Effects of preoperative physiotherapy on signs and symptoms of pulmonary collapse and infection after major abdominal surgery: secondary analysis of the LIPPSMAck-POP multicentre randomised controlled trial

Table 1 Treatment effects to postoperative clinical criteria and oxygen therapy requirements. Values are number (proportion)

	Control (n = 214)	Intervention (n = 218)	Adjusted RR (95% CI)	p value
Clinical criteria 14-day event rates
Auscultation abnormal	161 (75%)	151 (69%)	0.94 (0.84 to 1.06)	0.35
Sputum colour abnormal	90 (42%)	81 (37%)	0.94 (0.74 to 1.18)	0.57
Hypoxemia	74 (35%)	64 (29%)	0.93 (0.71 to 1.23)	0.63
Pyrexia	27 (13%)	28 (13%)	1.04 (0.63 to 1.73)	0.87
Collapse/consolidation on CXR/CT	57 (27%)	39 (18%)	0.74 (0.52 to 1.05)	0.09
Leukocytosis	134 (63%)	141 (65%)	1.05 (0.92 to 1.22)	0.43
Sputum culture positive for infection	15 (7.0%)	9 (4.1%)	0.67 (0.30 to 1.48)	0.33
Physician diagnosis of PPC in medical record	37 (17%)	28 (13%)	0.86 (0.51 to 1.35)	0.51
Respiratory antibiotics prescribed	53 (25%)	29 (13%)	0.60 (0.40 to 0.91)	0.02
Oxygen therapy and mechanical ventilation
No oxygen therapy	129 (60%)	172 (79%)	0.49 (0.31 to 0.78)	0.002
Standard oxygen therapy	48 (22%)	24 (11%)
High flow oxygen therapy	25 (12%)	10 (4.6%)
Invasive or non-invasive mechanical ventilation	12 (5.6%)	12 (5.5%)

Abbreviations: RR relative risk, CI confidence interval, PPC postoperative pulmonary complications, CXR chest X-ray, CT computerised tomography
Data are event rates within the first 14 postoperative hospital days compared using mixed effects general linear Poisson regression reported as relative risk with 95% CI, adjusted for age, respiratory comorbidity, and surgical category, with exposure time as the time to cessation of observations
Oxygen therapy was classified as the most intensive therapy provided to a patient, a rank-order scale, with comparison estimated as odds ratio using ordered linear regression adjusted as above

ISSN: 2047-0525