Robert O. Friedel

Electrocardiogram changes and plasma desipramine levels during treatment of depression.

Twenty‐six symptomatic subjects who met research diagnostic criteria for major affective disorder and were free of cardiovascular disease were treated for 3 wk with a fixed dosage schedule of desipramine (DMI) to a maximum of 200 mg/day. An electrocardiogram (ECG) and DMI plasma level determinations were obtained before treatment and weekly thereafter. DMI levels during the trial ranged from 13.4 to 882.2 ng/ml. DMI treatment was associated with increase in heart rate (p < 0.001), prolongation of the PR (p < 0.001), QRS (p < 0.001), and QTC intervals (p < 0.001), and increase in T wave amplitude (p < 0.001). Significant (p < 0.001) but relatively weak correlations were noted between DMI plasma levels and heart rate (r = 0.405), QRS interval (r = 0.346), QTC interval (r = 0.534), and T wave amplitude (r = −0.386). PR interval prolongation was independent of DMI levels (r = 0.171). DMI treatment induced no clinically significant ECG alterations or cardiovascular adverse effects. The relevance of DMI plasma level and the possible roles of other contributing factors in the production of these ECG changes are discussed.

Urinary Mhpg Excretion and Treatment with Desipramine or Amitriptyline: Prediction of Response, Effect of Treatment, and Methodological Hazards

Urinary 3-methoxy-4-hydroxyphenylglycol (MHPG) excretion was measured in 49 unipolar depressed outpatients to examine the relationship between pretreatment MHPG levels and therapeutic response to desipramine and amitriptyline and to determine the effects of these agents on MHPG excretion. Pretreatment MHPG excretion was greater in amitriptyline responders than amitriptyline nonresponders, but no different in desipramine responders compared to desipramine nonresponders. Pretreatment MHPG excretion did not differentiate desipramine from amitriptyline responders. Treatment for 3 weeks was associated with a decrement in MHPG excertion, particularly in the desipramine responders and combined desipramine and amitriptyline responders. Among patients within the postulated optimal desipramine plasma level range and patients with plasma amitriptyline plus nortriptyline levels > 70 ng/ml, high pretment MHPG excretion predicted therapeutic response and response was accompanied by a reduction in MHPG excretion. The interpretation of these findings is possibly confouded by the demonstration of a poor correlation (r = 0.61) between duplicate MHPG samples analyzed by a widely employed gas chromatography method and a gas chromatography/mass spectrometry technique. The methodological pitfalls encountered in the course of this investigation and their possible implications for similar studies are discussed.

ECG Effects of Comparable Plasma Concentrations of Desipramine and Amitriptyline

To compare the electrocardiographic effects of therapeutic doses of desipramine and amitripty-line, weekly electrocardiograms (ECGs) were obtained from 46 depressed outpatients treated blindly for 3 weeks to a maximum of 200 mg/day. There was no difference in mean weekly plasma tricyclic antidepressant levels achieved for the two drugs. Compared to baseline measures, treatment with both drugs was associated with an increase in heart rate and a reduction in T wave amplitude, whereas prolongation of the QRS and QTc intervals was significant only for desipramine patients. Comparisons between drugs revealed a greater prolongation of QRS interval duration with desipramine treatment. Changes in ECG measures were not correlated with plasma tricyclic antidepressant levels. The absence of QRS interval prolongation among amitriptyline patients is additional evidence for the importance of distinguishing between the ECG effects of therapeutic and toxic doses of the tricyclic antidepressant.

Urinary 3-Methoxy-4-Hydroxyphenethylene Glycol in the Prediction of Pain and Depression Relief with Doxepin: Preliminary Findings

A group of patients with both depression and chronic pain were treated with doxepin for 4 weeks. Elevated (above median) initial levels of urinary 3-methoxy-4-hydroxyphenethylene glycol (MHPG, a metabolite of central nervous system [CNS] norepinephrine metabolism) were found to be significantly related to final pain relief, but not to improvement in depression. Anxiety, as measured on the Zung Anxiety Scale, was an important covariable, as high anxiety levels were significantly associated with both higher MHPG levels and with pain relief. Possible biochemical and psychophysiological mechanisms associated with the modulation of pain are discussed.

Risperidone treatment of borderline personality disorder assessed by a borderline personality disorder-specific outcome measure: a pilot study.

To the Editors: Borderline personality disorder (BPD) has a prevalence of 1.4% and results in high rates of mortality (9%), morbidity, and comorbidity, and in disproportionately high uses of primary medical and psychiatric services. One of the most promising lines of pharmacotherapeutic research in BPD has evolved from the observed effects of low doses of neuroleptics. According to the American Psychiatric Association practice guideline, antipsychotic agents are recommended in patients with BPD who present with cognitive-perceptual symptoms, acute anger, hostility, assaultiveness, and self-injury. Placebocontrolled studies of these agents in BPD have consistently demonstrated significant symptom reduction. However, improvement in symptoms within and between the 4 behavioral dimensions of BPD (affective dysregulation, impulsivity, cognitive-perceptual impairment, and disturbed relationships) has varied among studies, even between those utilizing the same drug. This inconsistent pattern of efficacy of antipsychotic agents in BPD may be due, in part, to the treatment outcome measures used. No published study has utilized an instrument developed to evaluate treatment change in the specific symptoms and behavioral dimensions of BPD, possibly resulting in the observed inconsistency of findings and a bias in favor of type II errors. In general, condition-specific instruments seem to provide greater sensitivity to change. A self-report treatment outcome measure evaluating the symptoms and symptom domains of BPD may provide enhanced sensitivity and validity in evaluating treatment effects in this population and would complement the clinician-administered Zanarini Rating Scale for BPD (ZAN-BPD). Finally, there are conflicting results from the only 2 studies evaluating the efficacy of risperidone in BPD, a medication used off-label in patients with the disorder. Therefore, the primary objectives of this pilot study were to test the efficacy and tolerability of risperidone in subjects with BPD and to provide a preliminary evaluation of the utility of a new self-report treatment outcome measure for BPD, the Borderline Disorder Rating Scale (BDRS). Approval for this open-label trial was obtained from the University of Alabama at Birmingham (UAB) institutional review board, and written informed consent was obtained from all subjects. Female subjects were recruited from newspaper advertisements and from physicians practicing at UAB. Subjects met Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria for BPD (Structured Clinical Interview for DSM-IV Axis II Personality Disorders), scored 8 or higher on the Revised Diagnostic Interview for Borderlines, used an accepted means of contraception, and demonstrated an understanding of all tests and examinations used. Exclusion criteria were participation in a clinical drug trial within 1 month before study entry; currently pregnant or nursing; current diagnosis of any DSM-IVAxis I disorder (with the exception of dysthymia, posttraumatic stress disorder, generalized anxiety disorder, or substance abuse or dependence inactive for the prior 3 months); current schizotypal or antisocial personality disorder; active suicidal ideation with a plan; serious, unstable medical illnesses; use of any additional psychotropic medication during or within 4 weeks before the study, except for a maintenance dose of antidepressants; and treatment with any form of psychotherapy during the study. Comprehensive physical and psychiatric histories and examinations and laboratory studies were performed at entry into and upon completion of the study. One of the 2 primary treatment outcome measures used in the study was the BDRS, a 40-item scale developed by 4 of the authors (R.O.F., W.T.J., R.S.M., and A.S.) then at the UAB BPD Program. Four groups of 10 items each were developed to evaluate change in the 4 domains of BPD. A more complete description and a copy of the BDRS are available from the corresponding author. Further evaluation of its psychometric properties will be published elsewhere. Upon entry into the study, all subjects completed a battery of outcome measures that was readministered at the end of weeks 2, 4, 6, and 8. These included the BDRS and the second primary outcome measure, the Brief Symptom Inventory (BSI; 53 items); the State-Trait Anger Expression Inventory 2 (57 items); the Beck Depression Inventory II (26 items); the Barratt Impulsiveness Scale, 11th version (BIS-11; 30 items); the Quality of Life Enjoyment and Satisfaction Questionnaire (feelings, work, and social relations subscales); the Clinical Global Impression (CGI)—Severity and Improvement Clinician-Rated; the CGIImprovement Patient-Rated Scales; and the Social Desirability Scale. Safety and tolerability were assessed by clinicians at each visit using the treatment emergent signs and symptoms. The initial dose of risperidone for all subjects was 0.25 mg/d, given at bedtime to minimize any impairment due to somnolence. The dose was adjusted at each visit based on clinical perception of efficacy and tolerability, to a maximum dosage of 2 mg/d, a dose limit used to optimize tolerability as a result of experience in the UAB BPD Clinic. The dose was reduced at any time if, in the clinician’s judgment, adverse effects or poor tolerability indicated doing so. Letters to the Editors

Effects of γ-aminobutyric acid on 33Pi incorporation into rat brain phospholipids in vivo☆

Abstract The effects of γ-aminobutyric acid (GABA), bicuculline and strychnine on the incorporation in vivo of 33 Pi into phospholipids of rat brain were studied at 10 and 30 minutes after intracisternal injection of the radionuclide. GABA inhibited labeling of phospholipids in the three brain regions studied at both times. Bicuculline by itself had no significant effect on 33 Pi incorporation, but totally blocked the inhibitory effect of GABA in all three brain regions. Strychnine by itself inhibited phospholipid labeling in the brain stem and forebrain, had no significant effect on GABA inhibition of 33 Pi incorporation in the cerebellum and forebrain, and partially blocked the GABA effect in the brain stem. GABA inhibited 33 Pi incorporation into phosphatidic acid, phosphatidylinositol, phosphatidyl choline and phosphatidyl ethanolamine but had no effect on phosphatidyl serine. The data suggest that the inhibitory effects of GABA on CNS phospholipid labeling are mediated specifically through GABA receptor sites.