Lessons Learned from the ACCOMPLISH and ALLHAT Trials

Today we will discuss the ACCOMPLISH Trial from 2008 and ALLHAT Trial from 2002. Yes, a trial from over 20 years ago. Why are we bringing these back to the limelight? I wanted to review these trials specifically to show you that not all first-line guideline-recommended therapies are the same.

Lessons Learned from the ACCOMPLISH and ALLHAT Trials
Photo by Laurynas Mereckas / Unsplash

TL;DR version:

  • We practice medicine and prescribe therapies based on clinical trial data (also known as "evidenced-based medicine").
  • There are numerous therapies and medication classes to treat disease states, such as hypertension (HTN).
  • ACCOMPLISH trial: benazepril-amlodipine was superior to benazepril-hydrochlorothiazide in reducing CV events in patients with HTN at high risk for such events.
  • ALLHAT trial: chlorthalidone was superior to amlodipine and lisinopril for reducing SBP; chlorthalidone reduced rates of HF and HF hospitalizations/death compared to amlodipine; chlorthalidone reduced rates of stroke, combined CV disease, HF, and hospitalized/treated angina compared to lisinopril.
  • What does this mean? All medications used to treat the same disease are not equal. Want to know more? Keep reading below!

Disclaimer: This post is meant to provide an overview of a clinical topic that may include (but is not limited to) information on pathophysiology, diagnosis, treatment, clinical pharmacology, medication management, adverse effects, and clinical pearls. References are included at the bottom of each post. This post is not to be used as medical advice for direct patient care, but as a guide for learning and discussion.

A Blast from the Past

We practice medicine based on the data we have available to us. This data comes from clinical trials. The clinical trials are collected, analyzed, and presented in various resources such as guidelines, meta-analyses, textbooks, lectures, continuing educations, etc. Taking recommendations from the data and implementing them into your clinical practice is called "evidence-based medicine."

New clinical trials are published every minute. It seems impossible to keep up with how much information is being discovered every day (looking at you, COVID-19). Every so often, we need to remember we started: the landmark clinical trials.

There are dozens of key clinical trials in the world of HTN. The current AHA/ACC 2017 Hypertension Guidelines state: "For initiation of antihypertensive drug therapy, first-line agents include thiazide diuretics, CCBs, and ACE inhibitors or ARBs" (Table 8.1.6, Classification of Recommendation I, Level of Recommendation A). They provide some guidance of when clinicians should prescribe some medications over others, such as co-morbidities and race, but how else can we differentiate what the "best" first-line option is to use? All antihypertensive reduce blood pressure, right?

Today we will discuss the ACCOMPLISH Trial from 2008 and ALLHAT Trial from 2002. Yes, a trial from over 20 years ago. Why are we bringing these back to the limelight? Both landmark trials give great evidence that not all first-line guideline-recommended therapies are the same.

Il s’agit d’une photographie réalisée pour illustrer la page de recherche du site internet de l’agence web Olloweb Solutions
Photo by Agence Olloweb / Unsplash


Objective: To determine if combination therapy of angiotensin converting enzyme inhibitor (ACEI) plus amlodipine is superior to ACEI plus thiazide diuretic in reducing cardiovascular (CV) outcomes in high-risk patients with HTN

Primary efficacy measure: Composite of fatal and non-fatal CV events

Secondary efficacy measure: Composite of CV death, non-fatal myocardial infarction (MI), and non-fatal stroke

Population: Patients with HTN at high-risk for CV events

  • Mean age 68 years; male 60%
  • Mean BP 145/80 mmHg
  • Patients on lipid lowering therapy – 68%; antiplatelet 65%

Inclusion Criteria:

  • Age >60
  • High-risk for CV events (defined as history of coronary events, MI, revascularization, stroke, impaired renal function, diabetes mellitus (DM), peripheral artery disease, left-ventricular (LV) hypertrophy)
  • SBP >160 mmHg

Exclusion Criteria:

  • Symptomatic heart failure or LV ejection fraction <40%
  • Acute CV event within previous 1-3 months
  • Secondary HTN

Intervention groups:

  • Benazepril/amlodipine 20/5 mg
  • Benazepril/hydrochlorothiazide 20/12.5 mg
  • *Both therapy groups were titrated on the intervention medication to achieve a target blood pressure of <140/90 mmHg or <130/80 mmHg for DM/CKD patients
  • **The use of antihypertensive medications from other classes were allowed

Duration: mean follow-up period of 15 months


  • 5744 in benazepril/amlodipine and 5762 in benazepril/HCTZ
  • Baseline characteristics were similar between groups
  • Avg daily dose benazepril/amlodipine was 36.3/7.7 mg and benazepril/HCTZ was 36.1/19.3 mg
  • Avg BP for benazepril/amlodipine was 131.6/74.4 mmHg and benazepril/HCTZ was 132.5/74.4 mmHg (p<0.001)

Composite of fatal and non-fatal CV events: 552 (9.61%) vs 679 (11.8%), HR 0.80 (95% CI 0.72-0.90); p<0.001); ARR 2.17%; NNT 46

CV Death: 107 (1.86%) vs 134 (2.33%); HR 0.80 (0.62-1.03)
Fatal and non-fatal MI: 125 (2.18%) vs 159 (2.76%); HR 0.78 (0.62-0.99); ARR 0.58%; NNT 172
Fatal and non-fatal stroke: 112 (1.95%) vs 133 (2.31%); HR 0.84 (0.65-1.08)
Composite of CV death, non-fatal MI, and non-fatal stroke: 288 (5.01%) vs 364 (6.32%); HR 0.79 (0.67-0.92); ARR 1.30%; NNT 77


  • Additional blood pressure lowering agents from other classes were allowed
  • Cannot apply results to thiazide-type diuretics such as chlorthalidone

Recommendation: In high-risk patients with HTN, we recommend the use of benazepril/amlodipine over benazepril/HCTZ for reducing CV event rates, particularly fatal and non-fatal MI

Special population considerations

  • Target BP of <140/90 mmHg (<130/80 mmHg for DM/CKD patients) must be considered when interpreting the results
  • The most common baseline risk-factor for CV event was DM (60%)

Hats waiting for a head where they can perform their duties
Photo by Héctor J. Rivas / Unsplash

ALLHAT Trial (JAMA 2002)

Objective: To determine the effect of CCBs, ACEIs and alpha-blockers compared to thiazide-type diuretics on morbidity and mortality in high-risk patients with HTN

Primary efficacy measure: Composite of fatal coronary heart disease (CHD) or non-fatal MI

Secondary efficacy measure: All-cause mortality, fatal/non-fatal stroke, combined CHD, combined CV disease

Population: Patients with stage 1 or 2 HTN at high-risk for cardiac events

  • Mean age 67; male 53%
  • Mean BP 146/84 mmHg

Inclusion Criteria:

  • Age ≥ 55
  • HTN plus ≥ 1 risk factor for CHD disease (previous MI or stroke > 6 months prior, LV hypertrophy, T2DM, current cigarette smoker, HDL < 35 mg/dL, documented ASCVD)

Exclusion Criteria:

  • History of hospitalization due to heart failure (HF)
  • Symptomatic HF
  • LVEF < 35%

Intervention drugs:

  • Chlorthalidone, Lisinopril, Amlodipine, Doxazosin
  • *Study medication was titrated to a target BP of <140/90 mmHg in all groups
  • **The doxazosin group was stopped early due to increase rated of HF
  • ***Patients randomized 1.7:1:1 (higher in diuretic group to maximize power)

Duration: mean follow-up period of 4.9 years


  • 15255 in chlorthalidone group, 9054 in lisinopril, 9048 in amlodipine
  • No significant difference was seen for the primary outcome for either amlodipine or lisinopril compared to chlorthalidone
  • Average SBP at 5 years was significantly lower in the chlorthalidone group compared to amlodipine (133.9 mmHg vs 134.7 mmHg; p=0.03) and when compared with lisinopril (133.9 mmHg vs 125.9 mmHg; p<0.001).
  • Baseline characteristics were nearly identical across the three groups

Amlodipine vs chlorthalidone
Composite of fatal CHD and non-fatal MI:
798 (8.82%) vs 1362 (8.93%), RR 0.98 (0.90-1.07)

Heart failure: 706 (7.80%) vs 870 (5.70%), RR 1.38 (1.25-1.52); ARR 2.10%; NNT 48

Heart failure hospitalizations/heart failure death: 578 (6.39%) vs 724 (4.75%), RR 1.35 (1.21-1.50); ARR 1.64%; NNT 61

Lisinopril vs chlorthalidone

Composite of fatal CHD and non-fatal MI: 796 (8.79%) vs 1362 (8.93%), RR0.99 (0.91-1.08)

Stroke: 457 (5.05%) vs 675 (4.42%), RR 1.15 (1.02-1.30); p=0.02; ARR 0.62%; NNT 161

Combined Cardiovascular Disease: 2514 (27.8%) vs 3941 (25.8%), RR 1.10 (1.05-1.16); p<0.001; ARR 1.93%; NNT 52

Heart failure: 612 (6.76%) vs 870 (5.70%), RR 1.19 (1.07-1.31); p<0.001; ARR 1.06%; NNT 95

Hospitalized/treated angina: 1019 (11.3%) vs 1567 (10.3%), RR 1.11 (1.03-1.20); p=0.01; ARR 0.98%; NNT 102


  • Use of open-label add-on antihypertensive therapy (however, statistical differences were still seen for multiple outcomes between treatment groups)

Results summary:

  • No difference in rates of the primary outcome between any of the study medications
  • Rates of HF and HF hospitalizations/death were significantly lower in the chlorthalidone group compared to amlodipine
  • Rates of stroke, combined CV disease, HF, and hospitalized/treated angina were significantly lower in the chlorthalidone group compared to lisinopril
  • Average SBP at 5 years was significantly lower in the chlorthalidone group compared to both amlodipine and lisinopril

Recommendation: In high-risk HTN patients, we recommend the use of chlorthalidone over amlodipine and lisinopril as first-line therapy

Special population considerations:

  • Cannot extrapolate effect of chlorthalidone to other thiazides such as hydrochlorothiazide (which, arguably, is used more in clinical practice with hopes that this effect can be extrapolated).

What now? | Follow on Instagram: @timmossholder
Photo by Tim Mossholder / Unsplash

Final Thoughts

It is important to know the basics about differentiating when to use one medication over another to treat a disease state. These basics are drilled into our heads throughout pharmacy school on exams, patient discussions, and rotations.

Once you enter clinical practice and see a vast array of patients, you will quickly learn how much more there is to add to the "basics" taught in pharmacy school.

I hope you were able to finish this article with a new sense of appreciation for clinical trials and outcomes! The primary outcome is not ~always~ the most important. The secondary outcomes (when powered appropriately) can provide a lot of context for medication use and generate many hypotheses for future studies!

Bonus Thought: Is HCTZ Equivalent to Chlorthalidone?

A recent disclosure by Dr. John M. Mandrola published on Medscape a few days ago mentioned an update on the Diuretic Comparison Project (DCP). The DCP was started in 2016 using data from the Veterans Administration system. The DCP set out to compare CV outcomes with the either HCTZ or chlorthalidone in patients treated with HTN.

5 year results from the DCP were presented at the American Heart Association (AHA) Scientific Sessions 2022 in Chicago.

  • Eligible patients: patients taking HCTZ with a SBP greater than 120 mmHg. They both the patients and their clinicians had to give consent. One group continued with HCTZ and the other was switched to chlorthalidone.
  • Primary outcome: first occurrence of a major CV outcome (stroke, MI, noncancer death, hospitalization for HF, or urgent coronary revascularization).
  • Baseline population: mostly male; mean age of 72 years
  • Results: after 5 years of follow-up, there was no difference in the primary outcome (HR, 1.04; CI, 0.94-1.16). No component of the primary endpoint differed significantly.
  • Safety: hypokalemia occurred in 6% of the chlorthalidone group vs 4.4% of the HCTZ group—a difference that met statistical significance. Potassium levels below 3.1 mEq/L were significantly more common in the chlorthalidone group.

The meaning of this? HCTZ seems to be equivalent to chlorthalidone – which from a cost, availability, and prevalence of use perspective – this is great news for the medical community. Many practitioners have used HCTZ (a thiazide diuretic) in practice from the positive data behind chlorthalidone (a thiazide-type diuretic).

Of course, the DCP is not the end-all-be-all for the HCTZ vs chlorthalidone debate, but it is definitely refreshing to see some head-to-head data from such a large patient database over a long 5 year trial period.


  1. Jamerson K, Weber MA, Bakris GL, et al. Benazepril plus amlodipine or hydrochlorothiazide for hypertension in high-risk patients. NEJM. 2008;359)23):2417-28. https://www.nejm.org/doi/full/10.1056/nejmoa0806182
  2. Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic: The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA. 2002;288(23):2981-97. https://jamanetwork.com/journals/jama/fullarticle/195626
  3. Poppen, A. High-Powered Medicine: Landmark Clinical Trial Reviews. Published Jan 2021. https://www.amazon.com/High-Powered-Medicine-Landmark-Clinical-Reviews/dp/B08TS8GDVS#detailBullets_feature_div
  4. Frank A. Lederle, William C. Cushman, Ryan E. Ferguson, et al; Chlorthalidone Versus Hydrochlorothiazide: A New Kind of Veterans Affairs Cooperative Study. Ann Intern Med.2016;165:663-664. [Epub 16 August 2016].