Investigating the causes of chondrocyte death in septic arthritis
Clement, Rhys Gareth Ellis
Septic arthritis is a feared disease that can lead to the destruction of joints and result in significant morbidity. It is caused by an infection within a joint and up to 65% of cases are secondary to Staphylococcus aureus. Antibiotics are the mainstay of treatment and these have been so instrumental in improving the initial outcomes that there have been no major treatment advances since their introduction. The majority of cohort studies investigating septic arthritis consider a treatment success to be the resolution of infection and in the antibiotic era most patients initially appear to have a satisfactory outcome at this point. There is a deficiency within the literature because very few studies have examined the effect on the function of the infected joint over a longer time period. A retrospective cohort study spanning a period of 14 years that included 142 skeletally mature patients with a microbiologically confirmed episode of septic arthritis affecting the native hip or knee was performed. Eighty-two patients (57.7%) had Staphylococcus aureus isolated as the infecting organism. The main outcome measure was a ‘failure of the joint’ defined as a requirement for total joint arthroplasty, development of chronic infection, resection arthroplasty, surgical arthrodesis, or natural ankylosis. The mean survival of affected knee joints at 5 years was 88.3% and at 10 years was 65.5%. At 1 year the mean survival of hip joints was 87.5% and by 5 years was 63.1%. A major finding of this study was the link between the infecting organism and poor outcomes. S. aureus takes a more destructive course with 37% of affected patients experiencing ‘failure of the joint’ compared to only 8% of patients whose infections were caused by other organisms. In addition, patient reported outcome measures were collected on patients with S. aureus septic arthritis of the knee that had not experienced ‘failure of the joint.’ The data demonstrated Western Ontario and McMaster Universities Arthritic Index pain and stiffness scores that were comparable to population norms for the non-infected knees but placed the individuals between the 75th and 90th percentile for the infected knee. These results suggest that a significant insult to articular cartilage occurs during septic arthritis resulting in the gradual development of pain, stiffness and disability. They also indicate that the perceived good results experienced early after the resolution of infection underestimates the damage. A potential explanation for the clinical results is that chondrocyte death could occur during septic arthritis. Chondrocytes are required to maintain healthy articular cartilage and in their absence degenerative change would be expected to accelerate. The predominant cause of damage appears to be related to the infecting organism and further research was required before this could be confirmed. The remainder of this thesis is focused on septic arthritis caused by S. aureus, which is the most common causative organism and appears to be the most destructive. S. aureus alpha haemolysin has recently been identified as a major cause of chondrocyte death in a bovine cartilage explant model of septic arthritis but this experimental model did not assess if there was a contribution to cell death caused by the host immune system and the results needed to be reproduced in an animal model. An in vivo murine model of S. aureus-induced septic arthritis initiated via direct joint inoculation was developed in this study. The main outcome measure was fluorescence-mode confocal laser scanning microscopy (CLSM) that permitted an assessment of chondrocyte viability. Secondary outcome measures allowed a comparison of the magnitude of the inflammatory response and included analysis of clinical parameters (weight, limb swelling and gait changes); measurement of circulating immune populations, cytokines and chemokines using flow cytometry; and histological analysis. C57Bl6 mice were randomised to receive an intra-articular injection of S. aureus 8325-4, S. aureus DU1090 suspended in phosphate buffered saline (PBS) or a PBS control. S. aureus DU1090 is an isogenic mutant of S. aureus 8325-4 that is identical other than an inability to produce alpha haemolysin. Chondrocyte death was assessed 2 days after injection. Chondrocyte death was significantly higher with S. aureus 8325-4 (96.2%) than with S. aureus DU1090 (28.9%) or PBS control (3.8%). There was no statistical difference in the weight loss, limb swelling, gait changes, levels of circulating inflammatory markers or histological scores between the two different strains of S. aureus. PBS injected animals did not develop any signs of septic arthritis.The time course of chondrocyte death following injection of S. aureus 8325-4 was studied in more detail and revealed mean chondrocyte death of 15.6% at 8 hours and 47.6% at 24 hours. The effect of the immune system in isolation was briefly studied following the injection of antibiotic killed S. aureus DU1090. The result was varied with some animals exhibiting a mild inflammatory reaction but no response in others. The mean chondrocyte death was 12.6% after 2 days increasing to 25.5% after 7 days but these results need to be confirmed through further experimentation. The results of the in vivo model allow the following conclusions to be made: 1. Chondrocyte death is an early event in septic arthritis following infection with either S. aureus 8325-4 or S. aureus DU1090. As negligible chondrocyte death was observed in the control group injected with PBS the cause was considered to be the result of S. aureus toxins and/or the immune response. 2. Chondrocyte death was significantly higher with S. aureus 8325-4 (96.2%) than with the isogenic mutant strain DU1090 (28.9%). Analysis of weight loss, limb swelling, gait changes, immune populations and histology did not demonstrate any significant differences between the two S. aureus strains inferring an immune response of equal magnitude. The only difference between the two strains was an inability of DU1090 to produce alpha haemolysin, which was thereby considered to be the most significant cause of chondrocyte death. 3. Mice infected with DU1090 experienced a considerable level of chondrocyte death when compared to PBS injected mice. The inference is that either the immune system or S. aureus toxins other than alpha haemolysin were responsible. 4. Chondrocyte death following the injection of dead S. aureus DU1090 was 12.6% after 2 days increasing to 25.5% after 7 days. A possible cause for this was side effects of the immune system response causing chondrocyte death and this damage appeared to progresses with time. Further research is required to confirm this finding. These results are of translational relevance. Firstly, toxins released by S. aureus have a rapid and fatal action on chondrocytes supporting the immediate and comprehensive washout of joints. Secondly, the confirmation of alpha haemolysin as the most significant damaging toxin to chondrocyte viability may enable the development of future targeted therapeutic strategies in order to reduce the extent of cartilage destruction. Thirdly, if the immune system response is confirmed as a cause of chondrocyte death following activation of an inflammatory response the possibility of developing selective immunomodulation therapies as part of the treatment regime could be explored.