Clin Physiol Funct Imaging (2016) 36, pp274–280
doi: 10.1111/cpf.12224
Cardiac function assessed by exercise echocardiography on the first morning after coronary artery bypass grafting Hans Henrik Dedichen1,2, Idar Kirkeby-Garstad1,3, Petter Aadahl1,3, Jonny Hisdal4 and Brage H. Amundsen5,6 1
Department of Circulation and Medical Imaging, Circulation Research University of Trondheim, Norwegian University of Science and Technology,Trondheim, Norway, 2Department of Cardiothoracic Surgery, St. Olav’s Hospital, Trondheim University Hospital, 3Department of Anesthesiology and Intensive Care Medicine, St. Olav’s Hospital, Trondheim University Hospital, Trondheim, Norway, 4Section of Vascular Investigations, Oslo University Hospital Aker, Oslo, Norway, 5 Department of Cardiology, St. Olav’s Hospital, Trondheim University Hospital, and 6MI lab, Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, Trondheim, Norway
Summary Correspondence Hans Henrik Dedichen, Department of Circulation and Medical Imaging, NTNU, Box 8905, 7491 Trondheim, Norway E-mail: [email protected]
Accepted for publication Received 14 May 2014; accepted 24 November 2014
Key words cardiac surgery; diastolic function; exercise tolerance; systolic function; tissue Doppler imaging
Cardiac surgery patients are urged to resume light physical activity on the first postoperative day, even if cardiac function may not have recovered fully after the operation. To elucidate the postoperative recovery process, we examined cardiac surgery patients with exercise echocardiography before and on the first day after the operation. Patients undergoing on-pump coronary artery bypass grafting were examined with echocardiography during semirecumbent cycle exercise. Patients exercised for five minutes at 10 W intensity and five minutes at 30 W intensity in bed with the upper body supported to approximately 30°. Fourteen patients were studied. Mitral annulus excursion and pulsed wave Doppler from the left ventricular outflow tract indicated postoperatively reduced cardiac stroke volume. Early diastolic tissue velocities of the mitral annulus were reduced, and early trans-mitral flow velocity was increased. The ratio between early mitral flow velocity and early diastolic mitral tissue velocity was increased postoperatively, indicating impaired left ventricular relaxation and increased left atrial pressure. Postoperative systolic mitral annulus tissue velocities were similar to preoperative velocities, indicating maintained systolic function. Postoperative exercise was associated with improvements in myocardial function indices and cardiac stroke volume similar to preoperative improvements. There were no signs of further deterioration in myocardial function during 30 W exercise. In summary, reduced left ventricular diastolic function after surgery resulted in reduced cardiac stroke volume, increased left atrial pressure and a higher rate of perceived exertion on the first postoperative day.
Introduction Cardiac surgery patients are urged to start physical activity at a time when myocardial function may still be transiently impaired after the operation. In our institution and in many other cardiac surgery units, most patients are mobilized on the first postoperative day (Westerdahl & Moller, 2010). Myocardial dysfunction may reduce exercise tolerance and limit early postoperative physical activity. Whereas postoperative myocardial function at rest has been extensively investigated (Ng et al., 2005; Apostolakis et al., 2009; Leppikangas et al., 2011; Juhl-Olsen et al., 2012; Hashemi et al., 2013), less is known about myocardial function during physical activity in the early postoperative phase. In a previous study, we used right heart catheterization to examine patients during semirecumbent cycling on the first 274
day after cardiac surgery. Mixed venous oxygen saturation was reduced, and blood lactate concentrations were increased after surgery, indicating reduced oxygen delivery relative to oxygen demand. Thermodilution measurements indicated that cardiac stroke volume was maintained at postoperative rest, but increased less during postoperative than preoperative exercise (Kirkeby-Garstad et al., 2006). However, the thermodilution technique may be unreliable in spontaneously breathing patients (Sasse et al., 1994). Transthoracic echocardiography provides information on cardiac function that is complementary to the hemodynamic data from invasive right heart catheterization. Echocardiography with tissue Doppler imaging and Doppler flow measurements is feasible and reproducible during cycling (Bougault et al., 2008) and can provide detailed quantitative information on biventricular systolic and diastolic function.
© 2014 Scandinavian Society of Clinical Physiology and Nuclear Medicine. Published by John Wiley & Sons Ltd 36, 4, 274–280
Cardiac function during exercise early after cardiac surgery, H. H. Dedichen et al. 275
Thus, the aim of this study was to use exercise echocardiography to explore if our previous findings regarding postoperative cardiac stroke volume were supported and if they may be explained by postoperative myocardial dysfunction. We hypothesized that the echocardiographic indices of cardiac function and stroke volume would be reduced on the first day after coronary artery bypass grafting (CABG) and that postoperative exercise would cause smaller increases in echocardiographic indices of myocardial function than preoperative exercise.
Methods We included 18 patients admitted for elective on-pump CABG at St. Olav’s University Hospital, Trondheim, in the period August 2010 to March 2012. Inclusion criteria were patients admitted for CABG without concomitant procedures. Exclusion criteria were unstable coronary artery disease, severe impairment of left ventricular function (EF 20 mmHg, aortic insufficiency ≥ grade 2, mitral valve insufficiency ≥ grade 2), clinical diagnosis of heart failure, chronic obstructive pulmonary disease more severe than GOLD 2 (Global initiative for chronic Obstructive Lung Disease), renal disease (serum creatinine >140 lmol l 1) and impaired mobility. The patients received standard treatment and care during the hospitalization. General anaesthesia for CABG was induced with diazepam, thiopental, fentanyl and pancuronium. Anaesthesia was maintained with isoflurane and fentanyl. During extra-corporal circulation (ECC), patients were sedated with propofol and cooled to approximately 35°C. The heart was arrested by cross-clamping of the aorta and infusion of cold St. Thomas’ cardioplegia. Standard revascularization was made by anastomosing the left internal thoracic artery to the left anterior descending artery and placing vein grafts from the ascending aorta to the circumflex artery and the right coronary artery. The study was performed in accordance with the ethical standards of the Helsinki declaration and approved by the regional ethics committee for western Norway (REK 2010/ 1543-6-1). Exercise protocol The patients were tested with semirecumbent two-leg bicycle exercise, and the same test protocol was used in pre- and postoperative tests. The preoperative test was made on the day of the operation with the patients fasted and prepared for surgery. The postoperative test was made on the first morning after surgery after removal of drains. The postoperative test was made on the first morning after surgery when the drains had been removed from the mediastinum. The patients were given three litres of oxygen per minute in a nasal catheter during the postoperative tests. Patients were exercising in bed with the upper body supported to approximately 30°. The
bicycle ergometer (Angio, Lode B.V., Groningen, the Netherlands) was secured to the bed in a position that was reasonably comfortable. The ergometer automatically adjusts braking force to pedalling frequency to keep the exercise intensity at the set level. The patients cycled for five minutes at 10 W intensity before the intensity was increased to 30 W, and the cycling continued without interruption for five minutes. Pedalling frequency was 50–60 cycles min 1. Measurements Heart rate (HR) was measured with five lead ECG connected to the patient monitor (Philips IntelliVue MP 70/MP 90, Andover, MA, USA). Arterial pressure was measured through a catheter inserted in the left radial artery before the preoperative test. Central venous pressure (CVP) was measured with a catheter in the superior caval vein inserted after the preoperative test. Both catheters were connected to pressure transducers (Edwards Lifesciences Pressure monitoring Transducers, Irvine, CA, USA) and calibrated with the zero-point five centimetre below the sternoclavicular joint. Readings were made before exercise and during the last minute of cycling on each workload. Blood samples were drawn from the radial artery catheter before exercise and during the last minute of exercise on each resistance level. The samples were stored on ice during the procedure and analysed immediately after the test for arterial oxygen saturation and arterial lactate concentration with ABL 800 blood gas analyzer (Radiometer, Brønshøj, Denmark). The patients indicated their feeling of exertion during cycling on 10 W and 30 W, using the Borg 6-20 scale (Borg, 1998). Echocardiography Echocardiography was performed with a Vivid 7 scanner with a 25 MHz phased array M3S transducer (GE Vingmed, Horten, Norway). All examinations were made by the same operator (BHA). B-mode 2D and colour tissue Doppler images in the two- and four-chamber views were recorded. Mitral annular tissue velocities were measured in the septal, lateral, inferior and anterior points and averaged. Mitral annular excursion (MAE) was also averaged over the same four points. Right ventricular tissue velocities and tricuspid annulus plane systolic excursion (TAPSE) were obtained at the lateral tricuspid annulus in a four-chamber view. Tissue velocities were measured using colour tissue Doppler with a frame rate of 100–160 frames s 1 both at rest and during exercise, as pulsed wave tissue Doppler was considered less reliable during exercise due to respiratory motion of the heart. Early (e’) and late (a’) diastolic as well as systolic (s’) tissue velocities of the tricuspid and mitral annuli were analysed. Blood flow velocities were measured by pulsed wave Doppler across the mitral and tricuspid valves, and in the left ventricular outflow tract, preferably in end-expiration. Doppler recordings were
© 2014 Scandinavian Society of Clinical Physiology and Nuclear Medicine. Published by John Wiley & Sons Ltd 36, 4, 274–280
276 Cardiac function during exercise early after cardiac surgery, H. H. Dedichen et al.
analysed for mitral early (E) and atrial (A) waves, left ventricle outflow tract – velocity time integral (LVOT-VTI), isovolumic relaxation time (IVRT) and tricuspid E and A waves. Velocity measurements were averaged over three cardiac cycles with stable values. Echo recordings were made during the last two of the five minutes on each workload, stored digitally and analysed in Echopac (BT 11, GE Vingmed). Statistics A sample size of 21 was calculated to detect a change in the ratio between early mitral flow and early mitral tissue velocity (E/e’) of two or more with a power of 80% and a significance level of 005. Intermediary analysis showed highly significant increase in E/e’ and according to the approval from the local ethics committee inclusion of patients was stopped. Graphical inspection confirmed normal distribution of data and parametric methods were applied. Data were stored and analysed in Stata IC 12 (Stata Corp, College Station, TX, USA). Data are summarized as mean SD. Paired samples t-test was used to estimate P-values for differences in preoperative and postoperative Borg score, Hb and arterial oxygen content. Linear mixed models with exercise intensity and surgery as explanatory variables were used for analysis of echocardiographic data.
Results Eighteen patients underwent preoperative testing. Four of them were not tested after the operation because of ongoing postoperative bleeding (1), dyspnoea (1), mild postoperative confusion (1), and in one case, surgery was cancelled due to factors not related to the patient. Subject characteristics of the 14 patients that completed the postoperative test are shown in Table 1. Perioperative data are displayed in Table 2. One patient stopped after completing the postoperative 10W test due to exhaustion. Thus, data were collected at (14 9 6) 1 = 83 occasions. Images of sufficient quality for analysis were obtained in over 90% of the occasions except for mitral atrial (71%, reduced due to E- and A- wave fusion) and
Table 1
Patient characteristics.
Age (years) Height (cm) Weight (kg) EF echo (%) EF angio (%) Treatment for hypertension Diabetes mellitus b-blocker
Mean SD n = 14
Range
58 8 178 6 87 13 54 3 61 19 8 (57) 1 (7) 14 (100)
46–70 166–186 69–113 45–55 35–78
Values are mean SD and range or number and (%). EF, ejection fraction.
Table 2
Perioperative data.
Number of bypasses ECC time (min) XC time (min) Lowest intraoperative body temperature (°C) Perioperative fluid balance (ml) Perioperative blood loss (ml)
Mean SD n = 14
Range
2–5 48–87 26–53 349–366
36 68 42 355
07 15 8 05
3906 1051 1221 444
2469–6335 740–2250
Values are mean SD and range or number and (%). ECC time, duration of extra-corporal circulation; XC time, duration of myocardial ischaemia during surgery. Fluid balance and blood loss is from 6 a.m. on the day of surgery to 6 a.m. the following morning. Table 3 E A e’ a’ MAE s’ LVOT-VTI IVRT
Feasibility of exercise echocardiography. 98 80 96 96 96 96 95 95
Tric E Tric A RV e’ TV a’ TAPSE RV s’
98 52 98 94 92 98
Per cent of possible observations of each variable.
tricuspid atrial wave (52%, fore same reason as for mitral A), see Table 3 for details. All patients that were tested postoperatively recovered well from surgery and were in sinus rhythm with satisfying heart rate on the first postoperative morning. There were no complications (arrhythmia, chest pain, hypotension) during testing. Echocardiography did not reveal any significant pericardial effusion in any of the patients. Borg score at 10W increased from 94 18 preoperatively to 113 15 postoperatively and at 30 W from 106 20 preoperatively to 136 19 postoperatively (both P
doi: 10.1111/cpf.12224
Cardiac function assessed by exercise echocardiography on the first morning after coronary artery bypass grafting Hans Henrik Dedichen1,2, Idar Kirkeby-Garstad1,3, Petter Aadahl1,3, Jonny Hisdal4 and Brage H. Amundsen5,6 1
Department of Circulation and Medical Imaging, Circulation Research University of Trondheim, Norwegian University of Science and Technology,Trondheim, Norway, 2Department of Cardiothoracic Surgery, St. Olav’s Hospital, Trondheim University Hospital, 3Department of Anesthesiology and Intensive Care Medicine, St. Olav’s Hospital, Trondheim University Hospital, Trondheim, Norway, 4Section of Vascular Investigations, Oslo University Hospital Aker, Oslo, Norway, 5 Department of Cardiology, St. Olav’s Hospital, Trondheim University Hospital, and 6MI lab, Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, Trondheim, Norway
Summary Correspondence Hans Henrik Dedichen, Department of Circulation and Medical Imaging, NTNU, Box 8905, 7491 Trondheim, Norway E-mail: [email protected]
Accepted for publication Received 14 May 2014; accepted 24 November 2014
Key words cardiac surgery; diastolic function; exercise tolerance; systolic function; tissue Doppler imaging
Cardiac surgery patients are urged to resume light physical activity on the first postoperative day, even if cardiac function may not have recovered fully after the operation. To elucidate the postoperative recovery process, we examined cardiac surgery patients with exercise echocardiography before and on the first day after the operation. Patients undergoing on-pump coronary artery bypass grafting were examined with echocardiography during semirecumbent cycle exercise. Patients exercised for five minutes at 10 W intensity and five minutes at 30 W intensity in bed with the upper body supported to approximately 30°. Fourteen patients were studied. Mitral annulus excursion and pulsed wave Doppler from the left ventricular outflow tract indicated postoperatively reduced cardiac stroke volume. Early diastolic tissue velocities of the mitral annulus were reduced, and early trans-mitral flow velocity was increased. The ratio between early mitral flow velocity and early diastolic mitral tissue velocity was increased postoperatively, indicating impaired left ventricular relaxation and increased left atrial pressure. Postoperative systolic mitral annulus tissue velocities were similar to preoperative velocities, indicating maintained systolic function. Postoperative exercise was associated with improvements in myocardial function indices and cardiac stroke volume similar to preoperative improvements. There were no signs of further deterioration in myocardial function during 30 W exercise. In summary, reduced left ventricular diastolic function after surgery resulted in reduced cardiac stroke volume, increased left atrial pressure and a higher rate of perceived exertion on the first postoperative day.
Introduction Cardiac surgery patients are urged to start physical activity at a time when myocardial function may still be transiently impaired after the operation. In our institution and in many other cardiac surgery units, most patients are mobilized on the first postoperative day (Westerdahl & Moller, 2010). Myocardial dysfunction may reduce exercise tolerance and limit early postoperative physical activity. Whereas postoperative myocardial function at rest has been extensively investigated (Ng et al., 2005; Apostolakis et al., 2009; Leppikangas et al., 2011; Juhl-Olsen et al., 2012; Hashemi et al., 2013), less is known about myocardial function during physical activity in the early postoperative phase. In a previous study, we used right heart catheterization to examine patients during semirecumbent cycling on the first 274
day after cardiac surgery. Mixed venous oxygen saturation was reduced, and blood lactate concentrations were increased after surgery, indicating reduced oxygen delivery relative to oxygen demand. Thermodilution measurements indicated that cardiac stroke volume was maintained at postoperative rest, but increased less during postoperative than preoperative exercise (Kirkeby-Garstad et al., 2006). However, the thermodilution technique may be unreliable in spontaneously breathing patients (Sasse et al., 1994). Transthoracic echocardiography provides information on cardiac function that is complementary to the hemodynamic data from invasive right heart catheterization. Echocardiography with tissue Doppler imaging and Doppler flow measurements is feasible and reproducible during cycling (Bougault et al., 2008) and can provide detailed quantitative information on biventricular systolic and diastolic function.
© 2014 Scandinavian Society of Clinical Physiology and Nuclear Medicine. Published by John Wiley & Sons Ltd 36, 4, 274–280
Cardiac function during exercise early after cardiac surgery, H. H. Dedichen et al. 275
Thus, the aim of this study was to use exercise echocardiography to explore if our previous findings regarding postoperative cardiac stroke volume were supported and if they may be explained by postoperative myocardial dysfunction. We hypothesized that the echocardiographic indices of cardiac function and stroke volume would be reduced on the first day after coronary artery bypass grafting (CABG) and that postoperative exercise would cause smaller increases in echocardiographic indices of myocardial function than preoperative exercise.
Methods We included 18 patients admitted for elective on-pump CABG at St. Olav’s University Hospital, Trondheim, in the period August 2010 to March 2012. Inclusion criteria were patients admitted for CABG without concomitant procedures. Exclusion criteria were unstable coronary artery disease, severe impairment of left ventricular function (EF 20 mmHg, aortic insufficiency ≥ grade 2, mitral valve insufficiency ≥ grade 2), clinical diagnosis of heart failure, chronic obstructive pulmonary disease more severe than GOLD 2 (Global initiative for chronic Obstructive Lung Disease), renal disease (serum creatinine >140 lmol l 1) and impaired mobility. The patients received standard treatment and care during the hospitalization. General anaesthesia for CABG was induced with diazepam, thiopental, fentanyl and pancuronium. Anaesthesia was maintained with isoflurane and fentanyl. During extra-corporal circulation (ECC), patients were sedated with propofol and cooled to approximately 35°C. The heart was arrested by cross-clamping of the aorta and infusion of cold St. Thomas’ cardioplegia. Standard revascularization was made by anastomosing the left internal thoracic artery to the left anterior descending artery and placing vein grafts from the ascending aorta to the circumflex artery and the right coronary artery. The study was performed in accordance with the ethical standards of the Helsinki declaration and approved by the regional ethics committee for western Norway (REK 2010/ 1543-6-1). Exercise protocol The patients were tested with semirecumbent two-leg bicycle exercise, and the same test protocol was used in pre- and postoperative tests. The preoperative test was made on the day of the operation with the patients fasted and prepared for surgery. The postoperative test was made on the first morning after surgery after removal of drains. The postoperative test was made on the first morning after surgery when the drains had been removed from the mediastinum. The patients were given three litres of oxygen per minute in a nasal catheter during the postoperative tests. Patients were exercising in bed with the upper body supported to approximately 30°. The
bicycle ergometer (Angio, Lode B.V., Groningen, the Netherlands) was secured to the bed in a position that was reasonably comfortable. The ergometer automatically adjusts braking force to pedalling frequency to keep the exercise intensity at the set level. The patients cycled for five minutes at 10 W intensity before the intensity was increased to 30 W, and the cycling continued without interruption for five minutes. Pedalling frequency was 50–60 cycles min 1. Measurements Heart rate (HR) was measured with five lead ECG connected to the patient monitor (Philips IntelliVue MP 70/MP 90, Andover, MA, USA). Arterial pressure was measured through a catheter inserted in the left radial artery before the preoperative test. Central venous pressure (CVP) was measured with a catheter in the superior caval vein inserted after the preoperative test. Both catheters were connected to pressure transducers (Edwards Lifesciences Pressure monitoring Transducers, Irvine, CA, USA) and calibrated with the zero-point five centimetre below the sternoclavicular joint. Readings were made before exercise and during the last minute of cycling on each workload. Blood samples were drawn from the radial artery catheter before exercise and during the last minute of exercise on each resistance level. The samples were stored on ice during the procedure and analysed immediately after the test for arterial oxygen saturation and arterial lactate concentration with ABL 800 blood gas analyzer (Radiometer, Brønshøj, Denmark). The patients indicated their feeling of exertion during cycling on 10 W and 30 W, using the Borg 6-20 scale (Borg, 1998). Echocardiography Echocardiography was performed with a Vivid 7 scanner with a 25 MHz phased array M3S transducer (GE Vingmed, Horten, Norway). All examinations were made by the same operator (BHA). B-mode 2D and colour tissue Doppler images in the two- and four-chamber views were recorded. Mitral annular tissue velocities were measured in the septal, lateral, inferior and anterior points and averaged. Mitral annular excursion (MAE) was also averaged over the same four points. Right ventricular tissue velocities and tricuspid annulus plane systolic excursion (TAPSE) were obtained at the lateral tricuspid annulus in a four-chamber view. Tissue velocities were measured using colour tissue Doppler with a frame rate of 100–160 frames s 1 both at rest and during exercise, as pulsed wave tissue Doppler was considered less reliable during exercise due to respiratory motion of the heart. Early (e’) and late (a’) diastolic as well as systolic (s’) tissue velocities of the tricuspid and mitral annuli were analysed. Blood flow velocities were measured by pulsed wave Doppler across the mitral and tricuspid valves, and in the left ventricular outflow tract, preferably in end-expiration. Doppler recordings were
© 2014 Scandinavian Society of Clinical Physiology and Nuclear Medicine. Published by John Wiley & Sons Ltd 36, 4, 274–280
276 Cardiac function during exercise early after cardiac surgery, H. H. Dedichen et al.
analysed for mitral early (E) and atrial (A) waves, left ventricle outflow tract – velocity time integral (LVOT-VTI), isovolumic relaxation time (IVRT) and tricuspid E and A waves. Velocity measurements were averaged over three cardiac cycles with stable values. Echo recordings were made during the last two of the five minutes on each workload, stored digitally and analysed in Echopac (BT 11, GE Vingmed). Statistics A sample size of 21 was calculated to detect a change in the ratio between early mitral flow and early mitral tissue velocity (E/e’) of two or more with a power of 80% and a significance level of 005. Intermediary analysis showed highly significant increase in E/e’ and according to the approval from the local ethics committee inclusion of patients was stopped. Graphical inspection confirmed normal distribution of data and parametric methods were applied. Data were stored and analysed in Stata IC 12 (Stata Corp, College Station, TX, USA). Data are summarized as mean SD. Paired samples t-test was used to estimate P-values for differences in preoperative and postoperative Borg score, Hb and arterial oxygen content. Linear mixed models with exercise intensity and surgery as explanatory variables were used for analysis of echocardiographic data.
Results Eighteen patients underwent preoperative testing. Four of them were not tested after the operation because of ongoing postoperative bleeding (1), dyspnoea (1), mild postoperative confusion (1), and in one case, surgery was cancelled due to factors not related to the patient. Subject characteristics of the 14 patients that completed the postoperative test are shown in Table 1. Perioperative data are displayed in Table 2. One patient stopped after completing the postoperative 10W test due to exhaustion. Thus, data were collected at (14 9 6) 1 = 83 occasions. Images of sufficient quality for analysis were obtained in over 90% of the occasions except for mitral atrial (71%, reduced due to E- and A- wave fusion) and
Table 1
Patient characteristics.
Age (years) Height (cm) Weight (kg) EF echo (%) EF angio (%) Treatment for hypertension Diabetes mellitus b-blocker
Mean SD n = 14
Range
58 8 178 6 87 13 54 3 61 19 8 (57) 1 (7) 14 (100)
46–70 166–186 69–113 45–55 35–78
Values are mean SD and range or number and (%). EF, ejection fraction.
Table 2
Perioperative data.
Number of bypasses ECC time (min) XC time (min) Lowest intraoperative body temperature (°C) Perioperative fluid balance (ml) Perioperative blood loss (ml)
Mean SD n = 14
Range
2–5 48–87 26–53 349–366
36 68 42 355
07 15 8 05
3906 1051 1221 444
2469–6335 740–2250
Values are mean SD and range or number and (%). ECC time, duration of extra-corporal circulation; XC time, duration of myocardial ischaemia during surgery. Fluid balance and blood loss is from 6 a.m. on the day of surgery to 6 a.m. the following morning. Table 3 E A e’ a’ MAE s’ LVOT-VTI IVRT
Feasibility of exercise echocardiography. 98 80 96 96 96 96 95 95
Tric E Tric A RV e’ TV a’ TAPSE RV s’
98 52 98 94 92 98
Per cent of possible observations of each variable.
tricuspid atrial wave (52%, fore same reason as for mitral A), see Table 3 for details. All patients that were tested postoperatively recovered well from surgery and were in sinus rhythm with satisfying heart rate on the first postoperative morning. There were no complications (arrhythmia, chest pain, hypotension) during testing. Echocardiography did not reveal any significant pericardial effusion in any of the patients. Borg score at 10W increased from 94 18 preoperatively to 113 15 postoperatively and at 30 W from 106 20 preoperatively to 136 19 postoperatively (both P
