Immature articular cartilage is more susceptible to blood-induced damage than mature articular cartilageAn in vivo animal study.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 48, No. 2, February 2003, pp 396–403 DOI 10.1002/art.10769 © 2003, American College of Rheumatology Immature Articular Cartilage Is More Susceptible to Blood-Induced Damage Than Mature Articular Cartilage An In Vivo Animal Study Michel J. J. Hooiveld, Goris Roosendaal, Marieke E. Vianen, H. Marijke van den Berg, Johannes W. J. Bijlsma, and Floris P. J. G. Lafeber showed a mild synovitis in animals of all ages, but catabolic inflammatory activity was found only in immature animals. Conclusion. Joints of skeletally immature dogs appeared to be more susceptible than joints of mature dogs to the adverse effects of a joint hemorrhage. These data suggest that for humans, specifically young children are at risk for joint damage after a joint hemorrhage. Objective. Cartilage of young but skeletally mature dogs is more susceptible to blood-induced damage than that of old dogs. The aim of the present study was to investigate whether cartilage of skeletally immature individuals is even more adversely affected by exposure to blood than that of mature individuals, as suggested by clinical practice experience with humans. Methods. Right knees of 3 groups of 6 beagle dogs (skeletally immature, young mature, and old animals) were injected with autologous blood on days 0 and 2. On day 4, cartilage matrix proteoglycan turnover (content, synthesis, and release), synovial inflammation, and cartilage-destructive properties of the synovial tissue were determined and compared with those of the left uninjected control knees. Results. Subsequent to intraarticular bleeding, cartilage proteoglycan content decreased in an agedependent manner, with the largest decrease occurring in cartilage of immature animals. Proteoglycan synthesis per cell also decreased in an age-dependent manner, with the largest decrease occurring in the immature animals. Cartilage proteoglycan release increased in all 3 groups, but the decrease was not age dependent. Interestingly, immature animals showed a large increase in cartilage DNA content upon exposure to blood, whereas mature animals did not. Histologic analysis Joint damage due to recurrent joint hemorrhages is a common feature in patients with hemophilia (1). The mechanism of blood-induced joint damage is not fully understood. In previous studies, we have shown that the combination of mononuclear cells with red blood cells, as present in whole blood, is responsible for the initial adverse changes in cartilage matrix turnover (2). These initial changes are at first independent of inflammatory mediators such as tumor necrosis factor ␣ (TNF␣) or interleukin-1␤ (IL-1␤). Therefore, synovitis, which often accompanies degenerative cartilage damage, is not primarily involved (3). In addition, it was demonstrated in in vivo and in vitro studies that only a transient exposure of cartilage to blood is needed to induce long-lasting changes to cartilage matrix turnover, independent of synovitis, resembling degenerative cartilage damage (4,5). In addition, it appeared that young age predisposed toward an increased risk of cartilage degeneration after joint hemorrhage (6). In an in vivo animal study, cartilage in young but skeletally mature individuals showed more severe adverse changes in cartilage matrix turnover than did cartilage in older animals. Furthermore, the adverse changes in cartilage matrix turnover were found to be long lasting in both groups, but were Michel J. J. Hooiveld, MSc, Goris Roosendaal, MD, PhD, Marieke E. Vianen, BA, H. Marijke van den Berg, MD, PhD, Johannes W. J. Bijlsma, MD, PhD, Floris P. J. G. Lafeber, PhD: University Medical Center Utrecht, Utrecht, The Netherlands. Address correspondence and reprint requests to Michel J. J. Hooiveld, MSc, University Medical Center Utrecht, Room F02.127, PO Box 85500, 3508 GA Utrecht, The Netherlands. E-mail: M.J.J.Hooiveld@AZU.NL. Submitted for publication June 21, 2002; accepted in revised form October 16, 2002. 396 BLOOD-INDUCED JOINT DAMAGE more profound in the younger group (6). These data support the idea that early treatment of joint hemorrhages, specifically in younger individuals, is to be advised. This corroborates current ideas on treatment of patients with hemophilia. It has been found that hemophilia patients show little arthropathy in adulthood if they receive immediate treatment of their first joint hemorrhages, which usually occur early in life (7). Moreover, prophylactic treatment with clotting factors in hemophilia patients is preferable to on-demand treatment (8). In very young (i.e., skeletally immature) patients, prophylactic treatment yielded a better clinical outcome after 22 years of followup (9). Although it appears that intraarticular bleedings have the most significant impact in young patients, the effects of a joint hemorrhage in skeletally immature individuals are not known. It is therefore useful to determine whether joints of immature individuals are even more sensitive to the destructive effects of blood. Knowledge in this respect could be meaningful in terms of the underlying mechanisms of blood-induced joint damage as well as in decision-making in clinical treatment of joint hemorrhages in very young patients. MATERIALS AND METHODS Animals. Eighteen female beagle dogs were obtained from the Animal Facilities of Utrecht University. They were housed in groups of 3 per pen and were fed a standard commercial diet and water ad libitum. Animals were divided into 3 groups: 6 skeletally immature animals (mean ⫾ SD age 0.7 ⫾ 0.02 years), 6 young but skeletally mature animals (2.4 ⫾ 0.3 years), and 6 old animals (7.7 ⫾ 0.7 years). Immaturity was confirmed by radiographs of the knee. The Utrecht University Animal Experiments Ethical Committee had given ethical approval for this study. Experimental procedure. In all dogs, freshly collected autologous blood obtained from the vena jugularis was injected into the right knee on day 0 and day 2, mimicking a single intraarticular bleeding, thus guaranteeing the presence of blood in the joint for 4 days. We chose not to use anticoagulants. We know from previous studies that fibrincontaining plasma has no effect on the cartilage, and we know that coagulated blood has effects similar to those of anticoagulated blood (2). We also know that blood does not (fully) coagulate after autologous intraarticular injection; 2 days after an injection, unclotted blood can be aspirated from the joint (Hooiveld MJJ, Lafeber FPJG: unpublished observations). Moreover, hemophilia patients are treated with coagulation factor as soon as possible (within several hours) after a joint bleed. During the injection procedure, the animals were briefly anesthetized using medatomidine/antesedan (Pfizer, Capelle a/d IJssel, The Netherlands). The contralateral control knee was not injected at all, in order to prevent possible traumatic bleeding. Previous studies showed that control injec- 397 tions with saline did not have any effect on cartilage or synovium (10). The intraarticular injection of blood was confirmed by placing a second needle on the contralateral side of the joint in which blood appeared during injection. When this occurred, the second needle was removed, and 4 ⫾ 2 ml of blood (mean ⫾ SD) was injected intraarticularly. On day 4, all 18 animals were killed using intravenous injection of 10 ml of pentobarbital (200 mg/ml; CEVA, Naaldwijk, The Netherlands). Synovium and cartilage were collected aseptically from both knee joints within 1 hour postmortem and were analyzed according to standard procedures as described below. In addition, the cartilage of both hip joints from different young mature animals was collected in order to test the catabolic activity of culture supernatants of synovial tissue on healthy mature homologous cartilage. Cartilage tissue preparation and analysis. Cartilage was cut as thick as possible, taking care to exclude the underlying bone, from both femoral condyles and the tibial plateau. Two samples from femoral condyles of each knee were fixed in 4% phosphate buffered formalin for histochemistry. Deparaffined sections were stained with Safranin O–fast green–iron hematoxylin and scored according to the modified Mankin criteria (11). Due to the short duration of the experiment, only early signs of degeneration were expected, so the emphases were on loss of Safranin O staining and articular surface irregularities of the cartilage. At least 3 sections from each of the 2 specimens of each knee were stained and scored independently in random order by two observers. Six explants from each femoral and tibial cartilage were used for biochemical analyses of proteoglycan synthesis, release, and content, as well as DNA content. Samples (mean ⫾ SD 5.4 ⫾ 2.8 mg) were cultured, each individually in 200 l culture medium, in 96-well tissue culture plates, according to standard procedures as described previously (12). The culture medium consisted of Dulbecco’s modified Eagle’s medium (DMEM; Gibco, Breda, The Netherlands) supplemented with ascorbic acid (85 M; Sigma, Uithoorn, The Netherlands), glutamine (2 mM; Gibco), penicillin (100 IU/ml; Yamanouchi Pharma, Leiderdorp, The Netherlands), streptomycin sulfate (100 g/ml; Fisiopharma, Palomonte, Italy), and 10% heatinactivated pooled beagle serum. The mean of the outcome values of the 6 samples, from each condyle and tibial plate, was taken as representative of that knee. DNA content. The DNA content of cartilage was determined as a measure of its cellularity. Cartilage samples were digested for 2 hours at 65°C with 3% papain (Sigma) as described (13). Part of the papain digest was used for measurements of DNA using the fluorescent dye Hoechst 33258 (Sigma). Calf thymus DNA (Sigma) was used as a reference standard. Proteoglycan synthesis. As a measure of chondrocyte matrix synthesis, the ex vivo synthesis of proteoglycans (one of the main matrix components) was determined by assessment of the rate of sulfate incorporation (12). Briefly, cartilage samples were cultured for 4 hours in the presence of 14.8 MBq Na235SO4 (carrier free; DuPont NEN, Hoofddorp, The Netherlands) added in 10 l DMEM to the culture medium. After this pulse-labeling period, samples were washed and digested with 3% papain for 2 hours at 65°C. Glycosaminoglycans (GAGs) were precipitated by addition of cetylpyridinium 398 chloride (Sigma). The 35SO42⫺-labeled GAGs in the precipitate were measured by liquid scintillation analysis. The rate of sulfate incorporation, calculated with reference to the specific activity of the medium, was normalized to the DNA content of the explants and is expressed as nmoles of sulfate incorporated per hour per g DNA of the cartilage sample (nmoles/h*g DNA). Proteoglycan release. The release of proteoglycans was determined as a measure of loss of matrix components (12). Cartilage explants were precultured for 1 day, and then the amount of GAGs released into refreshed culture medium during 4 subsequent days of culture was assessed by precipitation with Alcian blue (Fluka, Zwijndrecht, The Netherlands). The release of GAGs was measured photospectrometrically and is expressed as mg GAG released per gram wet weight of cartilage sample. Proteoglycan content. The amount of cartilage GAG was determined as a measure of the proteoglycan content of the cartilage samples (12). In the papain digest, GAGs were precipitated and stained with Alcian blue. The GAG content is expressed in mg GAG normalized to the wet weight of the cartilage explants. Synovial tissue preparation and analysis. Five pieces of synovial tissue from the suprapatellar pouch were taken from each knee. Three were fixed in 4% phosphate buffered formalin for histochemistry. Deparaffined sections were stained with hematoxylin and eosin. To determine the degree of inflammation, at least 3 sections from each of the 3 specimens of each knee were stained and scored. Synovial inflammation was defined by lymphocyte infiltration, hyperplasia of the synovial lining cells, and hypertrophy of the synovial membrane as described previously (12), a slight modification of the scoring system described by Goldenberg and Cohen (14). Two independent observers who were blinded to the source of the tissue scored the tissue samples separately and in random order. Two synovial tissue samples (mean ⫾ SD 60 ⫾ 3 mg) from each experimental and control knee were cultured to assess cartilage-destructive activity. Each of the two samples was cultured for 4 days in 5 ml culture medium. The supernatants of these cultures were harvested, rendered cell free by centrifugation at 351g for 10 minutes, and tested in 50% (volume/volume) concentration for catabolic activity on homologous, mature, healthy hip cartilage using procedures as described above. Release and synthesis of proteoglycans by hip cartilage were determined. Culture supernatants of each tissue sample were tested on 8 cartilage samples per culture condition. The average result in the 16 samples was considered representative of the synovial tissue of the knee from which the two synovium samples were taken. Calculations and statistical analysis. The mean values of cartilage from condyles and plateaus did not differ significantly; therefore, values for condyles and tibia of each knee were averaged (n ⫽ 12 samples per dog). The statistical significance of the differences between control joints and blood-injected joints was evaluated with a nonparametric test for correlated data (Mann-Whitney U test) using the absolute averaged values of each knee. The mean ⫾ SD values are provided for each group of 6 dogs. Data are graphically presented as the percentage change in experimental knees versus contralateral control knees. Age dependency was ana- HOOIVELD ET AL Figure 1. Percentage change in total glycosaminoglycan (GAG) content of cartilage from blood-injected knees compared with that from control knees. GAG content is expressed in mg GAG normalized to the wet weight of the cartilage explants (mg/gm). Bars show the mean and SD. Changes are shown for knee cartilage from immature animals (mean ⫾ SD age 0.7 ⫾ 0.02 years), young mature animals (2.4 ⫾ 0.3 years), and old animals (7.7 ⫾ 0.7 years). The absolute values (mean ⫾ SD) of the GAG content of the control knees for these 3 groups were 37.0 ⫾ 2.7 mg/gm, 53.4 ⫾ 1.8 mg/gm, and 48.3 ⫾ 1.6 mg/gm, respectively. In all groups, the total GAG content of the blood-injected knees was decreased compared with that of the control knees (all P ⬍ 0.05). Immature cartilage showed the largest decrease in this respect (up to 19% loss compared with controls; P ⬍ 0.05). The change in GAG content correlated linearly with age (r ⫽ 0.46, P ⬍ 0.05; n ⫽ 18). lyzed by Spearman’s correlation analysis of these differences. We used the percentage change because of the age-dependent physiologic differences in some of the parameters under control conditions (15,16). Two-tailed P values less than or equal to 0.05 were considered significant. RESULTS Experimental intraarticular blood injections. After 48 hours, just prior to the second intraarticular injection, swelling of the knee due to the first blood injection was no longer palpable, presumably due to absorption of the intraarticular blood. The joint swelling was also not palpable on day 4, just before the animal was killed. The dogs used the joints almost normally within 2 hours after the injections and did not show signs of pain or inconvenience. Biochemical changes in cartilage ex vivo. The total GAG content was statistically significantly diminished in all 3 groups, with the immature cartilage having BLOOD-INDUCED JOINT DAMAGE Figure 2. Percentage change in sulfate incorporation rate as a measure of proteoglycan (PG) synthesis of cartilage from blood-injected knees compared with that from control knees. The sulfate incorporation rate is expressed as nmoles of sulfate incorporated per hour per g DNA of the cartilage sample (nmoles/h*g DNA). Bars show the mean and SD. Changes are shown for knee cartilage from immature, young mature, and old animals. The absolute values (mean ⫾ SD) of PG synthesis of the control knees for these 3 groups were 102.8 ⫾ 22.0 nmoles/h*g DNA, 35.9 ⫾ 6.3 nmoles/h*g DNA, and 24.9 ⫾ 4.6 nmoles/h*g DNA, respectively. In all groups, PG synthesis was decreased after exposure to blood in vivo. The 36% decrease in PG synthesis in immature cartilage (ⴱ ⫽ P ⬍ 0.05 versus control knees) differed significantly from the decreases in the other groups (both P ⬍ 0.05). The degree of decrease in PG synthesis correlated with age (r ⫽ 0.45, P ⬍ 0.05; n ⫽ 18). the largest decrease (19% compared with controls) (Figure 1). There was a significant age-dependent loss of GAG content after exposure to blood, as calculated by correlation analysis (r ⫽ 0.46, P ⬍ 0.05; n ⫽ 18). Proteoglycan synthesis, calculated per cell (i.e., per g DNA), was statistically significantly decreased in all 3 groups, with the immature group showing the largest decrease (36% compared with controls; P ⬍ 0.05) (Figure 2). The decrease in the immature group differed significantly from those in the other 2 groups (both P ⬍ 0.05). In addition, the decreases in proteoglycan synthesis upon exposure to blood changed linearly with age (r ⫽ 0.45, P ⬍ 0.05; n ⫽ 18). The total DNA content in both mature groups did not change significantly after blood injections. However, it increased slightly in young mature cartilage, whereas it decreased in old cartilage. In immature cartilage, the DNA content increased significantly upon exposure to blood (29% compared with controls; P ⬍ 399 0.01) (Figure 3). This increase differed significantly from those in the other 2 groups (both P ⬍ 0.05). The release of GAGs from the cartilage was statistically significantly increased in all 3 groups (Figure 4). In contrast to findings for the other parameters, the most profound increase was observed in the young mature group (22%), significantly greater than those in the immature (5%) and old (10%) groups (both P ⬍ 0.05). Thus, the increase in GAG release upon exposure to blood did not change linearly with age. Histologic analyses of cartilage. All biochemical analyses demonstrated an adverse effect of blood on cartilage proteoglycan turnover; release increased, synthesis diminished, and content decreased as a consequence. This loss of proteoglycans from the cartilage could be confirmed by the loss of Safranin O staining in all blood-exposed condylar cartilage samples. Scoring of loss of cartilage integrity showed that the immature cartilage was graded (mean ⫾ SD) 2.0 ⫾ 1.4, versus 2.4 ⫾ 1.2 for both the young mature and old cartilage groups (Table 1). These histologic differences between the 3 groups were not statistically significant, although Figure 3. Percentage change in total DNA content of cartilage from blood-injected knees compared with that from control knees. Bars show the mean and SD. The absolute values (mean ⫾ SD) of DNA content for control knees were 0.18 ⫾ 0.03 mg/gm wet weight, 0.19 ⫾ 0.02 mg/gm wet weight, and 0.17 ⫾ 0.02 mg/gm wet weight for immature, young mature, and old cartilage, respectively. Both the young mature and the old groups showed no statistically significant change in DNA content after exposure to blood in vivo. The 29% increase in DNA content of immature cartilage (ⴱ ⫽ P ⬍ 0.01 versus control knees) differed significantly from changes in DNA content in the other groups (both P ⬍ 0.05). 400 HOOIVELD ET AL Figure 4. Percentage change in release of glycosaminoglycans (GAGs) in cartilage from experimental knees compared with that from control knees. GAG release is expressed as mg GAG released per gram wet weight of cartilage sample. Bars show the mean and SD. The absolute values (mean ⫾ SD) of GAG release for control knees were 4.6 ⫾ 0.4 mg/gm, 3.8 ⫾ 0.5 mg/gm, and 4.3 ⫾ 0.5 mg/gm for immature, young mature, and old cartilage, respectively. GAG release was enhanced in all groups after intraarticular blood injections (all P ⬍ 0.05). The largest release of GAGs (22%) was found in young mature cartilage. ⴱ ⫽ P ⬍ 0.05 versus each of the other groups. exposed to synovial culture supernatants from control and blood-injected knees of all 3 groups were measured to determine catabolic activity of the synovium. No statistically significant differences between the groups were found in the catabolic activity of the synovia of the control knees, either for release (Figure 5) or for synthesis (Figure 6). Basal catabolic activity, measured by the induced proteoglycan release, was not changed significantly by blood injection in the young mature and old animals. However, a statistically significant increase in GAG release was induced by the experimental synovial culture supernatant from the immature animals (38% compared with control; P ⬍ 0.05) (Figure 5). Proteoglycan synthesis remained unchanged in the synovial culture supernatants from the 3 groups (Figure 6). Histologic analyses of synovium. Histologically, a very mild but statistically significant synovial inflammation was found in blood-injected versus control knees in all 3 groups. The degree of inflammation did not differ significantly between the 3 groups (Table 2), although the highest score was found in the immature group. With respect to the change in inflammatory activity compared the immature and young mature cartilage showed the most severe changes. A correlation of the effect with age did not reach statistical significance, unlike the correlations with age of some of the biochemical parameters. Catabolic activity of synovium. Proteoglycan release and synthesis in homologous mature hip cartilage Table 1. Modified scores of cartilage damage* Cartilage (age) Immature (⬍6 months) Young mature (2–3 years) Old (7–9 years) Control knee Bloodinjected knee Change (blood-injected ⫺ control) 0.1 ⫾ 0.3 0.4 ⫾ 0.5 2.0 ⫾ 1.4† 2.4 ⫾ 1.2† 1.9 ⫾ 1.0 2.0 ⫾ 0.4 1.3 ⫾ 1.0 2.4 ⫾ 1.2† 1.1 ⫾ 0.9 * Values are the mean Mankin score (11) ⫾ SD. After short-term exposure of cartilage to blood in vivo, a slight increase in cartilage degeneration was found. Due to the short duration of the experiment, scoring was based on the loss of Safranin O stain from the cartilage and on articular surface irregularities. Between the 3 groups, no significant differences in cartilage degeneration were found. Within the 3 groups (i.e., between control and experimental cartilage), the differences were significant. † P ⬍ 0.05 versus control cartilage. Figure 5. Total glycosaminoglycan (GAG) release from mature homologous hip cartilage after exposure to control or experimental synovial culture supernatants, as a measure of catabolic activity of the synovium. Bars show the mean and SD. Between the 3 groups, no statistically significant changes were found after exposure of cartilage to culture supernatant of synovium from control knees. Compared with control synovium supernatants, experimental culture supernatants from immature synovium were able to induce increased GAG release from cartilage (38%). Synovial culture supernatants from young mature and old animals were unable to change GAG release. ⴱ ⫽ P ⬍ 0.05 versus control synovium supernatants. BLOOD-INDUCED JOINT DAMAGE 401 Figure 6. Sulfate incorporation as a measure of proteoglycan synthesis in homologous mature hip cartilage after exposure to control or experimental synovial culture supernatants. The sulfate incorporation rate is expressed as nmoles of sulfate incorporated per hour per g DNA of the cartilage sample (nmol/h*g). Bars show the mean and SD. Between the 3 groups, no statistically significant changes were found after exposure of cartilage to culture supernatant of synovium from control knees. In addition, the experimental synovium culture supernatants did not alter the proteoglycan synthesis of healthy cartilage. with the contralateral control, there was a tendency toward an age-dependent effect of exposure to blood. DISCUSSION In recent years, we have shown in in vivo and in vitro studies that a relatively short-term exposure of cartilage to blood leads to lasting adverse changes in chondrocyte activity and cartilage matrix integrity (2,3). Table 2. Modified scores of synovial inflammation* Synovium (age) Immature (⬍6 months) Young mature (2–3 years) Old (7–9 years) Control knee Bloodinjected knee Change (blood-injected ⫺ control) 0.8 ⫾ 0.4 0.5 ⫾ 0.5 2.8 ⫾ 0.8† 2.3 ⫾ 0.6† 2.0 ⫾ 0.9 1.8 ⫾ 0.3 0.8 ⫾ 0.7 2.1 ⫾ 0.9† 1.3 ⫾ 0.8 * Values are the mean inflammation score (modified from Goldenberg and Cohen ) ⫾ SD. Short-term experimental intraarticular bleeding caused a slight increase in synovial inflammation. Between the 3 groups, no significant differences in synovitis were found. Within the 3 groups (i.e., between control and experimental joints), the differences were significant. † P ⬍ 0.05 versus control joints. Results from the present in vivo study show that knee cartilage of (skeletally) immature dogs after 4 days is more susceptible to blood damage than that of young mature and old dogs. The impact of the experimental joint bleed appeared to be age related for several of the parameters measured. This finding suggests that in case of a hemorrhage, cartilage of immature individuals is at greatest risk. This may be true not only in the case of hemophilia, but also in the case of a single or a limited number of traumatic bleedings. The age-related decrease in proteoglycan content was striking in this respect. The age-related inhibition of synthesis of proteoglycans is in accordance with the observations on proteoglycan content; the youngest animals show the most profound loss of proteoglycan content. This was found when synthesis was normalized to the amount of cells (i.e., the DNA content). However, the DNA content showed an age-dependent increase. The DNA content significantly increased in the immature animals, whereas no statistically significant change was observed in young mature animals, and a slight decrease in DNA content was seen in cartilage of old animals. However, although proteoglycan synthesis is decreased in the immature cartilage as a result of blood injections, the cartilage from immature animals still synthesizes larger quantities of proteoglycans per g DNA than cartilage from mature animals. This is because under control conditions, there is an agedependent decrease in proteoglycan synthesis (15), and, specifically in immature cartilage, proteoglycan synthesis is relatively high (17). Interestingly, this implies that in cartilage of immature animals, chondrocytes are triggered and are capable of proliferating upon receipt of a harmful signal. Previous in vitro studies have shown that chondrocyte apoptosis is induced by exposure of cartilage to blood (18). This corroborates the loss of DNA in old individuals. It could be that the loss of healthy chondrocytes is compensated for in immature cartilage, because cells are still able to proliferate in this cartilage (19). Another possibility is that chondrocytes in immature cartilage change from a proteoglycan-synthesizing to a proliferative phenotype, as has been suggested before (20). However, this is difficult to study, since detailed electron microscopy analysis is needed. A simple increase in the number of cells (light microscopy analysis) is not to be expected in 4 days; most cells will still be in a mitotic phase with a doubled amount of DNA, but they will still be recognized as single cells. Cell counting revealed a 6.5% increase in cartilage from blood-injected joints compared with that from 402 control joints (mean ⫾ SD 132 ⫾ 10 chondrocytes/mm2 versus 124 ⫾ 8 chondrocytes/mm2); although an increase was found in 5 of the 6 animals, the mean increase was not statistically significant. Nevertheless, in time this could mean that specifically the immature cartilage could recover from the harmful effects of exposure to blood. Whether this is indeed the case, the outcome after a longer recovery period has to be investigated. Notwithstanding, the present data show that in the short term, despite an increase in DNA, the new chondrocytes in the immature cartilage cannot maintain matrix integrity shortly after exposure to blood, making the cartilage vulnerable to further damage, which is in accordance with observations from clinical practice. It appeared that the cartilage from the young mature animals, and not that from the immature animals, had the greatest increase in GAG release when exposed to blood in vivo. This finding does not explain the fact that GAG content was most profoundly diminished in the immature cartilage. One possible explanation could be that the immature cartilage lost its GAG content at an earlier stage (i.e., during the exposure in vivo), whereas this process of release was still present ex vivo in the young mature cartilage, although with lower intensity. This would be consistent with the abovementioned hypothesis, that although the cartilage of the immature animals is most vulnerable to a bleeding, it may have the better chance of recuperating. On the one hand, the profound release of proteoglycans upon exposure to blood in the young mature animals did not depend on synovial changes. On the other hand, the synovium of immature animals was significantly more triggered by exposure to blood than was the synovium of the other 2 groups, as indicated by its significantly greater potential to release proteoglycans from homologous explants. Proteoglycan synthesis remained unchanged after addition of synovial culture supernatants of all 3 groups. This suggests that exposure to blood in the immature animals specifically triggers activity of proteinases such as matrix metalloproteinases, known to be involved in cartilage breakdown (21,22), and, to a lesser extent, catabolic activity of cytokines such as IL-1 and TNF␣, which are known to have inhibitory effects on cartilage matrix synthesis (23,24). Our results indicate that at a very early stage in immature dogs, the affected synovium adds to the harmful effects of a joint bleed. This finding is in contrast to previous observations in mature animals, in which synovitis occurs only at later stages (4). These data suggest that in order to prevent a more rapid degeneration of cartilage in immature indi- HOOIVELD ET AL viduals, it may be necessary to start prophylactic treatment of severe hemophilia at a very early age. Fischer et al have reported that prophylaxis with clotting factors after the first two hemorrhages in life, in most cases occurring before adulthood, is more beneficial in the long term than starting prophylaxis after more than three hemorrhages (7). Although it remains arguable whether only one joint hemorrhage is sufficient to induce permanent joint damage, our data suggest that one short-term exposure of cartilage to blood is sufficient to induce a temporary catabolic state in the joint. Immature joints are more susceptible in this respect. In summary, age has not previously been considered as a predictor for susceptibility to blood-induced joint damage. We conclude from the present study that joints of skeletally immature individuals are more susceptible than joints of mature individuals to the adverse effects of a joint hemorrhage. These data suggest that specifically in immature individuals, the treatment of joint hemorrhages needs further attention, not only in patients with hemophilia, but also after joint trauma involving joint hemorrhages. REFERENCES 1. Arnold WD, Hilgartner MW. Hemophilic arthropathy: current concepts of pathogenesis and management. J Bone Joint Surg Am 1977;59:287–305. 2. Roosendaal G, Vianen ME, van den Berg HM, Lafeber FP, Bijlsma JW. Cartilage damage as a result of hemarthrosis in a human in vitro model. J Rheumatol 1997;24:1350–4. 3. Roosendaal G, Vianen ME, Marx JJM, van den Berg HM, Lafeber FPJG, Bijlsma JWJ. Blood-induced joint damage: a human in vitro study. Arthritis Rheum 1999;42:1025–32. 4. Roosendaal G, TeKoppele JM, Vianen ME, van den Berg HM, Lafeber FPJG, Bijlsma JWJ. Blood-induced joint damage: a canine in vivo study. Arthritis Rheum 1999;42:1033–9. 5. Roosendaal G, van Rinsum AC, Vianen ME, van den Berg HM, Lafeber FP, Bijlsma JW. Haemophilic arthropathy resembles degenerative rather than inflammatory joint disease. Histopathology 1999;34:144–53. 6. Roosendaal G, TeKoppele JM, Vianen ME, van den Berg HM, Lafeber FP, Bijlsma JW. Articular cartilage is more susceptible to blood induced damage at young than at old age. J Rheumatol 2000;27:1740–4. 7. Fischer K, van der Bom JG, Mauser-Bunschoten EP, Roosendaal G, Prejs R, de Kleijn P, et al. The effects of postponing prophylactic treatment on long-term outcome in patients with severe hemophilia. Blood 2002;99:2337–41. 8. Smith PS, Teutsch SM, Shaffer PA, Rolka H, Evatt B. Episodic versus prophylactic infusions for hemophilia A: a cost-effectiveness analysis. J Pediatr 1996;129:424–31. 9. Fischer K. Treatment strategies for severe haemophilia [dissertation]. Utrecht (The Netherlands): Utrecht Univ.; 2001. 10. DeGroot J. AGEs in the development of osteoarthritis: cartilage synthesis and degradation [dissertation]. Utrecht (The Netherlands): Utrecht Univ.; 2001. 11. Mankin HJ, Dorfman H, Lippiello L, Zarins A. Biochemical and metabolic abnormalities in articular cartilage from osteo-arthritic BLOOD-INDUCED JOINT DAMAGE 12. 13. 14. 15. 16. 17. human hips. II. Correlation of morphology with biochemical and metabolic data. J Bone Joint Surg Am 1971;53:523–37. Lafeber FP, van der Kraan PM, van Roy JL, Huber-Bruning O, Bijlsma JW. Articular cartilage explant culture: an appropriate in vitro system to compare osteoarthritic and normal human cartilage. Connect Tissue Res 1993;29:287–99. Marijnissen AC, van Roermund PM, TeKoppele JM, Bijlsma JW, Lafeber FP. The canine ‘groove’ model, compared with the ACLT model of osteoarthritis. Osteoarthritis Cartilage 2002;10:145–55. Goldenberg DL, Cohen AS. Synovial membrane histopathology in the differential diagnosis of rheumatoid arthritis, gout, pseudogout, systemic lupus erythematosus, infectious arthritis and degenerative joint disease. Medicine (Baltimore) 1978;57:239–52. DeGroot J, Verzijl N, Bank RA, Lafeber FPJG, Bijlsma JWJ, TeKoppele JM. Age-related decrease in proteoglycan synthesis of human articular chondrocytes: the role of nonenzymatic glycation. Arthritis Rheum 1999;42:1003–9. DeGroot J, Verzijl N, Wenting-Van Wijk MJG, Bank RA, Lafeber FPJG, Bijlsma JWJ, et al. Age-related decrease in susceptibility of human articular cartilage to matrix metalloproteinase–mediated degradation: the role of advanced glycation end products. Arthritis Rheum 2001;44:2562–71. Verbruggen G, Cornelissen M, Almqvist KF, Wang L, Elewaut D, Broddelez C, et al. Influence of aging on the synthesis and 403 18. 19. 20. 21. 22. 23. 24. morphology of the aggrecans synthesized by differentiated human articular chondrocytes. Osteoarthritis Cartilage 2000;8:170–9. Hooiveld M, Roosendaal G, Wenting M, van de Berg M, Bijlsma J, Lafeber F. Short term exposure of cartilage to blood results in chondrocyte apoptosis. Am J Pathol. In press. Buckwalter JA, Mankin HJ. Articular cartilage: tissue design and chondrocyte-matrix interactions. Instr Course Lect 1998;47: 477–86. Lafeber FP, van der Kraan PM, van Roy JL, Vitters EL, HuberBruning O, van den Berg WB, et al. Local changes in proteoglycan synthesis during culture are different for normal and osteoarthritic cartilage. Am J Pathol 1992;140:1421–9. Nagase H, Woessner JF Jr. Role of endogenous proteinases in the degradation of cartilage matrix. In: Woessner JF Jr, Howell DS, editors. Joint cartilage degradation: basic and clinical aspects. New York: Marcel Dekker; 1993. p. 159–85. Mort JS, Billington CJ. Articular cartilage and changes in arthritis: matrix degradation. Arthritis Res 2001;3:337–41. Goldring MB. Osteoarthritis and cartilage: the role of cytokines. Curr Rheumatol Rep 2000;2:459–65. Van den Berg WB. Uncoupling of inflammatory and destructive mechanisms in arthritis. Semin Arthritis Rheum 2001;30(5 Suppl 2):7–16.