Objective: Polymyalgia rheumatica (PMR) has an abrupt onset of inflammatory symptoms, making it a useful model for studying the effects of inflammation in bone. PMR requires corticosteroid treatment, which may itself have a detrimental effect on bone. This study used serially measured biochemical markers of bone turnover and bone density to address the relative contributions of systemic inflammation and corticosteroid therapy to bone loss.
Methods: Fifty untreated patients with PMR were randomized to receive oral prednisolone or intramuscular methylprednisolone. Biochemical bone markers (pyridinoline [PYR], deoxypyridinoline [DPYR], procollagen type I carboxy-terminal peptide [PICP]) and bone mineral density (BMD) were measured at baseline and at 6, 12, and 24 months.
Results: The median disease duration at presentation was 12 weeks (range 5-32 weeks). Levels of urinary crosslinks were increased in patients with untreated PMR compared with controls (PYR 74.9 +/- 30.0 nmoles/mmole creatinine, DPYR 14.6 +/- 6.4 nmoles/mmole creatinine [mean +/- SD]; P = 0.0001); the PICP level was normal (115.0 +/- 39.0 microg/liter). With treatment, the crosslinks levels fell and PICP levels rose within 6 months (P = 0.01). Bone resorption (PYR) correlated with untreated disease activity (erythrocyte sedimentation rate [ESR]) (r = 0.5, P = 0.003) and with interleukin-6 levels (r = 0.48, P = 0.05). There was a significant reduction in BMD of both the hip and the spine after 12 months of treatment (P = 0.0002), with no difference between treatment groups. As the steroid dosage was reduced, bone mass improved. Initial ESR influenced the percent change in BMD at 1 year (r = 0.35, P = 0.05), while cumulative steroid dose, mean ESR, and type of steroid used did not.
Conclusion: Inflammation in PMR increases bone resorption and appears to have a more detrimental effect on bone than does low-dose corticosteroid. If corticosteroids can be tapered and discontinued, bone loss in PMR can be a transient phenomenon.