Browse the corpus

Elevated blood pressure (BP), or hypertension (i.e. systolic BP ≥ 140 mmHg or diastolic BP ≥ 90 mmHg, or both) is independently associated with diabetes mellitus, and both conditions substantially increase cardiovascular disease risk (e.g. stroke, heart failure, myocardial infarction), chronic kidney disease, and all‐cause mortality [, , , ]. Globally, 589 million (one in nine) adults aged 20 to 79 years are living with diabetes mellitus [], and hypertension is frequently detected in an estimated 70% to 80% of people with diabetes mellitus []. The concurrence of hypertension and diabetes mellitus may accelerate the development and progression of macrovascular (i.e. injury to the large blood vessels of the heart or brain due to a blockage or stoppage in blood flow, e.g. myocardial infarction, stroke) and microvascular (i.e. damage to the small blood vessels of the kidneys or eyes, e.g. nephropathy, retinopathy) complications [, , , ]. Consequently, lowering BP is important for reducing cardiovascular morbidity and all‐cause mortality among individuals with elevated BP and diabetes mellitus [, , , , ].

Using antihypertensives to lower BP in people with diabetes mellitus and hypertension reduces the risk of cardiovascular complications [, , ]. Antihypertensives may reduce vascular stress (i.e. physical strain and injury to blood vessels) and endothelial dysfunction (i.e. damage to the thin layer of cells lining the blood vessels in the body), along with mitigating activation of the renin‐angiotensin‐aldosterone system which regulates BP and the volume of the blood, oxidative stress (i.e. an imbalance that leads to damage to the cells in the body), and systemic inflammation (i.e. a continual state of defence by the immune system) [, ]. A central question in treating hypertension for people living with diabetes is how low BP should be to achieve clinical benefit. In clinical practice, a target BP is used as the goal of antihypertensive therapy and guides treatment decisions regarding the intensity of the antihypertensive regimen []. Lower systolic BP targets (e.g. < 120 mmHg) have been proposed for people with diabetes mellitus, who are at increased risk of cardiovascular disease and chronic kidney disease []; however, evidence on the clinical benefits of such targets compared with standard targets (e.g. <140 mmHg) remains inconsistent [, , , ].

The goal of lower BP targets would be a further reduction in cardiovascular disease risk and renal complications, compared with a standard BP target. However, achieving lower BP targets may require higher doses or a greater number of antihypertensive medications, or both, potentially increasing treatment costs and medication burden. Excessive BP lowering may also increase the risk of adverse cardiovascular outcomes [, , , , , , , , ].

Clinical practice guidelines recommend lowering BP in people with elevated BP and diabetes mellitus; however, the optimal BP targets in this population remain uncertain [, , , , , , , , , , , ]. Lower BP targets have been recommended for individuals with elevated BP and diabetes mellitus because of their increased risk of cardiovascular disease and kidney disease outcomes [, ]. For example, guidelines, including those from Diabetes Canada, the European Society of Cardiology, and the American Heart Association/American College of Cardiology, recommend systolic BP targets below 130 mmHg in people with diabetes mellitus, particularly among individuals at high cardiovascular risk, including those with cardiovascular disease, subclinical organ damage, or chronic kidney disease [, , ]. In prior Cochrane reviews and meta‐analyses of randomised controlled trials (RCTs), treatment with BP targets < 130/85 mmHg among people with hypertension did not reduce mortality or cardiovascular morbidity compared with standard BP targets (systolic BP < 140 to 160 mmHg and diastolic BP < 90 to 100 mmHg). Similarly, sensitivity analyses in people with diabetes mellitus provided little to no evidence of differences for lower BP targets, compared to standard BP targets [, ].

Recent systematic reviews and meta‐analyses report heterogeneous findings regarding the effects of intensive BP lowering in people with diabetes mellitus. While some reviews suggest that lower systolic BP targets may reduce the risk of cardiovascular mortality, major cardiovascular events, including fatal and non‐fatal stroke [, , , ], evidence regarding all‐cause mortality, chronic kidney disease, and other cardiovascular events remains inconsistent [, , , ]. Current evidence also suggests that lowering systolic BP to between 130 and 140 mmHg may provide cardiovascular benefit, while intensive lowering below this threshold may offer limited additional benefit [, ]. Given the continued uncertainty regarding optimal BP targets in people with elevated BP and diabetes mellitus, this review aims to investigate the benefits and harms of lower BP targets (≤ 130/85 mmHg), compared with standard BP targets (systolic BP ≤ 140 to 160 mmHg and diastolic BP ≤ 90 to 100 mmHg). In addition, this review will complement related reviews evaluating BP targets in adults with hypertension [], chronic kidney disease [], and cardiovascular disease [].

To determine if lower blood pressure (BP) targets (i.e. ≤ 130/85 mmHg), compared with standard BP targets (i.e. ≤ 140 to 160/90 to 100 mmHg) are associated with reduction in mortality and morbidity in people with elevated BP and diabetes mellitus.

We will consider only RCTs (i.e. randomisation at the individual level), and include published trials, conference abstracts, and other unpublished trials, irrespective of the language of publication. Cross‐over trials will not be eligible for inclusion due to the possibility of carryover effects distorting effect estimates, as BP target interventions are likely to have sustained effects.

We will include non‐pregnant and non‐lactating adults (≥ 18 years) with diabetes mellitus and elevated BP (i.e. systolic BP ≥ 140 mmHg or diastolic BP ≥ 90 mmHg, or both) [], documented in a standard way on at least two occasions, or already receiving treatment for elevated BP or treatment for nephropathy that includes BP‐lowering drugs. We will include trials if people with elevated BP diabetes mellitus were randomised to one of the two BP targets described below, irrespective of their BP at enrolment, any concomitant disease including baseline cardiovascular risk, or geographical location. We will consider studies that include a subset of eligible participants for inclusion. If data for the eligible subgroup are not reported separately, we will attempt to contact the corresponding author to retrieve relevant data, including age‐stratified data for adults (≥ 18 years), if applicable. If trial authors are unable to provide subgroup‐specific data, or correspondence is not successful, we will categorise such studies as awaiting classification [].

We will include trials of individuals randomised to a lower (≤ 130/85 mmHg) compared with a standard (≤ 140 to 160/90 to 100 mmHg) systolic/diastolic BP target. We will include trials irrespective of dose, type, duration, frequency, delivery intensity, or route of administration of medications administered as part of the intervention. For trials that report mean arterial blood pressure (MAP), a MAP < 107 mmHg, 102 mmHg, or 98 mmHg will be considered similar to systolic/diastolic BP < 140/90 mmHg, 135/85 mmHg, or 135/80 mmHg, respectively []. Studies that include co‐interventions (e.g. antihypertensive or antidiabetic medications) will be eligible if the same co‐intervention is administered in both the intervention and comparison group.

Number of people with at least one serious adverse event (i.e. any unfavourable medical occurrence that is considered serious, is life‐threatening/results in death, requires hospitalisation or prolongation of hospitalisation, or results in significant disability []). Total cardiovascular serious adverse events.

Withdrawals due to adverse effects (i.e. number or proportion of participants who discontinued treatment due to adverse effects). Number of antihypertensive drugs needed per participant (i.e. average number of antihypertensive medications required per participant in the intervention and control groups).

For all‐cause mortality, number of people with at least one serious adverse event, total cardiovascular serious adverse events, and withdrawals due to adverse effects, we will extract cumulative events occurring at any time from randomisation to the primary endpoint of the trial. For systolic BP achieved, diastolic BP achieved, and number of antihypertensive drugs per participant, where outcomes are assessed at multiple time points, we will use the outcome measurement at the primary endpoint of the trial.

We will search the following databases without language or publication status restrictions: Cochrane Hypertension Group Specialised Register via the Cochrane Register of Studies (CRS‐Web); Cochrane Central Register of Controlled Trials (CENTRAL) via the Cochrane Register of Studies (CRS‐Web); US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (www.clinicaltrials.gov); World Health Organization International Clinical Trials Registry Platform (WHO ICTRP) (apps.who.int/trialsearch).

The Cochrane Hypertension Information Specialist (CIS) will model the search strategies for databases on the core search strategy designed for MEDLINE. Where appropriate, the CIS will combine the subject strategy adaptations with the sensitivity and precision‐maximising search strategy designed by Cochrane for identifying RCTs (as described in the Cochrane Handbook for Systematic Reviews of Interventions ) []. We will run searches of MEDLINE and Embase from 2013, to capture trials published since the last published version of this review []. As records from those databases, as well as from ClinicalTrials.gov and WHO ICTRP, are regularly added to CENTRAL and the Hypertension Specialised Register, we will search the other databases listed above from 2023 to present. We will also incorporate results from a broader, related review on BP targets not limited to people with diabetes that was updated in 2020 and which is currently being updated in parallel with this review []. The MEDLINE search strategy is presented in .

We will check the reference lists of included studies and any relevant systematic reviews identified for further references to relevant trials. We will check the included studies for retractions and errata via PubMed (www.ncbi.nlm.nih.gov/pubmed) and the Retraction Watch Database (retractiondatabase.org), report the search dates in the review, and exclude retracted studies. We will search Epistemonikos (www.epistemonikos.org) for related systematic reviews. We will attempt to contact authors of relevant papers regarding any further published or unpublished work. We will attempt to contact authors of trials reporting incomplete information to provide the missing information. We will search Clarivate Web of Science for papers that cite studies included in the review.

Three review authors (DL, GG, DS) will independently evaluate trial eligibility via title and abstract screening using Covidence []. The same three review authors will independently perform full‐text screening, with any disagreements resolved by discussion with the senior review author (WT). If limited information is reported on the study design, or eligible studies are available as conference abstracts only, we will attempt to contact trialists to request additional study information. We will report records identified, included, or excluded, in addition to the reasons for trial exclusion, in the PRISMA 2020 flow diagram [].

Three authors (DL, GG, DS) will independently extract data from the included trials using a data extraction form that we will pilot‐test with at least two included studies. If available, we will extract information on study characteristics (e.g. trial funding, study duration, study location/setting, publication date, inclusion criteria, and exclusion criteria), study methods (e.g. randomisation methods, allocation concealment, sample size), study participants (e.g. age, gender/sex, PROGRESS‐Plus characteristics (e.g. socioeconomic status, race/ethnicity/culture/language, geographical location) [], baseline BP, existing comorbidities, number of participants randomised, number of participants lost to follow‐up/withdrawn), intervention (e.g. lower, intensive, or tight BP targets (e.g. systolic BP < 120 mmHg)), comparator (e.g. normal, usual, or standard BP targets (e.g. systolic BP < 140 mmHg)), and outcome data. If applicable, we will contact trial authors to retrieve missing or additional study information. Any discrepancies will be resolved by discussion among review authors or by recourse to the senior review author (WT). One review author (DL) will enter all study data into RevMan software [], which a second review author (GG) will verify.

Two of three review authors (DL, GG, DS) will independently assess risk of bias in the included studies using Cochrane's RoB 1 tool, following the methods described in the Cochrane Handbook for Systematic Reviews of Interventions []. We will assess the risk of bias based on the following risk of bias domains: sequence generation, allocation concealment, blinding (participants, personnel, caregivers), incomplete outcome data, selective outcome reporting, and other sources of bias. We will contact trial authors to retrieve relevant missing information or for clarification. We will assess methodological quality across all six risk of bias domains and assign an overall judgement of low, high, or unclear risk of bias. We will use Microsoft Excel to document the risk of bias judgements []. Any disagreements between review authors will be resolved by the senior review author (WT) or by recourse to a third review author (JW).

We will report dichotomous data (i.e. all‐cause mortality, number of people with at least one serious adverse event, total cardiovascular serious adverse events, withdrawals due to adverse effects) using risk ratio (RR) and 95% confidence intervals (CIs). When there is a statistically significant difference between treatments for any outcome, we will estimate the absolute risk reduction (ARR) and absolute risk increase (ARI). In order to estimate the number of patients needed to treat to provide one additional benefit or to produce one additional harm, we will calculate the number needed to treat for an additional beneficial outcome (NNTB) and the number needed to treat for an additional harmful outcome (NNTH), respectively.

For continuous outcomes (i.e. systolic BP achieved, diastolic BP achieved, number of antihypertensive drugs needed per participant), we will extract the mean and standard deviation (SD) []. If effect measures are not reported on the same scale, we will convert data into the required direction and format to approximate the mean and SDs. If outcomes are measured on the same scale, we will summarise data using mean difference (MD) and 95% CIs. If different scales are used to measure the same outcome, we will estimate the standardised mean difference (SMD) and 95% CIs. If applicable, for trials with multiple time points, we will consider the assessment time point immediately after trial completion [].

In included trials with more than two intervention groups, we will only include arms that meet the a priori inclusion criteria. If intervention arms are sufficiently similar, we will form a single pair‐wise intervention group []. We will report all intervention groups in the table of 'Characteristics of included studies' table.

We will attempt to contact trial investigators for missing information or clarification as needed, including for unpublished data or studies available as abstract only []. We will contact the corresponding author via email, sending two email reminders if needed. If we receive no response, or trial investigators are unable to provide the missing information, we will compute summary data from relevant statistics reported. In cases where we are unable to retrieve data or compute data appropriately for an outcome of interest, and we determine that this leads to a high risk of bias, we will report the extent of missingness and investigate its effect on the certainty of evidence. We will also address the implications of missingness on the findings of the review in the 'Discussion' section.

To assess selective reporting bias, three review authors (DL, GG, DS) will independently search trial registries, associated study protocols, and clinical study reports to evaluate potential discrepancies in the methods or outcomes of included studies, if applicable. If at least 10 studies are included in the meta‐analysis, we will use funnel plots to assess the risk of publication bias. We will investigate funnel plot asymmetry through visual inspection and conduct the Egger's test (P values < 0.10 to be identified as statistically significant) where appropriate to explore the presence of small‐study effects [].

To test for heterogeneity of treatment effect between trials (i.e. statistical heterogeneity), we will use the Chi² and I² statistics []. We will consider a Chi² value less than 0.05 or an I² value greater than 50% as indicative of significant statistical heterogeneity. We will examine methodological heterogeneity by assessing differences in study design and risk of bias across included studies. We will evaluate clinical heterogeneity by considering diversity in participant characteristics, interventions and comparators, outcome definitions, and outcome assessment time points. Where substantial statistical, clinical, and/or methodological heterogeneity, is identified, we will explore potential reasons for it through sensitivity and subgroup analyses. We will use critical outcomes to perform the following subgroup analyses including at least 10 trials, irrespective of the risk of bias, when possible. BP targets: systolic BP achieved versus diastolic BP achieved. Age: < 65 years versus ≥ 65 years versus not reported. Duration of elevated BP/hypertension: < 10 years versus ≥ 10 years versus not reported. Number of antihypertensive medications: < 5 versus ≥ 5. Duration of diabetes mellitus: < 10 years versus ≥ 10 years versus not reported. Sex: male versus female versus mixed versus not reported.

For subgroup effects examined statistically, we will assess differences between subgroups using the formal test for subgroup differences available within RevMan and report the Chi² value, P value, and interaction test I² statistic, to determine whether these effects would lead to differing clinical interpretations [].