May 2023

Hydrochlorthiazide: the turtle of the race?
“Adalphane Acedrex (a Novartis made combo of Reserpine+Dihydralazine+Hydrochlorothiazide)- was available in the hospital formulary and ‘uncontrolled hypertension’ was a rare condition”, Prof Hase would get nostalgic and remark on his grand rounds discussing evolution of pharmacotherapy for hypertension.

One of the earliest classes of antihypertensives, thiazide diuretics, are also one of the most effective drugs, with latest trial evaluating chlorthalidone showing a BP decline of almost 10mmHg. However, whether chlortalidone should altogether replace hydrochlorothiazide is still a matter of debate. This is an attractive proposition given the higher potency and some pleotropic benefits attributed to chlorthalidone (the term ‘pleotropic effect’ awakens sleeping skeptic in us, as this term is typically invoked to defend something that can’t stand scientific rigor alone). Large observational studies and meta-analyses have differing conclusions waiting for a randomized trial to settle the debate. (see here, here and here)

In a pragmatic randomized controlled trial involving 13,523 patients over 65 years of age, getting treated at Veterans Affairs health system, a switch to chlorthalidone (hydrochlorothiazide was the default thiazide at baseline), 12.5 or 25 mg did not improve the primary composite outcome of nonfatal myocardial infarction, stroke, heart failure resulting in hospitalization, urgent coronary revascularization for unstable angina, and non–cancer-related death. Primary outcome occurred in 702 patients [10.4%] in chlorthalidone and 675 patients [10.0%] in the hydrochlorothiazide group (hazard ratio, 1.04; 95% confidence interval, 0.94 to 1.16; P=0.45). The incidence of hypokalemia was higher in the chlorthalidone group than in the hydrochlorothiazide group (6.0% vs. 4.4%, P<0.001).

This trial is long-awaited evidence to understand the difference between two commonly used thiazide diuretics for the treatment of hypertension and reassures those physicians who haven’t yet switched to chlorthalidone. The superior efficacy of chlorthalidone to reduce blood pressure doesn’t translate into effectiveness. This trial is an example of low cost, operational research in the setting of structured heath care systems like VA. Such pragmatic trials are feasible, can be completed quickly (phenomenal recruitment rate of 100 patients per week!), and are possible at a substantially lesser cost.

Many hypertension clinics have already switched to chlorthalidone given the higher potency and 24-hour action of chlorthalidone. Both of these reasons are not good enough to inform practice. First, potency may not necessarily translate into efficacy (which current trial confirms). For example, immediate release nifedipine or hydralazine are potent antihypertensives but not necessarily effective at the ultimate goal of BP control-CV protection. Duration of action as well may not be valid argument as the mechanism of BP reduction with thiazides is unclear and both volume depletion and fall in systemic vascular resistance play a role. Modest vasodilatation observed with these drugs is more pronounced with chlorthalidone but contribution of this to the long-term BP reduction may not be substantial to change hard clinical endpoints.

Baxdrostat: new drug to treat hypertension

Hypertension is ‘the disease of kidney’, hypothesis that is supported by the fact that one of the most effective antihypertensives exert their effect via kidneys. Here is a welcome addition to the antihypertensive armamentarium —baxdrostat-a highly selective aldosterone synthase inhibitor. Prior efforts to inhibit this key enzyme in aldosterone synthesis were thwarted by concomitant and undesirable inhibition of cortisol synthesis catalyzed by 11 beta-hydroxylase (which share 93% sequence similarity with aldosterone synthase).

In this phase 2 trial-BrigHTN, 248 patients with treatment resistant hypertension were randomized to receive baxdrostat (0.5 mg, 1 mg, or 2 mg) once daily for 12 weeks or placebo.

Treatment resistant HTN was defined as: stable doses of at least three antihypertensive medications (one of which was a diuretic) and had a mean blood pressure of at least 130/80 mm Hg (the average of three measurements obtained with the use of an automated in-office blood-pressure monitor).

Trial was stopped early as criteria for overwhelming efficacy were met-a substantial decrease in blood pressure was noted with treatment: difference between the 1mg group and the placebo group, −8.1 mm Hg, and difference between the 2mg group and the placebo group, −11.0 mmHg, but the difference between the 0.5mg group and placebo was not significant.

Resistant hypertension is associated with high morbidity and mortality; baxdrostat is a welcome addition to our armamentarium to manage this condition. Some caution is needed before we consider it as a breakthrough. First, office BP measurements (primary endpoint) were used to assess the efficacy-why on the earth a Pharma who is able to pioneer a drug development can’t afford to use gold standard 24-hour ABPM to assess efficacy, when it is already used widely in practice? Second, there unexpectedly large placebo effect (9-10mmHg). The reasons for this may be multiple and given the office BP measurements used white coat effect is possible. For the same reason, patients with pseudo resistant HTN might have got enrolled. Factors like adherence to salt restriction, drug adherence, optimization of other antihypertensives might have decreased blood pressure in placebo arm as well as the drug (Hawthorne effect). More elderly patients got randomized to placebo; did it put control arm at disadvantage? This difference will be more relevant in long term, when CV outcomes of these patients will be compared. Phase 3 trials, involving larger sample size, longer follow up, and hard CV outcomes will clarify the pace of this agent in hypertension treatment.

Rise and fall of the sympathetic denervation in the treatment of hypertension

Last few months are of sympathetic overactivity: RADIANCE HTN TRIO trial, with its longer term (6 month) follow up and follow up efficacy data of SIMPLICITY HTN 3 were published back-to-back in JAMA and Lancet.

In patient on triple drug combination of CCB, ARB and thiazide, additional reduction in the BP after renal denervation was modest at 2 months. 136 patients with resistant HTN were randomized to ultrasound guided renal denervation-uRDN-(n=69) or a sham procedure (n=67). Primary outcome of reduction in daytime ambulatory systolic blood pressure was more with intervention than the sham procedure (-8·0 mm Hg [IQR -16·4 to 0·0]vs -3·0 mm Hg [-10·3 to 1·8]; median between-group difference -4·5 mm Hg [95% CI -8·5 to -0·3]; adjusted p=0·022). This difference persisted at six months as documented in this prespecified follow up analysis; however was diminished as compared to that noted at 2 months: mean daytime ambulatory BP at 6 months was 138.3 (15.1) mm Hg with uRDN vs 139.0 (14.3) mm Hg with sham (additional decreases of -2.4 [16.6]vs -7.0 [16.7]mm Hg from month 2, respectively).

Authors, nonetheless, are happy about fewer medication usages (although this difference wasn’t statistically significant at 6 months), and lesser need of aldosterone antagonists in intervention arm (hardly a reason to celebrate).

Renal denervation trials are classic example of the “big bang effect ” of scientific innovations. SIMPLICITY 1 and SIMPLICITY 2 reported a phenomenal reduction in BP over 30mmHg (for a moment elevating blood pressure of hypertension pharma). This was a massive overestimate, confirmed by SIMPLICITY 3 that addressed the key flaw of not having a sham control group in previous studies. We thought of waving a goodbye to renal denervation after these results.

But authors of the simplicity 3 want us to walk back in time and believe that the large reductions in BP reported by previous studies are supported by their 36 months follow up data of simplicity 3 trial published by lancet. The change in 24 h ambulatory systolic blood pressure at 36 months was -15·6 mm Hg (SD 20·8) in the renal artery denervation group and -0·3 mm Hg (15·1) in the sham control group (adjusted treatment difference -16·5 mm Hg [95% CI -20·5 to -12·5]; p≤0·0001). 

There are more than one reasons not believe this tall claim: first, after unmasking at 6 months study didn’t remain a blinded trial and all the patient and physician related biases can cloud the conclusions. As highlighted by the editorial and a letter, Hawthorne effect- change in the behaviour of physicians (better monitoring, frequent reassessments) and patient (better compliance to dietary restrictions and medications) in the intervention arm can explain the apparent difference. Second, despite having uncontrolled hypertension, control arm didn’t receive additional medications to optimize control (did they punish participants refusing to cross over?). Several important questions are raised by this interesting letter as well.

Another confusing feature of this post trial follow up data was the choice of the comparator group, and the way BP measurements were done for this group. After 6 months, a large majority (101 of 171) of the patient crossed over to intervention, however as one might have expected, authors didn’t compare those who received denervation with those who didn’t. Instead, they considered cross over patients (along with those not crossed over) with their BP at 6month considered for final assessment (last observation carried forward). This further adds to difficulty in interpreting the effect of intervention versus placebo and makes assessment of true effect size difficult.

Lost and found theme has created a large number of Bollywood blockbusters, where lost brother or sister is brought back by the director after plot starts getting dull to add life to the story. Producer/director of SIMPLICITY 3 (Medtronic) tries this, but this is hardly enough to bring back this intervention in hypertension practice. There is no easy way to know the contribution of sympathetic over activity in an individual patient with hypertension and if such measures become available, we may be able to better define the role of this treatment. Until then, results of 36 months follow up can be taken with a pinch of NaCl.

Na restriction works to lower BP , may take potassium’s help

BP control doesn’t always need more drugs or interventions like denervation, sometimes smart and simple interventions can yield significant benefits. Two recent studies highlight this point clearly. Salt restriction can prevent age associated increase in blood pressure (age is no more a non-modifiable risk factor), controls BP in those who are already hypertensive and has huge potential to save lives. Salt restriction is the default advice that most hypertensive patients in our clinic receive but it easier said than done. With rapidly changing food environments, adhering to salt restriction is more and more challenging, controlling salt intake in clinical trial can be difficult. These group of investigators from China in their DECIDE-Salt trial played a smart trick.

48 residential elderly care facilities in China (1,612 participants including 1,230 men and 382 women, 55 years or older) were cluster-randomized using a 2 × 2 factorial design to provision of salt substitute (62.5% NaCl and 25% KCl) versus usual salt and to a progressively restricted salt intake (You can fool your taste buds and slow reduction in the salt intake can go unnoticed) versus usual supply of salt or salt substitute for 2 years. Salt substitute compared with usual salt lowered systolic blood pressure (–7.1 mmHg, 95% confidence interval (CI) –10.5 to –3.8), meeting the primary outcome of the trial, whereas restricted supply compared with usual supply of salt or salt substitute had no effect on systolic blood pressure.

This degree of reduction in BP is clinically relevant and is equivalent to adding one antihypertensive. Fewer cardiovascular events were noted in salt substitute group (hazard ratio (HR) 0.60, 95% CI 0.38–0.96), but this had no effect on mortality (HR 0.84, 95% CI 0.63–1.13). As expected, salt substitute increased biochemical hyperkalemia, but this was not associated with adverse clinical outcomes. Results confirm the findings of a previous large SSaSS trial which evaluated effect of this intervention in hypertensive rural Chinese population with past history of stroke. Greater reduction in BP (than SSaSS trial) is attributed to “better modulation” of diet in collective living setting where residents have limited control over the composition of the food they eat.

Important finding of the study was no effect of ‘progressively restricted salt intake’ strategy, which authors attribute to reliance on facility manager and cooks to deliver this (whom residents could influence), and the possibility that residents might have identified the food with restricted salt (buds used to a very high salt intake) and resorted to usual salt intake. Whatever may be the reason, this highlights the difficulty of implementing salt restriction at community level.

Predominantly men and Chinese population, lack of complete follow up data from one particular site, lack of 24-hour urine collection to ascertain the delivery of the intervention are the notable limitations. However, this study is a welcome and important addition to the efficacy and feasibility of sodium restriction as an intervention in hypertension.

STOPACE trial

Medical reversal -where in an established practice/treatment is proved either useless or, sometimes, harmful when subjected to rigorous testing by an RCT- is a regular occurrence in medicine. One classic example of this in nephrology was targeting higher hemoglobin in dialysis patients. This practice was proved harmful after many years when subjected to rigorous scrutiny by well designed RCTs. Stopping ACEi with advancing CKD may be another. This practice was based on hunch of the physicians and observational data . Even clinical practice guidelines are silent on what to do with RASi in advanced CKD. With no guidance and data, it was left to whims and fancies of treating physicians to decide regarding use of RASi in advanced CKD. We now have better evidence that stopping RASI may not be the right thing to do in advanced CKD.

STOP ACEi trial randomized 411 patients with advanced and progressive CKD (GFR<30ml/min) either to discontinue or to continue therapy with RAS inhibitors. The primary end point was the eGFR at 3 years. The difference in eGFR at 3 years was not different in the two groups (difference, −0.7ml/min; 95% CI, −2.5 to 1.0; P=0.42). Other important outcomes like ESKD, initiation of dialysis and CV events were no different in both groups.

The trial results are important in many aspects. It breaks the strongly held dogma that RASi must be withheld in advanced CKD. The trial enrolled real world advanced CKD: median age- 63 years, median serum creatinine-3.4mg/dl, median GFR-18ml/min, 29% patients had GFR<15ml/min, and 37% were diabetics.

The findings of the trial do not support the practice of STOPPING RASi in advanced CKD to improve kidney function. But what about effect of RASi on CV outcomes in advanced CKD? Well, there are no data in advanced CKD patients. Observational data suggests an association between increased CV events after discontinuing RASi.

With this study, it is prudent to continue RASi in advanced CKD and it’s time to move from “hunch” based medicine to evidence-based medicine!

EMPA-CKD trial

After iron and iodine, a day is not far when Flozins will be added to the list of food fortifications. That might be an exaggeration. But after securing a star status in management of DKD and heart failure, flozins are all set to stamp their authority in management of CKD patients.

EMPA CKD randomized 6609 CKD patients to empagliflozin or placebo. Enrolled patients had an eGFR 20 to 45 ml/min/ 1.73 m², or who had an eGFR of 45 – 90 ml/min/ 1.73 m² with a urinary ACR >200(mg/gm). After median 2 years of follow up, progression of kidney disease or death from cardiovascular causes occurred in 432 of 3304 patients (13.1%) in the empagliflozin group and in 558 of 3305 patients (16.9%) in the placebo group (hazard ratio, 0.72; 95% CI, 0.64 to 0.82; P<0.001).

Of the patients enrolled mean age was 63.8 years, non-diabetics were 54% and 35% had eGFR<3oml/min. Serious adverse events were not different in the groups. The patient population is more like what we see in CKD clinics. EMPA CKD trial in addition to DAPA CKD trial defines the role of flozins in CKD. Benefits were not significant in those without albuminuria (was it due to premature termination of the trial?) and in this patient subgroup further evaluation of these agents is needed.

Heterogeneity of response to treatment of hypertensive

One of my patients with CKD is a tailor and his chief complaint in the last clinic visit was his shrinking business in the era of ‘readymade’. Now a days, it is far more convenient to order online or buy readymade trousers from showrooms. I was wearing such a ‘readymade trouser’ in my cousin’s wedding, and everyone was staring at me instead of the newlyweds on the dais. I had just started feeling elated at my youthfulness, when my smarter half (wife) revealed to me that I was appearing a clown. ‘One size fits all’ approach is disastrous whether its clothing or medicine.

Everyone has experienced this phenomenon in our patients with hypertension; some people respond better to some medication than other. In this interesting and intelligently designed study, authors explored this question in a randomized double blind multiple cross over trial.

280 patients (median age 64 yrs, grade 1 hypertension) were randomized: Each participant was scheduled for treatment in random order with 4 different drugs (from 4 different classes): lisinopril, candesartan, hydrochlorothiazide, and amlodipine with repeated treatments for 2 classes. Each treatment period consisted of one week placebo washout, 2 weeks of dose escalation period, and at least 4 weeks of target dose period. 1468 completed treatment periods were evaluated and significant difference in BP responses were observed: Specifically for choices of lisinopril versus HCTZ, lisinopril versus amlodipine, candesartan versus HCTZ and candesartan versus amlodipine. Personalized approach can achieve BP reduction equivalent to half of that expected from single agent in monotherapy, and half of that expected after adding second agent. While some clinical characteristics are often used practically to decide which drug to start with, they are crude at the best. Laragh has proposed individualization based on plasma renin activity and classified drugs acting on renin axis – R drugs (RAS blockers, beta blockers, sympatholytics) or volume axis of hypertension V drugs (duiretics, CCBs, alfa blockers) and proposed to use drug from two difference classes in combination. This study shows that such approach is not only pathophysiologically appropriate but also produce clinically meaningful effect in BP management.

Since that embarrassing incident at my cousin’s marriage, I get all my trousers from my tailor friend, and he is just amazing. Hypertension, makes one of the good cases for personalized medicine-tailoring.