Mechanical Chondroplasty: Early Metabolic Consequences In Vitro

Mechanical Chondroplasty: Early Metabolic Consequences In Vitro Lee D. Kaplan, M.D., Bryan Royce, B.S., Brian Meier, B.S., Justin M. Hoffmann, B.S., Jonathan D. Barlow, B.S., Yan Lu, M.D., and Herman F. Stampfli, B.S., M.B.A.

Purpose: The purpose of this study was to determine the depth of penetration from mechanical chondroplasty and metabolic consequences of this procedure on the remaining articular cartilage. Methods: Mechanical chondroplasty was performed in vitro on a portion of fresh grade I or II articular cartilage from 8 human knee arthroplasty specimens. Treated and control (untreated) explants (approximately 30 mg) were cut from the cartilage. The explants were divided into 2 groups, day 1 and day 4, placed separately in a 48-well plate containing media, and incubated at 37°C for 24 hours. After the 24-hour incubation, the explants were weighed on day 1 and day 4, and explant media were removed and tested for total proteoglycan synthesis and aggrecan synthesis. At time 0, 2 sets (2.6 mm each) of treated and control cartilage slices were cut with a precision saw. One set was stained for confocal laser microscopy via a cytotoxicity stain to determine cell viability. The second set was stained with H&E to determine depth of penetration. Results: The mean depth of penetration was 252.8 ⫾ 78 ␮m. There was no significant difference (P ⬎ .25) between total proteoglycan synthesis for control versus treatment groups on day 1 or 4. Aggrecan synthesis was significantly reduced on day 1 when normalized for tissue weight (P ⫽ .019) and double-stranded deoxyribonucleic acid (P ⫽ .004). On day 4, no significant difference was detected. Confocal laser microscopy did not show cell death below the zone of treatment. Conclusions: There was no significant metabolic consequence caused by chondroplasty to the remaining articular cartilage, and the zone of injury was limited to the treatment area. Clinical Relevance: Mechanical chondroplasty causes no significant metabolic consequences to articular cartilage under these conditions. Key Words: Articular cartilage—Mechanical chondroplasty—Debridement—Knee—Osteoarthritis— Cartilage defects.

C

From the Department of Orthopedics and Rehabilitation, University of Wisconsin Hospital and Clinic (L.D.K., B.R., B.M., J.M.H., H.F.S.), and Schools of Medicine and Public Health (L.D.K., B.R., B.M., J.M.H., J.D.B., H.F.S.) and Veterinary Medicine (Y.L.), University of Wisconsin-Madison, Madison, Wisconsin, U.S.A. Supported by the National Football League Charities and the Department of Orthopedic Surgery and Rehabilitation, School of Medicine and Public Health, University of Wisconsin-Madison. The authors report no conflict of interest. Address correspondence and reprint requests to Lee D. Kaplan, M.D., Department of Orthopedics and Rehabilitation, K4/727 CLC, 600 Highland Ave, Madison, WI 53792, U.S.A. E-mail: [email protected] © 2007 by the Arthroscopy Association of North America 0749-8063/07/2309-6606$32.00/0 doi:10.1016/j.arthro.2007.04.005

hondroplasty is routinely used in arthroscopic surgery to debride and smooth articular cartilage lesions. Thermal (bipolar and monopolar) radiofrequency (RF) probes and mechanical shaver devices are commonly used to perform chondroplasty. A recent innovation, the fluid pressure– driven saline solution jet instrument, also debrides cartilage and provides clean edges.1,2 This new debridement tool was recently evaluated against a mechanical shaver and bipolar RF probe and was found to be superior to both in clinically relevant areas in that study1: (1) depth of residual damage after chondroplasty and (2) degree of surface smoothness. Precise instrument control of this is required to guide the high-velocity instrument in resurfacing cartilage, and further studies are warranted.1 RF probes use high-frequency alternating current that flows from the probe to the tissue and applies a force

Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 23, No 9 (September), 2007: pp 923-929

923

924

L. D. KAPLAN ET AL.

to the ions within the tissue as an unmodulated sine wave causing selective heating. The selective heating creates a smooth cartilage surface by “melting” or caramelizing tissue, which reduces catabolic enzymes3 and seals damaged articular cartilage.4 This technique has been controversial as to whether there are long-term detrimental effects.5-10 The metabolic activity of nonarthritic cartilage may be more sensitive to thermal energy than arthritic tissue.11,12 Cartilage debridement, via mechanical shavers, has been widely used without significant concern over some of the issues that have been scrutinized in RF chondroplasty. Currently, mechanical chondroplasty is widely used without clear delineation of its consequences. Partial- and full-thickness articular lesions are commonly encountered in arthroscopy. Curl et al.13 evaluated 31,516 arthroscopies and found 53,569 hyaline cartilage lesions in 19,827 patients over a 4-year duration. The goals of chondroplasty are to halt the progression of the articular lesion, smooth the remaining surface, and prevent debris from becoming an irritant within the joint. The purpose of this study was to determine the depth of penetration and early metabolic consequences of mechanical chondroplasty in vitro. Evaluation was done by directly measuring the depth of cartilage defects created by a mechanical shaver and quantifying proteoglycan (PG) synthesis of the remaining articular cartilage. We hypothesized that mechanical chondroplasty would cause metabolic damage even at routine debridement depths. METHODS Articular cartilage tissue extracted from 8 knees subjected to total knee arthroplasty was used in this study. Appropriate approval from our institutional review board was obtained. Knees from patients with a history of inflammatory arthritis, chondrocyte transplant, systemic lupus, human immunodeficiency virus, hepatitis, pigmented villonodular synovitis, or avascular necrosis were excluded. Cartilage was graded by use of the International Cartilage Repair Society grading system before chondroplasty, and only minimally damaged grade I or II cartilage was used.14 Control and shaver-treated cartilage was not removed from the subchondral bone until after chondroplasty. The best available cartilage from each knee condyle (medial or lateral femoral) was used in this study. Either the medial or lateral femoral condyle from each knee, with adequate usable cartilage area for the study, was equally divided into control and treatment

areas. The treatment and control areas for each knee were always from the same femoral condyle. Each condyle was immersed in saline solution to avoid dehydration. The Dyonics power control unit (Smith & Nephew Endoscopy, Andover, MA) was set at 3,800 rpm, and a Series 3000 BoneCutter 4.5-mm full-radius oscillating arthroscopic blade (Smith & Nephew Endoscopy) was used to shave the treatment area. The shaver was operated by 1 investigator for the entire study. The investigator moved the oscillating arthroscopic blade across the cartilage surface in 1 direction and used a single-pass technique. Care was taken to apply the same pressure to each cartilage specimen and limit the amount of time the shaver blade resided at a single location. After debridement, the treatment and control areas were cut into explants weighing approximately 30 mg and randomly divided into 2 groups: day 1 and day 4. Five replicates with and without chondroplasty for each day, or a total of twenty explants, were assayed for each knee. The explants were placed in a 48-well plate with 1 mL of Dulbecco’s modified Eagle’s medium containing 4.5-g/L glucose, sodium pyruvate, and L-glutamine and without phenol red. The Dulbecco’s modified Eagle’s medium was also supplemented with 10% fetal bovine serum, 100-IU/mL penicillin, and 100␮g/mL streptomycin. All samples were incubated at 37°C and 5% carbon dioxide (CO2) (Forma Series II water-jacketed incubator; Thermo Electron, Waltham, MA) until assayed. After sterile explant removal, an Isomet Plus precision saw (Buehler, Lake Bluff, IL) was used to cut 2 sets of 2.6-mm slices from the debrided (treated) and nondebrided (control) areas of cartilage on each knee. The first set of slices was used to determine cell viability by use of a LIVE/DEAD Viability/Cytotoxicity Kit (Molecular Probes, Eugene, OR). An MCR1024 confocal laser microscope (Bio-Rad, Hemel Hempstead, England) equipped with a krypton/argon laser and the necessary filter systems (fluorescein [522DF32] and rhodamine [585EFLP]) was used via the triple-labeling technique as previously described.15,16 The second set of slices was used to assess the depth of chondroplasty penetration via H&E histologic staining. The microscope was calibrated by use of a micrometer measured through the objective lens (⫻4) used for this project. The pixel length measured on images was converted to micrometers. The mean depth of cleft (i.e., mean depth of 3 deepest clefts on cartilage surface) was determined on each histologic image of the osteochondral

CHONDROPLASTY DEPTH AND METABOLIC ACTIVITY sections by use of Adobe Photoshop (Adobe Systems, San Jose, CA). The dimethylmethylene blue (DMMB) assay was previously described.17 Metabolic activity was assessed by use of the DMMB assay for large PG synthesis (aggrecan) and sulfur 35 (35S) assay for total PG synthesis. For the DMMB assay, media from each explant well were collected after the 24-hour incubation at 37°C and 5% CO2 on day 1 and 4. Duplicate 50-␮L aliquots of the media were placed into a 96well plate along with a chondroitin 6-sulfate standard curve. A 200-␮L aliquot of DMMB solution17 was added to each well, and the absorbance was measured with a Multiskan Accent spectrophotometer (Thermo Electron) at 520 nm. The sample concentration was measured against the chondroitin standard curve and reported in micrograms. Cartilage explants were then weighed and placed in fresh media. The day 1 explants received 10 ␮Ci of 35 S and were incubated at 37°C and 5% CO2 for 4 hours. After 4 hours, the media from each explant well were collected and stored at ⫺20°C until assayed. A 0.5-mL aliquot of elution buffer (EB) was added to each explant well, and the plate was placed on a titer plate shaker (Lab-Line Instruments, Melrose Park, IL) and shaken for 2 days at 4°C. The EB consisted of 382 g of guanidine, 6.06 g of Tris buffer, and 7.1 g of sodium sulfate in 1 L of distilled water. The EB from the explants was collected and stored at ⫺20°C until assayed. The collected media (200 ␮L) and 100 ␮L of the explant EB were added to a solid-phase PD-10 extraction column (Sigma-Aldrich, St Louis, MO) after conditioning. The combined sample was eluted by use of five 1-mL aliquots of fresh EB and collected in five separate liquid scintillation vials. Scintillation cocktail was added to each vial. Radioactive sulfur incorporation into PGs produced during the 4-hour incubation was measured with a 1900CA Tri-Carb liquid scintillation counter (Packard Instruments, Downers Grove, IL). On day 3, the media were changed for the day 4 explants, and the 48-well plates were returned to the 37°C incubator for 24 hours, after which the aforementioned DMMB, 35S, and EB steps were completed. All of the cartilage explants were then digested by use of a 0.25-mg/mL solution of papain. Once digested, the samples were analyzed for doublestranded deoxyribonucleic acid (dsDNA) content by use of the PicoGreen assay (Molecular Probes) at a 1:50 dilution. Differences in DMMB/mass for the 8 samples based on group (control v treatment) and day (1 v 4)

925

were tested with a paired t test. A similar analysis was performed for DMMB/PicoGreen, DMMB/mass, 35S/ PicoGreen, and 35S/mass. Differences across time were not tested. All P values reported are 2-sided; P ⬍ .05 was used for statistical significance. The statistical analysis was performed via SAS for Windows, version 6.12 (SAS Institute, Cary, NC). RESULTS The metabolism of the treatment group (shaven cartilage explants) and control group (unshaven cartilage explants) was compared by use of both the DMMB and 35S assay methods. The results of each method were normalized for explant weight (in milligrams) and dsDNA by use of the PicoGreen assay. The 35S assay for day 1 and day 4 (Table 1) showed no significant difference in rates of chondrocyte metabolism between the media from treatment and control explants when normalized for either explant mass (weight) or number of cells for each explant (dsDNA). A statistically significant decrease in metabolism was noted on day 1 between the treatment and control explant media via the DMMB method when normalized for both explant mass/weight (P ⫽ .019) (Fig 1, Table 1) and number of cells for each explant/dsDNA (P ⫽ .004) (Fig 2, Table 1) by use of a PicoGreen assay. This difference, however, was resolved by day 4, when there were no significant differences between the treatment and control groups when measured by either assay method. The mean in vitro mechanical chondroplasty depth of penetration was 252.8 ⫾ 78 ␮m (95% confidence interval, 169.5 to 336.1) for cartilage slices assessed with the H&E stain. Confocal laser microscopy images obtained by use of cell viability stain did not show any visible cell death beyond the mechanical chondroplasty area. Figure 3A shows a representative image of an unshaven (control) cartilage slice, and Figure 3B shows a mechanically shaven (treatment) cartilage slice. DISCUSSION The articular cartilage used for this study was grade I or II based on the International Cartilage Repair Society grading system and was therefore relatively smooth.14 Other studies have chemically induced articular damage or used arthritic tissue.1,7,18 We used a smooth articular surface for 2 reasons: (1) to enable a baseline of metabolic activity that was not altered or variable by use of osteoarthritic tissue and (2) to

926

L. D. KAPLAN ET AL. TABLE 1.

PG (dsDNA or Mass) via DMMB or

Treated Group

35

S Assay

Control Group

Mean

SD

SE

Mean

SD

SE

P Value

8.775 7.317

3.482 4.28

1.231 1.513

7.446 6.562

3.723 1.969

1.316 0.696

.141 .611

0.083 0.106

0.033 0.102

0.012 0.036

0.077 0.079

0.032 0.035

0.011 0.012

.492 .413

0.073 0.041

0.018 0.028

0.006 0.01

0.089 0.046

0.018 0.024

0.006 0.008

.019* .16

0.0007 0.0005

0.0002 0.0003

0.00007 0.0001

0.0009 0.0005

0.0002 0.0003

0.00007 0.00009

.004* .282

35

S/mass Day 1 Day 4 35 S/PicoGreen Day 1 Day 4 DMMB/mass Day 1 Day 4 DMMB/PicoGreen Day 1 Day 4 *Statistically significant.

establish a precise depth of penetration, without the inherent measurement concerns found with clefts in osteoarthritic tissue. The mechanical shaver used in this study created a mean depth of penetration of 252.8 ␮m, as compared with 504 ␮m in the study by Green et al.1 The increased depth of penetration was attributed to collagen fibers being “pulled” into the shaver. This increased depth may also be a result of the area treated and the model used in that study, where osteoarthritis was chemically induced.1 As a comparison, the depth of penetration for a fluid pressure– driven saline solution

jet instrument in the study of Green et al. was 414 ␮m, and a depth of penetration from RF chondroplasty ranging from 150 ␮m to greater than 1,000 ␮m has been documented.6,7,19,20 RF chondroplasty creates a smooth cartilage surface by “melting” the tissue. There may be inherent problems associated with heating cartilage tissue.6,7,9,21 In the clinical setting a minimal depth of penetration that accomplishes the debridement tasks is desirable because shaving and laser abrasion are detrimental to the biologic microenvironment of chondrocytes.22 Our study found no significant cell death beyond the chondroplasty zone,

FIGURE 1. PG content (in micrograms) in explant media (with and without chondroplasty) for day 1 and 4 via DMMB assay. The PG content in the media was measured for 24 hours and normalized for explant weight (in milligrams). A statistically significant decrease (P ⫽ .019 [asterisk]) in PG content was observed for the chondroplasty group (treatment) compared with the non-chondroplasty group (control) on day 1; however, the significance between the treated and control groups was not observed on day 4.

CHONDROPLASTY DEPTH AND METABOLIC ACTIVITY

927

FIGURE 2. PG content (in micrograms) in explant media (with and without chondroplasty) for day 1 and 4 via DMMB assay. The PG content was measured for 4 hours on day 1 and 4 and normalized to dsDNA for each explant by use of a PicoGreen assay. A statistically significant decrease (P ⫽ .004 [asterisk]) in metabolism was observed for the chondroplasty group (treatment) compared with the non-chondroplasty group (control) on day 1; however, the significance between the treated and control groups was not observed on day 4.

whereas Edwards et al.,23 in their pony model, found 166 ␮m of cell death for a mechanical shaver and 609 to 735 ␮m for 2 different RF probes. Different chondroplasty instrumentation, whether thermal or mechanical, will produce different depths of penetration. Variability with thermal instruments has been attributed to many factors including study design, power setting, speed, number of passes, and force applied.24 Many of the same variability factors are applicable for mechanical chondroplasty. To establish study uniformity and reproducibility, the power setting and speed for the mechanical shaver were not changed during the study. The same size and type of arthroscopic surgery blade were used throughout the study. One investigator performed all chondroplasties using a single-pass technique and minimal shaver residence time on the cartilage surface. The shaver speed remained at 3,800 rpm throughout the study. This technique was preferred because multiple shaver passes, increased residence time, or higher shaver speeds may cause higher chondrocyte damage and death. There was a significant reduction in PG for the treated explants versus control explants on day 1 by use of the DMMB assay. Both methods of data normalization (explant weight per milligram [mass] and number of chondrocytes [dsDNA]) showed the same significant result. This suggests that the explants contained a proportionate amount of chondrocytes per unit cartilage and were representative of the cartilage being tested.

The 35S results showed no significant difference for total PG between the treated and control explants on day 1; however, the treated explants showed a trend of reduced metabolism. This 35S trend, along with the significantly reduced metabolism on day 1 with the DMMB assay, suggests that there is reduced metabolism immediately after chondroplasty. The difference between the DMMB assay conditions (incubation for 24 hours) versus 35S assay conditions (incubation for 4 hours) may be one explanation for the lack of significance with the 35S assay, with less PG being released to the media in 4 hours and increased assay variability as the assay approaches the 35S detection limit. One explanation for the reduction of PG production/ aggrecan accumulation on day 1 may be that chondrocytes were “stressed” after mechanical chondroplasty. A “shock” to cells, even if they remain viable, alters their responses to some biochemical stimuli.25 Quinn et al.25 found that PG synthesis rate, collagen, and other matrix macromolecules were reduced by static compression and the acquisition of newly synthesized PG was delayed. With blunt trauma, there is extensive cell death and matrix disruption adjacent to the lesion edge with no new matrix synthesis and chondrocyte proliferation behind this region.26 Wounding by a sharp instrument produces limited cell death, chondrocyte proliferation, and up-regulation of new matrix synthesis immediately adjacent to the lesion edge, with very little disruption of the collagen architecture.26

928

L. D. KAPLAN ET AL.

FIGURE 3. Representative confocal laser microscope images of cartilage slices stained with LIVE/DEAD cytotoxicity stain to determine cell viability: (A) cartilage slice without chondroplasty (control) and (B) cartilage slice with chondroplasty (treated). Viable cartilage stains green, whereas nonviable cartilage stains red.

Another possible explanation for the reduction in PG on day 1 was that mechanical chondroplasty altered the cartilage dynamics and the chondrocytes required time to recover. The shaver removed approximately 253 ␮m of cartilage, which included the superficial zone chondrocytes (SZC) (⬍100 ␮m) and part of the middle zone chondrocytes (MZC). The remaining MZC and deep zone chondrocytes (DZC) are less dense and active than SZC.25 These chondrocytes (MZC and DZC) were required to increase cellular activity and synthesize PG. Chondrocytes are heterogeneous at different cellular locations (SZC v

MZC v DZC) and have differences in receptor and growth factor requirements.27 Limitations of this study include that it was an in vitro study and did not have synovial factors in the milieu. The study also evaluated metabolism over a limited time frame and was performed on minimally osteoarthritic tissue. One variable within this study was the amount of pressure exerted on the cartilage. The total pressure exerted on the cartilage was the weight of the shaver and the pressure applied by the researcher. To control this variable, the applied pressure was repeatedly measured by use of an analytic scale and the chondroplasty technique was practiced on cartilage before the study. This study used the best available cartilage from the medial or lateral condyle. The shaver-treated and control explants were removed from the same femoral condyle. An ideal study may (1) use a single femoral condyle location for all knee specimens, (2) remove control and treatment explants from the same topographic position on each femoral condyle and, (3) use knee cartilage from a narrow (patient) age range. There are metabolic differences with tissue location, topographic position, and patient age.28,29 Quinn et al.29 identified 4 distinct types of cartilage morphology: (1) femoral condyles, (2) patellar groove, (3) mediocentral tibial plateau, and (4) lateral tibia (meniscus-covered). The joint location that performs similar biomechanical functions has similar cell and matrix morphologies.29 This suggests that the lateral and medial condyle may not have identical tissue but similar tissue. The quality of osteoarthritic cartilage is highly variable because patients consider knee replacement surgery only as a last resort. Improved data reproducibility may be achieved by exclusively using either medial or lateral cartilage; however, the day 1 metabolic activity produced statistically significant results with the combination of medial and lateral femoral condyles. We, however, always compared treated (chondroplasty) and control explants from the same condyle. This study only addressed early metabolic changes and the depth of penetration from mechanical chondroplasty. Given the limited time course assessed, there was no attempt to extrapolate these results beyond day 4. Although another model may better mimic in vivo conditions, it was critical for us to limit the variables to evaluate the 2 primary endpoints of metabolic activity and depth of penetration. Our study suggests that mechanical chondroplasty reduced metabolic activity immediately after chondroplasty; however, the reduced metabolism was not sustained because the amount of PG released to the media

CHONDROPLASTY DEPTH AND METABOLIC ACTIVITY was similar to controls by day 4. Therefore mechanical chondroplasty does not appear to cause sustained metabolic damage at routine debridement depths under these conditions. CONCLUSIONS There was no significant change in metabolic activity caused by chondroplasty to the remaining articular cartilage that was sustained for the duration of the study. Cell viability, below the zone of chondroplasty, remained unaffected by chondroplasty for the duration of the study. Acknowledgment: The authors thank Glen Leverson, Ph.D., for his assistance with statistical analysis.

REFERENCES 1. Green LM, King JS, Bianski BM, Pink MM, Jobe CM. In vitro effects of 3 common arthroscopic instruments on articular cartilage. Arthroscopy 2006;22:300-307. 2. King JS, Green LM, Bianski BM, Pink MM, Jobe CM. Shaver, bipolar radiofrequency, and saline jet instruments for cutting meniscal tissue: A comparative experimental study on sheep menisci. Arthroscopy 2005;21:844-850. 3. Yasura K, Nakagawa Y, Kobayashi M, Kuroki H, Nakamura T. Mechanical and biochemical effect of monopolar radiofrequency energy on human articular cartilage: An in vitro study. Am J Sports Med 2006;34:1322-1327. 4. Uribe JW. Electrothermal chondroplasty—Bipolar. Clin Sports Med 2002;21:675-685. 5. Barber FA, Iwasko NG. Treatment of grade III femoral chondral lesions: Mechanical chondroplasty versus monopolar radiofrequency probe. Arthroscopy 2006;22:1312-1317. 6. Caffey S, McPherson E, Moore B, Hedman T, Vangsness CT Jr. Effects of radiofrequency energy on human articular cartilage: An analysis of 5 systems. Am J Sports Med 2005;33: 1035-1039. 7. Edwards RB III, Lu Y, Nho S, Cole BJ, Markel MD. Thermal chondroplasty of chondromalacic human cartilage. An ex vivo comparison of bipolar and monopolar radiofrequency devices. Am J Sports Med 2002;30:90-97. 8. Encalada I, Richmond JC. Osteonecrosis after arthroscopic meniscectomy using radiofrequency. Arthroscopy 2004;20: 632-636. 9. Kaplan L, Uribe JW. The acute effects of radiofrequency energy in articular cartilage: An in vitro study. Arthroscopy 2000;16:2-5. 10. Lu Y, Edwards RB III, Nho S, Heiner JP, Cole BJ, Markel MD. Thermal chondroplasty with bipolar and monopolar radiofrequency energy: Effect of treatment time on chondrocyte death and surface contouring. Arthroscopy 2002;18:779-788. 11. Kaplan LD, Chu CR, Bradley JP, Fu FH, Studer RK. Recovery of chondrocyte metabolic activity after thermal exposure. Am J Sports Med 2003;31:392-398. 12. Kaplan LD, Ionescu D, Ernsthausen JM, Bradley JP, Fu FH, Farkas DL. Temperature requirements for altering the morphology of osteoarthritic and nonarthritic articular cartilage: In

13. 14. 15.

16.

17. 18.

19.

20.

21.

22. 23.

24. 25.

26. 27.

28.

29.

929

vitro thermal alteration of articular cartilage. Am J Sports Med 2004;32:688-692. Curl WW, Krome J, Gordon ES, Rushing J, Smith BP, Poehling GG. Cartilage injuries: A review of 31,516 knee arthroscopies. Arthroscopy 1997;13:456-460. Smith GD, Taylor J, Almqvist KF, et al. Arthroscopic assessment of cartilage repair: A validation study of 2 scoring systems. Arthroscopy 2005;21:1462-1467. Lu Y, Edwards RB III, Cole BJ, Markel MD. Thermal chondroplasty with radiofrequency. An in vitro comparison of bipolar and monopolar radiofrequency devices. Am J Sports Med 2001;29:42-49. Lu Y, Edwards RB III, Kalscheur VL, Nho S, Cole BJ, Markel MD. Effect of bipolar radiofrequency energy on human articular cartilage. Comparison of confocal laser microscopy and light microscopy. Arthroscopy 2001;17:117-123. Farndale RW, Sayers CA, Barrett AJ. A direct spectrophotometric microassay for sulfated glycosaminoglycans in cartilage cultures. Connect Tissue Res 1982;9:247-248. Loeser RF, Todd MD, Seely BL. Prolonged treatment of human osteoarthritic chondrocytes with insulin-like growth factor-I stimulates proteoglycan synthesis but not proteoglycan matrix accumulation in alginate cultures. J Rheumatol 2003; 30:1565-1570. Amiel D, Ball ST, Tasto JP. Chondrocyte viability and metabolic activity after treatment of bovine articular cartilage with bipolar radiofrequency: An in vitro study. Arthroscopy 2004; 20:503-510. Kaab MJ, Bail HJ, Rotter A, Mainil-Varet P, apGwynn I, Woler A. Monopolar radiofrequency treatment of partialthickness cartilage defects in the sheep knee joint leads to extended cartilage injury. Am J Sports Med 2005;33:14721478. Edwards RB III, Lu Y, Rodriguez E, Markel MD. Thermometric determination of cartilage matrix temperatures during thermal chondroplasty: Comparison of bipolar and monopolar radiofrequency devices. Arthroscopy 2002;18:339-346. Hunziker EB, Quinn TM. Surgical removal of articular cartilage leads to loss of chondrocytes from cartilage bordering the wound edge. J Bone Joint Surg Am 2003;85:85-92. Edwards RB III, Lu Y, Uthamanthil RK, Bogdanski JJ, Muir P, Athanasiou KA, Markel MD. Comparison of mechanical debridement and radiofrequency energy for chondroplasty in an in-vivo equine model for partial thickness cartilage injury. Osteoarthritis Cartilage 2007;15:169-178. Mitchell ME, Kidd D, Lotto ML, et al. Determination of factors influencing tissue effect of thermal chondroplasty: An ex vivo investigation. Arthroscopy 2006;22:351-355. Quinn TM, Maung AA, Grodzinsky AJ, Hunziker EB, Sandy JD. Physical and biological regulation of proteoglycan turnover around chondrocytes on cartilage explants. Ann N Y Acad Sci 1999;878:420-441. Redman SN, Dowthwaite GP, Thomson BM, Archer CW. The cellular responses of articular cartilage to sharp and blunt trauma. Osteoarthritis Cartilage 2004;12:106-116. Loeser RF, Shanker G. Autocrine stimulation by insulin-like growth factor 1 and insulin-like growth factor 2 mediates chondrocyte survival in vitro. Arthritis Rheum 2000;43:15521559. Bayliss MT, Osborne D, Woodhouse S, Davidson C. Sulfation of chondroitin sulfate in human articular cartilage. The effect of age, topographical position, and zone of cartilage on tissue composition. J Biol Chem 1999;274:15892-15900. Quinn TM, Hunziker EB, Hauselmann H-J. Variation of cell and matrix morphologies in articular cartilage among locations in the adult human knee. Osteoarthritis Cartilage 2005;13: 672-678.