Selective COX-2 inhibitors and renal injury in salt-sensitive hypertension.

In view of the ongoing controversy of cardiorenal safety of selective COX-2 inhibitors (coxibs), the present study was designed to examine the effects of 2 different coxibs, celecoxib and rofecoxib, compared with a traditional NSAID, diclofenac, and placebo on renal morphology and function in salt-sensitive hypertension. Salt-sensitive (DS) and salt-resistant (DR) Dahl rats were fed with NaCl-enriched diet (4% NaCl) for 8 weeks. Diclofenac (DS-diclofenac), rofecoxib (DS-rofecoxib), celecoxib (DS-celecoxib), or placebo was added to chow from weeks 6 to 8. Immunostaining for monocytes/macrophages (ED1) and cytotoxic T lymphocytes (CD8) was performed. In addition, renal morphology and proteinuria were assessed. Renal cortex mRNA was isolated for determination of COX-2, eNOS, and CRP mRNA by real-time reverse-transcriptase polymerase chain reaction. Untreated hypertensive animals showed glomerular injury including collapsing glomerulopathy, mesangial sclerosis, mesangiolysis, extracapillary proliferation, protein drops, and an especially high grade of glomerulosclerosis (P<0.05 versus DR-placebo) and CD8-positive and ED1-positive cells (P<0.01 versus DR-placebo), which was improved by celecoxib but not by diclofenac and rofecoxib. C-reactive protein mRNA in renal cortex was increased in DS-placebo animals (P<0.05 versus DR-placebo) and normalized by celecoxib (P<0.05 versus DS-placebo), whereas eNOS mRNA was decreased in the DS-rofecoxib group (P<0.05 versus DR-placebo, DS-celecoxib, and DS-diclofenac). Proteinuria was observed in hypertensive animals (P<0.0001 versus DR-placebo), increased by rofecoxib (P<0.05 versus DS-placebo), and normalized by celecoxib (P=0.0015 versus DS-placebo). This head-to-head comparison of selective and nonselective COX inhibitors demonstrates differential effects of coxibs on renal morphology and function in salt-dependent hypertension.