Rosiglitazone (Avandia) and increased risk of CV events

The Buzz: FDA advisory committees recommend removing rosiglitazone
Citation: AAFP News Now, 7/20/2010
Summary: Members of two FDA advisory committees have expressed concerns over the safety of rosiglitazone (Avandia) and have recommended removing the medication from the list of approved drugs. While investigation into the safety of rosiglitazone has been ongoing since 2007, two recent studies from JAMA and the Archives of Internal Medicine
reported significant elevation in CV risk associated with its use. Of note, GlaxoSmithKline, the maker of Avandia, has said that it will record a $2.36 billion legal charge in the second quarter of fiscal year 2010 for settlements related to rosiglitazone.
Commentary: While FDA has yet to make a decision, we should consider getting our patients off rosiglitazone, and certainly not write new prescriptions for it. Pioglitazone (Actos) is considered safer than rosiglitazone, though it should be considered only in patients refractory to or intolerant of first- and second-line therapies.
By: Spencer Blackman MD

UPDATE (9/25/2010): Rosiglitazone has been pulled from the European market, and will only be offered to US patients in whom "every other" diabetes medication has been tried.

Update on guidelines for osteoporosis screening

The Buzz:  Our models for osteoporosis screening are not as robust as we might think.

Citation:  “Screening for Osteoporosis:  An Update for the U.S. Preventive Services Task Force”.  Annals of Internal Medicine. 153 (2); July 20, 2010. 99-111.

Summary:  In 2002, USPSTF recommended bone density testing for women 65 or older, as well as women 60-64 with increased risk for fractures.  No recommendation was made for or against screening other women or men. This review article updates the evidence since their 2002 USPSTF guidelines. Key comments include: 

• No trials exist to establish the effectiveness and harms of osteoporosis screening in decreasing fractures, morbidity or mortality .
• The  Osteoporosis Self-Assessment Screening Tool (OST), based only on age & weight, is the simplest tool for risk stratification.  While more complex models have been validated, no model has been shown to be superior to another.
• Only 2 good quality trials evaluated DEXA scans in men and found comparable results to those in women. 
• Calcaneal Quantitative Ultrasonography (QUS) can predict fractures of femoral neck, hip and spine, but variation exists across studies and correlation with DEXA is low.
• Only 1 study evaluated the optimum screening interval for osteoporosis. Using a cohort of post-menopausal women, it found that repeating BMD up to 8 years after the initial screening did not significantly improve risk estimates for fractures.
• For women, bisphosphonates (such as alendronate, etc), parathyroid hormone (PTH), raloxifine and estrogen decreased the risk of vertebral fractures.  Bisphosphonates decreased the risk of non-vertebral fractures.  Medications were shown to be effective in patients whose BMD was -2.5 or less.   BMD > -2.5 showed a trend of decreasing risk, but was not found to be statistically significant.
• For men, only 1 study evaluated effectiveness of a medication, PTH, and found that  PTH was associated with a trend towards decreasing fractures, but was not found to be statistically significant.
• Case reports have linked serious GI events, atrial fibrillation, osteonecrosis of the jaw, severe musculoskeletal pain, and esophageal adenocarcinoma to use of bisphosphonates.  However, when the data for these adverse events were examined, the association was tentative.
• Raloxifine and estrogen have been shown to increase thromboembolic events.  Estrogen increases the risk of stroke.  Estrogen with progestin increases the risk of coronary heart disease as well as breast cancer. 

 Commentary:  No direct evidence exists that screening for osteoporosis actually improves outcomes, although indirect evidence suggests a benefit.  

By: Jennifer Young, MD

Should we use aggressive medical treatment to stop the progression of diabetic retinopathy?

The Buzz: In type 2 diabetics, intensive glucose and lipid control (but not intensive BP control) prevents the progression of diabetic retinopathy (DR), but the approach may harm more patients than it benefits.
Citation: NEJM 363;3 July 15, 2010
Summary: As part of a sub-study of the large multi-center RCT ACCORD trial, researchers evaluated 10,251 diabetic patients randomly assigned to intensive (A1c < 6.0%) vs standard therapy (A1c < 7.0-7.9%), and were followed to monitor progression of DR, including the need for laser surgery or vitrectomy. Decreased rates of progression of DR were seen in the intensive glucose (7.3 vs 10.4%, P = 0.003) and lipid (6.5 vs 10.2%, P = 0.006) groups, but not the intensive BP group (10.4 vs 8.8%, P = 0.29). However, the trial was halted early, and a large portion of patients with higher LDL levels and alb/creat scores and lower visual acuity scores did not receive 4 year follow-up.

Commentary: These findings need to be interpreted in conjunction with the ACCORD trial, since DR is only one CV endpoint that is seen in diabetics.  In fact, the trial was stopped early when results suggested intensive glucose lowering was found to have no benefit in MI/CVA prevention and increased all-cause mortality. These findings were not replicated in the newer ADVANCE trial, which used different glucose lowering agents.  However, despite this newer trial and the benefit seen in DR, it is likely most prudent to maintain an A1c goal of 7.0-7.9%. In addition, this study suggests adding fenofibrate therapy to a statin may be useful to slow DR progression, particularly if the patient is male and has high trigylcerides or low HDL, as CV risk reduction with fenofibrate was found in these subsets.

By: Elizabeth Haskins, MD

Dietary sugar intake and risk of hypertension

The Buzz: Could the rise in rates of hypertension in industrialized nations be correlated with excess dietary sugar? 

Citation: Journal of American Society of Nephrology 21: 2010, “Increased Fructose Associates with Elevated Blood Pressure.” 

Summary: The prevalence of hypertension has risen dramatically in the last century, going from 5-10% of the population in at the turn of the century, to approximately 31% of the adult population today.  Consumption of dietary sugar has also spiked, though the two have been inconsistently linked. To further investigate that connection, this cross-sectional analysis used data from 4528 adults without history of hypertension participating in  the National Health and Nutrition Examination Survey (NHANES 2003-2006), reporting  that consumption of excess sugar equivalent of  2.5 soft drinks per day was “independently and significantly associated with higher odds of elevated blood pressure (odds ratio 2.10; 95% CI 1.20 to 3.61).”  Speculated mechanisms included stimulation of the sympathetic nervous system, inhibition of endothelial nitric oxide synthase activity, and stimulation of uric acid. Study authors noted a few limitations, mainly that cause-effect relationships cannot be deduced from cross-sectional analysis and that self-reporting of consumption could have led to mis-classification or under-reporting.  However, the large sample of participants was representative of the US adult population and researchers made attempts to isolate the impact of fructose intake by controlling for a large number of confounding factors. 

Commentary: This research provides more weight to the suggestion that added sugars have a deleterious effect on cardiovascular health. 

By: Jeannie Bianchi, L.Ac.

Controversy Alert: Statins for Primary Prevention of CVD

The Buzz: Should statins be used to prevent CVD in patients without known disease? 
Citation: Arch Intern Med 170:12 June 20, 2010 
Summary: Three quarters of the patients who are taking statins are using them for primary prevention, i.e. to delay or prevent the onset of atherosclerosis and to reduce the incidence of heart attacks, strokes and other sequelae. The debate over the evidence for this use has been simmering for some time, and a recent issue of Archives of Internal Medicine presents two new papers which add significant fuel to the fire.

The first, by Ray et al, is a meta-analysis of 11 RCTs involving 65,229 patients (including the recent JUPITER trial) which found no significant reduction in risk associated with the use of statins. The strengths of this study include its large size, exclusion of patients with known CVD, and apparent lack of authors' conflict of interest. However, the average period of follow-up among the studies included in the report was just 3.7 years, and it is possible that longer term use may confer additional benefit no found in this report, though evidence for this is lacking.

The second, by de Lorgeril et al, is an analysis of the 2008 JUPITER study, a controversial trial which was ended early after just two years and reported a significant reduction in CVD-related events with the use of rosuvastatin (Crestor). In this current reappraisal of JUPITER, the authors point out a number of serious flaws with the methodology and results, including:

• The premature termination of the JUPITER trial due to the "clear benefit" in the treatment arm was based on unclear criteria, and truncated trials have been shown to be associated with greater effect sizes than those which are not stopped early.
• Significant conflicts of interest existed in the JUPITER trial. It was sponsored by the makers of Crestor, 9 of 14 authors have financial ties to the sponsor, and the principle investigator is a co-holder of the patent for the CRP test, which would be used much more frequently if the results of the trial are to be believed.
• The all-cause mortality curves were converging when the trial was stopped, suggesting the difference between the two groups may have disappeared with more time
• The authors found a number of inconsistencies that suggested major limitations to the data set. For example, the number of participants who had an MI who died (the case-fatality rate) was extremely low in the study (5-18%) as compared to the expected rate of 40-50%. Moreover, treatment with rosuvastatin appeared to triple the case-fatality rate, an effect which does not seem credible.

The papers were accompanied by an editorial which concluded that "we do not know" if there is any benefit of statins in primary prevention of CVD.
Commentary: These papers call into serious question whether we should be prescribing statins to patients without known CVD, especially if it decreases the importance we place on improvements in lifestyle, including smoking cessation, exercise, and diet. 
By: Spencer Blackman, MD

Genetic test predicts lifespan? Not ready for prime time.

The Buzz: Researchers report a study identifying genetic markers for longevity
Citation: New York Times; July 2, 2010
Summary: A study published in this week's issue of Science reports the identification of genetic markers associated with living to 100, and claims a 77% acuracy rate. Researchers analyzed the DNA of 1,055 centenarians and identified a pattern of 150 genetic markers which seemed to confer longevity. They then looked at another group of centenarians and found 77% of them had the same pattern. While the study authors admit the biology behind these findings has yet to be explained, they feel this may be an importnat step towards understanding the genetics of longevity. Of special interest, the centenarians had equal numbers of disease-related genetic markers, suggesting these 150 may be protective factors that delay the onset of diseases of old age.
Commentary: The bottom line - this is simply a statistical analysis that correlates a certain genetic make-up with longevity. The test is not available to the general public and is likely not to be any time soon. But given the amount of press it is getting, understanding the implications, and lack thereof, may be helpful in talking to our patients.
By: Spencer Blackman MD