The maximum left atrial volume index (LAVI) has been shown to be of prognostic values, but previous studies have largely been limited to older patients with specific cardiovascular conditions. We examined the independent prognostic values of LAVI in a large unselected series of predominantly younger patients in sinus rhythm followed up for a long period. We evaluated 483 consecutive patients (mean age 47.3 years) using transthoracic echocardiography. The median LAVI was 24 ml/m 2 . A primary combined end point of cardiovascular death, stroke, heart failure, myocardial infarction, and atrial fibrillation was sought. We had complete follow-up data for 97.3% of the 483 patients. During a median follow-up of 6.8 years, 86 patients (18.3%) reached the primary end point. Older age, male gender, diabetes, hypertension, hypercholesterolemia, chronic renal failure, a history of myocardial infarction or stroke, a mitral E deceleration time of ≤150 ms, and LAVI of ≥24 ml/m 2 were univariate predictors of the primary end point. Event-free survival was significantly lower for patients with a LAVI of ≥24 ml/m 2 . Age, a history of stroke, hypertension, chronic renal failure, and male gender were independent clinical predictors. A LAVI of ≥24 ml/m 2 was the only independent echocardiographic predictor (hazard ratio 1.72, 95% confidence interval 1.34 to 2.13, p = 0.018), with the chi-square of the Cox model increased significantly with the addition of the LAVI (p <0.001). The LAVI independently predicted an increased risk of cardiovascular death, heart failure, atrial fibrillation, stroke, or myocardial infarction during a median follow-up of 6.8 years. In conclusion, the prognostic values were incremental to the clinical risks and were valid in a younger, general patient population.
The left atrium modulates left ventricular (LV) filling through its multiple functions: a reservoir during LV systole, a conduit for blood transiting in early diastole, an active contractile chamber in late diastole, and a suction source that refills itself in early systole. The left atrium also acts as a volume sensor, with the atrial wall releasing natriuretic peptides in response to stretch-generating potent effects. The interest in the assessment of the left atrial (LA) size and function has been considerable. A growing body of evidence has demonstrated that an enlarged left atrium is indicative of significant ventricular, atrial, or valvular disease. Therefore, the LA size has been increasingly recognized as a marker of adverse cardiovascular outcomes. However, most of the studies that examined the prognostic values of LA size had included older patients, and many were limited by their relatively small sample size, short follow-up duration, restricted end point events, or a selected study population with specific cardiovascular conditions. Furthermore, it is unclear whether the prognostic value of LA size is independent of clinical or other echocardiographic parameters. Therefore, we examined the independent prognostic values of LA volume in a large consecutive series of unselected patients, in sinus rhythm, with a diverse clinical background, who had been followed up for a long period.
Methods
A consecutive series of patients in normal sinus rhythm, who had undergone transthoracic echocardiography at the Liverpool Hospital Cardiology Department in 2000, was prospectively identified. The patient demographics, clinical characteristics and indications for, and results of, the studies were prospectively entered into a database (Clinical Reporting System). The following clinical characteristics were sought: diabetes mellitus, hypertension, hypercholesterolemia, current cigarette smoking, history of myocardial infarction, obesity (body mass index ≥30 kg/m 2 ), chronic renal failure (glomerular filtration rate ≤30 ml/min/m 2 ), history of stroke, and family history of premature (<60 years of age) coronary or cerebrovascular disease. Information on the patients’ cardioactive medications was also recorded. Patients with prosthetic valves, more than mild aortic stenosis or regurgitation, more than mild mitral regurgitation or any mitral stenosis, in atrial fibrillation, or with a history of paroxysmal atrial fibrillation were excluded. The other exclusion criteria were inadequate image quality, congenital heart disease, pericardial effusion >1 cm in depth, and atrial myxoma. For patients who underwent repeat studies, only the first study was analyzed and included. A total of 483 patients were included in the present study. The reasons for referral included assessment of LV function (71%), drug therapy monitoring (7%), chronic kidney disease (6%), assessment for cardiac sources of emboli (6%), assessment of cardiac function during pregnancy (4%), preoperative assessment (3%), cardiac assessment in trauma patients (1%), cardiac assessment in hematologic conditions (1%), and miscellaneous (1%). The Human Research Ethics Committee approved the study, and all patients gave written informed consent.
All patients underwent comprehensive transthoracic echocardiography in the left lateral decubitus position according to standard protocols using either a Philips SONOS 5500 or 7500 system. All measurements were performed off-line and taken as the average of ≥3 representative cycles. The maximum LA volume was measured at end-ventricular systole from apical 2- and 4-chamber views just before mitral valve opening. The minimum LA volume was measured at end-ventricular diastole using the biplane Simpson method. The pulmonary veins, LA appendage, and recesses of the atrioventricular valves were excluded in both measurements. Foreshortening was carefully avoided. The maximum LA volume was indexed to the body surface area to give the LA volume index (LAVI). All analyses were performed with the patient population dichotomized into 2 groups according to the median maximum LAVI and into quartiles. Our inter- and intraobserver variabilities of the maximum LA volume measurements have been previously reported. The LV end-systolic and end-diastolic volumes were measured according to the biplane Simpson method in the apical 4- and 2-chamber views. The endocardium was traced, excluding the papillary muscles. The LV ejection fraction was also calculated. The LV mass was measured at the parasternal short-axis view at the mid-papillary level according to a standard formula. The LV volumes and mass were indexed to the body surface area to give the respective indexes. LV hypertrophy was defined as a LV mass index of ≥102 g/m 2 for men and ≥88 g/m 2 for women. The LV inflow pattern was examined with pulsed wave Doppler with a 2-mm sample volume placed at the tips of the mitral valve leaflets. The mitral E and A wave velocities and deceleration times were measured according to standard techniques. A restrictive inflow pattern was defined as a mitral E/A ratio of ≥2 and a deceleration time of ≤150 ms.
The patients were prospectively followed up for development of clinical events. The following clinical events were sought: death from any cause, death from cardiovascular causes, myocardial infarction, congestive heart failure, atrial fibrillation, stroke, the development of symptomatic peripheral vascular disease, and myocardial revascularization procedures, including percutaneous coronary intervention and coronary artery bypass surgery. A combined primary end point of major adverse cardiovascular events (MACE) was defined as cardiovascular death, new-onset atrial fibrillation, stroke, myocardial infarction, or congestive heart failure. Patients were censored at the first clinical event. Follow-up was initially performed by contacting the Registry of Births, Deaths and Marriages in the State of New South Wales, Australia, to obtain information on the vital status of the patients and, for deceased patients, the cause of death. The hospital databases and records were also examined for similar information. All surviving patients were then sent a letter inviting them to participate in the follow-up. Follow-up was performed by telephone interview using a standard scripted questionnaire. Permission was also sought from the patients to contact their family physicians and to examine other relevant records to obtain further medical information, as necessary. Stroke was defined as a definite focal neurologic deficit of acute onset consistent with a vascular event lasting for >24 hours. Heart failure was considered present when the patients had developed symptoms and signs consistent with LV failure and had been confirmed by the investigations deemed necessary by the treating physicians; no distinction was made between systolic and diastolic heart failure. New-onset atrial fibrillation was defined as the development of atrial fibrillation, as confirmed using 12-lead electrocardiography. Myocardial infarction was defined according to standard criteria. Cardiac deaths were those due to myocardial infarction, arrhythmia, congestive heart failure, and death related to a cardiac event. Sudden death was considered cardiac death. Follow-up was complete for 470 patients (97.3%). These 470 patients formed the study population of the present study; only 13 patients were lost to follow-up.
All results are reported as the mean ± SD, unless stated otherwise. Continuous variables were compared using the unpaired Student t test, and categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. Kaplan-Meier event-free survival curves were plotted with the parameters compared using the log-rank test. A Cox proportional hazard regression model (forward stepwise selection, likelihood ratio method) was used to identify the independent predictors of the development of the combined end point of MACE, with the significant univariate clinical and echocardiographic variables entered as covariates. The clinical variables were entered first into the model, and the changes in the −2 log likelihood and chi-square of the model with the addition of the echocardiographic parameters were evaluated. The Statistical Package for Social Sciences for Windows, version 17 (SPSS, Chicago, Illinois), was used for the statistical analyses, and statistical significance was defined as a 2-tailed p value of <0.05.
Results
The mean age of our 470 patients was 47.3 ± 16.8 years (range 14 to 96), with 259 men (55%). The baseline clinical and echocardiographic characteristics are summarized in Table 1 . The median maximum LAVI was 24 ml/m 2 (interquartile range 18.6 to 30.3).
| Characteristic | Value |
|---|---|
| Clinical | |
| Diabetes mellitus | 79 (16.8%) |
| Hypertension | 197 (41.9%) |
| Obesity | 121 (25.7%) |
| Hypercholesterolemia | 178 (37.9%) |
| Current smoker | 126 (26.8%) |
| Previous myocardial infarction | 63 (13.4%) |
| Previous of stroke/transient ischemic attack | 41 (8.7%) |
| Family history of premature coronary/cerebrovascular disease | 101 (21.5%) |
| Chronic renal failure | 44 (9.4%) |
| Body mass index (kg/m 2 ) | 27.6 ± 5.6 |
| Antiplatelet drugs | 169 (36%) |
| β Blockers | 97 (20.6%) |
| Calcium antagonists | 53 (11.3%) |
| Angiotensin-converting inhibitors/angiotensin receptor blockers | 144 (30.7%) |
| Statin | 146 (31.1%) |
| Diuretics | 53 (11.3%) |
| Warfarin | 19 (4%) |
| Echocardiographic | |
| Left ventricular end-diastolic volume index (mL/m 2 ) | 50 ± 15 (20–127) |
| Left ventricular end-systolic volume index (ml/m 2 ) | 25 ± 10 (10–97) |
| Left ventricular ejection fraction (%) | 50.1 ± 10.5 (19–74) |
| Maximum left atrial volume index (mL/m 2 ) | 25.2 ± 8.5 (10.8–51.5) |
| Left atrial ejection fraction (%) | 54.5 ± 9.5 (27.1–80.2) |
| Mitral E wave velocity (cm/s) | 72.8 ± 18.1 (26.9–123.7) |
| Mitral A wave velocity (cm/s) | 70.8 ± 17.6 (35.2–133) |
| Mitral E wave deceleration time (ms) | 153.7 ± 35.8 (73.7–284) |
| Left ventricular mass index (g/m 2 ) | 81.3 ± 22.1 (25.3–158.6) |
During a median follow-up of 6.8 years, 144 patients (31%) experienced at least one clinical end point event. A total of 71 patients (15%) died; 21 of these deaths were due to cardiovascular causes. In addition, 33 patients (7%) experienced a stroke, 29 patients (6%) experienced myocardial infarction, 20 patients (4%) developed new-onset atrial fibrillation, 16 patients (3%) developed congestive heart failure, 23 (5%) had undergone percutaneous coronary intervention, 11 (2%) had undergone coronary artery bypass surgery, and 18 patients (4%) developed symptomatic peripheral vascular disease. The primary combined end point of MACE (cardiac death, congestive heart failure, myocardial infarction, new-onset atrial fibrillation, and stroke) was reached by 86 patients (18%). Survival free of MACE was significantly lower for the patients with a maximum LAVI of at or greater than the median value of 24 ml/m 2 ( Figure 1 ). When patients were subdivided into quartiles, the main differences in event-free survival were seen between the lower and upper 2 quartiles ( Figure 2 ). No statistically significant differences were found in event-free survival between the third and fourth quartiles of maximum LAVI (p = 0.97).
Table 2 lists the univariate clinical and echocardiographic predictors of MACE in our 470 patients. To initially identify the independent clinical predictors of MACE, the significant univariate clinical predictors were entered into the initial Cox model. Table 3 lists the independent clinical predictors of MACE. To identify the incremental prognostic value of the echocardiographic parameters for MACE during follow-up, significant univariate echocardiographic predictors were then entered into the Cox model as dichotomous variables. The presence of an enlarged LAVI was the only independent echocardiographic predictor for MACE. The −2 log likelihood of the overall model with the addition of an enlarged maximum LAVI (≥24 ml/m 2 ) into the Cox regression model decreased significantly from 866.1 to 860.4 (p <0.001), showing a better model fit. Similarly, the chi-square of the Cox regression model increased significantly with the addition of the enlarged maximum LAVI as a covariate from 94.2 to 99.2 (p <0.001).
| Variable | MACE (n = 86) | No MACE (n = 384) | p Value |
|---|---|---|---|
| Clinical | |||
| Age (years) | 58 ± 15 | 45 ± 16 | <0.001 |
| Men | 57 (66.3%) | 202 (52.6%) | 0.021 |
| Diabetes mellitus | 27 (31.4%) | 52 (13.5%) | <0.001 |
| Hypertension | 62 (72.1%) | 135 (35.2%) | <0.001 |
| Obesity | 25 (29.1%) | 96 (25%) | NS |
| Hypercholesterolemia | 48 (55.8%) | 130 (33.9%) | <0.001 |
| Current smoker | 22 (25.6%) | 104 (27.1%) | NS |
| History of myocardial infarction | 23 (26.7%) | 40 (10.4%) | <0.001 |
| History of stroke/transient ischemic attack | 20 (23.2%) | 21 (5.5%) | <0.001 |
| Family history of premature coronary/cerebrovascular disease | 21 (24.4%) | 80 (20.8%) | NS |
| Chronic renal failure | 19 (22.1%) | 25 (6.5%) | <0.001 |
| Echocardiographic | |||
| Left ventricular dysfunction (ejection fraction <50%) | 41 (47.7%) | 168 (43.8%) | NS |
| Left ventricular hypertrophy | 23 (26.7%) | 69 (17.9%) | 0.064 |
| Restrictive left ventricular filling | 2 (2.3%) | 5 (1.3%) | NS |
| Mitral E wave deceleration time ≤150 ms | 34 (39.5%) | 197 (51.3%) | 0.048 |
| Maximum left atrial volume index (ml/m 2 ) | 27.7 ± 8.8 | 24.6 ± 8.3 | 0.003 |
| Enlarged left atrial volume index ⁎ | 55 (63.9%) | 180 (46.9%) | 0.004 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
