. 2017 Aug 11;7(8):e015712. doi: 10.1136/bmjopen-2016-015712

Table 1.

Characteristics of included studies

First author, year	Study population	Country	Participants at follow-up	Physical activity assessment	Longest follow-up	Main findings
Giaccardi, 2011¹⁸	All patients ≥65 years undergoing CABG and/or valve procedures (total sample: 74.1±5.8 years old); 43% female	Italy	158	Harvard Alumni Questionnaire	4 weeks postoperatively	Physical activity had an independent association with postoperative atrial fibrillation within 30 days.
Markou, 2007¹²	Elective CABG patients (active: 64.4±9.4, inactive: 63.8±9.0 years old); % female not reported	The Netherlands	428	The Corpus Christi Heart Project	1 year	Inactive versus active group had significantly more perioperative MIs, but not reoperations, ICU LOS, HLOS or postoperative complications at 1 year. Inactive group was more likely than active group to be physically active at 1 year.
Nery, 2007¹³	All patients undergoing CABG (active: 63±11, inactive 66±14 years old); 42% female	Brazil	55	Structured Questionnaire confirmed by Minnesota Leisure Time Physical Activity Questionnaire	1 year	Inactive versus active groups had significantly longer HLOS and more postoperative events at 1 year.
Markou, 2008¹⁴	Elective CABG patients (64.3±9.04 years old); 18% female	The Netherlands	568	The Corpus Christi Heart Project	1 year	Inactive versus active groups were more likely to be more physically active 1 year postoperatively.
Martini, 2010¹⁵	Elective CABG patients (active: 60±10, inactive: 62±10 years old); 34% female	Brazil	185	Baecke Usual Physical Activity Questionnaire	Two years	Inactive versus active group did not have significantly different MACCE outcomes at 2 years.
Nery, 2010¹⁶	Elective CABG patients (active: 60±10, inactive: 62±10 years old); 34% female	Brazil	202	Baecke Usual Physical Activity Questionnaire	Hospital discharge	Inactive versus active groups had more postoperative events within 30 days and a longer HLOS.
Rengo, 2010¹⁷	Acute or elective CABG patients ≥70 years (active: 72.3±3.2, inactive: 76.1±3.9 years old); 34% female	Italy	587	Physical Activity Scale for the Elderly	Mean 44.3±21.0 months	Physical activity had an independent and dose association with cardiac and all-cause mortality 5 years postoperatively.
Cacciatore, 2012¹⁹	All patients ≥65 years undergoing CABG and/or valve procedures (72.9±4.8 years old); 48% female	Italy	250	Physical Activity Scale for the Elderly	Hospital discharge	Physical activity was independently associated with reduced prolonged ICU LOS. Physical activity was not independently associated with postoperative ADLs.
Noyez, 2013²⁰	Elective CABG and/or valve patients (69.7±10.1 years old);	The Netherlands	3150	The Corpus Christi Heart Project	30 days postoperatively	Physical activity was not independently associated with hospital or 30 day mortality. Inactive vs. Active group had a significantly longer ICU LOS.
Min, 2015²¹	Elective CABG and/or valve patients ≥65 years (74.7±5.9 years old)	USA	62	The Health and Retirement Survey	4–6 months	Inactive versus active groups had significantly higher postoperative physical activity up to 6 months postoperatively.
van Laar²²	Patients ≥75 years undergoing elective isolated aortic valve replacement (79.5±2.8 years old); 59% female	The Netherlands	115	The Corpus Christi Heart Project	2 years postoperatively	Inactive versus active groups had significantly higher mortality rates 2 years postoperatively.

ADL, activities of daily living; CABG, coronary artery bypass graft surgery; HLOS, hospital length of stay; ICU LOS, intensive care unit length of stay; MACCEs, major adverse cerebrovascular and cardiac events; MI, myocardial infarction.