Faculty Bibliography

Trichotillomania is characterized by chronic hair-pulling resulting in noticeable hair loss. In a preliminary study, cerebrospinal fluid (CSF) measures in 8 medication-free, female, trichotillomania patients were compared to those of matched, normal controls. There was no difference between patients and controls in measures of CSF cortisol, 5-hydroxyindoleacetic acid (5-HIAA), homovanillic acid (HVA) and 3-methoxy-4-hydroxyphenylglycol (MHPG). CSF measures did not correlate with measures of trichotillomania symptomatology. However, degree of response to treatment with serotonin re-uptake inhibitors significantly correlated with baseline CSF 5-HIAA. This suggests that central serotonin turnover is specifically relevant to treatment response to serotonin re-uptake inhibitors in trichotillomania.

Subject heterogeneity increases generalizability of study findings so long as site differences do not interact with treatment effects. Of 51 baseline characteristics of 234 patients, 31 show significant site differences. Of 33 baseline demographic, treatment history, and presenting symptom items, site differences were seen on 16. This heterogeneity emphasizes the importance of examining site interactions with other effects. The first opportunity to examine such interactions is in the prediction of patient stabilization following the index acute episode. Among the 33 demographic, treatment history, and presenting symptom variables, 11 significantly predicted this short-term outcome. Only 2 of these 33 variables showed site-by-baseline interactions in predicting stabilization, most likely due to chance sampling fluctuations. Thus, site heterogeneity is adding generality without confounding the predictions.

Cerebrospinal fluid levels of norepinephrine and six monoamine metabolites were measured in 28 medication-free depressed patients. Patients with a major depressive episode with melancholia (n = 15) had significantly lower levels of the three dopamine metabolites: homovanillic acid (HVA), dihydroxyphenylacetic acid (DOPAC), and conjugated dihydroxyphenylacetic (CONJDOPAC), when compared with a combined group of patients with a major depressive episode or dysthymic disorder (n = 13). In patients with major depressive episode with melancholia, levels of HVA and of the serotonin metabolite 5-hydroxyindoleacetic acid significantly correlated with the severity of depression. In the total group of 28 depressed patients, cerebrospinal fluid (CSF) levels of norepinephrine significantly correlated with symptoms of anxiety. In both patients with major depressive episode and major depressive episode with melancholia, those who were non-suppressors on the dexamethasone suppression test had significantly higher CSF levels of the norepinephrine metabolite 3-methoxy-4-hydroxyphenylglycol compared to those who were suppressors

The monoamine metabolites 5-hydroxindoleacetic acid (5-HIAA), homovanillic acid (HVA) and 3-methoxy-4-hydroxyphenylethyleneglycol (MHPG) were measured in the lumbar cerebrospinal fluid (CSF) of 27 chronic schizophrenic patients who at some time had attempted suicide, and were compared with values from 27 chronic schizophrenic patients without a history of attempted suicide. There were no significant differences between either the violent or non-violent attempters and those without a history of attempted suicide in the mean lumbar CSF concentrations of the serotonin metabolite 5-HIAA, the dopamine metabolite HVA, or the norepinephrine metabolite MHPG. Significantly more of the suicide attempters had a previous major depressive episode, had received a course of ECT, and had significantly more psychiatric admissions than those who had never attempted suicide

Computed tomographic scans of 30 chronic schizophrenic patients and 26 matched medical controls were blindly assessed for ventricular brain ratio, cortical atrophy, third-ventricle diameter, and cerebellar atrophy. Schizophrenic patients had significantly larger third ventricles than the medical controls. There was no difference in the other brain morphologic variables. Phenomenology, drug response, CSF levels of 5-hydroxyindoleacetic acid, 3-methoxy-4-hydroxyphenylglycol, homovanillic acid, and a wide variety of clinical variables did not correlate with any measure of brain morphology. Clinicopathologic correlates of brain morphology may be limited to those patients with significant atrophy