复旦大学：《循证医学》课程教学资源（参考教材）Unit 09 药物不良反应的分析评价（主讲教师：朱畴文）.pdf_大学文库

Unit9:药物不良反应的分析评价主讲教师：朱畴文助理教师：高虹一、教学目的：掌握和熟悉药物不良反应及其研究、评价和应用的原则及方法二、教学内容： 1了解药物不息反应研究（药物流行病学）的历中和讲展学握药物不良反应的定义重要意义和有关监测研究的主要基本方法 3.初步掌握药物不良反应的确定方法（宏观和微观评价】 4.熟悉流行病学中各种相关性或因果研究方法的特点（带复习性质） 5.了解药物不良反应监测、药物流行病学的应用及前景三、教学重点：药物不良反应的研究方法四、教学难点：药物不良反应的判断五、中文和英文关键词 Adverse Drug Reaction药物不良反应 Pharmacoepidemiology药物流行病学 Causal As ciation因果关联六、阅读文献： 1.Nissen SE,Wolski K.Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes.N Engl J Med 2007:356(24):2457-71. 2.罗格列酮大事记七、讨论思考题一)微观判断：使用Naranjo模式和我国卫生部ADR中心评分法，推断如下临床病例发生药物不良反应的可能性一位29岁的女性在过去一周已经接受nitrofurantoin(呋喃妥因)治疗2 次。在地服下第3次治疗的第1片药的4小小时后。她出现发热(39℃)、呼吸窘迫和呕吐。她随即停了药天后她的全科医师开了胸片检查发现为过敏性肺炎”的表现。她的白细胞精增多(12*10^9/L)伴9%嗜酸性细胞。数日后，患者完全康复。 Naranjo's Adverse Drug Reaction Probability Scale 间题是否不知道分值 1 .以前有类似的报道吗？不良事件是在应用可疑药物之后出现的吗？ 2 0 3 当撒药后或应用特定的对抗药后不良反应有所 +1 0 0 好转吗？ 4 当再次用药后，不良反应又出现吗？ +2 、 0 有其他非药物因素可引起该不良反应吗使用安慰剂后，不良反应再次出现了吗 1 0 -1 7 药物血（或其他体液）浓度达到中毒浓度了吗？ + 0 0 8 增加（或减少）药物剂量，不良反应加重（或减轻）+1 0 0 了吗？

Unit 9：药物不良反应的分析评价主讲教师：朱畴文助理教师：高虹一、教学目的：掌握和熟悉药物不良反应及其研究、评价和应用的原则及方法二、教学内容： 1. 了解药物不良反应研究（药物流行病学）的历史和进展 2. 掌握药物不良反应的定义、重要意义和有关监测研究的主要基本方法 3. 初步掌握药物不良反应的确定方法（宏观和微观评价） 4. 熟悉流行病学中各种相关性或因果研究方法的特点（带复习性质） 5. 了解药物不良反应监测、药物流行病学的应用及前景三、教学重点：药物不良反应的研究方法四、教学难点：药物不良反应的判断五、中文和英文关键词 Adverse Drug Reaction 药物不良反应 Pharmacoepidemiology 药物流行病学 Causal Association 因果关联六、阅读文献： 1. Nissen SE, Wolski K. Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes. N Engl J Med 2007;356(24):2457-71. 2. 罗格列酮大事记七、讨论思考题：（一）微观判断：使用 Naranjo 模式和我国卫生部 ADR 中心评分法，推断如下临床病例发生药物不良反应的可能性一位 29 岁的女性在过去一周已经接受 nitrofurantoin（呋喃妥因）治疗 2 次。在她服下第 3 次治疗的第 1 片药的 4 小时后，她出现发热（39oC）、呼吸窘迫和呕吐。她随即停了药，一天后她的全科医师开了胸片检查，发现为“过敏性肺炎”的表现。她的白细胞稍增多（12*10^9/L）伴 9%嗜酸性细胞。数日后，患者完全康复。 Naranjo’s Adverse Drug Reaction Probability Scale 问题是否不知道分值 1 .以前有类似的报道吗? +1 0 0 2 不良事件是在应用可疑药物之后出现的吗？ +2 -1 0 3 当撤药后或应用特定的对抗药后不良反应有所好转吗？ +1 0 0 4 当再次用药后，不良反应又出现吗? +2 -1 0 5 有其他非药物因素可引起该不良反应吗? -1 +2 0 6 使用安慰剂后,不良反应再次出现了吗? -1 +1 0 7 药物血(或其他体液)浓度达到中毒浓度了吗？ +1 0 0 8 增加(或减少)药物剂量，不良反应加重(或减轻) 了吗? +1 0 0

n engl j med 356;24 www.nejm.org june 14, 2007 2457 The new england journal of medicine established in 1812 june 14, 2007 vol. 356 no. 24 Effect of Rosiglitazone on the Risk of Myocardial Infarction and Death from Cardiovascular Causes Steven E. Nissen, M.D., and Kathy Wolski, M.P.H. A bs tr ac t From the Cleveland Clinic, Cleveland. Address reprint requests to Dr. Nissen at the Department of Cardiovascular Medicine, Cleveland Clinic, 9500 Euclid Ave., Cleveland, OH 44195, or at nissens@ccf. org. This article (10.1056/NEJMoa072761) was published at www.nejm.org on May 21, 2007. N Engl J Med 2007;356:2457-71. Copyright © 2007 Massachusetts Medical Society. Background Rosiglitazone is widely used to treat patients with type 2 diabetes mellitus, but its effect on cardiovascular morbidity and mortality has not been determined. Methods 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 manufacturer (GlaxoSmithKline). Criteria for inclusion in our meta-analysis included a study duration of more than 24 weeks, the use of a randomized control group not receiving rosiglitazone, and the availability of outcome data for myocardial infarction and death from cardiovascular causes. Of 116 potentially relevant studies, 42 trials met the inclusion criteria. We tabulated all occurrences of myocardial infarction 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 limitations, patients and providers should consider the potential for serious adverse cardiovascular effects of treatment with rosiglitazone for type 2 diabetes. The New England Journal of Medicine Downloaded from nejm.org on February 5, 2016. For personal use only. No other uses without permission. Copyright © 2007 Massachusetts Medical Society. All rights reserved

The NEW ENGLAND JOURNAL f MEDICINE AHIAZOLIDINEDIONE DRUGS ARE WIDE any mvocardial infarctions or deaths from cardic ly used to lower hlood glucose levels in na and therefo tients with type 2 diabetes mellitus.In the in the analysis because the effect measure could United States,three such agents have been intro- not be calculated.Of the remaining 42 studies, 38 reported at least one myocardi: and 23 reported at le st one ed e ra comparator groups with regimens that did not receptors are ligand-activated nuclear transcrip- include rosiglitazone. tion factors that modulate gene expression,lower Multiple groups of patients who received rosig ing blood glucose primarily by increasing insulin when appli cable.The control g roup wa wide used as pa rec ing any d falt reg ries.One group includes five of the stud. or glimepiride (Avandaryl,GlaxoSmithKline). ies submitted to the FDA for the March 22,1999. The original approval of rosiglitazone was advisory board hearing that recommended ap- based on the ability of the drug to reduce blo proval of rosiglitazone.Group-level data from glucose and glycat d hem oglobin levels s are availab stu cd ve stud cun ite s D rom the rted in fashime trials cluding cardiovascular morbidity and mortality. However,the effect of any antidiabetic therapy on cardiovascular outcomes is particularly im- of these five trials were also published in peer reviewed journals.In these e trials,1967 es are tients we or active com rators to ass sess the effect of this Other studies that meta agent on cardiovascular outcomes.The sourc analysis were initially identified in the glaxo material for this analysis consisted of publicly SmithKline clinical-trial registry.As noted in available data from the original registration pack Table 1,we included 35 studies in this category age submitted to the F od and Drug Administr tion (FDA),a er serie of trial 26 ssible.the d on al indications for the drug. cthhepebicionnsoFantheake between published and unpublished data,data METHODS derived from the manufacturer's web site were used.In this group of 35 trials,9507 7 patients sts the 42 trials d11 48 trials me defined inclusion criteria of having a randomized comparator group,a similar duration of treat Assessment with Ramipiril and Rosiglitazone ment in all groups,and more than 24 weeks of Medication (DREAM)NCT00095654 trial20 and drug exposure.Six of the 48 trials did not report the A Diabetes Outcome Prevention Trial (ADOPT) 2458 N ENGLJ MED 356:24 WWW.NEJM.ORG JUNE 14.2007

T h e n e w e ng l a nd j o u r na l o f m e dic i n e 2458 n engl j med 356;24 www.nejm.org june 14, 2007 Thiazolidinedione drugs are widely used to lower blood glucose levels in patients with type 2 diabetes mellitus. In the United States, three such agents have been introduced: troglitazone, which was removed from the market because of hepatotoxicity, and two currently available agents, rosiglitazone (Avandia, GlaxoSmithKline) and pioglitazone (Actos, Takeda). The thiazolidinediones are agonists for peroxisomeproliferator–activated receptor γ (PPAR-γ). PPAR-γ receptors are ligand-activated nuclear transcription factors that modulate gene expression, lowering blood glucose primarily by increasing insulin sensitivity in peripheral tissues.1,2 Rosiglitazone was introduced in 1999 and is widely used as monotherapy or in fixed-dose combinations with either metformin (Avandamet, GlaxoSmithKline) or glimepiride (Avandaryl, GlaxoSmithKline). The original approval of rosiglitazone was based on the ability of the drug to reduce blood glucose and glycated hemoglobin levels.3 Initial studies were not adequately powered to determine the effects of this agent on microvascular or macrovascular complications of diabetes, including cardiovascular morbidity and mortality.3 However, the effect of any antidiabetic therapy on cardiovascular outcomes is particularly important, because more than 65% of deaths in patients with diabetes are from cardiovascular causes.4 Therefore, we performed a meta-analysis of trials comparing rosiglitazone with placebo or active comparators to assess the effect of this agent on cardiovascular outcomes. The source material for this analysis consisted of publicly available data from the original registration package submitted to the Food and Drug Administration (FDA), another series of trials performed by the sponsor after approval, and two large, prospective, randomized trials designed to study additional indications for the drug. Me thods Analyzed Studies Table 1 lists the 42 trials included in this metaanalysis. We screened 116 phase 2, 3, and 4 trials for inclusion. Of these, 48 trials met the predefined inclusion criteria of having a randomized comparator group, a similar duration of treatment in all groups, and more than 24 weeks of drug exposure. Six of the 48 trials did not report any myocardial infarctions or deaths from cardiovascular causes and therefore were not included in the analysis because the effect measure could not be calculated. Of the remaining 42 studies, 38 reported at least one myocardial infarction, and 23 reported at least one death from cardiovascular causes. In these trials, 15,565 patients were randomly assigned to regimens that included rosiglitazone, and 12,282 were assigned to comparator groups with regimens that did not include rosiglitazone. Multiple groups of patients who received rosiglitazone within a single trial were pooled together, when applicable. The control group was defined as patients receiving any drug regimen other than rosiglitazone. The trials fall into three categories. One group includes five of the studies submitted to the FDA for the March 22, 1999, advisory board hearing that recommended approval of rosiglitazone. Group-level data from these five studies are available in publicly disclosed briefing documents archived on the FDA Web site.6 Data from these same trials are also reported in a summary fashion on a clinicaltrial registry Web site maintained by the drug manufacturer, GlaxoSmithKline.5 Reports of four of these five trials were also published in peerreviewed journals.7-9 In these five trials, 1967 patients were randomly assigned to receive rosiglitazone, and 793 patients were assigned to receive various comparator drugs (Table 1). Other studies that we included in the metaanalysis were initially identified in the GlaxoSmithKline clinical-trial registry.5 As noted in Table 1, we included 35 studies in this category, 9 of which were published in peer-reviewed journals and 26 of which remain unpublished.10-18 Whenever possible, the results obtained on the GlaxoSmithKline Web site were cross-checked 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, 9507 patients were randomly assigned to receive rosiglitazone, and 5960 patients were assigned to receive various comparator drugs. A third data source consisted of two large, recently published trials, the Diabetes Reduction Assessment with Ramipiril and Rosiglitazone Medication (DREAM) NCT00095654 trial20 and the A Diabetes Outcome Prevention Trial (ADOPT) The New England Journal of Medicine Downloaded from nejm.org on February 5, 2016. For personal use only. No other uses without permission. Copyright © 2007 Massachusetts Medical Society. All rights reserved

ROSIGLITAZONE AND CARDIOVASCULAR OUTCOMES (ClinicalTrials.gov number,NCT00279045).In ity across trials,allowing for the use of a fixed- the DREAM study,2635 patients were randomly effects model.For additional analyses,the active assigned to receive rosiglitazone and 2634 patients ocp2latorecontolgroupswcrcesnbegroupedintm ollowing tour classes for comparison wit terosigl and pl oup with the In the ADOPT trial,1456 patients were random-use of methods similar to those used in the ly assigned to receive rosiglitazone and 2895 pa-pooled analyses.Data were analyzed with the use tients were assigned to receive either metformin of Comprehensive Meta-Analysis software,version or glyburide 2.2 (Biostat). RESULTS 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 levels or fasting b d gluco levels for patien nen led in the revyog vrg causes.With the exc DRAM su Overall thereu 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- proximately 8.2%for both study groups. TION AND DEATH able.it was no 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 were86 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 separately in to STATISTICAL ANALYEIS intervals.and P values for m cardial 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. The summary odds ratio for myocardial infard tion was 1.43 in the P=0.03 odd als in which patients had noadvers with the contro w2si6495%CL,0.98t0 cular events in either group were excluded from 2.74:P=0.06).Table 4also lists odds ratios and analyses.All reported P values are two-sided 95%confidence intervals for the pooled group of Statistical het of Cochran's Q statistic or the odds ratios for myocardial in N ENGL J MED 356;24 WWW.NEJM.ORG JUNE 14,2007 2459 The New England Journal of Medicine

Rosiglitazone and Cardiovascular Outcomes n engl j med 356;24 www.nejm.org june 14, 2007 2459 (ClinicalTrials.gov number, NCT00279045).21 In the DREAM study, 2635 patients were randomly assigned to receive rosiglitazone and 2634 patients were assigned to receive placebo. The DREAM study was designed to determine whether rosiglitazone could prevent the development of type 2 diabetes in patients at high risk for this disorder. In the ADOPT trial, 1456 patients were randomly assigned to receive rosiglitazone and 2895 patients were assigned to receive either metformin or glyburide. The ADOPT study was designed to assess the durability of glycemic control with rosiglitazone therapy, as compared with therapy with metformin or glyburide. Outcome Measures We reviewed data summaries provided in the FDA review documents, the GlaxoSmithKline clinical-trial registry Web site, and published trial results and then abstracted from the adverse-event tabulations information on myocardial infarction and death from cardiovascular causes. With the exception of the DREAM study, the included trials did not describe adjudication of myocardial infarction or death from cardiovascular causes. Time-to-event data for cardiovascular events were not available in any of these trials, which precluded the calculation of hazard ratios. Because only summary data were available, it was not possible to discern whether the same patient had both events. Therefore, an outcome measure based on the composite of death or myocardial infarction could not be constructed. Accordingly, these two outcomes are reported separately. Statistical Analysis Many trials had few cardiovascular events, so the odds ratios and 95% confidence intervals were calculated with the use of the Peto method.22-24 Because all trials had similar durations of followup for all treatment groups, the use of odds ratios represents a valid approach to assessing the risk associated with the use of rosiglitazone. Trials in which patients had no adverse cardiovascular events in either group were excluded from analyses. All reported P values are two-sided. Statistical heterogeneity across the various trials was tested with the use of Cochran’s Q statistic. A P value of more than the nominal level of 0.10 for the Q statistic indicated a lack of heterogeneity across trials, allowing for the use of a fixedeffects model. For additional analyses, the active comparator control groups were subgrouped into the following four classes for comparison with rosiglitazone: metformin, sulfonylurea, insulin, and placebo. Odds ratios and 95% confidence intervals were calculated for each subgroup with the use of methods similar to those used in the pooled analyses. Data were analyzed with the use of Comprehensive Meta-Analysis software, version 2.2 (Biostat). R esult s Baseline Characteristics Table 2 reports the doses of rosiglitazone and comparator drugs, baseline demographic characteristics, study periods, and glycated hemoglobin levels or fasting blood glucose levels for patients enrolled in the trials. The patients were relatively young, averaging less than 57 years of age for both the rosiglitazone group and the control group. Overall, there was a moderate predominance of men. Diabetes control was relatively poor, with a mean baseline glycated hemoglobin level of approximately 8.2% for both study groups. Myocardial Infarction and Death Table 3 reports the myocardial infarction events and deaths from cardiovascular causes that were reported in the 42 clinical trials we reviewed. There were 86 myocardial infarctions in the rosiglitazone group and 72 in the control group. There were 39 deaths from cardiovascular causes in the rosiglitazone group and 22 in the control group. Table 4 lists the odds ratios, 95% confidence intervals, and P values for myocardial infarction and death from cardiovascular causes for the rosiglitazone group and the control group. The summary odds ratio for myocardial infarction was 1.43 in the rosiglitazone group (95% confidence interval [CI], 1.03 to 1.98; P=0.03). The odds ratio for death from cardiovascular causes in the rosiglitazone group, as compared with the control group, was 1.64 (95% CI, 0.98 to 2.74; P=0.06). Table 4 also lists odds ratios and 95% confidence intervals for the pooled group of trials that were smaller and of shorter duration; results for the DREAM and ADOPT studies are shown separately. Table 5 lists odds ratios for myocardial inThe New England Journal of Medicine Downloaded from nejm.org on February 5, 2016. For personal use only. No other uses without permission. Copyright © 2007 Massachusetts Medical Society. All rights reserved

Rosiglitazone and Cardiovascular Outcomes n engl j med 356;24 www.nejm.org june 14 , 2007 2461 49653/1255,15 3 26 Rosiglitazone and sulfonylurea§ 175 Sulfonylurea§ 173 49653/1275 3 26 Rosiglitazone and glyburide 56 Glyburide 58 49653/1285 3 28 Rosiglitazone 39 Placebo 38 49653/1345 3 28 Rosiglitazone 561 Placebo 276 49653/1355 3 104 Rosiglitazone and glipizide 116 Glipizide 111 49653/1365 3 26 Rosiglitazone 148 Placebo 143 49653/1455,16 3 26 Rosiglitazone and gliclazide 231 Gliclazide 242 49653/1475,17 3 26 Rosiglitazone and sufonylurea¶ 89 Sulfonylurea¶ 88 49653/1625,18 3 26 Rosiglitazone and glyburide 168 Glyburide 172 49653/2345 3 26 Rosiglitazone and glimepiride 116 Glimepiride 61 49653/3305 3 52 Rosiglitazone 1,181 Placebo 382 49653/3315 3 52 Rosiglitazone 706 Placebo 325 49653/1375 3 32 Rosiglitazone and metformin 204 Glyburide and metformin 185 SB-712753/0025 3 24 Rosiglitazone and metformin 288 Metformin 280 SB-712753/0035 3 32 Rosiglitazone and metformin 254 Metformin 272 SB-712753/0075 3 32 Rosiglitazone with or without metformin 314 Metformin 154 SB-712753/0095 3 24 Rosiglitazone, metformin, and insulin 162 Insulin 160 49653/1325,19 2 24 Rosiglitazone and sulfonylurea‖ 442 Sulfonylurea‖ 112 AVA1001935 2 24 Rosiglitazone 394 Placebo 124 Subtotal 9,507 5,960 Recently published large, prospective, randomized trials DREAM20 NCT00095654 3 156 Rosiglitazone 2,635 Placebo 2,634 ADOPT21 NCT00279045 3 208 Rosiglitazone 1,456 Metformin or glyburide 2,895 Total 15,565 12,282 * Studies are listed according to the number designated by the sponsor, GlaxoSmithKline, and are available on the company’s Web site.5 ClinicalTrials.gov numbers are listed for trials included in that registry. DREAM denotes Diabetes Reduction Assessment with Ramipiril and Rosiglitazone Medication, and ADOPT A Diabetes Outcome Prevention Trial. † The administered drug was either glyburide or gliclazide. ‡ The administered drug was glyburide, gliclazide, or glipizide. § The administered drug was glyburide, glipizide, gliclazide, chlorpropamide, glimepiride, or tolbutamide. ¶The type of sulfonylurea was unspecified. ‖ The administered drug was glyburide, glipizide, gliclazide, chlorpropamide, gliquidone, or tolbutamide. The New England Journal of Medicine Downloaded from nejm.org on February 5, 2016. 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 nd j o u r na l o f m e dic i n e 2462 n engl j med 356;24 www.nejm.org june 14 , 2007 Table 2. Doses, Baseline Demographic Characteristics, Study Periods, and Glycated Hemoglobin Levels.* Study Drug Daily Dose Population Study Period Age Male Sex Race† Baseline Glycated Hemoglobin Level yr percent 100684 Rsg/Gly 4 or 8 mg Korean patients with type 2 DM Dec. 2003–July 2005 55.2 53.5 100 A NA Gly 5–15 mg 54.5 45.6 100 A NA 49653/143 Rsg/Gly 8 mg Type 2 DM poorly controlled on glyburide July 2000– Jan. 2003 52 45.3 44:56 B:H 9.2 Gly Usual care 53 48.3 38:62 B:H 9.4 49653/211 Rsg 4 mg Type 2 DM with CHF July 2001–Nov. 2003 64.3 84.3 99 7.7 Plc — 63.9 79.0 99 7.8 49653/284 Rsg/Met 4 or 8 mg/1 g Type 2 DM June 2001–Feb. 2003 55.5 51.1 72 8.1 Met 1–2 g 55.6 51.0 71 7.9 712753/008 Rsg/Met 8 mg/1 g Type 2 DM poorly controlled on Met June 2003–Dec. 2005 54.6 63.2 70 NA Rsg/Met 4 mg/2 g 56.0 65.2 78 NA Met 2g 56.9 53.4 69 NA AVM100264 Rsg/Met 4 or 8 mg/2 g Overweight patients with type 2 DM poorly controlled on Met July 2004–Jan. 2006 58.5 52.7 94 8.0 Met/Su 2 g/titrated 59.3 52.5 95 8.0 BRL49653C/185 Rsg/ELM/Met 4 mg/1.5 g Type 2 DM May 2000–May 2002 58.0 65.2 76 7.5 Rsg/ELM 4 mg 59.0 60.2 78 7.4 Met/ELM 1.5 g 60.0 56.4 78 7.5 ELM — 57.0 60.9 83 7.4 BRL 49653/334 Rsg 4 or 8 mg Type 2 DM or insulin resistance syndrome March 2002–Nov. 2004 67.7 44.8 99 6.3 Plc — 67.3 47.7 100 6.3 BRL 49653/347 Rsg/insulin 4 mg Type 2 DM poorly controlled on insulin Nov. 2002–April 2004 52.6 48.1 57 9.0 Rsg/insulin 2 or 4 mg 52.7 60.0 57 8.9 Insulin/Plc Usual care 53.8 46.2 57 9.1 49653/011 Rsg 8 mg Type 2 DM Sept. 1996–Sept. 1997 60.7 66.9 73 8.8 Rsg 4 mg 59.6 64.5 75 9.0 Plc — 58.8 65.8 74 9.0 49653/015 Rsg/Su 4 mg Type 2 DM Aug. 1996–March 1998 60.6 53.2 98 9.2 Rsg/Su 2 mg 61.0 62.8 86 9.2 Su 61.9 57.3 97 9.2 The New England Journal of Medicine Downloaded from nejm.org on February 5, 2016. For personal use only. No other uses without permission. Copyright © 2007 Massachusetts Medical Society. All rights reserved

Rosiglitazone and Cardiovascular Outcomes n engl j med 356;24 www.nejm.org june 14, 2007 2463 49653/020 Rsg 8 mg Type 2 DM Oct. 1996–May 1998 60.9 57.6 97 8.2 Rsg 4 mg 60.4 68.2 99 8.1 Gly Titrated 60.1 70.4 99 8.2 49653/024 Rsg 4 mg once daily Type 2 DM Jan. 1997–Feb. 1998 57.5 58.6 76 8.9 Rsg 2 mg twice daily 56.8 59.1 78 8.9 Rsg 8 mg once daily 58.9 65.7 80 8.9 Rsg 4 mg twice daily 56.5 59.9 71 9.0 Plc — 57.7 68.8 79 8.9 49653/079 Rsg 4 mg Type 2 DM poorly controlled on maximum dose of Gly April 1997–March 1998 59.1 63.6 70 9.1 Rsg/Gly 4 mg/20 mg 57.7 69.4 70 9.2 Gly 20 mg 58.5 66.7 69 9.3 49653/080 Rsg 8 mg Type 2 DM Nov. 1996–May 2000 55.1 75.0 73 8.9 Gly 2.5–5.0 mg 56.1 70.1 76 9.4 49653/082 Rsg/insulin 8 mg Type 2 DM poorly controlled on insulin July 1997–Aug. 1998 57.7 54.3 71 9.0 Rsg/insulin 4 mg 57.1 56.6 72 9.1 Insulin Usual care 55.6 55.8 68 8.9 49653/085 Rsg/insulin 4 or 8 mg Type 2 DM May 2000–June 2001 61.3 54.0 99 NA Insulin Usual care 61.5 46.8 100 NA 49653/093 Rsg/Met 8 mg/2.5 g Type 2 DM poorly controlled on Met June 1997–April 1998 57.8 60.0 58 8.7 Rsg 8 mg 58.8 53.7 59 8.7 Met 2.5 g 59.5 67.0 60 8.8 49653/094 Rsg/Met 8 mg/2.5 g Type 2 DM poorly controlled on Met April 1997–March 1998 58.3 68.2 77 8.9 Rsg/Met 4 mg/2.5 g 57.5 62.1 80 8.9 Met 2.5 g 58.8 74.3 81 8.6 49653/095 Rsg/insulin 8 mg Type 2 DM poorly controlled on insulin Aug. 1997–Dec. 1998 57.4 58.9 73 9.1 Rsg/insulin 4 mg 57.8 63.9 68 8.8 Insulin Usual care 58.9 45.3 73 9.1 49653/097 Rsg 8 mg Type 2 DM Aug. 1997–Jan. 2001 55.8 72.1 74 8.9 Gly Titrated 56.0 70.8 84 8.8 49653/125 Rsg/Su 4 mg Type 2 DM May 1999–Aug. 2000 54.6 45.7 56 A 9.1 Su Usual care 57.3 42.4 59 A 8.9 49653/127 Rsg/Gly 8 mg/<20 mg Type 2 DM poorly controlled on Gly Jan. 1999–Dec. 1999 60.0 51.0 75 9.1 Gly <20 mg 59.4 66.0 75 8.9 The New England Journal of Medicine Downloaded from nejm.org on February 5, 2016. For personal use only. No other uses without permission. Copyright © 2007 Massachusetts Medical Society. All rights reserved