Triamterene

Salbutamol (Albuterol) and diuretics are commonly prescribed together in patients with airflow obstruction and are associated with electrocardiographic effects. We have now investigated whether the use of potassium-sparing drugs might prevent the ECG sequelae of such combined therapy. Ten healthy subjects received seven days of randomized treatments with: placebo, bendrofluazide (5 mg), bendrofluazide plus triamterene 50 mg (conventional dose), or triamterene 200 mg (high dose), and bendrofluazide plus spironolactone (100 mg). Potassium and ECG responses to inhaled salbutamol, 2 mg, were measured after each treatment period. The T-wave flattening in response to bendrofluazide and salbutamol (0.24[Cl, 0.19 to 0.29]mV) was attenuated by the addition of triamterene, 200 mg (0.33[CI, O.28 to 0.37]mV; p<0.05) and spironolactone 100 mg (0.42[Cl, 0.37 to 0.47]mV; p<0.01), but not by triamterene 50 mg (0.25[CI, 0.20 to 0.30]mV). Spironolactone and high dose triamterene also diminished the frequency of U waves and ST depression. The ECG effects mirrored hypokalemic responses which were also blunted by high dose (p<0.01) but not low dose triamterene, as well as by spironolactone (p<0.001). Thus, the use of high dose triamterene and spironolactone protected against the hypokalemic and ECG sequelae of combined beta-agonist/diuretic therapy, whereas a conventional dose of triamterene had no effect. These findings may be important in the prevention of a potentially dangerous interaction in susceptible patients taking this combination of drugs.

Beta-adrenoceptor agonists and diuretics are commonly used drugs which are often prescribed together in patients with airflow obstruction. Inhaled beta-agonists are known to produce hypokalemia which is mediated by stimulation of [beta.sub.-2] adrenoceptor-linked membrane-bound sodium/potassium ATP-'ase, resulting in intracellular influx of potassium into muscle cells.[1] The use of high dose nebulized beta-agonists may be associated with cardiac arrhythmias in patients with chronic airflow obstruction,[2] and has also been implicated in some cases of sudden death in asthma.[3] In a case-control study from New Zealand, the use of fenoterol by metered-dose inhaler was also associated with an increased risk of death in severe asthma.[4] We have previously shown[5] that prior treatment with the potassium-losing diuretic bendrofluazide (5 mg) potentiates the hypokalemic and ECG effects of high-dose (2 mg) inhaled salbutamol. This combination may result in a potentially dangerous interaction, particularly in patients with coexisting ischemic heart disease, in whom hypokalemia may predispose to the development of ventricular arrhythmias.[6,7] Hypokalemia is also an important predictor of mortality in patients who have heart failure.[8] Furthermore, supplementation of bendrofluazide with potassium and magnesium does not alter these responses.[9] The purpose of the present study was to assess whether the hypokalemic and ECG responses to bendrofluazide and salbutamol are attenuated by the addition of triamterene or spironolactone, both of which are known to have potassium-sparing properties.

Methods

Patients

After approval of the local ethics committee, written informed consent was obtained from all subjects. Ten healthy subjects were studied with a mean (+ or -] SEM) age of 28 [+ or -] 2 years (range 23 to 41 years). A 12-lead ECG showed no abnormalities, and levels of blood electrolytes and creatinine were normal.

Study Design

The following oral treatments (daily doses) were given for seven days using a randomized (Latin-square) single-blind cross-over design: (a) bendrofluazide 5 mg; (b) bendrofluazide plus triamterene 50 mg; (c) bendrofluazide plus triamterene 200 mg; (d) bendrofluazide plus spironolactone 100 mg; (e) placebo. A washout period of seven days was used between treatments. After the end of each treatment period, subjects attended the laboratory between 9 and 10 AM having taken their tablets 3 hours previously. The subjects were then given two doses (each of 1 mg) of inhaled salbutamol.

Subjects lay supine for the duration of the study. A cannula was inserted into an antecubital vein and was kept patent by bolus injections of heparinized saline solution. A run-in period of 30 minutes was used to establish a true baseline, with which to compare successive responses to salbutamol. Two blood samples were withdrawn without a tourniquet at 15-minute intervals during the run-in period, for assay of plasma potassium and magnesium, and the mean of these values was used for the purposes of analysis. The blood was immediately centrifuged, separated, and stored at - 20 [degrees] C. A single lead ECC was recorded when the heart rate had settled to its lowest plateau level. Blood pressure was measured at 30-second intervals until four consistent readings were obtained, and the mean was used as a baseline.

Following the run-in period, two doses of inhaled salbutamol (each 1 mg) were given at 30-minute intervals. Measurements were then made 25 minutes after each dose of salbutamol. Inhaled salbutamol was delivered by a 750 ml pear-shaped spacer device using the method described by Gleason and Price.[10] The spacer device was used so as to eliminate individual dfferences in inhaler technique. Specially prepared high dose pressurized aerosol cannisters were used delivering 0.5 mg salbutamol per actuation. Blood samples for the measurement of potassium and magnesium were subsequently withdrawn, and the mean of four consistent blood pressure recordings was taken. An ECG trace (lead 2) was recorded for a period of ten beats when the heart rate had settled (on the monitor) to its lowest level. The following parameters were measured from the mean of five complexes: RR interval, QT interval, T wave amplitude, U wave amplitude, and ST segment depression. The QT interval was measured using the method described by Schamroth,[11] to account for the presence of U waves. The Bazett formula[12] was used to correct the QT interval for changes in heart rate (Q-Tc). Heart rate was calculated from the RR interval.

Measurements

The ECG (lead 2) was recorded using a monitor and printer, set at 50 mm/s paper speed and 0.5 mV/cm gain. Systolic and diastolic blood pressure were recorded with a semiautomated sphygmomanometer. All biochemical analyses were performed in batches at the end of the study and were assayed in duplicate. Plasma potassium was measured by flame photometry. The within and between assay values for analytical imprecision ([CV.sub.w], and [CV.sub.B]) were 0.45 percent and 0.58 percent, respectively Plasma magnesium was assayed by spectrophotometry ([CV.sub.w] 1.35 percent; [CV.sub.B] 1.80 percent). Normal reference ranges for our laboratory were potassium, 3.5 to 5.0 mmol/L and magnesium, 0.70 to 1. 15 mmol/L.

Statistical Analysis

The data were analyzed using a software package (Stratographics). Analysis of variance (ANOVA) was performed using all five treatments to establish whether there was a significant overall effect, and then followed by comparison of treatment pairs, also by ANOVA. Treatment groups and subjects were used as within-factors for the ANOVA. Linear regression analysis was performed using the least squares method,[13] with ANOVA to test the significance of the regression line. A significance level of less than 5 percent (two-tailed) was considered significant for all tests.

Results

The results are presented as means and 95 percent confidence intervals. The effects of salbutamol on potassium following each treatment are shown in Figure 1. Baseline potassium before salbutamol) was significantly lower after treatment with bendrofluazide (3.29[CI, 3.12 to 3.45]mmol/L) compared with placebo (3.79[CI, 3.63 to 3.95]mmol/L; p<0.001). The addition of triamterene 200 mg (3.67[Cl, 3.51 to 3.83]mmol/L; p<0.01) and spironolactone 100 mg (3.60[CI, 3.44 to 3.76]mmol/L; p<0.05), but not triamterene 50 mg (3.48[3.31-3.64]mmol/L) attenuated tbe effect of bendrofluazide on baseline potassium. Salbutamol alone also produced significant hypokalemia p<0.001). Plasma potassium fell to a lower absolute level in response to salbutamol after treatment with bendrofluazide (2.92[Cl, 2.71 to 3.12]mmol/L) compared with placebo (3.34[Cl, 3.13 to 3.55]mmol/L; p<0.05). The hypokalemia with bendrofluazide plus salbutamol was attenuated by triamterene 200 mg (3.43[Cl 3.22 to 3.64]mmol/L; p<0.01) and spironolactone (3.53[Cl, 3.22 to 3.74]mmol/L; p<0.001), but not by triamterene 50 mg (3.10[Cl, 2.90 to 3.31]mmol/L). Neither bendrofluazide nor salbutamol had any significant effect on plasma magnesium levels.

T wave responses are shown in Figure 2. Baseline T wave amplitude after bendrofluazide (0.41[Cl, 0.36 to 0.47]mV) was increased by the addition of triamterene 200 mg 0.51[Cl, 0.46 to 0.56]mV; p<0.01), and spironolactone 0.55[Cl, 0.49 to 0.60]mV; p<0.05), but not by triamterene 50 mg ( 0.43[Cl, 0.38 to 0.48]mV). Salbutamol alone also produced significant T wave flattening (p<0.001). The combination of bendrofluazide plus salbutamol resulted in a lower absolute T wave amplitude compared with salbutamol alone (0.34[Cl, 0.29 to 0.39]mV vs 0.24[Cl, 0.19 to 0.29]mV; p<0.05). The absolute level of potassium was higher after supplementation of bendrofluazide with triamterene 200 mg (0.33[Cl, 0.28 to 0.37]mV; p<0.05), and spironolactone (0.42[Cl, 0.37 to 0.47]mV; p<0.01), whereas triamterene 50 mg had no effect (0.25[Cl, 0.20 to 0.30]mV). There was a significant correlation between hypokalemic and T wave responses (r = 0.62, p<0.001).

Salbutamol produced prolongation of Q-Tc from 0.375(Cl, 0.367 to 0.384)s to 0.398(CI, 0.388 to 0.408)s (p<0.01), but was not significantly altered by any of the treatments. There was a significant correlation between potassium and QT responses (r = - 0.58, p<0.001). U waves occurred in four subjects (mean amplitude 0.2 mV) with the combination of bendrofluazide and salbutamol. The frequency of U waves was reduced to two subjects (0. 15 mV) by triamterene 200 mg, and occurred in one subject (0.05 mV) with spironolactone. The ST segment depression developed in three cases (0. 15 mV) with bendrofluazide and salbutamol, in two cases (0.08 mV) with the addition of triamterene 200 mg, and in one subject (0.05 mV) with spironolactone. Triamterene 50 mg, had no effect on either U wave or ST segment changes. The ECG changes were not associated with ventricular extrasystoles or arrhythmias. Examples of ECGs from three subjects are shown in Figure 3.

There was also a small chronotropic response to salbutamol from 62(CI, 60 to 65) beats/min to 72(CI, 69 to 75) beats/min (p<0.001). There were no significant changes in systolic or diastolic blood pressure.

Discussion

The results of the present study show that the addition of high dose triamterene (200 mg) or spironolactone (100 mg) to bendrofluazide attenuates baseline potassium and ECG effects and also protects against these changes following beta-adrenoceptor stimulation by doses of salbutamol lower than those currently recommended for nebulized therapy (2.5 to 5 mg). A conventional dose of triamterene (50 mg) had no effect on these responses. This dose of triamterene was chosen to be identical to that found in proprietary combination tablets, with thiazide or loop diureties. Dose-ranging studies have shown that the potassium sparing effect of spironolactone in response to bendrofluazide reaches a plateau at 100 mg,[14] which was the dose used for the current study Our data (Fig 1) suggest that spironolactone and high dose triamterene have a separate effect in preventing beta-agonist induced hypokalemia, as well as protecting against the kaliuretic effect of bendrofluazide. However, it remains uncertain as to whether this is a specific effect on the beta-receptor linked potassium transport mechanism, or whether it is simply as a consequence of raising total body levels of potassium. In terms of T wave effects, it would appear that spironolactone and triamterene predominantly increased baseline T wave amplitude without having much effect on beta-agonist induced T wave flattening.

Our results for baseline potassium were similar to those of Jackson et al[l5] who showed that triamterene, 200 mg, and spironolactone 50 mg, have equipotent potassium sparing activity. In another study, McKenna and co-workers[16] found that spironolactone 50 mg, but not triamterene, 100 mg, prevented chlorthalidone induced hypokalemia. It was, therefore, interesting to find that the potassium sparing effects of high dose triamterene and spironolactone were maintained even after beta-adrenoceptor stimulation, and that these effects were closely mirrored in ECG responses. This is consistent with the known association between hypokalemia and most of the ECG changes which occurred, apart from Q-Tc prolongation.[17,18]

A direct sympathomimetic effect on cardiac [beta.sub.-2] receptors,[19] may also be contributary to ECG changes. In a recent study,[20] reversible inferior T wave inversion was found in 34 percent of patients admitted with acute severe asthma. These changes were related to parameters of clinical severity, although serum potassium was not one of the factors included in the analysis. Similar T wave changes occur in asymptomatic men with normal coronary arteries, in response to infused adrenaline and atrial pacing.[21] Beta-blockade with propranolol did not prevent pacing-induced changes but did abolish responses to infused adrenaline, suggesting a dissociation between catecholamine and raterelated effects on the ECG. However, the chronotropic response to salbutamol in the present study was relatively small, and was, therefore, unlikely to be an important factor in the genesis of ECG sequelae.

Hypomagnesemia and hypocalcemia may cause prolongation of QT interval, although plasma magnesium levels were unchanged by bendrofluazide and salbutamol. Magnesium is a predominantly intracellular cation, and normal blood levels may not reflect an intracellular deficit.[22,23] Beta-agonists have been previously reported to produce hypomagnesemia and hypocalcemia,[24,25] although changes are very small and are probably within the limits of biologic variability.[26]

Diuretic therapy may be associated math a reduction of muscle potassium,[22] and depletion of total body potassium in the presence of unchanged blood levels.[27,28] The addition of triamterene (37.5 mg) to thiazide diuretic has been shown to increase intracellular (muscle) potassium and magnesium with unchanged blood levels.[29] Since the resting myocardial membrane potential is determined by the ratio of extracellular to intracellular potassium concentration (approximate ratio of 1:35), changes in extracellular potassium concentration will have a much larger effect on this ratio than comparable changes in intracellular potassium. This dissociation between intracellular and extracellular compartments may explain the failure of low dose triamterene to attenuate ECG responses in the present study. In contrast, spironolactone 100 mg, produces increased potassium levels in both blood and muscle.[30] We felt that the ECG was the most practical way of providing clinically relevant information on the effects of bendrofluazide and salbutamol, rather than subject our subjects to repeated muscle biopsies, which was considered to be unethical. Dyckner and Wester[23] have shown that supplementation with magnesium (by infusion) is required in addition to potassium, in order to restore intracellular levels of potassium in diuretic-treated patients. Furthermore, replacement of both cations is needed to reduce the rate ofventricular ectopy, both at rest[23] and in response to exercise.[31]

The extracellular potassium ion concentration is the single most important determinant of myocardial membrane stability.[32-34] Hypokalemia is known to be an important risk factor in the genesis of ventricular arrhythmias in patients with ischemic heart disease.[6,7] Patients with chronic airflow obstruction are usually smokers and commonly have coexistent ischemic heart disease. Such patients are often taking diuretics for cor pulmonale. During acute episodes of airflow obstruction, the myocardium may also be sensitized by hypoxemia, and hypokalemia may be potentiated by concomitant theophylline therapy[35] and by high adrenergic drive.[36] Furthermore, corticosteroid therapy is known to increase numbers of beta-receptors[37,38] and might, therefore, result in increased sensitivity to beta-agonists. The results of the present study suggest that the use of high dose triamterene or spironolactone may prevent a potentially dangerous interaction in susceptible patients taking the combination of diuretic and beta-agonist. Prospective studies are now indicated to assess the importance of these effects in patients with chronic airflow obstruction, especially during acute episodes of airflow obstruction.

References

[1] Lipworth BJ, McFarlane LC, Coutie WJ, McDevitt DG. Evaluation of the metabolic responses to inhaled salbutamol in the measurement of beta-2 adrenoceptor blockade. Eur J Clin Pharmacol 1989; 37:297-300 [2] Higgins RM, Cookson WOCM, Lane DG, John SM, McCarthy GL. Cardiac arrhythmias caused by nebulized beta-agonist therapy. Lancet 1987; 2:863-64 [3] Sears MR, Rea HH, Fenwick J, Gillies AJD, Holst PE, O'Donnell TV et al. 75 deaths in asthmatics prescribed home nebulizers. Br Med J 1987; 294:477-80 [4] Crane J, Flatt A, Jackson R, Ball M, Pearce N, Burgess C, et al. Prescribed fenoterol and death from asthma in New Zealand, 1981-1983: case-control study. Lancet 1989; 1:917-22 [5] Lipworth BJ, McDevitt DG, Struthers AD. Prior treatment with diuretic augments the electrocardiographic effects of inhaled albuterol. Am J Med 1989; 86:653-57 [6] Nordrehaug JE, Johannessen KE, Von Der lappe G. Serum potassium concentration as a risk factor of ventricular arrhythmias early in acute myocardial infarction. Circulation 1985; 71:645-49 [7] Stewart DE, lkram H, Espiner E, Nicholls GM. Arrhythmogenic potential of diuretic induced hypokalemia in patients with mild hypertension and ischemic heart disease. Br Heart J 1985; 54:290-97 [8] Dargie HJ, Cleland JGF, Leckie BJ, Inglis CG, East BW, Ford I. Relation of arrhythmias and electrolyte abnormalities to survival in patients with severe chronic heart failure. Circulation 1987; 75(suppl 4):98-107 [9] Lipworth BJ, Medevitt DG, Struthers AD. Electrocardiographic changes induced by inhaled salbutamol following treatment with bendrofluazide: effects of replacement therapy with potassium, magnesium and triamterene. Clin Sci 1990; 78:255-59 [10] Gleason JGA, Price JF. Nebuhaler technique. Br J Dis Chest 1988; 82:172-74 [11] Schamroth L. The Q-T interval. In: An introduction to electrocardiography. Oxford: Blackwell Scientific Publications, 1982: 141-44 [12] Bazett HC. An analysis of the time relations of electrocardiograms. Heart 1920; 7:353-70 [13] Draper NR, Smith H, eds. Fitting a straight line by least squares. In: Applied regression analysis. New York: John Wiley, 1981:1-69 [14] Ramsay LE, Hettiarachchi J, Fraser R, Morton JJ. Amiloride, spironolactone and potassium chloride in thiazide treated hypertensive patients. Clin Pharmacol Ther 1980; 27:53343 [15] Jackson PR, Ramsay LE, Wakefield V Relative potency of spironolactone, triamterene and potassium chloride in thiazide-induced hypokalemia. Br J Clin Pharmacol 1982; 14:257-63 [16] McKenna TJ, Donohoe JF, Brien TG, Healy JJ, Canning B St J, Muldowney DB. Potassium sparing agents during diuretic therapy in hypertension. Br Med J 1977; 2:734-41 [17] Weaver WF, Burchell H. Serum potassium and the electrocardiograph and hypokalemia. Circulation 1960; 21:505-21 [18] Surawicz B. Relation between electrocardiogram and electrolytes. Am Heart J 1967; 73:814-34 [19] Hall JA, Petch MC, Brown MJ. Intracoronary injections of salbutamol demonstrate the presence of functional beta-2 adrenoceptors in the human heart. Circ Res 1989; 65:546-53 [20] Hassan AB, Efthimou J, Ormerod O, Benson MK. Reversible electrocardiographic T wave abnormalities in severe acute asthma. Thorax 1989; 44:879-80 [21] Taggart P Donaldson R, Green J, et al. Interrelation of heart rate and autonomic activity in asymptomatic men with unobstructed coronary arteries: studies with atrial pacing, adrenaline infusion, and autonomic blockade. Br Heart J 1982; 47:19-25 [22] Dorup 1, Skajaa K, Clausen T, Kjeldsen K. Reduced concentrations of potassium, magnesium and sodium-potassium pumps in human skeletal muscle during treatment with diuretics. Br Med J 1988; 296:455-58 [23] Dyckner T, Wester PO. Ventricular extrasystoles and intracellular clectrolytes before and after potassium and magnesium infusions in patients on diuretic treatment. Am Heart J 1979; 97:12-17 [24] Whyte KF, Addis GJ, Whitesmith R, Reid JL. Adrenergic control of plasma magnesium in man. Clin Sci 1987; 72:135-38 [25] Bos WJW, Postma DS, Van Doormaal JJ. Magnesiuric and calciuric effects of terbutaline in man. Clin Sci 1988; 74:595-97 [26] Lipworth BJ, McDevitt DC. Beta-adrenoceptor responses to inhaled salbutamol in normal subjects. Eur J Clin Pharmacol 1989; 36:239-45 [27] Edmonds CJ, Jasani B. Total body potassium in hypertensive patients during prolonged diuretic therapy. Lancet 1972; 1:8-12 [28] Wilkinson PR, Issler H, Hesp R, Raftery EB. Total body and serum potassium during prolonged thiazide therapy for essential hypertension. Lancet 1975; 2:759-62 [29] Widmann L, Dyckner T, Wester PO. Effects of triamterene on senim and skeletal muscle electrolytes in diuretic-treated patients. Eur J Clin Pharmacol 1988; 33:577-79 [30] Dyckner T, Wester PO, Widman L. Effects of spironolactone or serum and muscle electrolytes in patients on long-term diuretic therapy for congestive heart failure and/or arterial hypertension. Eur J Clin Pharmacol 1986; 30:535-40 [31] Hollifield JW Potassium and magnesium abnormalities, diuretics and arrhythmias in hypertension. Am J Med 1984; 77:28-32 [32] Surawicz B, Lepeschin E, Hefflich HC, Hoffinan BF. Effect of potassium and calcium deficiency on the monophasic action potential, electrocardiogram and contractility of isolated rabbit hearts. Am J Physiol 1959; 196:1302-07 [33] Dangman KJ, Danilo P Jr, Hordof AJ, Mary-Rabine L, Reder RF, Rosen MR. Electrophysiologic characteristics of human ventricular and Purkinje fibres. Circulation 1982; 65:362-68 [34] Roden DM, Iansmith DHS. Effects of low potassium or magnesium concentrations on isolated cardiac muscle. Am J Med 1987; 82 (suppl 3):18-23 [35] Whyte KF, Reid C, Addis GJ, Whitesmith R, Reid JL. Salbutamol induced hypokalemia: the effect of theophyrine alone and in combination with adrenaline. Br J Clin Pharmacol 1988; 25:571-78 [36] Brown MJ, Brown DC, Murphy MB. Hypokalemia from beta-2 receptor stimulation by circulating epinephrine. N Engl J Med 1983; 309:1414-19 [37] Parker CW, Huber M, Baumann ML. Alterations in cyclic AMP metabolism in human bronchial asthma: III. Leukocyte and lymphocyte responses to steroids. J Clin Invest 1973; 52:1342-48 [38] Brodde OE, Brinkmann M, Schemuth R, O'Hara N, Daul A. Terbutaline induced desensitisation of human lymphocyte beta-adrenoceptors: 2 accelerated restoration of beta-adrenecoceptor responses by prednisolone and ketotifen. J Clin Invest 1985; 76:1096-1101

COPYRIGHT 1990 American College of Chest Physicians
COPYRIGHT 2004 Gale Group