Eur J Clin Pharmacol (1995) 48 : 229-233 © Springer-Verlag 1995

L. Gerrard • N. M . W h e e l d o n • D. G. M e D e v i t t

Effect of combined atenolol and nifedipine administration on psychomotor performance in normal subjects

Received: 9 May 1994/Accepted in revised form: 28 November 1994

Abstract The aim of the present study was to evaluate the central effects of single doses of the fi-adrenocep- tor antagonist atenolol and the calcium antagonist nifedipine retard, alone and in combination, in normal subjects. Twelve normal males received single oral doses of atenolol 100 rag, nifedipine retard 20 mg, atenolol 100 mg and nifedipine retard 20 mg in combination, diazepam 5 mg (active control), and each of two match- ing placebos in a double-blind, randomised fashion. Psychomotor performance was assessed using digit symbol substitution, letter cancellation (LCT), contin- uous attention, choice reaction time, finger tapping, immediate recall and short-term memory. Two flash fusion and critical flicker fusion thresholds were mea- sured and subjective assessments made using visual analogue scales (VAS). Diazepam 5 mg significantly worsened LCT scores at 4h, significantly impaired alert- ness at 2 h and 4 h, and tended to increase reaction time and impair continuous attention and physiologi- cal measurements. Atenolol 100 mg alone significantly reduced alertness at 2 h and 4 h, and also tended to impair physiological measurements. Nifedipine retard 20 mg produced no significant psychomotor effects. Combined atenolol and nifedipine retard administra- tion produced a small but significant improvement in continuous attention and a reduction in body sway, with no adverse effects being evident on performance or subjective awareness. The results suggest that no significant adverse effects on psychomotor performance are produced by single doses of atenolol 100 mg and nifedipine retard 20 mg when given together in normal

This paper was presented and accepted for publication in abstract form at the British Pharmacological Society, University of Bradford, July 1993.

L. Gerrard. N. M. Wheeldon([]) 1 - D. G. McDevitt University Department of Clinical Pharmacology, Ninewells Hospital & Medical School, Dundee DD1 9SY, Scotland, UK

Present address: 1Cardiothoracic Unit, Northern General Hospital, Sheffield $5 7AU, UK

subjects. The combination may therefore be useful in the treatment of hypertensive patients requiring dual therapy, and in whom adverse central effects are of par- ticular importance.

Key words Atenolol, Nifedipine; psychomotor perfor- mance, diazepam, combination therapy

The adverse effects of antihypertensive agents are of particular clinical importance, since the majority of untreated hypertensive patients are asymptomatic and, in general, control of raised blood pressure does not lead to an improvement in wellbeing. Indeed, drug effects on psychomotor performance may impair the ability to perform work and leisure activities, and lead to poor compliance with treatment. We have previously shown that the fi-adrenoceptor antagonist atenolol and the calcium antagonist nifedipine, conventional first- line agents for the treatment of mild hypertension, each produce small but significant adverse effects on psy- chomotor performance when used as single agents [1-3]. However, approximately half of all hypertensive patients require two or more drugs to control their blood pressure [4], and combination therapy allows rel- atively small doses of each agent to be used, with the aim of reducing the frequency and severity of adverse effects. We have previously demonstrated that dual administration of the fi-blocker atenolol and the thi- azide diuretic bendrofluazide produced significant adverse central effects, which were not evident with each drug used alone [5]. Whether similar interactions occur between other combinations of antihypertensive agents is not known. Calcium antagonists and fi-blockers have been shown to produce a synergistic effect on elevated blood pressure [6-8]. In addition, fi-adrenoceptor blockade has been found to reduce the vasodilatory side-effects of nifedipine [9] and less impairment of physical performance has been shown to occur with a calcium antagonist/fi-blocker combination than with a

230

fl-blocker alone [10] . This combination is commonly employed in clinical practice, although the central effects of such administration have not been previously investigated.

vidually on a monitor at a rate of 1 per second. A 'critical stimu- lus' of two letters were displayed at the top of the screen and sub- jects were required to make a response within 750 ms by pressing a key whenever the second letter followed the first in the series of presentations. The test lasted for 15 min and the number of cor- rect responses and errors was recorded.

Subjects and methods

Subjects

Twelve healthy male volunteers, aged between 20 and 30 years, were studied. All were required to have had a normal cardiorespiratory examination, full blood count, biochemical profile, 12-lead elec- trocardiogram and urinalysis prior to inclusion in the study. Alcohol was prohibited for at least 24 hours previously and no nicotine or caffeine consumption was allowed on study days. Subjects were not taking any concurrent medication. All volunteers provided written informed consent for the study, which was approved by the Tayside Medical Ethics Committee.

Protocol

Prior to entry into the study proper, each subject attended the lab- oratory on 2 separate days to become familiar with the schedule. They were trained to reach a plateau in their level of performance on each test before the effects of treatments were studied. Thereafter subjects attended at the same time of day on each of 6 study days separated by washout periods of at least 1 week. Subjective assess- ments and performance (except body sway) were tested in individ- ual sound-proofed cubicles, and the physiological measurements occurred in a room with controlled lighting to permit dark adap- tation. Each battery of psychometric tests lasted approximately 1 hour, and all measurements other than those of memory were made for 1 hour prior to drug ingestion, and during the periods 1.5-2.5 h and 3.5-4.5 h after ingestion. Subjects received single oral doses of atenolol 100 rag, nifedipine retard 20 mg, atenolol 100 mg and nifedipine retard 20 mg in combination, diazepam 5 mg (as an active control) and each of two matching placebos. Treatments were administered in a randomised, double-blind, crossover design, with the exception that one placebo was given in each half of the study (as a measure of potential practice effects).

Resting heart rate and systolic and diastolic blood pressures were recorded during each test session.

Measurements

Choice reaction time. Subjects were required to press a key corre- sponding to one of four light emitting diodes, which were illumi- nated in a rapid, random sequence of 30 presentations. The reaction time for the last 20 responses was recorded, and the mean calculated.

Finger tapping. Subjects tapped a pressure-sensitive transducer as rapidly as possible for 1 rain. The total number of taps and the number of involuntary rest pauses (IRP) exceeding 200 ms were recorded.

Short-term memory. Before drug ingestion subjects examined 12 photographs of miscellaneous objects over a 1 rain period. Four hours after drug ingestion they were allowed 45 s to recall and write down as many of the objects as possible.

Immediate recallmemory. Two hours after drug ingestion, 16 unre- lated words (two syllable nouns) were generated by microcomputer and were displayed individually on a monitor at the rate of one word every 3 s. Immediately afterwards subjects were allowed 45 s to write down the words recalled.

Body sway. Subjects were monitored standing erect and motionless using a Wright Codac ataxiameter to which the subject was con- nected by a 48 cm thread from the waist. Measurements were made of body sway in the anterioposterior direction, both with the sub- jects eyes open (1 rain) and closed (1 rain).

Physiological measurements

Critical flicker fitsion (CFF). The flicker fusion threshold was obtained using a central flickering source viewed monocularly at a distance of 60 cm through an artifical pupil. The flicker frequency commenced at 15-20 Hz and depending on whether the subject per- ceived the source as flickering or fused, the frequency was increased or decreased in 4 Hz steps. Subsequently, the flicker fusion fre- quency was finely tuned by steps of 0.25 Hz until the threshold fre- quency was determined. Each stimulus was presented for 2 s and the fusion threshold was defined as the mean of the last 20 pre- sentations varying within 0.25 Hz.

Performance tests

Six letter cancellation test (LCT). One of a series of 50 sheets con- taining 1200 randomised letters arranged in 40 columns was pre- sented to each subject and six target letters were identified. Subjects were allowed 5 rain to cancel as many target letters as possible and the numbers attempted, and those correctly and incorrectly can- celled were recorded. Different sheets were used on each study day.

Digit symbol substitution test (DSST). Subjectswerepresentedwith one of a series of 30 sheets (different on each day) containing 400 randomised digits and a code relating each digit to a symbol. Four minutes was allowed for subjects to substitute symbols for each digit, and the totai number of correct substitutions made was recorded.

Continuous attention (TESTAX). A random series of letters gen- erated by a BBC microcomputer was presented to subjects indi-

Two flash fusion (2FF). Via an artifical pupil, subjects were pre- sented with pairs of 10 ms flashes of light, separated by a time period ranging between 12 and 120 ms. Initially, flashes were sep- arated by 50 ms and their separation increased or decreased by 8 ms depending on the subjects response. Subjects were asked to report whether the light sources appeared fused or as two separate flashes. The step-size was reduced until steps of 1 ms were reached, and the flash fusion point was defined as the mean of the last 20 responses varying within 1 ms separation.

Subjective assessments

Visual analogue scales (VAS). Subjects reported their mood and feelings of well-being on a series of twelve, 100 mm ungraded, lin- ear visual analogue scales. Scores were determined by the distance from the subject's record to the end of the scale representing pos- itive subjective feelings. The assessments were: A: I am extremely

sleepy (0) - extremely wide awake (100); B: I am, extremely tense (0) - absolutely relaxed (100); C: I am, extremely agitated (0) - absolutely calm (100); D: I am, extremely lethargic (0) - extremely energetic (100); E: I am, mentally very dulled (0) extremely alert (100); F: I have, no ability to concentrate (0) - complete ability to concentrate (100); G: with regard to carrying out general duties I feel that I am, absolutely useless (0) - extremely efficient (100); H: I am, extremely irritable (0) not at all irritable (100); I: I am, extremely aggressive (0) - extremely passive (100); J: I feel, extremely withdrawn (0) - extremely sociable (100); K: I am, in the depths of depression (0) - ecstatically happy (100); L: I feel, extremely anx- ious (0) - absolutely carefree (100).

Blood pressure

At time 0, and at 2 and 4 h post-ingestion, resting heart rate and systolic (SBP) and diastolic (DBP) blood pressure recordings were made using a Dinamap semi-automatic sphygmomanometer (Critikon, F1., USA). The mean of three consistent recordings was used for analysis.

Statistical analysis

Data were analysed by analysis of variance for repeated measures using the SPSS-PL software package (Chicago, Ill., USA). Prior to assessing drug response, consideration was given to possible phase, practice and placebo effects. Phase effects were investigated by com- parison of pretreatment data from each study day, and practice effects by comparing the results from the first and second placebos. Placebo effects were assessed by comparing pre and post-ingestion data for the placebo treatment days and subsequent analysis of drug effects was adjusted to allow for the mean placebo response. Treatment effects were calculated as change from baseline and repeated measures ANOVA was performed on the difference between drug and placebo responses at 2 and 4 h post-ingestion. For the visual analogue scales, principal component analysis was

231

used to reduce the number of variables required to explain drug effects. Seven of the 12 subjective assessments reflected mood (B, C, H, I, J, K, L) and 5 related to the level of alertness (A, D, E, F, a) .

Results (Tables I to 3)

P h a s e effect

C o m p a r i s o n s o f p r e - i n g e s t i o n resul ts for each va r i ab le on each o f the t r e a t m e n t days showed no s ign i f ican t differences, a n d hence no ev idence o f a p h a s e effect was a p p a r e n t across the six s tudy sessions.

C o m p a r i s o n o f p l a c e b o t r e a t m e n t s

U s i n g A N O V A the effects o f the 1 st a n d 2nd p l a c e b o s were c o m p a r e d at 2 a n d 4 h pos t - i nge s t i on . N o s ign i f ican t differences o c c u r r e d b e t w e e n the effect o f the two p l a c e b o s o n any o f the m e a s u r e d var iables .

P l a c e b o effects

G i v e n t ha t the two p l a c e b o s d id n o t differ s ign i f ican t ly in the i r effects, the m e a n o f the two sets o f resul ts was used for analysis . A t 2 h there was a dec rease (P < 0.05) in the first p r i n c i p a l c o m p o n e n t o f the VAS (scales A , D , E , F a n d G). A t 4 h T E S T A X p e r f o r m a n c e was s ign i f ican t ly i m p a i r e d , w i th an inc rease in the t o t a l n u m b e r o f e r rors ( P < 0.001). In view o f these

Table 1 Effects of treatments on performance testing and physiological variables (compared to placebo). Performance tests

4 hours post-ingestion Diazepam 5 mg Atenolol 100 mg Nifedipine retard 20 mg At + NIF

LCT (% correct) -5.43 ( - 9 . 3 1 - - 1.54)** -5.48 DSST score 2.88 (-3.50-9.25) 7.46 Continuous attention 4.04 (-2.21-10.29) 0.38 (errors) Mean reaction time(s) 0.49 (-0.09 1.06) 0.27 Finger taps 5.92 ( - 5.70-17.53) 0.17 Involuntary rest pauses -6.33 ( 13.04-0.37) 1.58 Body sway (eyes open) -0.04 (-3.49-3.40) -0.74 CFF threshold (Hz) -7.80 (-26.19-10.6) -8.46 2FF threshold (s) -6.99 ( 25.39-11.41) -6.17

-11.31-0.36) -3.85 (-8.51-0.81) -2.92 - 1.11-16.02) 2.29 ( - 7. I8-11.76) 2.46 -2.4~3.17) 0.29 ( -2 .5 3.08) -1.21

-0.18 0.72) 0.44 (-0.11-0.99) 0.35 -14.05 14.38) 4.75 (-17.59 27.09) 0.75 -5.28-8.45) -0.67 (-10.20-8.86) -2.42 -3.35-1.87) -0.37 (-2.44-1.70) -2 .62 -20.09 3.17) 1.38 (-3.16-5.92) -0.96 -20.84-8.51) 10.33 (-23.38-2.73) 1.43

- 7 . 1 4 - 1 . 2 9 )

-4.27-9.19) -2.34-0.08)*

-0.29-0.99) - 11.15-9.65) - 10.48-5.64) - 5.02-0.23)* - 3.29 1.37) -4.44 7.31)

* P < 0.02 AT+NIF, combined atenolol 100 mg and nifedipine retard 20 mg *~ P< 0.01

Table 2 Subjective assessment of treatment effects, shown by Scores in visual analogue scales

Diazepam 5 mg Atenolol 100 mg Nifedipine retard 20 mg AT + NIF

2 hours post-ingestion VAS (first component) 1.25 (0.49-2.0)*** 0.49 (0.14-0.84)** 0.44 (-0.18 1.06) 0.44 ( -0 .16 1.03)

4 hours post-ingestion VAS (first component) 1.0 (0.24-1.75)** 0.37 (0.02 0.73)* 0.41 (-0.22-1.03) 0.41 (-0.19-1.0)

AT + NIF, Combined atenolol 100 mg nifedipine retard 20 rag. Mean changes (95% CI) from baseline compared to placebo *P < 0.05; **P < 0.02; ***P < 0.01

232

Table 3 Haemodynamic effects of treatment

SBP DBP HR Time (h) 2 4 2 4 2 4

Placebo (mean) - 3.58* 2.83 -2 .17 - 3.88* - 4.75* - 0.67 ( - 7 . 1 0 - - 0 . 0 6 ) ( -6 .35 0.69) (-5.52-1.19) ( - 7 . 2 3 - - 0 . 5 3 ) ( - 8 . 9 8 - - 0 . 5 2 ) ( -4 .90 3.56)

Diazepam 5 mg 2.50 3.00 - 1.08 3.04 -2 .50 -0 .33 (-7.58-2.58) (-2.08-8.08) (-5.30-3.14) ( - l .18-7 .26) ( -8 .38 3.38) (-6.22-5.55)

Atenolol 100 mg - 12.67' - 13.42" -3 .00 - 6 . 7 1 ' -8.67** - 11.75'* ( 2 2 . 6 6 - - 2 . 6 7 ) ( - 2 3 . 4 1 - - 3 . 4 2 ) (-9.13-3.13) ( -12 .84 -0 .58) ( - 1 3 . 8 1 - - 3 . 5 2 ) ( - 1 6 . 8 % - 6 . 6 1 )

Nifedipine retard 20 mg -0 .92 -5 .17 -2 .42 0.54 11.42 6.58** (-6.21-4.38) (-10.46-0.13) (-6.92-2.08) (-3.96-5.04) -(5.61-17.23) (0.77-12.39)

AT + NIF -11 .17 '* -13.58"* -7.75** -6.63* -6.75* -13.08"* ( - 1 8 . 9 2 - - 3 . 4 1 ) ( - 2 1 . 3 4 - - 5 . 8 3 ) ( - 1 3 . 4 9 - - 2 . 0 1 ) ( -12 .36 -0.89) ( - 1 3 . 2 6 - - 0 . 2 4 ) ( - 1 9 . 6 - - 6 . 5 7 )

AT + NIF, Combined atenoloi 100 mg and nifedipine retard 20 mg Mean (95% CI) absolute values *P < 0.05; **P < 0.01

significant effects, all drug effects were adjusted by com- parison with the mean results obtained from the two placebos.

Diazepam effects

Diazepam produced a significant impairment in LCT scores at 4 h (P < 0.02) and significantly worsened scor- ing in the VAS scales representing alertness (A,D,E,F and G), at 2 h (P<0.01) and at 4 h (P<0.02). Diazepam tended to increase reaction time and to impair continuous attention and physiological measurements.

Atenolol effects

At both 2 h (P < 0.02) and 4 h (P < 0.05) atenolol pro- duced a significant worsening in the VAS scales repre- senting alertness (A,D,E,F and G) and also tended to impair physiological measurements.

Nifedipine effects

Nifedipine retard produced no significant adverse effects on psychomotor performance or subjective awareness at 2 or 4 h post-ingestion.

Combined atenolol and nifedipine effects

The combination produced a small but significant reduction in body sway (P < 0.05) and an improvement in continuous attention (P < 0.05), with no significant adverse effects being found on psychomotor perfor- mance or subjective awareness.

Memory testing

No significant effects were seen on immediate recall or short-term memory by any of the treatments used.

Blood pressure and heart rate responses

At 2 h and 4 h post-ingestion atenolol 100 mg pro- duced a significant (P < 0.05) fall in SBP compared to placebo. Heart rate was also decreased at both time points (P < 0.01), whereas DBP was only reduced at 4 h (P < 0.05). Nifedipine retard 20 mg increased heart rate at 4 h (P < 0.05), although it had no significant effect on blood pressure.

The combination of atenolol 100 mg and nifedipine retard 20 mg reduced heart rate at 2 h (P < 0.05) and at 4 h post-ingestion (P < 0.01). SBP was reduced at both time points (P < 0.01), as was DBP at 2 h (P < 0.01) and at 4 h (P < 0.05).

Atenolol effects vs combined atenolol & nifedipine effects

At 2 h no significant difference was found between the two drug regimens. At 4 h there was a small but significant increase (P < 0.05) in the number of TES- TAX scores missed after atenolol compared to the com- bination.

Nifedipine effects vs combined atenolol & nifedipine effects

At both 2 h and 4 h there were no significant differences between the two regimes.

Discussion

The results of the present study have not demonstrated any evidence of an adverse interactive effect of com- bined atenolol and nifedipine administration on psy- chomotor performance and subjective awareness, when compared both with placebo and with the individual drugs given alone. Indeed, compared to placebo

233

significant improvements in continuous attention and body sway occurred, and the adverse effects on sub- jective awareness found with atenolol alone were no longer evident when the two agents were combined.

Practice and phase effects were excluded, since there were no significant differences on comparing the responses to each of the two placebos, or between the pretreatment baselines on each study day. This sug- gested that subjects had been adequately trained to reach a plateau in their level of performance before the effects of the treatments were investigated. Significant placebo effects did occur on subjective awareness and continuous attention, and the drug effects were adjusted to account for this. The active control (diazepam 5 rag) produce significant impairment of LCT performance and significantly worsened the first principal compo- nent of the VAS representing alertness, thus establish- ing the sensitivity of the system.

Atenolol 100 mg alone significantly reduced alert- ness, and tended to impair physiological measurements, but had no effect on performance tests. This is in agree- ment with our previous findings [1,5]. We have also pre- viously demonst ra ted that this dose of atenolol produces EEG changes consistent with sedation [2]. Nifedipine retard 20 mg used alone produced no significant effects on performance, physiological mea- surements and subjective awareness, which is also in keeping with our previously published observations [3]. Significant improvements in continuous attention and body sway occurred using the two agents in combina- tion, with no evidence of any adverse central effects.

Is it possible that effects on systemic blood pressure were involved in these responses? This is possible, since atenolol alone produced a significantly greater fall in BP than nifedipine retard alone. However, the combi- nation produced an equivalent hypotensive effect to atenolol alone in the absence of adverse central effects, suggesting that this was not the primary mechanism involved.

Although no measurable psychomotor effects were found after nifedipine retard alone, heart rate was significantly increased. This is a well-recognised effect of nifedipine [3], and suggests reflex activation of the sympathetic nervous system due to drug-induced peripheral vasodilatation. The response was abolished in the presence of concomitant/~-adrenoceptor block- ade. Empirically, the 'anxiolytic' effects of/~-blockers and the 'anxiogenic' effect of nifedipine may offset each other to some extent, with the result that the drug com- bination produces little measureable change in psy- chomotor performance.

We accept that studies of this kind have certain lim- itations, mainly relating to the number of subjects it is practicable to study. In addition, we recognise that sin- gle-dose studies in normal young volunteers may not

necessarily reflect effects in older hypertensive patients on chronic therapy. However, the study was very care- fully performed in a double-blind, randomised fashion, and it had the strength of including both double placebo and an active control. We do accept, however, that the effects of diazepam 5 mg in this study were not as great as one might have expected.

The results suggest that no significant adverse effects on psychomotor performance are produced by single doses of atenolol 100 mg and nifedipine retard 20 mg when given together to normal subjects. In this respect, the combination may be useful in the treatment of hypertensive patients requiring dual therapy, and in whom adverse central effects are of particular impor- tance, e.g. the elderly or patients involved in skilled activity. Further research in these patient groups is, be necessary to substantiate these findings.

Acknowledgements The. authors acknowledge the assistance of Mrs. C. Hau in the statistical analysis and Mrs. J. Thomson in typing the manuscript.

References

I. Currie D, Lewis RV, McDevitt DG, Nicholson AN, Wright NA (1988) Central effects of [3-adrenoceptor antagonists. I- Performance and subjective assessments of mood. Br J Clin Pharmacol 26:121-128

2. Nicholson AN, Wright NA, Zetlin MB, Currie D, McDevitt DG (1988) Central effects of 13-adrenoceptor antagonists II-elec- troencephalogram and body sway. Br J Clin Pharmaco126:1-13

3. McDevitt DG, Currie D, Nicholson AN, Wright NA, Zetlin MB (1991) Central effects of the calcium antagonist, nifedip- inc. Br J Clin Pharmacol 32:541-549

4. Rahn KH (1990) Principles in the combination of antihyper- tensive drugs. In: Pharmacology of antihypertensive therapeu- tics. Ganten D, Mulrow PJ (eds) Springer, Berlin Heidelberg, New York, pp 677-685

5. Gerrard L, Wheeldon NM, McDevitt DG (1994) Central effects of combined bendroftuazide and atenolol administration. Eur J Clin Pharmacol 45:539-543

6. Aoki K, Kondo S, Mochizaki A, Yoshida T, Kato S, Kato K, Takikawa K (1978) Antihypertensive effect of cardiovascular calcium antagonists in hypertensive patients in the absence and presence of beta-adrenergic blockade. Am Heart J 96:218 226

7. Franz IW, Wiewel D (1984) Antihypertensive effects on blood pressure at rest and during exercise of calcium antagonists, 13- receptor blockers, and their combination in hypertensive patients. J Cardiovasc Pharmacol 6:S1037-I042

8. Kazda S, Knorr A (1990) Calcium antagonists. In: Ganten D, Mulrow PJ (eds) Pharmacology of antihypertensive therapeu- tics. Springer, Berlin Heidelberg, New York, pp 301-375

9. Husted SE, Kraemmer NH, Christensen CK, Lederballe- Pederson O (1982) Long-term therapy of arterial hypertension with nifedipine given alone or in combination with a beta- adrenoceptor blocking agent. Eur J Clin Pharmaco122: 101 103

10. Kindermann W, Widmann W, Reider TH, Kullmer TH (1987) Physical fitness during antihypertensive treatment: results of combined treatment with a calcium antagonist and beta blocker. Fortschr Med 105 : 75-81