Laboratory tests as predictors of the antihypertensive effects of amlodipine, bisoprolol, hydrochlorothiazide and losartan in men: results from the randomized, double-blind, crossover GENRES Study.

OBJECTIVE: Individual blood pressure responses to antihypertensive therapy are difficult to predict. To improve optimization of antihypertensive therapy, we analyzed correlations of relevant laboratory tests with blood pressure responses to four antihypertensive monotherapies. METHODS: In the GENRES study, 208 Finnish men aged 35-60 years with moderate hypertension used amlodipine 5 mg, bisoprolol 5 mg, hydrochlorothiazide 25 mg and losartan 50 mg daily, each for 4 weeks as a monotherapy in a double-blind, randomized, placebo-controlled crossover study; that is, each subject received each type of monotherapy in a random order. The treatment periods were preceded and separated by 4-week placebo periods. Ambulatory 24-h and office blood pressure measurements were carried out after all study periods. Data from several biochemical tests were correlated to antihypertensive drug responses. RESULTS: Serum total calcium concentration was negatively correlated with blood pressure responses to amlodipine (P values 0.001-0.002). Plasma renin activity was positively correlated with blood pressure responses to losartan (P values 0.001-0.005) and bisoprolol (P values 0.03-0.17), and negatively with blood pressure responses to hydrochlorothiazide (P values 0.01-0.07). Daily urinary excretion of sodium was negatively correlated with ambulatory blood pressure responses to amlodipine (P values 0.001-0.01). CONCLUSIONS: In this carefully controlled study, marked individual variations in antihypertensive drug responsiveness were found to correlate to several baseline laboratory parameters. The negative correlation between serum calcium levels and amlodipine responses is intriguing and suggests an underlying mechanistic association. Collectively, our data imply that laboratory tests may have some value in prediction of the efficacy of various antihypertensive drug therapies, although great patient-to-patient variation remains an obstacle for exact predictive classification.

Predictors of antihypertensive drug responses: initial data from a placebo-controlled, randomized, cross-over study with four antihypertensive drugs (The GENRES Study).

BACKGROUND: Only a minority of hypertensive individuals is adequately controlled for their hypertension, partially because reliable predictors for efficient antihypertensive drug therapy are lacking. METHODS: In a prospective, randomized, double-blind, cross-over, placebo-controlled study (The GENRES Study), 208 moderately hypertensive Finnish men (aged 35 to 60 years) were treated for 4 weeks with antihypertensive drugs from four different classes: amlodipine (5 mg), bisoprolol (5 mg), hydrochlorothiazide (25 mg), or losartan (50 mg) daily. Each individual received each of the four monotherapies in a randomized order. Four-week placebo periods were included before and between drug treatment periods. Antihypertensive responses were assessed with 24-h ambulatory and office measurements and analyzed according to age, body mass index, triceps skin fold thickness, waist-to-hip ratio, duration of hypertension, number of previous antihypertensive drugs, number of affected parents, and blood pressure (BP) levels, and profiles during placebo periods. RESULTS: The median BP responses in 24-h ambulatory recordings (systolic/diastolic) were 11/8 mm Hg for bisoprolol, 9/6 mm Hg for losartan, 7/5 mm Hg for amlodipine, and 5/2 mm Hg for hydrochlorothiazide. The highest pairwise within-subject correlations in BP responses were seen for the combinations of bisoprolol-losartan and amlodipine-hydrochlorothiazide. The BP responses to bisoprolol and losartan did not vary according to the variables. Amlodipine and hydrochlorothiazide responses were positively correlated with age, placebo BP level, and lower night-time dipping on placebo. CONCLUSIONS: Baseline clinical and BP parameters may be used to predict the efficacy of antihypertensive therapies. The GENRES Study material should provide an excellent platform for future pharmacogenetic analyses of antihypertensive drug responsiveness.

Beta1-adrenergic receptor polymorphisms, QTc interval and occurrence of symptoms in type 1 of long QT syndrome.

BACKGROUND: The most prevalent LQT1 form of inherited long QT syndrome is caused by mutations of the KCNQ1 gene resulting repolarizing I(Ks) potassium current to decrease and the QT interval to prolong. As abrupt sympathetic activation triggers ventricular arrhythmias that may cause syncopal attacks and sudden death in LQT1 patients, we investigated whether two known beta1-adrenergic receptor polymorphisms were associated with the duration of QT interval or history of symptoms in LQT1. METHODS: We determined beta1-adrenergic receptor polymorphisms (Ser49Gly and Arg389Gly) in 168 LQT1 patients. We also reviewed each patient's clinical records on the history of long QT syndrome-related symptoms and measured QT intervals from baseline ECG in each subject and from an exercise test ECG in 55 LQT1 patients. RESULTS: Patients with the homozygous Arg389Arg genotype tended to have shorter and those with the Ser49Ser genotype longer QT intervals than patients with other genotypes, but neither polymorphism studied alone affected the risk of symptoms. In contrast, adjusted odds ratio for the history of symptoms was 4.9 (95% CI 1.18 to 20.3) in patients homozygous for both Ser49 and Arg389. These double homozygous patients showed similar QT intervals as the rest of the LQT1 cohort. CONCLUSIONS: In this relatively small study, double homozygosity for Arg389 and Ser49 of the human beta1-adrenergic receptor associated with the risk of symptoms in LQT1. The association between these beta1-adrenergic receptor polymorphisms and the symptom history in LQT1 is not mediated via QT interval duration.

Effects of epinephrine on right ventricular monophasic action potentials in the LQT1 versus LQT2 form of long QT syndrome: preferential enhancement of "triangulation" in LQT1.

AIMS: To explore effects of epinephrine and phenylephrine on the behavior of right ventricular monophasic action potentials (MAPs) in symptomatic LQT1 and LQT2 patients. METHODS AND RESULTS: We recorded endocardial MAPs from right interventricular septum at baseline and during epinephrine and phenylephrine infusions in six symptomatic DNA-verified LQT1 (QTc 528 +/- 83) and five LQT2 patients (QTc 527 +/- 72) and in five control patients (QTc 381 +/- 22). We measured MAP durations at 90% and at 50% levels of repolarization and their difference (MAP50 to MAP90, a measure of MAP morphologic "triangulation"), during atrial pacing to characterize rate dependence of MAPs and repolarization phase 3 durations, respectively. Restitution kinetics were determined during atrioventricular sequential pacing, using the approach of empirical restitution rate. Epinephrine prolonged MAP50-to-MAP90 duration and increased the rate dependence of MAP90 duration and increased restitution rate in type LQT1, but not in LQT2 patients nor in control subjects. Phenylephrine did not change MAP behavior. During epinephrine administration, both LQT1 and LQT2 patients had a ratio of the restitution rate of MAP to diastolic interval >1.0 at short diastolic intervals. CONCLUSION: Symptomatic LQT1 patients with prolonged baseline QTc intervals showed beta-adrenergic-induced changes in MAPs (triangulation) known to be arrhythmogenic, thus giving insight to the difference in clinical triggers of life-threatening arrhythmias between LQT1- and LQT2-affected individuals.

Functional characterization of the common amino acid 897 polymorphism of the cardiac potassium channel KCNH2 (HERG).

OBJECTIVE: To determine whether the amino acid 897 threonine (T) to lysine (K) polymorphism of the KCNH2 (HERG) potassium channel influences channel performance or patient phenotype. METHODS: The phenotypic effects of this polymorphism were investigated in vitro by electrophysiological experiments in HEK-293 cells and in vivo by exercise electrocardiography in a group of LQTS patients carrying the same genetically proven KCNQ1 mutation. RESULTS: When expressed in HEK-293 cells, the 897T isoform of the KCNH2 channel exhibited changes in inactivation and deactivation properties, and a smaller current density than the more common 897K isoform. Western blot experiments indicated that the decreased current density associated with 897T was caused by reduced channel expression. During a maximal exercise test in 39 LQT1 patients carrying an identical KCNQ1 mutation (G589D) and showing a prolonged QT interval (>440 ms), QT intervals were longer in patients carrying the 897T allele than in those homozygous for the 897K allele. CONCLUSIONS: The K897T variation has an effect on channel function and clinical phenotype. Our data warrant further investigations into the significance of this polymorphism in drug-induced and inherited LQTS.