- Email: [email protected]
Changes in the long head of the biceps tendon in rotator cuff tear shoulders
Clinical Biomechanics 20 (2005) 162–165 www.elsevier.com/locate/clinbiomech
Changes in the long head of the biceps tendon in rotator cuff tear shoulders James E. Carpenter *, Jason D. Wening, Amy G. Mell, Joseph E. Langenderfer, John E. Kuhn, Richard E. Hughes MedSport and Orthopaedic Research Laboratories, Department of Orthopaedic Surgery, The University of Michigan, 400 N. Ingalls Building Ann Arbor, MI 48109-0486, USA Received 8 August 2003; accepted 24 September 2004
Abstract Background. Morphologic changes in the long head of the biceps tendon have been described in association with rotator cuff disease, yet mechanical significance of these changes remains unclear. Methods. An experiment was designed to test the hypotheses that the cross-sectional area and material properties of the long head of the biceps tendon are different in shoulders with full thickness rotator cuff tears and shoulders with intact rotator cuff tendons. Seven pairs of cadaver shoulders were tested. In each pair one shoulder had a full thickness rotator cuff tear and the other did not. Thus, a matched design was used. Cross sectional areas were measured. Tendon material properties were measured using an optical strain system. Findings. We were unable to detect a statistically significant difference in the long head of the biceps area or material properties between tendons in shoulders with and without rotator cuff tears. An a priori power analysis was conducted indicating the sample size was sufficient to detect a difference of 70 MPa in the elastic modulus measurement. Interpretation. Our data indicate there is no difference in the long head of the biceps cross sectional area or material properties. Therefore, the long head of the biceps tendon appears to retain its material properties in the presence of a rotator cuff tear. The clinical significance of this finding is that the long head of the biceps can be retained in the presence of a rotator cuff tear without concern that mechanical properties have substantially deteriorated. Ó 2004 Elsevier Ltd. All rights reserved. Keywords: Biceps; Shoulder; Biomechanics; Rotator cuff tear
1. Introduction Changes in the long head of biceps (LHB) tendon have been widely reported in association with rotator cuff (RC) tears. Widening of the tendon is commonly described at the time of operation for rotator cuff repairs. Murthi et al. (2000) reported pathologic changes in the LHB to be directly proportional to the extent of rotator
*
Corresponding author. Address: MedSport, 24 Frank Lloyd Wright Dr. Ann Arbor, MI 48106, USA. E-mail address: [email protected] (J.E. Carpenter). 0268-0033/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2004.09.009
cuff disease (Murthi et al., 2000). If the changes are primarily those of chronic inflammation and degeneration, then deterioration or weakening of the tendon might be predicted. Such deterioration may help explain the high risk for LHB rupture in shoulders with full thickness cuff tears. However, it is possible that instead of degeneration, the changes seen are due to only a morphologic response to increased loading of the LHB tendon in the setting of a rotator cuff tear. The clinical relevance of LHB mechanical changes lies in surgical treatment of rotator cuff tears. It may be acceptable to remove a degenerated LHB tendon when performing a rotator cuff repair because it may
J.E. Carpenter et al. / Clinical Biomechanics 20 (2005) 162–165
be prone to rupture. However, the LHB can play an important role in stabilizing the shoulder (Itoi et al., 1993, Itoi et al., 1994, Malicky et al., 1996, Warner and McMahon, 1995). This function may be even more important in the presence of a rotator cuff tear (Kido et al., 2000). Maintenance of mechanical properties of the LHB tendon in the face of a cuff tear would support the argument for preserving the LHB tendon when performing RC tear repair. The purpose of this study was to test the hypothesis that the cross-sectional area and material properties of the LHB differ in cadavers shoulders with and without RC tears.
2. Methods 2.1. Specimens Over a five-year period, 414 fresh frozen human cadaver shoulders were collected through the University of Michigan Anatomical Donations Program. Of these, 57 (14%) shoulders were found to have full thickness rotator cuff tears and 357 were without full thickness cuff tears. Details of these tears have been reported (Wening et al., 2002). A subgroup of this sample consisted of unilateral cuff tear shoulders with a contralateral intact rotator cuff that could serve as a matched control (n = 7 shoulder pairs). 2.2. Geometric measurements and mechanical testing During the collection period each shoulder was stored frozen at 20 °C. Later, each shoulder was thawed and soft tissue was removed, leaving the humerus and scapula along with the rotator cuff tendons (infraspinatus, teres minor, subscapularis, and supraspinatus), the capsule, and the biceps tendon intact. The rotator cuff was inspected for a full thickness tear and re-frozen. They were thawed one last time immediately prior to mechanical testing. During measurement and testing the tissue was kept hydrated with saline. The long head of the biceps tendon was transected approximately 10 mm proximal to the biceps muscle and the labrum was disarticulated from the glenoid. Being careful not to damage the specimen, the biceps tendon was pulled through the inter-tubercular groove. Tendon thickness was measured using a custom built device consisting of an indenter probe attached to a high-resolution displacement transducer that moves perpendicularly to a flat plate. This device was calibrated prior to each use resulting in accuracy on the order of 10 lm. The specimens were cut longitudinally to form a 3 mm by 15 mm dumbbell shape using a tendon cutter custom made from dermatome blades. Due to the small thickness, the cross sectional area was assumed to be a
163
rectangle and calculated as the product of width and thickness. The biceps tendon specimen was mounted in a servohydraulic-testing machine (MTS Systems Corporation, Minneapolis, Minn., USA) using custom designed cylindrical cryoclamps. The ends of the biceps tendon were placed in the center of the hollowed out cylinders containing saline. Liquid nitrogen was used to freeze the saline and secure the tendons. After mounting, the load cell was zeroed, and then a nominal pre-load was applied to take the slack out of the tendon. Using Verhoeff stain six equally spaced parallel lines were marked on the dumb bell shaped region perpendicular to the tendon fibers. The tendon was then preconditioned with ten cyclic cycles of 2% strain followed by relaxation. The specimen was finally loaded to failure in uniaxial tension at a rate of 100 mm/min. Data were collected by a microcomputer and an optical strain measurement system, which has been previously described in detail (Derwin et al., 1994). This optical system consists of a video camera with a 10– 140 mm zoom lens connected to a video cassette recorder illustrated schematically in Fig. 1. The camera was focused to obtain close-up images of the biceps tendon, with the stain lines lying vertically in the image. The audio channel was used to time-code the videotape so frames could be synchronized to the load cell measurements. The entire test of each specimen was recorded. Video post-processing was performed to determine the strain from the Verhoeff stain lines on the tendon. Video frames were captured using a frame grabbing board in a Macintosh computer, resulting in 256 greyscale images. Regions of interest, which contained the stain lines, were interactively defined by the analyst for
Fig. 1. Schematic of non-contact, optical strain measurement system.
164
J.E. Carpenter et al. / Clinical Biomechanics 20 (2005) 162–165
each captured image. The same set of lines were selected in each image when possible. Image processing consisted of six steps: (1) the image was smoothed by convolving it with the function f(x) = exp( bx2), where b was a constant based on image quality; (2) the gradient of the image was calculated; (3) the gradient was thresholded; (4) the gradient of the image was sampled along rows of pixels parallel to the direction of strain; (5) intersections with edge of each stain line were determined by computing a weighted average of locations having highest slope; and (6) the average of left and right edges of each line were computed to estimate the location of each stain line. Strain was computed as e = (Ln L0)/L0 where Ln is length at frame n and L0 is length after preconditioning. Standard procedures were used for calculating the tendonÕs material properties. Ultimate stress was computed by dividing the ultimate load by the cross-sectional area. It should be noted that the cross-sectional area was calculated using the 3 mm width of the dumbbell shape cutter and the thickness prior to cutting the tendon. The modulus was determined as the slope of the stress versus strain curve in the linear region between 20% and 50% of the ultimate load. 2.3. Statistical analysis A nonparametric sign test was used to test the hypothesis that the median elastic modulus of the biceps tendon in shoulders with RC tears was different from the modulus of the biceps tendon in shoulders without rotator cuff tears. The statistical power of the matched analysis was computed a priori from a pilot study of two unpaired shoulders. We assumed that the elastic modulus of biceps tendons within individuals was uncorrelated, and we used the variance estimate from the pilot study to compute the power. We found an 80% chance of detecting a difference of 70 MPa when controlling for Type I error at the p = 0.05 level of significance. Power calculations were performed using nQuery Advisor (Statistical Solutions, Inc., Cork, Ireland). The effect size was selected to provide 80% power given the available number of shoulders; it was not chosen based on forces seen clinically.
3. Results Analysis of the geometric properties of the LHB tendon demonstrated a cross-sectional area of 0.20 cm2 in RC tear shoulders versus 0.15 cm2 in tendons from shoulders without RC tears. This difference was not statistically significant (p = 0.5). The mechanical testing demonstrated a tissue modulus of 483 MPa in RC tear shoulders versus 629 MPa in tendons from shoulders
Table 1 Results of mechanical testing Measurement (SD) 2
X-section area (cm ) Modulus (Mpa) Max load (N)a Max stress (Mpa)a Max strain a
Cuff tear (n = 7)
No cuff tear (n = 7)
0.20 (0.13) 483 (225) 303 (74.8) 43.3 (19.4) 0.25 (0.19)
0.15 (0.04) 629 (230) 305 (96.9) 45.1 (19.6) 0.18 (0.13)
Cuff tear (n = 5), No cuff tear (n = 4).
Fig. 2. Cross-sectional area versus modulus for N = 7 pairs. The capital letter represents the shoulder with the cuff tear while the lowercase letter represents the contralateral side. Each set of shoulders (N = 7) is represented with a different letter, A–E for the cuff tear side and a–e for the contralateral side.
without RC tears (Table 1). The difference between these values, as well as the other material properties, was not statistically significant (p = 0.5). A scatter plot of crosssectional area versus modulus for all specimens is presented in Fig. 2.
4. Discussion The findings from our study suggest that the LHB biceps tendon maintains its material properties in the presence of a full thickness rotator cuff tear. It is known that rate of rupture of the LHB tendon is much higher in shoulders with cuff tears than those without tears. Logically, this increased risk for tears is due to either a deterioration of the LHB mechanical strength or higher tendon loads due to alterations in the mechanical environment. From this studyÕs results, it appears that the loading environment is more responsible for this risk for rupture than is a deterioration of tendon properties. The clinical significance of this finding is that, from a purely mechanical viewpoint, the LHB can be retained in the presence of a RC tear without concern that mechanical properties have substantially deteriorated. Unfortunately, with this biomechanical study we are not able to assess the contribution of the LHB tendon
J.E. Carpenter et al. / Clinical Biomechanics 20 (2005) 162–165
to the dysfunction or pain of the shoulder with a full thickness rotator cuff tear. Clinically, these factors may outweigh the importance of the mechanical properties. Previous biomechanical analysis of the LHB tendon reported a modulus of 421 MPa compared to our findings of 483 MPa and 629 MPa in the tear and no tear groups respectively (McGough et al., 1996). The condition of the rotator cuff tendons in the shoulders from that study was not reported. These differences are relatively small and can be explained by differences in testing techniques and location along the tendon. The difficulties in mechanical testing highlight some of the limitations of this study. First of all, we were not able to measure structural properties of the full LHB tendon. We chose a site for testing at the proximal portion of the bicipital groove as an area of high mechanical loading. In order to achieve failure data for the tendon tissue and minimize the effects of tendon gripping, we choose to test a narrowed down section of the LHB tendon. Thus we measured the properties of the central portion of the tendon only. It is possible that other areas of the tendon might have demonstrated significant differences. Our study protocol required freezing and thawing tissue, which could theoretically alter tissue mechanical properties. However, others have shown that freezing and thawing tendon up to five times does not alter mechanical properties (Smith et al., 1996). Moreover, any freeze–thaw effect would apply equally to all shoulders, which would be accounted for using the data analysis methods for a matched design. Perhaps the most significant limitation of the study was the small sample size. We anticipated this potential pitfall and conducted a pilot study and performed a statistical power analysis. Our pilot testing suggested that we did have 80% power to determine a 70 MPa difference if one had been present. Unfortunately, the variability of the data from our actual study was greater than it was in our pilot study. We would have preferred to use a larger sample, but we have found the pairing of a shoulder with a cuff tear and a shoulder without a cuff tear on the same cadaver to be uncommon. In our sample of 414 shoulders, only 3.4% satisfied this requirement. Therefore collecting a larger number of these specimens is not practical. An important alternative explanation for the fact that we found no difference between the material properties of the LHB tendon in shoulders with and without full thickness cuff tear is that the LHB tendon is altered in both shoulders. Murthi et al. (2000) has demonstrated significant pathologic changes in the LHB in 89% of
165
shoulders with full thickness cuff tears as well as in 75% of shoulders undergoing surgical treatment of impingement syndrome, with no cuff tears. Therefore, the LHB tendons in our contralateral control shoulders without cuff tears may not be normal tendons. Ideally this could be addressed by testing a group of normal tendons as controls. Unfortunately, it is not likely that a control group of disease-free, normal tendons that would match the age of the shoulder in this study exists.
Acknowledgments The authors would like to thank Laurie Huston, Mark Stock and Dennis Kayner. Financial support was provided by the National Defense Science and Engineering Graduate Fellowship (JDW).
References Derwin, K.A., Soslowsky, L.J., Green, W.D.K., Elder, S.H., 1994. A new optical system for the determination of deformations and strains: Calibration characteristics and experimental results. Journal of Biomechanics 27, 1277–1285. Itoi, E., Kuechle, D.K., Newman, S.R., Morrey, B.F., An, K.-N., 1993. Stabilising function of the biceps in stable and unstable shoulders. The Journal of Bone and Joint Surgery (Br) 75-B, 546– 550. Itoi, E., Newman, S.R., Kuechle, D.K., Morrey, B.F., An, K.-N., 1994. Dynamic anterior stabilisers of the shoulder with the arm in abduction. The Journal of Bone and Joint Surgery (Br) 76-B, 834– 836. Kido, T., Itoi, E., Konno, N., Sano, A., Urayama, M., Sato, K., 2000. The depressor function of biceps on the head of the humerus in shoulders with tears of the rotator cuff. Journal of Bone and Joint Surgery (Br) 82, 416–419. Malicky, D.M., Soslowsky, L.J., Blasier, R.B., Shyr, Y., 1996. Anterior glenohumeral stabilization factors: progressive effects in a biomechanical model. Journal of Orthopaedic Research 14, 282– 288. McGough, R.L., Debski, R.E., Taskiran, E., Fu, F.H., Woo, S.L.-Y., 1996. Mechanical properties of the long head of the biceps tendon. Knee Surgery, Sports Traumatology, Arthroscopy 3, 226–229. Murthi, A., Vosburgh, C., Neviaser, T., 2000. The incidence of pathologic changes of the long head of the biceps tendon. Journal of Shoulder and Elbow Surgery 9, 382–385. Smith, C.W., Young, I.S., Kearney, J.N., 1996. Mechanical Properties of Tendons: Changes With Sterilization and Preservation. Journal of Biomechanical Engineering 118, 56–61. Warner, J., McMahon, P., 1995. The role of the long head of the biceps brachii in superior stability of the glenohumeral joint. Journal of Bone and Joint Surgery 77, 366–372. Wening, J., Hollis, R., Hughes, R.E., Kuhn, J.E., 2002. Quantitative morphology of full thickness rotator cuff tears. Clinical Anatomy 15, 18–22.
Changes in the long head of the biceps tendon in rotator cuff tear shoulders James E. Carpenter *, Jason D. Wening, Amy G. Mell, Joseph E. Langenderfer, John E. Kuhn, Richard E. Hughes MedSport and Orthopaedic Research Laboratories, Department of Orthopaedic Surgery, The University of Michigan, 400 N. Ingalls Building Ann Arbor, MI 48109-0486, USA Received 8 August 2003; accepted 24 September 2004
Abstract Background. Morphologic changes in the long head of the biceps tendon have been described in association with rotator cuff disease, yet mechanical significance of these changes remains unclear. Methods. An experiment was designed to test the hypotheses that the cross-sectional area and material properties of the long head of the biceps tendon are different in shoulders with full thickness rotator cuff tears and shoulders with intact rotator cuff tendons. Seven pairs of cadaver shoulders were tested. In each pair one shoulder had a full thickness rotator cuff tear and the other did not. Thus, a matched design was used. Cross sectional areas were measured. Tendon material properties were measured using an optical strain system. Findings. We were unable to detect a statistically significant difference in the long head of the biceps area or material properties between tendons in shoulders with and without rotator cuff tears. An a priori power analysis was conducted indicating the sample size was sufficient to detect a difference of 70 MPa in the elastic modulus measurement. Interpretation. Our data indicate there is no difference in the long head of the biceps cross sectional area or material properties. Therefore, the long head of the biceps tendon appears to retain its material properties in the presence of a rotator cuff tear. The clinical significance of this finding is that the long head of the biceps can be retained in the presence of a rotator cuff tear without concern that mechanical properties have substantially deteriorated. Ó 2004 Elsevier Ltd. All rights reserved. Keywords: Biceps; Shoulder; Biomechanics; Rotator cuff tear
1. Introduction Changes in the long head of biceps (LHB) tendon have been widely reported in association with rotator cuff (RC) tears. Widening of the tendon is commonly described at the time of operation for rotator cuff repairs. Murthi et al. (2000) reported pathologic changes in the LHB to be directly proportional to the extent of rotator
*
Corresponding author. Address: MedSport, 24 Frank Lloyd Wright Dr. Ann Arbor, MI 48106, USA. E-mail address: [email protected] (J.E. Carpenter). 0268-0033/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2004.09.009
cuff disease (Murthi et al., 2000). If the changes are primarily those of chronic inflammation and degeneration, then deterioration or weakening of the tendon might be predicted. Such deterioration may help explain the high risk for LHB rupture in shoulders with full thickness cuff tears. However, it is possible that instead of degeneration, the changes seen are due to only a morphologic response to increased loading of the LHB tendon in the setting of a rotator cuff tear. The clinical relevance of LHB mechanical changes lies in surgical treatment of rotator cuff tears. It may be acceptable to remove a degenerated LHB tendon when performing a rotator cuff repair because it may
J.E. Carpenter et al. / Clinical Biomechanics 20 (2005) 162–165
be prone to rupture. However, the LHB can play an important role in stabilizing the shoulder (Itoi et al., 1993, Itoi et al., 1994, Malicky et al., 1996, Warner and McMahon, 1995). This function may be even more important in the presence of a rotator cuff tear (Kido et al., 2000). Maintenance of mechanical properties of the LHB tendon in the face of a cuff tear would support the argument for preserving the LHB tendon when performing RC tear repair. The purpose of this study was to test the hypothesis that the cross-sectional area and material properties of the LHB differ in cadavers shoulders with and without RC tears.
2. Methods 2.1. Specimens Over a five-year period, 414 fresh frozen human cadaver shoulders were collected through the University of Michigan Anatomical Donations Program. Of these, 57 (14%) shoulders were found to have full thickness rotator cuff tears and 357 were without full thickness cuff tears. Details of these tears have been reported (Wening et al., 2002). A subgroup of this sample consisted of unilateral cuff tear shoulders with a contralateral intact rotator cuff that could serve as a matched control (n = 7 shoulder pairs). 2.2. Geometric measurements and mechanical testing During the collection period each shoulder was stored frozen at 20 °C. Later, each shoulder was thawed and soft tissue was removed, leaving the humerus and scapula along with the rotator cuff tendons (infraspinatus, teres minor, subscapularis, and supraspinatus), the capsule, and the biceps tendon intact. The rotator cuff was inspected for a full thickness tear and re-frozen. They were thawed one last time immediately prior to mechanical testing. During measurement and testing the tissue was kept hydrated with saline. The long head of the biceps tendon was transected approximately 10 mm proximal to the biceps muscle and the labrum was disarticulated from the glenoid. Being careful not to damage the specimen, the biceps tendon was pulled through the inter-tubercular groove. Tendon thickness was measured using a custom built device consisting of an indenter probe attached to a high-resolution displacement transducer that moves perpendicularly to a flat plate. This device was calibrated prior to each use resulting in accuracy on the order of 10 lm. The specimens were cut longitudinally to form a 3 mm by 15 mm dumbbell shape using a tendon cutter custom made from dermatome blades. Due to the small thickness, the cross sectional area was assumed to be a
163
rectangle and calculated as the product of width and thickness. The biceps tendon specimen was mounted in a servohydraulic-testing machine (MTS Systems Corporation, Minneapolis, Minn., USA) using custom designed cylindrical cryoclamps. The ends of the biceps tendon were placed in the center of the hollowed out cylinders containing saline. Liquid nitrogen was used to freeze the saline and secure the tendons. After mounting, the load cell was zeroed, and then a nominal pre-load was applied to take the slack out of the tendon. Using Verhoeff stain six equally spaced parallel lines were marked on the dumb bell shaped region perpendicular to the tendon fibers. The tendon was then preconditioned with ten cyclic cycles of 2% strain followed by relaxation. The specimen was finally loaded to failure in uniaxial tension at a rate of 100 mm/min. Data were collected by a microcomputer and an optical strain measurement system, which has been previously described in detail (Derwin et al., 1994). This optical system consists of a video camera with a 10– 140 mm zoom lens connected to a video cassette recorder illustrated schematically in Fig. 1. The camera was focused to obtain close-up images of the biceps tendon, with the stain lines lying vertically in the image. The audio channel was used to time-code the videotape so frames could be synchronized to the load cell measurements. The entire test of each specimen was recorded. Video post-processing was performed to determine the strain from the Verhoeff stain lines on the tendon. Video frames were captured using a frame grabbing board in a Macintosh computer, resulting in 256 greyscale images. Regions of interest, which contained the stain lines, were interactively defined by the analyst for
Fig. 1. Schematic of non-contact, optical strain measurement system.
164
J.E. Carpenter et al. / Clinical Biomechanics 20 (2005) 162–165
each captured image. The same set of lines were selected in each image when possible. Image processing consisted of six steps: (1) the image was smoothed by convolving it with the function f(x) = exp( bx2), where b was a constant based on image quality; (2) the gradient of the image was calculated; (3) the gradient was thresholded; (4) the gradient of the image was sampled along rows of pixels parallel to the direction of strain; (5) intersections with edge of each stain line were determined by computing a weighted average of locations having highest slope; and (6) the average of left and right edges of each line were computed to estimate the location of each stain line. Strain was computed as e = (Ln L0)/L0 where Ln is length at frame n and L0 is length after preconditioning. Standard procedures were used for calculating the tendonÕs material properties. Ultimate stress was computed by dividing the ultimate load by the cross-sectional area. It should be noted that the cross-sectional area was calculated using the 3 mm width of the dumbbell shape cutter and the thickness prior to cutting the tendon. The modulus was determined as the slope of the stress versus strain curve in the linear region between 20% and 50% of the ultimate load. 2.3. Statistical analysis A nonparametric sign test was used to test the hypothesis that the median elastic modulus of the biceps tendon in shoulders with RC tears was different from the modulus of the biceps tendon in shoulders without rotator cuff tears. The statistical power of the matched analysis was computed a priori from a pilot study of two unpaired shoulders. We assumed that the elastic modulus of biceps tendons within individuals was uncorrelated, and we used the variance estimate from the pilot study to compute the power. We found an 80% chance of detecting a difference of 70 MPa when controlling for Type I error at the p = 0.05 level of significance. Power calculations were performed using nQuery Advisor (Statistical Solutions, Inc., Cork, Ireland). The effect size was selected to provide 80% power given the available number of shoulders; it was not chosen based on forces seen clinically.
3. Results Analysis of the geometric properties of the LHB tendon demonstrated a cross-sectional area of 0.20 cm2 in RC tear shoulders versus 0.15 cm2 in tendons from shoulders without RC tears. This difference was not statistically significant (p = 0.5). The mechanical testing demonstrated a tissue modulus of 483 MPa in RC tear shoulders versus 629 MPa in tendons from shoulders
Table 1 Results of mechanical testing Measurement (SD) 2
X-section area (cm ) Modulus (Mpa) Max load (N)a Max stress (Mpa)a Max strain a
Cuff tear (n = 7)
No cuff tear (n = 7)
0.20 (0.13) 483 (225) 303 (74.8) 43.3 (19.4) 0.25 (0.19)
0.15 (0.04) 629 (230) 305 (96.9) 45.1 (19.6) 0.18 (0.13)
Cuff tear (n = 5), No cuff tear (n = 4).
Fig. 2. Cross-sectional area versus modulus for N = 7 pairs. The capital letter represents the shoulder with the cuff tear while the lowercase letter represents the contralateral side. Each set of shoulders (N = 7) is represented with a different letter, A–E for the cuff tear side and a–e for the contralateral side.
without RC tears (Table 1). The difference between these values, as well as the other material properties, was not statistically significant (p = 0.5). A scatter plot of crosssectional area versus modulus for all specimens is presented in Fig. 2.
4. Discussion The findings from our study suggest that the LHB biceps tendon maintains its material properties in the presence of a full thickness rotator cuff tear. It is known that rate of rupture of the LHB tendon is much higher in shoulders with cuff tears than those without tears. Logically, this increased risk for tears is due to either a deterioration of the LHB mechanical strength or higher tendon loads due to alterations in the mechanical environment. From this studyÕs results, it appears that the loading environment is more responsible for this risk for rupture than is a deterioration of tendon properties. The clinical significance of this finding is that, from a purely mechanical viewpoint, the LHB can be retained in the presence of a RC tear without concern that mechanical properties have substantially deteriorated. Unfortunately, with this biomechanical study we are not able to assess the contribution of the LHB tendon
J.E. Carpenter et al. / Clinical Biomechanics 20 (2005) 162–165
to the dysfunction or pain of the shoulder with a full thickness rotator cuff tear. Clinically, these factors may outweigh the importance of the mechanical properties. Previous biomechanical analysis of the LHB tendon reported a modulus of 421 MPa compared to our findings of 483 MPa and 629 MPa in the tear and no tear groups respectively (McGough et al., 1996). The condition of the rotator cuff tendons in the shoulders from that study was not reported. These differences are relatively small and can be explained by differences in testing techniques and location along the tendon. The difficulties in mechanical testing highlight some of the limitations of this study. First of all, we were not able to measure structural properties of the full LHB tendon. We chose a site for testing at the proximal portion of the bicipital groove as an area of high mechanical loading. In order to achieve failure data for the tendon tissue and minimize the effects of tendon gripping, we choose to test a narrowed down section of the LHB tendon. Thus we measured the properties of the central portion of the tendon only. It is possible that other areas of the tendon might have demonstrated significant differences. Our study protocol required freezing and thawing tissue, which could theoretically alter tissue mechanical properties. However, others have shown that freezing and thawing tendon up to five times does not alter mechanical properties (Smith et al., 1996). Moreover, any freeze–thaw effect would apply equally to all shoulders, which would be accounted for using the data analysis methods for a matched design. Perhaps the most significant limitation of the study was the small sample size. We anticipated this potential pitfall and conducted a pilot study and performed a statistical power analysis. Our pilot testing suggested that we did have 80% power to determine a 70 MPa difference if one had been present. Unfortunately, the variability of the data from our actual study was greater than it was in our pilot study. We would have preferred to use a larger sample, but we have found the pairing of a shoulder with a cuff tear and a shoulder without a cuff tear on the same cadaver to be uncommon. In our sample of 414 shoulders, only 3.4% satisfied this requirement. Therefore collecting a larger number of these specimens is not practical. An important alternative explanation for the fact that we found no difference between the material properties of the LHB tendon in shoulders with and without full thickness cuff tear is that the LHB tendon is altered in both shoulders. Murthi et al. (2000) has demonstrated significant pathologic changes in the LHB in 89% of
165
shoulders with full thickness cuff tears as well as in 75% of shoulders undergoing surgical treatment of impingement syndrome, with no cuff tears. Therefore, the LHB tendons in our contralateral control shoulders without cuff tears may not be normal tendons. Ideally this could be addressed by testing a group of normal tendons as controls. Unfortunately, it is not likely that a control group of disease-free, normal tendons that would match the age of the shoulder in this study exists.
Acknowledgments The authors would like to thank Laurie Huston, Mark Stock and Dennis Kayner. Financial support was provided by the National Defense Science and Engineering Graduate Fellowship (JDW).
References Derwin, K.A., Soslowsky, L.J., Green, W.D.K., Elder, S.H., 1994. A new optical system for the determination of deformations and strains: Calibration characteristics and experimental results. Journal of Biomechanics 27, 1277–1285. Itoi, E., Kuechle, D.K., Newman, S.R., Morrey, B.F., An, K.-N., 1993. Stabilising function of the biceps in stable and unstable shoulders. The Journal of Bone and Joint Surgery (Br) 75-B, 546– 550. Itoi, E., Newman, S.R., Kuechle, D.K., Morrey, B.F., An, K.-N., 1994. Dynamic anterior stabilisers of the shoulder with the arm in abduction. The Journal of Bone and Joint Surgery (Br) 76-B, 834– 836. Kido, T., Itoi, E., Konno, N., Sano, A., Urayama, M., Sato, K., 2000. The depressor function of biceps on the head of the humerus in shoulders with tears of the rotator cuff. Journal of Bone and Joint Surgery (Br) 82, 416–419. Malicky, D.M., Soslowsky, L.J., Blasier, R.B., Shyr, Y., 1996. Anterior glenohumeral stabilization factors: progressive effects in a biomechanical model. Journal of Orthopaedic Research 14, 282– 288. McGough, R.L., Debski, R.E., Taskiran, E., Fu, F.H., Woo, S.L.-Y., 1996. Mechanical properties of the long head of the biceps tendon. Knee Surgery, Sports Traumatology, Arthroscopy 3, 226–229. Murthi, A., Vosburgh, C., Neviaser, T., 2000. The incidence of pathologic changes of the long head of the biceps tendon. Journal of Shoulder and Elbow Surgery 9, 382–385. Smith, C.W., Young, I.S., Kearney, J.N., 1996. Mechanical Properties of Tendons: Changes With Sterilization and Preservation. Journal of Biomechanical Engineering 118, 56–61. Warner, J., McMahon, P., 1995. The role of the long head of the biceps brachii in superior stability of the glenohumeral joint. Journal of Bone and Joint Surgery 77, 366–372. Wening, J., Hollis, R., Hughes, R.E., Kuhn, J.E., 2002. Quantitative morphology of full thickness rotator cuff tears. Clinical Anatomy 15, 18–22.