This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) statement. ²⁴ The review protocol was registered prospectively on PROSPERO (registration ID: 529588). Electronic searches were conducted in AMED, MEDLINE, and CINAHL via EBSCOhost, as well as Web of Science and Scopus. The search period covered January 2009 to December 2025. This broader time frame was selected following preliminary scoping, which indicated that restricting the search to the most recent 10 years yielded too few eligible studies.

Search results were exported to Microsoft Excel for deduplication and screening. Titles and abstracts were screened first, followed by full-text assessment of potentially eligible studies using a predefined screening form. Reference lists of included studies and relevant systematic reviews were also hand-searched to identify any additional eligible records. Two reviewers conducted the search, deduplication, screening, and data extraction processes. Any uncertainties regarding eligibility were discussed with the other authors. The full search strategy is presented in Appendix A, and the full-text screening template is presented in Appendix B. The study selection process is shown in the PRISMA flow diagram in Figure 1.

RCTs that compared prehabilitation, postoperative physiotherapy and perioperative interventions to standard care or no treatment were included. Only RCTs on adults (18 years or older) undergoing laparoscopic surgery were included in the review. The primary outcomes of the review were the rate of PPCs and parameters assessing pulmonary function, while the secondary outcomes were arterial blood gases (ABGs), LOS, six-minute walk distance (6MWD) and hospitalisation costs.

Studies were eligible for inclusion if they met the following criteria: 1.

Participants were adults aged 18 years or older undergoing laparoscopic abdominal surgery.

Studies were excluded if participants underwent thoracic or cardiac surgery, if the surgical approach was non-laparoscopic, or if the intervention was not principally physiotherapy based. Multimodal programmes were excluded where the specific contribution of physiotherapy could not be distinguished from other components. Reviews, case reports, and studies without accessible full text were also excluded.

Data extraction was performed using a structured electronic table (See Appendix). The extracted data included author, year of publication, study design, sample size, participant characteristics, surgery type, intervention and comparator details, intervention duration, follow-up time points, and reported outcomes. Primary outcomes were postoperative pulmonary complications and pulmonary function measures. Secondary outcomes were arterial blood gases, length of stay, six-minute walk distance, and hospitalisation costs. Where available, group means, standard deviations, and between-group comparisons were extracted. Data extraction was undertaken by one reviewer, with uncertainties discussed with the other authors.

A narrative synthesis was undertaken because of substantial heterogeneity across studies in surgical populations, intervention type, intervention timing and duration, comparator conditions, and outcome definitions. Summary tables were used to present study characteristics and reported outcomes (see appendix). Findings were synthesised according to intervention timing, namely preoperative, perioperative, and postoperative physiotherapy interventions, and were interpreted with consideration of methodological quality and risk of bias.

Methodological quality was assessed using the Physiotherapy Evidence Database (PEDro) scale, which is based on the Delphi list. ²⁵ Each trial was scored from 0 to 10, with higher scores indicating better methodological quality. In line with previous interpretations, scores below 4 were considered poor, scores of 4 to 5 fair, scores of 6 to 8 good, and scores of 9 to 10 excellent. ²⁶ The PEDro assessment was used to support interpretation of study quality rather than as a sole determinant of study value. A summary of the PEDro assessment is presented in Figure 2.

Risk of bias was assessed using the Cochrane Risk of Bias tool for randomised controlled trials. ²⁷ The following domains were evaluated: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective outcome reporting, and other potential sources of bias. Each domain was rated as low, high, or unclear risk of bias. Overall judgements were then made for each study based on the ratings across domains. The traffic light plot is presented in Figure 3, and the summary plot is presented in Figure 4.

A total of 3811 records were identified through database searching, with an additional two records identified through citation searching. After removal of 14 duplicates, 3795 titles and abstracts were screened. Seventy-three full-text articles were assessed for eligibility, of which nine randomised controlled trials met the inclusion criteria and were included in the review. Although the search period extended to December 2025, the included studies were published between 2010 and 2022. The study selection process is presented in Figure 1. The characteristics of the included studies are summarised in Table 1.

A total of 1321 patients undergoing LS participated in the 9 studies included in the review. The populations of the studies included 552 males and 719 females. One study did not report the sex of participants. ²⁸ The mean age of participants ranged from 38–69.7 years. 82% of participants in one study had a body mass index (BMI) of between 30–40 kg/m2 whilst 18% of participants had a BMI of >40 kg/m2. ²⁹ The mean BMI ranged from 21.45–48.6 kg/m2 in 7 studies and one study did not report on BMI. ²⁸ The surgical procedures included were as follows Colorectal Surgery (n = 506), Bariatric surgery (n = 467), Cholecystectomy (n = 224), Hernia Surgery (n = 68), Appendectomy: (n = 43), Laparoscopic diagnostic (n = 8) and Splenectomy (n = 5). The baseline characteristics of patients were comparable and there were no statistically significant differences between groups apart from a study in which the BMI of the control group was significantly higher than the intervention group. ³⁰ Baseline participant and surgical characteristics are presented in Table 2.

The studies included examined the effects of either preoperative, ^{28– 31} postoperative ^{32– 34} or perioperative ^{35, 36} physiotherapy interventions. Interventions used within these studies consisted of preoperative exercise interventions, incentive spirometry, early mobilisation, postural drainage, inspiratory muscle training, breathing and coughing exercises (see Table 2). The duration of preoperative interventions ranged from 5–30 days before surgery whilst the postoperative interventions ranged from 3–4 days, however, it was unclear how long the intervention was performed in one trial. ³⁴ The duration of studies with perioperative interventions ranged from 9–90 days. The controls in only one trial received no treatment. ³² In three trials the control groups received standard care. ^{30, 35, 36} The controls in the remaining trials received IS in addition to standard care, ³⁴ mobilisation only, ³³ postoperative physiotherapy only ²⁹ and postoperative incentive spirometry only. ^{28, 31}

Primary outcome findings are summarised in Table 3.

Postoperative pulmonary complications

Seven out of nine studies ^{28– 31, 34– 36} reported on the incidence of PPCs following the intervention period. A definition of PPCs was provided in most trials however there was a lack of definition in two trials. ^{28, 30} The mean PPC rate across all studies was 6.91% for the intervention groups and 15.79% for the control groups. One trial examining a preoperative exercise intervention reported a PPC rate of 26.92% in the intervention group and 62.5% in the control group, and this difference was statistically significant (p = 0.034). ²⁹ Three trials examining preoperative IS and IMT reported no PPCs across either group. ^{28, 30, 31} One trial examining a perioperative breathing intervention reported a rate of 12.88% in the IG and 32.58% in the CG which was statistically significant. ³⁵ Another trial which examined a perioperative breathing intervention reported a PPC rate of 4.17% in the IG and 11.67% in the CG which was also statistically significant. ³⁶ One trial examining a postoperative ambulation intervention to ambulation in addition to IS reported a PPC rate of 3.6% in the IG and 7.1% in the CG which was not statistically significant. ³⁴

Pulmonary function parameters

Six out of nine trials reported on pulmonary function parameters. ^{28– 33} One trial which examined a preoperative exercise therapy intervention reported highly significant differences in Slow Vital Capacity (SVC), Inspiratory Capacity (IC), Maximum Inspiratory Pressure (MIP), Maximum Expiratory Pressures (MEP) between groups preoperatively and postoperatively on days two and five. ²⁹ One trial examining preoperative IS found no statistically significant differences in IC between groups preoperatively or day one postoperatively. ³¹ Conversely, another trial examining preoperative IS reported significantly different values in Forced Expiratory Volume in one second (FEV1) and FVC for the IG compared to the CG preoperatively and at all time intervals postoperatively. There was a similar reduction in Peak Expiratory Flow Rate (PEFR) in both groups except at 48 hrs postoperatively when the reduction was significantly higher in the CG. ²⁸ One trial examining preoperative IMT and IS reported higher values of FEV1, and FVC at 1 day preoperatively and 12 hrs postoperatively but there was no statistically significant difference between groups. There was a statistically significant difference between groups at 1 day preoperatively in MIP and MEP in favour of the IG. MIP and MEP remained higher 12 hours postoperatively in the IG however there was no statistical significance between groups. ³⁰ Another trial compared one CG that received no treatment and three IGs which performed postoperative diaphragmatic breathing exercises (DBE) Flow incentive spirometry (FIS) and Volume incentive spirometry (VIS) interventions. A statistically significant difference was observed one day pre-operation and postoperative day two in FVC between the groups receiving DBE, VIS and the CG with the FVC being significantly higher in the IGs. There were no statistically significant differences in FVC between IGs or between the group receiving FIS and the CG on the one-day pre-operation and postoperative 2nd day. There was no significant difference reported in FEV1 or PEFR between any of the IGs or between the IGs and the CG or at preoperative and postoperative 2nd day. ³¹ A trial examining a postoperative mobilisation and chest physiotherapy intervention reported that values for vital capacity (VC), expiratory reserve volume (ERV), inspiratory reserve volume (IRV), tidal volume (TV), FEV1, FEV1/FVC, PEFR, forced expiratory flow at 25–75% (FEF25–75) all increased in the IG however only VC, TV, FEV1/FVC, and PEFR resulted in a significant improvement compared to the CG. ³³

Secondary outcome findings are summarised in Table 4.

Arterial blood gases

Three trials reported on various postoperative ABG values. Duymaz et al. (2020) reported a significant increase in pressure of oxygen (PO2), pressure of carbon dioxide (PCO2) and acid-base balance (pH) scores on the day of discharge in the IG receiving postoperative chest physiotherapy and mobilisation compared to the CG receiving mobilisation only. ³³ A study reported significantly higher mean oxygenation (PaO2/FiO2 ratio) values at 1 and 12 hrs postoperatively in the IG receiving preoperative IS and IMT than in the CG receiving preoperative usual care. ³⁰ Similarly, ³⁶ reported significantly higher mean arterial partial pressure of oxygen (PaO2) values at postoperative days 1 and 4 in the IG receiving perioperative breathing exercises compared to the CG receiving standard care.

6MWD

Three trials reported on 6MWD. ²⁹ reported a significant increase in 6MWD preoperatively, day 5 and 1 month postoperatively in the IG receiving preoperative exercise therapy compared to the CG receiving postoperative therapy only. ³⁵ reported a significant improvement in 6MWD 1 day preoperatively, 3 days, 30 days, and 90 days after surgery in the IG receiving perioperative breathing exercises compared to the CG receiving standard care. The CG also saw a significant decrease in their mean 6MWD compared to their baseline value. ³³ reported a significant increase in 6MWD on the day of discharge in the IG receiving postoperative chest physiotherapy and mobilisation compared to the CG receiving mobilisation only.

Hospitalisation costs

Two trials reported on Hospitalisation Costs. Both ³⁵ and ³⁶ reported significantly lower mean hospitalisation costs in the IG compared to CG in their respective studies examining perioperative breathing interventions.

Length of stay (LOS)

Two trials examining perioperative breathing interventions reported on LOS. ³⁵ reported there was a significantly lower median LOS in the IG compared to the CG. Although ³⁶ also reported a lower median LOS duration there was no significant difference.

PEDro Assessment of methodological quality

Figure 2. shows a summary of the PEDro assessment. The total PEDro scores ranged from 5–8/10 achieving a rating of “fair” to “good” according to the scale. All trials fulfilled criteria for 1 (Eligibility Criteria), 2 (Random Allocation), 10 (Between-Group Analysis) and 11 (Point Estimates and variability). All trials were unable to blind participants due to the nature of the interventions used. There was also a lack of reporting regarding blinding of the therapists providing the interventions therefore both criteria 5 (Blind subjects) and 6 (Blind therapists) were not satisfied. Two trials ^{29, 31} did not describe any methods to conceal allocation and did not satisfy criteria 3 (Concealed Allocation). In one trial most baseline data was comparable between groups however criteria 4 (Baseline Comparability) was not fulfilled as baseline BMI was higher in the CG compared to the IG. ³⁰ Criteria 4 was satisfactory in all other trials. Five trials were unable to satisfy criteria 7 (Blind Assessors) as there was a lack of description of methods used to blind the outcome assessor(s). ^{28– 32} Two trials were unable to satisfy criteria 8 (Adequate Follow-up) due to a lack of reporting on the number of participants from whom the key outcome measures were obtained. ^{28, 33} Four trials ^{28, 30, 33, 36} were unable to satisfy criteria 9 as it was specifically stated that dropouts were not included in the analysis or information regarding whether all participants were included in the final analysis was not reported.

Figure 3. and 4. Illustrate a summary of the ROB assessment. Random sequence generation was performed in eight studies using a computerised random number generator however the process was unclear in the other two of the studies. Allocation concealment was adequately described in seven studies with these studies opting to use numbered, opaque, sealed envelopes before group assignment. All studies were rated as high risk of bias in the blinding of participants and personnel domain as interventions would have involved either exercise or incentive spirometry making blinding of the participants impossible. Outcome assessors were blinded to allocated interventions leading to an overall low risk of detection bias. Two studies showed a low risk of selection bias however this was unclear in the remaining studies as there was insufficient information to permit other judgements. All studies were at low risk of attrition bias as most trials had no missing outcome data and any missing data balanced in numbers across groups with similar reasons for missing data. Overall studies were assessed to be at high risk of bias one or more key domains were rated as high risk of bias.

This review has several limitations. First, the included studies were heterogeneous in terms of surgical populations, intervention type, intervention timing and duration, comparator conditions, and outcome measures. This limited direct comparison between studies and precluded meta-analysis. Second, the methodological quality of the included trials ranged from fair to good, and the overall risk of bias was high in several studies, particularly because blinding of participants and treating personnel was not feasible for many physiotherapy interventions. Third, some included trials were relatively old and may not fully reflect current perioperative practice. Finally, the review processes of screening, data extraction, PEDro scoring, ^{46, 47} and risk of bias assessment were conducted primarily by two reviewers, which may have increased the risk of reviewer bias despite consultation with the project supervisor when uncertainties arose. These limitations mean that the findings should be interpreted as suggestive rather than definitive.

The findings of this review suggest that physiotherapy may have a meaningful role in supporting recovery after laparoscopic abdominal surgery, particularly when respiratory interventions are delivered across the perioperative period. Patients undergoing laparoscopic surgery often present with important clinical risk factors, including obesity, multiple comorbidities, reduced physical capacity, and varying levels of postoperative symptom burden. ^{11, 15} These factors should be considered when selecting and individualising physiotherapy interventions.

From a clinical perspective, perioperative breathing exercise programmes appear promising for reducing PPCs and improving recovery outcomes in selected laparoscopic populations. Prehabilitation may also be beneficial, although the available evidence remains limited. By contrast, the evidence supporting routine postoperative incentive spirometry alone remains inconsistent. Clinicians should therefore consider the overall clinical context, patient risk profile, and available resources when deciding which interventions to prioritise. Future research should aim to identify which subgroups are most likely to benefit and should use standardised intervention reporting and outcome measures to improve translation into practice.