Avandia-heart-lawyers.com

Effect of Rosiglitazone on the Risk of Myocardial Infarction Steven E. Nissen, M.D., and Kathy Wolski, M.P.H.
Background
Rosiglitazone is widely used to treat patients with type 2 diabetes mellitus, but its From the Cleveland Clinic, Cleveland. Ad­
effect on cardiovascular morbidity and mortality has not been determined.
the Department of Cardiovascular Medi­cine, Cleveland Clinic, 9500 Euclid Ave., Cleveland, OH 44195, or at nissens@ccf.
org.
We conducted searches of the published literature, the Web site of the Food and Drug Administration, and a clinical-trials registry maintained by the drug manu- This article (10.1056/NEJMoa072761) was facturer (GlaxoSmithKline). Criteria for inclusion in our meta-analysis included a published at www.nejm.org. study duration of more than 24 weeks, the use of a randomized control group not N Engl J Med 2007;356.
receiving rosiglitazone, and the availability of outcome data for myocardial infarc- Copyright 2007 Massachusetts Medical Society. tion and death from cardiovascular causes. Of 116 potentially relevant studies, 42 trials met the inclusion criteria. We tabulated all occurrences of myocardial infarc- tion and death from cardiovascular causes. Results
Data were combined by means of a fixed-effects model. In the 42 trials, the mean
age of the subjects was approximately 56 years, and the mean baseline glycated hemoglobin level was approximately 8.2%. In the rosiglitazone group, as compared with the control group, the odds ratio for myocardial infarction was 1.43 (95% confidence interval [CI], 1.03 to 1.98; P = 0.03), and the odds ratio for death from cardiovascular causes was 1.64 (95% CI, 0.98 to 2.74; P = 0.06).
Conclusions
Rosiglitazone was associated with a significant increase in the risk of myocardial
infarction and with an increase in the risk of death from cardiovascular causes that had borderline significance. Our study was limited by a lack of access to original source data, which would have enabled time-to-event analysis. Despite these limita- tions, patients and providers should consider the potential for serious adverse car- diovascular effects of treatment with rosiglitazone for type 2 diabetes.
Downloaded from www.nejm.org on May 22, 2007 . For personal use only. No other uses without permission. Copyright 2007 Massachusetts Medical Society. All rights reserved. T h e n e w e ng l a n d j o u r na l o f m e dic i n e Thiazolidinedione drugs are wide- of drug exposure. Six of the 48 trials did not re- ly used to lower blood glucose levels in pa- port any myocardial infarctions or deaths from tients with type 2 diabetes mellitus. In the cardiovascular causes and therefore were not in- United States, three such agents have been intro- cluded in the analysis because the effect measure duced: troglitazone, which was removed from could not be calculated. Of the remaining 42 the market because of hepatotoxicity, and two studies, 38 reported at least one myocardial in- currently available agents, rosiglitazone (Avan- farction, and 22 reported at least one death from dia, GlaxoSmithKline) and pioglitazone (Actos, cardiovascular causes. In these trials, 15,560 pa- Takeda). The thiazolidinediones are agonists for tients were randomly assigned to regimens that peroxisome-proliferator–activated receptor γ included rosiglitazone, and 12,283 were assigned (PPAR-γ). PPAR-γ receptors are ligand-activated to comparator groups with regimens that did not nuclear transcription factors that modulate gene include rosiglitazone.
expression, lowering blood glucose primarily by Multiple groups of patients who received rosig- increasing insulin sensitivity in peripheral tis- litazone within a single trial were pooled to- sues.1,2 Rosiglitazone was introduced in 1999 gether, when applicable. The control group was and is widely used as monotherapy or in fixed- defined as patients receiving any drug regimen dose combinations with either metformin (Avan- other than rosiglitazone. The trials fall into damet, GlaxoSmithKline) or glimepiride (Avan- three categories. One group includes five of the studies submitted to the FDA for the March 22, The original approval of rosiglitazone was 1999, advisory board hearing that recommended based on the ability of the drug to reduce blood approval of rosiglitazone. Group-level data from glucose and glycated hemoglobin levels.3 Initial these five studies are available in publicly dis- studies were not adequately powered to deter- closed briefing documents archived on the FDA mine the effects of this agent on microvascular Web site.6 Data from these same trials are also or macrovascular complications of diabetes, in- reported in a summary fashion on a clinical- cluding cardiovascular morbidity and mortality.3 trial registry Web site maintained by the drug However, the effect of any antidiabetic therapy manufacturer, GlaxoSmithKline.5 Reports of four on cardiovascular outcomes is particularly im- of these five trials were also published in peer- portant, because more than 65% of deaths in reviewed journals.7-9 In these five trials, 1967 patients with diabetes are from cardiovascular patients were randomly assigned to receive rosig- causes.4 Therefore, we performed a meta-analy- litazone, and 793 patients were assigned to re- sis of trials comparing rosiglitazone with pla- ceive various comparator drugs (Table 1).
cebo or active comparators to assess the effect Other studies that we included in the meta- of this agent on cardiovascular outcomes. The analysis were initially identified in the Glaxo- source material for this analysis consisted of SmithKline clinical-trial registry.5 As noted in publicly available data from the original regis- Table 1, we included 35 studies in this category, tration package submitted to the Food and Drug 9 of which were published in peer-reviewed jour- Administration (FDA), another series of trials nals and 26 of which remain unpublished.10-18 performed by the sponsor after approval, and two Whenever possible, the results obtained on the large, prospective, randomized trials designed to GlaxoSmithKline Web site were cross-checked study additional indications for the drug.
with the publication. In cases of disagreement between published and unpublished data, data derived from the manufacturer’s Web site were used. In this group of 35 trials, 9502 patients Analyzed Studies
were randomly assigned to receive rosiglitazone, Table 1 lists the 42 trials included in this meta- and 5961 patients were assigned to receive vari- analysis. We screened 116 phase 2, 3, and 4 trials ous comparator drugs.
for inclusion. Of these, 48 trials met the pre- A third data source consisted of two large, defined inclusion criteria of having a random- recently published trials, the Diabetes Reduction ized comparator group, a similar duration of Assessment with Ramipiril and Rosiglitazone treatment in all groups, and more than 24 weeks Medication (DREAM) trial20 and the A Diabetes Downloaded from www.nejm.org on May 22, 2007 . For personal use only. No other uses without permission. Copyright 2007 Massachusetts Medical Society. All rights reserved. Rosiglitazone and Cardiovascular Outcomes Outcome Prevention Trial (ADOPT) (ClinicalTrials. for the Q statistic indicated a lack of heterogene- gov number, NCT00279045).21 In the DREAM ity across trials, allowing for the use of a fixed- study, 2635 patients were randomly assigned to effects model. For additional analyses, the active receive rosiglitazone and 2634 patients were as- comparator control groups were subgrouped into signed to receive placebo. The DREAM study was the following four classes for comparison with designed to determine whether rosiglitazone rosiglitazone: metformin, sulfonylurea, insulin, could prevent the development of type 2 diabetes and placebo. Odds ratios and 95% confidence in- in patients at high risk for this disorder. In the tervals were calculated for each subgroup with the ADOPT trial, 1456 patients were randomly as- use of methods similar to those used in the signed to receive rosiglitazone and 2895 patients pooled analyses. Data were analyzed with the use were assigned to receive either metformin or of Comprehensive Meta-Analysis software, version glyburide. The ADOPT study was designed to 2.2 (Biostat).
assess the durability of glycemic control with rosiglitazone therapy, as compared with therapy Baseline Characteristics
Outcome Measures
Table 2 reports the doses of rosiglitazone and We reviewed data summaries provided in the comparator drugs, baseline demographic charac- FDA review documents, the GlaxoSmithKline teristics, study periods, and glycated hemoglobin clinical-trial registry Web site, and published levels or fasting blood glucose levels for patients trial results and then abstracted from the ad- enrolled in the trials. The patients were relatively verse-event tabulations information on myocar- young, averaging less than 57 years of age for both dial infarction and death from cardiovascular the rosiglitazone group and the control group. causes. With the exception of the DREAM study, Overall, there was a moderate predominance of the included trials did not describe adjudication men. Diabetes control was relatively poor, with a of myocardial infarction or death from cardio- mean baseline glycated hemoglobin level of ap- vascular causes. Time-to-event data for cardio- proximately 8.2% for both study groups.
vascular events were not available in any of these trials, which precluded the calculation of hazard Myocardial Infarction and Death
ratios. Because only summary data were avail- Table 3 reports the myocardial infarction events able, it was not possible to discern whether the and deaths from cardiovascular causes that were same patient had both events. Therefore, an out- reported in the 42 clinical trials we reviewed. come measure based on the composite of death There were 86 myocardial infarctions in the rosig- or myocardial infarction could not be construct- litazone group and 72 in the control group. ed. Accordingly, these two outcomes are reported There were 39 deaths from cardiovascular causes in the rosiglitazone group and 22 in the control group. Table 4 lists the odds ratios, 95% confi- Statistical Analysis
dence intervals, and P values for myocardial in- Many trials had few cardiovascular events, so the farction and death from cardiovascular causes odds ratios and 95% confidence intervals were for the rosiglitazone group and the control group. calculated with the use of the Peto method.22-24 The summary odds ratio for myocardial infarc- Because all trials had similar durations of follow- tion was 1.43 in the rosiglitazone group (95% up for all treatment groups, the use of odds ra- confidence interval [CI], 1.03 to 1.98; P = 0.03). tios represents a valid approach to assessing the The odds ratio for death from cardiovascular risk associated with the use of rosiglitazone. Tri- causes in the rosiglitazone group, as compared als in which patients had no adverse cardiovas- with the control group, was 1.64 (95% CI, 0.98 to cular events in either group were excluded from 2.74; P = 0.06). Table 4 also lists odds ratios and analyses. All reported P values are two-sided. 95% confidence intervals for the pooled group of Statistical heterogeneity across the various trials trials that were smaller and of shorter duration; was tested with the use of Cochran’s Q statistic. results for the DREAM and ADOPT studies are A P value of more than the nominal level of 0.10 shown separately.
Downloaded from www.nejm.org on May 22, 2007 . For personal use only. No other uses without permission. Copyright 2007 Massachusetts Medical Society. All rights reserved. T h e n e w e ng l a n d j o u r na l o f m e dic i n e Rosiglitazone
Duration
Meta-Analysis.*
the
Registry
Rosiglitazone
original
Reference
included
Table 1. Clinical
Additional
Downloaded from www.nejm.org on May 22, 2007 . For personal use only. No other uses without permission. Copyright 2007 Massachusetts Medical Society. All rights reserved. Rosiglitazone and Cardiovascular Outcomes pro trials
randomized
published
administered administered administered type administered spective,
Recently
Downloaded from www.nejm.org on May 22, 2007 . For personal use only. No other uses without permission. Copyright 2007 Massachusetts Medical Society. All rights reserved. T h e n e w e ng l a n d j o u r na l o f m e dic i n e Glycated
Baseline Hemoglobin
Levels.*
Hemoglobin
Population
Glycated
Periods,
Characteristics,
Demographic
Baseline
Table 2. Doses,
Downloaded from www.nejm.org on May 22, 2007 . For personal use only. No other uses without permission. Copyright 2007 Massachusetts Medical Society. All rights reserved. Rosiglitazone and Cardiovascular Outcomes Downloaded from www.nejm.org on May 22, 2007 . For personal use only. No other uses without permission. Copyright 2007 Massachusetts Medical Society. All rights reserved. T h e n e w e ng l a n d j o u r na l o f m e dic i n e Glycated
Baseline Hemoglobin
Population
Table 2. (Continued.)
Downloaded from www.nejm.org on May 22, 2007 . For personal use only. No other uses without permission. Copyright 2007 Massachusetts Medical Society. All rights reserved. Rosiglitazone and Cardiovascular Outcomes Downloaded from www.nejm.org on May 22, 2007 . For personal use only. No other uses without permission. Copyright 2007 Massachusetts Medical Society. All rights reserved. T h e n e w e ng l a n d j o u r na l o f m e dic i n e Table 3. Myocardial Infarctions and Cardiovascular Deaths in Rosiglitazone Trials.
Rosiglitazone Group
Control Group
Downloaded from www.nejm.org on May 22, 2007 . For personal use only. No other uses without permission. Copyright 2007 Massachusetts Medical Society. All rights reserved. Rosiglitazone and Cardiovascular Outcomes Table 3. (Continued.)
Rosiglitazone Group
Control Group
Table 5 lists odds ratios for myocardial in- events in patients with diabetes.4 Because expo- farction and death from cardiovascular causes sure of such patients to rosiglitazone is wide- associated with rosiglitazone for subgroups de- spread, the public health impact of an increase in fined according to the comparator drug. Similar cardiovascular risk could be substantial if our results were obtained when the analysis exclud- data are borne out by further analysis and the ed trials with an active comparator group. The results of larger controlled trials.
heterogeneity P values were 0.53 for myocardial infarction and 0.68 for death from cardiovascu- source data to construct a composite outcome lar causes across subgroups. As compared with that included myocardial infarction or death from placebo or other antidiabetic regimens, the esti- cardiovascular causes, the increase in the odds mated odds ratios in all cases were greater than ratios for both of these end points suggests that 1.0, suggesting that observed adverse effects dur- observed adverse effects associated with rosiglit- ing rosiglitazone treatment were not unique to azone were probably not due to chance alone. This meta-analysis included a group of trials that In an analysis that was not prespecified, we were of relatively short duration (24 to 52 weeks). also studied the effects of rosiglitazone on death The odds ratio for these shorter-term trials was from any cause. The odds ratio for death from any similar to the overall results of the meta-analy- cause was 1.18 (95% CI, 0.89 to 1.55; P = 0.24). sis. Thus, in susceptible patients, rosiglitazone therapy may be capable of provoking myocardial infarction or death from cardiovascular causes after relatively short-term exposure. In contrast, Our data show that, as compared with placebo or long-term therapies that improve cardiovascular with other antidiabetic regimens, treatment with outcomes, such as statins and antihypertensive rosiglitazone was associated with a significant drugs, often take several years to provide benefits. increase in the risk of myocardial infarction and Notably, the estimates for the odds ratios for with an increase in the risk of death from cardio- myocardial infarction and death from cardiovas- vascular causes that was of borderline signifi- cular causes appear elevated for rosiglitazone in cance. The similar odds ratio for comparison comparison with placebo or other commonly pre- with placebo suggests that the increased risk as- scribed antidiabetic therapies (Table 5).
sociated with rosiglitazone was not a function of The mechanism for the apparent increase in the protective effects of active comparator drugs. myocardial infarction and death from cardiovas- However, these findings are based on limited ac- cular causes associated with rosiglitazone remains cess to trial results from publicly available sourc- uncertain. One potential contributing factor may es, not on patient-level source data. Furthermore, be the adverse effect of the drug on serum lipids. results are based on a relatively small number of The FDA-approved rosiglitazone product label events, resulting in odds ratios that could be af- reports a mean increase in low-density lipopro- fected by small changes in the classification of tein (LDL) cholesterol of 18.6% among patients events. Nonetheless, our findings are worrisome treated for 26 weeks with an 8-mg daily dose, as because of the high incidence of cardiovascular compared with placebo.25 In observational stud- Downloaded from www.nejm.org on May 22, 2007 . For personal use only. No other uses without permission. Copyright 2007 Massachusetts Medical Society. All rights reserved. T h e n e w e ng l a n d j o u r na l o f m e dic i n e Table 4. Rates of Myocardial Infarction and Death from Cardiovascular Causes.
Odds Ratio
Rosiglitazone Group
Control Group
Myocardial infarction
Death from cardiovascular causes
ies and lipid-lowering trials, elevated levels of adverse cardiovascular events, including myocar- LDL cholesterol were associated with an increase dial infarction, during phase 2 and 3 testing.28 in adverse cardiovascular outcomes. Thus, an After publication of an analysis of cardiovascu- increase in LDL cholesterol of the magnitude lar outcomes, muraglitazar was not approved by observed in the rosiglitazone group may have the FDA, and further development was subse- contributed to adverse cardiovascular outcomes, quently halted by the manufacturer. Development although the rapidity and magnitude of the ap- programs for many other PPAR agonists have parent hazard was not consistent with an effect been terminated after evidence of toxicity emerged during preclinical studies or initial trials in hu- Several other properties of rosiglitazone may mans. According to a former FDA official, more contribute to adverse cardiovascular outcomes. than 50 Investigational New Drug applications Rosiglitazone and other thiazolidinediones are for novel PPARs have been filed, but no additional known to precipitate congestive heart failure in drugs have successfully reached the market in susceptible patients.26 Congestive heart failure is more than 6 years.29 In some cases, these drugs a physiological state that is associated with an have failed because of evidence of direct myocar- increased intravascular volume. Volume overload dial toxicity in studies in animals,29 but few data increases stress on the left ventricular wall, a on toxicity are available in the public domain factor that determines myocardial oxygen de- because of the common industry practice of not mand. In susceptible patients, an increase in publishing safety findings for failed products.
myocardial oxygen demand could theoretically PPAR agonists such as rosiglitazone have very provoke ischemic events. The administration of complex biologic effects, resulting from the ac- thiazolidinediones, including rosiglitazone, also tivation or suppression of dozens of genes.30 The produces a modest reduction in the hemoglobin patterns of gene activation or suppression differ level.25 In susceptible patients, a reduced hemo- substantially among various PPAR agonists, even globin level may result in increased physiological within closely related compounds. The biologic stress, thereby provoking myocardial ischemia. effects of the protein targets for most of the A study of rosiglitazone that was conducted in genes influenced by PPAR agonists remain large- rats reported an increase in the rate of death after ly unknown. Accordingly, many different and experimentally induced myocardial infarction.27 seemingly unrelated toxic effects have emerged Rosiglitazone is not the first PPAR agonist during development of other PPAR agents.29 that has been reported to increase adverse car- Some drugs have provoked multispecies, multi– diovascular events. Muraglitazar, an investiga- organ system cancers; others have resulted in tional dual PPAR-α and PPAR-γ agonist, increased rhabdomyolysis or nephrotoxicity.29 Troglitazone Downloaded from www.nejm.org on May 22, 2007 . For personal use only. No other uses without permission. Copyright 2007 Massachusetts Medical Society. All rights reserved. Rosiglitazone and Cardiovascular Outcomes was withdrawn from the market for rare, but Table 5. Risk of Myocardial Infarction and Death from Cardiovascular Causes
sometimes fatal, liver toxicity. Accordingly, it for Patients Receiving Rosiglitazone versus Several Comparator Drugs.
must be assumed that a variety of unexpected toxic effects are possible when PPAR agonists are Odds Ratio
Comparator Drug
The question as to whether the observed risks Myocardial infarction
of rosiglitazone represent a “class effect” of thiazolidinediones must also be considered. Pio- glitazone is a related agent also widely used to treat type 2 diabetes mellitus. However, unlike rosiglitazone, pioglitazone has been studied in a prospective, randomized trial of cardiovascular outcomes, called Prospective Pioglitazone Clini- Death from cardiovascular causes
cal Trial in Macrovascular Events (PROACTIVE).31 The primary end point, a broad composite that included coronary and peripheral vascular events, showed a trend toward benefit from pioglita- zone (hazard ratio, 0.90; P = 0.095). A secondary end point consisting of myocardial infarction, stroke, and death from any cause showed a sig- nificant effect favoring pioglitazone (hazard ra- of these trials were small and short-term, re- tio, 0.84; P = 0.027). Notably, pioglitazone ap- sulting in few adverse cardiovascular events or pears to have more favorable effects on lipids, deaths. Accordingly, the confidence intervals for particularly triglycerides, than does rosiglita- the odds ratios for myocardial infarction and death from cardiovascular causes are wide, re- These emerging findings raise an important sulting in considerable uncertainty about the question about the appropriateness of the cur- magnitude of the observed hazard. Furthermore, rent regulatory pathways for the development of we did not have access to original source data drugs to treat diabetes. The FDA considers dem- for any of these trials. Thus, we based the analysis onstration of a sustained reduction in blood on available data from publicly disclosed sum- glucose levels with an acceptable safety profile maries of events. The lack of availability of adequate for approval of antidiabetic agents. source data did not allow the use of more statis- However, the ultimate value of antidiabetic tically powerful time-to-event analysis. A meta- therapy is the reduction of the complications of analysis is always considered less convincing diabetes, not improvement in a laboratory mea- than a large prospective trial designed to assess sure of glycemic control. Although reductions in the outcome of interest. Although such a dedi- blood glucose levels have been shown to reliably cated trial has not been completed for rosiglita- reduce microvascular complications of diabetes, zone, the ongoing Rosiglitazone Evaluated for the effect on macrovascular complications has Cardiac Outcomes and Regulation of Glycaemia proved to be unpredictable.33 After the failure of in Diabetes (RECORD) trial may provide useful muraglitazar and the apparent increase in adverse insights.34 cardiovascular outcomes with rosiglitazone, the Despite these limitations, our data point to use of blood glucose measurements as a surro- the urgent need for comprehensive evaluations to gate end point in regulatory approval must be clarify the cardiovascular risks of rosiglitazone. The manufacturer’s public disclosure of sum- Our study has important limitations. We mary results for rosiglitazone clinical trials is pooled the results of a group of trials that were not sufficient to enable a robust assessment of not originally intended to explore cardiovascular cardiovascular risks. The manufacturer has all outcomes. Most trials did not centrally adjudicate the source data for completed clinical trials and cardiovascular outcomes, and the definitions of should make these data available to an external myocardial infarction were not available. Many academic coordinating center for systematic anal- Downloaded from www.nejm.org on May 22, 2007 . For personal use only. No other uses without permission. Copyright 2007 Massachusetts Medical Society. All rights reserved. T h e n e w e ng l a n d j o u r na l o f m e dic i n e ysis. The FDA also has access to study reports the potential risks of rosiglitazone in the treat- and other clinical-trial data not within the pub- ment of type 2 diabetes.
lic domain. Further analyses of data available to Dr. Nissen reports receiving research support to perform the FDA and the manufacturer would enable a clinical trials through the Cleveland Clinic Cardiovascular Coor- more robust assessment of the risks of this drug. dinating Center from Pfizer, AstraZeneca, Daiichi Sankyo, Roche, Takeda, Sanofi-Aventis, and Eli Lilly. Dr. Nissen consults Our data suggest a cardiovascular risk associated for many pharmaceutical companies but requires them to do- with the use of rosiglitazone. Until more precise nate all honoraria or consulting fees directly to charity so that estimates of the cardiovascular risk of this treat- he receives neither income nor a tax deduction. No other poten- tial conflict of interest relevant to this article was reported.
ment can be delineated in patients with diabetes, We thank Craig Balog for statistical programming support.
patients and providers should carefully consider REFERENCES
1. Vamecq J, Latruffe N. Medical signifi-
1695-702. [Erratum, JAMA 2000;284: treatment in Chinese patients with type 2 cance of peroxisome proliferator-activated 1384.] diabetes and exposure to hepatitis B or C. 11. Weissman P, Goldstein BJ, Rosenstock Diabetes Technol Ther 2003;5:33-42.
2. Campbell IW. The clinical signifi-
J, et al. Effects of rosiglitazone added to 20. The DREAM (Diabetes Reduction As-
cance of PPAR gamma agonism. Curr Mol submaximal doses of metformin com- sessment with ramipril and rosiglitazone pared with dose escalation of metformin Medication) Investigators. Effect of rosig- 3. Center for Drug Evaulation and Re-
in type 2 diabetes: the EMPIRE Study. litazone on the frequency of diabetes in search. Approval package: Avandia (ro- Curr Med Res Opin 2005;21:2029-35.
patients with impaired glucose tolerance siglitazone maleate) tablets. Company: 12. Wolffenbuttel BH, Gomis R, Squatrito or impaired fasting glucose: a randomised
SmithKline Beecham Pharmaceuticals. S, Jones NP, Patwardhan RN. Addition of controlled trial. Lancet 2006;368:1096- Application no. 21-071. Approval date: low-dose rosiglitazone to sulphonylurea 105. [Erratum, Lancet 2006;368:1770.] 5/25/1999. (Accessed May 15, 2007, at therapy improves glycaemic control in 21. Kahn SE, Haffner SM, Heise MA, et
http:/ www.fda.gov/cder/foi/nda/99/21071_ Type 2 diabetic patients. Diabet Med al. Glycemic durability of rosiglitazone, 4. American Diabetes Association. Com-
13. St John Sutton M, Rendell M, Dando-
N Engl J Med 2006;355:2427-43. [Erratum, plications of diabetes in the United States. na P, et al. A comparison of the effects of N Engl J Med 2007;356:1387-8.] (Accessed May 15, 2007, at http://www. rosiglitazone and glyburide on cardiovas- 22. Bradburn MJ, Deeks JJ, Berlin JA, Lo-
cular function and glycemic control in calio AR. Much ado about nothing: a com- patients with type 2 diabetes. Diabetes parison of the performance of meta-ana- 5. GlaxoSmithKline. Rosiglitazone stud-
lytical methods with rare events. Stat Med ies. (Accessed May 15, 2007, at http://ctr. 14. Raskin P, Rendell M, Riddle MC, et al. 2007;26:53-77.
A randomized trial of rosiglitazone thera- 23. Sweeting MJ, Sutton AJ, Lambert PC.
py in patients with inadequately con- What to add to nothing? Use and avoid- 6. Center for Drug Evaluation and Re-
trolled insulin-treated type 2 diabetes. ance of continuity corrections in meta- search. Medical review(s). Application Diabetes Care 2001;24:1226-32.
analysis of sparse data. Stat Med 2004;23: number: 021071. (Accessed May 15, 2007, 15. Vongthavaravat V, Wajchenberg BL, 1351-75. [Erratum, Stat Med 2006;25:2700.]
at http://www.fda.gov/cder/foi/nda/99/ Waitman JN, et al. An international study 24. Sutton A, Cooper N, Lambert P, Jones
of the effects of rosiglitazone plus sul- D, Abrams K, Sweeting M. Meta-analysis 7. Lebovitz HE, Dole JF, Patwardhan R, phonylurea in patients with type 2 diabe-
of rare and adverse event data. Pharmaco- Rappaport EB, Freed MI. Rosiglitazone tes. Curr Med Res Opin 2002;18:456-61.
monotherapy is effective in patients with 16. Baksi A, James RE, Zhou B, Nolan JJ. 25. Avandia (rosiglitazone maleate) tab-
type 2 diabetes. J Clin Endocrinol Metab Comparison of uptitration of gliclazide lets: prescribing information. Research 2001;86:280-8. [Errata, J Clin Endocrinol with the addition of rosiglitazone to gli- Triangle Park, NC: GlaxoSmithKline, clazide in patients with type 2 diabetes 2007 (package insert). (Accessed May 15, 8. Phillips LS, Grunberger G, Miller E, et inadequately controlled on half-maximal 2007, at http://www.fda.gov/cder/foi/label/
al. Once- and twice-daily dosing with doses of a sulphonylurea. Acta Diabetol 2007/021071s023lbl.pdf.) rosiglitazone improves glycemic control 2004;41:63-9.
26. Nesto RW, Bell D, Bonow RO, et al.
in patients with type 2 diabetes. Diabetes 17. Barnett AH, Grant PJ, Hitman GA, et Thiazolidinedione use, fluid retention,
Care 2001;24:308-15. [Erratum, Diabetes al. Rosiglitazone in Type 2 diabetes mel- and congestive heart failure: a consensus litus: an evaluation in British Indo-Asian statement from the American Heart As- 9. Jones TA, Sautter M, Van Gaal LF, patients. Diabet Med 2003;20:387-93.
Jones NP. Addition of rosiglitazone to 18. Kerenyi Z, Samer H, James R, Yan Y, tion: October 7, 2003. Circulation 2003;
metformin is most effective in obese, in- Stewart M. Combination therapy with 108:2941-8.
sulin-resistant patients with type 2 diabe- rosiglitazone and glibenclamide com- 27. Lygate CA, Hulbert K, Monfared M,
tes. Diabetes Obes Metab 2003;5:163-70.
pared with upward titration of gliben- Cole MA, Clarke K, Neubauer S. The 10. Fonseca V, Rosenstock J, Patwardhan clamide alone in patients with type 2 dia-
R, Salzman A. Effect of metformin and betes mellitus. Diabetes Res Clin Pract not alter remodeling but increases mor- rosiglitazone combination therapy in pa- 2004;63:213-23.
tality in rats post-myocardial infarction. tients with type 2 diabetes mellitus: a ran- 19. Zhu XX, Pan CY, Li GW, et al. Addition Cardiovasc Res 2003;58:632-7.
domized controlled trial. JAMA 2000;283: of rosiglitazone to existing sulfonylurea 28. Nissen SE, Wolski K, Topol EJ. Effect
Downloaded from www.nejm.org on May 22, 2007 . For personal use only. No other uses without permission. Copyright 2007 Massachusetts Medical Society. All rights reserved. Rosiglitazone and Cardiovascular Outcomes of muraglitazar on death and major ad- response elements using integrated com- lipidemia. Diabetes Care 2005;28:1547- verse cardiovascular events in patients putational genomics. J Lipid Res 2006; 54.
with type 2 diabetes mellitus. JAMA 2005; 47:1583-7.
33. Riveline JP, Danchin N, Ledru F, Var-
31. Dormandy JA, Charbonnel B, Eckland roud-Vial M, Charpentier G. Sulfonylureas
29. El-Hage J. Peroxisome proliferator- DJ, et al. Secondary prevention of macro-
activated receptor (PPAR) agonists: pre- vascular events in patients with type 2 dia- mental data to clinical use: available data clinical and clinical cardiac safety consid- betes in the PROactive Study (PROspective in humans and clinical applications. Dia- erations. Rockville, MD: Center for Drug pioglitAzone Clinical Trial In macroVas- betes Metab 2003;29:207-22.
Evaluation and Research, 2006. (Accessed cular Events): a randomised controlled 34. Home PD, Pocock SJ, Beck-Nielsen H,
May 15, 2007, at http://www.fda.gov/cder/ trial. Lancet 2005;366:1279-89.
et al. Rosiglitazone Evaluated for Cardiac present/DIA2006/El-Hage_CardiacSafety. 32. Goldberg RB, Kendall DM, Deeg MA, Outcomes and Regulation of Glycaemia
et al. A comparison of lipid and glycemic in Diabetes (RECORD): study design and 30. Lemay DG, Hwang DH. Genome-wide effects of pioglitazone and rosiglitazone protocol. Diabetologia 2005;48:1726-35.
identification of peroxisome proliferator in patients with type 2 diabetes and dys- Copyright 2007 Massachusetts Medical Society Downloaded from www.nejm.org on May 22, 2007 . For personal use only. No other uses without permission. Copyright 2007 Massachusetts Medical Society. All rights reserved.

Source: http://www.avandia-heart-lawyers.com/pdf/NEJM_avandia.pdf