Localization of mRNA of procollagen alpha 1 type I in torn supraspinatus tendons: In situ hybridization using digoxigeninlabeled oligonucleotide probe. Clin Orthop Relat Res 1994: 18-21. 17. Lo IKY, Boorman R, Marchuk L, et al. Matrix molecule mRNA levels in the bursa and rotator cuff of patients with full-thickness rotator cuff tears. Arthroscopy 2005;21:645-651. 18. Seear PJ, Sweeney GE. Stability of RNA isolated from postmortem tissues of Atlantic salmon (Salmo salar L.). Fish Physiol Biochem 2008;34:19-24. 19. Rathbun JB, Macnab I. The microvascular pattern of the rotator cuff. J Bone Joint Surg Br 1970;52:540-553. 20. Ide J, Tokiyoshi A, Hirose J, Mizuta H. Arthroscopic repair of traumatic combined rotator cuff tears involving the subscapularis tendon. J Bone Joint Surg Am 2007;89:2378-2388. 21. Gotoh M, Hamada K, Yamakawa H, et al. Interleukin-1induced subacromial synovitis and shoulder pain in rotator cuff diseases. Rheumatology 2001;40:995-1001. 22. Nago M, Mitsui Y, Gotoh M, et al. Hyaluronan modulates cell proliferation and mRNA expression of adhesion-related procollagens and cytokines in glenohumeral synovial/capsular fibroblasts in adhesive capsulitis. J Orthop Res 2010;28:726-731. 23. Ben-Shahar Y, Robichon A, Sokolowski MB, Robinson GE. Influence of gene action across different time scales on behavior. Science 2002;296:741-744. 24. Sugaya H, Maeda K, Matsuki K, Moriishi J. Functional and structural outcome after arthroscopic full-thickness rotator cuff repair: Single-row versus dual-row fixation. Arthroscopy 2005; 21:1307-1316. 25. Matthews TJ, Smith SR, Peach CA, Ress JL, Urban JP, Carr AJ. In vivo measurement of tissue metabolism in tendons of the rotator cuff. J Bone Joint Surg Br 2007;89:633-638. 26. Hashimoto T, Nobuhara K, Hamada T. Pathologic evidence of degeneration as a primary cause of rotator cuff tear. Clin Orthop Relat Res 2003:111-120. 27. Matthews TJ, Hand GC, Ress JL, Athanasou NA, Carr AJ. Pathology of the torn rotator cuff tendon: Reduction in potential for repair as tear size increases. J Bone Joint Surg Br 2006;88:489-495. 28. Schmutz S, Fuchs T, Regenfelder F, et al. Expression of atrophy mRNA relates to tendon tear size in supraspinatus muscle. Clin Orthop Relat Res 2009;467:457-464. 29. Gazielly DF, Gleyze P, Montagnon C. Functional and anatomical results after rotator cuff repair. Clin Orthop Relat Res 1994:43-53. 30. Oizumi N, Suenaga N, Hisada Y, Minami A, Fukuda K. Postoperative changes of MRI findings over time after rotator cuff repair. Shoulder Joint 2007;31:261-264. 31. Ohnada Y, Matsumoto D, Fukuda K, Oizumi N, Sugimoto H. Postoperative changes of MRI findings after rotator cuff repair within 1 year and over 1 year. Shoulder Joint 2007;34:427-430. 32. Neuberger A, Slack HG. The metabolism of collagen from liver, bone, skin and tendon in the normal rat. Biochem J 1953;53:47-52. 33. Marchuk L, Sciore P, Reno C, Frank CB, Hart DA. Postmortem stability of total RNA isolated from rabbit ligament, tendon and cartilage. Biochim Biophys Acta 1998;1379:171177.