Left ventricular diastolic dysfunction (LVDD) has been reported to have strong correlation with exercise capacity. However, this relationship has not been studied extensively in community-dwelling older adults. Data on pulse and tissue Doppler echocardiographic estimates of resting early (E) and atrial (A) transmitral peak inflow and early (Em) mitral annular velocities, and six-minute walk test were obtained from 89 community-dwelling older adults (mean age, 74; range, 65-93 years; 54% women), without a history of heart failure. Overall, 47% had cardiovascular morbidity and 60% had normal diastolic function (E/A 0.75-1.5 and E:Em <10). Among the 36 individuals with LVDD, 83%, 14% and 3% had grade I (E/A <0.75, regardless of E/E m ), II (E/A 0.75-1.5 and E/E m ≥10) and III (E/A>1.5 and E/E m ≥10) LVDD, respectively. Those with LVDD were older (77 versus 73 years; p = 0.001) and had a trend for higher prevalence of cardiovascular morbidity (58% versus 40%; p = 0.083). LVDD negatively correlated with six-minute walk distance (1013 versus 1128 feet; R = −0.25; p = 0.017). This association remained significant despite adjustment for cardiovascular morbidity (R = −0.35; p = 0.048), but lost significance when adjusted for age (R = −0.32; p = 0.105), age and cardiovascular morbidity (R = −0.38; p = 0.161), and additional adjustment for sex, race, body mass index, and systolic blood pressure (R = −0.44; p = 0.365). In conclusion, most community-dwelling older adults without heart failure had normal left ventricular diastolic function or grade-I LVDD. Although LVDD was associated with decreased performance on a six-minute walk test, that association was no longer evident after adjustment for age, body mass index and cardiovascular morbidity.
Left ventricular (LV) diastolic function is an independent predictor of exercise capacity in heart failure (HF). LV diastolic function has also been shown to be associated with functional capacity in younger adults referred for exercise testing. Aging is strongly associated with decreases in LV diastolic function and functional capacity. However, the relation between LV diastolic function and functional capacity has not been extensively studied in relatively healthy community-dwelling older adults. Further, the extent to which the age-associated decrease in LV diastolic function contributes to the decrease in functional capacity in community-dwelling older adults remains uncertain. The purpose of the present study was to examine the association between age, LV diastolic function, and physical function in a cohort of community-dwelling ambulatory older volunteers without HF.
Methods
Community-dwelling ambulatory older adults (≥65 years) without a history of HF, valvular heart disease, or atrial fibrillation were eligible for participation. Participants were recruited by community advertisements from 2003 through 2006. After obtaining informed consent participants were interviewed by trained research assistants and then underwent electrocardiography, echocardiography, and a 6-minute walk test. Participants with electrocardiographic or echocardiographic evidence of atrial fibrillation, LV ejection fraction <45%, and severe aortic or mitral insufficiency were excluded. Local institutional review boards approved the study protocol.
All echocardiographic measurements were performed by trained technicians under direct supervision of a trained echocardiographer (G.J.P.) using Philips ATL 5000 HDI ultrasound (Bothell, Washington) equipment. The technician and echocardiographer were blinded to participants’ cardiovascular morbidities. LV mass (indexed to height and weight) and LV ejection fraction were measured according to recommendations of the American Society of Echocardiography. Early (E) and atrial (A) transmitral maximal inflow velocities at tips of mitral valve leaflets by pulse Doppler and early (E m ) and atrial mitral annular velocity by pulse tissue Doppler at the base of the lateral wall 5 to 10 mm below the mitral annulus area in the apical 4-chamber view were obtained using standard techniques. Participants were stratified into grades of LV diastolic dysfunction (LVDD) as follows: grade 0 or normal diastolic function (E/A 0.75 to 1.5 and E/E m <10), grade I (E/A <0.75 regardless of E/E m ), grade II (E/A 0.75 to 1.5 and E/E m ≥10), and grade III (E/A >1.5 and E/E m ≥10).
The 6-minute walk test, a validated and reliable measurement of functional exercise capacity, was performed indoors along a flat firm surface by a trained technician according to a written protocol developed according to recommendations made by the American Thoracic Society guideline. Briefly, participants were instructed to walk back and forth for 6 minutes. They could slow down or stop and rest as necessary and then resume walking when they could. They were also gently encouraged to walk as fast as possible throughout the test as required. At the end of the walk, they were asked to report any symptoms such as pain, shortness of breath, or fatigue. Participants were categorized as having a short 6-minute walk test distance if they had walked <1,059 feet (median) in 6 minutes.
We categorized participants into 2 groups based on presence of LVDD. We compared baseline characteristics of participants including findings from echocardiographic and 6-minute walk test data by the presence or absence of LVDD using chi-square and Student’s t tests as appropriate. Linear regression models were used to determine relations between LVDD and 6-minute walk test, adjusting for age alone, cardiovascular morbidities alone, age and cardiovascular morbidities and additional adjustment for gender, race, body mass index, and systolic blood pressure. All tests were 2-tailed, and a p value <0.05 was considered statistically significant. All statistical analyses were conducted using SPSS 15 (SPSS, Inc., Chicago, Illinois).
Results
Participants (n = 89) had a mean age ± SD of 74 ± 6 years, 54% were women, 16% were nonwhite, and 47% had ≥1 cardiovascular morbidity. Overall 53 (60%) participants had normal LV diastolic function and 36 participants had LVDD. Of those with LVDD, 83% (30 of 36) had grade I, 14% (5 of 36) had grade II, and 3% (1 of 36) had grade III LVDD. Except for older age and a higher prevalence of diabetes in those with LVDD, most other baseline characteristics including baseline history of physical function were balanced between groups ( Table 1 ). Participants had a mean LV ejection fraction of 57 ± 3%, mean E/A ratio of 0.86 ± 0.23, peak E m wave of 10 ± 2 cm/s, and E/E m ratio of 7.0 ± 2.3 ( Table 2 ). As expected, E/A ratio and peak E m were lower in those with LVDD. However, E/E m ratio was similar between groups.
| All (n = 89) | LVDD | p Value | ||
|---|---|---|---|---|
| No | Yes | |||
| (n = 53) | (n = 36) | |||
| Age (years) | 74 ± 6 | 73 ± 5 | 77 ± 6 | 0.001 |
| Women | 48 (54%) | 26 (62%) | 22 (45%) | 0.263 |
| Nonwhite | 14 (16%) | 7 (13%) | 7 (19%) | 0.428 |
| Married | 41 (46%) | 27 (51%) | 14 (39%) | 0.263 |
| Education college or higher | 69 (78%) | 43 (81%) | 26 (72%) | 0.323 |
| Income ≥$25,000 | 47 (53%) | 26 (49%) | 21 (51%) | 0.390 |
| Smoker | ||||
| Current | 2 (2%) | 1 (2%) | 1 (3%) | 0.574 |
| Former | 29 (33%) | 15 (29%) | 14 (39%) | |
| Self-reported general health good or fair | 51 (57%) | 27 (51%) | 24 (67%) | 0.141 |
| Physical function/symptoms | ||||
| Dyspnea on exertion in previous 12 months | 13 (15%) | 8 (15%) | 5 (14%) | 0.846 |
| Able to climb 1 flight of stairs | 86 (97%) | 51 (96%) | 35 (97%) | 0.798 |
| Able to climb >2 flights of stairs daily | 39 (44%) | 19 (36%) | 20 (56%) | 0.066 |
| Able to walk about 3 city blocks | 87 (98%) | 51 (96%) | 36 (100%) | 0.238 |
| Walked about 3 city blocks in previous week | 13 (15%) | 8 (15%) | 5 (14%) | 0.874 |
| Less active than someone of similar age | 59 (66%) | 34 (64%) | 25 (69%) | 0.604 |
| Body mass index (kg/m 2 ) | 25 ± 4.6 | 25 ± 4.6 | 26 ± 4.7 | 0.294 |
| Pulse (beats/min) | 64 ± 10.0 | 64 ± 10 | 65 ± 10 | 0.489 |
| Systolic blood pressure (mm Hg) | 136 ± 18.8 | 134 ± 17 | 139 ± 21 | 0.197 |
| Diastolic blood pressure (mm Hg) | 70 ± 10.4 | 68 ± 10 | 72 ± 11 | 0.163 |
| Cardiovascular morbidities | 42 (47%) | 21 (40%) | 21 (58%) | 0.083 |
| Hypertension | 33 (37%) | 16 (30%) | 17 (47%) | 0.103 |
| Coronary artery disease | 7 (8%) | 4 (8%) | 3 (8%) | 0.892 |
| Diabetes mellitus | 6 (6.7%) | 1 (2%) | 5 (14%) | 0.027 |
| Stroke or transient ischemic attack | 4 (4.5%) | 2 (4%) | 2 (6%) | 0.705 |
| Peripheral vascular disease | 3 (3%) | 2 (4%) | 1 (3%) | 0.786 |
| Medications | ||||
| Angiotensin-converting enzyme inhibitor or angiotensin II receptor blocker | 9 (10%) | 6 (11%) | 3 (8%) | 0.646 |
| β Blocker | 12 (13%) | 10 (19%) | 2 (6%) | 0.071 |
| Diuretic | 8 (9%) | 5 (9%) | 3 (8%) | 0.859 |
| Electrocardiogram | ||||
| Left ventricular hypertrophy | 2 (2%) | 1 (2%) | 1 (3%) | 0.781 |
| First-degree heart block | 8 (10%) | 2 (4%) | 6 (18%) | 0.035 |
⁎ Participants were stratified as having normal left ventricular diastolic function (grade 0, early/atrial transmitral maximal inflow velocity 0.75 to 1.5 and early transmitral maximal inflow velocity/early mitral annular velocity <10) or left ventricular diastolic dysfunction. Those with left ventricular diastolic dysfunction were further classified as having grade I (early/atrial transmitral maximal inflow velocity <0.75 regardless of early transmitral maximal inflow velocity/early mitral annular velocity), grade II (early/atrial transmitral maximal inflow velocity 0.75 to 1.5 and early transmitral maximal inflow velocity/early mitral annular velocity ≥10), or grade III (early/atrial transmitral maximal inflow velocity >1.5 and early transmitral maximal inflow velocity/early mitral annular velocity ≥10).
| All (n = 89) | LVDD | p Value | ||
|---|---|---|---|---|
| No (n = 53) | Yes (n = 36) | |||
| Echocardiogram | ||||
| Ejection fraction (%) | 57 ± 3 | 57 ± 3 | 56 ± 3 | 0.703 |
| Early/atrial transmitral maximal inflow velocity ratio | 0.86 ± 0.23 | 0.96 ± 0.18 | 0.71 ± 0.21 | <0.001 |
| Peak early transmitral maximal inflow velocity wave (cm/s) | 65 ± 16 | 61 ± 12 | 68 ± 19 | 0.033 |
| Peak atrial transmitral maximal inflow velocity wave (cm/s) | 78 ± 17 | 72 ± 14 | 86 ± 17 | <0.001 |
| Peak early mitral annular velocity wave (m/s) | 9.8 ± 2.4 | 10.6 ± 2.1 | 8.7 ± 2.2 | <0.001 |
| Peak early transmitral maximal inflow velocity/peak early mitral annular velocity | 6.9 ± 2.3 | 6.8 ± 1.9 | 7.1 ± 2.8 | 0.548 |
| Left ventricular end-diastolic dimension (mm) | 45 ± 5 | 45 ± 5 | 44 ± 5 | 0.342 |
| Relative wall thickness | 0.43 ± 0.10 | 0.42 ± 0.08 | 0.46 ± 0.12 | 0.062 |
| Left ventricular mass index (g/m 2 ) | 34 ± 8 | 34 ± 7 | 35 ± 9 | 0.416 |
| Left atrial dimension (mm) | 37 ± 6 | 37 ± 6 | 37 ± 5 | 0.906 |
| 6-minute walk test | ||||
| Distance completed at 3 minutes (feet) | 542 ± 113 | 563 ± 119 | 512 ± 98 | 0.035 |
| Total distance completed at 6 minutes (feet) | 1,081 ± 226 | 1,128 ± 234 | 1,012 ± 198 | 0.017 |
| Symptoms during walk | 10 (11%) | 8 (15%) | 2 (6%) | 0.162 |
Associations of age with parameters of LV diastolic function are displayed in Figure 1 . E m at rest was negatively associated with age (R = −0.41, p <0.001) regardless of cardiovascular morbidities ( Figure 1 ). However, the overall significant association between age and E/A ratio (R = −0.24 p = 0.026) was significant only in those without cardiovascular morbidity (R = 0.44, p = 0.002) and not in those with cardiovascular morbidity (R = 0.04, p = 0.803; Figure 1 ). E/E m at rest was positively associated with increased age in the overall population (R = 0.29, p = 0.007) and in those with cardiovascular morbidity (R = 0.32, p = 0.038) but not in those without (R = 0.13, p = 0.371; Figure 1 ).
