1 ALLHAT Role of Diuretics in the Prevention

1 ALLHAT Role of Diuretics in the Prevention of Heart Failure The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial Davis BR, Piller LB, Cutler JA, et al. Circulation 2006. 113: 2201 -2210.

2 ALLHAT Introduction and Background l Heart failure is a major public health problem, especially in persons 65 years of age and older (= number one reason for hospitalizations in this age group). l Age-adjusted incidence per 100, 000 personyears during 1990 -1999 was 564 for men and 327 for women, age 65 -74 years (NEJM, 2002, Framingham) l Five-year age-adjusted survival rate was only 59% among men and 45% for women. l In 91% of HF cases, hypertension is an antecedent (Framingham, JAMA, 1996)

3 ALLHAT Hypertension Control and Heart Failure • In a meta-analysis of 12 trials of patients with hypertension it was found that, compared to placebo, drug therapy for hypertension prevents over 50% of HF events (Moser, JACC, 1996). • In another meta-analysis, diuretics and betablockers (BB) were equally effective in preventing HF events (Psaty, JAMA, 1997).

4 ALLHAT Hypertension Control and Heart Failure • A meta–analysis of active comparator trials found no significant difference between ACEinhibitors and diuretics for preventing HF; ACEinhibitors were more efficacious than CCBs (BPLTT Collaboration, Lancet, 2002). • The INSIGHT trial found that a long-acting nifedipine regimen was associated with a > 2 x higher incidence of HF events compared to a diuretic combination (HCTZ/amiolride) (Brown, Lancet, 2000).

5 ALLHAT Objectives • Characterize HF in ALLHAT by its antecedent risk factors and underlying conditions. • Examine occurrence of HF by treatment group overall, in subgroups, and over time. • Explore relation of initial occurrence of HF to prerandomization type of BP medication used. • Explore follow-up BP and use of additional drugs as mediating/modifying factors. • Examine post-HF mortality overall and by treatment group.

Randomized Design of ALLHAT Hypertension Trial ALLHAT 42, 418 high-risk hypertensive patients 90% previously treated 10% untreated STEP 1 AGENTS Chlorthalidone Amlodipine Lisinopril Doxazosin 12. 5 -25 mg 2. 5 -10 mg 10 -40 mg 1 -8 mg N=9, 054 N=9, 061 N=15, 255 N=9, 048 Other AHT Drugs STEP 2 AND 3 AGENTS (5 years) Atenolol 28. 0% Clonidine 10. 6% Reserpine 4. 3% Hydralazine 10. 9% 6

7 ALLHAT Decision to Stop Doxazosin Arm • NHLBI Director accepted recommendation of independent review group to terminate doxazosin arm (early in year 2000), due to: – Futility of finding a significant difference for primary outcome – Statistically significant 25 percent higher rate of major secondary endpoint, combined CVD outcomes, along with twofold higher rate of HF • Detailed HF analyses published (Davis et al. Ann Intern Med 2002).

8 ALLHAT Heart Failure Data Collection • Hospitalized nonfatal – discharge summary • Hospitalized fatal – death certificate, discharge summary • Nonhospitalized fatal – death certificate • Nonhospitalized nonfatal (treated) – clinician report • 100% review of discharge summaries and death certificates by CTC Medical Reviewers – Queries to clinics if diagnosis questionable

9 ALLHAT Criteria for HF Evaluation* Must have one from each category: Category “A” Category “B” Paroxysmal nocturnal dyspnea Rales Dyspnea at rest 2+ ankle edema NYHA Classification III Tachycardia Orthopnea Cardiomegaly by CXR characteristic of CHF S 3 gallop Jugular venous distention *ALLHAT Manual of Operations, 5. 3. 4; adopted from the SHEP trial

10 ALLHAT Validity of HF Outcome Verified • Traditional risk factors in agreement with previous studies, e. g. , Framingham • HF Validation Study confirmed original observed treatment differences – Independent central review using both ALLHAT and Framingham criteria

11 ALLHAT Heart Failure Validation Study Criteria % Agreement ALLHAT 71% Framingham Heart Study 80% Reviewers’ judgement 84%

12 ALLHAT Inclusion/Exclusion Criteria for Antihypertensive Trial • Men and women > 55 years old • If untreated: 140/90, 180/110 mm Hg (2 visits) • If treated: ≤ 160/100 mm Hg (visit 1), ≤ 180/110 mm Hg (visit 2) – No washout required • At least one additional cardiovascular risk factor • Exclude if symptomatic HF or EF < 35%, creatinine 2 mg/d. L, require diuretics, CCB, ACEI, or AB’s for non-BP indication

13 Step 1 Treatment Protocol ALLHAT Step 1 Agent Initial Dose* Dose 1* Dose 2* Dose 3* Chlorthalidone 12. 5 25 Amlodipine 2. 5 5 10 Lisinopril 10 10 20 40 Doxazosin 1 2 4 8 * mg/day Step 2/3 drugs –– atenolol, reserpine, clonidine, hydralazine “Non-study” drugs –– all other antihypertensive medications

14 ALLHAT Baseline Characteristics Hospitalized/Fatal HF During Trial Yes No Difference p N 1, 773 31, 584 Age (mean) 70. 3 66. 7 +3. 6 <0. 001 Men, % 55. 2% 53. 0% +2. 2% 0. 008 Pre-RZ Treatment, % 93. 1% 90. 0% +3. 1% 0. 004 SBP (mean mm Hg) 148. 2 146. 2 +2. 0 <0. 001 DBP (mean mm Hg) 81. 8 84. 1 -2. 3 <0. 001 Pulse (mean bpm) 74. 6 73. 5 +1. 1 <0. 001 18. 3% 22. 1% -3. 8 <0. 001 49. 4 35. 4 +14. 0% <0. 001 LVH by ECG, % 18. 4% 16. 3% +2. 1% <0. 001 History of CHD, % 37. 6% 24. 7% +12. 9 <0. 001 30. 3 29. 7 +0. 6 <0. 001 Cigarette smoking, % Diabetes, % BMI (mean)

Hospitalized/ Fatal Heart Failure by ALLHAT Treatment Group ALLHAT Cumulative Event Rate . 1. 08 RR 95% CI A-C 1. 35 1. 21 -1. 50 L-C 1. 11 0. 99 -1. 24 A-L 1. 23 1. 09 – 1. 38 Chlorthalidone Amlodipine Lisinopril . 06. 04. 02 0 0 1 2 3 Years 4 5 6 7 15

16 ALLHAT Heart Failure Before and After 1 Year • Observed HF differences were larger earlier in the follow-up. • The lisinopril group had a lower HF rate than the amlodipine group, but event curves did not separate until later. • A test of the proportional hazards assumption for Cox regression revealed that RRs were not constant over time. Therefore, a Cox regression that used a time-dependent indicator variable (<=1 year versus >1 year) was utilized.

Hospitalized/ Fatal Heart Failure by ALLHAT Treatment Group Within 1 Year and >1 Year ALLHAT Cumulative Hosp/Fatal HF Rate Baseline to Year 1 . 02 17 RR 95% CI A-C 2. 22 1. 69 – 2. 91 L-C 2. 08 1. 58 – 2. 74 A-L 1. 07 0. 82 – 1. 38 > Year 1 . 1 Chlorthalidone Amlodipine Lisinopril RR 95% CI A-C 1. 22 1. 08 – 1. 38 L-C 0. 96 0. 85 – 1. 10 A-L 1. 27 1. 10 – 1. 46 . 08 . 06. 01. 04 . 02 00 0 0 . 5 Years to Hosp/Fatal HF 1 1 2 3 4 5 6 Years to Hosp/Fatal HF 7

18 ALLHAT Hospitalized/fatal HF in Subgroups Amlodipine / Chlorthalidone Relative Risks from Baseline to 1 Year of Follow-up Relative Risk (95% CI) Total Age < 65 Age ≥ 65 Non-Black Men Women Diabetic Non-Diabetic Favors Amlodipine Chlorthalidone 2. 22 (1. 69 - 2. 91) 2. 89 (1. 62 - 5. 17) 2. 06 (1. 51 - 2. 80) 2. 12 (1. 49 - 3. 01) 2. 37 (1. 55 - 3. 63) 2. 27 (1. 56 - 3. 30) 2. 17 (1. 46 - 3. 21) 2. 71 (1. 83 - 4. 02) 1. 83 (1. 25 - 2. 67) 0. 50 1 2 3 4 5 6

19 ALLHAT Hospitalized/fatal HF in Subgroups Amlodipine / Chlorthalidone Relative Risks After 1 Year of Follow-up Relative Risk (95% CI) Total Age < 65 Age ≥ 65 Non-Black Men Women Diabetic Non-Diabetic Favors Amlodipine Chlorthalidone 1. 22 (1. 08 - 1. 38) 1. 38 (1. 10 - 1. 73) 1. 17 (1. 02 - 1. 35) 1. 20 (1. 04 - 1. 39) 1. 28 (1. 03 - 1. 58) 1. 28 (1. 09 - 1. 50) 1. 16 (0. 97 - 1. 39) 1. 23 (1. 04 - 1. 46) 1. 21 (1. 02 - 1. 43) 0. 50 1 2 3 4 5 6

ALLHAT Hospitalized/fatal HF in Subgroups Lisinopril / Chlorthalidone Relative Risks from Baseline to 1 Year of Follow-up Relative Risk (95% CI) Total Age < 65 Age ≥ 65 Non-Black Men Women Diabetic Non-Diabetic Favors Lisinopril Favors Chlorthalidone 2. 08 (1. 58 - 2. 74) 2. 53 (1. 39 - 4. 59) 1. 98 (1. 45 - 2. 70) 2. 04 (1. 43 - 2. 90) 2. 15 (1. 39 - 3. 33) 1. 80 (1. 22 - 2. 67) 2. 40 (1. 63 - 3. 54) 1. 99 (1. 31 - 3. 05) 2. 16 (1. 50 - 3. 10) 0. 50 1 2 3 4 5 20

ALLHAT Hospitalized/fatal HF in Subgroups Lisinopril / Chlorthalidone Relative Risks After 1 Year of Follow-up Relative Risk (95% CI) Total Age < 65 Age ≥ 65 Non-Black Men Women Diabetic Non-Diabetic Favors Lisinopril Favors Chlorthalidone 0. 96 (0. 85 - 1. 10) 0. 95 (0. 74 - 1. 23) 0. 97 (0. 84 - 1. 13) 0. 90 (0. 77 - 1. 06) 1. 10 (0. 88 - 1. 37) 1. 02 (0. 86 - 1. 21) 0. 89 (0. 73 - 1. 09) 1. 01 (0. 84 - 1. 22) 0. 93 (0. 77 - 1. 12) 0. 50 1 2 21

22 ALLHAT HF Development and Relation to Other Outcomes • HF development associated with: – 6. 6 -fold increase in death rate – 11. 7 -fold increase in CV death rate • Previous MI → 5. 7 -fold increased HF risk • Of participants with hospitalized HF: – 72% hospitalized once – 23. 3% hospitalized 2 -3 times – 4. 7% hospitalized 4+ times

23 ALLHAT Why are hazard ratios not constant throughout? Hypotheses? • Withdrawal from BP meds used prior to enrollment • Time course for effect of first-step (primary) drug – Diuretic – immediate? – ACEI – delayed? • Addition of step-up meds (esp. anti-HF meds) • Differences in BP

24 ALLHAT Prior Use of Antihypertensive Agents • Prior medication use associated with HF risk, especially during first year – RR 1. 42 (1. 18 – 1. 71) • Relative benefits of chlorthalidone consistent with or without prior antihypertensive medication use

25 ALLHAT Specific Prior Antihypertensive Agents • Data not collected within ALLHAT – Available for 1115 / 1773 HF cases CCB’s 47% ACEI 37% Diuretics 39% • Case-only analysis – No evidence for any statistically significant interaction between prior drug type (e. g. , diuretic) and treatment effect for HF, overall or during the first year

26 ALLHAT Immediate vs Delayed Effects • Do diuretics have a more immediate effect on HF prevention than ACEI or ARB? – Effect of diuretics begins at trial onset – Several ACEI vs placebo studies suggest that ACEI effect is not immediate – VALUE trial – valsartan vs amlodipine – HF similar in first 2 years, strong trend afterward favoring valsartan

27 ALLHAT Use of Step-up BP Meds Addition of Step 2 and Step 3 meds could have contributed to lessening or cessation of divergence of HF curves after 1 year.

ALLHAT Open-Label ACEI and Atenolol Use 28

29 ALLHAT Open-Label Diuretic and CCB Use

ALLHAT Diuretic, ACEI, or Atenol Use 30

31 ALLHAT BP Results by Treatment Group Compared to chlorthalidone: SBP significantly higher in the amlodipine group (~1 mm Hg) and the lisinopril group (~2 mm Hg). DBP significantly lower in the amlodipine group (~1 mm Hg).

32 ALLHAT BP Differences l Adjustment for follow-up SBP as timedependent covariates in a Cox regression model only slightly modified the relative risks 2. 22 2. 16 first year, 1. 22 1. 18 after 1 year l Amlodipine/chlorthalidone 2. 08 2. 01 first year, 0. 96 0. 93 after 1 year l Lisinopril/chlorthalidone

33 Cumulative Event Rate ALLHAT All-Cause Mortality Chlorthalidone Amlodipine Lisinopril . 6. 5. 4. 3. 2. 1 0 0 1 2 3 4 5 Years from Hospitalized HF to Death 6 7

34 ALLHAT Post-HF Mortality • Mortality rates after hospitalized HF high relative to those seen in ALLHAT overall – 25% vs 5% at 2. 5 years, respectively • No significant treatment group differences for post-HF mortality • The reason that the treatment difference for hospitalized HF did not translate into an effect on total mortality is that only 5. 6% of all deaths were attributed to HF.

ALLHAT Heart Failure and Total Mortality • Lisinopril-chlorthalidone absolute difference in hospitalized HF over 6 years was 0. 4%. – The excess of cases in the lisinopril group = 36 patients. – Case-fatality rate over average follow-up of 2. 5 years = 25%. – Thus, 9 excess cases of fatal HF would be expected in the lisinopril group. This is fewer than 1% of all deaths in the lisinopril group (n=1314). • Similar calculations for the amlodipine group: – 154 excess cases of hospitalized HF – Estimated number of fatal HF cases was 39, 3% of the amlodipine deaths (n=1256). 35

36 ALLHAT Effect on Total Mortality • HF differences in the trial would not have affected differences in total mortality – Also noted in the BPLTTC analyses – Absolute HF risk low – Increase in RR outweighed by even small reduction in higher absolute risks for stroke and CHD – Differences in # of HF events during trial result in only very small differences in # of deaths – ALLHAT post-trial mortality surveillance to examine this further

37 ALLHAT Conclusions 1 • Chlorthalidone superior to amlodipine in both time periods • Chlorthalidone superior to lisinopril during the first year • True for subgroups – age, race, sex, diabetes history • Other factors could not individually account for all of the observed treatment differences – Prior antihypertensive meds – Other open-label BP meds – Follow-up BP differences

38 ALLHAT Conclusions 2 • Developing HF is associated with a high mortality rate (~50% at 5 years) • It may take time for HF differences to translate into detectable mortality differences between treatments • Diuretics are clearly preferred over CCBs overall and over ACE inhibitors, at least in the short term, in preventing HF.

39 Extra Slides

40 ALLHAT Placebo-Controlled Trials • Most placebo-controlled trial have used diuretics and/or β-blockers as active regimens • Diuretics & ACEI shown to prevent HF in patients with hypertension – SHEP, HOPE • CCB vs placebo trials less conclusive – Syst-Eur • Meta-analyses – active therapy of hypertension can prevent >40% of HF events – Psaty, Smith, Siscovick, et al.

41 ALLHAT Active-Controlled Trials • VALUE • STOP Hypertension-2 • ANBP 2 • INVEST • CONVINCE – CCB or diuretic/β-blocker – BP reduced similarly, HF 30% more with CCB

42 ALLHAT BPLTTC Meta-Analyses • CCB-based therapies – NS 20% increase in HF incidence compared with placebo – 33% higher risk of HF compared with diuretic/β-blocker • ACEI-based therapies – 18% fewer HF events than with CCB or placebo – 7% NS higher risk than with diuretic/ β-blocker • CCBs less effective in preventing HF than other regimens • ACEI no more effective in preventing HF than diuretic/ β-blocker

43 Randomized Design of ALLHAT High-risk hypertensive patients ≥ 55 years Consent / Randomize (42, 418) Eligible for lipidlowering Amlodipine Chlorthalidone Doxazosin Lisinopril Not eligible for lipid-lowering Consent / Randomize (10, 355) Pravastatin Usual care Follow for CHD and other outcomes until death or end of study (up to 8 yr).

44 Event Reduction in SHEP, Syst-Eur, and HOPE SHEP: Systolic Hypertension in the Elderly, n=4, 736; chlorthalidone Syst-Eur: Systolic Hypertension in Europe, n=4, 695; nitrendipine HOPE: Heart Outcomes Prevention Evaluation Study, n=9, 297; ramipril