Hiago Vinicius Costa SilvaI; Adriana Claudia LunardiII; Ana Carolina Pereira Nunes PintoI; Juliana Ribeiro Fonseca Franco de MacedoV; Elinaldo da Conceição dos SantosI
DOI: 10.21470/1678-9741-2022-0319
ABSTRACT
Introduction: Cardiac surgery is a frequent surgical procedure and may present a high risk of complications. Among the prophylactic strategies studied to decrease the rates of negative outcomes, respiratory care seems to reduce pulmonary complications. Incentive spirometry (IS) is a low-cost, respiratory exercise technique, used for the prevention and treatment of postoperative pulmonary complications (PPC). The aim of this review was to evaluate whether IS is superior to respiratory care, mobilization exercises, and noninvasive ventilation on PPC, and clinical outcomes.CABG = Coronary artery bypass grafting
CENTRAL = Cochrane Central Register of Controlled Trials
CI = Confidence interval
CINAHL® = Cumulative Index of Nursing and Allied Health
CPAP = Continuous positive airway pressure
ECC = Extracorporeal circulation
FEF = Forced expiratory flow
FEV1 = Forced expiratory volume in one second
FVC = Forced vital capacity
MD = Mean differences
MIP = Maximal inspiratory pressure
MIP = Maximal inspiratory pressure
NR = Not registered
PaO2 = Partial pressure of oxygen
PEDro = Physiotherapy Evidence Database
PEF = Peak of expiratory flow
PO = Postoperative
PPC = Postoperative pulmonary complications
RCT = Randomized controlled trial
INTRODUCTION
Cardiac surgery is a frequent surgical procedure. Each year, Australian hospitals perform > 12,000 cardiac surgeries, and a single Brazilian hospital has already performed > 2,900 of these procedures[1,2]. In the United States of America, the cost of cardiac surgery is approximately 1% to 2% of the health budget[3]. The majority of patients undergo coronary artery bypass grafting (CABG), and 74.6% of surgeries are scheduled[4]. Complex cardiac surgery and prolonged hospital length of stay (LOS) may present a high risk of complications and mortality; postoperative mortality has been documented at 4% (valve operations) within the first seven days and 6.4% (overall mortality) within the first postoperative month[4].
Approximately 10.2% to 27.3% of CABG patients present at least one complication, 70.6% after valve surgery, and 84.2% after combined surgery (CABG + valve surgery)[5,6]. Regarding the complications, 2.2% are major adverse cardiovascular events[7], 7.5% are reintubated during the intensive care unit (ICU) stay, which increases the rate of complications[8], 23.2% remain hospitalized in an ICU for more than two days after surgery, and 59.7% remain hospitalized for more than seven days[6]. It seems that when the complication rate increases, hospital LOS and mortality also increase (12% in the ICU and 15.1% in the 30-day period), mainly in older adults[5,9].
Among the prophylactic strategies to decrease these rates of negative outcomes, respiratory care seems to reduce pulmonary complications and minimize postoperative pulmonary dysfunction[10]. As one of the respiratory care techniques, incentive spirometry (IS) is a low-cost, widespread, respiratory exercise technique, used for the prevention and treatment of postoperative pulmonary complications (PPC) in patients undergoing cardiac surgery[11]. IS is a device that provides visual feedback when the patient inhales at a predetermined flow or volume. The patient is required to place the lips firmly around the mouthpiece and to inhale slowly to raise the ball (flow-oriented) or piston/plate (volume-oriented) in the chamber toward the defined target[12].
It has been suggested that patients undergoing cardiac surgery who are more adherent to IS therapy may benefit from a reduced LOS and a reduction in the mortality rate[13]. On the other hand, scientific evidence has suggested that IS does not improve clinical outcomes in different surgical patients[14]. In order to strengthen the scientific findings, our systematic review, performed with strict methodological criteria, is intended to clarify these specific gaps, exclusively in patients undergoing cardiac surgery and assist clinicians in decision making. Our aim was to assess whether IS is superior to respiratory care, mobilization exercises, and noninvasive ventilation (NIV) on PPC, adverse events, mortality, hospital and/or ICU LOS, lung function, oxygenation, and maximal inspiratory pressure (MIP) in patients undergoing cardiac surgery.
METHODS
Design
We conducted a systematic review following the reporting recommendations proposed by the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (or PRISMA)[15]. The protocol was registered in the International Prospective Register of Systematic Reviews (or PROSPERO) (#CRD42020161009), is available online at https://www.crd.york.ac.uk/prospero/export_record_pdf.php), and was previously published[16].
Eligibility Criteria
Types of Studies, Participants, and Interventions
We searched for randomized and quasi-randomized controlled trials published in any year, in any language. The studies included in this review were required to have enrolled patients aged 18 years or older, who were breathing spontaneously, undergoing cardiac surgeries, and which evaluated the effects of postoperative flow or volume-oriented IS on our pre-defined clinical outcomes. The treatment comparison was made with standard care, such as respiratory care (maximal inspiratory breathing exercises, coughing and deep breathing, supported/assisted coughing, huffing technique, diaphragmatic breathing, fractional inspiration, active cycle of breathing, and autogenic drainage), NIV, and other therapies (mobilization exercise, blow bottles, and verbal encouragement). The mobilization exercises considered in this review were early mobilization programs, active/passive exercises of upper/lower limbs, and physical therapy.
The controlled trials had to have evaluated at least one of the following outcomes:
Secondary Outcomes
LOS: The number of days spent in hospital after cardiac surgical procedure.
Length of ICU stay: The number of days spent in the ICU after cardiac surgical procedure.
Lung function: Variables evaluated were peak of expiratory flow (PEF), forced expiratory volume in one second (FEV₁), forced vital capacity (FVC), and vital capacity (VC).
Oxygenation: Arterial partial pressure of oxygen (PaO2) and peripheral and central arterial oxygen saturation (SO2) were accepted.
MIP (cmH₂O): MIP measured with digital or analog manovacuometer or manometer was accepted.
Database and Search Strategy
The search strategy was sensitive (Supplement 1) to capture all potentially qualifying studies through the Medical Literature Analysis and Retrieval System Online (or MEDLINE®), Embase®, Cochrane Central Register of Controlled Trials (or CENTRAL), Physiotherapy Evidence Database (PEDro), Cumulative Index of Nursing and Allied Health (or CINAHL®), Latin American and Caribbean Health Sciences Literature (or LILACS), Scientific Electronic Library Online (or SciELO), and Scopus® databases, as well as in the OpenGrey database, the main clinical trial registration sites, conferences, congresses, and symposiums in the area described in the protocol[16]. When necessary, we contacted the authors of the clinical trials to request additional data. The snowball technique, which consists of searching the reference lists of the included studies, was used to optimize the search. The search was performed on July 22 and 24, 2022.
Study Selection and Data Extraction
Two authors independently selected the studies identified by the search strategy based on eligibility criteria. Duplicate publications were excluded, after which the authors selected the studies by titles and abstracts. Non-randomized trials and studies lacking predefined outcomes were excluded. In some cases, it was necessary to read the full texts. Where reports with the same participants but different outcome measurements or using different time points for the assessments were found, both reports were included. However, the two reports were considered as parts of only one study.
The Rayyan app was used to optimize the process of screening and selecting the studies[17]. Disagreements between authors regarding the inclusion of the study were resolved by a third author. Two authors extracted data independently, and disagreements were also resolved by a third author.
Methodological Rigor of Included Studies and Certainty of Evidence
We assessed the methodological characteristics of the trials using the PEDro scale[18]. We used PEDro scores available at https://pedro.org.au/. Where PEDro scores were not available, two previously trained authors evaluated the clinical trials using the PEDro scale. The PEDro methodological rigor scale ranges between 1 and 10, with higher scores indicating higher quality studies. The studies are classified according to the scores as follows: < 4 are considered “poor”, 4 to 5 are considered “fair”, 6 to 8 are considered “good”, and 9 to 10 are considered “excellent”[19]. We assessed the certainty of evidence using the Classification of Recommendations, Assessment, Development and Evaluation (GRADE)[20], through the software GRADEpro in the main outcomes[21].
Data Analysis
When at least two studies were sufficiently homogeneous in terms of participants, interventions, and outcome measures, we pooled their results in a meta-analysis. Meta-analyses were performed using an inverse variance method and random effects model in Review Manager version 5.3 (The Nordic Cochrane Center, Copenhagen, Denmark)[22]. Continuous variables were analyzed using the weighted mean differences (MD) and for studies that evaluated the same outcome with different instruments, we used the standardized mean differences (SMD) with 95% confidence interval (CI)[23]. Dichotomous variables were analyzed using risk ratios (RR) with 95% CI.
Trials were pooled according to similarity of intervention, populations, and the outcomes measured. Separate meta-analyses were conducted to examine the effects of IS in the following comparisons:
In case of trials that examined the effects of multiple interventions that were of interest for this review, to avoid double counting the participants, we included two reasonably independent comparisons. However, we split the “shared” group sample size (respiratory care) into two smaller sample sizes. For example, Stock et al. (1984)[24] had three groups in its clinical trial: intervention group (with 12 participants), control group 1 (with 13 participants), and control group 2 (with 13 participants). In this situation, the analysis was performed twice; in the first analysis, the intervention group (with six participants [half the original sample size]) was analyzed vs. control group 1. In the second analysis, the intervention group (with six participants [half the original sample size]) was compared with control group 2.
Therefore, in the included clinical trials with three comparison groups (flow-IS group vs. volume-IS vs. respiratory care), and where data were analyzed twice in our study, we initially identified the name of the main author, and then the year of publication, followed by the letter “a” (Amin et al 2021a: flow-IS group vs. respiratory care) and in the second mention, we identified the name of the main author, and then the year of publication, followed by the letter “b” (Amin et al 2021b: volume-IS group vs. second standard care)[25].
Assessment of Heterogeneity
As planned, where appropriate data were available, we carried out subgroup analyses so as to investigate the influence of each comparison on the size of the treatment. Among the preplanned subgroup analyses, it was possible to perform subgroup analyses considering the type of device used (flow-oriented or volume-oriented) in the main comparisons (IS vs. respiratory care; IS vs. NIV; and IS vs. other therapies).
To estimate the heterogeneity across the studies in each meta-analysis, the I2 statistic was used. As suggested in the Cochrane Handbook for Systematic Reviews of Interventions, if heterogeneity was substantial (I2 ≥ 50%), a sensitivity analysis was considered[26]. Although we intended to perform separate analyses for studies with no blinding or deficiency in blinding of outcome assessors, with inappropriate randomization methods, with a large number (> 20%) of patients lost to follow-up, with imputation of standard deviation, or when adherence was not reported, we could not perform sensitivity analyses because we did not find enough studies with appropriate blinding, randomization, or follow-up.
RESULTS
Twenty-three reports of 22 studies were included in this systematic review[27-48]. Twenty-two publications were reported in full; from one clinical trial, only the abstract was reported. One study with two publications was included in this systematic review. The reports of this study were named as Jenkins et al. (1989)[30] and Jenkins et al. (1990)[31], however, as planned, they were considered as part of only one study. The authors of the clinical trial published in abstract format were contacted in an attempt to request additional data[39], however, we did not receive any answers. In this case, we used the data available in the abstract. Twenty-one studies were randomized controlled trials (RCTs) and one was a quasi-randomized trial. The flow chart of this systematic review is shown in Figure 1.
Included Studies
Overall, we included 21 randomized trials and one quasi-randomized controlled trial in this systematic review. The studies involved 1,677 patients, with ages ranging from 38.3 to 65 years[31,45], sample sizes ranging from 16 to 270 participants[43,46], and study follow-up time ranging from two days to hospital discharge (Table 1)[33,39,40,42]. Regarding the characteristics of the surgery and intervention, 74% of patients underwent CABG, 48% of patients received treatment using volume-oriented IS, 39% of patients used flow-oriented IS, and three studies did not have enough information to determine whether the type of spirometer was flowor volume-oriented (Table 2)[39,46,48]. The hospital LOS ranged from 6.5 to 12.5 days, and the length of ICU stay ranged from 2.61 to 6.87 days (Table 3). PaO2 ranged on average from 59.4 mmHg to 99 mmHg[24,28], and SO2 from 79 to 97.7%[35,39].
| Study | Total sample size | Total average age (year) | Study follow-up time |
|---|---|---|---|
| Iverson et al., 1978[27] | 145 | - | - |
| Gale and Sanders, 1980[28] | 109 | - | Until 3rd PO day, or longer if abnormal signs were present in the chest or on the chest X-ray |
| Dull and Dull, 1983[29] | 49 | 57.9 | Until 3rd PO day |
| Stock et al., 1984[24] | 38 | 57.3 | Until 3rd PO day |
| Jenkins et al., 1989[30] | 110 | 55 | Until 5th PO day |
| Jenkins et al., 1990[31] | 110 | 38.3 | Until 5th PO day |
| Oikkonen et al., 1991[32] | 52 | 55 | Until 7th PO day |
| Crowe and Bradley, 1997[33] | 185 | 64.4 | Until hospital discharge |
| Savci et al., 2006[34] | 60 | 56.2 | Until 5th PO day |
| Romanini et al., 2007[35] | 40 | 56.75 | Until 3rd PO day |
| Renault et al., 2009[36] | 36 | 56.8 | Until 7th PO day |
| Dias et al., 2011[37] | 35 | 62.3 | Until 5th PO day |
| El-Kader, 2011[38] | 36 | 48.6 | Until 10th PO day |
| Al-Mutairi et al., 2012a[39] | 72 | 57 | Until 2nd PO day |
| Al-Mutairi et al., 2012b[40] | 108 | 62 | Until 2nd PO day |
| Mueenudheen et al., 2012[41] | 32 | 53.87 | Until 3rd PO day |
| Rizwan et al., 2012[42] | 32 | 38.34 | Until 2nd PO day |
| Zangerolamo et al., 2013[43] | 16 | 64.65 | Until discharge from the intensive care unit |
| Yazdannik et al., 2016[44] | 50 | 57.25 | Until 3rd PO day |
| Manapunsopee et al., 2019[45] | 90 | 65 | Until 4th PO day |
| Alam et al., 2020[46] | 270 | 46.9 | Until 3rd PO day |
| Amin et al., 2021[47] | 72 | 62.56 | Until 7th PO day |
| Barkhordari-Sharifabad et al., 2021[48] | 40 | 64.1 | Until 4th PO day |
PO=postoperative
| Study | Type of surgery | ECC time (minutes) | Intervention with IS | Intervention with standard care | |||||
|---|---|---|---|---|---|---|---|---|---|
| IS | Standard care | Technique | Frequency of use | Standard care 1 | Standard care 2 | ||||
| Technique | Frequency of use | Technique | Frequency of use | ||||||
| Iverson et al., 1978[27] | CABG | - | - | Volume-oriented IS | Three to five times every 3 hours | IPPB | 15-min. treatment every 3 hours with 15 to 20 cmH₂0 | Blow bottles | Three to five times every 3 hours |
| Gale and Sanders, 1980[28] | CABG | - | - | Volume-oriented IS | 10 deep breaths in a treatment time of 20 min. | IPPB | 20 min. with inspiratory pressure of 20 cmH₂O | - | - |
| Dull and Dull, 1983[29] | CABG + VR | - | Volume-oriented IS | 10 repetitions of maximal inhalation, four times a day | Early mobilization | Early mobilization twice a day | Early mobilization + respiratory care | 10 repetitions of maximal inhalation four times a day | |
| Stock et al., 1984[24] | CABG + VR + CABG and VR or aneurysmectomy | - | Volume-oriented IS | 15 min. - every 2 hours during waking hours, from the second to the 72nd hour after extubation | Coughing + deep breathing | 15 min. - every 2 hours during waking hours, from the second to the 72nd hour after extubation | CPAP | Pressure of 7.5 cmH₂O - two or three maximal inspirations every 3 to 5 min. | |
| Jenkins et al., 1989[30] | CABG | - | Flow‑oriented IS | 3 to 5 consecutive breaths, at least twice on days 1 and 2 and at least once daily on days 3 to 5 | Respiratory care | 3 to 5 consecutive deep breaths, at least twice on days 1 and 2 and at least once daily on days 3-5 | Airway clearance exercises + physical therapy | 3 to 5 consecutive deep breaths, at least twice on days 1 and 2 and at least once daily on days 3-5 | |
| Jenkins et al., 1990[31] | CABG | - | Flow‑oriented IS | Inspiration leaving the balls floating in the first and second chambers - 3 to 5 breaths | Respiratory care | Deep inspiration. Between 3 and 5 consecutive deep breaths | Verbal encouragement | Treatment consisted solely of encouraging (verbally) patients to huff and cough and early mobilization | |
| Oikkonen et al., 1991[32] | CABG | 96 ± 6 | 108 ± 6 | Volume-oriented IS | Inhalation, exceeding 3 seconds and repeated at least 5 times per training | IPPB | Pressure of 10 to 15 cmH₂O - 5 to 10 min. in each session | - | - |
| Crowe and Bradley, 1997[33] | CABG | - | - | Volume-oriented IS | Provided once or twice per day, encouraged by other members of the health care team | Respiratory care | Lung expansion maneuvers and secretion-removal maneuvers. Provided once or twice per day | - | - |
| Savci et al., 2006[34] | CABG | - | - | Volume-oriented IS | Twice a day. From the 3rd day, once a day for 15 min. | Respiratory care | 1-2 breathing control breaths | - | - |
| Romanini et al., 2007[35] | CABG | 57.50 ± 11.53 | 54 ± 9.26 | Volume-oriented IS | 10 min., an interval of 5 min. and 10 min. again. | IPPB | 10 min., an interval of 5 min. and 10 min. again | - | - |
| Renault et al., 2009[36] | CABG | 84.77 ± 32.29 | 80.94 ± 25.34 | Flow‑oriented IS | 2 times a day (ICU), and once a day (inpatient unit) | Respiratory care | 3 sets of 10 breathing exercises + assisted cough and huffing + early mobilization | - | - |
| Dias et al., 2011[37] | CABG and VR | - | - | Volume-oriented IS | Twice a day for 5 days | Bronchial hygiene + mobilization | Twice a day for 5 days | Bronchial hygiene + mobilization + inspiratory exercise | Twice a day for 5 days |
| El-Kader, 2011[38] | CABG | - | - | Volume-oriented IS | Application of 5 min., five times a day | CPAP | Application of 15 min. and pressure = l0 cmH₂O every day | IPPB | Application of IPPB 15 min./day |
| Al-Mutairi et al., 2012a[39] | Any heart surgery | - | - | IS | Used IS 15 times per hour for 3 days | CPAP | 4-6 cmH₂O | - | - |
| Al-Mutairi et al., 2012b[40] | Any heart surgery | - | - | Volume-oriented IS | 15 times per hour for 3 days | CPAP (2 hours) | 4-6 cmH₂O for half hour every 2 hours for 3 days | CPAP (4 hours) | 4-6 cmH₂O for half hour every 4 hours for 3 days |
| Mueenudheen et al., 2012[41] | CABG | - | - | Flow‑oriented IS | 3 sets of 10 breaths with a pause of 1 min. between each set | Respiratory care | 3 sets of 10 consecutive breaths with a pause of 1 min. between each set | - | - |
| Rizwan et al., 2012[42] | Mitral valve replacement surgery | - | - | Flow‑oriented IS | 3 sets of 10 deep breaths with 30-60 seconds to rest | Respiratory care | 3 sets of 10 deep breaths with 30-60 seconds to rest | - | - |
| Zangerolamo et al., 2013[43] | CABG | 64.3 ± 11.1 | 57.5 ± 10 | Flow‑oriented IS | 3 sets of 10 repetitions | Respiratory care | For each exercise, three sets of 10 repetitions. 3 sessions per day | - | - |
| Yazdannik et al., 2016[44] | CABG | - | - | Flow‑oriented IS | 10 times breathing with IS every 2 hours in the daytime for three days | Respiratory care | Only usual exercise | - | - |
| Manapunsopee et al., 2019[45] | CABG | 109.0 ± 55.0 | 129.0 ± 57.0 | Flow‑oriented IS + respiratory care | 10 times per hour - slow maximal inhalations | Respiratory care | Breathing exercise 10 times per hour | - | - |
| Alam et al., 2020[46] | CABG | - | - | IS | Breathing exercise with IS | Standard physiotherapy + Acapella | Breathing exercise + Acapella | - | - |
| Amin et al., 2021[47] | CABG | - | - | Flow‑oriented IS | 3 sets of 5 repeated deep breaths - four times a day | Volume-oriented IS | 3 sets of 5 repeated deep breaths - four times in a day | Respiratory care | 3 sets of 5 deep breaths - 4 times in a day |
| Barkhordari-Sharifabad et al., 2021[48] | CABG | - | - | IS | 10 deep breaths, every 2 hours during awakening | Respiratory care | 10 times with 2-hour interval when the patient woke up | - | - |
CABG=coronary artery bypass grafting; CPAP=continuous positive airway pressure; ECC=extracorporeal circulation; ICU=intensive care unit; IPPB=intermittent positive pressure breathing; IS=incentive spirometry; VR=valve replacement
| Study | Outcomes observed | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| PPC (n,%) | Adverse events (n,%) | Mortality (n,%) | Length of hospital stay (days) (mean ± SD) | Length of intensive care unit stay (days) (mean ± SD) | ||||||
| ISG | CG | ISG | CG | ISG | CG | ISG | CG | ISG | CG | |
| Iverson et al., 1978a[27] | 35 (60.3)* | 23 (54.7)* and 1 (2.3)** | 0 | 1 (2.4) | 0 | 1 (2.4) | - | - | - | - |
| Iverson et al., 1978b[27] | 35 (60.3)* | 18 (40)* | 0 | 0 | 0 | 0 | - | - | - | - |
| Gale and Sanders, 1980[28] | 51 (98)* | 57 (100)* | - | - | - | - | - | - | - | - |
| Dull and Dull, 1983[29] | - | - | - | - | - | - | - | - | - | - |
| Stock et al., 1984a[24] | 11 (92)* | 12 (92)* | 0 (0) | 1 (8) | - | - | - | - | - | - |
| Stock et al., 1984b[24] | 11 (92)* | 9 (67)* | 0 (0) | 2 (15) | - | - | - | - | - | - |
| Jenkins et al., 1989a[30] | 28 (74)* and 2 (5.2)** | 26 (74)* and 4 (11.4)** | - | - | - | - | - | - | - | - |
| Jenkins et al., 1989b[30] | 28 (74)* and 2 (5.2)** | 28 (75)* and 5 (13.5)** | - | - | - | - | - | - | - | - |
| Jenkins et al., 1990a[31] | 28 (74)* and 2 (5.2)** | 26 (74)* and 4 (11.4)** | - | - | - | - | - | - | - | - |
| Jenkins et al., 1990b[31] | 28 (74)* and 2 (5.2)** | 28 (75)* and 5 (13.5)** | - | - | - | - | - | - | - | - |
| Oikkonen et al., 1991[32] | 21 (80.7)* | 16 (61.5)* | 11 (24.3) | 13 (50) | - | - | - | - | - | - |
| Crowe and Bradley, 1997[33] | 9 (10)* and 8 (8.9)** | 10 (10.5)* and 10 (10.5)** | - | - | - | - | 9 ± 3.1 | 9.7 ± 4.9 | - | - |
| Savci et al., 2006[34] | 9 (30)* | 10 (33.3)* | - | - | - | - | - | - | - | - |
| Romanini et al., 2007[35] | - | - | - | - | - | - | - | - | - | - |
| Renault et al., 2009[36] | - | - | - | - | - | - | - | - | 2.61 ± 0.69 | 3.22 ± 1.06 |
| Dias et al., 2011a[37] | - | - | 0 (0) | 0 (0) | - | - | - | - | - | - |
| Dias et al., 2011b[37] | - | - | 0 (0) | 0 (0) | - | - | - | - | - | - |
| El-Kader, 2011[38] | - | - | - | - | - | - | - | - | - | - |
| Al-Mutairi et al., 2012a[39] | - | - | - | - | - | - | - | - | - | - |
| Al-Mutairi et al., 2012b1[40] | - | - | - | - | 0 (0) | 1 (2.8) | 9.5 ± NR | 8.7 ± NR | - | - |
| Al-Mutairi et al., 2012b2[40] | - | - | - | - | 0 (0) | 2 (5.6) | 9.5 ± NR | 9 ± NR | - | - |
| Mueenudheen et al., 2012[41] | - | - | - | - | - | - | - | - | - | - |
| Rizwan et al., 2012[42] | - | - | - | - | - | - | - | - | - | - |
| Zangerolamo et al., 2013[43] | - | - | - | - | 1 (12.5) | 2 (25) | 6.5 ± 1.69 | 8.25 ± 2.6 | 5 ± 1.6 | 6.87 ± 2.74 |
| Yazdannik et al., 2016[44] | - | - | - | - | - | - | - | - | - | - |
| Manapunsopee et al., 2019[45] | 3 (6.4)* and 1 (2.1)** | 5 (11.6)* | 17 (36.2) | 10 (23.2) | - | - | 6.75 ± 7.77 | 12.5 ± 18.82 | - | - |
| Alam et al., 2020[46] | - | - | - | - | - | - | - | - | - | - |
| Amin et al., 2021[47] | - | - | - | - | - | - | - | - | - | - |
| Barkhordari-Sharifabad et al., 2021[48] | - | - | - | - | - | - | - | - | - | - |
CG=control group; ISG=incentive spirometry group; NR=not registered; PPC=postoperative pulmonary complications; SD=standard deviation
* Atelectasis
** Pneumonia
Iverson et al. 1978a=comparison of intervention group vs. first control group; Iverson et al. 1978b=comparison of intervention group vs. second control group; Stock et al. 1984a=comparison of intervention group vs. first control group; Stock et al. 1984b=comparison of intervention group vs. second control group; Jenkins et al. 1989a=comparison of intervention group vs. first control group; Jenkins et al. 1989b=comparison of intervention group vs. second control group; Jenkins et al. 1990a=comparison of intervention group vs. first control group; Jenkins et al. 1990b=comparison of intervention group vs. second control group; Dias et al. 2011a=comparison of intervention group vs. first control group; Dias et al. 2011b=comparison of intervention group vs. second control group; Al-Mutairi et al. 2012b1=comparison of intervention group vs. first control group; Al-Mutairi et al. 2012b2=comparison of intervention group vs. second control group
Considering the primary outcomes analysis, among the included studies, nine clinical trials reported PPC rate[24,27,28,30-34,45], five reported adverse events rate[24,27,32,37,45], and three reported mortality rate (Table 3)[26,40,43].
With respect to the secondary outcomes analysis, four trials reported LOS[33,40,43,45], two reported ICU LOS[36,43], eight reported parameters of lung function[24,29,30,33,34,36,41,47], ten reported PaO2, nine reported SO2, and one reported reintubation rate. No trials evaluated the use of antibiotics (which was an outcome of interest for this review[16]).
For these continuous outcomes, results were reported differently across studies, and we performed transformations where it was adequate. In two clinical trials, PaO2 was converted from kilopascals to millimeters of mercury and in one clinical trial the standard deviation was estimated using the Revman calculator[28,31,32]. For some studies the standard deviation was also estimated using the Revman calculator[24,37,39,47]. In one clinical trial, LOS was registered as median, with minimum and maximum, and this was converted to mean and standard deviation for our analysis[45,49] (Table 4). For some studies, transformations were not possible. For instance, one clinical trial recorded forced expiratory flow without standard deviation[29], and insufficient information to estimate the standard deviation. Therefore, we did not pool the results in the meta-analysis. When results were presented using different measures, such as those from studies reporting lung function, which reported values both as a percentage of predicted values and as absolute values in liters, then results were pooled using the SMD.
| Lung function | Oxygenation | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Study | FEV1% | FVC% | PEF (L/min) | VC (%) | FEF (%) | PaO2 (mmHg) | SO2 (%) | |||||||
| ISG | CG | ISG | CG | ISG | CG | ISG | CG | ISG | CG | ISG | CG | ISG | CG | |
| Iverson et al., 1978a[27] | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| Iverson et al., 1978b[27] | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| Gale and Sanders, 1980[28] | - | - | - | - | - | - | 1.8 ± 1.4 L | 1.4 ± 1.5 L | - | - | 60.6 ± 13.7 | 59.4 ± 12.07 | - | - |
| Dull and Dull, 1983a[29] | 37 ± NR | 40 ± NR | 35 ± NR | 35 ± NR | - | - | - | - | 55 ± NR | 70 ± NR | - | - | - | - |
| Dull and Dull, 1983b[29] | 37 ± NR | 40 ± NR | 35 ± NR | 41 ± NR | - | - | - | - | 55 ± NR | 53 ± NR | - | - | - | - |
| Stock et al., 1984a[24] | 0.76 ± 0.173 L | 0.63 ± 0.288 L | 0.96 ± 0.243 L | 0.76 ± 0.324 L | - | - | - | - | - | - | 93 ± 17 | 99 ± 21 | - | - |
| Stock et al., 1984b[24] | 0.76 ± 0.173 L | 0.59 ± 0.252 L | 0.96 ± 0.243 L | 0.73 ± 0.283 L | - | - | - | - | - | - | 93 ± 17 | 94 ± 17 | - | - |
| Jenkins et al., 1989a[30] | 1.8 ± 0.5 L | 1.7 ± 0.4 L | 2.4 ± 0.5 L | 2.2 ± 0.5 L | 2.55 ± 0.6 | 2.50 ± 0.6 | - | - | - | - | 60 ± 6.7 | 67.5 ± 7.5 | - | - |
| Jenkins et al., 1989b[30] | 1.8 ± 0.5 L | 1.8 ± 0.5 L | 2.4 ± 0.5 L | 2.3 ± 0.6 L | 2.55 ± 0.6 | 2.53 ± 0.67 | - | - | - | - | 60 ± 6.7 | 66.7 ± 7.5 | - | - |
| Jenkins et al., 1990a[31] | - | - | - | - | - | - | 2.6 ± 0.1 L | 2.5 ± 0.1 L | - | - | 60 ± 7.5 | 60 ± 7.5 | - | - |
| Jenkins et al., 1990b[31] | - | - | - | - | - | - | 2.6 ± 0.1 L | 2.7 ± 0.2 L | - | - | 60 ± 7.5 | 60 ± 7.5 | - | - |
| Oikkonen et al., 1991[32] | - | - | - | - | - | - | - | - | - | - | 75 ± 7.5 | 82.5 ± 7.5 | - | - |
| Crowe and Bradley, 1997[33] | 81 ± 4 | 83 ± 4 | 82 ± 6 | 85 ± 4 | - | - | - | - | - | - | - | - | 90 ± NR | 79 ± NR |
| Savci et al., 2006[34] | 57.26 ± 14.6 | 64.98 ± 12.95 | 57.6 ± 14.17 | 63.17 ± 11.65 | 4.54 ± 1.44 | 5.90 ± 1.96 | 53.18 ± 13.6 | 57.76 ± 9.47 | - | - | 76.08 ± 13.69 | 79.69 ± 18.26 | 95.74 ± 2.13 | 94.59 ± 4.33 |
| Romanini et al., 2007[35] | - | - | - | - | - | - | - | - | - | - | - | - | 91.15 ± 83.2 | 94.7 ± 86.4 |
| Renault et al., 2009[36] | 1.12 ± NR L | 1.37 ± NR L | 1.37 ± NR L | 1.27 ± NR L | - | - | - | - | - | - | - | - | - | - |
| Dias et al., 2011a[37] | - | - | 46.7 ± 52.08 | 51.3 ± 37.09 | - | - | - | - | - | - | - | - | 97.2 ± NR | 97 ± NR |
| Dias et al., 2011b[37] | - | - | 46.7 ± 52.08 | 54.3 ± 49.64 | - | - | - | - | - | - | - | - | 97.2 ± NR | 97.7 ± NR |
| El-Kader, 2011a[38] | - | - | - | - | - | - | - | - | - | - | 82.91 ± 2.3 | 71.66 ± 4 | - | - |
| El-Kader, 2011b[38] | - | - | - | - | - | - | - | - | - | - | 82.91 ± 2.3 | 74.5 ± 4.8 | - | - |
| Al-Mutairi et al., 2012a[39] | - | - | - | - | - | - | 1.59 ± 3.9 L | 1.88 ± 4.6 L | - | - | - | - | 96.53 ± 181.5 | 96.83 ± 182.1 |
| Al-Mutairi et al., 2012b1[40] | - | - | - | - | - | - | - | - | - | - | - | - | 95.6 ± 0.4 | 97.17 ± 0.43 |
| Al-Mutairi et al., 2012b2[40] | - | - | - | - | - | - | - | - | - | - | - | - | 95.6 ± 0.4 | 96 ± 0.6 |
| Mueenudheen et al., 2012[41] | 52 ± 10 | 54 ± 9 | 46 ± 9 | 45 ± 8 | - | - | - | - | - | - | 79.77 ± 9.22 | 82.85 ± 10.53 | - | - |
| Rizwan et al., 2012[42] | - | - | - | - | - | - | - | - | - | - | 76.41 ± 15.85 | 75.35 ± 11.55 | 94.72 ± 2.9 | 94.7 ± 1.95 |
| Zangerolamo et al., 2013[43] | - | - | - | - | - | - | 1.68 ± 0.48 L | 1.4 ± 0.4 L | - | - | - | - | - | - |
| Yazdannik et al., 2016[44] | - | - | - | - | - | - | - | - | - | - | 72.7 ± 7.1 | 82.3 ± 4.8 | 96.8 ± 1.4 | 90.5 ± 2.1 |
| Manapunsopee et al., 2019[45] | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| Alam et al., 2020[46] | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
| Amin et al., 2021a[47] | 0.96 ± 1.2 L | 0.86 ± 1.1 L | 1.10 ± 1.4 L | 0.90 ± 1.1 L | 1.44 ± 1.8 L | 1.08 ± 1.4 L | - | - | - | - | - | - | - | - |
| Amin et al., 2021b[47] | 1.26 ± 15.2 L | 0.86 ± 1.1 L | 1.53 ± 1.9 L | 0.90 ± 1.1 L | 2.22 ± 44.7 L | 1.08 ± 1.4 L | - | - | - | - | - | - | - | - |
| Barkhordari-Sharifabad et al., 2021[48] | - | - | - | - | - | - | - | - | - | - | - | - | 96.5 ± 1.50 | 95.2 ± 2.38 |
CG=control group; FEF=forced expiratory flow; FEV₁=forced expiratory volume in one second; FVC=forced vital capacity; ISG=incentive spirometry group; NR=not registered; PaO2=partial pressure of oxygen; PEF=peak of expiratory flow; SO2=oxygen saturation; VC=vital capacity
Iverson et al. 1978a=comparison of intervention group vs. first control group; Iverson et al. 1978b=comparison of intervention group vs. second control group; Dull and Dull 1983a=comparison of intervention group vs. first control group; Dull and Dull 1983b=comparison of intervention group vs. second control group; Stock et al. 1984a=comparison of intervention group vs. first control group; Stock et al. 1984b=comparison of intervention group vs. second control group; Jenkins et al. 1989a=comparison of intervention group vs. first control group; Jenkins et al. 1989b=comparison of intervention group vs. second control group; Jenkins et al. 1990a=comparison of intervention group vs. first control group; Jenkins et al. 1990b=comparison of intervention group vs. second control group; Dias et al. 2011a=comparison of intervention group vs. first control group; Dias et al. 2011b=comparison of intervention group vs. second control group; El-Kader 2011a=comparison of intervention group vs. first control group; El-Kader 2011b=comparison of intervention group vs. second control group; Al-Mutairi et al. 2012b1=comparison of intervention group vs. first control group; Al-Mutairi et al. 2012b2=comparison of intervention group vs. second control group; Amin et al. 2021a=comparison of intervention group vs. first control group; Amin et al. 2021b=comparison of intervention group vs. second control group
Assessment of Methodological Rigor
Among the included studies, in general, the PEDro score ranged from 2 to 7 points, with a mean and standard deviation of 4.5±1.1. For seven trials, the scores were not available on the PEDro platform, therefore, the scores were independently graded by two authors[27,39,42-44,47,48]. After the evaluation of the two authors, three inconsistencies were observed, one on item 11 and two on item 8[27,42,44]. In this situation, a third author was consulted to arbitrate. Considering the PEDro scale, the following percentages of studies did not meet the criteria: on item 1, 21.7%; on item 2, 8.7%; on item 3, 95.7%; on item 4, 13%; on items 5 and 6, 100%; on item 7, 78.2%; on item 8, 52.1%; on item 9, 95.7%; and on item 11, 8.7%. On item 10, all studies met the criteria. In the classification of the PEDro scale, 16 (69.6%) studies were judged as having “fair”[24,27,28,30,31,35,37-44,47,48], four (17.4%) as “good”[32-34,45], three (13%) as “poor”[29,36,46], and zero (0%) as “excellent” quality (Table 5). Considering the low methodological rigor of the studies included in this review, we were not able to perform sensitivity analysis including only high-quality studies.
| Study | Total score | Item 1 | Item 2 | Item 3 | Item 4 | Item 5 | Item 6 | Item 7 | Item 8 | Item 9 | Item 10 | Item 11 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Iverson et al., 1978[27] | 4* | N | N | N | Y | N | N | N | Y | N | Y | Y |
| Gale and Sanders, 1980[28] | 4 | Y | Y | N | N | N | N | N | Y | N | Y | Y |
| Dull and Dull, 1983[29] | 3 | N | Y | N | Y | N | N | N | N | N | Y | N |
| Stock et al., 1984[24] | 5 | N | Y | N | Y | N | N | N | Y | N | Y | Y |
| Jenkins et al., 1989[30] | 5 | Y | Y | N | Y | N | N | Y | N | N | Y | Y |
| Jenkins et al., 1990[31] | 4 | Y | Y | N | Y | N | N | N | N | N | Y | Y |
| Oikkonen et al., 1991[32] | 6 | Y | Y | N | Y | N | N | Y | Y | N | Y | Y |
| Crowe and Bradley, 1997[33] | 6 | Y | Y | N | Y | N | N | Y | Y | N | Y | Y |
| Savci et al., 2006[34] | 6 | Y | Y | N | Y | N | N | Y | Y | N | Y | Y |
| Romanini et al., 2007[35] | 4 | N | Y | N | Y | N | N | N | N | N | Y | Y |
| Renault et al., 2009[36] | 2 | Y | N | N | Y | N | N | N | N | N | Y | N |
| Dias et al., 2011[37] | 4 | Y | Y | N | Y | N | N | N | N | N | Y | Y |
| El-Kader, 2011[38] | 5 | N | Y | N | Y | N | N | N | Y | N | Y | Y |
| Al-Mutairi et al., 2012a[39] | 5* | Y | Y | N | Y | N | N | N | Y | N | Y | Y |
| Al-Mutairi et al., 2012b[40] | 4 | Y | Y | N | N | N | N | N | Y | N | Y | Y |
| Mueenudheen et al., 2012[41] | 4 | Y | Y | N | Y | N | N | N | N | N | Y | Y |
| Rizwan et al., 2012[42] | 4* | Y | Y | N | Y | N | N | N | N | N | Y | Y |
| Zangerolamo et al., 2013[43] | 4* | Y | Y | N | Y | N | N | N | N | N | Y | Y |
| Yazdannik et al., 2016[44] | 4* | Y | Y | N | Y | N | N | N | N | N | Y | Y |
| Manapunsopee et al., 2019[45] | 7 | Y | Y | Y | Y | N | N | Y | N | Y | Y | Y |
| Alam et al., 2020[46] | 3 | Y | Y | N | N | N | N | N | N | N | Y | Y |
| Amin et al., 2021[47] | 5* | Y | Y | N | Y | N | N | N | Y | N | Y | Y |
| Barkhordari-Sharifabad et al., 2021[48] | 5* | Y | Y | N | Y | N | N | N | Y | N | Y | Y |
N=no; Y=yes
* =Score assessed by authors
Item 1=eligibility criteria were specified (item 1 is not included in the total score calculation); item 2=subjects were randomly allocated to groups (in a crossover study, subjects were randomly allocated an order in which treatments were received); item 3=allocation was concealed; item 4=the groups were similar at baseline regarding the most important prognostic indicators; item 5=there was blinding of all subjects; item 6=there was blinding of all therapists who administered the therapy; item 7=there was blinding of all assessors who measured at least one key outcome; item 8=measures of at least one key outcome were obtained from more than 85% of the subjects initially allocated to groups; item 9=all subjects for whom outcome measures were available received the treatment or control condition as allocated or, where this was not the case, data for at least one key outcome was analyzed by “intention to treat”; item 10=the results of between-group statistical comparisons are reported for at least one key outcome; item 11=the study provides both point measures and measures of variability for at least one key outcome
Comparisons of Interventions
We rated the certainty of the evidence for each outcome in all comparisons using the GRADE approach[21]. The details of each evaluation can be found in Supplement 2.
Incentive Spirometry vs. Respiratory Care
Primary Outcomes
There may be a small difference or no difference on PPC rate between IS and respiratory care (RR 0.91; 95% CI 0.72 to 1.14) (low certainty of evidence) (Supplement 3 - Figure 2A). The evidence is of very low certainty for the other primary outcomes. Only one trial evaluated the mortality rate[43]. This trial also used flow IS and compared its effects to the effects of respiratory care (Supplement 3 - Figure 2B). In the same way, only one trial evaluated the adverse events[45]. This trial used flow IS and compared its effects to the effects of respiratory care (Supplement 3 - Figure 2C).
Secondary Outcomes
We found low certainty of evidence that there may be a small or no difference on FEV1 between IS and respiratory care (SMD -0.16; 95% CI -0.48 to 0.16) (Supplement 3 - Figure 3D). The evidence is of very low certainty for all the other secondary outcomes of this comparison. For these outcomes, no differences in LOS (MD -1.38; 95% CI -2.96 to 0.21), length of ICU stay (MD -0.78; 95% CI -1.61 to 0.06), PEF (MD -0.60; 95% CI -1.97 to 0.78), FVC (SMD 0.14; 95% CI -0.40 to 0.67), VC (SMD 0.38; 95% CI -0.59 to 1.34), and SO2 (MD 2.54; 95% CI -1.74 to 6.82) were observed, comparing IS and respiratory care (Supplement 3 - Figures 3A, 3B, 3C, 3E, 3F, and 3H). However, in the subgroup analysis of VC, flow IS was superior compared to respiratory care. The Amin et al. (2021)[47] study was not included in the PEF meta-analysis as it did not have sufficient extractable data. The Barkhordari-Sharifabad et al. (2021)[48] study was not included in the SO2 meta-analysis as it was unclear whether it used flow-oriented IS or volume-oriented IS.
The meta-analysis showed that IS leads to lower recovery of PaO2 than respiratory care (MD -4.48; 95% CI -8.32 to -0.63) (very low certainty of evidence). In the subgroup analyses, flow-oriented IS was inferior to recovery PaO2 compared to respiratory care (Supplement 3 - Figure 3G). Two trials evaluated MIP[36,45], however, only one had sufficient extractable data (Supplement 3 - Figure 3I)[45].
Incentive Spirometry vs. Other Therapies
Primary Outcomes
The evidence for the primary outcomes of IS vs. other therapies is of very low certainty. We found no differences on PPC between IS and other therapies (RR 1.04; 95% CI 0.73 to 1.49) (Supplement 3 - Figure 4A). Only one study evaluated adverse events (Supplement 3 - Figure 4B).
Secondary Outcomes
The evidence for the secondary outcomes is also of very low certainty. No difference was observed between IS and other therapies regarding FEV1 (MD 0.08; 95% CI -0.08 to 0.25), FVC (SMD 0.15; 95% CI -0.25 to 0.55), and PaO2 (MD -3.63; 95% CI -9.18 to 1.93) (very low certainty of evidence) (Supplement 3 - Figures 5B, 5C, and 5E). Only one study evaluated PEF[30], and another study evaluated VC (Supplement 3 - Figure 5A, 5D)[31].
Incentive Spirometry vs. NIV
Primary Outcomes
Four trials compared the effects of IS vs. NIV on PPC, and three trials on mortality and adverse events. The evidence for the primary outcomes of IS vs. NIV is of very low certainty. All trials used volume-oriented IS. No differences were found between volume-oriented IS and NIV on PPC (RR 1.14; 95% CI 0.84 to 1.55), mortality (RR 0.49; 95% CI 0.08 to 2.93), and adverse events (RR 1.10; 95% CI 0.62 to 1.95) (Supplement 3 - Figures 6A, 6B, and 6C).
Secondary Outcomes
The evidence for secondary outcomes is also of very low certainty. No differences were found between IS and NIV on PaO2 (MD 2.95; 95% CI -4.69 to 10.58) or on SO2 (MD -0.99; 95% CI -2.12 to 0.14) (Supplement 3 - Figures 7D, 7E). Only one trial compared the effects of IS and NIV on FEV1, FVC, VC, and MIP. All trials used volume-oriented IS (Supplement 3 - Figures 7A, 7B, 7C, and 7F). A single study recorded the reintubation rate, with zero reintubation in the IS group and one reintubation in the standard care group[32].
DISCUSSION
To the best of our knowledge, this is the first systematic review with meta-analysis to investigate the effects of IS exclusively in patients undergoing cardiac surgery, performing sub-analysis to pool the studies according to the type of IS used as respiratory care. The results showed that the use of IS was not superior to respiratory care, other therapies, and NIV on the outcomes evaluated. On the other hand, IS was inferior to respiratory care for recovery PaO2. In the subgroup analysis, flow-oriented IS was inferior to respiratory care on recovery PaO2. However, flow-oriented IS was superior to respiratory care on VC. Overall, the methodological rigor of the clinical trials included in this review was “fair” and the certainty of evidence ranged from “very low” to “low”.
In general, although our meta-analysis showed that IS is not different from respiratory care, other therapies, or NIV, except for PaO2 (in IS vs. respiratory care) for which we cannot make any positive or negative statements about effectiveness after cardiac surgery, the majority of the included studies present severe methodological problems and inadequate sample size. In addition, over the years studies have investigated the effects of IS on PPC, adverse events, and mortality after surgical procedures on the thorax, showing different results, some in agreement with and others contrary to our findings[11,50,51].
Our results are in line with a previous Cochrane systematic review that included seven RCTs with a total of 592 patients to assess the effects of IS for preventing pulmonary complications after CABG[51]. This review found no evidence of a benefit from IS in reducing pulmonary complications and in decreasing the negative effects on pulmonary function in patients undergoing CABG. Of note, besides including only patients that had undergone CABG, this review is outdated and did not perform the certainty of evidence evaluation. The inclusion of a broader and updated body of knowledge and GRADE assessments in our review is of particular importance, as it facilitates decision making of physiotherapists working in the frontline.
A clinical trial investigated the effects of IS after cardiac surgery in 90 patients; 47 patients were treated with flow-oriented IS + deep breathing exercise, and 43 patients received only deep breathing exercise (control group)[45]. Patients who received IS + deep breathing exercise had no reduction in atelectasis, pneumonia, pneumothorax, and pleural effusion. However, the control group had fewer adverse events (dyspnea) (P-value = 0.03)[45]. On the other hand, one thing is certain, although, to date, the clinical efficacy on PPC is not proven, IS is widely used and investigated[52].
A preliminary trial[53] that investigated the effectiveness of IS (flow-oriented device) on respiratory motion in healthy subjects suggested that two weeks of respiratory training using IS is useful for improving respiratory motion and pulmonary function. A clinical trial[54] with 260 surgical patients (non-cardiac patients) showed that IS (flow-oriented and volume-oriented) and diaphragmatic breathing exercise better preserve pulmonary function and diaphragm excursion. If these findings are also demonstrated in patients after cardiac surgery using IS, this method will represent an easily accessible and low-cost device to be used in the treatment of these patients.
A broad range of different types of IS devices and treatment protocols were used in the studies included in this review. However, we were unable to determine which of them is more effective. Although we planned to perform other subgroup analyses, we were also unable to identify whether the type of surgery, the severity of the disease, or details of the intervention, such as frequency, duration, and time the intervention started could influence the effect of intervention. Due to the heterogeneity of the RCTs regarding the combinations of interventions and comparisons, different comparisons had to be made, and we were only able to perform a few comprehensive meta-analyses. Therefore, the precision of effect estimates was jeopardized.
Furthermore, due to several methodological limitations in the included studies and conflicting results, further well-designed trials, with long-term follow-up, and which report the rate of core outcome results, such as PPC, adverse events, mortality, lung function, and LOS, are needed, as well as in the ICU. New RCTs should be standardized to provide more homogeneous and reliable data to properly compare the results. For example, studies should evaluate the same IS device, delivered using standardized protocols, for treating similar types of surgeries.
Of note, some limitations should be underscored. In addition, there is a need for clear and complete reporting of outcome data for the interventions being compared. All trials included in this review had important methodological limitations. Although blinding of participants and personnel may be very difficult from a practical perspective, several other factors such as the lack of blinding of outcome assessors, loss to follow-up, and the absence of intention-to-treat analyses were common methodological limitations in the available studies.
Overall, due to the serious risk of bias and imprecision, the overall certainty of the available evidence is very low, and several questions persist. Thus, it is unclear whether IS used alone or in combination with other therapies is effective when compared to other interventions used alone or in combination.
Moreover, although some studies concluded that IS was safe, the available information on adverse events was insufficient to perform a comprehensive meta-analysis that could provide more accurate results on the safety of IS. The evidence is currently insufficient to support or refute the routine use of IS after cardiac surgeries. The results of the six ongoing RCTs are necessary to provide more precise and reliable information on which to base further trials and protocols, and to guide clinical decision-making processes on the use of IS after cardiac surgeries.
We believe the strengths of this systematic review include transparency, rigid methods, assessment of the quality of evidence for each outcome, and extensive and careful searches, with no restrictions on language or publication date. We searched the gray literature database and ongoing studies and performed a rigorous critical assessment of the current body of evidence. Furthermore, the assessment of certainty of evidence using the GRADE approach is paramount in pointing out limitations in current trials and upon which to base further high quality RCTs. Another strong point of this review was the separate analysis by the type of IS (flowor volume-oriented IS), when possible. This high-quality review underlines that there is an urgent need to conduct high-quality RCTs in this field.
Limitations
We consider as limitations of this systematic review the inclusion of biased clinical trials, such as those with lack of blinding of outcome assessors, or without adequate randomization; substantial heterogeneity among studies that made them unsuitable for meta-analysis; or studies with small samples that do not allow us to provide accurate estimates of the effects. As another limitation, we were unable to explain the heterogeneity in the meta-analysis of the PaO2 and SO2 outcomes.
CONCLUSION
This meta-analysis revealed that IS was not superior to standard respiratory care for PPC and clinical outcomes, therefore its use should not be widely recommended until high-quality further studies are performed to ensure this clinical guidance.
REFERENCES
1.
2.
3.
4.
5.
6.
8.
9.
10.
11.
12.
13.
14.
15. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 2009;6(7):e1000097. doi:10.1371/journal.pmed.1000097.
16. Santos EC, Pinto ACPN, Macedo JRFF, Lunardi AC. Effect of incentive spirometry after cardiac surgery: protocol for a systematic review. Fisioter Bras. 2020;21(1):117-25. doi:10.33233/fb.v21i1.3625.
17. Ouzzani M, Hammady H, Fedorowicz Z, Elmagarmid A. Rayyan-a web and mobile app for systematic reviews. Syst Rev. 2016;5(1):210. doi:10.1186/s13643-016-0384-4.
18. Maher CG, Sherrington C, Herbert RD, Moseley AM, Elkins M. Reliability of the PEDro scale for rating quality of randomized controlled trials. Phys Ther. 2003;83(8):713-21.
19. Gonzalez GZ, Moseley AM, Maher CG, Nascimento DP, Costa LDCM, Costa LO. Methodologic quality and statistical reporting of physical therapy randomized controlled trials relevant to musculoskeletal conditions. Arch Phys Med Rehabil. 2018;99(1):129-36. doi:10.1016/j. apmr.2017.08.485.
20. Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, Flottorp S, et al. Grading quality of evidence and strength of recommendations. BMJ. 2004;328(7454):1490. doi:10.1136/bmj.328.7454.1490.
21. GRADEpro GDT: GRADEpro Guideline Development Tool . McMaster University, 2020 (developed by Evidence Prime, Inc.). Available from gradepro.org.
22. Review Manager 5 (RevMan 5) . Version 5.3. Copenhagen: Nordic Cochrane Centre, Cochrane, 2014.
23. Murad MH, Wang Z, Chu H, Lin L. When continuous outcomes are measured using different scales: guide for meta-analysis and interpretation. BMJ. 2019;364:k4817. doi:10.1136/bmj.k4817.
24. Stock MC, Downs JB, Cooper RB, Lebenson IM, Cleveland J, Weaver DE, et al. Comparison of continuous positive airway pressure, incentive spirometry, and conservative therapy after cardiac operations. Crit Care Med. 1984;12(11):969-72. doi:10.1097/00003246-198411000-00010.
25. Higgins JPT, Eldridge S, Li T (editors). Chapter 23: Including variants on randomized trials. In: Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for Systematic Reviews of Interventions version 6.2 (updated February 2021).
26. Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for Systematic Reviews of Interventions. 2nd ed. Chichester UK: The Cochrane Collaboration and John Wiley & Sons Ltd., 2019. 703 p.
27. Iverson LI, Ecker RR, Fox HE, May IA. A comparative study of IPPB, the incentive spirometer, and blow bottles: the prevention of atelectasis following cardiac surgery. Ann Thorac Surg. 1978;25(3):197-200. doi:10.1016/s0003-4975(10)63521-7.
28. Gale GD, Sanders DE. Incentive spirometry: its value after cardiac surgery. Can Anaesth Soc J. 1980;27(5):475-80. doi:10.1007/ BF03007047.
29. Dull JL, Dull WL. Are maximal inspiratory breathing exercises or incentive spirometry better than early mobilization after cardiopulmonary bypass? Phys Ther. 1983;63(5):655-9. doi:10.1093/ptj/63.5.655.
30. Jenkins SC, Soutar SA, Loukota JM, Johnson LC, Moxham J. Physiotherapy after coronary artery surgery: are breathing exercises necessary? Thorax. 1989;44(8):634-9. doi:10.1136/thx.44.8.634.
31. Jenkins SC, Soutar SA, Loukota JM, Johnson LC, Moxham J. A comparison of breathing exercises, incentive spirometry and mobilisation after coronary artery surgery. Physiother Theory Pract. 1990;6:117-26. doi:10.3109/09593989009037789.
32. Oikkonen M, Karjalainen K, Kähärä V, Kuosa R, Schavikin L. Comparison of incentive spirometry and intermittent positive pressure breathing after coronary artery bypass graft. Chest. 1991;99(1):60-5. doi:10.1378/ chest.99.1.60.
33. Crowe JM, Bradley CA. The effectiveness of incentive spirometry with physical therapy for high-risk patients after coronary artery bypass surgery. Phys Ther. 1997;77(3):260-8. doi:10.1093/ptj/77.3.260.
34. Savcı S, Sakınç S, İnce DI, Arikan H, Can Z, Buran Y, et al. Active cycle of breathing techniques and incentive spirometer in coronary artery bypass graft surgery. Fizyoter Rehabil. 2006;17(2):61-9.
35. Romanini W, Muller AP, Carvalho KA, Olandoski M, Faria-Neto JR, Mendes FL, et al. The effects of intermittent positive pressure and incentive spirometry in the postoperative of myocardial revascularization. Arq Bras Cardiol. 2007;89(2):94-9, 105-10. doi:10.1590/s0066- 782x2007001400006.
36. Renault JA, Costa-Val R, Rosseti MB, Houri Neto M. Comparison between deep breathing exercises and incentive spirometry after CABG surgery. Rev Bras Cir Cardiovasc. 2009;24(2):165-72. doi:10.1590/ s0102-76382009000200012.
37. Dias CM, Vieira Rde O, Oliveira JF, Lopes AJ, Menezes SL, Guimarães FS. Three physiotherapy protocols: effects on pulmonary volumes after cardiac surgery. J Bras Pneumol. 2011;37(1):54-60. doi:10.1590/s1806-37132011000100009.
38. El-Kader SMA. Blood gases response to different breathing modalities in phase I of cardiac rehabilitation program after coronary artery bypass graft. Eur J Gen Med. 2011;8(2):85-91. doi.org/10.29333/ejgm/82706.
39. Almutairi F, Fallows S, Mason-Whitehead E. Continuous positive airway pressure (cpap) had better outcomes when compared with incentive spirometry (IS) to re-open collapse alveoli after cardiac surgery: randomized study. Am J Respir Crit Care Med. 2012;185:A4869. doi:10.1164/ajrccm-conference.2012.185.1_MeetingAbstracts.A4869.
40. Al-Mutairi FH, Fallows SJ, Abukhudair WA, Islam BB, Morris MM. Difference between continuous positive airway pressure via mask therapy and incentive spirometry to treat or prevent post-surgical atelectasis. Saudi Med J. 2012;33(11):1190-5.
41. Mueenudheen TP, Moiz JA, Gupta VP. A comparative study on the effects of incentive spirometry and deep breathing exercise on pulmonary functions after uncomplicated coronary artery bypass grafting surgery. Indian J Physiother Occup Ther. 2012;6(2):63-7.
42. Rizwan A, Jalwan J, Mukherjee S. To compare the immediate effect of deep breathing exercise and incentive spirometry on ABG after mitral valve replacement surgery. Indian J Physiother Occup Ther. 2012;6(4):12-7.
43. Zangerolamo TB, Barrientos TG, Baltieri L, Moreno M, Pazzianotto-Forti EM. Effects of flow-oriented incentive spirometry after myocardial revascularization. Rev Bras Cardiol. 2013;26(3):180-5.
44. Yazdannik A, Bollbanabad HM, Mirmohammadsadeghi M, Khalifezade A. The effect of incentive spirometry on arterial blood gases after coronary artery bypass surgery (CABG). Iran J Nurs Midwifery Res. 2016;21(1):89-92. doi:10.4103/1735-9066.174761.
45. Manapunsopee S, Thanakiatpinyo T, Wongkornrat W, Chuaychoo B, Thirapatarapong W. Effectiveness of incentive spirometry on inspiratory muscle strength after coronary artery bypass graft surgery. Heart Lung Circ. 2020;29(8):1180-6. doi:10.1016/j.hlc.2019.09.009.
46. Alam M, Hussain S, Shehzad MI, Mushtaq A, Rauf A, Ishaq S. Comparing the effect of incentive spirometry with acapella on blood gases in physiotherapy after coronary artery bypass graft. Cureus. 2020;12(2):e6851. doi:10.7759/cureus.6851.
47. Amin R, Alaparthi GK, Samuel SR, Bairapareddy KC, Raghavan H, Vaishali K. Effects of three pulmonary ventilation regimes in patients undergoing coronary artery bypass graft surgery: a randomized clinical trial. Sci Rep. 2021;11(1):6730. doi:10.1038/s41598-021-86281-4.
48. Zerang F, Amouzeshi A, Barkhordari-Sharifabad M. Comparison of the effect of incentive spirometry and deep breathing exercises on hemodynamic parameters of patients undergoing coronary artery bypass graft surgery: a clinical trial. J Vasc Nurs. 2022;40(3):134-9. doi:10.1016/j.jvn.2022.08.002.
49. Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and variance from the median, range, and the size of a sample. BMC Med Res Methodol. 2005;5:13. doi:10.1186/1471-2288-5-13.
50. Freitas ER, Soares BG, Cardoso JR, Atallah AN. Incentive spirometry for preventing pulmonary complications after coronary artery bypass graft. Cochrane Database Syst Rev. 2007;(3):CD004466. Update in: Cochrane Database Syst Rev. 2012;9:CD004466. doi:10.1002/14651858. CD004466.pub2.
51. Freitas ER, Soares BG, Cardoso JR, Atallah ÁN. Incentive spirometry for preventing pulmonary complications after coronary artery bypass graft. Cochrane Database Syst Rev. 2012;2012(9):CD004466. doi:10.1002/14651858.CD004466.pub3.
52. Eltorai AEM, Baird GL, Pangborn J, Eltorai AS, Antoci V Jr, Paquette K, et al. Financial impact of incentive spirometry. Inquiry. 2018;55:46958018794993. doi:10.1177/0046958018794993.
53. Kotani T, Akazawa T, Sakuma T, Nagaya S, Sonoda M, Tanaka Y, et al. Effects of incentive spirometry on respiratory motion in healthy subjects using cine breathing magnetic resonance imaging. Ann Rehabil Med. 2015;39(3):360-5. doi:10.5535/arm.2015.39.3.360.
54. Alaparthi GK, Augustine AJ, Anand R, Mahale A. Comparison of diaphragmatic breathing exercise, volume and flow incentive spirometry, on diaphragm excursion and pulmonary function in patients undergoing laparoscopic surgery: a randomized controlled trial. Minim Invasive Surg. 2016;2016:1967532. doi:10.1155/2016/1967532.
Authors’Roles & Responsibilities
HVCS = Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved; final approval of the version to be published
ACL = Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; drafting the work or revising it critically for important intellectual content; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved; final approval of the version to be published
ACPNP = Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; drafting the work or revising it critically for important intellectual content; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved; final approval of the version to be published
JRFFM = Drafting the work or revising it critically for important intellectual content; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved; final approval of the version to be published
ECS = Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; drafting the work or revising it critically for important intellectual content; agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved; final approval of the version to be published
Article receive on Wednesday, April 5, 2023
Article accepted on Tuesday, May 16, 2023
All scientific articles published at www.rbccv.org.br are licensed under a Creative Commons license
All rights reserved 2017 / © 2024 Brazilian Society of Cardiovascular Surgery
DEVELOPMENT BY 
English PDF
Print
Send this article by email
How to cite this article
Submit a comment
Mendeley
Pocket
Share on Facebook
Share on Twitter