Ejd;01jun99

Psychopharmacology (1999) 143 : 286 –292 P. Bitsios · E. Szabadi · C.M. Bradshaw
Comparison of the effects of venlafaxine, paroxetine and desipramineon the pupillary light reflex in man Received : 29 July 1998 / Final version : 17 November 1998 Abstract Rationale : The time-course of the pupillary
effects. Conclusions : The increase in resting pupil diam- light reflex response is determined by the successive eter could be indicative of parasympathetic inhibition activation of the parasympathetic and sympathetic and / or sympathetic activation. The shortening of the innervations of the iris, the latency and the amplitude recovery time of the light reflex response is consistent reflecting parasympathetic and the recovery time with sympathetic potentiation resulting from nora- mainly sympathetic activity. Objective : To compare the drenaline uptake blockade in the iris. The prolonga- effects of single doses of three antidepressants (ven- tion of the latency and decrease of the amplitude of lafaxine : serotonin / noradrenaline reuptake inhibitor, the light reflex response are indicative of a parasym- paroxetine : selective serotonin reuptake inhibitor, and patholytic effect of venlafaxine. However, as venlafax- desipramine : tricyclic antidepressant) on resting pupil ine has negligible affinity for muscarinic cholinoceptors, diameter and the pupillary light reflex response.
this effect cannot be attributed to the blockade of Methods : Fifteen healthy male volunteers participated cholinoceptors in the iris. A possible explanation for in five weekly sessions, each of which was associated this finding is that it reflects a central rather than a with one treatment (venlafaxine 75 mg or 150 mg, peripheral effect of the drug : the blockade of nora- paroxetine 20 mg, desipramine 100 mg, or placebo) drenaline uptake in the brain could lead to the poten- according to a double-blind, double-dummy, balanced, tiation of the noradrenergic inhibition of central cross-over design. An infrared binocular television parasympathetic (Edinger-Westphal) neurones. These pupillometer was used for the recording of the resting results demonstrate the ability of therapeutically rele- pupil diameter and the pupillary light reflex in dark- vant single doses of venlafaxine to potentiate nora- ness, in previously dark-adapted eyes. Resting pupil drenergic responses in man, consistent with the diameter in darkness was recorded before and after treatment. The pupillary light reflex was elicited aftertreatment, with six light flashes (green, 565 nm peak Key words Light reflex · Pupil · Venlafaxine ·
wavelength) of 200 ms duration and of incremental Paroxetine · Desipramine · Human volunteers illuminance (measured in the plane of the cornea) :3.0 × 10–3, 8.5 × 10–3, 2.5 × 10–2, 7.0 × 10–2, 0.18,0.43 mW cm92. The parameters studied were : latency,amplitude and 75 % recovery time. Results : Analyses of variance followed by post hoc tests (least significantdifference test or Dunnett’s test; P < 0.05) revealed that Venlafaxine is an antidepressant with serotonin and both doses of venlafaxine produced a significant noradrenaline uptake inhibiting properties (SNRI) increase in resting pupil diameter, decrease in ampli- with a weak effect on dopamine uptake (Muth et al.
tude and shortening of the 75 % recovery time of the 1986; Bolden-Watson and Richelson 1993). Venla- light reflex response; venlafaxine 150 mg prolonged the faxine does not inhibit monoamine oxidase A or B latency, while the other treatments had no significant (Muth et al. 1986) and does not have anticholinergic,antiadrenergic, antiserotonergic or antihistaminergicproperties (Preskorn 1994). It is believed that at P. Bitsios · E. Szabadi (*) · C.M. Bradshaw lower doses venlafaxine acts mainly as a serotonin re- Department of Psychiatry, University of Nottingham, uptake inhibitor, whereas at higher doses, it exerts an Floor A, South Block, Queen’s Medical Centre,Nottingham NG7 2UH, UK additional noradrenergic re-uptake inhibition (Muth et al. 1986; Richelson 1994). The latter has been thought to account for venlafaxine’s increased efficacyin severely depressed patients when higher doses are used (DeMontigny and Preskorn 1995; Preskorn 1995).
There is little evidence, however, that venlafaxine Fifteen healthy male volunteers aged 20–28 years (mean ± SEM inhibits noradrenaline re-uptake at clinically used doses 22.00 ± 1.0) and weighing 57–105 kg (mean ± SEM 76.0 ±5.8) par- in humans, or that noradrenergic uptake blockade ticipated in the study. Subjects were all medication-free and wererequested to stop smoking and to avoid drinking alcohol, coffee occurs mainly at the upper limits of the recommended and other caffeine-containing beverages for at least 12 h before the dosage range. In a recent single dose study, venlafax- experimental session. All of them were using tobacco and caffeine ine 150 mg, but not venlafaxine 75 mg, potentiated the occasionally and all were occasional social alcohol consumers. They venoconstrictor response to noradrenaline in healthy were all tested in the morning hours (9 : 00 a.m.–14 : 00 p.m.). Thestudy protocol was approved by the University of Nottingham volunteers, consistent with the postulated noradrener- Medical School Ethics Committee. All volunteers gave their writ- gic re-uptake inhibiting effect of the drug at higher ten consent following a verbal explanation of the study and after dosage levels (Abdelmawla et al. 1997a).
reading a detailed information sheet.
The human pupil is a suitable, non-invasive system to test noradrenergic responses in humans, in vivo.
Mydriasis evoked by systemically administered drugs may be the result of increased sympathetic or reduced Venlafaxine 75 mg and 150 mg, paroxetine 20 mg, desipramine parasympathetic influence on the iris, or both, and 100 mg, and placebo were administered orally in matching capsules.
drug-induced miosis may be the result of decreased In each session the subjects ingested one capsule on two occasions : sympathetic or increased parasympathetic influence one of the capsules contained the active drug and the other the on the iris, or both. The pupillary light reflex response placebo, except in the placebo session when both capsules containedplacebo. The first capsule, containing desipramine, paroxetine or (see Fig. 1.) may help to elucidate the effects of a drug placebo was ingested 180 min and the second capsule, containing on the sympathetic and parasympathetic inputs to venlafaxine 75 mg, venlafaxine 150 mg or placebo, 100 min priorthe iris, since the time-course of the light reflex response is determined by the successive activation of theparasympathetic and sympathetic inputs, latency andamplitude reflecting parasympathetic activation, and recovery time reflecting mainly sympathetic activation(Loewenfeld 1993).
Subjects participated in five weekly sessions. Subjects were allocated The aim of this study was to provide evidence for to drugs and sessions according to a double-blind, balanced, cross- the noradrenaline re-uptake inhibiting property of ven- over design. A double-dummy procedure (see above, Drugs) was lafaxine in healthy volunteers using the pupillary light adopted in order to account for the different absorption kinetics ofthe active drugs. It has been reported that the peak concentration reflex as a test system. We used desipramine, a tricyclic after a single oral dose of desipramine (Sallee and Pollock 1990) antidepressant with potent noradrenaline re-uptake and paroxetine (Kaye et al. 1990) is attained approximately 3 h after inhibiting properties but little action on serotonin re- ingestion, whereas the peak concentration is obtained approxi- uptake (Richelson 1994), and paroxetine, a selective mately 2 h after the ingestion of venlafaxine (Kamerus et al. 1992).
serotonin re-uptake inhibitor with little effect on nora-drenaline re-uptake (Richelson and Nelson 1984), as controls. Some of these results have been communi-cated to the British Association for Psychopharma- An infrared binocular television pupillometer (TVP 1015B Applied Science Laboratories, Waltham, Mass., USA) was used for the Fig. 1 Example of a light
reflex response. Ordinate :
pupil diameter (mm), abscissa :
running time (s). Horizontal
bar : light stimulus; 1 initial
pupil diameter; 2 75 %
recovery; 3 pupil diameter at
maximal constriction; A onset
of light stimulus; B onset of
response; C time of maximal
constriction; D time at which
75 % recovery is attained;
1–3 amplitude; AB latency;
CD 75 % recovery time
recording of resting pupil diameter and the pupillary light reflex indarkness, in previously dark-adapted eyes. The experimental ses-sion started with three 45-s recordings of resting pupil diameter,the average of which served as the pre-treatment baseline restingpupil diameter. Three hours after ingestion of the first capsule(100 min after ingestion of the second capsule), recordings of rest-ing pupil diameter were repeated and then the pupillary light reflexresponse was studied. The stimuli were six light flashes (green,565 nm peak wavelength) of 200 ms duration, and of incrementalilluminance (measured in the plane of the cornea) : 3.0 × 1093, 8.5 × 1093, 2.5 × 1092, 7.0 × 1092, 0.18, 0.43 mW cm91. The lightflashes were delivered at 25 s intervals, via a light emitting diodepositioned 1 cm from the cornea of the subject’s right eye. Therecordings took place in a dark, sound-attenuated room and thesubjects fixed their gaze on a dim red spot of light positioned approx-imately 2.5 m in front of them. Stimulus presentation was controlledby a microcomputer, and pupillary measures were digitized andstored on a floppy disk for off-line analysis. The parameters stud-ied were : latency (i.e. time elapsing from the onset of the stimulusto the onset of the response, s), amplitude of light reflex response(i.e. the difference between the initial and the minimal diameters ofa pupillary response to a light flash, mm) and 75 % recovery time(i.e. time taken from the peak of the response to obtain 75 % recov-ery, s) (see Fig. 1). Each light reflex response was visually inspected.
If an eye-blink occurred either during the presentation of the stim-ulus, reducing the amount of light reaching the retina, or at thepeak of pupillary constriction, rendering the amplitude immea-sureable, the response in question was not included in the analysis.
Fig. 2 Change in resting pupil diameter in darkness (mm) from pre-
treatment baseline, in the presence of the five treatments. P :
placebo, Px 20 : paroxetine 20 mg, V75 : venlafaxine 75 mg, V150 :
venlafaxine 150 mg, DMI100 : desipramine 100 mg. The heights of
the columns correspond to the means obtained in the group of 15subjects; vertical bars are SEM. Asterisks denote statistical Data obtained from the left pupil were analyzed. The post / pre- treatment difference was calculated for resting pupil diameter indarkness for each subject and for the group of 15 subjects. One-way analysis of variance with repeated measures with treatment as Table 1 Effects of treatments on pupil diameter (mm) : differences
the within-subject factor followed by post hoc tests (least significant difference test) were used to compare the effects of treatment onresting pupil diameter in darkness. Separate two-way analyses of variance with repeated measures, with treatment (five levels) and light intensity (six levels) as the within-subject factors, were used toanalyze each light reflex measure (i.e. latency, amplitude, 75 % recov- ery time). In the case of a significant effect of treatment, multiple comparisons between placebo and the four active treatments were undertaken using Dunnett’s test (df = 56, k = 5, criterion P < 0.05).
The relationship between response amplitude and recovery time wasanalyzed using the product moment correlation coefficient, and bestfit linear functions were derived with the method of least squares.
lafaxine 75 mg and more so with venlafaxine 150 mg.
Unpaired t-test was used to compare the slope values for each active Analysis of variance of the latency data revealed significant main effects of treatment (F = 5.19; df: 4,56,P < 0.001) and light intensity (F = 83.5; df: 5,70;P < 0.001) but no significant interaction (F < 1).
Comparisons between placebo and the four active treat-ments using Dunnett’s test showed that only the effect The effects of the treatments on resting pupil diameter of venlafaxine 150 mg was significant (t = 3.70). in darkness are shown in Fig. 2 and Table 1. It can be It can be seen that amplitude was smaller with ven- seen that all treatments caused an increase in resting lafaxine 75 mg than under the placebo condition, and pupil diameter. One-way analysis of variance showed more so with venlafaxine 150 mg. Analysis of variance that the effect of treatment was significant (F = 4.2; of the amplitude data revealed significant main effects df: 4,56; P < 0.005). Post hoc comparisons with the of treatment (F = 21.8; df: 4,56; P < 0.001) and light least significant difference test showed that only the intensity (F = 339.6; df: 5,70; P < 0.001) but no increases caused by 75 and 150 mg of venlafaxine were significant interaction (F = 1; df: 20,280; P > 0.1).
Comparisons between placebo and the four active treat- The latency, amplitude and 75 % recovery time of the ments using Dunnett’s test showed that the effects of light reflex response (group means) are displayed in Fig.
venlafaxine 75 and 150 mg were significant (t = 5.46 3. It can be seen that latency was prolonged with ven- Fig. 3 Parameters of the light
reflex response obtained at six
graded illuminance levels,
measured in the plane of the
cornea, in the presence of the
five treatments. Open circles :
placebo; open triangles :
paroxetine; open squares :
desipramine; closed triangles :
venlafaxine 75 mg; closed
squares : venlafaxine 150 mg
It can be seen that recovery time was shorter with both venlafaxine treatments. Analysis of variance ofthe recovery time data revealed significant main effectsof treatment (F = 4.84; df: 4,56; P < 0.002) and lightintensity (F = 24.9; df: 5,70; P < 0.001) but nosignificant interaction (F < 1). Comparisons betweenplacebo and the four active treatments using Dunnett’stest showed that the effects of venlafaxine 75 and150 mg were significant (t = 2.79 and 3.45, respec-tively). The possibility that the changes in light reflex recov- ery time were secondary to changes in amplitude wasalso explored. The relationship between reflex responseamplitude and 75 % recovery time, at each light inten-sity value studied, is shown in Fig. 4; the results of thelinear regression analysis are displayed in Table 2. Itcan be seen from Fig. 4 that the regression linesobtained after placebo and paroxetine did not deviatefrom each other, whereas the regression lines obtainedafter desipramine and venlafaxine deviated from thatobtained after placebo. Statistical comparison of theslopes of the regression lines obtained in the presence Fig. 4 Relationship between the amplitude and 75 % recovery time
of the antidepressants and placebo showed that the of the light reflex responses evoked by the range of light stimulus slopes obtained after the two doses of venlafaxine intensities shown in Fig. 3. Open circles : placebo; closed circles : significantly differed from that obtained after placebo active treatment; the dotted lines were fitted with linear regression (unpaired t-test : venlafaxine 75 mg versus placebo analysis; for statistical analysis see text and Table 1 t = 3.09, df: 10, P < 0.02; venlafaxine 150 mg versusplacebo t = 3.71, df: 10, P < 0.01). There were no sta- size (Loewenfeld 1993). Tricyclic antidepressants have tistically significant differences between the slopes variable effects on resting pupil diameter, depending on obtained after paroxetine and placebo (t = 0.34, df: 10, the balance between their ability to block noradrena- P > 0.1) or desipramine and placebo (t = 0.86, df: 10, line uptake / post-junctional muscarinic cholinoceptors, effects which would tend to increase pupil diameter,and their ability to block post-junctional α1-adreno-ceptors, an effect which would tend to decrease pupil diameter (Szabadi and Bradshaw 1986). The tricyclicantidepressant used in the present experiment, Resting pupil diameter reflects the balance between the desipramine, is a potent inhibitor of noradrenaline opposing sympathetic and parasympathetic innerva- reuptake and has relatively low affinities for muscarinic tions of the iris, sympathetic activation increasing and cholinoceptors and α1-adrenoceptors (Richelson parasympathetic activation decreasing resting pupil 1994). The predicted effect of desipramine on resting Table 2 Results of linear
regression analysis (leastsquares, product moment *Slope significantly different from placebo condition, P < 0.02 (see text) pupil size would be mydriasis, reflecting mainly nora- mydriasis caused by venlafaxine was due to the block- drenaline reuptake. In the present experiment, a single ade of 5HT (re)uptake. Furthermore, the selective nora- dose (100 mg) of desipramine failed to have a drenaline reuptake inhibitor (NARI) reboxetine, which significant effect on pupil diameter, consistent with a has virtually no affinity for 5HT uptake (Brunello and previous observation in our laboratory (Theofilopoulos Racagni 1998), also causes mydriasis (Szabadi et al.
et al. 1995). It should be noted, however, that in some 1998), supporting the hypothesis that the mydriatric other studies a mydriatic effect of desipramine could effect of venlafaxine is due to the blockade of nora- be observed (Szabadi et al. 1980, 1998; Shur and Checkley 1982; Kerr and Szabadi 1985). It seems, there- It is well known that there is a correlation between fore, that the mydriasis observed after the administra- the level of physiological arousal and pupil diameter, tion of desipramine is not a consistent finding. The sedation being accompanied by miosis and activation basis for the inconsistency may lie in the fact that by mydriasis (Loewenfeld 1993). In this respect, it is of desipramine also blocks α1-adrenoceptors, and the rela- interest that venlafaxine has been reported to have a tionship between the mydriasis resulting from nora- desipramine-like activating effect on the EEG, espe- drenaline reuptake blockade and the miosis from cially at a dosage of 50 mg and above (Saletu et al.
α1-adrenoceptor blockade may differ between different 1992; Patat et al. 1998). Furthermore, venlafaxineexperiments.
shows some alerting effect in both psychophysiological Both single doses (75 and 150 mg) of venlafaxine tests (e.g. critical flicker fusion frequency) and subjec- caused an increase in dark-adapted resting pupil diam- tive ratings of level of alertness (Saletu et al. 1998). It eter. This observation is in agreement with an earlier is likely that the alerting effect of venlafaxine is due to report demonstrating the mydriatic effect of smaller sin- central noradrenergic activation (Foote and Aston- gle doses (12.5, 25 and 50 mg) of venlafaxine (Saletu Jones 1995) resulting from the blockade of noradren- et al. 1992). As venlafaxine has practically no affinity aline uptake in the brain. Interestingly, central for muscarinic cholinoceptors (Preskorn 1994), but has noradrenergic activation may also contribute to the the ability to block noradrenaline uptake (Muth et al.
mydriasis by enhancing the tonic noradrenergic inhi- 1986; Richelson 1994), the most likely explanation for bition of the Edinger-Westphal nucleus (see below).
venlafaxine-evoked mydriasis is that it reflects the Venlafaxine also had distinctive effects on the kinetic blockade of noradrenaline uptake into sympathetic parameters of the light reflex response, prolonging nerve terminals in the iris, which in turn results in the the latency, reducing the amplitude and shortening potentiation of the influence of the sympathetic input the recovery time of the response. The shortening of the recovery time could have been secondary to the reduc- As venlafaxine blocks not only noradrenaline but tion in the amplitude of the reflex response, since it is also 5-hydroxytryptamine (re)uptake in the brain well documented that smaller light reflex responses take (Muth et al. 1986; Richelson 1994), it should be con- a shorter time to recover (Smith 1988; Theofilopoulos sidered whether the mydriasis evoked by venlafaxine et al. 1995). However, it is unlikely that the reduction might have been, at least partly, mediated by a sero- in amplitude can fully explain the shortening of the tonergic mechanism. Indeed, there is evidence that such recovery time in the present experiment, since the sta- mechanisms may be involved in pupillary control. Thus tistical analysis of the relationship between amplitude it has been reported that both 5-HT2 receptor antag- and recovery time showed that the two doses of ven- onists (Millson et al. 1991, 1992) and 5HT1A receptor lafaxine shortened the recovery time over and above agonists (Fanciullacci et al. 1995; Phillips et al. 1998) that predicted on the basis of a reduction in amplitude cause miosis. On the other hand, fenfluramine, a drug alone. Thus, the shortening of the recovery time seems known to release 5HT from pre-synaptic terminals, to be consistent with a genuine alteration in the auto- increases pupil diameter (Kramer et al. 1973). However, nomic components of the light reflex response. In fact, therapeutically relevant single doses of the SSRIs the effect on the recovery time is consistent with the fluvoxamine (50–100 mg : Wilson et al. 1983; Flett blockade of noradrenaline uptake in the iris, leading et al. 1992) and paroxetine (20 mg : present study), to sympathetic potentiation. There is evidence that the when compared to placebo, have no significant effect recovery time is modulated by sympathetic activity : on pupil diameter. Therefore, it is unlikely that the variables which increase sympathetic activity [e.g. heat stressor (Leung et al. 1992; Mortlock et al. 1996)] prolongation of the latency and reduction of the ampli- shorten the recovery time, whereas variables which tude of the light reflex response, and, to some extent, decrease the sympathetic influence on the iris [e.g. the the mydriasis). It is of interest that venlafaxine, like the centrally acting sympatholytic drug clonidine (Morley selective noradrenaline reuptake inhibitor reboxetine et al. 1991) and the α1-adrenoceptor antagonist pra- (Szabadi et al. 1998), displays a similar pseudo- zosin (Mortlock et al. 1996)] prolong the recovery time.
anticholinergic effect on salivary gland activity Again, this effect was not shared by the SSRI paroxe- (Abdelmawla et al. 1997b), probably resulting from the tine, consistent with previous findings with another potentiation of the central noradrenergic inhibition of SSRI, fluvoxamine (Flett et al. 1992), indicating that the salivary nuclei in the brain stem.
central 5HT reuptake blockade by venlafaxine is In the present study, both single doses (75 and unlikely to be involved in the shortening of the recov- 150 mg) of venlafaxine exerted effects consistent with ery time. Furthermore, the selective noradrenaline the blockade of noradrenaline uptake both in the reuptake inhibitor reboxetine, which has no effect on periphery and the central nervous system. It is of inter- 5HT uptake, causes a similar shortening of the recov- est that in another report (Abdelmawla et al. 1997a), ery time of the light reflex response (Theofilopoulos et in which the effects of the same doses of venlafaxine on noradrenergic responses of the dorsal hand vein The prolongation of the latency and reduction of were studied, only the higher dose potentiated the amplitude of the light reflex response, observed after response to noradrenaline. These observations indicate the administration of venlafaxine, are surprising a difference between the effects of the drug on responses findings, since these effects are generally attributed to evoked by endogenously released and exogenously parasympathetic inhibition (Smith 1988), and ven- applied noradrenaline : for potentiation of responses to lafaxine has virtually no affinity for muscarinic exogenously applied noradrenaline higher dosages of cholinoceptors (Preskorn 1994). An alternative expla- venlafaxine are required than for the potentiation of nation may be that this “pseudo-anticholinergic” effect the effects of endogenously released noradrenaline.
of venlafaxine is due to noradrenergic potentiation in Therefore, it is likely that patients treated even with the brain, again resulting from noradrenaline uptake lower dosages of venlafaxine may experience some blockade. There is evidence that the pre-ganglionic enhancement of the pharmacodynamic effects of parasympathetic cholinergic neurones in the Edinger- Westphal nucleus of the mid-brain are under tonicinhibitory noradrenergic control from the locus Acknowledgements We are grateful to Wyeth Laboratories for
financial support. P.B. was supported by a scholarship from the
coeruleus (for review, see Szabadi and Bradshaw 1996) : State Scholarship Foundation of Greece (I.K.Y.).
the blockade of noradrenaline uptake at the inhibitorynoradrenergic synapses in the Edinger-Westphalnucleus would “switch off ” the parasympathetic neu-rones, resulting in a pseudo-anticholinergic effect in the periphery. Indeed, the opposite effect, i.e. disinhibitionresulting in the enhancement of the light reflex Abdelmawla A, Langley RW, Szabadi E, Bradshaw CM (1997a) response, can be observed when the inhibitory input Comparison of venlafaxine, desipramine and paroxetine on thedorsal hand vein in man. J Psychopharmacol 11 : A28 from the locus coeruleus is “switched off ” by the α2- Abdelmawla A, Bitsios P, Langley RW, Szabadi E, Bradshaw CM adrenoceptor agonist clonidine (Szabadi and Bradshaw (1997b) Comparison of venlafaxine, desipramine and paroxe- 1996). It is of interest that the selective noradrenaline tine on autonomic functions in man. J Pychopharmacol reuptake inhibitor reboxetine, which like venlafaxine Bitsios P, Langley RW, Szabadi E, Bradshaw CM (1997) has little affinity for muscarinic cholinoceptors Comparison of venlafaxine, desipramine and paroxetine on the (Brunello and Racagni 1998), has a similar inhibitory pupillary light reflex in man. J Psychopharmacol 11 : A28 (“pseudo-anticholinergic”) effect on the light reflex Bolden-Watson C, Richelson E (1993) Blockade by newly-devel- response (Theofilopoulos et al. 1995).
oped antidepressants of biogenic amine uptake into rat brain In conclusion, the pupillary effects of single doses of Brunello N, Racagni G (1998) Rationale for the development of venlafaxine are consistent with the ability of the drug noradrenaline reuptake inhibitors. Hum Psychopharmacol to block noradrenaline reuptake and thereby potenti- ate the pharmacological effects of endogenously DeMontigny E, Preskorn SH (1995) Comparison of the tolerabil- released noradrenaline. The blockade of noradrenaline ity of bupropion, fluoxetine, imipramine, nefazadone, paroxe-tine, sertraline, and venlafaxine. J Clin Psychiatry 56 : 2–21 reuptake at the noradrenergic sympathetic effector Fanciullacci M, Sicuteri R, Alessandri M, Geppetti P (1995) junction in the iris could explain the mydriasis and the Buspirone, but not sumatriptan, induces miosis in humans : rel- shortening of the recovery time of the light reflex evance for a serotoninergic pupil control. Clin Pharmacol Ther response, whereas the blockade of noradrenaline reup- take at the inhibitory noradrenergic synapses on Flett SR, Szabadi E, Bradshaw CM (1992) A comparison of the effects of fluvoxamine and amitriptyline on autonomic Edinger-Westphal neurones may be responsible for the functions in healthy volunteers. Eur J Clin Pharmacol 42 : pseudo-anticholinergic pupillary effects of the drug (i.e.
Foote SL, Aston-Jones GS (1995) Pharmacology and physiology Phillips MA, Szabadi E, Bradshaw CM (1998) Effects of the anxi- of central noradrenergic systems. In : Bloom FE, Kupfer DJ olytics lesopitron and buspirone on the human pupil. J (eds) Psychopharmacology : the fourth generation of progress.
Preskorn SH (1994) Antidepressant drug selection : criteria and Kaye CM, Haddock PF, Langley G, Mellows TCG, Zussman TBD, Greb WH (1989) A review of the metabolism of paroxetine in Preskorn SH (1995) Clinical pharmacology of SSRIs : the basis for their optimal use. Professional Communication Caddo, Kerr FA, Szabadi E (1985) Comparison of the effects of chronic administration of ciclazindol and desipramine on pupillary Richelson E (1994) Pharmacology of antidepressants; characteris- responses to tyramine, methoxamine and pilocarpine in healthy tics of ideal drug. Mayo Clin Proc 69 : 1069–1081 volunteers. Br J Clin Pharmacol 19 : 639–647 Richelson E, Nelson A (1984) Antagonism by antidepressants of Klamerus KJ, Maloney K, Rudolph RL (1992) Introduction of neurotransmitter receptors of normal human brain in vitro.
composite parameters to the pharmacokinetics of venlafaxine and its active metabolites O-desmethyl metabolites. J Clin Saletu B, Grunberger J, Anderer P, Linzmayer L, Semlitsch HV, Magni G (1992) Pharmacodynamics of venlafaxine evaluated Kramer R, Rubicek M, Turner P (1973) The role of norfenfluramine by EEG brain mapping, psychometry and psychophysiology.
in fenfluramine-induced mydriasis. J Pharm Pharmacol 25 : Sallee FR, Pollock BG (1990) Clinical pharmacokinetics of Leung NK-C, Bradshaw CM, Szabadi E (1992) Effect of high ambi- imipramine and desipramine. Clin Pharmacokinet 18 : 181–184 ent temperature on the kinetics of the pupillary light reflex in Shur E, Checkley S (1982) Pupil studies in depressed patients : an healthy volunteers. Br J Clin Pharmacol 33 : 458–460 investigation of the mechanism of action of desipramine. Br J Loewenfeld IE (1993) The pupil. Wayne State University Press, Smith SA (1988) Pupillary function in autonomic failure. In : Millson DS, Haworth SJ, Rushton A, Wilkinson D, Hobson E, Bannister R (ed) Autonomic failure, 2nd edn. Oxford University Harry J (1991) The effects of a 5-HT2 receptor antagonist (ICI 169,369) on changes in waking EEG, pupillary responses and Szabadi E, Bradshaw CM (1986) Antidepressant drugs and the state of arousal in human volunteers. Br J Clin Pharmacol autonomic nervous system. In : Deakin JFW (ed) The biology of depression. Gaskell, London, pp 190–220 Millson DS, Jessup CL, Swaisland A, Haworth S, Rushton A, Harry Szabadi E, Bradshaw CM (1996) Autonomic pharmacology of α2- JD (1992) The effects of a selective 5-HT2 receptor antagonist adrenoceptors. J Psychopharmacol 10 : 6–18 (ICL 170,809) on platelet aggregation and pupillary responses Szabadi E, Gaszner P, Bradshaw CM (1980) The peripheral anti- in healthy volunteers. Br J Clin Pharmacol 33 : 281–288 cholinergic activity of tricyclic antidepressants : comparison of Morley MJ, Bradshaw CM, Szabadi E (1991) Effects of clonidine amitriptyline in human volunteers. Br J Psychiatry 137 : 433–439 and yohimbine on the pupillary light reflex and carbachol- Szabadi E, Bradshaw CM, Boston PF, Langley RW (1998) The evoked sweating in healthy volunteers. Br J Clin Pharmacol human pharmacology of reboxetine. Hum Psychopharmacol Mortlock S-A, Langley RW, Bradshaw CM, Szabadi E (1996) Effect Theofilopoulos N, McDade G, Szabadi E, Bradshaw CM (1995) of locally applied prazosin on the kinetics of the pupillary light Effects of reboxetine and desipramine on the kinetics of the reflex. Br J Clin Pharmacol 42 : 236–238 pupillary light reflex response. Brit J Clin Pharmacol Muth EA, Haskins JT, Moyer JA (1986) Antidepressant biochem- ical profile of the novel bicyclic compound Wy-45,030, an ethyl Wilson WH, Higano H, Papadatos Y, Kelwala S, Ban TA (1983) cyclohexanol derivative. Biochem Pharmacol 35 : 4493–4497 A double-blind placebo-controlled study to compare the auto- Patat A, Troy S, Burke J, Trocherie S, Danjou P, Le Coz F, Allain nomic effects of fluvoxamine with those of amitriptyline H, Gandon J-M (1998) Absolute bioavailability and electroen- and doxepine in healthy volunteers. Br J Clin Pharmacol cephalographic effects of conventional and extended-release for- mulations of venlafaxine in healthy subjects. J Clin Pharmacol38 : 256–267

Source: http://bitsioslab.med.uoc.gr/papers/34.pdf