Monte-Carlo, November 28 - December 1, 2001

I - M 


G. Maina, U. Albert, C. Bergesio, F. Bogetto
Department of Neuroscience, Anxiety and Mood Disorders Unit, University of Turin

Aim of the study Phobias, including pharmacophobia, are an inherent feature of Panic Disorder (PD); moreover, PD patients are especially sensitive to physical symptoms and medication side effects. These essential features of PD exert a negative influence on the compliance of patients to the pharmacological therapy, especially at the beginning of the treatment. The aim of the present study was to investigate compliance to pharmacological treatment and drop‑out rates due to severe adverse events at the beginning of therapy in a group of patients with PD, and compare them to those of a control group of patients suffering from OCD.

Materials and Methods 198 patients with a principal diagnosis of PD (DSM-IV) entered a 12-week acute treatment period with SSRIs. Target doses and time to reach the target dose followed the APA Treatment Guidelines for Panic Disorder. Patients were assessed weekly during the first 4 weeks in order to evaluate compliance and drop-out rates due to adverse events. For the purposes of the present study patients were considered poor-compliant whenever they skipped a drug dose. During the first 4 weeks patients received medication from the study staff and were instructed to return all drug packaging to the study site in order to allow drug accountability. Adverse events were recorded using the UKU Side Effect Scale; drop‑outs due to side effects were recorded. Poor‑compliance and drop-out rates were compared to those of a group of 198 patients with a principal diagnosis of OCD matched for age and sex and treated with SSRIs according to international treatment guidelines. The chi-square test was used for statistical comparisons.

Results  The rate of poor-compliance at the end of the first 4 weeks of treatment was 24.2% for PD patients; 10.6% of patients discontinued before the fourth week because of adverse events. When examining gender-differences, a significant difference emerged only for drop-out rates, with males discontinuing significantly more than females (16.9% vs. 7.1 %). OCD subjects showed a poor‑compliance rate significantly lower that that reported for PD patients (10.7% vs. 24.2%); this significant difference was mainly due to the difference between PD females and 0CD females (26.8% vs. 5.4%). No differences between PD and OCD patients were detected with regard to drop‑out rates due to side effects.

Conclusions  Our results confirm that 1) PD patients have a higher poor‑compliance rate compared to OCD patients; this difference is even stronger for females; we might speculate that this high poor-compliance rate is due to pharmacophobia; 2) drop out rates during the first 4 weeks of treatment are not higher in PD patients than in OCD subjects. This might indicate that PD patients, once they get treatment, do not report more severe adverse events as compared to other anxiety disorder subjects.


 R Mallick, J Chen, R Entsuah
Wyeth-Ayerst Research, Collegeville, PA, USA

     The study objective was to evaluate baseline [BL] impairment in work and corresponding improvement following treatment of depression.

    Data from 8 clinical studies of 2046 patients meeting DSM-III-R or DSM-IV criteria for moderate-severe depressive disorder were pooled. Patients were randomly assigned to receive venlafaxine/venlafaxine XR [VEN/VEN XR] 75–375 mg/d (n=851); a selective serotonin reuptake inhibitor [SSRI] (fluoxetine, paroxetine, or fluvoxamine at various dosages, n=749); or placebo (n=446). Work and Activities (WA) item scores (0 to 4) were evaluated at BL and after 8 weeks of treatment. Of patients who were completely impaired (WA item score 4) at BL, the proportion who had improved to full work functionality (WA item score 0) was compared across treatment groups.

    At BL, median WA item score for all patients was 3. Overall, 15% (312/2046) had a BL score of 4; the placebo group had significantly fewer patients with complete work impairment (9.6% compared with 16.0% for the SSRI group and 17.5% for the VEN/VEN XR group). After 8 weeks’ treatment, 28.9% (43/149) of patients who were completely impaired in the VEN/VEN XR group had improved to full functionality compared with 15.8% (19/120) of patients in the SSRI group (VEN/VEN XR/SSRI OR=2.2, P=0.01) and 16.3% (7/43) of patients in the placebo group (VEN/VEN XR/placebo OR=2.1, P=0.11; SSRI/placebo OR=0.97, P=1.0).

    A significant proportion of patients who were completely work impaired at BL improved dramatically to full functionality by treatment end. VEN/VEN XR was associated with the highest proportion of such dramatically improved patients.


R Mallick, P Ninan, R Shrivastava

Wyeth-Ayerst Research, Collegeville, PA, USA

In patients with major depressive disorder, we compared the effect of venlafaxine, SSRIs, and placebo on specific depressive symptoms and patient characteristics predictive of work impairment and of improvement with treatment.

    Data were pooled from 8 clinical studies of 2046 intent-to-treat patients meeting DSM-III-R or -IV criteria for moderate-to-severe depression given venlafaxine (75–375 mg/d), an SSRI (fluoxetine, paroxetine, or fluvoxamine at various dosages), or placebo. Work impairment was measured using the Work and Activities (WA) item of the HAM-D17. Logistic regression identified the baseline characteristics  significantly (at p<0.05) predictive of behavioral work impairment (WA=3 or 4) at baseline.

    At baseline, 76% of patients were behaviorally work impaired; 15%  had ceased employment (WA=4). Moderate-to-severe CGI‑S scores, above-average baseline retardation, general somatic symptoms, genital symptoms, late insomnia, and above-average age were significantly predictive of behavioral work impairment at baseline. Patients with above-average baseline retardation and suicidality, work dysfunction, and above-average age were significantly less likely to attain early, sustained return to full work function; those with a longer-than-average depressive episode and those treated with venlafaxine were significantly more likely to do so.

    The finding that a longer depressive episode increased the odds of a sustained return to full work function may represent successful spontaneous coping over time unrelated to treatment. However, treatment with venlafaxine was independently predictive of enabling patients with severe baseline depressive symptoms and work dysfunction to return earlier to sustained full work function, in relation to both SSRIs and placebo.


T. Mamiya, M. Suzuki and M. Ukai

Department of Chemical Pharmacology, Faculty of Pharmacy, Meijo University, Yagotoyama, 150, Tempaku-ku, Nagoya, 468-8503, JAPAN


Learned helplessness is widely employed as a model of depression with high validity criteria.   Several pharmacological and neurochemical findings support a role for opioid systems in affective disorders in the experimental and clinical studies.  Because it is unknown whether kappa-opioid receptor agonist is involved in the learned helplessness model of depression, in this study we evaluate the effects of U-50,488H on learned helplessness in mice. 

Male albino ddY mice (5-8 weeks old) were used.  1. Inescapable electric shock pretreatment:  Mice were treated with inescapable electric footshocks.  2.  Conditioned avoidance response:  Active conditioned avoidance response was observed 24 h after inescapable electric shock.  Each animal was placed singly in the box and subjected to 30 active conditioned avoidance training.  To avoid the electric shock, the mouse had to press the lever in the box.  An escape failure was considered when the mouse failed to escape the shock.  Active conditioned avoidance training were performed for 4 consecutive days, and the number of escape failures was recorded.  All experiments were performed in accordance with the Guidelines for Animal Experiments of Meijo University. 

Non-shocked saline-treated animals displayed fewer escape failures than shocked animals.  Mice pre-exposed to inescapable electric shocks were treated with U-50,488H.  Stimulation of kappa -opioid receptors by U-50,488H (10 mg/kg i.p.) attenuated the escape failures induced by shock pre-exposure.  This improvement by U-50,488H was blocked by MR2266, an opioid receptor antagonist.  Although the precise nature of the interaction in the kappa -opioid systems of the brain is unknown, kappa -opioid receptor agonists may activate the impaired neuronal systems.  These results suggest that kappa-opioid receptor agonist plays an inhibitory role in the learned helplessness in mice. 

Acknowledgment:  This study was supported in part by Grants-in-Aid for Scientific Research and Scientific Frontier Research Project from Ministry of Education, Culture, Sports, Science and Technology, Japan. 


 C. Mastrocinque, S. Pini, L. Dell’Osso, S. Navari, M. Morandi, N. Mosti, M. Abelli, G.B. Cassano
Stefano Pini, Department of Psychiatry, University of Pisa, I-56100 Pisa, Italy 
Phone/Fax +39 050 835420 - e-mail: s.pini@psico.med.unipi.it

    Objective: Bipolar I disorder often co-occurs with panic disorder, but little is known about the relationships between the clinical features of bipolar illness and this comorbid condition. The authors assessed comorbid lifetime and current axis I panic disorder  in 166 patients with bipolar I disorder with psychotic features and the relationships of this comorbid disorder to selected clinical and historical illness variables.

    Method: 166 outpatients with bipolar I disorder with psychotic features were evaluated using structured diagnostic interviews and clinician-administered and self-rated questionnaires to determine the diagnosis of bipolar disorder, comorbid axis I disorder diagnoses, and demographic and clinical characteristics.

    Results: Of the 166 bipolar patients, 34 (20.5%) fulfilled DSM-IV criteria for lifetime diagnosis of panic disorder (no panic attacks during the month previous to the evaluation), 17 (10.2%) for current panic disorder (panic attacks during the last month) and  115 (69.3%) had no panic disorder comorbidity. Comparisons among these three groups showed that the group with lifetime panic disorder comorbidity had an earlier age at onset of syndromal bipolar disorder, higher levels of phobic anxiety and more frequent depressive polarity at the index episode. Current axis I panic comorbidity was associated with a history of substance abuse, more severe thought disturbances and activation and more frequent mixed polarity at index episode, this latter reaching borderline statistical significance.

    Conclusions: Patients with bipolar disorder with psychotic features often have comorbid panic disorder. Consideration of both lifetime and current axis I panic disorder comorbidity in bipolar disorder might represent a clinically useful way of characterizing or subtyping the illness. Further research into the prognostic and treatment response implications of panic disorder comorbidity in bipolar disorder is important.


A Mehnert (1), M Jones (2), L Fellows (3). On Behalf of the RODOS Investigator Group.

(1) Janssen Research Foundation, Beerse, Belgium; (2) Department of Psychology, Macquarie University, Sydney, Australia; (3) Graylands Selby-Lemnos and Special Care Health Services, Mount Claremont, Western Australia.

    Objective: To compare the patterns of drug usage, costs and outcomes associated with the inpatient treatment of bipolar disorder with risperidone (RIS) or olanzapine (OLA).

    Methods: A retrospective cohort study using patient chart review was undertaken. Patients were included if the following inclusion criteria were fulfilled: (1) diagnosis of bipolar disorder, (2) discharged or at least 120 days of follow-up, (3) £ 65 years of age and (4) treated with either RIS or OLA. Treatment outcomes and cost of treatments were calculated and statistically analysed.

    Results: Data on 62 patients was available (RIS=30, OLA=32) from hospitals in Australia, The Netherlands, Norway and Spain. Demographic characteristics for the two samples were similar. The median time to onset of efficacy was 9 days for RIS and 12 days for OLA. The average daily dose for RIS was 3.0 ± 1.5mg and with OLA 11.0 ± 4.1mg. Tolerability of study drugs was generally good in both groups, with 10% of RIS-treated and 16% of OLA-treated patients discontinuing therapy. Patients receiving OLA were more likely to discontinue due to lack of efficacy (9%) than those receiving RIS (0%), although this difference did not reach statistical significance (p=0.09). Usage of concomitant neuroleptics and other medications was similar in the two groups. Daily costs of the studied medication were over twice as high for OLA (4.6 ± 1.7 USD) than for RIS (2.0 ± 1.0 USD, p<0.0001). Daily costs for all in-patient drug use were also significantly higher in the OLA group (6.2 ± 2.4 USD) than the RIS group (2.9 ± 1.9 USD, p<0.0001).

    Conclusions: RIS and OLA were both effective in the bipolar patients studied. In-patient costs for patients receiving RIS were significantly lower than costs for patients receiving OLA in this sample of patients with bipolar disorder.



A Mehnert (1), M Jones (2), L Fellows (3). On Behalf of the RODOS Investigator Group. (1) Janssen Research Foundation, Beerse, Belgium; (2) Department of Psychology, Macquarie University, Sydney, Australia; (3) Graylands Selby-Lemnos and Special Care Health Services, Mount Claremont, Western Australia.

    Objective: To compare the outcomes associated with atypical antipsychotic treatment in patients with schizophrenia and mood disorder.

    Methods: A retrospective cohort study using patient chart review was undertaken. Patients were included if the following inclusion criteria were fulfilled: (1) diagnosis of schizophrenia or mood disorder, (2) discharged or at least 120 days of follow-up, (3)£ 65 years of age, (4) treated with either risperidone (RIS) or olanzapine (OLA).

    Results: Data on 1901 patients with schizophrenia and 112 patients with mood disorders were available from hospitals in 9 countries: Australia, Austria, Denmark, Germany, Great Britain, The Netherlands, Norway, Spain and Sweden. Schizophrenia patients typically had an earlier age of onset of symptoms (24.5 years) than those with mood disorder (30.1 years, p=0.0002). Patients with mood disorder received lower doses of RIS or OLA (p<0.0001), and were more often discharged within 120 days (p=0.02). A slightly shorter time to achieve efficacy (10 days versus 12 days) was seen in patients with mood disorders (p=0.01). Median time to discharge was significantly shorter in patients with mood disorders (22 days) than patients with schizophrenia (35 days, p<0.0001). Adjusting for background factors has little impact on the dose comparison, time-to discharge or time-to efficacy analyses.

    Conclusions: Atypical antipsychotics were effective in treating acute psychotic exacerbations in schizophrenia and mood disorders, with a median time to efficacy of 10-12 days. Patients with mood disorders generally required lower doses of atypical antipsychotic and were discharged faster than schizophrenic patients.


CM Miner, EB Brown, JS Gonzales, R Munir
Lilly Research Laboratories,

Lilly corporate Ctr. Indianapolis, IN 46285 USA


 Background: The chronic, recurrent nature of major depression often necessitates maintenance treatment. Since compliance with daily therapy may be difficult, a once-weekly fluoxetine formulation has been developed. This study addresses the impact on quality of life, patient satisfaction, and compliance upon switching patients whose symptoms had responded to daily dosing with other SSRIs to fluoxetine once‑weekly.

 Methods: Patients met DSM-IV criteria for major depression prior to beginning treatment for their current episode, received 6-52 weeks of treatment with citalopram (20-40 mg/day, in=83), paroxetine (20 mg/day, n=77), or sertraline (50-100 mg/day, n=86), and responded to that treatment (CGI-Severity score <2, modified HAM-D-17 score <10). The current SSRI was continued for 1 week, then switched to open‑label enteric-coated fluoxetine 90 mg taken once weekly for 12 weeks. Quality of life was assessed with the Short Form Health Survey and patient satisfaction with a questionnaire.

 Results: Significant improvement in quality of life measures, including vitality, general mental health, and role limitations due to emotional problems, after switching were identified. At endpoint, most patients (84.7%) indicated a preference for once‑weekly fluoxetine over their once‑daily SSRI, despite the fact that, at the study start most (76.4%) were satisfied with their daily SSRI. Compliance rates were high with no evidence of decreased compliance over time (99.6%, 97.2% and 99.5% at 4, 8 and 12 weeks).

 Conclusions: Fluoxetine once weekly is a viable treatment option for long-term therapy. Patients showed significant improvement in quality of life measures after the medication regimen was switched, while a comparable antidepressant response was maintained. This suggests variables beyond general depressive symptom assessment should be considered when evaluating treatment response and ultimately setting treatment goals.




C . Miner, E. Brown, J. Gonzales, and S. McCray Lilly Research Laboratories, Lilly Corporate Ctr. Indianapolis, IN 46285 USA


 Objective: Daily fluoxetine treatment during the luteal phase of the menstrual cycle is efficacious for PMDD, but weekly fluoxetine treatment during the luteal phase has not previously been examined. We will present data from a randomized, placebo controlled clinical trial of enteric-coated fluoxetine 90 mg given once or twice during the luteal phase for treatment of PMDD.

 Methods: Women recorded PMDD symptoms daily using the Daily Record of Severity of Problems (DRSP) throughout the trial. Following a screening period of two cycles, patients received single‑blind placebo on day 14 and day 7 before expected onset of menses for one cycle. Placebo nonresponders were then randomized to take placebo only (days 14 and 7), enteric‑coated fluoxetine 90 mg only (days 14 and 7), or placebo (day 1 14) and enteric‑coated fluoxetine 90 mg (day 7) in a double‑blind manner for three cycles. This was followed by one cycle of single‑blind placebo on days 14 and 7 before expected onset of menses.

 Results: The primary efficacy measure is change from mean baseline luteal phase DRSP scores to mean treated luteal phase DRSP scores over three months of treatment. Additionally, effects of enteric‑coated fluoxetine 90 mg on specific symptoms of PMDD, quality of life and safety will be analyzed.

 Conclusions: Enteric‑coated fluoxetine 90 mg once and/or twice a cycle may provide safe and effective treatment of PMDD.