Aldosterone glucuronidation inhibition as a potential mechanism for arterial dysfunction associated with chronic celecoxib and diclofenac use in patients with rheumatoid arthritis

Objective Adverse cardiovascular (CV) effects of non-steroidal anti-inflammatory drugs (NSAIDs) are largely independent of their cyclooxygenase (COX) enzyme selectivity, but could be a consequence of aldosterone 18β-glucuronidation inhibition (AGI), which varies between NSAIDS. This study assesses the chronic effects of celecoxib (selective COX-2 inhibitor) versus diclofenac (non-selective NSAID) therapy on arterial dysfunction in patients with rheumatoid arthritis (RA). Methods AGI was assessed in vitro using human kidney cortical microsomes. Arterial function was measured clinically as the extent (augmentation index, AIX%) and timing (reflected wave transit time, RWTT, msec) of arterial wave reflection using radial applanation pulse wave analysis (SphygmoCor PWA device) in 39 RA patients without overt CV disease aged 40-65. A higher AIX% (and lower RWTT) indicates arterial dysfunction. Clinical assessment on a single occasion included a fasting blood sample, patient questionnaire and medical record review. Multivariable analysis was used to adjust for sex, mean blood pressure, arthritis duration, cumulative ESR-years and current DMARD therapy. Results The inhibition constant (Ki) for celecoxib was lower than that of diclofenac (Ki, 3.5 vs. 8.4 μM). Chronic celecoxib use was associated with a higher AIX% (34.8 vs. 32.3) and lower RWTT (130.1 vs. 132.7 msec) compared with diclofenac. Adjusted mean differences were AIX% 4.7 (95%CI 0.6 to 8.9; p=0.03) and RWTT -3.6 (95%CI -10.0 to 2.7; p=0.26). Conclusion Celecoxib has a greater potency for AGI than diclofenac and its use is associated with a significantly higher AIX%. Our findings support AGI as a plausible mechanism for the CV toxicity of NSAIDs.