Xanthine derivatives and beta – agonists are commonly used together in the treatment of bronchial obstruction as their effects are additive. In addition to being potent bronchodilators, these drugs also have a variety of extrapulmonary effects including the potential to cause arrhythmias. Indeed, although this is not universally accepted, combination bron-chodilator therapy has been suggested as a possible cause of unexpected sudden death in asthma, by induction of arrhythmias. In order to study the possible arrhythmogenic effects of combination therapy, we decided to compare monotherapy with betag-agonists with two combination regimens, beta- agonists plus theophylline and beta2-agonists plus enprofylline. Enprofylline is a xanthine derivative about four to five times more potent as a bronchodilator than theophylline. Theophylline but not enprofylline is an adenosine antagonist, and a role for adenosine as an antiadrenergic and antiarrhythmic agent in the heart has been suggested. Enprofylline plus beta£-agonists could therefore be a less arrhythmogenic combination than theophylline and betaa-ago-nists.
Patients and Methods
Twenty patients, and women and nine men, gave their written informed consent to participate in this double-blind, placebo-controlled study. The study protocol was approved by the Committee of Medical Ethics, University of Lund. Clinical data of the patients before study entry are given in Table 1. No patient had a history, treatment, physical signs or findings on an ECG-recording suggesting clinically relevant ischemic heart disease. Two patients, one man aged 62 and one woman aged 58, had mild hypertension easily controlled with 2.5 mg bendroflumethiazide and 100 mg spironolactone daily, respectively. Join Canadian Neighbor Pharmacy and you will be able to read latest news, order drugs, have the right for discount.
The study was performed over seven consecutive weeks, one run-in week followed by three, two-week treatment periods. During the run-in week, orally administered theophylline was withdrawn, but no other changes were made to individual patients’ drug regimens.
The study drug, ie, placebo (P), enprofylline (E, plain tablets 300 mg), and theophylline (X Theo-Dur, 300 mg) were given in random order, each for two weeks, according to a balanced Latin square design. The initial dose was 150 mg three times daily for three days and thereafter, 300 mg three times daily for the rest of the period. In the second week of each period, Holter monitoring was performed over 72 consecutive hours. Spirometry was performed on three occasions (in the mornings of the “Holter-daysbefore drug intake), and two blood samples drawn for analysis of plasma xanthine concentrations, one sample in the morning before drug intake and one sample about four hours later.
Self-assessment of lung (unction (peak expiratory flow rate, PEFR 1/min; mini-Wight peak flow meter) was performed twice daily (morning and evening) at home in each period. The xanthines were analyzed by a HPLC method according to Edholm (Draco reports 844-RD0077 and 844-RD0119). Adverse effects were noted by the patients on a diary card.
Recording and Analysis of Cardiac Arrhythmias
During each period of treatment, three consecutive 24-hour ambulatory ECG-recordings were made. One ECG lead was recorded using the Oxford MR 10 recorder and a bipolar, presternal lead configuration was used in most cases in order to minimize muscle noise and motion artifacts. When the presternal position produced very low amplitude QRS complexes, one of the electrodes was moved to a left precordial position. The recordings were played back on a tape deck and the signal digitized using a sampling rate of 100 Hz and 8 bits digital resolution, stored on a computer disk memory and then analyzed for heart rate and arrhythmias by a computer system developed at our hospital. The computer report of arrhythmias and a “full-disclosure” printout of the ECG (6.25 mm/s, 5 min 20 s of ECG on a paper of size DIN A4) provided a basis for visual quantification of arrhythmias, which was made on an hour-by-hour basis. Episodes of particular interest were printed out at 12.5 mm/s for more detailed study. The following classes of arrhythmias were reported for each hour:
1. Single premature ventricular beats (PVBs)
2. Ventricular runs (VR, defined as two or more consecutive PVBs)
3. Ventricular bigeminy
4. Single premature supraventricular beats (PSVBs)
5. Supraventricular runs (SVR, defined as two or more consecutive PSVBs)
6. Supraventricular bigeminy
For runs (classes 2 and 5), the number of episodes and the number of beats in the longest episode were counted. For episodes of bigeminy (classes 3 and 6), the number of episodes, the largest number of ectopics in the episodes, and the total number of ectopics in bigeminy were counted.
A recorded hour was only considered eligible for further statistical analysis if the ECG was of such good technical quality that it was readable without ambiguity for at least 50 minutes.
Statistical Model and Analysis
Nonparametric tests were used to compare the effects of the different regimens. The number of arrhythmias during one hour was assumed to follow a positively skewed discrete probability distribution and no parametric assumptions about the shape of the distributions were made. Although observations from the distributions were assumed to be statistically independent, it seems plausible, however, that a certain amount of dependence will occur between the number of arrhythmias at adjacent hours.
As it was considered important to detect an increase of arrhythmias, a nonparametric test with sufficient power to detect location-shifts was required and a Wilcoxon rank-sum test was considered to be appropriate. If tied observations occurred, midranks were used.
When no significant difference was established using the Wilcoxon test, a chi-square test was performed to allow the detection of other possible deviations from the null-hypothesis of equal distributions.
Comparisons of the distributions were also made for each patient separately, and since the tests were statistically independent, an overall significance level could be calculated. Thus, if each test is performed on the 0.01 level, the overall level for 20 patients will be less than 0.05.
The FEVj and PEFR values obtained with placebo, enprofylline, and theophylline were compared by an analysis of variance and Students t-test for paired differences.
Table 1—Patient Characteristics
|No of patients included||20|
|No of women||11|
|No of men||9|
|Age (Yr; mean, range)||51, 31-76|
|Duration of obstructive|
|lung disease (Yr; mean, range)||19, 1-40|
|Therapy of obstructive lung disease before study entry|
|(a) Oral beta*-agonists||18|
|(b) Inhaled beta^agonists||2|
|(a + b)||16|
|(c) Oral theophylline||12|
|(d) Inhaled steroids||11|
|(e) Oral steroids||1|
|(f) Inhaled DSCG||1|
|(g) Inhaled anticholinergics||1|