Ref (type)	Population	Outcome, Interventions	Results and statistical analysis	Effect size	Favours
Pneumonia
RCT	102 people undergoing cholecystectomy In review	Pneumonia 7/51 (14%) with deep-breathing exercises plus directed cough plus postural drainage therapy 19/51 (37%) with control	RD –23.6% 95% CI –40% to –7%	Effect size not calculated	deep-breathing exercises plus directed cough plus postural drainage therapy
RCT	81 people (35 men), mean age 64.1 years (range 18–84 years), undergoing elective upper abdominal surgery In review	Pneumonia 0/40 (0%) with deep-breathing exercises plus directed cough 1/41 (2%) with no prophylaxis	RD –2.4% 95% CI –9% to +4%		Not significant
RCT 3-armed trial	137 people undergoing cholecystectomy In review	Pneumonia 4/45 (9%) with deep-breathing exercises plus directed cough plus postural drainage therapy 1/45 (2%) with control	RD +6.7% for lung expansion v control 95% CI –3% to +16%		Not significant
RCT 3-armed trial	137 people undergoing cholecystectomy In review	Pneumonia 4/47 (9%) with deep-breathing exercises plus directed cough plus postural drainage therapy plus bronchodilator aerosol 1/45 (2%) with control	RD +6.3% for lung expansion v control 95% CI –3% to +15%		Not significant
RCT	50 people undergoing upper or lower abdominal surgery In review	Pneumonia 1/29 (3%) with deep-breathing exercises plus directed cough 0/21 (0%) with control	RD +3.4% 95% CI –7% to +13%		Not significant
RCT	40 people undergoing cholecystectomy In review	Pneumonia 0/20 (0%) with incentive spirometry 1/20 (5%) with control	RD –5.0% 95% CI –18% to +8%		Not significant
RCT	204 people (166 men), mean age 63 years (SD 11.8), undergoing elective intra-abdominal vascular surgery In review	Pneumonia (per CDC criteria) 2/99 (2%) with nasal continuous positive-airway pressure for 12 hours after surgery 5/105 (5%) with control (O2 by nasal cannula to keep saturation above 95%)	RD –2.8% 95% CI –8% to +2%		Not significant
RCT	368 people (158 men), mean age 53.3 years (range 19–92 years), undergoing elective open abdominal surgery In review	Pneumonia 0.6% with deep-breathing exercises with or without positive respiratory pressure throughout expiration 7% with no prophylaxis Absolute numbers not reported	P <0.05	Effect size not calculated	deep-breathing exercises with or without positive respiratory pressure throughout expiration
Postoperative pulmonary complications (unspecified)
RCT	81 people (35 men), mean age 64.1 years (range 18–84 years), undergoing elective upper abdominal surgery In review	Postoperative pulmonary complications (unspecified) 3/40 (8%) with deep-breathing exercises plus directed cough 8/41 (20%) with control	RD –12% 95% CI –27% to +3%		Not significant
RCT 4-armed trial	172 people undergoing upper or lower abdominal surgery In review	Postoperative pulmonary complications (unspecified) 9/42 (21%) with incentive spirometry 21/44 (48%) with control	RD –26.3% for incentive spirometry v control 95% CI –46% to –7%	Effect size not calculated	incentive spirometry
RCT 4-armed trial	172 people undergoing upper or lower abdominal surgery In review	Postoperative pulmonary complications (unspecified) 9/41 (22%) with deep-breathing exercises plus directed cough 21/44 (48%) with control	RD –25.8% for deep breathing exercise plus directed cough v control 95% CI –45% to –6%	Effect size not calculated	deep-breathing exercises plus directed cough
RCT 4-armed trial	172 people undergoing upper or lower abdominal surgery In review	Postoperative pulmonary complications (unspecified) 10/45 (22%) with intermittent positive-pressure breathing (IPPB) 21/44 (48%) with control	RD –25.5% for intermittent positive-pressure breathing v control 95% CI –45% to –6%	Effect size not calculated	intermittent positive-pressure breathing
RCT	50 people undergoing upper or lower abdominal surgery In review	Postoperative pulmonary complications (unspecified) 5/29 (17%) with deep-breathing exercises plus directed cough 3/21 (14%) with control	RD +3.1% 95% CI –17% to +23%		Not significant
RCT	40 people undergoing upper abdominal surgery In review	Postoperative pulmonary complications (unspecified) 8/20 (40%) with incentive spirometry 6/20 (30%) with control	RD +10.0% 95% CI –19% to +39%		Not significant
RCT 3-armed trial	200 people undergoing upper abdominal surgery In review	Postoperative pulmonary complications (unspecified) 29/50 (58%) with intermittent positive-pressure breathing (IPPB) 48/100 (48%) with control	RD +10.0% for intermittent positive-pressure breathing v control 95% CI –7% to +27%		Not significant
RCT 3-armed trial	200 people undergoing upper abdominal surgery In review	Postoperative pulmonary complications (unspecified) 28/50 (56%) with intermittent positive-pressure breathing (IPPB) with more intensive therapy 48/100 (48%) with control	RD +8.0% for IPPB with more intensive therapy v control 95% CI –9% to +25%		Not significant
RCT	64 people undergoing upper or lower abdominal surgery In review	Postoperative pulmonary complications (unspecified) 1/32 (3%) with continuous positive-airway pressure 0/32 (0%) with control	RD +3.1% 95% CI –5% to +11%		Not significant
RCT	230 people having open heart surgery	Postoperative pulmonary complications (unspecified) 5/115 (4%) with physiotherapy plus deep-breathing exercises 3/115 (3%) with physiotherapy without deep-breathing exercises	P = 0.72		Not significant
RCT	876 people (430 men) median age 55 years (IQR 32–72), undergoing abdominal surgery In review	Postoperative pulmonary complications (unspecified) 16% with incentive spirometry 15% with control (chest physiotherapy only)	Significance not assessed
RCT	155 low-risk people, median age 36 years (IQR 29–44), undergoing abdominal surgery In review Subgroup analysis	Postoperative pulmonary complications (unspecified) 6/79 (8%) with incentive spirometry 8/76 (11%) with deep-breathing exercises	P = 0.50		Not significant
RCT	301 high-risk people, median age 68 years (IQR 58–76), undergoing abdominal surgery In review Subgroup analysis	Postoperative pulmonary complications (unspecified) 29/152 (19%) with incentive spirometry 20/149 (13%) with incentive spirometry plus conventional chest physiotherapy	P = 0.18		Not significant
RCT	82 people undergoing inguinal hernia surgery In review	Postoperative pulmonary complications (unspecified) 8/40 (20%) with deep-breathing exercises 5/42 (12%) with control	RD +8.1% 95% CI –8% to +24%		Not significant
RCT	54 women undergoing hysterectomy In review	Postoperative pulmonary complications (unspecified) 1/27 (4%) with deep-breathing exercises plus directed cough 0/27 (0%) with control	RD +3.7% 95% CI –6% to +13%		Not significant
Atelectasis or infiltrate on chest radiograph
Systematic review	People undergoing any upper abdominal surgery; number of people in analysis not reported In review 14 RCTs in this analysis	Atelectasis or infiltrate on chest radiograph with incentive spirometry with no treatment Absolute results not reported	OR 0.44 95% CI 0.18 to 0.99	Moderate effect size	prophylactic lung expansion
Systematic review	People undergoing any upper abdominal surgery; number of people in analysis not reported In review 4 RCTs in this analysis	Atelectasis or infiltrate on chest radiograph with incentive spirometry with deep-breathing exercise Absolute results not reported	OR 0.91 95% CI 0.57 to 1.4		Not significant
Systematic review	People undergoing any upper abdominal surgery; number of people in analysis not reported In review 3 RCTs in this analysis	Atelectasis or infiltrate on chest radiograph with incentive spirometry with intermittent positive-pressure breathing Absolute results not reported	OR 0.73 95% CI 0.39 to 1.36		Not significant
Systematic review	People undergoing any upper abdominal surgery; number of people in analysis not reported In review 2 RCTs in this analysis	Atelectasis or infiltrate on chest radiograph with intermittent positive-pressure breathing (IPPB) with deep-breathing exercise Absolute results not reported	OR 0.94 95% CI 0.28 to 3.17		Not significant