Amitriptyline plasma concentrations and clinical response

Is dose escalation of antidepressants a rational strategy after a medium-dose treatment has failed? A systematic review

BACKGROUND

Maximizing the dose of antidepressants is widely recommended in cases of non-response to medium-dose treatment. However, scientific evidence supporting high-dose treatment is scarce. Systematic studies comparing dose escalation with alternative strategies for refractory depression (i. e. augmentation or change of compound) are lacking. The aim of this publication is to review available direct and indirect evidence concerning dose increase of antidepressants after a medium-dose trial has failed.

METHOD

We performed a systematic literature search of Medline (1966-2003) and reviewed studies and publication references for available evidence.

DATA SOURCES AND STUDY SELECTION

Studies of the following types were included: 1) dose increase studies in treatment refractory patients, 2) comparative dose studies, 3) therapeutic drug monitoring studies.

RESULTS

Available data suggest differential efficacy of various pharmacological classes at more than medium-dosage. Direct evidence shows no increase of efficacy with high-dose selective serotonin reuptake inhibitor (SSRI) treatment; however, indirect evidence suggests enhanced therapeutic efficacy with high-dose tricyclic antidepressants. Few clinical data show ultra-high-dose treatment with the irreversible monoamine-oxidase-(MAO-) inhibitor tranylcypromine to be effective for refractory depression. Data concerning other selective compounds are insufficient to allow any definitive conclusion on the benefit of high-dose treatment.

CONCLUSIONS

Based on available data highdose antidepressant treatment of patients refractory to medium-dose treatment is recommended for tricyclic compounds but not for SSRI. Some data suggest beneficial efficacy of ultra-high doses of the irreversible MAOI tranylcypromine. Research on other substance groups is limited and inconclusive. Prospective studies comparing dose escalation with alternative strategies for treatment of non-responding patients are needed.

Amoxapine: a review of its pharmacology and efficacy in depressed states

Amoxapine is an N-demethylated dibenzoxazepine closely related in the neuroleptic loxapine. Its tricyclic structure appears to give it antidepressant properties resembling imipramine and amitriptyline. In uncontrolled trials it was shown to have antidepressant activity in usual doses up to 200 to 400mg daily. In placebo and double-blind controlled studies comparing amoxapine with the standard tricyclic antidepressants imipramine and amitriptyline, it was shown to be comparable in efficacy with a possibly somewhat faster onset of improvement of selected symptoms of depression in some studies. Because of the small study groups and lack of placebo control, many reports do not show statistically significant differences of treatment over standard drugs. To date there have been no studies comparing amoxapine with electroconvulsive therapy. Side effects were qualitatively similar to standard drugs with a suggestion that in standard doses or overdose myocardial effects are mild. However, the final place of amoxapine in the therapy of depressed states is still to be decided.

Monitoring of tricyclic antidepressant therapy

During the three-year period 1978-1980, 2141 plasma samples from 1055 patients receiving therapy with amitriptyline (77%) or nortriptyline (23%) were analysed using GLC with nitrogen selective detection. Compared to the recommended therapeutic ranges, wild inter-individual differences were observed in plasma drug concentration, even when corrections for dosage were made. Concentrations ranged from below the limit of sensitivity of the assay (5 microgram.1(-1)) to greater than 1 mg.1(-1). The reporting of toxic symptoms subjective side-effects) was found not to reliably predict high drug concentrations. Serious complications, however, were associated with high plasma drug concentrations. Neither nortriptyline nor amitriptyline displayed dose-dependent pharmacokinetics over the concentration ranges studied. Treatment with either drug produced age-related increases in drug concentration, which were more pronounced in female patients. With amitriptyline therapy, there was an age-related decrease in the plasma nortriptyline:amitriptyline ratio, suggesting that demethylation may be more influenced by increasing age than hydroxylation. Plasma drug monitoring of tricyclic antidepressant therapy is the only reliable means of ensuring that all patients receive a fair opportunity to benefit from these drugs.

Relationship between the plasma concentration of clomipramine and desmethylclomipramine in depressive patients and the clinical response

Thirty one in-patients suffering from depression were treated orally with clomipramine (C1) at various dosage, for 28 days, after a "wash-out" period of three days. In 17 patients receiving 75 mg per day of C1, steady state plasma levels of C1 were reached at Day 14, and steady state plasma levels of its active metabolite, desmethylclomipramine (DMC1), were reached at Day 21. In contrast, in 7 other patients receiving a dosage increasing to 150 mg per day at Day 7, mean plasma levels of C1 and DMC1 continued to rise during the entire treatment period. At the steady state, a correlation was found between C1 dosage expressed as mg kg body weight and the plasma concentration of C1 and DMC1. Factors such as tobacco and alcohol consumption seem to modify the C1/DMC1 ratio. A comparison of clinical response with plasma levels of C1, DMC1 and C1 + DMC1 showed a significant negative linear correlation.

A double blind comparison of zimelidine and amitriptyline in endogenous depression

In a randomized double-blind group comparison study of 40 patients with endogenous depression zimelidine appeared to be as effective an antidepressant as amitriptyline at 4 and 6 weeks using the Hamilton Rating Scale (HRS) and the Montgomery and Asberg Depression Rating Scale (MADRS). At 2 weeks there was a significantly better response (P less than 0.05) on zimelidine compared to amitriptyline on the clinician's global scale and 4 out of 10 items on the MADRS suggesting an early onset of action. A significant better response to zimelidine was seen on the item somatic anxiety (HRS) while the effect on sleep and appetite was better in the amitriptyline group. There were significantly more side effects, raw and corrected, in the amitriptyline-treated group. High steady state plasma concentrations of norzimelidine (greater than 800 nmol/l) which were significantly correlated with age (r = 0.8) were associated with a significantly poorer response suggesting that a lower dose than 200 mg in older patients may be appropriate.