Clinical Briefs

By Louis Kuritzky, MD

Risk of Cardiovascular Events Associated with Selective COX-2 Inhibitors

It has been suggested that the COX-1 sparing activity of celecoxib (CXB) and rofecoxib (RXB) might reduce the rate of serious NSAID-related gastrointestinal (GI) bleeding. Recently, concern has been expressed that despite favorable GI attributes, cardiovascular adversities might be somehow magnified by the "coxibs." Two of the largest randomized trials performed with CXB and RXB (n ³ 8000 each trial) have demonstrated an increased incidence of thrombotic cardiovascular events. These adversities are the subject of review by Mukherjee and colleagues.

Aspirin and traditional NSAIDs, in addition to reducing inflammation by blockade of COX-2, also affect platelet function through COX-1, making platelet aggregation less likely. Indeed, COX-2 specific agents may decrease vasodilatory and antiaggregatory PGI-2 production, leading to a net increase in likelihood of vascular thrombosis.

In large trials of CXB (the CLASS trial) and RXB (the VIGOR trial), both of which were comparator-controlled with active traditional NSAIDs, the annualized rates of myocardial infarction were significantly greater than the placebo rate found in a recent large meta-analysis. In the VIGOR trial, there was a greater than 2-fold relative risk of confirmed thrombotic cardiovascular events in rofecoxib recipients, compared with naproxen. The CLASS and VIGOR trials are similar, but not identical, in that the CLASS trial included a substantial minority of participants who took doses of aspirin sufficient to prevent myocardial infarction. Mukherjee et al conclude that caution is necessary when using CXB and RXB in persons at risk for cardiovascular morbidity.

Mukherjee D, et al. JAMA. 2001;286: 954-959.

Evaluation of Decision Rules for Referring Women for Bone Densitometry by Dual-Energy X-ray Absorptiometry

Dual-energy x-ray absorptiometry (DEXA) is currently the most popular method of assessing bone mineral density (BMD). Since there are many persons at risk for osteoporosis, the question of which individuals best merit DEXA screening is clinically important. Numerous decision rules have been offered to aid selection of women who merit consideration for DEXA, including the simple calculated osteoporosis risk estimation (SCORE), osteoporosis risk assessment instrument (ORAI), age, body size, no estrogen (ABONE), and National Osteoporosis Foundation (NOF) decision rules. This report compared the sensitivity, specificity, and area under the receiver operating characteristic curve of the above rules in an effort to see which one(s) best predict use in high-risk patients.

Study subjects (n = 2365) included postmenopausal women older than age 45 who underwent DEXA at the femoral neck. Both the SCORE and ORAI decision rules performed superior to the NOF. For simplicity of use, the ORAI instrument, which scores subjects on the basis of age, weight, and use of estrogen, may provide the best balance of sensitivity, specificity, and clinical applicability, though SCORE was similarly effective.

Cadarette SM, et al. JAMA. 2001;286: 57-63.

Albuminuria in Diabetic and Nondiabetic Individuals

Persons with diabetes are burdened with a cluster of risk factors that magnify risk for cardiovascular disease (CVD), including hypertension, renal dysfunction, and dyslipidemia. The presence of microalbuminuria—defined as 30-300 mg/24 hours urine albumin excretion—normal being less than 30 mg/24—is an additional acknowledged CVD risk factor in diabetics. It is less clear whether microalbuminuria is a risk factor of the same magnitude among nondiabetic populations, and whether levels of urine albumin excretion below the currently recognized threshold of 30 mg/24 hours might still be associated with exaggerated CV risk.

To study these issues further, Gerstein and colleagues obtained albumin/creatinine ratio measurements in persons aged 55 or older with either a history of CVD, or diabetes and at least one other cardiovascular risk factor (combined n = 9043). The study information was obtained from data obtained in the HOPE study population, who were followed for a median of 4.5 years.

For persons with urine albumin excretion below threshold for designation as microalbuminuria, there was a graded risk between albumin-creatinine ratio (ACR) and CVD risk. For every 0.4 mm increase in ACR, all-cause mortality was increased by 6.8%. Adverse associations between albumin excretion and CVD were present in diabetic and nondiabetic persons. Gerstein et al suggest that measurement of albuminuria is inexpensive and can be used to help identify persons of greater risk for CVD.

Gerstein HC, et al. JAMA. 2001;286: 421-426.