Relationship of Timing of Diclofenac and Celecoxib Use and Fracture Callus Morphology and Morphometry in a Rat Model

Abstract

Background: Fractures are increasingly common worldwide partly due to road traffic collisions, workplace mechanization, and aging. The healing of fractures goes through a series of overlapping stages starting with the inflammatory phase. The use of non-steroidal anti-inflammatory drugs (NSAIDs) to manage the pain related to fractures has, however, been associated with delayed fracture healing presumably due to their effect on the cyclooxygenase (COX) pathway. Osteopontin and Osteocalcin are seldom studied in the context of fracture healing with the use of NSAIDs. Selective COX inhibitors like celecoxib may have a more profound slowing of the fracture healing process when compared to the non-selective molecules including diclofenac. The effect of NSAIDs may be more pronounced in the early fracture healing period when inflammation occurs. Altering the timing of NSAID use by avoiding them in this period may result in differences in fracture healing. The process of aging is also associated with a reduction in healing capacity and a slowing of the fracture healing process. Objective: This study sought to determine the histomorphometry of the rat tibia fracture callus with the use of celecoxib and diclofenac while also investigating the effect of their administration at different timings and the effect of aging. The distribution of osteopontin and osteocalcin in the healing fracture callus was also investigated. Materials and Methods: The study utilized male rats (Rattus norvegicus) which were divided into two groups. One group received 5mg/kg/day of diclofenac while the second group got a similar dosage of celecoxib. The animals were further divided into two sub-groups; one to receive their study medication on the day after fracture induction (early) and the other group to get their medication from the eighth day (late). A further division allocated animals to either young (5 months) or old (15 months) groups. With three animals in each group providing baseline data on day 7, five randomly selected animals from each sub-group were euthanized on day 21 and the rest on day 42. Ethical approval was obtained; the animals were kept in standardized conditions and animal care guidelines were followed throughout the study period. The healing tibia specimens were harvested and processed for light microscopy and immunohistochemistry. After photography, the morphology of the fracture callus was described and the histological grade determined. Imaging software was used to determine the proportions of cartilage and bone. Stereological techniques were used to determine the numeric density of cells and immunohistochemistry was utilized to demonstrate the distribution of osteocalcin and osteopontin. Data was collected using MS Excel before collating and analyzed using SPSS v17.0. After determining the normality of data, the student T-test was used for pair-wise comparisons with the one-way ANOVA utilized for multivariate analysis. The Tukey HSD was used for post hoc analysis with the p set at 0.05. Results: On day 7 the fracture callus was composed mainly of mesenchymal fibrous tissue with little cartilage. While on day 21 the fracture callus was composed of almost equal proportions of mesenchymal fibrous tissue, cartilage, and bone, by day 42 most of this had progressed to mainly bone. Over the same period, the histological grade, the proportion of bone, and the osteoblast density increased while the proportion of cartilage reduced. The celecoxib group‘s fracture callus was characterized by mesenchymal fibrous tissue and cartilage while the diclofenac group was composed mainly of bone and cartilage. The diclofenac group had a higher proportion of bone on day 21 (37.08% v 29.80%, p=0.021). The diclofenac group also had higher histological grades (7.20 v 6.10, p=0.004) and proportion of bone (56.71% v 46.45%, p=0.001) than the celecoxib group on day 42. The cell densities did not vary based on the drug administered on both days. Animals that received medication early had more immature fracture calluses characterized by

fibrous tissue and cartilage. The early group also had a lower proportion of bone on day 42 (47.68% v 55.47%, p=0.017). Older rats showed slower healing with the persistence of cartilage and less bone when compared to the younger ones. The older group also had a lower proportion of bone (47.68% v 55.47%, p=0.017) on day 42. Osteocalcin was expressed in the areas of immature bone with better staining in the younger animals. Osteopontin was localized in bone with robust expression in osteoblasts and osteoclasts. When the different NSAIDs and the different timings were analyzed using multivariate analysis, the use of diclofenac late resulted in higher proportions of bone in the young (p=0.003) and older animals (p=0.049). Analyzing the different drugs in the different age groups; the young diclofenac group had the highest histological grade (p=0.004) and proportion of bone (p=0.003). Conclusion: The use of celecoxib resulted in slowed fracture healing when compared to diclofenac. This was more evident when it was administered in the first week after a fracture and in the older animals. The use of celecoxib especially in the first week after a fracture resulted in a more cellular fracture callus which would precipitate poor biomechanics. Caution is urged when using selective COX-2 inhibitors after fractures in aging individuals.