Efficacy of Diuretics and beta-Blockers in Diabetic Hypertensive Patients

Volume 23 Supplement 2
Data, Results, and Consequences of Major Trials With Focus on Type 2 Diabetes
Proceedings from a Symposium

ORIGINAL ARTICLE

Efficacy of Diuretics and -Blockers in Diabetic Hypertensive Patients

Michel Lièvre, PHD
François Gueyffier, MD, PHD
Tord Ekbom, MD
Robert Fagard, MD
Jeffrey Cutler, MD

Eleanor Schron, RN, MS
Michel Marre, MD
Jean-Pierre Boissel, MD
INDANA Steering Committee

OBJECTIVE — To review the effectiveness of diuretic or -blocker–based treatment of hypertension in diabetic patients.

RESEARCH DESIGN AND METHODS — A meta-analysis on individual patient data was performed on four trials of the treatment of hypertension in which diabetic patients were included and treated with first-line diuretics or -blockers. The main outcomes were the relative risk of death, fatal or nonfatal stroke, fatal or nonfatal coronary events, and major cardiovascular events.

RESULTS — There were 92 diabetic patients who received first-line -blockers and 1,008 who received diuretics. In the control groups, diabetic patients had nearly twice the risk of any outcome when compared with nondiabetic patients. The same blood pressure reduction was achieved under treatment in the diabetic and nondiabetic patients, except for systolic pressure, which decreased more in the nondiabetic patients at 1 year. In the 15,843 nondiabetic patients, the risk of all four outcomes was reduced significantly in the treated group. In the 2,254 diabetic patients, the risk reduction was significant only for fatal and nonfatal stroke (36%, P = 0.011) and major cardiovascular events (20%, P = 0.032), but not for death (5%, P = 0.65) and fatal or nonfatal coronary events (15%, P = 0.23). However, no heterogeneity was detected between diabetic patients and nondiabetic patients for any outcome. The numbers of outcomes avoided for 1,000 patients treated for 5 years were higher in diabetic patients (e.g., 38 major cardiovascular events) than with nondiabetic patients (e.g., 28 major cardiovascular events).

CONCLUSIONS — These results show that hypertensive diabetic patients benefit from first-line treatment with diuretics. No conclusion can be drawn for -blockers, owing to the small sample size.

Diabetes Care 23 (Suppl. 2):B65–B71, 2000

To obtain a precise estimation of the benefit for diabetic patients of antihypertensive therapy based on diuretics or -blockers, in terms of cardiovascular risk, a meta-analysis of individual data from the controlled clinical trials in which diabetic patients were included was performed.

Although epidemiological studies show that hypertension is very frequent in diabetic patients (1–4), most of the randomized clinical trials that have assessed the effects of antihypertensive treatment on cardiovascular risk have excluded diabetic patients (5–10) or did not report their inclusion (11–14). No such placebo-controlled trial focusing on diabetic patients has been undertaken. A possible reason for their exclusion is that until recently, trials in hypertensive patients assessed treatment strategies mainly based on diuretics and -blockers that were susceptible to impair control of blood glucose and lipid levels (15). In addition, two observational studies found an increased risk of death in hypertensive patients treated with diuretics (16,17).

However, diabetic patients have been included in some trials. Subgroup findings from the recently published Systolic Hypertension in the Elderly Program (SHEP) (18) show a 34% reduction of major cardiovascular disease compared with placebo in diabetic patients over 60 years of age treated with a diuretic-based strategy. In most reports of trials that included diabetic patients, subgroup findings are not available.

This work is part of the Individual Data Analysis of Antihypertensive Drug Interventions (INDANA) project (19).

RESEARCH DESIGN AND METHODS — The INDANA project is a collaboration of representatives from most (presently 11) of the large randomized controlled trials comparing systematic antihypertensive drug treatment with placebo, no treatment, or a nonsystematic treatment of hypertension. Its results are derived from centralized individual files of the baseline and follow-up data available for all 53,799 patients enrolled in the trials (19).

Trials
All publications of randomized controlled clinical trials of antihypertensive treatments based on diuretics and/or -blockers that displayed results on clinical outcomes and were included in the meta-analysis of Collins et al. (20) or published after this meta-analysis were screened for the mention of diabetic patients. This first search was done to assess the completeness of the INDANA database regarding studies having included diabetic patients. In addition, summarized data (numbers of patients and events) were extracted from the INDANA database by diabetic status and by trial according to the intention-to-treat principle. Diabetes status was determined from the presence or absence of history of diabetes, as reported in each trial, or a baseline fasting blood glucose >126 mg/dl (7.0 mmol/l), according to the most recent criteria for the diagnosis of diabetes (21).

Outcomes
Four outcomes were initially considered for the meta-analysis: 1) death from any cause; 2) fatal or nonfatal stroke, excluding transient ischemic attack; 3) fatal or nonfatal coronary event, including sudden death, which was defined as unexpected and unexplained death occurring within 24 h after symptom onset; and 4) major cardiovascular events, which combines the second and third outcomes and cardiovascular-related mortality, including death from thromboembolism. Cardiovascular death and noncardiovascular death were subsequently added in an explanatory approach.

Statistics
To estimate the relationship between diabetes and the incidence of each of the study outcomes, a logistic regression analysis was performed on the pooled control groups of the trials of the INDANA database having admitted diabetic patients. The results are given as the odds ratio of each outcome in diabetic patients compared with nondiabetic patients, adjusted for age, sex, current cigarette smoking (yes or no), baseline systolic blood pressure (sBP), BMI, history of stroke, history of myocardial infarction, total cholesterol, creatinine clearance (estimated by Cockroft's formula), uricemia, and heart rate.

Blood pressure decrease from baseline was calculated from individual data in the INDANA database.

The relative risk of each outcome in the treated group compared with the control group was estimated according to the logarithm of the relative risk method (22). Relative risks are presented with their 95% CI. The Cochran homogeneity ² statistic enabled estimation of the degree of homogeneity between individual trial results on one hand and between diabetic patients versus nondiabetic patients on the other hand (22). A low homogeneity P value (<0.1) denotes a lack of homogeneity and makes the interpretation of the combined result difficult (23). In addition, we performed a sensitivity analysis by excluding from the meta-analysis one trial that differed by its design from the other trials included in the main meta-analysis.

For each outcome, the number of events avoided for 1,000 patients treated for 5 years was calculated as follows: (number of outcomes in the control group/patients X years of follow-up) X (1 – RR) X 5,000, where RR is the common relative risk for the parameter.

Data management and logistic regression were done using SAS software (SAS Institute, Cary, NC). Meta-analysis computations were performed using Easy-MA software (Clinical Pharmacology Department, Lyon, France. (24)

RESULTS

Main analysis
Four trials had diabetic patients in the INDANA database: the Hypertension Detection and Follow-up Program (HDFP) (25), the European Working Party on High Blood Pressure in the Elderly (EWPHE) trial (26), the Swedish Trial in Old Patients with Hypertension (STOP-H) (27), and SHEP (28). EWPHE trial data were used only for the analysis of mortality, the reporting of nonfatal events being affected by a censoring bias in this study. There were 46 patients with an unknown diabetic status: 13 from the STOP trial, and 33 from the SHEP trial. These patients were not included in the meta-analysis. We identified three other trials not represented in the INDANA database but which included diabetic patients: 1) the Veterans Administration Cooperative Study in patients with diastolic blood pressures (dBPs) averaging 115–129 mmHg, in which 19 of 143 patients were diabetic patients (29); 2) the study of Wolff and Lindeman (30), in which 14 of 87 patients had diabetes; and 3) the Hypertension-Stroke Cooperative Study Group trial (31), in which 162 of 452 patients were diabetic patients. Subgroup findings for diabetic patients were not provided in the reports of these trials, except for stroke in the Hypertension-Stroke Cooperative Study Group trial (31). In other trial reports, the presence of diabetic patients was either excluded by protocol or not mentioned. Thus, the meta-analysis was performed on four trials for mortality (HDFP, EWPHE, STOP-H, and SHEP: 18,097 patients, 2,254 diabetic patients, and 15,843 nondiabetic patients) and on three trials for other outcomes (HDFP, STOP-H, and SHEP: 17,257 patients, 2,162 diabetic patients, and 15,095 nondiabetic patients). The principal characteristics of the trials included in the meta-analysis are shown in Table 1. In all of these trials, diuretics were used as the only first-line treatment (1,008 patients), except in STOP-H, in which 92 patients received a -blocker and 42 a diuretic.

Figure 1—Meta-analysis of studies of antihypertensive drugs in which diabetic patients were included: fatal or nonfatal stroke. Relative risks, 95% CIs, and sample sizes (Sizes) for each trial subgroup, each subgroup (diabetic and nondiabetic), and total are shown. For each subgroup, mean relative risk and P value are indicated. Cochran Q het, P value for heterogeneity test; D+, diabetic patients; ND, nondiabetic patients.

In these four trials, diabetic patients compared with nondiabetic patients were older by 2.2 years (P < 0.001); had a higher sBP (P = 0.002), slightly lower dBP (P = 0.042), and slightly higher heart rate (P = 0.026) and total cholesterol (P = 0.01); had similar creatinine clearance; and had higher BMI and blood glucose (P < 0.001) (Table 2). Baseline characteristics were very similar in treated and untreated patients, irrespective of diabetic status (Table 2).

In the control groups, the risk of outcome in diabetic patients was nearly twice that in nondiabetic patients, except for the risk of death from any cause, which was increased by 50% (Table 3). The results of the logistic regression showed that the adjusted risk of clinical events was increased significantly in diabetic patients compared with nondiabetic patients, with odds ratios of 1.74 for death from any cause (95% CI 1.19–2.03, P = 0.0001), 1.80 for fatal or nonfatal stroke (1.46–2.22, P = 0.0001), 1.93 for fatal and nonfatal coronary events (1.63–2.30, P = 0.0001), 2.03 for major cardiovascular events (1.76–2.35, P = 0.0001), and 1.48 for noncardiovascular death (1.19–1.84, P = 0.0004).

Compared with the control groups, sBP decreased less in the diabetic treated patients than in the nondiabetic treated patients at 1 year (P = 0.019), but not at 2 and 5 years (Table 4). These results do not take into account the changes in systolic pressure in the HDFP Study at 1 and 2 years of follow-up, for which individual data were not available in the INDANA database.

In nondiabetic patients (Figs. 1–4), the risk of all four outcomes was reduced significantly in the treated group by 17% (95% CI 8–25, P < 0.001) for death from any cause, by 37% (24–47, P < 0.001) for fatal and nonfatal stroke, by 23% (11–33, P < 0.001) for fatal and nonfatal coronary events, and by 28% (20–36, P < 0.001) for major cardiovascular events. In diabetic patients (Figs. 1–4), there was a nonsignificant 5% decrease in the risk of death from any cause (–18 to 23%, P = 0.65) and a nonsignificant 15% decrease in the risk of fatal or nonfatal coronary events (–11 to 35%, P = 0.23). There were significant decreases in the risks of fatal and nonfatal stroke (36%; 10–55, P = 0.011) and of major cardiovascular events (20%; 2–34, P = 0.032) in diabetic patients. No heterogeneity was detected between the trials nor between the diabetic and nondiabetic groups. The risk of cardiovascular death was significantly reduced by 25% in the nondiabetic patients (13–35%, P < 0.001) and nonsignificantly reduced by 15% in the diabetic patients (–13 to 36%, P = 0.27). The risk of noncardiovascular death was nonsignificantly decreased by 8% in the nondiabetic patients (–7 to 21%, P = 0.29) and nonsignificantly increased by 6% in the diabetic patients (–51 to 25%, P = 0.73). The number of outcomes avoided by treating 1,000 patients for 5 years are shown in Table 3.

Figure 2—Meta-analysis of studies of antihypertensive drugs in which diabetic patients were included: major cardiovascular events. Relative risks, 95% CIs, and sample sizes (Sizes) for each trial subgroup, each subgroup (diabetic and nondiabetic), and total are shown. For each subgroup, mean relative risk and P value are indicated. Cochran Q het, P value for heterogeneity test; D+, diabetic patients; ND, nondiabetic patients.

Sensitivity analysis
Because the HDFP Study (25) differed from the other trials by its design (referred care instead of placebo, with 44% patients having received antihypertensive therapy at any time during follow-up), we performed a sensitivity analysis by excluding this trial from the meta-analysis. This resulted in more than halving the number of diabetic patients included (1,006 2,254), which means a considerable decrease in the study power. In the nondiabetic patients, results were similar in the main and sensitivity analyses. In the diabetic patients, the difference between the treated and control groups was no longer significant for any outcome. However, the relative risk reductions were compatible with those observed in the main analysis, with the 95% CI of the relative risk in the sensitivity analysis entirely included in the 95% CI risk in the main analysis for all outcomes, except fatal and nonfatal coronary events. In the diabetic patients, the decreases in risk were 1% for death from any cause (–35 to 24%, P = 0.93), 33% for fatal or nonfatal coronary events (–6 to 58%, P = 0.09), 30% for fatal and nonfatal stroke (–9 to 55%, P = 0.11), 25% for major cardiovascular events (–1 to 44%, P = 0.058), 23% for cardiovascular death (–13 to 48%, P = 0.21). The risk of noncardiovascular death was nonsignificantly increased by 34% in the diabetic patients (–116 to 17%, P = 0.23)

Figure 3—Meta-analysis of studies of antihypertensive drugs in which diabetic patients were included: death from any cause. Relative risks, 95% CIs, and sample sizes (Sizes) for each trial subgroup, each subgroup (diabetic and nondiabetic), and total are shown. For each subgroup, mean relative risk and P value are indicated. Cochran Q het, P value for heterogeneity test; D+, diabetic patients; ND, nondiabetic patients.

CONCLUSIONS — Based on individual data, this meta-analysis of four trials consisting of 18,097 patients (2,254 diabetic patients) allowed a comparison of the benefit due to antihypertensive treatment in diabetic and nondiabetic hypertensive patients.

We considered all the trials included in previous overviews (20,32). Only one other long-term randomized placebo-controlled trial in hypertension has been published since, the Systolic Hypertension in Europe Trial (Syst-Eur) (33), which tested a calcium antagonist. The results of the present meta-analysis were probably not altered much by the absence of data on the 195 patients included in the three other studies (29–31). We have no information, however, on the presence of diabetic patients in four studies totaling 1,605 patients (11–14). Thus, we may have missed more diabetic patients. Diabetes status was defined according to the new criteria for the diagnosis of diabetes (21). Because biological assays were not performed for all patients at baseline, were not centralized, and were not ensured to have been always performed under fasting conditions, some patients may not have been classified correctly. However, restricting the analysis to patients with diabetes status as reported in each trial did not change the results much (although statistical significance was not reached) probably because of a lack of power (data not shown). Because the HDFP Study (25) differed by its design from the other trials included in the meta-analysis, it might have introduced some heterogeneity in the results. However, the exclusion of this trial in a sensitivity analysis only resulted in a considerable decrease in power but did not change the results in a meaningful way

Figure 4—Meta-analysis of studies of antihypertensive drugs in which diabetic patients were included: fatal or nonfatal coronary events. Relative risks, 95% CIs, and sample sizes (Sizes) for each trial subgroup, each subgroup (diabetic and nondiabetic), and total are shown. For each subgroup, mean relative risk and P value are indicated. Cochran Q het, P value for heterogeneity test; D+, diabetic patients; ND, nondiabetic patients.

There were no significant differences in relative risk reduction in diabetic patients compared with nondiabetic patients, but there were trends toward smaller effects of antihypertensive treatment on all outcomes, especially death from any cause. Only reductions in the risk of fatal and nonfatal stroke and of major cardiovascular events were significant in diabetic patients. This may be explained by the smaller number of diabetic patients, thus a lower statistical power, in the diabetic patients compared with the nondiabetic patients. The near absence of treatment effect on mortality in the diabetic patients was a combination of a moderate decrease in cardiovascular mortality and a small increase in noncardiovascular mortality. Because confidence intervals were very wide, the most likely explanation for the very small effect of the treatment on mortality is that it occurred by chance. An alternative explanation could be a true increase in noncardiovascular mortality resulting from a longer exposure due to a reduction in cardiovascular mortality. In fact, diabetic patients were at a greater risk of noncardiovascular death than nondiabetic patients. It has been shown in a simulation that the elimination of the excess cardiovascular risk in diabetic patients would only result in a trivial increase in life expectancy, with the patients having avoided cardiovascular events being at a greater risk of noncardiovascular death (34).

Finally, in the absence of any heterogeneity between the diabetic and nondiabetic subgroups, the effects of antihypertensive treatment can nevertheless be considered similar in both categories of patients. However, in terms of events avoided for 1,000 patients treated during 5 years, except for mortality, the treatment of hypertension appears to be more efficient in diabetic patients than in nondiabetic patients because the former are at a much higher risk of events. These decreases in clinical event rates might have been even greater if similar net decreases in blood pressure had been obtained in the diabetic and nondiabetic patients.

We have no explanation for the smaller decrease in sBP in diabetic patients compared with nondiabetic patients, and it must be stressed that this was observed only at 1 year, a measurement time at which the results of HDFP regarding sBP were not available. The same decreases in blood pressure were observed in the control groups in both diabetic and nondiabetic patients. Thus, the difference was not due to a more aggressive treatment of hypertension in the control group in diabetic patients. Treatment might therefore have been less effective in diabetic patients than in nondiabetic patients. However, it is not possible to ascribe this difference to a lower effectiveness of the treatments in diabetic patients or to the use of lower medication dosage in diabetic patients, for fear of the metabolic effects of diuretics and -blockers.

In the trials included in our meta-analysis, diuretics were used as first-line antihypertensive drugs in 92% of the treated patients, and -blockers were used in only 8%. This precludes any conclusion being drawn from our results regarding -blockers. By extrapolating the results of a trial in which captopril reduced the risk of death, dialysis, and transplantation in patients with type 1 diabetes and renal failure (35), and by extrapolating the results of a meta-analysis (36) suggesting that ACE inhibitors may preserve renal function in diabetic (mainly type 1 diabetic) patients better than other treatments, ACE inhibitors have been proven to be the best choice for treatment of both hypertensive type 1 and type 2 diabetic patients (37). However, no advantage of ACE inhibitors over other antihypertensive treatments in terms of decrease in the risk of clinical events has yet been demonstrated in type 2 diabetic patients, who represent the largest group of hypertensive diabetic patients. Secondary outcome results of two recent trials in diabetic patients might suggest that ACE inhibitors should be preferred to calcium channel blockers in diabetic patients (38,39). In addition, subgroup results of the Multicenter Isradipine Diuretic Atherosclerosis Study suggest that calcium channel blockers might be detrimental to diabetic patients compared with diuretics (40), although no patients with overt diabetes were included in this study. On the other hand, in the diabetic subgroup of the Syst-Eur study (41), patients benefited from treatment with nitrendipine compared with placebo. Finally, the Hypertension in Diabetes Study failed to demonstrate a difference between captopril and atenolol in reducing the risk of macrovascular complication in diabetic patients (42), and the Captopril Prevention Project Study (43) gave results that were very difficult to interpret. In this mega-trial (10,985 patients allocated to captopril or conventional therapy with diuretics, -blockers, or both), a beneficial effect of captopril compared with conventional treatment was shown in the small (572 patients) diabetic subgroup for most of the cardiovascular outcomes except stroke. However, no difference was shown between the two treatment groups in the whole trial (again, except for stroke) that was significantly more frequent in the captopril group. The open nature of the trial, with randomization by sealed envelopes and imbalances between the captopril and conventional groups at entry, render its interpretation difficult. For obvious ethical reasons, none of these recent trials included a placebo group. Randomized controlled trials have proven that diabetic hypertensive patients benefit from first-line treatment with diuretics. To what extent ACE inhibitors or other treatments may be more beneficial in diabetic patients should be assessed in comparison with diuretics in trials performed specifically in diabetic patients.

APPENDIX

Members of the INDANA Steering Committee
J.P. Boissel, F. Boutitie, F. Gueyffier (Clinical Pharmacology Department, Claude Bernard University [EA643], Lyon Hospitals, France); J. Cutler, L. Friedman, E. Schron (National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, MD); S. Pocock (London School of Hygiene and Tropical Medicine, London, U.K.); J. Coope (General Practice, Bollington, U.K.); T. Ekbom (Department of Community Health Sciences, Dalby/Lund, Sweden); R. Fagard (Hypertension and Cardiovascular Rehabilitation Unit, Leuven, Belgium); K. Kerlikowske (Veterans Administration Medical Center, San Francisco, CA); M. Perry (Washington University School of Medicine, St. Louis, MO); and R. Prineas (Department of Epidemiology and Public Health, Miami, FL).

References

1. Diabetes Drafting Group: World Health Organization Multinational Study of Vascular Disease in Diabetics: Prevalence of small vessel and large vessel disease in diabetic patients from 14 centres. Diabetologia 28:615–640, 1985

2. Assmann G, Schulte H: The Prospective Cardiovascular Munster (PROCAM) Study: prevalence of hyperlipidemia in persons with hypertension and/or diabetes mellitus and the relationship to coronary heart disease. Am Heart J 116:1713–1724, 1988

3. Stamler J, Vaccaro O, Neaton JD, Wentworth D, for the Multiple Risk Factor Intervention Trial Research Group: Diabetes, other risk factors, and 12-yr cardiovascular mortality for men screened in the Multiple Risk Factor Intervention Trial. Diabetes Care 16:434–444, 1993

4. Hypertension in Diabetes Study (HDS). I. Prevalence of hypertension in newly presenting type 2 diabetic patients and the association with risk factors for cardiovascular and diabetic complications. J Hypertens 11:309–317, 1993

5. Coope J, Warrender TS: Randomized trial of treatment of hypertension in elderly patients in primary care. Br Med J 293:1145–1151, 1986

6. Smith WM: US Public Health Service Hospitals Cooperative Study Group: Treatment of mild hypertension: results of a ten-year intervention trial. Circ Res 40 (Suppl. 1):98–105, 1977

7. Australian National Blood Pressure Management Committee: The Australian therapeutic trial in mild hypertension. Lancet i:1261–1267, 1980

8. Helgeland A: Treatment of mild hypertension: a five year controlled drug trial: the Oslo Study. Am J Med 69:725–732, 1980

9. Medical Research Council Working Party: MRC Trial of Treatment of Mild Hypertension: principal results. BMJ 291:97–104, 1985

10. MRC Working Party: Medical Research Council Trial of Treatment of Hypertension in Older Adults: principal results. Br Med J 304:405–412, 1992

11. Carter AB: Hypotensive therapy in stroke survivors. Lancet i:485–489, 1970

12. Veterans Administration Cooperative Study Group on Antihypertensive Agents: Effects of treatment on morbidity in hypertension. II. Results in patients with diastolic blood pressure averaging 90 through 114 mm Hg. JAMA 213:1143–1152, 1970

13. Barraclough M, Joy MD, MacGregor GA, Foley TH, Lee MR, Rosendorff Holland WW, Cranston WI, Rhea JN: Control of moderately raised blood pressure: report of a co-operative randomized trial. Br Med J 3:434–436, 1973

14. Veterans Administration/National Heart Lung and Blood Institute Study Group for Evaluating Treatment in Mild Hypertension: Evaluation of drug treatment in mild hypertension: plan and preliminary results of a two-year feasibility trial for a multicenter intervention study to evaluate the benefits versus the disadvantages of treating mild hypertension. Ann N Y Acad Sci 304:267–292, 1978

15. Flamenbaum W: Metabolic consequences of antihypertensive therapy. Ann Intern Med 98:875–880, 1983

16. Klein R, Moss SE, Klein BEK, De Meta DL: Relation of ocular and systemic factors to survival in diabetes. Arch Intern Med 149:266–272, 1989

17. Warram JH, Labbef LMB, Valsania P, Christlieb AR, Krolewski AS: Excess mortality associated with diuretic therapy in diabetes mellitus. Arch Intern Med 151:1350–1356, 1991

18. Curb JD, Pressel SL, Cutler JA, Savage PJ, Applegate WB, Black H, Camel G, Davis BR, Frost PH, Gonzalez N, Guthrie G, Oberman A, Ruttan GH, Stamler J, for the Systolic Hypertension in the Elderly Program Cooperative Research Group: Effect of diuretic-based antihypertensive treatment on cardiovascular disease risk in older diabetic patients with isolated systolic hypertension. JAMA 276:1886–1892, 1996

19. Gueyffier F, Boutitie F, Boissel JP, Coope J, Cutler J, Ekbom T, Fagard R, Friedman L, Perry HM, Pocock S, Prineas R, Schron E: INDANA: a meta-analysis on individual patient data in hypertension: protocol and preliminary results. Thérapie 50:353–362, 1995

20. Collins R, Peto R, MacMahon S, Hebert P, Fiebach NH, Eberlein KA, Godwin J, Qizilbash N, Taylor JO, Hennekens CH: Blood pressure, stroke, and coronary heart disease. II. Short-term reductions in blood pressure: overview of randomized drug trials in their epidemiological context. Lancet 335:827–838, 1990

21. The Expert Committee on the Diagnosis and Classification of Diabetes Mellitus: Report on the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 20:1183–1197, 1997

22. Whitehead A, Witehead J: A general parametric approach to the meta-analysis of randomized clinical trials. Stat Med 10:1665–1677, 1991

23. Boissel JP, Blanchard J, Panak E, Peyrieux JC, Sacks H: Considerations for the meta-analysis of randomized clinical trials: summary of a panel discussion. Control Clin Trials 10:254–281, 1989

24. Cucherat M, Boissel JP, Leizorovicz A, Haugh MC: EasyMA: a program for the meta-analysis of clinical trials. Comput Methods Programs Biomed 53:187–190, 1997

25. Hypertension Detection and Follow-up Program Cooperative Group: Five-year findings of the Hypertension Detection and Follow-up Program. I. Reduction in mortality in persons with high blood pressure, including mild hypertension. JAMA 242:2562–2571, 1979

26. Amery A, Birkenhager W, Brixko P, Bulpitt C, Clement D, Deruyttere M: Mortality and morbidity results from the European Working Party on High Blood Pressure in the Elderly trial. Lancet i:1349–1354, 1985

27. Dahlöf B, Lindholm LH, Hansson L, Schersten B, Ekbom T, Wester PO: Morbidity and mortality in the Swedish Trial in Old Patients with Hypertension (STOP-hypertension). Lancet 338:1281–1285, 1991

28. SHEP Cooperative Research Group: Prevention of stroke by antihypertensive drug treatment in older persons with isolated systolic hypertension: final results of the Systolic Hypertension in the Elderly Program (SHEP). JAMA 265:3255–3264, 1991

29. Veterans Administration Cooperative Study Group on Antihypertensive Agents: Effects of treatment on morbidity in hypertension: results in patients with diastolic blood pressures averaging 115 through 129 mm Hg. JAMA 202:116–122, 1967

30. Wolff FW, Lindeman RD: Effects of treatment in hypertension: results of a controlled study. J Chron Dis 19:227–240, 1966

31. Hypertension-Stroke Cooperative Study Group: Effect of antihypertensive treatment on stroke recurrence. JAMA 229:409–418, 1974

32. Lièvre M, Leizorovicz A: Treatment of high blood pressure in patients aged over 60 years: lessons from randomized clinical trials. Cardiology in the Elderly 3:217–222, 1995

33. Staessen JA, Fagard R, Thijs L, Celis H, Arabidze GG, Birkenhäger WH, Bulpitt CJ, De Leeuw PW, Dollery CT, Fletcher AE, Forette F, Leonetti G, Nachev C, O'Brien ET, Rosenfeld J, Rodicio JL, Tuomilehto J, Zanchetti A: Randomized double-blind comparison of placebo and active treatment for older patients with isolated systolic hypertension. Lancet 350:757–764, 1997

34. Eastman RC, Keen H: The impact of cardiovascular disease on people with diabetes: the potential for prevention. Lancet 350 (Suppl. 1):29–32, 1997

35. Lewis EJ, Hunsicker LG, Bain RP, Rhode RD, for the Collaborative Study Group: The effect of angiotensin-converting-enzyme inhibition on diabetic nephropathy. N Engl J Med 329:1456–1462, 1993

36. Kasiske BL, Kalil RSN, Ma JZ, Liao M, Keane WF: Effect of antihypertensive therapy on the kidney in patients with diabetes: a meta-regression analysis. Ann Intern Med 118:129–138, 1993

37. Mogensen CE, Keane WF, Bennett PH, Jerums G, Parving HH, Passa P, Steffes MW, Striker GE, Viberti GC: Prevention of diabetic renal disease with special reference to microalbuminuria. Lancet 346:1080–1084, 1995

38. Estacio RO, Jeffers BW, Hiatt WR, Biggerstaff SL, Gifford N, Schrier RW: The effect of nisoldipine as compared with enalapril on cardiovascular outcomes in patients with type-2 diabetes and hypertension. N Engl J Med 338:645–662, 1998

39. Tatti P, Pahor M, Byington RP, Di Mauro P, Guarisco R, Strollo G, Strollo F: Outcome results of the Fosinopril Versus Amlodipine Cardiovascular Events Randomized Trial (FACET) in patients with hypertension and NIDDM. Diabetes Care 21:597–603, 1998

40. Byington RP, Craven TE, Furberg CD, Pahor M: Isradipine, raised glycosylated haemoglobin, and risk of cardiovascular events. Lancet 350:1075–1076, 1997

41. Tuomilehto J, Rastenyte D, Birkenhäger WH, Thijs L, Antikanen R, Bulpitt CJ, Fletcher AE, Forette F, Goldhaber A, Palatini P, Sarti C, Fagard R: Effects of calcium-channel blockade in older patients with diabetes and systolic hypertension. N Engl J Med 340:677–684, 1999

42. U.K. Prospective Diabetes Study Group: Efficacy of atenolol and captopril in reducing the risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 39. BMJ 317:713–720, 1998

43. Hansson L, Lindholm L, Niskanen L, Lanke J, Hedner T, Niklason A: Effect of angiotensin-converting-enzyme inhibition compared with conventional therapy on cardiovascular morbidity and mortality in hypertension: the Captopril Prevention Project (CAPP) randomised trial. Lancet 353:611–616, 1999

From the Clinical Pharmacology Department (M.L., F.G., J.-P.B.), Claude Bernard University (Equipe d'Accueil 643), Lyon Hospitals, Lyon, France; the Department of Community Health Sciences (T.E.), Dalby/Lund, Sweden; the Hypertension and Cardiovascular Rehabilitation Unit (R.F.), Leuven, Belgium; the National Heart, Lung, and Blood Institute (J.C., E.S.), National Institutes of Health, Bethesda, Maryland; and Medicine B (M.M.), University Hospital, Angers, France. A complete listing of the INDANA Steering Committee members is given in the appendix.

Address correspondence and reprint requests to Michel Lièvre, PhD, Service de Pharmacologie Clinique, Faculté de Médecine HRT Laënnec, rue Guillaume Paradin, BP 8071, 69376 Lyon Cedex 08, France. E-mail: michel.lievre@upcl.univ-lyon1.fr.

Received for publication 9 July 1999 and accepted in revised form 9 December 1999.

Abbreviations: dBP, diastolic blood pressure; EWPHE, European Working Party on High Blood Pressure in the Elderly; HDFP, Hypertension Detection and Follow-up Program; INDANA, Individual Data Analysis of Antihypertensive Drug Interventions; sBP, systolic blood pressure; SHEP, Systolic Hypertension in the Elderly Program; STOP-H, Swedish Trial in Old Patients with Hypertension; Syst-Eur, Systolic Hypertension in Europe Trial.

A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances.

This article is based on a presentation at a symposium. The symposium and the publication of this article were made possible by an unrestricted educational grant from Aventis Pharma.

For Technical Issues contact webmaster@diabetes.org