- Email: [email protected]
Apoptosis in Adhesions and the Adhesion-Tendon Gliding Interface: Relationship to Adhesion-Tendon Gliding Mechanics
SCIENTIFIC ARTICLE
Apoptosis in Adhesions and the Adhesion-Tendon Gliding Interface: Relationship to Adhesion-Tendon Gliding Mechanics Ya Fang Wu, MD, Jin Bo Tang, MD Purpose Adhesion formation is closely related to tendon-gliding function. We aimed to investigate apoptosis (programmed cell death) in adhesions and tendons and study its relationship to the mechanics of adhesions and healing tendons. Methods The flexor digitorum profundus tendons of 30 long toes in 15 chickens were completely transected and repaired surgically. At postoperative weeks 4, 6, and 8, tendon-gliding excursions were tested and adhesion scores were recorded. Tendons and surrounding adhesions were then harvested for analysis of apoptosis using in situ terminal deoxynucleotidyl transferase dUTP (deoxyuridine triphosphate) nick end labeling assay. Three-dimensional image reconstruction was used to provide an overall view of cellular distribution in tendons and adhesions. Finally, we analyzed the correlation between the apoptotic index measured at the adhesions and the gliding excursions. Ten uninjured tendons served as normal controls. Results Apoptosis was found to be a dominant cellular event in the adhesion tissues at both the adhesion-tendon gliding interface and the adhesion core. The apoptotic index in the adhesions was generally above 20% to 50%. The apoptotic index was significantly higher in the adhesions than in the junction region of the cut tendon ends at weeks 4, 6, and 8. A higher apoptotic index in the adhesions significantly correlated to lower tendon excursions at week 6. Conclusions Apoptosis in adhesions and at the adhesion-tendon interface is a prominent event in the tendon-healing process. The tendons exhibiting a lower tendon-gliding amplitude, meaning more severe adhesions, tended to have a greater apoptotic index in their adhesions during a certain period of the tendon-remodeling process. Clinical relevance Apoptosis in the adhesions and at the adhesion-tendon interface may contribute remarkably to the fate of adhesions and the restoration of the tendon gliding surface, which may be closely related to the tendon function. (J Hand Surg 2013;38A:1071–1078. Copyright © 2013 by the American Society for Surgery of the Hand. All rights reserved.) Key words Tendon healing, apoptosis, adhesions, gliding interface, gliding excursions. DHESION FORMATION AFTER intrasynovial tendon surgery is a major clinical concern because it increases scarring of the tendon and limits active finger motion, leading to poor clinical outcomes.1–3
A
From the Hand Surgery Research Center, Department of Hand Surgery, Affiliated Hospital of Nantong University, Nantong, Jiangsu, China. Received for publication September 11, 2012; accepted in revised form March 4, 2013. Supported by grants from the Health Bureau of Jiangsu Province and the Department of Hand Surgery, Nantong University, Natural Science Foundation of China (No. 81030035 and 81271985).
After the early stages of tendon healing, the gliding movement of the tendon gradually improves and the smooth surface of the tendon is restored. Apoptosis (programmed cell death) was found to be associated Corresponding author: Jin Bo Tang, MD, Department of Hand Surgery, Affiliated Hospital of Nantong University, 20 West Temple Road, Nantong 226001, Jiangsu, China; e-mail:[email protected]. 0363-5023/13/38A06-0002$36.00/0 http://dx.doi.org/10.1016/j.jhsa.2013.03.012
No benefits in any form have been received or will be received related directly or indirectly to the subject of this article.
© ASSH 䉬 Published by Elsevier, Inc. All rights reserved. 䉬 1071
1072
APOPTOSIS IN ADHESIONS AND INTERFACE
with tendon healing, especially immediately after trauma and during the remodeling phase of tendon healing.4,5 However, we do not yet know how apoptosis occurs within adhesion tissues. After intrasynovial chicken tendons were completely cut and repaired, a bimodal distribution of cellular apoptosis was observed at postoperative days 3 to 7 and at days 28 to 56.4,5 High apoptotic reactions were also found during the late remodeling process of healing rat patellar tendon.6 A previous study investigated apoptotic changes in both healing tendon and subcutaneous tissues, using a 50% partial tendon laceration model in mice.7 In that study, apoptosis in the immediate vicinity of the lacerated tendon and subcutaneous tissue wound was found to increase gradually over 84 days after surgery. Though the subcutaneous tissues and tendon followed similar changes over time, more apoptotic cells were seen in the subcutaneous tissues than in the tendon. Little is known about the apoptotic changes of cells either in adhesions or at the adhesion-tendon gliding interface. Apoptosis might be an important factor affecting the mechanical properties of both adhesions and tendons and the remodeling of the tendon surface. We proposed to investigate apoptosis in different areas of adhesions during the remodeling of the healing tendon. We also sought to characterize the relationship between apoptotic changes in cells in adhesions, the severity of adhesion, and the gliding mechanics of the tendon. We hypothesized that more apoptosis would occur in adhesions than in the tendon substance during the recovery of the tendon gliding surface. MATERIALS AND METHODS Forty long toes (30 as surgical models and 10 as normal controls) of 20 white Leghorn chickens weighing 1.5 to 2.5 kg each were used. Institutional animal research regulations were followed during the study. The chickens were anesthetized with intramuscular injection of ketamine (50 mg/kg body weight). Operative procedures The surgery on the 30 long toes was performed using a thigh tourniquet and aseptic technique. A zigzag incision was made in the plantar skin between the proximal interphalangeal (PIP) joint and the distal interphalangeal joint (DIPJ) of the toe.8 –11 Through a 1-cm longitudinal incision in the sheath, the flexor digitorum profundus (FDP) tendon was transected completely and was surgically repaired with 5-0 sutures (Ethilon, Ethicon, Somerville, NJ) using a modified Kessler stitch. Running 6-0 epitendinous sutures were added. The sheath was not closed, and the skin was closed with
interrupted suture. The toe was immobilized in semiflexion using adhesive tape and a light casting until the end of postoperative week 3. The chickens were then allowed to walk freely. After 4, 6, and 8 weeks, the chickens were killed—5 chickens (10 tendons) at each time point. Five chickens (10 uninjured FDP tendons) served as day 0 normal controls. Test of tendon gliding and grading of the adhesions The toes were harvested from the knee joint level and secured to a tensile testing machine (Model 4411; Instron Corporation, Canton, MA). The metatarsophalangeal joint was fixed, and the other joints of the toes were left free. The FDP tendon was exposed from the knee joint to the ankle joint, and its proximal end was connected to a force transducer. After a preload of 0.1N had been applied to the FDP tendon, the digit was flexed until the force reached 15N. The gliding excursions of the FDP tendon at forces of 5N, 10N, and 15N were recorded. Then the tendon was exposed through a volar longitudinal incision in the toe. We carefully separated the normal subcutaneous tissues from adhesions and left the adhesions around the tendon. The length and density of the adhesions were evaluated, and the adhesions were scored according to established grading criteria.12–14 In Situ assay for apoptosis After biomechanical evaluation, as described, the same 10 tendons harvested from each time point were assayed for apoptosis. A 2-cm segment of tendon with adhesions from each toe was harvested with the laceration site at its center, and the samples were kept in 4% paraformaldehyde at 4°C for 48 hours. After gradient alcohol dehydration, the tendons were embedded in paraffin and cut longitudinally in 4m slices from volar to dorsal. We used the in situ terminal deoxynucleotidyl transferase dUTP (deoxyuridine triphosphate) nick end labeling (TUNEL) assay (11684817910; Roche, Mannheim, Germany) to measure apoptosis. After deparaffinization and rehydration, the tissue sections were pretreated with 10 g/mL proteinase K solution for 15 minutes at room temperature. Thereafter, slides were rinsed in PBS (phosphatebuffered saline) and incubated with TUNEL reaction mixture for 1 hour at 37°C in a humidified chamber. After washing, converter-peroxidase solution was applied, and the slides were incubated for 30 minutes at 37°C. The slides were washed and incubated for 10 minutes at room temperature after addition of 3,3=-diaminobenzidine (D-5905;
JHS 䉬 Vol A, June
APOPTOSIS IN ADHESIONS AND INTERFACE
1073
Sigma Aldrich, St. Louis, MO). As negative controls, sections were incubated in label solution without terminal dUTP. The sections were counterstained with hematoxylin and mounted under glass coverslips. Nuclei positive for apotosis stained brown, and those negative for apoptosis stained blue. Quantitation of apoptotic cells in adhesions and at the adhesion-tendon gliding interface Under 400⫻ magnification, we observed the adhesion core and the adhesion at the tendon gliding interface separately at the same horizontal level. In these 2 parts of the adhesions, we separately observed (1) the adhesions over the junction regions of the tendon ends about 2 mm each proximal and distal to the cut site and (2) the adhesions over the extended region of the tendon, approximately 2 mm to 5 mm in each direction from the tendon cut site.4 In each region, 3 fields of view were selected for calculating cells per tendon. An observer blinded to the observational time points counted the total cell numbers and the number of positively stained cells. An apoptotic index (AI) was calculated as the number of TUNEL-positive cells divided by the total number of cells.15–17 In each tendon sample, we also recorded the number of apoptotic cells in the surface and core of the tendon corresponding horizontally to the adhesion tissues to facilitate comparison to their AI to that of adhesions. Distribution of apoptotic cells in 3-dimensional views We applied the software (Reconstruct, Version 1.1.0.0) for 3-dimensional reconstruction to align sections stained with in situ TUNEL assay of these samples to observe the distribution of apoptotic cells in tendon and adhesions.7,18 Serial sections were calibrated and imported into the software program. Wire maps were produced by tracing individual structures of tendon and adhesions. Individual cells were mapped according to their stereological position. Boissonnat surface shading and color selection and transparency features were introduced after the 3-dimensional image reconstruction. Tendon area was highlighted in purple, adhesion in deep blue, apoptotic cells in cyan, and cells without morphological apoptotic changes in black. Statistical analysis For each time point, results of gliding excursions and AI were reported as an average and SD. One-way analysis of variance followed by post hoc Tukey test was used to compare the gliding excursions and AI between 3 different time points. The paired t-test was
FIGURE 1: The gliding excursion of tendon at the 5N, 10N, and 15N load, respectively.
used to compare the AI at 4 areas of the tendonadhesion complex in individual samples at the same time points. The correlation between the AI in the adhesions and the gliding excursion at the load of 15N was calculated using the Pearson test for each time point. We also analyzed whether the AI in the adhesions over the tendon junction region differed significantly according to scores of the adhesions using 1-way analysis of variance followed by Tukey test. The significance level of all comparisons was set at P ⬍ .05. RESULTS Gliding excursion and adhesion score The gliding excursions of the FDP tendon of the normal toes were 16.3 ⫾ 1.2 mm, 19.4 ⫾ 1.3 mm, and 21.4 ⫾ 0.9 mm at the gliding force of 5N, 10N, 15N, respectively, which were significantly greater than those at postoperative weeks 4, 6, and 8 with respective loads (P ⬍ .01). From weeks 4 to 8, gliding excursion gradually increased. However, significant differences were only found between weeks 4 and 8 (P ⬍ .05, all comparisons) (Fig. 1). The adhesion scores were 4.0 ⫾ 0.8, 3.7 ⫾ 0.5, and 3.6 ⫾ 0.8 for the tendons at weeks 4, 6, and 8, respectively. Apoptosis The AI at all the postoperative time points was significantly greater than that of the noninjury control group, among which the AI was 3% ⫾ 2%. One sample at week 4 was used to illustrate the apoptosis seen in adhesions and tendons (Fig. 2). Apoptosis at the adhesion core: At all time points, the greatest AI was at the adhesion core in the junction region.
JHS 䉬 Vol A, June
1074
APOPTOSIS IN ADHESIONS AND INTERFACE
FIGURE 2: Representative histological staining pictures from a week 4 sample. Left: histological sections illustrate the location of the adhesion core, adhesion-tendon interface, tendon surface, and tendon core in junction and extended regions. The regions of adhesions peripheral to the core and extended regions of the tendons are correspondingly named the core or the extended regions of adhesions, respectively. The apoptotic reactions were generally higher in the junction region than in the extended region for both adhesions and tendons and from adhesions to tendons (⫻ 400).
FIGURE 3: Apoptotic index (AI) of the adhesion core, tendon-adhesion surface, tendon surface, and tendon core. *Data significantly greater than those at week 6. †Data significantly greater than those at weeks 4 and 6.
The mean values of the AI in the adhesion core and adhesion-tendon interface were above 0.35 at all 3 time points. The AI in the adhesion core was significantly greater than that recorded at the tendon core (P ⬍ .05). At week 8, the mean AI was as high as 0.68 in the junction region and 0.58 in the extended region of the adhesion core. The AI in adhesions was significantly greater at week 8 than those at week 4 or 6 (P ⬍ .01). Apoptosis at the tendon-adhesion interface: Compared with the tendon core, the AI at the adhesion-tendon interface was significantly higher at all the time points in the junction and extended regions (Figs. 2, 3). At both
junction and extended regions, the AI at the tendonadhesion interface was significantly greater at week 8 than at either week 4 or week 6 (P ⬍ .05). Apoptosis in the tendon: Two regions, the tendon surface and core, were examined. In both regions, the AI was lower than that in the adhesions (Fig. 3). The AI in the junction region of the tendon was generally greater than in the extended region of the tendons (Fig. 3). Analysis of 3-dimesional reconstructed images We used 3-demesional reconstructed images to obtain global views of differences in distribution of apoptotic
JHS 䉬 Vol A, June
APOPTOSIS IN ADHESIONS AND INTERFACE
1075
FIGURE 4: A–C Histological sections show the tendon and surrounding adhesions from volar to dorsal aspect of the adhesiontendon samples at postoperative weeks 4, 6, and 8. D–F Corresponding 3-dimensional images of A–C built based on multiple histology slides depicting distribution of the apoptotic cells (cyan) in tendon (purple), and adhesions (deep blue) at postoperative weeks 4, 6, and 8. The cells without signs of apoptosis are shown in black (white arrow). Note the generally denser apoptotic cells at the tendon-adhesion interface and in the adhesion core. The blue arrowheads in A and D indicate the level of tendon transection.
cells in the tendon and adhesions and confirmed the substantially greater density of positive cells in the adhesion core and adhesion-tendon gliding interface compared with the tendon (Fig. 4). Correlation analysis of AI with tendon gliding excursions and adhesion scores The correlation plots of the tendon gliding excursions at the pulling force of 15N and AI of the adhesion-tendon interface and adhesion core are shown in Table 1. Overall, the AI correlated with the gliding excursion negatively. There was a significant negative correlation
between AI of the adhesions and the gliding excursion at week 6 (Table 1, Fig. 5). No statistically significant correlation was found for such comparisons at weeks 4 or 8. The samples with greater severity scores of peritendinous adhesions had significantly greater AI (P ⬍ .01) (Fig. 6). DISCUSSION The major focus of our present study was the investigation of apoptosis within adhesions around the tendon. At each time point, we documented more dramatic
JHS 䉬 Vol A, June
1076
APOPTOSIS IN ADHESIONS AND INTERFACE
TABLE 1. Correlation of Tendon Gliding Excursions With Aptotic Index in the Adhesion Interface or the Core at 3 Postoperative Time Points Adhesion-Tendon Interface Time Points (wk)
Sample Size
Junction Region
Adhesion Core
Extended Region
Junction Region
Extended Region
4
10
r ⫽ ⫺.117, P ⫽ .747
r ⫽ ⫺.102, P ⫽ .779
r ⫽ ⫺.077, P ⫽ .833
r ⫽ ⫺.053, P ⫽ .884
6
10
r ⫽ ⫺.817, P ⫽ .004*
r ⫽ ⫺.745, P ⫽ .013*
r ⫽ ⫺.781, P ⫽ .008*
r ⫽ ⫺.734, P ⫽ .016*
8
10
r ⫽ ⫺.353, P ⫽ .257
r ⫽ ⫺.598, P ⫽ .068
r ⫽ ⫺.219, P ⫽ .543
r ⫽ ⫺.566, P ⫽ .088
r, correlation coefficient. *Correlation with statistical significance.
FIGURE 5: Scatter plots of the apoptotic index (AI) in tendon-adhesion interface versus the gliding excursion under the load of 15N at week 6.
FIGURE 6: Comparison of the apoptotic index (AI) in the adhesions of different scores. *Data significantly greater than those of grade 3 or 4 at the adhesion core.
apoptotic changes in the adhesions than in the healing tendon. Regarding the time course of apoptotic changes, the AI was much higher in postoperative week 8 than in week 4 or 6. However, the tendon gliding
excursion increased from week 4 to week 8. The increased apoptotic reaction may relate to the clearance of excess cells participating in tissue repair during the earlier healing period, contributing to the recovery of tendon gliding.19,20 With regards to changes documented in different regions of adhesions, the apoptosis was more significant in the junction region than in the extended region. We further sought to characterize the apoptosis in the tendon and surrounding adhesions at the same horizontal level. The severity of apoptosis gradually increases from the tendon core, to the tendon surface, the adhesion-tendon interface, and the adhesion core. A previous study documented that healing tendons exhibit generally greater apoptosis in the tendon surface than in the core.4 The present result is in agreement with that previous report. More importantly, we found that AI was notably higher in the adhesions than in the tendon. No significant difference was found in AI between the adhesion core and the adhesion-tendon interface. This finding indicates that the tissues at the adhesion-tendon gliding interface did not undergo
JHS 䉬 Vol A, June
APOPTOSIS IN ADHESIONS AND INTERFACE
significantly more apoptosis than that seen at the center of the adhesions. The remodeling phase of tendon healing is responsible for cell number decrease, collagen realignment, and the restoration of a smooth surface to facilitate tendon gliding.1,21,22 During tendon mobilization, the healing tendon and adhesions bear the internal and external tensions caused by weight-bearing, muscle stretching during movements, and shear against surrounding tissues or the other pressures acting on tendon and adhesions. The adhesions in particular are subjected to greater shear force than the tendon. The remarkable increases in apoptosis in the adhesion and the tendonadhesion interface may be associated with the effect of shear deformation induced by mobilization. We may assume that more apoptosis in the tendon-adhesion interface would accelerate the recovery of the gliding tendon’s smooth surface and that less apoptosis would delay recovery. In addition, dissection revealed that the adhesions were relatively compact at weeks 4 and 6 and that the density decreased at week 8, which may also relate to the greater apoptosis in the adhesions during this period. We attempted to correlate the biomechanical findings with molecular apoptotic changes in each sample. Our correlation analysis showed negative correlations between gliding excursion and AI in the adhesions of the healing tendon with significant negative correlation at week 6. Using the same samples, we found that grade 5 adhesions had greater AI than grade 3 or 4 adhesions. We may assume that during the certain period of tendon remodeling process, more severe adhesions underwent greater apoptosis. More apoptosis would ultimately favor shrinkage of the adhesion tissues, supporting recovery of tendon gliding. It is also possible that the tendon with more severe adhesions is usually subjected to more vigorous external forces; hence, the greater shear load on the tendons may accelerate apoptosis in the adhesions. Another possible reason for a greater degree of apoptosis in more severe adhesions may be the relatively insufficient nutrient supply to the dense adhesions, which leads to accelerated apoptosis. The present study is limited in that we observed only apoptosis in adhesions, and cell proliferation was not investigated. The observed time points were weeks 4, 6, and 8, covering the initial part of tendon remodeling, which can last as long as 1 year. We believe that it would be worthwhile to extend the observation to other indicators of cellular activity and to later time points. In addition, the cells participating in formation of the adhesions and collagen production and breakdown in adhesions are not clear, which should be further investi-
1077
gated as another important aspect in this complex biological process. Another limitation is the possible false-positive results of the TUNEL assay. Nevertheless, this assay is a standard approach to quantify cellular apoptosis.23–25 Overall, our findings suggest that strong apoptotic reactions in the adhesions and at the tendon-adhesion interface favor the recovery of gliding function of healing tendon. REFERENCES 1. Manske PR. Flexor tendon healing. J Hand Surg Br. 1988;13(3): 237–245. 2. Tang JB. Clinical outcomes associated with flexor tendon repair. Hand Clin. 2005;21(2):199 –210. 3. Dy CJ, Hernandez-Soria A, Ma Y, Roberts TR, Daluiski A. Complications after flexor tendon repair: a systematic review and metaanalysis. J Hand Surg Am. 2012;37(3):543–551. 4. Wu YF, Chen CH, Cao Y, Avanessian B, Wang XT, Tang JB. Molecular events of cellular apoptosis and proliferation in the early tendon healing period. J Hand Surg Am. 2010;35(1):2–10. 5. Wu YF, Zhou YL, Mao WF, Avanessian B, Liu PY, Tang JB. Cellular apoptosis and proliferation in the middle and late intrasynovial tendon healing periods. J Hand Surg Am. 2012;37(2):209 –216. 6. Lui PP, Cheuk YC, Hung LK, Fu SC, Chan KM. Increased apoptosis at the late stage of tendon healing. Wound Repair Regen. 2007;15(5): 702–707. 7. Wong JK, Lui YH, Kapacee Z, Kadler KE, Ferguson MW, McGrouther DA. The cellular biology of flexor tendon adhesion formation: an old problem in a new paradigm. Am J Pathol. 2009;175(5):1938 –1951. 8. Kobayashi M, Toguchida J, Oka M. Development of polyvinyl alcohol-hydrogel (PVA-H) shields with a high water content for tendon injury repair. J Hand Surg Br. 2001;26(5):436 – 440. 9. Cao Y, Tang JB. Strength of tendon repair decreases in the presence of an intact A2 pulley: biomechanical study in a chicken model. J Hand Surg Am. 2009;34(10):1763–1770. 10. Fu SC, Hung LK, Lee YW, Mok TY, Chan KM. Tendon adhesion measured by a video-assisted gliding test in a chicken model. J Hand Surg Eur Vol. 2011;36(1):40 – 47. 11. Wu YF, Zhou YL, Tang JB. Relative contribution of tissue oedema and the presence of an A2 pulley to resistance to flexor tendon movement: an in vitro and in vivo study. J Hand Surg Eur. 2012; 37(4):310 –315. 12. Xu Y, Tang JB. Effects of superficialis tendon repairs on lacerated profundus tendons within or proximal to the A2 pulley: an in vivo study in chickens. J Hand Surg Am. 2003;28(6):994 –1001. 13. Tang JB, Xie RG, Cao Y, Ke ZS, Xu Y. A2 pulley incision or one slip of the superficialis improves flexor tendon repairs. Clin Orthop Relat Res. 2007;456:121–127. 14. Tang JB, Cao Y, Zhu B, Xin KQ, Wang XT, Liu PY. Adenoassociated virus-2-mediated bFGF gene transfer to digital flexor tendons significantly increases healing strength. an in vivo study. J Bone Joint Surg Am. 2008;90(5):1078 –1089. 15. Losa M, Barzaghi RL, Mortini P, et al. Determination of the proliferation and apoptotic index in adrenocorticotropin-secreting pituitary tumors: comparison between micro- and macroadenomas. Am J Pathol. 2000;156(1):245–251. 16. Tanaka F, Kawano Y, Li M, et al. Prognostic significance of apoptotic index in completely resected non–small-cell lung cancer. J Clin Oncol. 1999;17(9):2728 –2736. 17. Rai NK, Suryabhan, Ansari M, Kumar M, Shukla VK, Tripathi K. Effect of glycaemic control on apoptosis in diabetic wounds. J Wound Care. 2005;14(6):277–281.
JHS 䉬 Vol A, June
1078
APOPTOSIS IN ADHESIONS AND INTERFACE
18. Fiala JC. Reconstruct: a free editor for serial section microscopy. J Microsc. 2005;218:52– 61. 19. Beredjiklian PK. Biologic aspects of flexor tendon laceration and repair. J Bone Joint Surg Am. 2003;85(3):539 –550. 20. Lin TW, Cardenas L, Soslowsky LJ. Biomechanics of tendon injury and repair. J Biomech. 2004;37(6):865– 877. 21. Strickland JW. Development of flexor tendon surgery: twenty-five years of progress. J Hand Surg Am. 2000;25(2):214 –235. 22. James R, Kesturu G, Balian G, Chhabra AB. Tendon: biology, biomechanics, repair, growth factors, and evolving treatment options. J Hand Surg Am. 2008;33(1):102–112.
23. Teodoro AJ, Oliveira FL, Martins NB, Maia GA, Martucci RB, Borojevic R. Effect of lycopene on cell viability and cell cycle progression in human cancer cell lines. Cancer Cell Int. 2012;12(1): 36. 24. Zheng T, Kang MJ, Crothers K, et al. Role of cathepsin S-dependent epithelial cell apoptosis in IFN-gamma-induced alveolar remodeling and pulmonary emphysema. J Immunol. 2005; 174(12):8106 – 8115. 25. Garrity MM, Burgart LJ, Riehle DL, Hill EM, Sebo TJ, Witzig T. Identifying and quantifying apoptosis: navigating technical pitfalls. Mod Pathol. 2003;16(4):389 –394.
JHS 䉬 Vol A, June
Apoptosis in Adhesions and the Adhesion-Tendon Gliding Interface: Relationship to Adhesion-Tendon Gliding Mechanics Ya Fang Wu, MD, Jin Bo Tang, MD Purpose Adhesion formation is closely related to tendon-gliding function. We aimed to investigate apoptosis (programmed cell death) in adhesions and tendons and study its relationship to the mechanics of adhesions and healing tendons. Methods The flexor digitorum profundus tendons of 30 long toes in 15 chickens were completely transected and repaired surgically. At postoperative weeks 4, 6, and 8, tendon-gliding excursions were tested and adhesion scores were recorded. Tendons and surrounding adhesions were then harvested for analysis of apoptosis using in situ terminal deoxynucleotidyl transferase dUTP (deoxyuridine triphosphate) nick end labeling assay. Three-dimensional image reconstruction was used to provide an overall view of cellular distribution in tendons and adhesions. Finally, we analyzed the correlation between the apoptotic index measured at the adhesions and the gliding excursions. Ten uninjured tendons served as normal controls. Results Apoptosis was found to be a dominant cellular event in the adhesion tissues at both the adhesion-tendon gliding interface and the adhesion core. The apoptotic index in the adhesions was generally above 20% to 50%. The apoptotic index was significantly higher in the adhesions than in the junction region of the cut tendon ends at weeks 4, 6, and 8. A higher apoptotic index in the adhesions significantly correlated to lower tendon excursions at week 6. Conclusions Apoptosis in adhesions and at the adhesion-tendon interface is a prominent event in the tendon-healing process. The tendons exhibiting a lower tendon-gliding amplitude, meaning more severe adhesions, tended to have a greater apoptotic index in their adhesions during a certain period of the tendon-remodeling process. Clinical relevance Apoptosis in the adhesions and at the adhesion-tendon interface may contribute remarkably to the fate of adhesions and the restoration of the tendon gliding surface, which may be closely related to the tendon function. (J Hand Surg 2013;38A:1071–1078. Copyright © 2013 by the American Society for Surgery of the Hand. All rights reserved.) Key words Tendon healing, apoptosis, adhesions, gliding interface, gliding excursions. DHESION FORMATION AFTER intrasynovial tendon surgery is a major clinical concern because it increases scarring of the tendon and limits active finger motion, leading to poor clinical outcomes.1–3
A
From the Hand Surgery Research Center, Department of Hand Surgery, Affiliated Hospital of Nantong University, Nantong, Jiangsu, China. Received for publication September 11, 2012; accepted in revised form March 4, 2013. Supported by grants from the Health Bureau of Jiangsu Province and the Department of Hand Surgery, Nantong University, Natural Science Foundation of China (No. 81030035 and 81271985).
After the early stages of tendon healing, the gliding movement of the tendon gradually improves and the smooth surface of the tendon is restored. Apoptosis (programmed cell death) was found to be associated Corresponding author: Jin Bo Tang, MD, Department of Hand Surgery, Affiliated Hospital of Nantong University, 20 West Temple Road, Nantong 226001, Jiangsu, China; e-mail:[email protected]. 0363-5023/13/38A06-0002$36.00/0 http://dx.doi.org/10.1016/j.jhsa.2013.03.012
No benefits in any form have been received or will be received related directly or indirectly to the subject of this article.
© ASSH 䉬 Published by Elsevier, Inc. All rights reserved. 䉬 1071
1072
APOPTOSIS IN ADHESIONS AND INTERFACE
with tendon healing, especially immediately after trauma and during the remodeling phase of tendon healing.4,5 However, we do not yet know how apoptosis occurs within adhesion tissues. After intrasynovial chicken tendons were completely cut and repaired, a bimodal distribution of cellular apoptosis was observed at postoperative days 3 to 7 and at days 28 to 56.4,5 High apoptotic reactions were also found during the late remodeling process of healing rat patellar tendon.6 A previous study investigated apoptotic changes in both healing tendon and subcutaneous tissues, using a 50% partial tendon laceration model in mice.7 In that study, apoptosis in the immediate vicinity of the lacerated tendon and subcutaneous tissue wound was found to increase gradually over 84 days after surgery. Though the subcutaneous tissues and tendon followed similar changes over time, more apoptotic cells were seen in the subcutaneous tissues than in the tendon. Little is known about the apoptotic changes of cells either in adhesions or at the adhesion-tendon gliding interface. Apoptosis might be an important factor affecting the mechanical properties of both adhesions and tendons and the remodeling of the tendon surface. We proposed to investigate apoptosis in different areas of adhesions during the remodeling of the healing tendon. We also sought to characterize the relationship between apoptotic changes in cells in adhesions, the severity of adhesion, and the gliding mechanics of the tendon. We hypothesized that more apoptosis would occur in adhesions than in the tendon substance during the recovery of the tendon gliding surface. MATERIALS AND METHODS Forty long toes (30 as surgical models and 10 as normal controls) of 20 white Leghorn chickens weighing 1.5 to 2.5 kg each were used. Institutional animal research regulations were followed during the study. The chickens were anesthetized with intramuscular injection of ketamine (50 mg/kg body weight). Operative procedures The surgery on the 30 long toes was performed using a thigh tourniquet and aseptic technique. A zigzag incision was made in the plantar skin between the proximal interphalangeal (PIP) joint and the distal interphalangeal joint (DIPJ) of the toe.8 –11 Through a 1-cm longitudinal incision in the sheath, the flexor digitorum profundus (FDP) tendon was transected completely and was surgically repaired with 5-0 sutures (Ethilon, Ethicon, Somerville, NJ) using a modified Kessler stitch. Running 6-0 epitendinous sutures were added. The sheath was not closed, and the skin was closed with
interrupted suture. The toe was immobilized in semiflexion using adhesive tape and a light casting until the end of postoperative week 3. The chickens were then allowed to walk freely. After 4, 6, and 8 weeks, the chickens were killed—5 chickens (10 tendons) at each time point. Five chickens (10 uninjured FDP tendons) served as day 0 normal controls. Test of tendon gliding and grading of the adhesions The toes were harvested from the knee joint level and secured to a tensile testing machine (Model 4411; Instron Corporation, Canton, MA). The metatarsophalangeal joint was fixed, and the other joints of the toes were left free. The FDP tendon was exposed from the knee joint to the ankle joint, and its proximal end was connected to a force transducer. After a preload of 0.1N had been applied to the FDP tendon, the digit was flexed until the force reached 15N. The gliding excursions of the FDP tendon at forces of 5N, 10N, and 15N were recorded. Then the tendon was exposed through a volar longitudinal incision in the toe. We carefully separated the normal subcutaneous tissues from adhesions and left the adhesions around the tendon. The length and density of the adhesions were evaluated, and the adhesions were scored according to established grading criteria.12–14 In Situ assay for apoptosis After biomechanical evaluation, as described, the same 10 tendons harvested from each time point were assayed for apoptosis. A 2-cm segment of tendon with adhesions from each toe was harvested with the laceration site at its center, and the samples were kept in 4% paraformaldehyde at 4°C for 48 hours. After gradient alcohol dehydration, the tendons were embedded in paraffin and cut longitudinally in 4m slices from volar to dorsal. We used the in situ terminal deoxynucleotidyl transferase dUTP (deoxyuridine triphosphate) nick end labeling (TUNEL) assay (11684817910; Roche, Mannheim, Germany) to measure apoptosis. After deparaffinization and rehydration, the tissue sections were pretreated with 10 g/mL proteinase K solution for 15 minutes at room temperature. Thereafter, slides were rinsed in PBS (phosphatebuffered saline) and incubated with TUNEL reaction mixture for 1 hour at 37°C in a humidified chamber. After washing, converter-peroxidase solution was applied, and the slides were incubated for 30 minutes at 37°C. The slides were washed and incubated for 10 minutes at room temperature after addition of 3,3=-diaminobenzidine (D-5905;
JHS 䉬 Vol A, June
APOPTOSIS IN ADHESIONS AND INTERFACE
1073
Sigma Aldrich, St. Louis, MO). As negative controls, sections were incubated in label solution without terminal dUTP. The sections were counterstained with hematoxylin and mounted under glass coverslips. Nuclei positive for apotosis stained brown, and those negative for apoptosis stained blue. Quantitation of apoptotic cells in adhesions and at the adhesion-tendon gliding interface Under 400⫻ magnification, we observed the adhesion core and the adhesion at the tendon gliding interface separately at the same horizontal level. In these 2 parts of the adhesions, we separately observed (1) the adhesions over the junction regions of the tendon ends about 2 mm each proximal and distal to the cut site and (2) the adhesions over the extended region of the tendon, approximately 2 mm to 5 mm in each direction from the tendon cut site.4 In each region, 3 fields of view were selected for calculating cells per tendon. An observer blinded to the observational time points counted the total cell numbers and the number of positively stained cells. An apoptotic index (AI) was calculated as the number of TUNEL-positive cells divided by the total number of cells.15–17 In each tendon sample, we also recorded the number of apoptotic cells in the surface and core of the tendon corresponding horizontally to the adhesion tissues to facilitate comparison to their AI to that of adhesions. Distribution of apoptotic cells in 3-dimensional views We applied the software (Reconstruct, Version 1.1.0.0) for 3-dimensional reconstruction to align sections stained with in situ TUNEL assay of these samples to observe the distribution of apoptotic cells in tendon and adhesions.7,18 Serial sections were calibrated and imported into the software program. Wire maps were produced by tracing individual structures of tendon and adhesions. Individual cells were mapped according to their stereological position. Boissonnat surface shading and color selection and transparency features were introduced after the 3-dimensional image reconstruction. Tendon area was highlighted in purple, adhesion in deep blue, apoptotic cells in cyan, and cells without morphological apoptotic changes in black. Statistical analysis For each time point, results of gliding excursions and AI were reported as an average and SD. One-way analysis of variance followed by post hoc Tukey test was used to compare the gliding excursions and AI between 3 different time points. The paired t-test was
FIGURE 1: The gliding excursion of tendon at the 5N, 10N, and 15N load, respectively.
used to compare the AI at 4 areas of the tendonadhesion complex in individual samples at the same time points. The correlation between the AI in the adhesions and the gliding excursion at the load of 15N was calculated using the Pearson test for each time point. We also analyzed whether the AI in the adhesions over the tendon junction region differed significantly according to scores of the adhesions using 1-way analysis of variance followed by Tukey test. The significance level of all comparisons was set at P ⬍ .05. RESULTS Gliding excursion and adhesion score The gliding excursions of the FDP tendon of the normal toes were 16.3 ⫾ 1.2 mm, 19.4 ⫾ 1.3 mm, and 21.4 ⫾ 0.9 mm at the gliding force of 5N, 10N, 15N, respectively, which were significantly greater than those at postoperative weeks 4, 6, and 8 with respective loads (P ⬍ .01). From weeks 4 to 8, gliding excursion gradually increased. However, significant differences were only found between weeks 4 and 8 (P ⬍ .05, all comparisons) (Fig. 1). The adhesion scores were 4.0 ⫾ 0.8, 3.7 ⫾ 0.5, and 3.6 ⫾ 0.8 for the tendons at weeks 4, 6, and 8, respectively. Apoptosis The AI at all the postoperative time points was significantly greater than that of the noninjury control group, among which the AI was 3% ⫾ 2%. One sample at week 4 was used to illustrate the apoptosis seen in adhesions and tendons (Fig. 2). Apoptosis at the adhesion core: At all time points, the greatest AI was at the adhesion core in the junction region.
JHS 䉬 Vol A, June
1074
APOPTOSIS IN ADHESIONS AND INTERFACE
FIGURE 2: Representative histological staining pictures from a week 4 sample. Left: histological sections illustrate the location of the adhesion core, adhesion-tendon interface, tendon surface, and tendon core in junction and extended regions. The regions of adhesions peripheral to the core and extended regions of the tendons are correspondingly named the core or the extended regions of adhesions, respectively. The apoptotic reactions were generally higher in the junction region than in the extended region for both adhesions and tendons and from adhesions to tendons (⫻ 400).
FIGURE 3: Apoptotic index (AI) of the adhesion core, tendon-adhesion surface, tendon surface, and tendon core. *Data significantly greater than those at week 6. †Data significantly greater than those at weeks 4 and 6.
The mean values of the AI in the adhesion core and adhesion-tendon interface were above 0.35 at all 3 time points. The AI in the adhesion core was significantly greater than that recorded at the tendon core (P ⬍ .05). At week 8, the mean AI was as high as 0.68 in the junction region and 0.58 in the extended region of the adhesion core. The AI in adhesions was significantly greater at week 8 than those at week 4 or 6 (P ⬍ .01). Apoptosis at the tendon-adhesion interface: Compared with the tendon core, the AI at the adhesion-tendon interface was significantly higher at all the time points in the junction and extended regions (Figs. 2, 3). At both
junction and extended regions, the AI at the tendonadhesion interface was significantly greater at week 8 than at either week 4 or week 6 (P ⬍ .05). Apoptosis in the tendon: Two regions, the tendon surface and core, were examined. In both regions, the AI was lower than that in the adhesions (Fig. 3). The AI in the junction region of the tendon was generally greater than in the extended region of the tendons (Fig. 3). Analysis of 3-dimesional reconstructed images We used 3-demesional reconstructed images to obtain global views of differences in distribution of apoptotic
JHS 䉬 Vol A, June
APOPTOSIS IN ADHESIONS AND INTERFACE
1075
FIGURE 4: A–C Histological sections show the tendon and surrounding adhesions from volar to dorsal aspect of the adhesiontendon samples at postoperative weeks 4, 6, and 8. D–F Corresponding 3-dimensional images of A–C built based on multiple histology slides depicting distribution of the apoptotic cells (cyan) in tendon (purple), and adhesions (deep blue) at postoperative weeks 4, 6, and 8. The cells without signs of apoptosis are shown in black (white arrow). Note the generally denser apoptotic cells at the tendon-adhesion interface and in the adhesion core. The blue arrowheads in A and D indicate the level of tendon transection.
cells in the tendon and adhesions and confirmed the substantially greater density of positive cells in the adhesion core and adhesion-tendon gliding interface compared with the tendon (Fig. 4). Correlation analysis of AI with tendon gliding excursions and adhesion scores The correlation plots of the tendon gliding excursions at the pulling force of 15N and AI of the adhesion-tendon interface and adhesion core are shown in Table 1. Overall, the AI correlated with the gliding excursion negatively. There was a significant negative correlation
between AI of the adhesions and the gliding excursion at week 6 (Table 1, Fig. 5). No statistically significant correlation was found for such comparisons at weeks 4 or 8. The samples with greater severity scores of peritendinous adhesions had significantly greater AI (P ⬍ .01) (Fig. 6). DISCUSSION The major focus of our present study was the investigation of apoptosis within adhesions around the tendon. At each time point, we documented more dramatic
JHS 䉬 Vol A, June
1076
APOPTOSIS IN ADHESIONS AND INTERFACE
TABLE 1. Correlation of Tendon Gliding Excursions With Aptotic Index in the Adhesion Interface or the Core at 3 Postoperative Time Points Adhesion-Tendon Interface Time Points (wk)
Sample Size
Junction Region
Adhesion Core
Extended Region
Junction Region
Extended Region
4
10
r ⫽ ⫺.117, P ⫽ .747
r ⫽ ⫺.102, P ⫽ .779
r ⫽ ⫺.077, P ⫽ .833
r ⫽ ⫺.053, P ⫽ .884
6
10
r ⫽ ⫺.817, P ⫽ .004*
r ⫽ ⫺.745, P ⫽ .013*
r ⫽ ⫺.781, P ⫽ .008*
r ⫽ ⫺.734, P ⫽ .016*
8
10
r ⫽ ⫺.353, P ⫽ .257
r ⫽ ⫺.598, P ⫽ .068
r ⫽ ⫺.219, P ⫽ .543
r ⫽ ⫺.566, P ⫽ .088
r, correlation coefficient. *Correlation with statistical significance.
FIGURE 5: Scatter plots of the apoptotic index (AI) in tendon-adhesion interface versus the gliding excursion under the load of 15N at week 6.
FIGURE 6: Comparison of the apoptotic index (AI) in the adhesions of different scores. *Data significantly greater than those of grade 3 or 4 at the adhesion core.
apoptotic changes in the adhesions than in the healing tendon. Regarding the time course of apoptotic changes, the AI was much higher in postoperative week 8 than in week 4 or 6. However, the tendon gliding
excursion increased from week 4 to week 8. The increased apoptotic reaction may relate to the clearance of excess cells participating in tissue repair during the earlier healing period, contributing to the recovery of tendon gliding.19,20 With regards to changes documented in different regions of adhesions, the apoptosis was more significant in the junction region than in the extended region. We further sought to characterize the apoptosis in the tendon and surrounding adhesions at the same horizontal level. The severity of apoptosis gradually increases from the tendon core, to the tendon surface, the adhesion-tendon interface, and the adhesion core. A previous study documented that healing tendons exhibit generally greater apoptosis in the tendon surface than in the core.4 The present result is in agreement with that previous report. More importantly, we found that AI was notably higher in the adhesions than in the tendon. No significant difference was found in AI between the adhesion core and the adhesion-tendon interface. This finding indicates that the tissues at the adhesion-tendon gliding interface did not undergo
JHS 䉬 Vol A, June
APOPTOSIS IN ADHESIONS AND INTERFACE
significantly more apoptosis than that seen at the center of the adhesions. The remodeling phase of tendon healing is responsible for cell number decrease, collagen realignment, and the restoration of a smooth surface to facilitate tendon gliding.1,21,22 During tendon mobilization, the healing tendon and adhesions bear the internal and external tensions caused by weight-bearing, muscle stretching during movements, and shear against surrounding tissues or the other pressures acting on tendon and adhesions. The adhesions in particular are subjected to greater shear force than the tendon. The remarkable increases in apoptosis in the adhesion and the tendonadhesion interface may be associated with the effect of shear deformation induced by mobilization. We may assume that more apoptosis in the tendon-adhesion interface would accelerate the recovery of the gliding tendon’s smooth surface and that less apoptosis would delay recovery. In addition, dissection revealed that the adhesions were relatively compact at weeks 4 and 6 and that the density decreased at week 8, which may also relate to the greater apoptosis in the adhesions during this period. We attempted to correlate the biomechanical findings with molecular apoptotic changes in each sample. Our correlation analysis showed negative correlations between gliding excursion and AI in the adhesions of the healing tendon with significant negative correlation at week 6. Using the same samples, we found that grade 5 adhesions had greater AI than grade 3 or 4 adhesions. We may assume that during the certain period of tendon remodeling process, more severe adhesions underwent greater apoptosis. More apoptosis would ultimately favor shrinkage of the adhesion tissues, supporting recovery of tendon gliding. It is also possible that the tendon with more severe adhesions is usually subjected to more vigorous external forces; hence, the greater shear load on the tendons may accelerate apoptosis in the adhesions. Another possible reason for a greater degree of apoptosis in more severe adhesions may be the relatively insufficient nutrient supply to the dense adhesions, which leads to accelerated apoptosis. The present study is limited in that we observed only apoptosis in adhesions, and cell proliferation was not investigated. The observed time points were weeks 4, 6, and 8, covering the initial part of tendon remodeling, which can last as long as 1 year. We believe that it would be worthwhile to extend the observation to other indicators of cellular activity and to later time points. In addition, the cells participating in formation of the adhesions and collagen production and breakdown in adhesions are not clear, which should be further investi-
1077
gated as another important aspect in this complex biological process. Another limitation is the possible false-positive results of the TUNEL assay. Nevertheless, this assay is a standard approach to quantify cellular apoptosis.23–25 Overall, our findings suggest that strong apoptotic reactions in the adhesions and at the tendon-adhesion interface favor the recovery of gliding function of healing tendon. REFERENCES 1. Manske PR. Flexor tendon healing. J Hand Surg Br. 1988;13(3): 237–245. 2. Tang JB. Clinical outcomes associated with flexor tendon repair. Hand Clin. 2005;21(2):199 –210. 3. Dy CJ, Hernandez-Soria A, Ma Y, Roberts TR, Daluiski A. Complications after flexor tendon repair: a systematic review and metaanalysis. J Hand Surg Am. 2012;37(3):543–551. 4. Wu YF, Chen CH, Cao Y, Avanessian B, Wang XT, Tang JB. Molecular events of cellular apoptosis and proliferation in the early tendon healing period. J Hand Surg Am. 2010;35(1):2–10. 5. Wu YF, Zhou YL, Mao WF, Avanessian B, Liu PY, Tang JB. Cellular apoptosis and proliferation in the middle and late intrasynovial tendon healing periods. J Hand Surg Am. 2012;37(2):209 –216. 6. Lui PP, Cheuk YC, Hung LK, Fu SC, Chan KM. Increased apoptosis at the late stage of tendon healing. Wound Repair Regen. 2007;15(5): 702–707. 7. Wong JK, Lui YH, Kapacee Z, Kadler KE, Ferguson MW, McGrouther DA. The cellular biology of flexor tendon adhesion formation: an old problem in a new paradigm. Am J Pathol. 2009;175(5):1938 –1951. 8. Kobayashi M, Toguchida J, Oka M. Development of polyvinyl alcohol-hydrogel (PVA-H) shields with a high water content for tendon injury repair. J Hand Surg Br. 2001;26(5):436 – 440. 9. Cao Y, Tang JB. Strength of tendon repair decreases in the presence of an intact A2 pulley: biomechanical study in a chicken model. J Hand Surg Am. 2009;34(10):1763–1770. 10. Fu SC, Hung LK, Lee YW, Mok TY, Chan KM. Tendon adhesion measured by a video-assisted gliding test in a chicken model. J Hand Surg Eur Vol. 2011;36(1):40 – 47. 11. Wu YF, Zhou YL, Tang JB. Relative contribution of tissue oedema and the presence of an A2 pulley to resistance to flexor tendon movement: an in vitro and in vivo study. J Hand Surg Eur. 2012; 37(4):310 –315. 12. Xu Y, Tang JB. Effects of superficialis tendon repairs on lacerated profundus tendons within or proximal to the A2 pulley: an in vivo study in chickens. J Hand Surg Am. 2003;28(6):994 –1001. 13. Tang JB, Xie RG, Cao Y, Ke ZS, Xu Y. A2 pulley incision or one slip of the superficialis improves flexor tendon repairs. Clin Orthop Relat Res. 2007;456:121–127. 14. Tang JB, Cao Y, Zhu B, Xin KQ, Wang XT, Liu PY. Adenoassociated virus-2-mediated bFGF gene transfer to digital flexor tendons significantly increases healing strength. an in vivo study. J Bone Joint Surg Am. 2008;90(5):1078 –1089. 15. Losa M, Barzaghi RL, Mortini P, et al. Determination of the proliferation and apoptotic index in adrenocorticotropin-secreting pituitary tumors: comparison between micro- and macroadenomas. Am J Pathol. 2000;156(1):245–251. 16. Tanaka F, Kawano Y, Li M, et al. Prognostic significance of apoptotic index in completely resected non–small-cell lung cancer. J Clin Oncol. 1999;17(9):2728 –2736. 17. Rai NK, Suryabhan, Ansari M, Kumar M, Shukla VK, Tripathi K. Effect of glycaemic control on apoptosis in diabetic wounds. J Wound Care. 2005;14(6):277–281.
JHS 䉬 Vol A, June
1078
APOPTOSIS IN ADHESIONS AND INTERFACE
18. Fiala JC. Reconstruct: a free editor for serial section microscopy. J Microsc. 2005;218:52– 61. 19. Beredjiklian PK. Biologic aspects of flexor tendon laceration and repair. J Bone Joint Surg Am. 2003;85(3):539 –550. 20. Lin TW, Cardenas L, Soslowsky LJ. Biomechanics of tendon injury and repair. J Biomech. 2004;37(6):865– 877. 21. Strickland JW. Development of flexor tendon surgery: twenty-five years of progress. J Hand Surg Am. 2000;25(2):214 –235. 22. James R, Kesturu G, Balian G, Chhabra AB. Tendon: biology, biomechanics, repair, growth factors, and evolving treatment options. J Hand Surg Am. 2008;33(1):102–112.
23. Teodoro AJ, Oliveira FL, Martins NB, Maia GA, Martucci RB, Borojevic R. Effect of lycopene on cell viability and cell cycle progression in human cancer cell lines. Cancer Cell Int. 2012;12(1): 36. 24. Zheng T, Kang MJ, Crothers K, et al. Role of cathepsin S-dependent epithelial cell apoptosis in IFN-gamma-induced alveolar remodeling and pulmonary emphysema. J Immunol. 2005; 174(12):8106 – 8115. 25. Garrity MM, Burgart LJ, Riehle DL, Hill EM, Sebo TJ, Witzig T. Identifying and quantifying apoptosis: navigating technical pitfalls. Mod Pathol. 2003;16(4):389 –394.
JHS 䉬 Vol A, June