“Intensive vs. Standard BP control?”

Journal Club

27 July 2016

Facilitated by Loai Albarqouni

1.  Background

A new guidance for the diagnosis and management of hypertension in adults has been recently issued in Australia by the National Heart Foundation. The landmark change in the guidance is the recommended lower BP target in a selected high cardiovascular risk population (<120 mmHg instead of the previous <140 mmHg). This recommendation was largely based on the SPRINT trial which was published by the end of last year.

 2.  Paper presented

A Randomized Trial of Intensive versus Standard Blood-Pressure Control.  By the SPRINT research group. N Engl J Med 2015; 373:2103-21161

Design: Open-label randomised controlled trial (ClinicalTrial.gov: NCT01206062)

Research Question: In adult non-diabetic patients (≥ 50 years) with a systolic BP between 130-180 mmHg and an increased CV risk, does intensive BP treatment (<120 mmHg) compared to standard BP treatment (<140 mmHg) improve clinical outcomes (MI, ACS, HF, or CV-related mortality)?

PICO

P: Adult patients (≥ 50 years) with a systolic BP between 130-180 mmHg and an increased CV risk. Patients with Diabetes and those have a previous history of stroke were excluded.

I: Treatment algorithm to target patient’s SBP less than 120 mmHg.

C: Treatment algorithm to target patient’s SBP between 135-139 mmHg (<140 mmHg).

O: Primary outcome: the composite outcome of myocardial infarction, acute coronary syndrome, stroke, heart failure, or death from cardio­vascular causes. Secondary outcomes: individual components of 1ry and death from any cause. Renal outcomes as well as adverse events were also reported.

Follow-up: Median 3.26 years. Planned for a mean follow-up of 5 years but stopped early for benefit.

Setting: 102 clinical sites in the US.

Funding: NIH, NHLBI, the National Institute of Diabetes and Digestive and Kidney Diseases, the National Institute of Neurological Disorders and Stroke, and the National Institute on Aging.

Study Appraisal

This article was appraised using the RAMbo method:

Randomisation: The study is an open-label randomised trial, but details about the randomisation and allocation process are not clearly described in the main article. In the protocol, authors reported that an internet-based web browser randomisation procedure was used, which implies adequate randomisation and allocation concealment. Additionally, Table 1 shows comparable baseline characteristics were achieved between the randomisation groups (no significant differences between groups). We can say that the randomisation process seems to be adequately addressed.

Attrition: The percentage of the patients those lost to follow-up was comparable in both groups (about 10%) and patients were analysed according to the randomisation groups (intention to treat analysis).

Overall, both groups were treated equally apart from the intervention, but we are a little bit concerned about the reported discrepancies in the utilisation of antihypertensive medications classes between the two groups which might affect the observed difference in the outcomes. In other words, it arguable whether the observed benefit in the intensive group was solely resulted from the lower BP target or could be resulted also from the difference in the utilisation of antihypertensive medications2.

Follow-up period is adequate but we discussed also the decision to stop the trial earlier (see below). We thought that it would be better if they continued the trial to adequately observed the full spectrum of potential adverse events resulted from the intensive treatment strategy.

Overall, the follow up was well maintained and adequate.

Measurement (done by blinded assessors, or were objective outcomes):

Because of the nature of the intervention, treatment groups and treating clinicians could not be blinded which might lead to performance and detection bias, however, thought to be minimal and negligible, if any. In addition, the outcomes were “partly” objective and adjudicated by a committee whose members were unaware of the study group assignments.

In general, we think there might be some sources of bias but these are minimal and did not affect the internal validity of the study.

3.  Summary of results

During the study, 9361 eligible patients were randomised to the treatment groups and followed for a median of 3.26 years. Table 1 in the article shows the baseline characteristics of the included patients:  Mean age was 68 years old and 64% were men. The average Framingham 10-yr CVD risk score was 20.1 (±11) with about 39% of the patients have 10yr CVD risk score < 15% (intermediate risk).  About 50% of the patients on Aspirin and less than half of them (43-45%) were on statins. The average baseline systolic BP was 139.7 (±15) mmHg, with about two third of them started the trial with systolic BP less than 145 mmHg.

The main finding of the trial was the lower rate of the primary outcome (cardiovascular events including CVD mortality) in the intensive group compared to standard group (1.65% per year vs. 2.19% per year; HR= 0.73[0.6-0.9]; p=0.003). Table below shows the results of the trial (primary outcomes, secondary outcomes, renal and adverse events) expressed in relative and absolute measures.

screenshot

4.  Discussion

A serious adverse event was defined as an event that was fatal or life-threatening, that resulted in clinically significant or persistent disability, that required or prolonged a hospitalization, or that was judged by the investigator to represent a clinically significant hazard or harm to the participant that might require medical or surgical intervention to prevent one of the other events listed above. ARD: Absolute Risk Difference, NNT/NNH: Number needed to treat or harm.

The presented article is an interesting example of high quality well-conducted randomised controlled trial. The authors concluded that in adult non-diabetic patients 50 years or older with an increased CV risk, targeting a systolic BP of less than 120 mmHg, as compared with less than 140 mmHg, resulted in lower rates of fatal and nonfatal major cardiovascular events and death from any cause, although significantly higher rates of some adverse events were observed in the intensive-treatment group.

Despite the high internal validity of the trial, we have discussed few issues that could affect the validity of the results (basically the external validity of the results):

  1. Measurements of blood pressure: the author stated that “Dose adjustment was based on a mean of three blood-pressure measurements at an office visit while the patient was seated and after 5 minutes of quiet rest; the measurements were made with the use of an automated measurement system”, however, this standardised procedure is rarely practised in most of the GPs and clinics, so there could be a difference of about 10 mmHg between BP readings obtained in routine clinical practice and those observed in the trial (e.g. the observed 121.4 mmHg mean BP in intensive group might be equivalent to 130-135 mmHg in routine clinical practice)3,4. [20038756] [27334926]
  2. We discussed also the dose titration algorithm followed in “standard” treatment group, since author reported that in patients in standard-treatment group, the dose of medication was reduced if systolic BP went below the respecified target (130 mmHg on single visit or 135 mmHg in two consecutive visits). We doubted that this the standard way of treating hypertensive patients as the name imply.
  3. The authors stated that the data and safety monitoring board of the trial recommended that the trial should be stopped early for the overwhelmed benefits. Although it is a common practice to stop the trial early in case the benefits or more importantly the harms were significantly higher in one side, it can lead to an overestimated risk estimates. In a systematic review5 [20332404] comparing the results of 91 truncated trials (stopped early) vs. 424 matching non-truncated RCTs (not stopped early) asking the same questions, the relative risk in truncated RCTs found to be 29% (23-35%) smaller than non-truncated RCTs (overestimated). In other words, if we applied that to the results of the primary outcomes, the results will become insignificant: Hazard ratio 0.95, 95CI% (0.81-1.13).
  4. We also highlighted that even in the controlled closely-monitored environment; target blood pressure of less than 120 mmHg was only achieved in less than half of the patients and after one/or two year of intensive closely-monitored treatment algorithm (on average, patients in intensive group had used one more antihypertensive medications than those in standard [3.0 vs. 1.9]). This reflects the challenging task of achieving this targets in real clinical practice.
  5. We also extensively discussed the resulted benefits and harms reported in this trial. We agreed that the beneficial outcomes of the intensive treatment should be cautiously interpreted in the light of the substantial observed adverse events (e.g. acute kidney injury and serious adverse events).
  6. The new national guidance on Hypertension:  

A new guidance on the management of hypertension in adults was issued by the Australian national heart foundation, which recommended a paradigm shift of the systolic BP target of the treatment from the standard less than 140 mmHg to intensive 120 mmHg for a selected high CVD risk group of patients. The guideline development panel should be applauded for their evidence-based approach and the use of the appropriate evidence to reach the recommendations. Authors have clearly specified the subgroup of patients for whom intensive treatment might be beneficial (non-diabetic patients with an increased risk of CVD), and highlights the blood pressure measurement variations between techniques used in SPRINT trial and those commonly used in clinics and how challenging will it be to reduce the BP to the new levels. However, we have a few comments on the guideline recommendation relevant to the newly recommended low target BP.

In the summary of the recommendations, which is also the most commonly read section of the guideline, the authors give a strong (level II) recommendation advising clinicians to adopt the new BP target: “In selected high cardiovascular risk populations where a more intense treatment can be considered, aiming to a target of <120 mmHg systolic blood pressure can improve cardiovascular outcomes”. According to the GRADE recommendation definitions that the guideline has followed, strong recommendation denotes “Benefits clearly outweigh drawbacks/harms”. However, in the light of the evidence from SPRINT trial, we thought that benefits and harms are very close to each other’s, if harms do not outweigh benefits (63 patients needed to be treated intensively over 3.5 years to reduce one CV event while treating 43 patients intensively can lead to one serious adverse events, which is fatal or life-threatening).

The guideline panels have briefly reported that an informed shared-decision process should be considered, however, the guideline failed to report a detailed information about how to integrate the patients’ preferences in the decision process at point-of-care, which is vitally important in this scenario: a preference-sensitive scenario. This is commonly seen in guidelines, as Chong et al reviewed 51 evidence-based CPG and found that only 5% of the word count and 6% of references in the guidelines related to patients’ preferences6 [19387746]. One of the basic suggestions is to clearly articulate the “preference sensitive” nature of the recommendation. In addition, guideline should clearly communicate both the benefit and harms of the intervention to the clinicians and patients in a way that they can make a balanced informed decision. Moreover, guideline should also provide practical tools to the clinicians and patients to enhance the shared-decision making process, for instance, a decision support tool might be helpful in this scenario (see below for an example of a simple decision tool based on the evidence from SPRINT trial)7.

So overall, this is a clear example of a preference-sensitive situation where a balanced informed shared decision making approach is essential.Pic 1

Pic 2

References

  1. Sprint Research Group., Wright JT, Jr., Williamson JD, et al. A Randomized Trial of Intensive versus Standard Blood-Pressure Control. The New England journal of medicine. 2015;373(22):2103-2116.
  2. Fretheim A, Odgaard-Jensen J, Brors O, et al. Comparative effectiveness of antihypertensive medication for primary prevention of cardiovascular disease: systematic review and multiple treatments meta-analysis. BMC medicine. 2012;10:33.
  3. Myers MG, Godwin M, Dawes M, Kiss A, Tobe SW, Kaczorowski J. Measurement of blood pressure in the office: recognizing the problem and proposing the solution. Hypertension. 2010;55(2):195-200.
  4. Schiffrin EL, Calhoun DA, Flack JM. Do We Need a New Definition of Hypertension After SPRINT? American journal of hypertension. 2016.
  5. Bassler D, Briel M, Montori VM, et al. Stopping randomized trials early for benefit and estimation of treatment effects: systematic review and meta-regression analysis. Jama. 2010;303(12):1180-1187.
  6. Chong CA, Chen IJ, Naglie G, Krahn MD. How well do guidelines incorporate evidence on patient preferences? Journal of general internal medicine. 2009;24(8):977-982.
  7. Images created by Iconarray.com. Risk Science Center and Center for Bioethics and Social Sciences in Medicine University of Michigan.