Abstract
This study aimed to determine the effects of inspiratory muscle training (IMT) on exercise capacity, respiratory muscle strength, length of hospital stay (LOS), and quality of life (QOL) following coronary artery bypass graft surgery.
The search was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses and the Cochrane Handbook and included the databases MEDLINE, EMBASE, CINAHL, Scopus, and CENTRAL. The review included randomized controlled trials utilizing IMT during phase 1 or 2 postoperative cardiac rehabilitation (PoCR) versus alternative treatment (active or passive control) in patients following coronary artery bypass graft surgery.
Fifteen studies were included (11 phase 1 studies, 4 phase 2 studies) with no reported adverse events. In phase 1 PoCR, IMT reduced the LOS (−1.02 days; 95% CI = −2.00 to −0.03) and increased exercise capacity (6-minute walk distance) (+75.46 m; 95% CI = 52.34 to 98.57), and maximal inspiratory pressure (MIP) (10.46 cm H2O; 95% CI = 2.83 to 18.10), but had no effect on maximal expiratory pressure. In phase 2 PoCR, IMT increased 6-minute walk distance (45.84 m; 95% CI = 10.89 to 80.80), MIP (−23.19 cm H2O; 95% CI = −31.31 to −15), maximal expiratory pressure (20.18 cm H2O; 95% CI = 9.60 to 30.76), and QOL (−11.17; 95% CI = −17.98 to −4.36), with no effect on peak oxygen uptake. There was a high risk of bias for MIP (75% of the phase 1 studies) and 6MWT (1 of 4 phase 2 studies). The quality of the evidence ranged from very low to moderate.
IMT significantly improves exercise capacity, respiratory muscle strength, LOS, and QOL in phase 1 and 2 PoCR.
IMT may benefit patients during phase 1 and 2 of PoCR, considering the safety, low cost, and potential benefits.
Introduction
Coronary artery bypass graft (CABG) surgery is a well-established intervention to treat patients with chronic coronary disease, improving several clinical parameters, including survival and quality of life (QOL)1–3 but despite today’s advanced technologies, it may still be associated with cardiac, cerebrovascular,4 and pulmonary5 complications in the postoperative period. The etiology of pulmonary impairments after cardiac surgery includes several causal factors such as sternotomy, pleurotomy,6 cardiopulmonary bypass, topical cooling for myocardial protection,7 and chest drains (especially intercostal)8 and reduced respiratory muscle strength,9 which has been associated with poor functional capacity10 all of which lead to a prolonged recovery period.11,12
Inspiratory muscle training (IMT) has been found to be beneficial in several populations, including heart failure,13 major abdominal surgery,14 and chronic obstructive pulmonary disease.15 A recent guideline from the American Physical Therapy Association has recommended that IMT either alone (quality of evidence: I; recommendation: A [strong]) or combined with aerobic exercise (AE) (quality of evidence: II; recommendation: B [moderate]) be provided to patients with heart failure.16
Moreover, according to previous systematic reviews,14,17 IMT has been recognized as an important intervention during preoperative cardiac, lung, and abdominal surgeries with a moderate risk of bias to improve pulmonary complications, length of hospital stay (LOS), intensive care unit duration, inspiratory muscle performance, and pulmonary function. The other previously mentioned reviews18,19 included studies with the use of IMT in the preoperative period, not providing scientific evidence for patients who only use it in the postoperative period since the load and protocol of training are different from the preoperative period. However, to the best of our knowledge, there is limited evidence examining the effects of IMT, specifically during the postoperative period after cardiac surgery, especially since most patients will not have the opportunity to perform IMT in the preoperative period as recommended.18,20 Two systematic reviews21,22 have examined the effects of postoperative IMT, but did not examine several critically important postoperative factors, including the effects of IMT during inpatient (phase 1) versus outpatient (phase 2) periods, the methods of IMT training, additional outcome measures such as QOL and on peak oxygen uptake (peak Vo2, in milliliters of oxygen per kilogram per minute (mL/kg·min−1) (phase 2), and certainty of the evidence assessment.21,22
Therefore, the present systematic review with meta-analysis was performed to determine the effect of the IMT on exercise capacity, respiratory muscle strength, LOS, and QOL in patients following CABG surgery in phase 1 and 2 of postoperative cardiac rehabilitation (PoCR). We also examined IMT methods, the risk of bias, and the quality of the evidence.
Methods
Design
The systematic review was registered with PROSPERO (No. CRD42021277225) and reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses and the Cochrane Handbook.23 The AMSTAR checklist was also utilized to assess the overall quality of the systematic review24 (Fig. 1).
Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) flowchart. *Consider, if feasible to do so, reporting the number of records identified from each database or register searched (rather than the total number across all databases/registers). **If automation tools were used, indicate how many records were excluded by a human and how many were excluded by automation tools. IMT = inspiratory muscle training. Adapted from: Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:71. doi:http://dx.doi.org/10.1136/bmj.n71. For more information, visit http://www.prisma-statement.org/.
Eligibility Criteria
The randomized controlled trials included in this study assessed the effects of IMT as a sole intervention or combined with exercise, comparing it with other interventions (active-control) or placebo (passive-control). The control group must not have received IMT as an alternative intervention.
The included studies were divided into subgroups according to the method of IMT training (spring load, tapered flow resistance load, and fixed flow resistive load), load (>30% or <30% of the maximal inspiratory pressure [MIP] and maximal expiratory pressure [MEP]) (minimum recommended load from the American Physical Therapy Association guideline),16 and PoCR period when IMT was provided (phase 1 = inpatient and phase 2 = outpatient).
Inclusion Criteria
Studies involving patients older than 18 years, both sexes, undergoing CABG surgery.
Exclusion Criteria
Studies investigating only 1 acute session of IMT.
Data Items
The following outcomes were considered: exercise capacity (assessed with the 6-minute walking distance [6MWD, meters]) or peak Vo2, respiratory muscle strength (assessed with MIP and MEP, cm H2O), LOS (days), and QOL following CABG surgery. Only studies assessing the chronic effect (at least 3 days of intervention) of IMT were included.
Selection Process, Data Collection Process, and Presentation
First, 2 independent reviewers extracted relevant data from each eligible study, and a third one resolved any discrepancies. Selection, data collection, and extraction were performed using the Covidence platform.25 The study’s authors were contacted by email if the outcome data were unclear, not reported, or missing data were observed. Details regarding data synthesis, outcome analysis, quality of the studies, risk-of-bias assessment, and a summary of study findings with the quality of evidence can be found in Figs. 2 and 3.
Summary of evidence including forest plot, risk-of-bias assessment, and certainty of evidence for studies comparing inspiratory muscle training (IMT) intervention versus control during phase 1 of postoperative cardiac rehabilitation following coronary artery bypass graft surgery. (1) Exercise capacity: 6-Minute Walk Test (6MWT, meters) and peak oxygen uptake (peak Vo2, mL of O2·kg·min−1). (2) Respiratory muscle strength: maximum inspiratory pressure (MIP, cm H2O) (2a) and maximum expiratory pressure (MEP, cm H2O) (2b). (3) Length of hospital stay (LOS, days). IMT methods: pressure threshold load (PTL), tapered flow resistance load (TFRL), and flow resistive load (FRL). Risk of bias (RoB2 tool): D1 = randomization process, D2 = deviations from the intended interventions, D3 = missing outcome data, D4 = measurement of the outcome, and D5 = selection of the reported result. The data are “traffic light” RoB2 plots of the domain-level judgments for each individual result, considering low risk (+), some concerns (!), and high risk (−). Quality of evidence: Grading of Recommendations Assessment, Development and Evaluation (GRADE). CPT = conventional physical therapy; MCID = minimum clinically important difference; ND = not described by the authors (patients were submitted to care routines not described by the authors); PEP = positive expiratory pressure; PTL = pressure threshold load; RFL = resistive flow load. aExplanations related to the quality of evidence due to the following criteria: risk of bias, inconsistency, indirectness, and imprecision.
Summary of evidence including forest plot, risk-of-bias assessment, and certainty of evidence for studies comparing inspiratory muscle training (IMT) intervention versus control during phase 2 of postoperative cardiac rehabilitation following coronary artery bypass graft surgery. (1) Exercise capacity: 6-Minute Walk Test (6MWT, meters) and peak oxygen uptake (peak Vo2, mL of O2·kg·min−1). (2) Respiratory muscle strength: maximum inspiratory pressure (MIP, cm H2O) (2a) and maximum expiratory pressure (MEP, cm H2O) (2b). (3) Quality of life (QOL): Minnesota Living with Heart Failure Questionnaire (MLHFQ). Risk of bias 2 (RoB2 tool): D1 = randomization process, D2 = deviations from the intended interventions, D3 = missing outcome data, D4 = measurement of the outcome, and D5 = selection of the reported result. The data are “traffic light” RoB2 plots of the domain-level judgments for each individual result, considering low risk (+), some concerns (!), and high risk (−). Quality of evidence: Grading of Recommendations Assessment, Development and Evaluation (GRADE). MCID = minimum clinically important difference; ND = not described by the authors (patients were submitted to care routines not described by the authors); PTL = pressure threshold load; RE = resistive exercise. aExplanations related to the quality of evidence due to the following criteria: risk of bias, inconsistency, indirectness and imprecision.
Information Sources and Search Strategy
A systematic search was performed in the following databases for articles published from inception to March 2023 in 5 databases: MEDLINE (PubMed), EMBASE, CINAHL, Scopus, and CENTRAL. The full search strategy for each database is available (Suppl. Material 1). There was no publication date, age, or setting restrictions; articles published in English, Portuguese, French, Dutch, and Spanish were included.
Risk-of-Bias and Certainty Assessments
The 2 independent reviewers evaluated the quality of the included studies by assessing the risk of bias utilizing the latest template version of the RoB 2 tool.26 This tool assesses study quality according to the following 5 domains: randomization process (D1), deviations from the intended interventions (D2), missing outcome data (D3), measurement of the outcome (D4), and selection of the reported result (D5) in 3 categories (low risk, some concerns, and high risk). The Grading of Recommendations Assessment, Development and Evaluation method was used to assess the certainty of the evidence utilizing the GRADEpro GDT platform. Results are reported in 4 categories (very low, low, moderate, and high).27
Effect Measures and Synthesis Methods
The meta-analysis was performed using a random effects model when at least 2 studies were similar for 1 of the prespecified outcomes, according to the population, intervention, comparison, and outcomes framework. All outcome measures were reported as continuous variable, and either the mean difference or postintervention values were utilized to assess the effect measures according to baseline characteristics. Sensitivity analyses were performed considering the following aspects: load utilized (> and < 30% of the MIP) in phase 1 and 2 PoCR. Statistical heterogeneity of the treatment effect among studies was assessed using the Cochran Q test, and the inconsistency I2 test (a value of >50% was considered to indicate high heterogeneity) was used to quantify heterogeneity as a percentage. All analyses were conducted using RevMan (version 5.3; The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark). Forest plots were presented for all meta-analyses, and statistical significance was established at a P value of <.05.
Transformation to Clinical Practice
We calculated the treatment success probability based on an outcome measure’s minimum clinically important difference (MCID). This probability was expressed as the number of patients who experienced a clinically relevant effect. The probability of success was considered favorable when the lowest and highest confidence interval (95% CI ) values were more significant than the MCID. We selected the MCID for each outcome from the available literature, including the 6MWT (25 and 36.1 m),28,29 peak Vo2 (1.0 mL of O2·kg·min−1),30 and MIP (17.2–17.6 cm H2O).31 We found no MCID for MEP, LOS, and QOL.
Results
Study Selection and Characteristics
A total of 10,335 articles were identified in the initial systematic search, and 28 potentially eligible articles were retrieved as full text. One additional article was identified on a non-structured website search and was examined for possible inclusion. Overall, 15 articles (published between 2009 and March 2023) with 447 patients met the inclusion criteria and were included in this meta-analysis (Fig. 1). Eleven studies examined the effects of IMT in phase 1, and 4 examined the effects of IMT in phase 2 of cardiac rehabilitation following CABG surgery. The reasons for excluding of studies during the full-text review are available in Supplementary Material 2.
Results of Individual Studies
Overall, 226 participants were randomly assigned to intervention groups and 221 to control groups. Moreover, 336 patients were enrolled into phase 1 and 111 in phase 2 PoCR. The mean age of all participants was 58.58 years (range = 55–67), with 69.46% being men. Nine of the studies in phase 1 used spring load devices, 1 utilized a tapered flow resistance load device, and 1 a fixed flow resistive load device that adjusted inspiratory resistance based on patient effort (not exceeding moderate effort).32 All phase 2 studies used spring load devices.33–36 No adverse events were reported in any of the studies. The characteristics of the participants are presented in Table 1.
Characteristics of Studies Included in the Meta-Analysisa
| Study | CR Phase | Surgical Procedure | Population | Intervention Group | Comparison Group | Outcome Assessed | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Mean (SD) Age, y | No. of Participants | No. (%) Men | Frequency | Intensity | Duration | Volume (Sets/Rep.) | Progression | Type | Frequency | Intensity | Duration | Volume (Sets/Rep.) | Type of Intervention | ||||||||
| Interv. Group | Control Group | Interv. Group | Control Group | Interv. Group | Control Group | IMT Method | Additional Therapy | ||||||||||||||
| Praveen and Praveen32 | 1 | CABG | 57.2 (5.62) | 55.6 (5.26) | 15 | 15 | NI | NI | 3×/d | NI | 15 d | 3/10 | If RPE was <5 on Borg Scale, resistance of IMT was increased incrementally | FRL | CPT (deep-breathing exercises, directed cough, and early mobilization) | 3×/d | NI | 15 d | NI | CPT (not detailed) | MIP and MVV |
| Barros et al37 | 1 | CABG | 62.1 (8.10) | 67.08 (7.11) | 23 | 23 | 19 (82.6) | 6 (40) | 1×/d | 40% of MIP | 7 d | 3/10 | Not changed | SL | CPT (bronchial hygiene maneuver, postural drainage, and tracheal aspiration when necessary) | NI | NI | 8 d | NI | CPT | MIP, MEP, TV, dyspnea, PEF, pain, and hospital stay |
| Savci et al42 | 1 | CABG | 62.82 (8.69) | 57.48 (11.48) | 22 | 21 | 19 (82.36) | 19 (90.47) | 2×/d | 15%– 45% of MIP | 30 min for 10 d | NI | Resistance was increased incrementally between 15% and 45% on basis of patients’ tolerance | SL | CPT (mobilization, upper and lower limb active exercise, chest physical therapy, and walking) | 1×/d | NI | 30 min, for 10 d | NI | CPT | MIP, 6MWT, FC, hospital stay, QOL, and psychosocial status |
| Matheus et al40 | 1 | CABG | 61.83 (13.53) | 63.3 (10.2) | 23 | 24 | 18 (78,2) | 16 (66,67) | 2×/d | 40% of MIP | 3 d | 3/10 | Not changed | SL | CPT (lung reexpansion with fractional patterns, respiratory incentive, orthostatism, and ambulation) | 2×/d | NI | 3 d | NI | CPT | MIP, MEP, TV, VC, and PEF |
| Cordeiro et al38 | 1 | CABG + valve | 57 (10) | 56.4 (9.6) | 25 | 25 | 11 (44) | 16 (64) | 2×/d | 40% of MIP | 6 d | 3/10 | Not changed | SL | No specific intervention | NI | NI | 8 d | NI | No specific intervention | MIP and 6MWT |
| Elmarakby43 | 1 | CABG | 56.90 (3.75) | 56.95 (3.75) | 17 | 16 | 17 (52) | 16 (48) | 2×/d | 30%–80% of MIP | 15 min for 21 d (14 in preoperative and 7 in postoperative) | 1/30 | Workload was increased by 2 cm H2O if patient scored level of perceived exertion as <5/10 | SL | CPT (early mobilization, deep-breathing exercises, and cough instructions) | 1×/d | NI | 20 d | NI | CPT | MIP, Spo2, and A-a gradient |
| Zanini et al41 | 1 | CABG | 58 (5) | 61 (5) | 10 | 10 | 9 (90) | 7 (70) | 2×/d | 20% of MIP + 4 cm H2O | 7 d | 8–10/10 | 20% of MIP increasing from 1–4 cm H2O until day 6 | SL | CPT (upper and lower limb active exercise, ambulation, bronchial hygiene therapy, and deep breathing) + EPAP | 2×/d | NI | 6 d | EPAP: 5–10/10; 5–15 cm H2O | CPT + EPAP | MIP, MEP, CPET, lung capacity parameters, and 6MWT |
| Cordeiro et al48 | 1 | CABG | 61 (9.6) | 62 (10) | 21 | 21 | 14 (67) | 13 (62) | 2×/d | 10%–35% of MIP based on AT | 7 d | 3/15 | Began with 10% of MIP and increased 10% at each level of test | SL | CPT (breathing exercises, kinesiotherapy, cycloergometry, and ambulation) + NIV | 2×/d | 40% of MIP | 8 d | 3/15 | SL + CPT | MIP, MEP, 6MWT, VC, and PEF |
| Cordeiro et al12 | 1 | CABG | 55 (10) | 57 (8) | 19 | 19 | 11 (58) | 10 (55) | 2×/d | 40% of MIP | 6 d | 3/10 | Not changed | SL | CPT (ambulation, breathing exercises, cycle ergometry, and kinesiotherapy) | NI | NI | 9 d | NI | CPT | MIP, MEP, PEF, and 6MWT |
| de Aquino et al39 | 1 | CABG | 60.07 (8.52) | 60.52 (9.74) | 41 | 42 | 28 (68) | 37 (88) | 1×/d | 15% of MIP and MEP | 5 d | 3/10 | Not changed | SL | CPT (knee flexion and extension + upper and lower limb), CPT (PMT = peripheral muscle training) involved knee flexion and extension exercises with 0.5 kg weights and finger flexions with 3.0 lb weights | 1×/d | NI | 5 d | NI | CPT | MIP, MEP, QOL, CPT, PMT, FC, pain, and 6MWT |
| Fortes et al44 | 1 | CABG + valve | 61.5 (12.3) | 59.7 (13.1) | 15 | 15 | 11 (73.33) | 12 (80) | 2×/d | 30% of MIP | 6 d | 1/30 | New evaluation was performed to redefine MIP load on postoperative day 3 | TRFL | CPT (deep inspiration, cough, pulmonary reexpansion, diaphragmatic breathing, active range of motion exercises involving the limbs (elbows, shoulders, hips, and knees), ambulation, and oxygen therapy when necessary) | 2×/d | NI | 6 d | NI | CPT | MIP, IMS dynamics, S-index, and PIF |
| Hermes et al34 | 2 | CABG | 55.2 (7.9) | 59.5 (8.7) | 12 | 12 | 7 (58.3) | 10 (83.8) | 2×/d | 30% of MIP | 60 min/session for 12 wk | 3/10 | Each week, training load was adjusted to maintain 30% of MIP | SL | CE | 2×/wk | Based on percentage of heart rate reserve; 50% of load of 1MR | 60 min/session for 12 wk | 3/10 | CE | MIP, MEP, peak Vo2, and QOL |
| Radi et al36 | 2 | CABG | NI | NI | 22 | 23 | NI | NI | NI | 60% of MIV | 30 min/session, 10–12 sessions | NI | NI | SL | CPT (not detailed) | NI | NI | NI | NI | CPT (not detailed) | MIV and 6MWT distance |
| Miozzo et al35 | 2 | CABG | 57.6 (7.9) | 57.4 (8.54) | 13 | 11 | 8 (88.9) | 7 (77.8) | NI | 50%–80% of MIP | 12 wk | 5/10 | 50% of MIP for 2 wk, progressing in following weeks to 80% | SL | Aerobic exercise | NI | 50% of reserve peak HR progressing to 80% over time | 40 min/session for 12 wk | NI | Aerobic exercise | 6MWT, MIP, MEP, PMT, and QOL |
| dos Santos et al33 | 2 | CABG | 55 (7) | 56.6 (5.5) | 12 | 12 | 8 (66.7) | 9 (75) | 2×/wk | 50%–80% of MIP | 12 wk | 5/10 | 50% of MIP for 2 wk, progressing in following weeks to 80% | SL | CE | 2×/wk | Minimum load of 9 cm H2O was kept constant | 12 wk | 3/10 | SL + CE | MIP, MEP, 6MWT, CPET, peak Vo2, and QOL |
| Study | CR Phase | Surgical Procedure | Population | Intervention Group | Comparison Group | Outcome Assessed | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Mean (SD) Age, y | No. of Participants | No. (%) Men | Frequency | Intensity | Duration | Volume (Sets/Rep.) | Progression | Type | Frequency | Intensity | Duration | Volume (Sets/Rep.) | Type of Intervention | ||||||||
| Interv. Group | Control Group | Interv. Group | Control Group | Interv. Group | Control Group | IMT Method | Additional Therapy | ||||||||||||||
| Praveen and Praveen32 | 1 | CABG | 57.2 (5.62) | 55.6 (5.26) | 15 | 15 | NI | NI | 3×/d | NI | 15 d | 3/10 | If RPE was <5 on Borg Scale, resistance of IMT was increased incrementally | FRL | CPT (deep-breathing exercises, directed cough, and early mobilization) | 3×/d | NI | 15 d | NI | CPT (not detailed) | MIP and MVV |
| Barros et al37 | 1 | CABG | 62.1 (8.10) | 67.08 (7.11) | 23 | 23 | 19 (82.6) | 6 (40) | 1×/d | 40% of MIP | 7 d | 3/10 | Not changed | SL | CPT (bronchial hygiene maneuver, postural drainage, and tracheal aspiration when necessary) | NI | NI | 8 d | NI | CPT | MIP, MEP, TV, dyspnea, PEF, pain, and hospital stay |
| Savci et al42 | 1 | CABG | 62.82 (8.69) | 57.48 (11.48) | 22 | 21 | 19 (82.36) | 19 (90.47) | 2×/d | 15%– 45% of MIP | 30 min for 10 d | NI | Resistance was increased incrementally between 15% and 45% on basis of patients’ tolerance | SL | CPT (mobilization, upper and lower limb active exercise, chest physical therapy, and walking) | 1×/d | NI | 30 min, for 10 d | NI | CPT | MIP, 6MWT, FC, hospital stay, QOL, and psychosocial status |
| Matheus et al40 | 1 | CABG | 61.83 (13.53) | 63.3 (10.2) | 23 | 24 | 18 (78,2) | 16 (66,67) | 2×/d | 40% of MIP | 3 d | 3/10 | Not changed | SL | CPT (lung reexpansion with fractional patterns, respiratory incentive, orthostatism, and ambulation) | 2×/d | NI | 3 d | NI | CPT | MIP, MEP, TV, VC, and PEF |
| Cordeiro et al38 | 1 | CABG + valve | 57 (10) | 56.4 (9.6) | 25 | 25 | 11 (44) | 16 (64) | 2×/d | 40% of MIP | 6 d | 3/10 | Not changed | SL | No specific intervention | NI | NI | 8 d | NI | No specific intervention | MIP and 6MWT |
| Elmarakby43 | 1 | CABG | 56.90 (3.75) | 56.95 (3.75) | 17 | 16 | 17 (52) | 16 (48) | 2×/d | 30%–80% of MIP | 15 min for 21 d (14 in preoperative and 7 in postoperative) | 1/30 | Workload was increased by 2 cm H2O if patient scored level of perceived exertion as <5/10 | SL | CPT (early mobilization, deep-breathing exercises, and cough instructions) | 1×/d | NI | 20 d | NI | CPT | MIP, Spo2, and A-a gradient |
| Zanini et al41 | 1 | CABG | 58 (5) | 61 (5) | 10 | 10 | 9 (90) | 7 (70) | 2×/d | 20% of MIP + 4 cm H2O | 7 d | 8–10/10 | 20% of MIP increasing from 1–4 cm H2O until day 6 | SL | CPT (upper and lower limb active exercise, ambulation, bronchial hygiene therapy, and deep breathing) + EPAP | 2×/d | NI | 6 d | EPAP: 5–10/10; 5–15 cm H2O | CPT + EPAP | MIP, MEP, CPET, lung capacity parameters, and 6MWT |
| Cordeiro et al48 | 1 | CABG | 61 (9.6) | 62 (10) | 21 | 21 | 14 (67) | 13 (62) | 2×/d | 10%–35% of MIP based on AT | 7 d | 3/15 | Began with 10% of MIP and increased 10% at each level of test | SL | CPT (breathing exercises, kinesiotherapy, cycloergometry, and ambulation) + NIV | 2×/d | 40% of MIP | 8 d | 3/15 | SL + CPT | MIP, MEP, 6MWT, VC, and PEF |
| Cordeiro et al12 | 1 | CABG | 55 (10) | 57 (8) | 19 | 19 | 11 (58) | 10 (55) | 2×/d | 40% of MIP | 6 d | 3/10 | Not changed | SL | CPT (ambulation, breathing exercises, cycle ergometry, and kinesiotherapy) | NI | NI | 9 d | NI | CPT | MIP, MEP, PEF, and 6MWT |
| de Aquino et al39 | 1 | CABG | 60.07 (8.52) | 60.52 (9.74) | 41 | 42 | 28 (68) | 37 (88) | 1×/d | 15% of MIP and MEP | 5 d | 3/10 | Not changed | SL | CPT (knee flexion and extension + upper and lower limb), CPT (PMT = peripheral muscle training) involved knee flexion and extension exercises with 0.5 kg weights and finger flexions with 3.0 lb weights | 1×/d | NI | 5 d | NI | CPT | MIP, MEP, QOL, CPT, PMT, FC, pain, and 6MWT |
| Fortes et al44 | 1 | CABG + valve | 61.5 (12.3) | 59.7 (13.1) | 15 | 15 | 11 (73.33) | 12 (80) | 2×/d | 30% of MIP | 6 d | 1/30 | New evaluation was performed to redefine MIP load on postoperative day 3 | TRFL | CPT (deep inspiration, cough, pulmonary reexpansion, diaphragmatic breathing, active range of motion exercises involving the limbs (elbows, shoulders, hips, and knees), ambulation, and oxygen therapy when necessary) | 2×/d | NI | 6 d | NI | CPT | MIP, IMS dynamics, S-index, and PIF |
| Hermes et al34 | 2 | CABG | 55.2 (7.9) | 59.5 (8.7) | 12 | 12 | 7 (58.3) | 10 (83.8) | 2×/d | 30% of MIP | 60 min/session for 12 wk | 3/10 | Each week, training load was adjusted to maintain 30% of MIP | SL | CE | 2×/wk | Based on percentage of heart rate reserve; 50% of load of 1MR | 60 min/session for 12 wk | 3/10 | CE | MIP, MEP, peak Vo2, and QOL |
| Radi et al36 | 2 | CABG | NI | NI | 22 | 23 | NI | NI | NI | 60% of MIV | 30 min/session, 10–12 sessions | NI | NI | SL | CPT (not detailed) | NI | NI | NI | NI | CPT (not detailed) | MIV and 6MWT distance |
| Miozzo et al35 | 2 | CABG | 57.6 (7.9) | 57.4 (8.54) | 13 | 11 | 8 (88.9) | 7 (77.8) | NI | 50%–80% of MIP | 12 wk | 5/10 | 50% of MIP for 2 wk, progressing in following weeks to 80% | SL | Aerobic exercise | NI | 50% of reserve peak HR progressing to 80% over time | 40 min/session for 12 wk | NI | Aerobic exercise | 6MWT, MIP, MEP, PMT, and QOL |
| dos Santos et al33 | 2 | CABG | 55 (7) | 56.6 (5.5) | 12 | 12 | 8 (66.7) | 9 (75) | 2×/wk | 50%–80% of MIP | 12 wk | 5/10 | 50% of MIP for 2 wk, progressing in following weeks to 80% | SL | CE | 2×/wk | Minimum load of 9 cm H2O was kept constant | 12 wk | 3/10 | SL + CE | MIP, MEP, 6MWT, CPET, peak Vo2, and QOL |
1MR = 1 maximum repetition; 6MWT = 6-Minute Walk Test; A-a gradient = alveolar-arterial gradient; AT = anaerobic training; CABG = coronary artery bypass grafting; CE = combined exercise (aerobic and resistance exercises); CPET = cardiopulmonary exercise test; CPT = conventional physical therapy; CR = cardiac rehabilitation; EPAP = expiratory positive airway pressure; FC = functional capacity; FRL = flow resistance load; IMS = inspiratory muscle strength; IMT = inspiratory muscle training; Interv. = intervention; MEP = maximal expiratory pressure; MIP = maximal inspiratory pressure; MIV = maximum inspiratory volume; MVV = maximal voluntary ventilation; NI = not indicated; NIV = noninvasive ventilation; peak HR = maximal heart rate; peak VO2 = peak oxygen uptake; PEF = peak expiratory flow; PIF = peak inspiratory flow; PMT = peripheral muscle strength training; QOL = quality of life; Rep. = repetitions; RPE = rate of perceived exertion; S-index = dynamic inspiratory muscle strength; SL = spring load; Spo2 = O2 saturation; TRFL = tapered resistance flow load; TV = tidal volume; VC = vital capacity.
Characteristics of Studies Included in the Meta-Analysisa
| Study | CR Phase | Surgical Procedure | Population | Intervention Group | Comparison Group | Outcome Assessed | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Mean (SD) Age, y | No. of Participants | No. (%) Men | Frequency | Intensity | Duration | Volume (Sets/Rep.) | Progression | Type | Frequency | Intensity | Duration | Volume (Sets/Rep.) | Type of Intervention | ||||||||
| Interv. Group | Control Group | Interv. Group | Control Group | Interv. Group | Control Group | IMT Method | Additional Therapy | ||||||||||||||
| Praveen and Praveen32 | 1 | CABG | 57.2 (5.62) | 55.6 (5.26) | 15 | 15 | NI | NI | 3×/d | NI | 15 d | 3/10 | If RPE was <5 on Borg Scale, resistance of IMT was increased incrementally | FRL | CPT (deep-breathing exercises, directed cough, and early mobilization) | 3×/d | NI | 15 d | NI | CPT (not detailed) | MIP and MVV |
| Barros et al37 | 1 | CABG | 62.1 (8.10) | 67.08 (7.11) | 23 | 23 | 19 (82.6) | 6 (40) | 1×/d | 40% of MIP | 7 d | 3/10 | Not changed | SL | CPT (bronchial hygiene maneuver, postural drainage, and tracheal aspiration when necessary) | NI | NI | 8 d | NI | CPT | MIP, MEP, TV, dyspnea, PEF, pain, and hospital stay |
| Savci et al42 | 1 | CABG | 62.82 (8.69) | 57.48 (11.48) | 22 | 21 | 19 (82.36) | 19 (90.47) | 2×/d | 15%– 45% of MIP | 30 min for 10 d | NI | Resistance was increased incrementally between 15% and 45% on basis of patients’ tolerance | SL | CPT (mobilization, upper and lower limb active exercise, chest physical therapy, and walking) | 1×/d | NI | 30 min, for 10 d | NI | CPT | MIP, 6MWT, FC, hospital stay, QOL, and psychosocial status |
| Matheus et al40 | 1 | CABG | 61.83 (13.53) | 63.3 (10.2) | 23 | 24 | 18 (78,2) | 16 (66,67) | 2×/d | 40% of MIP | 3 d | 3/10 | Not changed | SL | CPT (lung reexpansion with fractional patterns, respiratory incentive, orthostatism, and ambulation) | 2×/d | NI | 3 d | NI | CPT | MIP, MEP, TV, VC, and PEF |
| Cordeiro et al38 | 1 | CABG + valve | 57 (10) | 56.4 (9.6) | 25 | 25 | 11 (44) | 16 (64) | 2×/d | 40% of MIP | 6 d | 3/10 | Not changed | SL | No specific intervention | NI | NI | 8 d | NI | No specific intervention | MIP and 6MWT |
| Elmarakby43 | 1 | CABG | 56.90 (3.75) | 56.95 (3.75) | 17 | 16 | 17 (52) | 16 (48) | 2×/d | 30%–80% of MIP | 15 min for 21 d (14 in preoperative and 7 in postoperative) | 1/30 | Workload was increased by 2 cm H2O if patient scored level of perceived exertion as <5/10 | SL | CPT (early mobilization, deep-breathing exercises, and cough instructions) | 1×/d | NI | 20 d | NI | CPT | MIP, Spo2, and A-a gradient |
| Zanini et al41 | 1 | CABG | 58 (5) | 61 (5) | 10 | 10 | 9 (90) | 7 (70) | 2×/d | 20% of MIP + 4 cm H2O | 7 d | 8–10/10 | 20% of MIP increasing from 1–4 cm H2O until day 6 | SL | CPT (upper and lower limb active exercise, ambulation, bronchial hygiene therapy, and deep breathing) + EPAP | 2×/d | NI | 6 d | EPAP: 5–10/10; 5–15 cm H2O | CPT + EPAP | MIP, MEP, CPET, lung capacity parameters, and 6MWT |
| Cordeiro et al48 | 1 | CABG | 61 (9.6) | 62 (10) | 21 | 21 | 14 (67) | 13 (62) | 2×/d | 10%–35% of MIP based on AT | 7 d | 3/15 | Began with 10% of MIP and increased 10% at each level of test | SL | CPT (breathing exercises, kinesiotherapy, cycloergometry, and ambulation) + NIV | 2×/d | 40% of MIP | 8 d | 3/15 | SL + CPT | MIP, MEP, 6MWT, VC, and PEF |
| Cordeiro et al12 | 1 | CABG | 55 (10) | 57 (8) | 19 | 19 | 11 (58) | 10 (55) | 2×/d | 40% of MIP | 6 d | 3/10 | Not changed | SL | CPT (ambulation, breathing exercises, cycle ergometry, and kinesiotherapy) | NI | NI | 9 d | NI | CPT | MIP, MEP, PEF, and 6MWT |
| de Aquino et al39 | 1 | CABG | 60.07 (8.52) | 60.52 (9.74) | 41 | 42 | 28 (68) | 37 (88) | 1×/d | 15% of MIP and MEP | 5 d | 3/10 | Not changed | SL | CPT (knee flexion and extension + upper and lower limb), CPT (PMT = peripheral muscle training) involved knee flexion and extension exercises with 0.5 kg weights and finger flexions with 3.0 lb weights | 1×/d | NI | 5 d | NI | CPT | MIP, MEP, QOL, CPT, PMT, FC, pain, and 6MWT |
| Fortes et al44 | 1 | CABG + valve | 61.5 (12.3) | 59.7 (13.1) | 15 | 15 | 11 (73.33) | 12 (80) | 2×/d | 30% of MIP | 6 d | 1/30 | New evaluation was performed to redefine MIP load on postoperative day 3 | TRFL | CPT (deep inspiration, cough, pulmonary reexpansion, diaphragmatic breathing, active range of motion exercises involving the limbs (elbows, shoulders, hips, and knees), ambulation, and oxygen therapy when necessary) | 2×/d | NI | 6 d | NI | CPT | MIP, IMS dynamics, S-index, and PIF |
| Hermes et al34 | 2 | CABG | 55.2 (7.9) | 59.5 (8.7) | 12 | 12 | 7 (58.3) | 10 (83.8) | 2×/d | 30% of MIP | 60 min/session for 12 wk | 3/10 | Each week, training load was adjusted to maintain 30% of MIP | SL | CE | 2×/wk | Based on percentage of heart rate reserve; 50% of load of 1MR | 60 min/session for 12 wk | 3/10 | CE | MIP, MEP, peak Vo2, and QOL |
| Radi et al36 | 2 | CABG | NI | NI | 22 | 23 | NI | NI | NI | 60% of MIV | 30 min/session, 10–12 sessions | NI | NI | SL | CPT (not detailed) | NI | NI | NI | NI | CPT (not detailed) | MIV and 6MWT distance |
| Miozzo et al35 | 2 | CABG | 57.6 (7.9) | 57.4 (8.54) | 13 | 11 | 8 (88.9) | 7 (77.8) | NI | 50%–80% of MIP | 12 wk | 5/10 | 50% of MIP for 2 wk, progressing in following weeks to 80% | SL | Aerobic exercise | NI | 50% of reserve peak HR progressing to 80% over time | 40 min/session for 12 wk | NI | Aerobic exercise | 6MWT, MIP, MEP, PMT, and QOL |
| dos Santos et al33 | 2 | CABG | 55 (7) | 56.6 (5.5) | 12 | 12 | 8 (66.7) | 9 (75) | 2×/wk | 50%–80% of MIP | 12 wk | 5/10 | 50% of MIP for 2 wk, progressing in following weeks to 80% | SL | CE | 2×/wk | Minimum load of 9 cm H2O was kept constant | 12 wk | 3/10 | SL + CE | MIP, MEP, 6MWT, CPET, peak Vo2, and QOL |
| Study | CR Phase | Surgical Procedure | Population | Intervention Group | Comparison Group | Outcome Assessed | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Mean (SD) Age, y | No. of Participants | No. (%) Men | Frequency | Intensity | Duration | Volume (Sets/Rep.) | Progression | Type | Frequency | Intensity | Duration | Volume (Sets/Rep.) | Type of Intervention | ||||||||
| Interv. Group | Control Group | Interv. Group | Control Group | Interv. Group | Control Group | IMT Method | Additional Therapy | ||||||||||||||
| Praveen and Praveen32 | 1 | CABG | 57.2 (5.62) | 55.6 (5.26) | 15 | 15 | NI | NI | 3×/d | NI | 15 d | 3/10 | If RPE was <5 on Borg Scale, resistance of IMT was increased incrementally | FRL | CPT (deep-breathing exercises, directed cough, and early mobilization) | 3×/d | NI | 15 d | NI | CPT (not detailed) | MIP and MVV |
| Barros et al37 | 1 | CABG | 62.1 (8.10) | 67.08 (7.11) | 23 | 23 | 19 (82.6) | 6 (40) | 1×/d | 40% of MIP | 7 d | 3/10 | Not changed | SL | CPT (bronchial hygiene maneuver, postural drainage, and tracheal aspiration when necessary) | NI | NI | 8 d | NI | CPT | MIP, MEP, TV, dyspnea, PEF, pain, and hospital stay |
| Savci et al42 | 1 | CABG | 62.82 (8.69) | 57.48 (11.48) | 22 | 21 | 19 (82.36) | 19 (90.47) | 2×/d | 15%– 45% of MIP | 30 min for 10 d | NI | Resistance was increased incrementally between 15% and 45% on basis of patients’ tolerance | SL | CPT (mobilization, upper and lower limb active exercise, chest physical therapy, and walking) | 1×/d | NI | 30 min, for 10 d | NI | CPT | MIP, 6MWT, FC, hospital stay, QOL, and psychosocial status |
| Matheus et al40 | 1 | CABG | 61.83 (13.53) | 63.3 (10.2) | 23 | 24 | 18 (78,2) | 16 (66,67) | 2×/d | 40% of MIP | 3 d | 3/10 | Not changed | SL | CPT (lung reexpansion with fractional patterns, respiratory incentive, orthostatism, and ambulation) | 2×/d | NI | 3 d | NI | CPT | MIP, MEP, TV, VC, and PEF |
| Cordeiro et al38 | 1 | CABG + valve | 57 (10) | 56.4 (9.6) | 25 | 25 | 11 (44) | 16 (64) | 2×/d | 40% of MIP | 6 d | 3/10 | Not changed | SL | No specific intervention | NI | NI | 8 d | NI | No specific intervention | MIP and 6MWT |
| Elmarakby43 | 1 | CABG | 56.90 (3.75) | 56.95 (3.75) | 17 | 16 | 17 (52) | 16 (48) | 2×/d | 30%–80% of MIP | 15 min for 21 d (14 in preoperative and 7 in postoperative) | 1/30 | Workload was increased by 2 cm H2O if patient scored level of perceived exertion as <5/10 | SL | CPT (early mobilization, deep-breathing exercises, and cough instructions) | 1×/d | NI | 20 d | NI | CPT | MIP, Spo2, and A-a gradient |
| Zanini et al41 | 1 | CABG | 58 (5) | 61 (5) | 10 | 10 | 9 (90) | 7 (70) | 2×/d | 20% of MIP + 4 cm H2O | 7 d | 8–10/10 | 20% of MIP increasing from 1–4 cm H2O until day 6 | SL | CPT (upper and lower limb active exercise, ambulation, bronchial hygiene therapy, and deep breathing) + EPAP | 2×/d | NI | 6 d | EPAP: 5–10/10; 5–15 cm H2O | CPT + EPAP | MIP, MEP, CPET, lung capacity parameters, and 6MWT |
| Cordeiro et al48 | 1 | CABG | 61 (9.6) | 62 (10) | 21 | 21 | 14 (67) | 13 (62) | 2×/d | 10%–35% of MIP based on AT | 7 d | 3/15 | Began with 10% of MIP and increased 10% at each level of test | SL | CPT (breathing exercises, kinesiotherapy, cycloergometry, and ambulation) + NIV | 2×/d | 40% of MIP | 8 d | 3/15 | SL + CPT | MIP, MEP, 6MWT, VC, and PEF |
| Cordeiro et al12 | 1 | CABG | 55 (10) | 57 (8) | 19 | 19 | 11 (58) | 10 (55) | 2×/d | 40% of MIP | 6 d | 3/10 | Not changed | SL | CPT (ambulation, breathing exercises, cycle ergometry, and kinesiotherapy) | NI | NI | 9 d | NI | CPT | MIP, MEP, PEF, and 6MWT |
| de Aquino et al39 | 1 | CABG | 60.07 (8.52) | 60.52 (9.74) | 41 | 42 | 28 (68) | 37 (88) | 1×/d | 15% of MIP and MEP | 5 d | 3/10 | Not changed | SL | CPT (knee flexion and extension + upper and lower limb), CPT (PMT = peripheral muscle training) involved knee flexion and extension exercises with 0.5 kg weights and finger flexions with 3.0 lb weights | 1×/d | NI | 5 d | NI | CPT | MIP, MEP, QOL, CPT, PMT, FC, pain, and 6MWT |
| Fortes et al44 | 1 | CABG + valve | 61.5 (12.3) | 59.7 (13.1) | 15 | 15 | 11 (73.33) | 12 (80) | 2×/d | 30% of MIP | 6 d | 1/30 | New evaluation was performed to redefine MIP load on postoperative day 3 | TRFL | CPT (deep inspiration, cough, pulmonary reexpansion, diaphragmatic breathing, active range of motion exercises involving the limbs (elbows, shoulders, hips, and knees), ambulation, and oxygen therapy when necessary) | 2×/d | NI | 6 d | NI | CPT | MIP, IMS dynamics, S-index, and PIF |
| Hermes et al34 | 2 | CABG | 55.2 (7.9) | 59.5 (8.7) | 12 | 12 | 7 (58.3) | 10 (83.8) | 2×/d | 30% of MIP | 60 min/session for 12 wk | 3/10 | Each week, training load was adjusted to maintain 30% of MIP | SL | CE | 2×/wk | Based on percentage of heart rate reserve; 50% of load of 1MR | 60 min/session for 12 wk | 3/10 | CE | MIP, MEP, peak Vo2, and QOL |
| Radi et al36 | 2 | CABG | NI | NI | 22 | 23 | NI | NI | NI | 60% of MIV | 30 min/session, 10–12 sessions | NI | NI | SL | CPT (not detailed) | NI | NI | NI | NI | CPT (not detailed) | MIV and 6MWT distance |
| Miozzo et al35 | 2 | CABG | 57.6 (7.9) | 57.4 (8.54) | 13 | 11 | 8 (88.9) | 7 (77.8) | NI | 50%–80% of MIP | 12 wk | 5/10 | 50% of MIP for 2 wk, progressing in following weeks to 80% | SL | Aerobic exercise | NI | 50% of reserve peak HR progressing to 80% over time | 40 min/session for 12 wk | NI | Aerobic exercise | 6MWT, MIP, MEP, PMT, and QOL |
| dos Santos et al33 | 2 | CABG | 55 (7) | 56.6 (5.5) | 12 | 12 | 8 (66.7) | 9 (75) | 2×/wk | 50%–80% of MIP | 12 wk | 5/10 | 50% of MIP for 2 wk, progressing in following weeks to 80% | SL | CE | 2×/wk | Minimum load of 9 cm H2O was kept constant | 12 wk | 3/10 | SL + CE | MIP, MEP, 6MWT, CPET, peak Vo2, and QOL |
1MR = 1 maximum repetition; 6MWT = 6-Minute Walk Test; A-a gradient = alveolar-arterial gradient; AT = anaerobic training; CABG = coronary artery bypass grafting; CE = combined exercise (aerobic and resistance exercises); CPET = cardiopulmonary exercise test; CPT = conventional physical therapy; CR = cardiac rehabilitation; EPAP = expiratory positive airway pressure; FC = functional capacity; FRL = flow resistance load; IMS = inspiratory muscle strength; IMT = inspiratory muscle training; Interv. = intervention; MEP = maximal expiratory pressure; MIP = maximal inspiratory pressure; MIV = maximum inspiratory volume; MVV = maximal voluntary ventilation; NI = not indicated; NIV = noninvasive ventilation; peak HR = maximal heart rate; peak VO2 = peak oxygen uptake; PEF = peak expiratory flow; PIF = peak inspiratory flow; PMT = peripheral muscle strength training; QOL = quality of life; Rep. = repetitions; RPE = rate of perceived exertion; S-index = dynamic inspiratory muscle strength; SL = spring load; Spo2 = O2 saturation; TRFL = tapered resistance flow load; TV = tidal volume; VC = vital capacity.
Phase 1 Postoperative IMT
The method of spring load IMT used in phase 1 PoCR included the Threshold IMT (Philips Respironics, Inc., Murrysville, PA) device (7 studies)31,37–42 and the PowerBreathe Plus device (1 study; PowerBreathe International, Southam, England, UK).43 The PowerBreathe Kinetic Series device (POWERbreathe International, Southam, England, UK)44 was used in 1 study and was the only phase 1 study not reporting the intensity of IMT. The intensity of IMT used in the phase 1 PoCR studies ranged from 10% to 40% of the MIP. Five studies utilized an IMT intensity of ≥30% of the MIP,37,38,40,43,45 and 2 studies used an intensity of <30% of the MIP.39,41 Two studies initiated IMT in the preoperative period and continued during the postoperative period with an IMT load of 30% of the MIP and 15% to 45% of the MIP, respectively.42,43
The weekly frequency of IMT in phase 1 PoCR ranged from once to twice daily for 20 to 30 min each session, and the study duration was between 3 and 21 days using a training volume of 3 sets with 10 to 15 repetitions each set.12,37–41,45 The control group received conventional physical therapy based on the practice standard in each study setting.
Phase 2 Postoperative IMT
Four studies of IMT in phase 2 PoCR investigated the effects of 2 to 12 weeks of IMT combined with AE and other combined exercises.33–36 The spring load devices used in these studies included the PowerBreathe Plus device (POWERbreathe International, Southam, England, UK) in 2 studies that used the same protocol, beginning IMT at 50% of the MIP, which was gradually increased until 80% of the MIP was achieved and maintained for 12 weeks,33,35 the Threshold device in 1 study that maintained IMT at 30% of the MIP which was performed with 3 sets of 10 repetitions,34 and 1 study that did not report the type of spring load device employed. However, the intensity of IMT was maintained at 60% of the maximum inspiratory volume for 30 minutes per session for 12 weeks.36 The control group performed combined aerobic and resistance training33,34 and AE35 over a period of 3 months, and 1 study did not report the intervention of the control group.36
Reporting Bias
The risk-of-bias assessment of the 8 randomized controlled trials reporting on the effect of IMT on MIP in phase 1 PoCR42,43 found that 2 were judged to have a low risk of bias,30,38 and 6 were judged to have a high risk of bias.32,37,38,40,44 The risk-of-bias assessment of the 5 studies examining the effect of IMT on the 6MWT found that 2 were judged to have a low risk of bias,40,42 and 2 were judged to have a high risk of bias.41,45 The risk of bias of the 5 studies examining the effect of IMT on LOS was predominantly low, with only 1 study judged to have a high risk of bias37 (Fig. 2). Weighted bar plots of the distribution of risk-of-bias judgments are available in Supplementary Material 3.
The risk-of-bias assessment of the 3 randomized controlled trials reporting on the effect of IMT on MIP in phase 2 PoCR and the 2 studies examining the effect of IMT on MEP found a low risk of bias.33,35 The risk of bias of the 3 studies examining the effect of IMT on the 6MWT and the 2 studies examining the effect of IMT on peak Vo2 was low.18,20 The risk of bias in the 2 studies examining the effect of IMT on QOL was mostly low33,35 (Fig. 3). Weighted bar plots of the distribution of risk-of-bias judgments are available in Supplementary Material 4.
Results of Syntheses
Phase 1 Cardiac Rehabilitation Studies
Exercise capacity analysis was assessed through the 6MWT in phase 1 PoCR. The overall increase in the 6MWT was +71.13 m (95% CI = 37.61 to 104.66) (I2 = 75%) with an effect that was similar in studies employing IMT with loads of >30% of the MIP, +75.46 m (95% CI = 52.34 to 98.57) (I2 = 9%) but was not observed in the studies utilizing IMT with loads of <30% of the MIP12,38 (Fig. 2, panel 1). This response exceeded both prespecified MCIDs for the 6MWT, 25, and 36.1 m.28,29
We selected the MCID for each outcome from the available literature, including peak Vo2 (1.0 mL of O2·kg·min−1)30 and MIP (17.2–17.6 cm H2O).31
Eight studies assessed the MIP in the phase 1 PoCR period. The effects of IMT with a workload of >30% of the MIP significantly improved the MIP compared to the control group, with a mean difference of 12.40 cm H2O (95% CI = 3.00 to 21.80) (I2 = 22%) (Fig. 2, panel 2a), which was not observed with IMT performed with a workload of <30% of the MIP.
The results of all pooled data on the effects of IMT on MIP found that IMT significantly improved MIP with a mean difference of 10.46 cm H2O (95% CI = 2.83 to 18.19) (I2 = 70%).12,32,37,39–41,43,45 However, better results were observed when IMT was performed both preoperative and postoperative with preoperative (load of 60%–80% of the MIP) and postoperative (load of 15%–45%) of the MIP) periods, revealing a significant improvement compared to the control group (30.46 cm H2O; 95% CI = −35.05 to −25.87) (I2 = 16%) (load = 30%43; loads = 15% and 45%42) (Fig. 4). This response did not exceed the prespecified MCID for the MIP, which ranges from 17.2 to 17.6 cm H2O31; however, this MCID was obtained from an outpatient protocol.
Forest plot comparing inspiratory muscle training (IMT) intervention (performed in the preoperative and postoperative periods) versus control during phase 1 cardiac rehabilitation following coronary artery bypass graft surgery.
Five studies assessed the MEP in the phase 1 PoCR period and found no significant overall effect of IMT on MEP with no difference between IMT using workloads of >30% or < 30% of MIP (Fig. 2, panel 2b).12,32,37,39–41,43
Five studies examined the effects of IMT on LOS (days) in the phase 1 PoCR period. IMT significantly reduced LOS (−1.02 days; 95% CI = −2.00 to −0.04) (I2 = 58%) with a 44% greater effect in studies utilizing IMT with workloads >30% of the MIP (−1.47 days; 95% CI = −2.84 to −0.09) (I2 = 45%). The studies utilizing IMT with a workload <30% of MIP had no effect on LOS.39,41
Phase 2 Cardiac Rehabilitation Studies
Four studies evaluated the effect of IMT on exercise capacity using the 6MWT in 3 studies and peak Vo2 in 2. IMT significantly improved the 6MWT with a mean difference of 45.84 m (95% CI = 10.89 to 80.80) (I2 = 43%) (Fig. 3, panel 1a)33–35 but had no effect on peak Vo2 (Fig. 3, panel 1b).33,34
Three studies examined the effect of IMT on MIP in phase 2 PoCR and demonstrated a significant improvement in the MIP with a mean difference of −23.19 m (95% CI = −31.31 to −15.08) (I2 = 0%) (Fig. 3, panel 2a).18,20,46 This response did not exceed the prespecified MCID for the MIP, which ranges from 17.2 to 17.6 cm H2O,31 according to the obtained confidence interval lower limit (−15.08).
Two studies examined the effect of IMT on MEP in phase 2 PoCR and demonstrated a significant improvement in MEP with a mean difference of 20.18 m (95% CI = 9.60 to 30.76) (I2 = 0%) (Fig. 3, panel 2b).18,46
Two studies examined the effect of IMT on QOL using the Minnesota Living with Heart Failure Questionnaire and demonstrated a significant improvement in QOL with a mean difference of −11.17 (95% CI = −11.98 to −4.36) (I2 = 54%) (Fig. 3, panel 3).33,34
Certainty of the Evidence
Based on the Grading of Recommendations Assessment, Development and Evaluation method, the certainty of the evidence for IMT in patients following CABG surgery in phase 1 PoCR was found to be very low for MIP and low for 6MWT and LOS (Suppl. Material 5). However, in phase 2 PoCR, the level of the evidence was found to be moderate for peak Vo2 and 6MWT and very low for QOL (Suppl. Material 6).
Discussion
Our systematic review found that IMT during phase 1 and 2 PoCR in patients following CABG surgery improves exercise capacity, respiratory muscle strength, and QOL while also reducing LOS. The above improvements are meaningful, considering that patients after CABG surgery are often debilitated and have the potential to be less involved using a relatively simple and safe form of training—IMT. Additionally, a minimum load of at least 30% of MIP appears necessary to optimize the effects of IMT in patients after CABG surgery. Furthermore, we were able to access the certainty of the evidence and risk of bias for the above outcome measures with relatively good results and with important information to facilitate the development of subsequent studies of IMT in patients after CABG surgery.
The additional effects of the IMT associated with conventional physical therapy in phase 1 cardiac rehabilitation on exercise capacity and respiratory muscle strength were observed independent of the type of IMT device.37,40,43 The beneficial effects of IMT on exercise capacity may be partially explained by the fact that greater inspiratory muscle strength can decrease respiratory overload during exercise, increasing peripheral blood flow availability and contributing to greater exercise tolerance.47
In addition to the benefits observed on respiratory strength, Elmarakby43 also observed an increase in oxygen saturation and a decrease in alveolar-arterial oxygen gradient following IMT, reflecting an improvement in gas exchange with the potential to reduce atelectasis prevalence in phase 1 PoCR following CABG surgery. The above improvements may also be responsible for the improvement in exercise tolerance from IMT. The results of this study suggest that the determining factor for improving inspiratory muscle strength, exercise capacity, and LOS is related to IMT workloads of >30% of the MIP since the studies utilizing a spring-loaded device with a load of 20% to 24% of the MIP showed no significant increase in any observed outcomes.39,41 Regardless of the favorable effects of IMT in phase 1 and 2 PoCR, the risk of bias from the majority of the phase 1 cardiac rehabilitation studies is concerning contributing to a low and very low certainty of evidence.
Likewise, Dos Santos et al33 demonstrated that 12 weeks of moderate to high-intensity IMT associated with AE and resistance exercise in phase 2 cardiac rehabilitation provided additional benefits to the antioxidant profile of patients undergoing CABG surgery after 12 weeks of training. Similarly, Eibel et al45 saw that a 7-day cardiac rehabilitation program, including ventilatory muscle training, modulated the vascular function and oxidative stress profile, while Elmarakby43 observed an increase in oxygen saturation and a decrease in alveolar-arterial oxygen gradient following IMT. The physiological mechanisms described by these studies utilized in this systematic review may explain the other possible relationship between the increase in MIP with a greater exercise capacity and QOL.33,43
Early postoperative IMT may decrease the harmful effects of the early immobilization period, such as deconditioning, and facilitate patients in the next phase of cardiac rehabilitation (Savci et al).42 The addition of IMT in phase 2 cardiac rehabilitation may have complementary effects to those obtained with combined training on functional capacity. One potential explanation of this finding is that the IMT program improves systemic vasodilation and perfusion of peripheral muscles.47
In the present meta-analysis, the additional effects of the IMT associated with combined AE and RE in phase 2 cardiac rehabilitation on functional capacity (6MWT), respiratory muscle strength (MIP and MEP), and QOL were found in studies utilizing IMT with spring load devices, utilizing 50% to 80% of the MIP33 and 30% of MIP.34 Another study utilized 60% of the maximum inspiratory volume.36 Similarly, Dos Santos et al33 have speculated that the improvement in exercise capacity is due to attenuation of the respiratory muscle metaboreflex and subsequent increase in blood flow to exercising peripheral skeletal muscles.
Despite the above, IMT did not produce significant improvements in peak Vo2. One explanation for the lack of a beneficial effect of IMT combined with AE and RE on peak Vo2 may be due to the inclusion of only 2 studies with high heterogeneity (I2 = 84%).33,34 However, the favorable results observed in submaximal functional capacity, including 3 studies with low heterogeneity and significant improvement in the 6MWT33,35,36 are highly relevant for patients after CABG surgery.
Comparisons With Other Reviews
This review differs from the recent reviews published in 2023 by Zhang et al22 and Dsouza et al21; they included a search period ending in 2021 and 2020, respectively. Furthermore, these systematic reviews did not perform analyses concerning the training load or the cardiac rehabilitation phase. Zhang et al22 has also included studies utilizing expiratory positive airway pressure. Therefore, our review addressed other relevant topics for clinical practice.
IMT resulted in an increase in inspiratory muscle strength, functional capacity, and quality of life, and this increase was higher in studies that included patients utilizing training loads higher than 30%. The IMT, when performed in phase 2 cardiac rehabilitation in combination with AE and resistance exercise for 12 weeks intervention, demonstrated a greater gain in inspiratory and expiratory muscle strength.
Limitations
The limitations of this systematic review and meta-analysis include the small number of studies available, a moderate risk of bias, mainly in the phase 1 PoCR studies, and a relatively diverse certainty of the evidence. However, the meta-analytic findings in the majority of phase 1 and phase 2 PoCR studies are important for patients after CABG surgery, suggesting that IMT may improve the overall postoperative experience of patients following bypass surgery.
Conclusions
IMT significantly increases exercise capacity, respiratory muscle strength, LOS, and QOL following CABG surgery in phase 1 and 2 cardiac rehabilitation programs. IMT with a workload >30% of MIP, applied in both phases 1 and 2, provides a 20% to 30% greater effect on muscle strength. The overall evidence for these conclusions may be influenced by bias and the quality of evidence and should be interpreted cautiously, mainly in phase 1 studies. However, considering the safety and low cost of IMT, the results of this systematic review and meta-analysis suggest that it may be of substantial value to patients after CABG surgery. Further investigation of the effects of IMT in patients after CABG surgery is needed to determine if the above findings can be supported while attending to the risk of bias and certainty of the evidence issues we have observed.
Author Contributions
Clênia Oliveira Araújo (Conceptualization, Data curation, Investigation, Methodology, Writing—original draft), Carla Cristina de Araújo (Data curation, Formal analysis, Investigation, Methodology, Visualization), Francisco R.A. dos Santos (Methodology, Resources, Validation), Lawrence P. Cahalin (Writing—review & editing), Graziella França Bernardelli Cipriano (Project administration, Visualization, Writing—original draft, Writing—review & editing), and Gerson Cipriano Jr. (Conceptualization, Project administration, Validation, Visualization, Writing—original draft, Writing—review & editing)
Funding
There are no funders to report for this study.
Systematic Review Registration
Methods of the analysis and inclusion criteria were specified in advance and documented in a protocol registered on PROSPERO (CRD42021277225).
Disclosures
The authors completed the ICMJE Form for Disclosure of Potential Conflicts of Interest and reported no conflicts of interest.
Comments