* Correspondence Kamila S. de Freitas Gonçalves, WHO Collaborating Centre for Nursing Research Development, College of Nursing, University of São Paulo, R. Prof. Hélio Lourenço, 3900 ‐ Vila Monte Alegre, Ribeirão Preto 14040‐902, SP, Brazil.
Email: moc.duolci@yrlehsalimak
This is an open access article under the terms of the http://creativecommons.org/licenses/by-nc-nd/4.0/ License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non‐commercial and no modifications or adaptations are made.
The authors confirm that the data supporting the findings of this study are available within the article (and/or) its Supporting Information Materials.
Hypertension (HTN) is a multifactorial chronic disease. Considering the high prevalence rates of this disease, treatment of HTN is necessary, not only to reduce blood pressure (BP) levels but also to prevent the development of cardiovascular, cerebrovascular, and kidney diseases. This treatment can be through medication, which will be determined according to the BP values, obtained either in medical consultations or at home; presence of cardiovascular risk factors, and the presence of target organ damage identified during anamnesis. The aim of this systematic review and meta‐analysis is to summarize the effects of device‐guided slow breathing (DGSB) and nondevice‐guided slow breathing (NDGSB) on BP levels of patients with HTN.
This study is a systematic review and meta‐analysis of randomized clinical trials, pertaining to hypertensive patients, with or without comorbidity, over 18 years old, of both sexes, and with or without hypertensive medication. The selected studies showed comparisons between groups that performed DGSB and/or NDGSB with control conditions. The primary outcome was the value of systolic blood pressure (SBP) and diastolic blood pressure (DBP) after the interventions.
Twenty‐two studies involving 17,214 participants were included in the quantitative analysis. Considerable heterogeneity was revealed between studies. Using random effect model, it was found that DGSB did not significantly reduce SBP and DBP compared to usual care, both in terms BP values and in relation to their variations (SBP, mean difference [MD]: −2.13 mmHg, (95% confidence interval [CI]: −12.71 to 8.44), 288 individuals; I 2 = 93%, high heterogenity: DBP, MD: −0.90, 95% CI: −3.97 to 2.11, 288 individuals; I 2 = 63%, substantial heterogenity. SBP variations MD: −2.42, 95% CI: −7.24 to 2.40, 443 individuals; I 2 = 85% high heterogenity/DBP variations MD: −1.67, 95% CI: −4.57 to 1.24, 443 individuals; I 2 = 80%, high heterogenity).
Based on these results it appears that DGSB did not reduce BP in hypertensive patients and NDGSB is a new path for the future.
Keywords: breathing exercisese, device‐guided breathing, hypertension, physical therapy modalities, resperate, systematic review
Hypertension (HTN) is a multifactorial chronic disease and the main risk factor for the development of cardiovascular diseases (CVDs) and chronic kidney disease. 1 It affects 32% of adults and more than 60% of the elderly, being responsible for half of the deaths from CVD in Brazil. 2 In addition, its complications can lead to decreased work productivity and family income. 1 , 2 In high‐income countries such as Canada, the HTN prevalence has declined; in middle‐income countries, such as Latin America, Asia, the Middle East, and North Africa, detection and treatment of HTN have enhanced, whereas low detection and treatment rates persist in the poorest nations, such as those of sub‐Saharan Africa and Oceania. 1 , 3 , 4
Considering the high prevalence rates of the disease, the treatment of HTN is necessary, not only to reduce blood pressure (BP) levels but also to prevent the development of CVD, cerebrovascular diseases, and kidney diseases. This treatment can be through medication, which will be determined according to the BP values, obtained either in medical consultations or at home, cardiovascular risk factors, and the presence of target organ damage identified during anamnesis.
Nonpharmacological treatment has also been shown to be effective in reducing BP levels in patients with HTN, 1 , 2 , 3 , 4 which includes bodyweight control, establishing healthy eating habits, reducing salt consumption, alcohol consumption control, smoking cessation, stress control, aerobic and isometric physical exercises, and slow breathing guided or not by devices. 1 , 2 , 3 , 4
The physiotherapy prescription for the treatment of HTN may include both exercise and device‐guided slow breathing (DGSB) or nondevice‐guided slow breathing (NDGSB); these breathing exercises consist of slow and deep breathing, 6–10 breathing per minute, and can be performed with or without devices. Concerning the practice of exercises, isometric exercises have been shown to be effective in reducing BP levels, as well as aerobic exercises and dynamic exercises. 4 , 5 , 6 On the other hand, DGSB presents controversies about its application. Since it activates cardiac and pulmonary stretching receptors, decreases sympathetic activity, increases parasympathetic activity and vagal tone, changing heart rate and BP, it would be clinically sound to consider that it reduces BP levels. With the BP reduction, there is an increase in baroreflex sensitivity, which promotes improvements in the autonomic balance of hypertensive patients. 7
The American Heart Association reports that there is no strong evidence on the effectiveness of DGSB, whereas the 8th Brazilian Hypertension Guidelines report the degree of recommendation IIa, level of evidence A. 1 , 4 Already a review of 2016 8 reports that there is currently insufficient evidence of data grouped to recommend the routine use of DGSB in hypertensive patients, even though this device is cleared by the United States Food and Drug Administration and the United Kingdom's National Health Service. Cernes et al. 9 in their review stated that DGSB, as long as it is monitored by a health professional, can be recommended for hypertensive patients who cannot obtain full control of their BP with drug treatment or cannot tolerate potential side effects of treatment. de Barros et al. 10 conducted a controlled clinical study with 15 individuals in the control group and 17 in the experimental group in which they performed DGSB 15–20 min/day, 6–10 breathing/min, and concluded that DGSB, in a long term, did not reduce BP values, catecholamine levels, or muscle sympathetic nerve activity in hypertensive patients. However, this use of DGSB was indicated in the 7th Brazilian Hypertension Guidelines. 11
Recommendations for the use of DGSB or NDGSB in clinical practice should be guided by a systematic, high‐quality literature review. Recently, Chaddha et al. 12 published an article that fulfills this requirement. Their review compared DGSB with NDGSB (pranayama, a technique used in yoga) for 4 weeks in prehypertensive and hypertensive patients. The review included 17 studies, and systolic blood pressure (SBP) was reported in 1017 subjects and diastolic blood pressure (DBP) was reported in 964 subjects. Although interesting, it does not cover only hypertensive patients and compares DGSB to pranayama exclusively. Therefore, a systematic review (SR) of the antihypertensive effects of DGSB or NDGSB applied by physical therapists is necessary to provide the best evidence available to clinical physical therapists and hypertensive patients. In addition, it is also important to summarize the evidence on the effectiveness of the DGSB or NDGSB compared to usual care.
This SR was carried out with the objective of summarizing the effects of DGSB or NDGSB on BP levels of hypertensive patients when: compared with the control conditions (such as minimal intervention, usual care, placebo, and no treatment), compared to other interventions, and used as an adjunct to other treatments (medicated). Thus, the research question for this SR with randomized clinical trials (RCTs) was: What are the effects of prolonged use of device‐guided or NDGSB compared to usual care, on the BP values of hypertensive patients?
This SR was inspired by the recommendations of the Cochrane Handbook of Systematic Reviews 13 and the Preferred Reporting Items for Systematic Review and Meta‐Analysis (PRISMA; see eAddenda for Appendix S1). 14
The selected articles met the inclusion criteria according to the type of study, participants, and intervention for SR.
RCTs published up to January 2020 were included in this SR, without language restriction and year of publication.
Hypertensive patients, with or without comorbidity, over 18 years old, of both sex, with or without hypertensive medication treatment.
Interventions considered had to be DGSB and NDGSB compared to the control conditions (such as minimal intervention—only BP measurement—usual care, placebo, and no treatment); and interventions could be used as an adjunct to other treatments (medication). Any dosage of device‐guided breathing treatment was accepted. Regarding the follow‐up time, 4 and 8‐week studies were considered, and for meta‐analysis, only 8‐week studies were considered (it is the more common time used to treat and reach the BP reduction indicated in studies). 8 , 9
RCTs that also used other interventions along with DGSB/NDGSB, such as physical activity (aerobic exercises, Tai chi, resistance training, and isometric exercises), salt reduction and salt substitution, stress control techniques that use other types of deep breathing with meditation (e.g., Qigong, Yoga, progressive muscle relaxation and attention‐based stress reduction programs), dietary (dietary approach to stop hypertension, low‐carbohydrate diet, Mediterranean diet, high‐protein diet, low‐fat diet, vegetarian diet, paleolithic diet, and low index glycemic/load) and lifestyle (comprehensive lifestyle modification, smoking cessation, alcohol restriction, sleep, home heating, and weight loss) were excluded since it was not possible to identify the specific effect of DGSB/NDGSB.
The primary outcome was the values of SBP and DBP, expressed in mmHg, reached after the interventions, as well as their variations.
The secondary outcome was a reduction in the quantity/dosage of drugs administered to HTN control if the study subjects also used it.
A systematic search of all published RCTs on the effects of device and NDGSB on hypertensive patients, without language restriction, was carried out until January 2020 in nine databases: Pubmed/MEDLINE (Medical Literature Analysis and Retrieval System Online), Latin American and Caribbean Health Sciences Literature (LILACS), EMBASE, Cochrane Central Register of Controlled Trials (CENTRAL), Physiotherapy Evidence Database (PEDro), Cumulative Index to Nursing and Allied Health Literature (CINAHL), Scopus, Web of Science, Livivo, as well as searching clinical trial records databases, CT.GOV (Clinical Trials.Gov), and bases for gray OpenGrey literature, Gray Literature Report, and ProQuest Central (Citation, Abstract or Indexing and Dissertations and Theses). In all of these databases, potentially eligible studies were researched, including completed and ongoing RCTs, until January 2020. The complete search strategy used in PubMed/MEDLINE is shown in Appendix S2 (see eAddenda for Appendix S2).
Two reviewers independently analyzed all titles and abstracts retrieved with the search. When there was agreement on a particular record, the study was analyzed in full text by both reviewers, according to the eligibility criteria. In the presence of disagreement between the reviewers, a third reviewer was convened. When additional information was needed, authors of the potentially eligible studies were contacted.
Two reviewers independently extracted the following data from the included trials: author, publication date, country of publication, study type, sample size, participant characteristics (age, gender, use or not of antihypertensive medications, presence of comorbidities, categories of BP, details of intervention (type of device used in the DGSB—whether DGSB was performed with or without load, or how the NDGSB was performed, breaths per minute for DGSB and NDGSB, time of use of the device in a day, and for how many months), details for BP measurement (device used, type of measurement (home or office), protocol used for measurement including preparation), and outcome measures (systolic and diastolic BP). A third reviewer was called in case of disagreement. When necessary, the authors of RCTs included were contacted to provide additional information.
The quality assessment of the included studies was conducted using the Cochrane Risk of Bias Tool for Randomized Trials (RoB2), 15 which includes a randomization process, deviations from the intended interventions, conflicting result data, result measurement, selection of the reported result, and biases generally. The same two reviewers performed an independent assessment. Disagreements between reviewers were resolved by discussion and, if necessary, the opinion of a third reviewer was requested. The same two reviewers performed data extraction, using standardized forms regarding the methodological characteristics of the studies, interventions, and results. Disagreements were again resolved by discussion and, if necessary, the opinion of a third reviewer was requested.
All data from continuous variables referring to BP values in mmHg were synthesized using the mean difference (MD) method, with their respective 95% confidence intervals (CIs). Standard deviations (SDs) for analysis were also extracted.
The effects of interventions on BP values were analyzed separately. The data were evaluated according to the type of intervention (DGSB or NDGSB); however, only studies lasting at least 8 weeks were considered for meta‐analysis (results evaluated after 8 weeks of randomization). Whenever possible, study results, where there was an intention‐to‐treat analysis, were used.
The presence of statistical heterogeneity between RCTs was assessed using the I 2 statistic. The quality of the evidence was considered inconsistent if considerable heterogeneity between the groups (I 2 > 50%) was observed. When sufficient evidence is available, a funnel plot could be used to investigate possible publication bias.
The overall quality of the evidence for each outcome was assessed using the Grading of Recommendations, Development and Evaluation (GRADE) 16 system, regardless of whether or not the information was sufficient to summarize the data in quantitative analysis.
The quality of the evidence was categorized as follows: the evidence was of high quality if the results were consistent in ≥75% of the participants, with a low risk of bias, without publication bias, and with consistent direct and accurate data; further research is unlikely to alter the estimate or confidence in such results. The evidence was of moderate quality when only one of the five classification factors above was met; further research can alter the estimated effect and impact on confidence in the effect in this case. The evidence was of poor quality when two of the five classification factors were not met. In this situation, future research is likely to alter the estimated effect and have a significant impact on confidence in the effect. The evidence was of very low quality when three of the five classification factors were not met and, in this case, any estimate of effect is uncertain. 16
The statistics commonly used for meta‐analysis of continuous data are the MD or the standardized mean difference (SMD). Selection of summary statistics for continuous data is determined by whether studies all report the outcome using the same scale (when the MD can be used), as SBP and DBP, or using different scales (when the SMD is usually applied). For the MD approach, the SDs are used together with the sample sizes to compute the weight given to each study. Studies with small SDs are given relatively higher weights while studies with larger SDs are given relatively smaller weights. If the heterogeneity will be present, a CI around the random‐effects summary estimate is wider than a CI around a fixed‐effect summary estimate. This will happen whenever the I 2 statistic is greater than zero, The RevMan will be used for these analyses. 13