Relapse Prevention in Major Depressive Disorder After Successful Acute Electroconvulsive Treatment: a 6-month Double-blind Comparison of Three Fixed Dosages of Escitalopram and a Fixed Dose of Nortriptyline - Lessons from a Failed Randomised Trial of the Danish University Antidepressant Group (DUAG-7).

INTRODUCTION: Electroconvulsive treatment (ECT) is an effective treatment for severe depression but carries a risk of relapse in the following months. METHODS: Major depressive disorder patients in a current episode attaining remission from ECT (17-item Hamilton Depression Rating Scale (HAM-D17) score≤9) received randomly escitalopram 10 mg, 20 mg, 30 mg or nortriptyline 100 mg as monotherapies and were followed for 6 months in a multicentre double-blind set-up. Primary endpoint was relapse (HAM-D17≥16). RESULTS: As inclusion rate was low the study was prematurely stopped with only 47 patients randomised (20% of the planned sample size). No statistically significant between-group differences could be detected. When all patients receiving escitalopram were compared with those receiving nortriptyline, a marginal superiority of nortriptyline was found (p=0.08). One third of patients relapsed during the study period, and one third completed. DISCUSSION: Due to small sample size, no valid efficacy inferences could be made. The outcome was poor, probably due to tapering off of non-study psychotropic drugs after randomisation; this has implications for future study designs. ClinicalTrials.gov Identifier: NCT00660062.

Acute and continuation therapy in unipolar depression: observations from the run-in phase of a maintenance trial.

OBJECTIVE: The aim of the study was to analyze treatments and outcome in depressed patients. METHOD: Patients with recurrent depressive disorder (n = 289), recruited for a prophylaxis study, were followed up in hospital settings for 6 months with diagnostic and depression ratings at baseline and monthly depression ratings. Data on psychotropic drugs were retrieved from hospital case records. Independent associations between baseline, treatment and outcome variables were examined by logistic regression models. RESULTS: Depressive symptoms subsided gradually. After 6 months, 21% had dropped out, 43% were rated as remitted (HAM-D-17 <8) and 8% had not responded (HAM-D-17 >15). Patients once remitted rarely relapsed (<5%). All patients received antidepressant drugs, half of them more than one (2-4) as well as other psychotropic drugs. Patients responding poorly received more frequently multiple antidepressants, tricyclic antidepressants, hypnosedatives, lithium and/or antipsychotics. CONCLUSION: The 6-month outcome was generally poor. Choice of treatment appeared at least partly to be determined by the therapeutic outcome.

First-pass metabolism of imipramine and desipramine: impact of the sparteine oxidation phenotype.

Four rapid extensive metabolizers (EM), four slow EM, and three poor metabolizers (PM) of sparteine were given single intravenous doses of 50 mg imipramine and desipramine. All subjects had previously taken single oral doses (100 mg) of imipramine and desipramine. The first-pass metabolism of imipramine and desipramine ranged from 23% to 73% and 0% to 48%, respectively, and was more pronounced for both drugs in EM than in poor metabolizers. The study suggested saturable 2-hydroxylation of imipramine and desipramine during the first-pass through the liver, especially in EM.

First-pass metabolism of nortriptyline in man.

The kinetics of nortriptyline were studied after oral and intravenous (iv) administration of test doses of 50 mg 14C-nortriptyline. The systemic availability of orally administered nortriptyline varied from 0.46 to 0.59 in 6 subjects. The decrease in availability was due to metabolism after administration. Systemic clearance varied from 0.31 to 0.66 L/min. From these measurements indirect estimates of the hepatic blood flow could be made, and a variation from 0.6 to 1.5 L/min was found. Quantitative measurements of first-pass metabolism could also be obtained from urinary metabolite excretion data when the kinetics of metabolite formation and elimination were taken into account. From the second or third day after the test dose, the urinary excretion rate of total radioactivity declined monoexponentially with half-lives closely corresponding to the plasma half-lives of unchanged nortriptyline. Analysis of the data from the iv test according to a 2-compartment open model showed that there was a close correlation between the rate constant of distribution from central to peripheral compartment (k12) and the elimination rate constant in the central compartment (kel). Still, there was some variation in the kel/k12-ratio, and this variation corresponded to the variation of the estimated hepatic blood flow.

First-pass metabolism of imipramine in man.

The systemic availability of orally administered imipramine (IP) varied from 29 to 77% in 4 subjects. The decrease in availability was due to an excess in metabolism after oral administration. This first-pass metabolism did not correlate with plasma half-life, apparent clearance, or the rate of metabolite excretion in urine. There was close correlation with the excess in formation of demethylated metabolites after oral administration, which suggests that the first-pass metabolism is mediated by demethylation, but does not correlate to the total rate of demethylation.

Pharmacokinetics of the selective serotonin reuptake inhibitor paroxetine: nonlinearity and relation to the sparteine oxidation polymorphism.

Steady-state plasma concentrations of paroxetine were studied at five or more paroxetine dose levels (10 to 70 mg/day) in each of 13 extensive metabolizers of sparteine and at three or four dose levels (10 to 40 mg/day) in each of three poor metabolizers of sparteine, all treated for diabetic neuropathy symptoms. On a dose of 30 mg/day there was a 25-fold variation in steady-state concentrations (25 to 670 nmol/L). The upper extreme of this variation was made up by the poor metabolizers of sparteine and the lower extreme by some fast extensive metabolizers. Further, within the extensive metabolizer group, steady-state levels showed a significant, positive correlation with sparteine metabolic ratio at all dose levels. On increasing doses, a disproportionate increase in plasma drug levels was observed in the majority of patients. In nearly all extensive metabolizers the concentration-dose data were best described by a pharmacokinetic model assuming elimination by at least two kinetically distinct processes, one a high-affinity saturable process and one a low-affinity linear process. Estimates of clearance at low drug levels of the high-affinity process showed a significant negative correlation with the sparteine metabolic ratio. Clearance of the low-affinity process was not related to the metabolic ratio and was of the same magnitude in extensive and poor metabolizers. The data thus confirmed that the metabolism of paroxetine and sparteine cosegregates and indicated that the enzyme responsible for a high-affinity saturable paroxetine elimination process is identical with CYP2D6, the source of the sparteine oxidation polymorphism.

Imipramine demethylation and hydroxylation: impact of the sparteine oxidation phenotype.

Eighteen healthy volunteers, selected according to their ability to oxidize sparteine, took single oral doses of 100 mg imipramine and desipramine. For imipramine the following clearances (L X min-1) were found in six rapid extensive metabolizers (EM), six slow EM, and six poor metabolizers (PM), respectively (mean and range): apparent oral clearance: 2.55 (1.39 to 3.47), 2.28 (1.18 to 4.26), and 1.35 (0.96 to 1.64). Clearance via demethylation was: 1.42 (0.61 to 2.01), 1.60 (0.78 to 3.81), and 1.09 (0.76 to 1.64); clearance via other pathways was: 1.13 (0.74 to 1.75), 0.69 (0.40 to 1.59), and 0.26 (0 to 0.46). For desipramine the apparent oral clearance (L X min-1) was 0.19 (0.12 to 0.24) in PM compared with 1.64 (1.46 to 1.80) and 1.03 (0.77 to 1.13) in rapid EM and slow EM. Extremely long elimination t1/2s of desipramine were seen in PM: 81 to 131 hours compared with 13 to 23 hours in EM. 2-OH-imipramine and 2-OH-desipramine were detectable in plasma of only the 12 EM, where the ratio-to-parent compound was higher in rapid EM than in slow EM. This study confirms that 2-hydroxylation of imipramine and desipramine depends almost exclusively on the sparteine oxygenase, whereas the demethylation of imipramine depends mainly on a different isozyme.

The mephenytoin oxidation polymorphism is partially responsible for the N-demethylation of imipramine.

The metabolism of imipramine in six poor metabolizers of mephenytoin was compared with the metabolism of 16 extensive metabolizers of mephenytoin from an earlier study. Each subject was given single doses of 100 mg imipramine hydrochloride and 100 mg desipramine hydrochloride on separate occasions. Imipramine demethylation clearance was 0.74 L.min-1 (mean; range, 0.31-1.24) in poor metabolizers of mephenytoin compared with 1.43 L.min-1 (mean; range, 0.61-3.81) in extensive metabolizers of mephenytoin (p = 0.01, Mann-Whitney U test). It has previously been shown that the imipramine clearance by way of other pathways and desipramine oral clearance, both largely representing 2-hydroxylation, are considerably lower in poor metabolizers of sparteine than in extensive metabolizers of sparteine. In contrast, five subjects who were poor metabolizers of mephenytoin and extensive metabolizers of sparteine and a control group of 11 subjects who were extensive metabolizers of mephenytoin and sparteine showed no statistically significant difference with regard to these parameters. One subject who was a poor metabolizer of mephenytoin and sparteine had the lowest imipramine oral clearance of all 22 subjects studied. In conclusion, this and an earlier study show that the oxidation of imipramine is mediated by means of two different polymorphic P450 isozymes, 2-hydroxylation by way of the sparteine oxygenase (P450IID6) and demethylation by way of the mephenytoin oxygenase (P450IIC8).

Single-dose kinetics of clomipramine: relationship to the sparteine and S-mephenytoin oxidation polymorphisms.

The influence of the sparteine and the S-mephenytoin oxidation polymorphisms on the kinetics of clomipramine were investigated in 25 healthy volunteers: 10 extensive metabolizers of sparteine and mephenytoin (EMs/EMm), nine poor metabolizers of sparteine and extensive metabolizers of mephenytoin (PMs/EMm), five extensive metabolizers of sparteine and poor metabolizers of mephenytoin (EMs/PMm), and one poor metabolizer of sparteine and mephenytoin (PMs/PMm). A single oral dose of 100 mg clomipramine hydrochloride was given to each subject after an overnight fast. Serum and urine levels of clomipramine and its metabolites were monitored after 1, 2, 3, 4, 6, 8, 11, 14, 24, 36, 48, and 96 hours. Additional serum was monitored after 6, 9, 12, and 15 days in the poor metabolizers. 2-Hydroxyclomipramine was undetectable in most subjects before enzymatic hydrolysis of serum and urine. The total median clearance of clomipramine was 99 L.hr-1 (range, 68 to 210) in the EMs/EMm subjects, 56 L.hr-1 (range, 37 to 183) in the PMs/EMm subjects, 66 L.hr-1 (range, 37 to 89) in the EMs/PMm subjects, and 43 L.hr-1 in the PMs/PMm subject. It was significantly lower in PMs/EMm and EMs/PMm subjects compared with EMs/EMm subjects (p = 0.006 and 0.028, respectively; Mann-Whitney). In addition, the formation clearance of 2-hydroxyclomipramine and the hydroxylation indexes were significantly lower in PMs/EMm subjects, as was the demethylation index in EMs/PMm subjects compared with EMs/EMm subjects. Our data thus provide evidence that the 2- and 8-hydroxylation of clomipramine are catalyzed by CYP2D6 and that the N-demethylation is catalyzed in part by CYP2C.

Chloroguanide metabolism in relation to the efficacy in malaria prophylaxis and the S-mephenytoin oxidation in Tanzanians.

S-Mephenytoin and chloroguanide (proguanil) oxidation was studied in 216 tanzanians. The mephenytoin S/R ratio in urine ranged from <0.1 to 1.16. The distribution was skewed to the right, without evidence of a bimodal distribution. Ten subjects (4.6%, 2.2% to 8.3%, 95% CI) with an S/R mephenytoin ratio >0.9, were arbitrarily defined as poor metabolizers of mephenytoin. The chloroguanide/cycloguanil ratio ranged from 0.82 to 249. There was a significant correlation between the mephenytoin S/R ratio and the chloroguanide/cycloguanil ratios (rs = 0.73; p<0.00001). This indicates that cytochrome P4502C19 or CYP2C19 is a major enzyme that catalyzes the bioactivation of chloroguanide to cycloguanil. Chloroguanide is a pro-drug, and hence a low CYP2C19 activity may lead to prophylactic failure caused by inadequate formation of cycloguanil. Fifty-eight women who previously took either 200 mg chloroguanide daily (n = 26) or 200 mg chloroguanide daily plus 300 mg chloroquine weekly (n = 32) in a malaria chemoprophylaxis study showed that there was significant correlation between the number of earlier breakthrough parasitemia episodes and the chloroguanide/cycloguanil ratio (rs = 0.30; p = 0.02). The breakthrough rate did not correlate with the S/R mephenytoin ratio. However, other factors, such as exposure to mosquitoes and sensitivity of the plasmodium to cycloguanil, are probably more important.

Concentration-response relationship in imipramine treatment of diabetic neuropathy symptoms.

A single-blind imipramine dose titration study was conducted in 15 diabetic patients with neuropathy symptoms. The effect of treatment was evaluated by use of visual analog scales. Imipramine doses were individually adjusted until doses yielded plasma concentrations of imipramine plus desipramine that were well above 400 nmol/L or until all neuropathy symptoms had vanished. In all except one patient, there was marked relief of symptoms. In the responding patients (n = 14), much of the effect occurred at plasma levels of imipramine plus desipramine below 100 nmol/L, but a considerable interindividual variation was observed. Concentrations above 400 to 500 nmol/L were required to ensure maximal effect in all patients, and we did not find any indication of a decreased effect at high drug levels. The dose-dependent kinetics of imipramine was confirmed, and dose increments should therefore be carried out in small steps and preferably with monitoring of drug levels.

d-Propoxyphene kinetics after single oral and intravenous doses in man.

d-Propoxyphene kinetics was studied in 8 healthy male subjects after single oral doses of d-propoxyphene at 65, 130, and 190 mg and after slow intravenous infusion of 65 mg. Total urinary excretion (7 days) indicated complete oral absorption but systemic availability was reduced corresponding to fist-pass elimination of 30% to 70%. There was linearity between oral dose and the corresponding area under the plasma concentration/time curve of d-propoxyphene and the metabolite norpropoxyphene. The kinetic measurements showed 2- to 3-fold interindividual variations: oral clearance, 1.3 to 3.6 1/min; systemic clearance, 0.6 to 1.2 1/min; apparent volume of distribution, 700 to 1,800 1; d-propoxyphene half-life (t1/2), 8 to 24 hr; and norpropoxyphene t1/2, 18 to 29 hr. There were pronounced intraindividual dose-dependent variations in oral clearance in some subjects. The intravenous concentration curves indicated a 3-compartment distribution model.

Sparteine and mephenytoin oxidation: genetic polymorphisms in east and west Greenland.

The oxidation of sparteine and mephenytoin was examined in a group of subjects living in Greenland: 300 in East Greenland and 171 in West Greenland. The distribution of the ratio between the chromatographic peak areas of S- and R-mephenytoin in the urine, the S/R ratio was clearly bimodal in both populations. Thus 9.3% of the East Greenlanders had S/R ratios of 0.9 or more and were phenotyped as poor metabolizers of mephenytoin. In the West Greenlanders, 2.9% of the sample had S/R ratios of 0.90 or more and were accordingly phenotyped as poor metabolizers. The intraethnic difference with regard to the frequency of the mephenytoin poor metabolizer is probably attributable in part to a much higher proportion of admixed Caucasian genes in the West Greenlanders than in the East Greenlanders. In both the East and the West Greenlanders, the sparteine metabolic ratio displayed marked interindividual differences without a clear bimodal distribution. Poor metabolizers arbitrarily defined as subjects with an metabolic ratio of 20 or more made up 3.3% of the East Greenlanders and 2.3% of the West Greenlanders, but the difference between the two groups was not statistically significant.

The selective serotonin reuptake inhibitor citalopram relieves the symptoms of diabetic neuropathy.

The effect of the selective serotonin reuptake inhibitor citalopram on diabetic neuropathy symptoms was examined in a double-blind, placebo-controlled, crossover study for two 3-week periods. Citalopram was given as a fixed dose of 40 mg/day. Data from 15 patients could be included in the statistical analysis. Citalopram significantly relieved the symptoms of neuropathy as measured by both observer- and self-rating in comparison with placebo. The steady-state plasma concentration of citalopram was 10 to 890 nmol/L. There was no significant relationship between the plasma concentration of citalopram and the effect of treatment as measured by observer- or self-rating. Two of 17 patients, both receiving citalopram, left the study because of side effects (nausea and vomiting or gastric upset and diarrhea). Side-effect ratings were significantly higher during administration of citalopram than during administration of placebo, but citalopram was generally well tolerated. Compared with earlier results obtained with imipramine administered on the basis of plasma level monitoring, citalopram appeared to be less effective, but seemed to be better tolerated.

Specific effect of venlafaxine on single and repetitive experimental painful stimuli in humans.

BACKGROUND: Tricyclic antidepressants relieve neuropathic pain, and the analgesic properties of tricyclic antidepressants are substantiated in human experimental pain models. It has been speculated that drugs with a selective inhibition of presynaptic reuptake of both serotonin and noradrenaline could have an analgesic effect comparable to the analgesic effect of tricyclic antidepressants. OBJECTIVE: Our objective was to evaluate the analgesic effect of the serotonin-noradrenaline reuptake inhibitor venlafaxine in human experimental pain models. METHOD: The study was carried out as a randomized, placebo-controlled, double-blind, crossover experiment that included 16 healthy volunteers. A 37.5-mg dose of venlafaxine was given orally 4 times with 12-hour intervals, and pain tests were performed before and 3 hours after the second and fourth doses. Pain tests included the determination of pain detection and tolerance thresholds to pressure, pain detection and tolerance thresholds on single electrical transcutaneous stimulation of the sural nerve, pain temporal summation on repetitive electrical sural nerve stimulation, and pain experienced during the cold pressor test. RESULTS: Venlafaxine increased thresholds for pain tolerance after single electrical stimulation (P =.005) and pain summation (P =.01) on repetitive stimulation but did not alter the thresholds for pain detection after single electrical sural nerve stimulation. Venlafaxine did not alter pain experienced during the cold pressor test or increase the pressure pain thresholds. CONCLUSION: Venlafaxine increases the pain tolerance threshold to electrical sural nerve stimulation and the threshold at which pain increases (pain summation). The impact of venlafaxine on pain summation in this experimental pain model on repetitive stimulation may indicate a potential analgesic effect for clinical neuropathic pain.

Moclobemide, a substrate of CYP2C19 and an inhibitor of CYP2C19, CYP2D6, and CYP1A2: a panel study.

The reversible monoamine oxidase A inhibitor moclobemide was given in single (300 mg) and multiple doses (600 mg/day) to 11 male and four female healthy volunteers (age range, 23 to 27) who were either poor metabolizers of S-mephenytoin (n = 7) or extensive metabolizers of S-mephenytoin (n = 8). All were extensive metabolizers of sparteine. Poor metabolizers of S-mephenytoin had lower moclobemide clearance values (median, single dose: 16.1 versus 43.2 L.hr-1; steady state: 13.4 versus 22.1 L.hr-1) and longer moclobemide half-life values (median, single dose: 4.0 versus 1.8 hours; steady state: 5.1 versus 2.7 hours) than extensive metabolizers of S-mephenytoin. The plasma levels of a metabolite formed by C-hydroxylation (Ro 12-8095) were lower in poor metabolizers of S-mephenytoin than in extensive metabolizers of S-mephenytoin. Moclobemide thus partially undergoes oxidative metabolism by way of the polymorphic CYP2C19. A combined mephenytoin, sparteine, and caffeine test performed before, during, and after multiple dosing of moclobemide showed changes in the metabolic indexes compatible with a reversible inhibition of oxidation by way of the corresponding CYP enzymes--CYP2C19, CYP2D6, and CYP1A2--during moclobemide treatment.

Plasma levels and antidepressive effect of imipramine.

The relationship between the antidepressive effect of imipramine and the plasma concentrations of imipramine and the active metabolite desipramine was studied in 24 patients suffering from endogenous depression. After a placebo period of 7 days, the patients received imipramine, 75 mg 3 times a day. The dose was reduced in patients with pronounced side effects. Blood samples for drug assay were drawn in the morning, 15 hr after the last drug intake. Imipramine and desipramine in plasma were assayed by quantitative in situ thin-layer chromatography. Individual variations in plasma concentration were 20- to 30-fold in both imipramine and desipramine. Severity of depression was assessed on the Hamilton Rating Scale (HRS). Eleven of 12 patients who responded satisfactorily to the treatment (HRS post-treatment score less than 8) had plasma concentration of imipramine greater than or equal to 45 mug/L, and desipramine greater than 75 mug/L, whereas the 12 patients not responding satisfactorily (post-treatment score on HRS greater than or equal to 8) all had concentrations of imipramine or desipramine or both below these limits.

The relationship between paroxetine and the sparteine oxidation polymorphism.

The relationship between the selective serotonin reuptake inhibitor paroxetine and the sparteine oxidation polymorphism was investigated in a combined single-dose (30 mg) and steady-state (30 mg/day for 2 weeks) study including a panel of nine extensive metabolizers and eight poor metabolizers of sparteine. The median area under the plasma concentration-time curve (AUC) after the first paroxetine dose was about seven times higher in poor metabolizers than in extensive metabolizers (3910 versus 550 nmol.hr/L), whereas at steady state the median AUCss tau interphenotype difference was only twofold (4410 versus 2550 nmol.hr/L). Plasma half-life and steady-state plasma concentration were significantly longer and higher, respectively, in poor metabolizers than in extensive metabolizers (41 versus 16 hours and 151 versus 81 nmol/L). Paroxetine pharmacokinetics were linear in poor metabolizers and nonlinear only in extensive metabolizers. Sparteine metabolic ratio (MR = 12 hour urinary ratio of sparteine/dehydrosparteine), increased during treatment with paroxetine in subjects who were extensive metabolizers, and after 14 days treatment two extensive metabolizers were phenotyped as poor metabolizers and the remaining extensive metabolizers were changed into extremely slow extensive metabolizers with sparteine MRs of 5.7 to 16.5. The inhibition of sparteine metabolism was rapidly reversed after cessation of paroxetine administration. In the poor metabolizers there were no significant changes in MRs during the study. It is concluded that paroxetine and sparteine metabolism cosegregates, but the interphenotype difference in metabolism was less prominent at steady state than after a single dose, presumably because of saturation of the sparteine oxygenase (CYP2D6) in subjects who were extensive metabolizers. Paroxetine is a potent inhibitor of sparteine oxidation by CYP2D6 in vivo.

Codeine increases pain thresholds to copper vapor laser stimuli in extensive but not poor metabolizers of sparteine.

The analgesic efficacy and kinetics of a single oral dose of 75 mg codeine was investigated in 12 extensive metabolizers and 12 poor metabolizers of sparteine in a double-blind, placebo-controlled crossover study. The cosegregation of the O-demethylation of codeine to morphine with the sparteine oxidation polymorphism was confirmed. Hence morphine could not be detected in the plasma of any of the poor metabolizers, whereas detectable morphine plasma levels were found in 10 of 12 extensive metabolizers. Pain thresholds to laser stimuli were determined before drug intake and 90, 150, and 210 minutes after drug intake. Codeine significantly increased the pricking pain thresholds in the extensive metabolizers (p less than 0.05), whereas there were no significant changes in the poor metabolizers. No change in pain thresholds occurred with placebo in any of the two phenotypes. In the extensive metabolizers there was a significant positive correlation between the increase in pain threshold and plasma concentration of codeine. The study supports the hypothesis that morphine formation is essential for achievement of analgesia during codeine treatment.

Imipramine metabolites in blood of patients during therapy and after overdose.

Plasma or serum concentrations of imipramine and five of its nonconjugated metabolites (desipramine, 2-OH-imipramine, 2-OH-desipramine, imipramine-N-oxide, and didesipramine) were followed in three cases of imipramine overdose and during steady state in 24 patients on continuous imipramine treatment. In the overdose cases the imipramine and desipramine concentrations declined monoexponentially with t 1/2s of 12 to 21 and 31 to 37 hr. The 2-OH-imipramine and 2-OH-desipramine levels were lower and declined in parallel with their corresponding parent compounds. In the patients on continuous imipramine treatment, the steady-state levels of 2-OH-imipramine and 2-OH-desipramine were very low or immeasurable (less than 15 nmol/l) in five patients. In most patients (n = 18) the hydroxymetabolite levels were much higher with 2-OH-imipramine/imipramine ratios of 0.09 to 0.45 and 2-OH-desipramine/desipramine ratios of 0.36 to 0.86. In one patient there were particularly high ratios (2-OH-imipramine/imipramine, 0.85; 2-OH-desipramine/desipramine, 1.30). The patients with very low hydroxymetabolite levels had considerably higher desipramine levels than the others, indicating that the low metabolite levels were due to poor hydroxylation. In one of these poor hydroxylators a desipramine t 1/2 of about 120 hr was estimated after imipramine discontinuation. With increased imipramine dose the 2-OH-imipramine levels tended to rise little or not at all. Imipramine-N-oxide could only be detected in the overdose cases during the first 6 to 12 hr and didesipramine was generally present only when the desipramine levels were above 200 nmol/l.