Secondary Logo

Journal Logo

Cardiac Rehabilitation

Cardiac Rehabilitation After Heart Valve Surgery


Savage, Patrick D. MS; Rengo, Jason L. MS; Menzies, Keon E. MD, PhD, MPH; Ades, Philip A. MD

Author Information
Journal of Cardiopulmonary Rehabilitation and Prevention: July/August 2015 - Volume 35 - Issue 4 - p 231-237
doi: 10.1097/HCR.0000000000000104
  • Free

Individuals with heart valve (HV) disorders, in contrast with coronary artery bypass graft (CABG) surgery patients, often experience cardiac abnormalities and diminished functional capacity for several years before surgery.1 A period of postsurgical convalescence results in further declines in functional capacity for both HV and CABG patients. Peak aerobic capacity for CABG patients entering CR is exceedingly low,2 and there is evidence that values for individuals after HV surgery are particularly reduced.3–5

The Center for Medicare and Medicaid Services expanded cardiac rehabilitation (CR) coverage in 2006 to include patients following HV surgery. Current CR guidelines for HV patients are based primarily on results from randomized clinical trials in patients with coronary artery disease.6 Although patients after CABG have experienced improvements in aerobic fitness from CR exercise,7,8 there is a paucity of studies examining the outcomes for HV patients. Therefore, the primary aims of this study were to evaluate baseline peak aerobic capacity for HV patients participating in CR and to compare demographic and exercise training-related outcomes between patients undergoing HV and CABG surgery. We hypothesized that patients after HV surgery are less fit than patients after CABG but benefit similarly from the exercise training component of CR.


Five hundred seventy-six consecutive patients who underwent open heart surgery with a classic sternotomy (HV [n = 125], valve plus coronary artery bypass surgery [HV + CABG, n = 57], or CABG [n = 394]) and enrolled in CR between January 2006 and December 2012 were prospectively studied. The study protocol was approved by the Institutional Review Board at the University of Vermont and Fletcher Allen Health Care.

The number of CR sessions completed and, when appropriate, the self-reported reason for program discontinuation were recorded. Participation in CR was individualized. The number of CR sessions attended (up to a maximum of 36) was determined by medical necessity, insurance coverage, an individual's goals and objectives, and personal preference. Participants were considered “completers” of the program if they attended CR sessions and underwent a postprogram assessment.

Peak aerobic capacity was assessed with a symptom-limited graded exercise test on a treadmill before commencing CR. A postprogram exercise stress test was performed approximately 4 months from the baseline evaluation, regardless of the number of CR sessions attended. If an individual was unable to walk on a treadmill at a minimum of 2 miles per hour, a stress test was not performed (n = 52, 9%). Expired gas was continuously analyzed during the modified Balke exercise testing protocol using a Medgraphics Ultima CPX metabolic cart (Minneapolis, MN) and subjects exercised to voluntary exhaustion. Peak oxygen uptake (

O2) was considered to be the highest 30-second average during the test. Handgrip strength was measured using the dominant hand with the shoulder adducted and neutrally rotated, elbow at 90° flexion, and the forearm and wrist neutrally positioned using a Jamar handgrip dynamometer (Jamar, Bolingbrook, IL). The reported handgrip measure represented the mean of 3 consecutive attempts.

A diagnosis of hypertension and diabetes mellitus (DM) were recorded at the point of entry to CR, and smoking history was self-reported. Prescribed cardiac and preventive medications were reviewed and confirmed with the patient at entry to CR. Glycated hemoglobin (HbA1c) values were obtained during hospitalization. A comorbidity score was determined by assessing for peripheral vascular disease, cerebrovascular disease, chronic lung disease, or orthopedic limitations. If a comorbid condition was present, it was quantified by severity as follows: (1) present but not exercise-limiting; (2) present and impacted exercise performance; and (3) exercise-limiting. A total comorbidity score ranging from 0 to 12 was thus determined. Self-reported physical functioning was assessed using the Medical Outcomes Study 36-Item Short Form Health Survey questionnaire,9 which uses a 0 to 100 scale with 100 representing excellent physical functioning. Depressive symptoms were assessed using the Geriatric Depression Scale10 (scored 0 to 15), with higher scores indicating more depressive symptoms.

Specific details regarding surgery were gathered retrospectively via chart review. The type of valve abnormality, as well as the number of surgical arterial anastomoses, was obtained from the surgical report. Left ventricular ejection fraction was obtained from the preoperative echocardiogram or, if not available, from the left heart catheterization. Whether the patient was discharged to home or subacute rehabilitation was documented from the hospital discharge records. The time between the date of hospitalization and enrollment in CR was recorded.

The exercise training program has been described elsewhere11 and is similar to that performed at most rehabilitation programs around the United States.12 Subjects performed CR exercise at an intensity of 70% to 85% of their peak heart rate (65%-75% of peak

O2) and/or a Borg Scale rating of perceived exertion of between “light” and “somewhat hard” (12-14 on a scale of 6-20).13 In general, individuals exercised for 45 to 60 minutes per CR session on various modalities, including treadmills, elliptical trainers, and rowing, cycle, and arm ergometers. Typically, an exercise prescription consisted of 30 minutes of treadmill walking and 8 minutes on 2 other ergometers. Patients performed weight-training exercises consisting of 1 set of 10 repetitions for 6 different exercises targeting major muscle groups. Upper body strength training began 3 months after surgery. All patients were encouraged to exercise aerobically on non-CR days.

Statistical Analysis

For analysis purposes, the cohort was separated into 3 groups: HV, HV + CABG, and CABG. Analysis of variance was used to compare baseline variables between surgical groups. Contingency table analysis was used to compare nominal variables. For all valve patients, a stepwise linear multiple regression analysis was used to determine which variables independently correlated with change in peak

O2. Variables included in the regression analysis were age, sex, days from the index cardiac event and entry stress test, baseline body weight, entry body mass index, waist circumference, type of valve abnormality, left ventricular ejection fraction, the number of CR sessions attended, peak

O2, handgrip strength, and total comorbidity, depression, and physical function scores. A level of significance of P < .05 (2-tailed test) was used for hypothesis testing. Statistical analyses were carried out using the Stat View (SAS Institute Inc, Cary, NC) statistical package.


Demographic and clinical characteristics for the cohort are listed in Table 1. HV + CABG patients were significantly older than HV-only patients, in the CABG group and both groups of HV patients were significantly older than individuals in the CABG group. The number of days from index event to entry into CR was similar between groups. The percentage of females was greater in HV than in the HV + CABG and CABG groups. Individuals in the CABG group weighed significantly more than HV patients, but weight was similar between HV groups. For both body mass index and waist circumference, significantly higher values were observed in each of the CABG groups than for HV-only patients.

Table 1
Table 1:
• Clinical and Demographic Characteristics for the Entire Cohort and Separated By Surgical Groupsa

Combining valve groups, valvular disorders included 134 mitral, 39 aortic, and 8 combined abnormalities (mitral and aortic). The mean number of anastomoses was 3.2 ± 1.1 in the CABG group and 2.0 ± 1.0 in the HV + CABG groups. Left ventricular ejection fraction was significantly higher in both HV groups than for CABG patients. Prevalence of hypertension and smoking along with a diagnosis of type 2 diabetes mellitus (T2DM) were significantly lower in the HV group than in both of the CABG groups. Glycated hemoglobin values were significantly lower in both of the valve groups than for CABG patients. Cardiopreventive medication (β-blocker, calcium channel blocker, and angiotensin-converting-enzyme inhibitor) use was similar between groups except that significantly more patients in the CABG group were on statin therapy. The comorbid score was similar between the 3 groups. Upon hospital discharge, significantly more patients in the valve groups convalesced at a subacute rehabilitation facility than individuals in the CABG-only group.

Overall, 52 individuals or 9% of the cohort were unable to perform an entry stress test. Compared with CABG and HV patients, a significantly greater percentage of individuals in the HV + CABG group were unable to perform a baseline stress test. For individuals who were able to do an entry CR stress test, peak

O2 was lower in the HV + CABG group than for CABG and HV patients. Handgrip strength was lower for both groups of valve patients compared with individuals with CABG. The Medical Outcomes Study 36-Item Short Form Health Survey and Geriatric Depression Scale scores were similar among all surgical groups. The mean number of CR sessions attended was significantly greater in the HV + CABG group than in the CABG and HV groups.

Changes in outcome were assessed for individuals who had measures obtained at baseline and completion of CR (n = 313, 54.3% of total patients). Consequently, individuals who “completed” the program but did not have baseline peak

O2 measurements were excluded. A similar percentage of patients in each group (CABG = 53.6%; HV = 58.4%; HV + CABG = 52.6%; P = .55) had outcome data for the program. For the entire cohort, the peak

O2 increased 19.5% from 17.4 ± 4.4 to 20.8 ± 5.5 mLO2·kg−1·min−1 (Table 2) (P < .0001). Improvements in peak

O2 with CR exercise training were similar between the 3 groups of patients (Figure 1).

Figure 1
Figure 1:
Exercise capacity according to type of surgery at the start and the end of cardiac rehabilitation is presented. Percentage improvement from before (Pre) to after (Post) cardiac rehabilitation for each group is included above the bars.a P < .0001 for within-group comparisons. P = not significant for between-group comparisons.
Table 2
Table 2:
• Precardiac and Postcardiac Rehabilitation Values for the Entire Cohort and Separated By Surgical Groupa

Within the group of patients who had valve surgery, the peak

O2 increased 22.0% from 16.8 ± 5.2 to 20.3 ± 6.4 mLO2·kg−1·min−1 (P < .0001). The percentage increase in peak

O2 was similar between the 3 types of valvular abnormalities (ie, mitral [+19.2%], aortic [+24.4%], and mitral + aortic [+21.9%]; P = .27).

Forty-three (23.6%) of all valve patients went to subacute rehabilitation upon hospital discharge compared with 31 (7.9%) of the CABG patients. Compared with individuals who did not, a significantly greater percentage of individuals who went to subacute rehabilitation were unable to perform an entry stress test (4.3% vs 30.2%, respectively; P < .0001). For those valve patients who performed a baseline stress test, individuals who went to subacute rehabilitation had a lower baseline peak

O2 than patients who were discharged to home (13.0 ± 3.0 vs 17.6 ± 5.2 mLO2·kg−1·min−1, respectively; P < .0001). Similar improvements in peak

O2 were achieved for valve patients who went to subacute rehabilitation (2.7 ± 2.1 mLO2·kg−1·min−1) compared with those who did not (3.5 ± 2.8 mLO2·kg−1·min−1) (P < .22, between groups). In addition, among all individuals who went to subacute rehabilitation, similar improvements in peak

O2 were observed among CABG patients (1.7 ± 2.4 mLO2·kg−1·min−1) compared with all valve patients (2.6 ± 2.1 mLO2·kg−1·min−1) (P < .29, between groups).

Valve, HV + CABG, and CABG groups all achieved similar gains in strength as measured by handgrip dynamometer (Table 2). Self-reported physical function and depression questionnaire scores improved overall, and the changes were similar between groups (Table 2). Weight was unchanged, overall, and within each group (Table 2).

For all valve patients, factors that correlated with improvement in peak

O2 included—glycated hemoglobin (HbA1c) (r = −0.18; P < .004); diagnosis of T2DM (r = −0.07; P < .008); total comorbid score (r = −0.06; P < .02); age (r = −0.05; P < .03); and there was a trend toward improvement with the number of days between index event and entry to CR (r = −0.04; P < .07). Using stepwise multivariate analysis, HbA1c and age independently negatively correlated with change in peak

O2 (cumulative total r = 0.51, adjusted R2= 0.23; P < .002).


Our results demonstrate that patients who undergo HV surgery have a similar baseline peak aerobic capacity and achieve similar improvements in aerobic fitness as a result of CR exercise training as individuals who have undergone CABG. In addition, HV patients, regardless of the type of abnormality or whether CABG was performed concurrently, experience similar improvements in exercise capacity.

It has been previously reported that CABG patients participating in CR experience significant improvements in aerobic fitness.7,8 Although previous studies have demonstrated improvements in aerobic fitness for postsurgical HV patients participating in CR,14–17 the effect of exercise training following HV or HV + CABG surgery compared with patients who had undergone CABG surgery is less studied. In a study of exclusively HV patients from Belgium, Pardaens et al17 reported improvements in aerobic fitness that were similar to our results. Additionally, Pardaens et al17 reported that HV patients, regardless of preoperative risk or type of surgery (mini- or full sternotomy or port access), obtain a similar benefit from training. Together, these studies confirm that exercise training protocols used in CR are effective for patients after undergoing HV surgery, CABG or both.

It has previously been shown that exercise capacity is related to subsequent survival in individuals with coronary heart disease.18–20 Improvement in peak

O2 with CR exercise training has also been associated with decreased mortality rate.21 A previous study by Goel et al,22 reported a significant survival benefit with CR participation in patients undergoing HV + CABG. Further study is needed to determine if, absent coronary heart disease, a similar survival benefit exists with improvements in aerobic fitness for HV patients participating in CR.

Although our study groups experienced similar improvements in aerobic fitness as a result of participating in CR, significant differences at baseline existed between the groups. Valve + CABG patients were significantly older than HV-only or CABG patients. Valve and HV + CABG patients were more likely to convalesce, following hospital discharge, in a subacute rehabilitation facility. In addition, the length of time it took to enroll in CR was longest, and baseline peak

O2 was lowest in the HV + CABG group. Finally, valve + CABG patients were more often deemed unable to perform a baseline stress test due to extremely low aerobic fitness levels compared with the other 2 groups. Combined, these characteristics suggest that valve + CABG patients are more disabled at entry to CR than CABG or HV patients. Valve + CABG patients completed more CR sessions than the other groups; however, they experienced similar improvements in aerobic fitness, handgrip strength, and self-reported physical function and depression scores.

For the entire study cohort, individuals who attended subacute rehabilitation were more likely to be deemed too unfit to perform a baseline stress test. For patients who performed a baseline stress test, attendees of subacute rehabilitation had a lower peak aerobic capacity. For individuals who had baseline and exit measures, however, peak

O2 improved similarly for those individuals who went to subacute rehabilitation compared with those who did not. Consequently, despite being significantly less aerobically fit at entry to CR, individuals who convalesced in subacute rehabilitation facilities should be encouraged to participate in CR as they experience similar improvements in aerobic fitness as those patients who were not in such facilities.

Among valve patients, a diagnosis of T2DM was negatively correlated with changes in peak

O2, and HbA1c was the strongest independent factor associated with improvements in aerobic fitness. For individuals with coronary heart disease, it has been previously shown that a diagnosis of T2DM is associated with less improvement in peak

O2.11,23 Similarly, our results indicate that T2DM and HbA1c values are negatively associated with changes in aerobic fitness for individuals rehabilitating from valve-related surgery.

In addition to HbA1c, age was the other independent predictor of change in peak

O2 among the valve patients. In a previous report of CABG patients, age was not correlated with change in peak

O2,11 and, in this study, valve patients were significantly older than individuals in the CABG group. Alternatives to currently used exercise training protocols maybe indicated for older valve patients.

Our study has limitations. The results presented are from 1 CR center. The study design, while prospective, was observational and nonrandomized. Cardiac rehabilitation, however, is considered a standard of care,12 precluding randomizing individuals to a nonexercising control group. Without a control group, the extent of the observed improvements in aerobic fitness that are spontaneous and the amount that is the result of participating in CR are unknown. However, previous studies suggest that, without participating in CR, exercise capacity does not improve for individuals recovering from valve surgery.5,24 Despite the lack of a control group, our results are relevant as they represent what was observed in a clinical CR program. Our analysis is also limited in that we do not have outcome measures for the individuals who did not have both pre- and post-CR measures, and we do not have information regarding exercise training intensity.


Patients who undergo HV surgery gain similar improvements in aerobic fitness from participating in CR exercise training as individuals who have undergone CABG. The observed improvements in aerobic fitness are similar, regardless of the type of valve impairment or whether coronary artery bypass was performed concurrently. In addition, CABG and HV patients experience similar improvements in strength and self-reported physical function and depression scores.


This research was supported in part by the National Institutes of Health Center of Biomedical Research Excellence award P20GM103644 from the National Institute of General Medical Sciences (NIGMS).


1. Bissessor N, Stewart R, Wee YS, et al. Complex valve disease: pre-surgical functional capacity evaluation using peak oxygen consumption. J Heart Valve Dis. 2009;18(5):554–561.
2. Ades PA, Savage PD, Brawner CA, et al. Aerobic capacity in patients entering cardiac rehabilitation. Circulation. 2006;113(23):2706–2712.
3. Khan JH, McElhinney DB, Hall TS, Merrick SH. Cardiac valve surgery in octogenarians: improving quality of life and functional status. Arch Surg. 1998;133(8):887–893.
4. Niemelä K, Ikäheimo M, Takkunen J. Determination of the anaerobic threshold in the evaluation of functional status before and following valve replacement for aortic regurgitation. Cardiology. 1985;72(4):165–173.
5. Nakamura M, Chiba M, Ueshima K, et al. Effects of mitral and/or aortic valve replacement or repair on endothelium-dependent peripheral vasorelaxation and its relation to improvement in exercise capacity. Am J Cardiol. 1996;77(1):98–102.
6. Heran BS, Chen JM, Ebrahim S, et al. Exercise-based cardiac rehabilitation for coronary heart disease. Cochrane Database Syst Rev. 2011. doi:10.1002/14651858.CD001800.pub2.
7. Lan C, Chen SY, Hsu CJ, Chiu SF, Lai JS. Improvement of cardiorespiratory function after percutaneous transluminal coronary angioplasty or coronary artery bypass grafting. Am J Phys Med Rehabil. 2002;81(5):336–341.
8. Hsu CJ, Chen SY, Su S, et al. The effect of early cardiac rehabilitation on health-related quality of life among heart transplant recipients and patients with coronary artery bypass graft surgery. Transplant Proc. 2011;43(7):2714–2717.
9. Ware JE, Sherbourne CD. The Medical Outcomes Study: a 36 item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992;30:473–483.
10. Burke WJ, Roccaforte WH, Wengel SP. The short form of the Geriatric Depression Scale: a comparison with the 30-item form. J Geriatr Psychiatry Neurol. 1991;4:173–178.
11. Savage PD, Antkowiak M, Ades PA. Failure to improve cardiopulmonary fitness in cardiac rehabilitation. J Cardiopulm Rehabil Prev. 2009;29(5):284–291.
12. Williams MA, ed. AACVPR Guidelines for Cardiac Rehabilitation Programs. 5th ed. Champaign, IL: Human Kinetics; 2013.
13. Thompson WR, ed. ACSM's Guidelines for Exercise Testing and Prescription. 8th edition. Baltimore, MD: Lippincott Williams and Wilkins; 2010.
14. Gohlke-Bärwolf C, Gohlke H, Samek L, et al. Exercise tolerance and working capacity after valve replacement. J Heart Valve Dis. 1992;1(2):189–195.
15. Meurin P, Iliou MC, Ben Driss A, et al. Working Group of Cardiac Rehabilitation of the French Society of Cardiology. Early exercise training after mitral valve repair: a multicentric prospective French study. Chest. 2005;128(3):1638–1644.
16. Jairath N, Salerno T, Chapman J, Dornan J, Weisel R. The effect of moderate exercise training on oxygen uptake post-aortic/mitral valve surgery. J Cardiopulm Rehabil. 1995;15(6):424–430.
17. Pardaens S, Moerman V, Willems AM, et al. Impact of the preoperative risk and the type of surgery on exercise capacity and training after valvular surgery. Am J Cardiol. 2014;113(8):1383–1389.
18. Keteyian SJ, Brawner CA, Savage PD, et al. Peak aerobic capacity predicts prognosis in patients with coronary heart disease. Am Heart J. 2008;156(2):292–300.
19. Kavanagh T, Mertens DJ, Hamm LF, et al. Peak oxygen intake and cardiac mortality in women referred for cardiac rehabilitation. J Am Coll Cardiol. 2003;42(12):2139–2143.
20. Kavanagh T, Mertens DJ, Hamm LF, et al. Prediction of long-term prognosis in12 169 men referred for cardiac rehabilitation. Circulation. 2002;106(6):666–671.
21. Vanhees L, Fagard R, Thijs L, Amery A. Prognostic value of training-induced change in peak exercise capacity in patients with myocardial infarcts and patients with coronary bypass surgery. Am J Cardiol. 1995;76(14):1014–1019.
22. Goel K, Pack QR, Lahr B, et al. Cardiac rehabilitation is associated with reduced long-term mortality in patients undergoing combined heart valve and CABG surgery [published online ahead of print November 21, 2013]. Eur J Prev Cardiol. doi:1177/2047487313512219.
23. Vergès B, Patois-Vergès B, Cohen M, et al. Effects of cardiac rehabilitation on exercise capacity in type 2 diabetic patients with coronary artery disease. Diabet Med. 2004;21:889–895.
24. Habel-Verge C, Landry F, Desaulniers D, et al. Physical fitness improves after mitral valve replacement. CMAJ. 1987;136(2):142–147.

cardiac rehabilitation; coronary artery bypass surgery; heart valve surgery; peak aerobic capacity

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.