Effect of Single-Drug Therapy on Reduction of Left Atrial Size in Mild to Moderate Hypertension
Comparison of Six Antihypertensive Agents
Background—Cardiac effects of hypertension include increased left ventricular (LV) mass and LV hypertrophy, as well as increased left atrial size, a predictor of stroke and atrial fibrillation. Although literature on reduction of LV mass with antihypertensive therapy is extensive, little information is available on effects of treatment on left atrial size.
Methods and Results—Patients with mild to moderate hypertension (diastolic blood pressure 95 to 109 mm Hg) were randomly allocated to treatment with atenolol, captopril, clonidine, diltiazem, hydrochlorothiazide, or prazosin in a double-masked trial. Two-dimensional targeted M-mode echocardiography was used to assess left atrial size and LV mass at baseline, 8 weeks, and 1 and 2 years. Longitudinal analysis examined changes in left atrial size from the baseline study, statistically adjusting for age, race, pretreatment left atrial size and LV mass, and serial measurements of systolic blood pressure, body weight, urinary sodium excretion, and physical activity score. Without adjustment for covariates, only hydrochlorothiazide was associated with decreases in left atrial size from baseline at 8 weeks (−1.0±5.2 mm; P=0.052), 1 year (−2.0±5.1 mm; P=0.02), and 2 years (4.6±7.2 mm; P=0.002). After adjustment for effects of covariates, patients with normal left atrial size had greater reduction (−3.3 mm) in left atrial size at 2 years with hydrochlorothiazide than with any other drug. For patients with left atrial enlargement, left atrial size decreased significantly with hydrochlorothiazide, atenolol, clonidine, and diltiazem at 1 year and with all treatments at 2 years. However, reduction at 2 years was greater with hydrochlorothiazide than with captopril or prazosin.
Conclusions—Antihypertensive drugs differ in their effects on left atrial size. Hydrochlorothiazide was associated with greater overall reduction of left atrial size than other drugs effective for the treatment of hypertension. Reduction of left atrial size with therapy is in part independent of factors known to influence left atrial size, including LV mass and reduction of LV mass with treatment. The clinical benefit of reducing left atrial size with antihypertensive treatment remains to be determined.
The cardiac effects of hypertension include increases in left ventricular (LV) mass1 and left atrial size.1 2 3 4 Whereas LV mass has been shown to be a predictor of cardiovascular events, including myocardial infarction and death, left atrial size is associated with the likelihood of developing atrial fibrillation5 6 and stroke.7 Although the literature on reduction of LV mass with antihypertensive therapy is extensive,4 8 relatively little information is available on the effects of antihypertensive treatment on left atrial size. Moreover, the influence of covariates other than drug selection that may affect left atrial size,4 9 such as body weight, systolic blood pressure reduction, age, race, sodium excretion, and physical activity, have not been evaluated.
The Veterans Affairs Cooperative Study on Single-Drug Therapy in Mild-Moderate Hypertension was a randomized titration trial of monotherapy using 6 active drugs and placebo to test the comparative efficacy of different classes of drugs for the lowering of diastolic blood pressure.10 One objective of the study was to use echocardiography to assess the response of cardiac structure and function to antihypertensive monotherapy. Because the mechanisms responsible for left atrial dilation, and presumably its regression, are likely multifactorial, the echocardiographic data were analyzed with adjustment for potentially contributory factors by use of appropriate statistical methods.
Criteria for patient selection, randomization, and treatment have been detailed elsewhere.10 Briefly, 1105 men, recruited at 15 Veterans Affairs Medical Centers, with diastolic blood pressure 95 to 109 mm Hg after a 4- to 8-week washout period on placebo, were randomly allocated to double-blind treatment with 1 of 6 drugs. A seventh group, randomized to placebo, was not included in analyses for this report. The medications were atenolol 25, 50, or 100 mg daily; captopril 12.5, 25, or 50 mg twice daily; clonidine 0.1, 0.2, or 0.3 mg twice daily; diltiazem-SR 60, 120, or 180 mg twice daily; hydrochlorothiazide 12.5, 25, or 50 mg daily; and prazosin 2, 5, or 10 mg twice daily. Patients who achieved the diastolic blood pressure goal of <90 mm Hg without adverse drug effects at the end of the 8-week titration period were then entered into a maintenance phase for ≤2 years. Patients entering the study in the last year of the recruitment phase had a maximum follow-up of 1 year.
After 6 to 12 weeks’ withdrawal of prior antihypertensive therapy (542 patients) or 4 to 12 weeks of baseline observation in the 242 patients not receiving therapy at study entry, 2-dimensional targeted M-mode echocardiography was performed. To minimize variability, measurement of left atrial dimension, ventricular septum, LV cavity, and posterior wall dimensions from paper strip-chart recordings were performed by a single reader using an off-line image analysis system, according to American Society of Echocardiography criteria.11
To determine left atrial dimension, the maximum dimension was measured between the leading edge of the posterior aortic wall and the leading edge of the posterior wall of the left atrium at end systole. Interobserver error for left atrial dimension, determined from comparison of 121 echocardiograms read by the core laboratory sonographer for this study with measurements made by a “reference reader” (J.S.G.), was 3.2%. Intraobserver error was 2.8% for sonographer and 1.7% for reference reader measurements.
LV mass was calculated as described elsewhere12 and converted to Cornell-Penn convention values by use of published regression equations.13 Left atrial size was analyzed as a linear dimension as well as indexed by height in an attempt to account for allometric, physiological differences in left atrial size based on body size. Indexing by body surface area, as has sometimes been done, was considered inappropriate because we wished to determine the effects of body weight on left atrial size and its change with treatment, and body weight is the principal determinant of variance in body surface area. Left atrial enlargement was defined as left atrial dimension ≥43 mm. This partition value was defined by the Framingham study9 as the 90th percentile value in normotensive, nonobese men.
Clinical and Laboratory Assessments
Blood pressure was measured with a cuff sphygmomanometer after 5 minutes’ rest sitting upright with the arm supported. All visit blood pressures represented the mean of 3 seated measurements with the back and arm supported, taken after 5 minutes’ rest. Baseline blood pressure was taken as the mean of the blood pressure averages determined at the randomization visit and the preceding visit. Treatment blood pressures were the mean of the blood pressure averages for the first 2 consecutive visits at which goal blood pressure was achieved. Hence, the average of 6 measurements taken over 2 visits was used. Physical activity index was obtained by an administered questionnaire that queried work and recreational physical activity for the 6-month time period preceding recruitment into the study. Plasma renin was determined by 125I radioimmunoassay (Clinical Assays, Travenol Division, Genentech Diagnostics, Inc). Sodium excretion was determined from one 24-hour measurement of total urinary sodium excretion (mmol).
In a previous publication from this study,4 baseline analysis of the relationships between left atrial size and body mass index, systolic blood pressure, sodium excretion, plasma renin, race, age, and physical activity disclosed independent effects of age, race, and body mass index. In the present study, these covariates were also examined in analyses of treatment effects on changes in left atrial size, with weight replacing body mass index.
A number of randomized patients were not included in the analysis because of lack of an adequate blood pressure response, blood pressure that rose above safety limits, or inability to obtain a readable echocardiogram. Therefore, baseline patient characteristics were compared between those with a left atrial dimension measurement at each time point and those without. The t test for independent groups was used for continuous variables, and the χ2 test was used for categorical variables.
Baseline comparability of the 6 treatment groups was examined for those with left atrial dimension measurements at each time point. One-way ANOVA was used for continuous variables and the χ2 test for categorical variables.
Comparisons of changes from baseline to 8 weeks, 1 year, and 2 years in mean left atrial dimension and patient covariates among treatment groups were performed by use of ANOVA. Pairwise comparisons of groups using Tukey’s method were made if the ANOVA statistic was significant at the 0.05 level. Within-group changes from baseline were analyzed by use of the paired t test.
Patients were then stratified into 2 groups (normal or enlarged left atrium) based on their pretreatment left atrial dimension. At 8 weeks, 1 year, and 2 years, a 2-way ANOVA was conducted to evaluate the effect of the 6 treatments and pretreatment left atrial size on changes in left atrial size from baseline.
To estimate and compare left atrial size slopes for the 6 treatments, adjusting for patient covariates, mixed-model ANOVA was performed.14 This is a type of repeated-measures analysis that allows for incomplete data. The computations proceeded in 2 stages. First, an appropriate regression model was selected to characterize the left atrial size response curve over time for each patient. We assumed a model that was linear in time with an intercept that was random across patients and a compound symmetry covariance structure among the serial LV mass measurements. Second, the “average” regression curve was estimated for each treatment group to examine whether regression curve slopes and intercepts differed across treatment groups after adjustment for patient covariates. Differences between treatments were assessed by specification of all possible pairwise contrasts. With the use of a Bonferroni adjustment for 15 possible comparisons, a P value of 0.0033 was considered statistically significant for inferences regarding comparisons of treatments.
Mixed-model analysis may be biased if missing data are due to a nonrandom process. Whereas nonacquisition of a readable echocardiogram was unrelated to either the assigned treatment or left atrial size, withdrawals from the study were primarily due to protocol safety rules regarding lack of blood pressure control. To account for this type of withdrawal, we examined whether the changes in left atrial size at the end of titration differed between those who were withdrawn at that time point and patients who advanced to the maintenance phase, after adjusting for the covariates listed earlier in this article.15
At 1 and 2 years, there were insufficient numbers of patients remaining on placebo (19 and 8, respectively) with a left atrial measurement to permit valid statistical comparisons. However, a valid comparison of effects of active therapy on left atrial size with the effects of placebo was possible during the 8-week titration period by inclusion of all randomized patients with a left atrial measurement in the analysis, both responders and nonresponders to treatment. Two analyses were done for this. First, a simple 1-way ANOVA was done that compared changes in left atrial size from baseline to end titration across the 7 treatment groups. Second, a similar comparison was done that adjusted for covariates that were found to be determinants of left atrial size.
The statistical computer package SAS (version 6.12) was used to generate statistical analyses.16 Values for quantitative measures are expressed as mean±SD unless otherwise stated. All statistical tests were 2-tailed, and P≤0.05 was used to identify statistically significant results.
Eleven hundred five patients were randomized to 1 of 6 treatment groups; 683 of these patients completed the antihypertensive medication titration phase (8 weeks), and 493 and 395 patients completed 1 and 2 years of antihypertensive medication maintenance therapy, respectively. Left atrium measurements were available for 784 (71%) of the randomized patients at baseline, 503 (46%) at 8 weeks, 252 (23%) at 1 year, and 166 (15%) at 2 years. There was no difference between treatment groups in the proportion of patients with left atrial size measurements at each time point (Table 1⇓). Table 1⇓ summarizes the reasons measurements were not obtained at each time point for the original group of randomized patients. At 8 weeks, the acquisition rate was reduced because 10% of patients had withdrawn from the study owing to uncontrolled blood pressure or adverse effects of the medication, and 45% did not have a left atrium measurement. A postrandomization measurement was not obtained if a baseline reading was unavailable. At 1 year, the acquisition rate was reduced further primarily because 30% of patients did not achieve adequate blood pressure reduction during the titration phase and were ineligible to proceed to the maintenance phase. At 2 years, the acquisition rate was reduced because patients who entered the study in the last year of recruitment were unable to reach 2 years of follow-up. When treatment groups were compared, the acquisition rate at 1 year was highest for the diltiazem group because superior blood pressure reduction in this group resulted in fewer protocol-dictated withdrawals from the study, whereas the rate was lowest for prazosin because a higher percentage of patients withdrew because of adverse effects of the medication. Table 2⇓ summarizes baseline characteristics for those patients with and those without an analyzed left atrium measurement at 8 weeks, 1 year, and 2 years. At 8 weeks, the analyzed group had a higher percentage of blacks (56.3%) than those without an analyzed left atrium measurement (41.5%; P=0.001). Racial differences were also evident at 1 and 2 years.
At 1 year, the analyzed group had lower values for baseline systolic blood pressure (150.2 versus 152.7 mm Hg; P=0.01) and baseline diastolic blood pressure (98.4 versus 99.5 mm Hg; P<0.001) than the unanalyzed group. At 2 years, the analyzed group had lower values for baseline diastolic blood pressure (98.2 versus 99.4 mm Hg; P<0.001) and baseline LV mass (310.1 versus 333.9 g; P=0.002).
Patient Characteristics at Baseline
At baseline, average age of the patients was 58.8±10.0 years, blood pressure averaged 152.4±13.7/99.3±3.4 mm Hg, and average left atrial size was 41.3±6.4 mm (range, 20 to 63 mm). Height-indexed left atrial dimension averaged 23.5±3.6 mm/m (range, 11.6 to 36.5 mm/m). Blacks made up 52% of the study sample. Left atrial enlargement was noted in 315 patients (40%).
Table 3⇓ compares baseline characteristics across the 6 treatment groups for patients included in the analysis at each time point. Comparison of baseline treatment differences among those with 8-week measurements showed baseline treatment group differences for diastolic blood pressure (P=0.04), ranging from 98.6 mm Hg for the atenolol group to 100.1 mm Hg for the prazosin group. For patients with 1-year measurements, baseline comparison across treatment groups revealed a significant difference in the percentage of blacks (P=0.01), which ranged from 39.1% in the captopril group to 73.3% in the prazosin group. There was also a trend (P=0.07) at 1-year for baseline LV mass to differ between groups, ranging from 301.8 g in the captopril group to 357.2 g in the prazosin group. For patients with 2-year measurements, there were significant differences across treatment groups for the percentage of blacks (P=0.003) and the physical activity index (P=0.003), which was lowest for the prazosin group and highest for the atenolol group. The lower acquisition rate for whites was primarily due to the unavailability of suitable echocardiographic equipment at a few predominantly white sites. There was no difference between blacks and whites at sites that had equipment available.
Effects of Treatment on Left Atrial Size
Serial changes in left atrial size across treatment groups are shown in Table 4⇓ and displayed in Figure 1⇓. With hydrochlorothiazide, left atrial size decreased −1.0±5.2 mm from baseline (P=0.052) at 8 weeks, −2.0±5.1 at 1 year (P=0.02), and −4.6±7.2 at 2 years (P=0.002). Whereas left atrial size increased on atenolol at 8 weeks (1.1±4.2 mm; P=0.03), there was a trend (P=0.08) for decrease on atenolol (−1.5±5.4 mm) at 1 and 2 years (−1.9±5.4; P=0.07). No other changes from baseline occurred with any other drug. Although there were significant overall decreases in left atrial size on treatment at 8 weeks (P=0.03) and at 1 year (P=0.04), no pairwise differences were identified by Tukey’s procedure.
At 1 and 2 years, there were insufficient numbers of patients remaining on placebo for statistical analysis. However, at 8 weeks the mean change in left atrial size for placebo was a decrease of 0.89 mm compared with changes ranging from a decrease of 1.03 mm for hydrochlorothiazide to an increase of 1.08 mm for atenolol. On ANOVA, no pairs of treatment groups differed statistically. After adjustment for covariates, atenolol (P=0.022), prazosin (P=0.040), and diltiazem (P=0.050) differed significantly from placebo.
Serial Changes in Patient Covariates
Changes in patient covariates across treatment groups are found in Table 5⇓. At 8 weeks, only changes in systolic blood pressure and weight demonstrated a significant difference across treatment groups. The clonidine group underwent a greater reduction in systolic blood pressure (−18.5 mm Hg) than the atenolol (−10.7 mm Hg), captopril (−10.7 mm Hg), diltiazem (−12.1 mm Hg), or prazosin (−13.3 mm Hg) groups. The prazosin group gained 2.5 lb, which was significantly greater than changes on all other drugs.
At 1 year, the hydrochlorothiazide group had a greater reduction in systolic blood pressure (−16.4 mm Hg) than the captopril group (−8.1 mm Hg). There was a trend (P=0.09) for changes in physical activity score to differ across groups, ranging from a decrease of 13.7 on atenolol to an increase of 9.1 on hydrochlorothiazide.
At 2 years, urinary sodium excretion decreases on hydrochlorothiazide differed from increases on diltiazem, whereas atenolol decreases in physical activity differed from increases for hydrochlorothiazide.
Longitudinal Analysis of Treatment Effects on Left Atrial Size
Estimated left atrial size, adjusted in the mixed-model analysis for patient covariates, is displayed in Figure 2⇓. This model was adjusted for treatment; time; treatment-time interaction; baseline left atrial size and interaction of baseline left atrial size with time; interaction of baseline left atrial size with treatment; interaction of baseline left atrial size, time, and treatment; race, baseline LV mass and serial measurements of systolic blood pressure, age, weight, and interaction of weight and time; urinary sodium excretion; and physical activity score. After adjustment for covariates, the following interdrug differences were found for patients with normal baseline left atrial size (median value 37 mm): at 1 and 2 years, respectively, the 1.4- and 3.3-mm reductions on hydrochlorothiazide were significantly greater than the reductions on any other drug. For patients with enlarged baseline left atrial size (median value 47 mm), the following interdrug differences were found: at 1 and 2 years, hydrochlorothiazide patients had a larger decrease than captopril or prazosin patients. In addition to drug selection, older age (P=0.008) was associated with less average reduction in left atrial size, whereas blacks tended to have larger decreases (P=0.10). Increased body weight was associated with smaller reductions over time (P<0.001). Finally, baseline left atrial size altered the changes in left atrial size over time differentially across treatment groups (P=0.004). Serial systolic blood pressure measurements (P=0.36), baseline LV mass (P=0.10), physical activity (P=0.80), and urinary sodium excretion (P=0.30) were not significant predictors of changes in left atrial size. Separately, a comparison of patients who dropped out at the end of titration showed that such patients had similar changes in left atrial size at 8 weeks compared with those who advanced to the maintenance phase.
The results of this study show that in men with mild to moderate hypertension and high prevalence of left atrial enlargement, various classes of antihypertensive drugs have differing effects on left atrial size. Principally, hydrochlorothiazide was associated with greater reduction of left atrial size than other drugs effective for treatment of hypertension. Moreover, the reduction in left atrial size with antihypertensive treatment was in part independent of differences in factors known to influence left atrial size in hypertensives, such as body weight, blood pressure, age, race, and LV mass. Furthermore, although both captopril and atenolol were associated with reduction of LV mass equivalent to that produced by hydrochlorothiazide, captopril was ineffective for reduction of left atrial size and enlargement. Although atenolol produced smaller reductions of LV mass than captopril, it was nonetheless associated with a trend toward reduction of left atrial size, albeit less than with hydrochlorothiazide.
Despite the plethora of studies on effects of antihypertensive therapy on LV mass, little information is available on effects of treatment on left atrial size. Whereas 1study of verapamil17 demonstrated decrease in left atrial size, another18 failed to show an effect of verapamil therapy on the left atrium, despite reduction of LV wall thickness (but not LV mass). An echocardiographic study of labetalol19 in hypertensive patients also failed to disclose reduction of left atrial size despite reduction of LV mass. The mechanisms whereby hydrochlorothiazide is associated with reduction of left atrial size whereas other drugs are not are uncertain. Reduction of intravascular volume or selective effects on LV or left atrial compliance are possibilities.
Limitations and Advantages of the Present Study
By design, patients were required to remain on the initially assigned monotherapy and were terminated from the study if they required other or additional medication for blood pressure control or changed therapy because of side effects. Hence, the analyses were not performed on an “intention-to-treat” basis. This design aspect of the study is consistent with clinical practice, in which patients do not continue to be given antihypertensive medication in the absence of blood pressure lowering, despite the possibility of favorable effects on cardiac anatomy.
Nonetheless, the findings could be biased by differences in dropout rates and redistribution of relevant biological characteristics across treatment groups. For example, the percentage of blacks with left atrial measurements differed between drugs, and those differences did not remain constant over time, reflecting the influence of race on blood pressure responses to monotherapy.10 However, most baseline characteristics of patients with readable echocardiograms remained comparable across treatment groups, and covariate adjustment (including race) for changes in biological predictors of LV mass increased the statistical significance of treatment comparisons and showed virtually no impact of missing data on the estimates of treatment effects. Although the requirement for adherence to monotherapy contributed to study dropouts, the study design permitted assessment of patients uncomplicated by drug crossovers. Despite study dropouts, 252 patients (30 to 54 patients per treatment limb) evaluated at 1 year represents a substantial number of patients studied on single-drug therapy.
For patients who were categorized as having left atrial enlargement, the overall reduction in left atrial size seen for all treatment groups likely reflected not only treatment effects but also regression to the mean. However, for subjects with normal left atrial size at baseline, a reduction in left atrial size was only observed for hydrochlorothiazide. This, as well as the significantly greater reduction of left atrial size with hydrochlorothiazide than with other drugs, even in patients with left atrial enlargement at baseline, cannot be explained by regression to the mean. Hence, it is likely that hydrochlorothiazide differs from the other drugs tested with regard to its efficacy for reducing left atrial size.
We studied only men who had a high prevalence (46% by Cornell criteria, 64% by Framingham criteria) of LV hypertrophy.8 Although this is higher than the 12% to 26% noted in some studies, it is consistent with the 40% to 51% prevalence noted in others.20 21 It is likely that our male, veteran population represents individuals with greater severity of hypertension than those in some other studies. Although it is difficult to know how generalizable our findings or the findings of others are to the “universe” of patients with hypertension, our data cannot be directly extrapolated to women or to hypertensives with less severe disease.
Left atrial size has prognostic value for clinical outcome in hypertension and other cardiovascular disorders.5 7 22 23 Although it would have been of interest to know if reduction of left atrial size was associated with decrease in incident atrial fibrillation, the present study was not designed to assess the effects of reduction of left atrial size on that or other outcome variables. Hence, it remains uncertain whether drugs that are effective for reduction of left atrial size confer benefit over and above that associated with reduction of blood pressure and LV mass. Future studies will be required to determine if superior reduction of left atrial size with hydrochlorothiazide, in combination with its demonstrated efficacy for reduction of LV mass8 as well as blood pressure, is associated with improved outcome.
Financial support for this study was provided by the Cooperative Studies Program of the Department of Veterans Affairs Research and Development Service.
- Received December 3, 1997.
- Revision received February 27, 1998.
- Accepted March 17, 1998.
- Copyright © 1998 by American Heart Association
Henry W, Morganroth J, Pearlman A, Clark C, Redwood D, Itscoitz S, Epstein S. Relation between echocardiographically determined left atrial size and atrial fibrillation. Circulation. 1976;53:273–279.
The Stroke Prevention in Atrial Fibrillation Investigators. Predictors of thromboembolism in atrial fibrillation, II: echocardiographic features of patients at risk. Ann Intern Med. 1992;116:6–12.
Gottdiener J, Reda D, Massie BM, Materson BJ, Williams DW, Anderson RJ. Effect of single-drug therapy on reduction of left ventricular mass in mild to moderate hypertension: comparison of six antihypertensive agents: the Department of Veterans Affairs Cooperative Study Group on Antihypertensive Agents. Circulation. 1997;95:2007–2014.
Vaziri S, Lauer M, Larson M, Benjamin E, Levy D. Influence of blood pressure on left atrial size. Hypertension. 1995;25:1155–1160.
Materson BJ, Reda DJ, Cushman WC, Massie BM, Freis ED, Kochar MS, Hamburger RJ, Fye C, Lakshman R, Gottdiener J. Single-drug therapy for hypertension in men: a comparison of six antihypertensive agents with placebo: the Department of Veterans Affairs Cooperative Study Group on Antihypertensive Agents. N Engl J Med. 1993;328:914–921.
Sahn DJ, DeMaria A, Kisslo J, Weyman A. The committee on M-mode standardization of the American Society of Echocardiography: recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation. 1978;58:1072–1083.
Troy BL, Pombo J, Rackley CE. Measurement of left ventricular wall thickness and mass by echocardiography. Circulation. 1972;45:602–611.
Devereux RB. Detection of left ventricular hypertrophy by M-mode echocardiography: anatomic validation, standardization, and comparison to other methods. Hypertension. 1987;9(suppl II):II-19–II-26.
Diggle P, Liang K, Zeger S. Analysis of Longitudinal Data. Oxford, UK: Clarendon Press; 1994.
SAS/STAT User’s Guide, Version 6. 4th ed. Cary, NC: SAS Institute; 1989.
Ghali JK, Liao Y, Simmons B, Castaner A, Cao G, Cooper RS. The prognostic role of left ventricular hypertrophy in patents with or without coronary artery disease. Ann Intern Med. 1992;117:831–836.
Benjamin EJ, D’Agostino RB, Belanger AJ, Wolf PA, Levy D. Left atrial size and the risk of stroke and death: the Framingham Heart Study. Circulation. 1995;92:835–841.