Clinical Briefs

By Louis Kuritzky, MD

Efficacy and Safety of Low-Dose Aspirin in Polycythemia Vera

The increased red cell mass diagnostic of polycythemia vera (PCV) results in blood hyperviscosity, which is associated with increased thrombotic events. Initial enthusiasm for the concept of ASA thromboprophylaxis in PCV was dampened by a 1986 trial of aspirin (ASA) at a dose of 900 mg/d, in which an unacceptably high incidence of major GI bleeding was seen. In non-PCV populations, low-dose ASA has been shown to provide effective thromboprophylaxis, with lesser risk of major GI bleeding.

Plasma thromboxane (a direct stimulator of platelet activation) levels are elevated as much as 10 fold in PCV, a situation parallel to that seen in acute coronary syndromes, in which ASA has been proven to provide dramatic risk reduction. Since even low-dose ASA results in substantially reduced platelet thromboxane production, but less GI bleeding, the potential merit of such a clinical trial is straightforward.

PCV patients lacking any other direct indication for ASA (eg, previous MI) were enrolled in a double-blind placebo-controlled randomized trial to compare 100 mg ASA with placebo (n = 518). The 2 primary end points of the study were: 1) nonfatal MI + nonfatal stroke + CV death; 2) nonfatal MI + nonfatal stroke + PE + DVT + CV death. Secondary end points included individual thrombotic components of the above.

After a mean followup of 3 years, ASA reduced the primary end point #2 by 60%; primary end point #1 was reduced 59%, but did not achieve statistical significance. The lower ASA dose (100 mg/d) demonstrated excellent safety, with no statistically significant increased risk of either major or minor bleeding compared to placebo. Landolfi and associates recommend consideration of low-dose ASA for thromboprophylaxis in PCV.

Landolfi R, et al. N Engl J Med. 2004;350:114-124.

Coronary Artery Calcium Score Plus Framingham Score for Risk Prediction

The Framingham Risk Score (FRS) is a commonly recommended tool for estimating risk of coronary events (CHD) in asymptomatic persons (asymptomatic for CHD, that is). It provides an assessment of the likelihood of experiencing a CHD event in the next 10 years. Despite inclusion of age, sex, smoking, BP, lipids, and glucose, the FRS is imperfect in identifying those at CV risk, especially for those determined to be at intermediate risk’ (FRS = 10-19%). Another tool used for CHD risk stratification is coronary artery calcium scoring (CACS), as obtained by CT. The purpose of this trial was to ascertain whether combining the 2 enhances accuracy.

Asymptomatic persons older than 45 (n = 1461) with at least one CHD risk factor (but without prior MI or proven CAD) were enrolled. Diabetics were excluded because CACS has not proven effective in this population, who are by definition already recognized as high risk for CAD at presentation. Patients were followed up to 8.5 years (mean, 7 years).

For persons with a FRS of at least 10% (but < 20%), a CACS greater than 300 (highest quartile CHD risk) significantly modified risk prediction. For instance, a FRS 10-year risk prediction of 10% was increased to 13-19% when coupled with a CACS score > 300.

Greenland and associates suggest that for low-risk (FRS < 10%) and high risk (FRS > 20%) individuals, CACS adds little. Prognostication about the intermediate risk group (FRS = 10-19%) is enhanced by combining the tools. 

Greenland et al. JAMA. 2004;291:210-215.

Intra-Articular Hyaluronic Acid in the Treatment of Knee Osteoarthritis

The use of hyaluronic acid (HUA) injection in human subjects began in 1997, following a history of similar treatment in veterinary medicine. HUA is a constituent of normal synovial fluid, and has been conceptualized as a "joint lubricant." Because of mixed efficacy responses in clinical trials, clinician acceptance of this treatment modality for osteoarthritis of the knee (OA) has been somewhat tepid.

This metaanalysis included 22 trials, with almost 3000 patients. To quantify treatment effects, an "effect size" metric was used; 0.2-0.5 is a "small" effect size, comparable to the advantage of NSAIDs over acetaminophen in OA treatment trials. Analysis included all recipients of HUA injection, but was further separated out into groups based upon whether subjects had received standard, or highest molecular weight HUA.

Overall, HUA was found to provide a modest benefit (effect size = 0.32); Lo and associates discuss that even this result may be overoptimistic, since publication bias was discerned amongst HUA injection trials. According to this analysis, whether highest molecular weight HUA is more advantageous than other configurations remains indeterminate. Lo et al call for further independent trials to provide greater clarification of HUA efficacy. 

Lo GH, et al. JAMA. 2003;290:3115-3121.

Dr. Kuritzky, Clinical Assistant Professor, University of Florida, Gainesville, is Associate Editor of Internal Medicine Alert.