This follows a previous article with a general summary of the two medication studies for BPD presented here: http://www.pdan.org/drug-trials-encouraging-results-for-bpd/
SEROQUEL XR (OLANZAPINE) STUDY FOR BPD, 2010-2013
“Comparison of Low and Moderate Dosages of Extended- Release Quetiapine in Borderline Personality Disorder: A Randomized, Double-Blind, Placebo-Controlled Trial”, by Donald Black, Mary Zanarini, Ann Romine, Martha Shaw, Jeff Allen, and Charles Schulz. Am J. Psychiatry (Nov. 2014) Vol. 171, No. 11, pp 1174-1182.
The study looked at about 100 patients suffering from BPD. This study did not use anypsychotherapy. It researched the changes in severity of traits or symptoms of BPD along certain psychological scales.
The study took place from January 2010 to March 2013. It followed these patients through an 11-week period for each patient. The first 2 weeks consisted of office visits for screening and to assess the mental state of these patients, followed by 8 weeks of medication (without psychotherapy), during which the patients were seen once a week by a professional to assess their progress, followed by an 11th week office visit to assess their resulting mental and physical state.
After the initial interviews, 95 patients were qualified for the medication phase of the study. The patients were between the ages 18 and 45, living in the USA, with levels of moodiness, impulsivity, distrustfulness and difficult relationships, which satisfied the DSM criteria for BPD.
These 95 patients were randomly placed into three groups that were given certain dosages of Quetiapine extended-release, known to consumers as Seroquel XR.
A placebo is a sugar pill. The act of just taking a pill can make people feel better because they think the pill is helping them. in general, for almost any mental condition, it has been observed that about 30 percent of people in a study report feeling better if they thought they were taking real medication, even though it was a sugar pill. That is why researchers use a “control group” of people: to determine if the results are from the pill or in people’s mind.
Neither the doctors administering the medication, nor the patients knew who was taking a sugar pill and who was taking the real medication. This is called a “double blind” study, and is the gold standard to make sure the results come from the medication only.
Patients in the Low-Dosage Group received 50 mg/day the first week; then 150 mg/day for 7 weeks.
Patients in the Moderate-Dosage Group received 50 mg/day the first week, 150 mg/day for 3 weeks, then 300 mg/day for 4 weeks.
People in the third group received placebo pills containing no active medication.
The group sizes at the beginning of the study were:
- Placebo (control) Group: 29 persons
- Low-Dosage Group: 33 persons
- Moderate-Dosage Group: 33 persons
By the 11th week final assessment, 64 patients completed the study out of the 95 initial patients, for a retention rate of 67%.
The group sizes at the end of the study were:
- Placebo (control) Group: 23 persons
- Low-Dosage Group: 22 persons
- Moderate-Dosage Group: 19 persons
The overall completion rate for the 8-week double-blind treatment phase was 67% (67% for the low-dosage quetiapine group, 58% for the moderate-dosage quetiapine group, and 79% for the placebo group). Participants who experienced sedation were more likely to drop out.
Among participants who completed the study, 82% in the low-dosage quetiapine group were rated as “responders,” compared with 74% in the moderate-dosage group and 48% in the placebo group.
Treatment-emergent adverse events included sedation, change in appetite, and dry mouth.
Several diagnostic scales were used to study the severity of various symptoms throughout the duration of the clinical trial. The main scale used was the Zanarini-BPD scale (Z-scale). This scale looks at patients along nine dimensions (the 9 criteria of BPD) and asks the person filling in the questionnaire (the patient or therapist, both evaluations were used in the trial) to evaluate the severity of their symptoms on a scale of 0 to 4. Then one adds up the score in each dimension. Hence the total score can vary from 0 to 36.
Normal Level on Zanarini-BPD Scale
When one takes the Zanarini-BPD scale questionnaire, a “normal person” might give herself or himself a score of 0 to 2 in a few dimensions. In another study comparing BPD patients to psychiatric patients without BPD, those without BPD had a score of 5.2 on the Zanarini-BPD scale.
Clinical BPD is considered a total Z-scale score of 9 or greater on the scale. Most BPD treatment studies start with a population of people with initial scores around 16 to 19. After treatment, scores may fall to around 9-13 depending on the study. For BPD patient, anything under 9 would be considered a very substantial improvement.
RESULTS ON Z-SCALE FOR BPD
In this Seroquel XR study (table 2), the initial mean Z-scales of
- Placebo group : 14.6
- Low dosage : 15.8
- Moderate dosage: 17.7
At the end of the trial, researchers from a decrease of
- – 6.0 for Placebo
- – 9.8 for Low dosage
- – 7.9 for Moderate dosage
Thus, the end mean results were:
- Placebo: 8.6
- Low dosage: 6.0
- Moderate dosage: 9.8
In other words the Low-dosage group achieved by the end of the trial – on average for the group – remission from clinical BPD on the Z-scale. This is a rather remarkable feat for a group of 22 people.
The Moderate-dosage group achieved by the end of the trial a level that was near normal, below 10, though still above 9. A Z-scale level of 9 or greater is viewed as clinical BPD. Note that this group started with an average Z-scale that was almost 2 points higher than the Low-dosage group, hence their average decrease on the scale was impressive too.
A Z-scale level of 6.0 is also much closer to the level of 5.2 found in non-BPD patients in another study. Charles Schulz remarked: “There is a study comparing the BPD patients to psychiatric patients without BPD. Those without BPD have a score of 5.2 and the BPD patients in the low dosage group had a score of 6.0. So in our Seroquel XR study, at a Zanarini-scale of 5.2 the Low-dosage patients were very close to non-BPD people.”
Study: Mary C. Zanarini, (2003). Zanarini Rating Scale For Borderline Personality Disorder (ZAN-BPD): A Continuous Measure of DSM-IV Borderline Psychopathology. Journal of Personality Disorders: Vol. 17, No. 3, pp. 233-242.
RESULT ON MODIFIED OVERT AGGRESSION SCALE (MOAS)
A similar analysis can be conducted for the MOAS test (Modified Overt Aggression Scale). The euthymic level (pertaining to a normal mood in which the range of emotions is neither depressed nor highly elevated) for MOAS is 0 to 2.
The initial mean MOAS scales were:
- Placebo group : 11.3
- Low dosage : 18.7
- Moderate dosage: 22.6
A strong decrease in verbal & physical aggression (MOAS) was noted. The following mean changes per week were noted in Table 4:
- – 0.37 for Placebo,
- – 1.92 for Low dosage,
- – 1.82 for Moderate dosage
Converting to changes over the 8 medication weeks of the trial, one gets total mean reductions of
- – 2.96 for Placebo
- – 15.36 for Low dosage
- – 14.56 for Moderate dosage
Hence the final mean MOAS scores were:
- Placebo : 8.34
- Low dosage : 3.34
- Moderate dosage: 8.04
So by the end of the study, the Low-dosage group was close to normal level on the overt aggression scale.
MOAS level for the Moderate-dosage started nearly 4 points higher than Low-dosage group. It achieved a similar reduction on the scale of about 15 points. The Moderate-dosage group h remained higher than normal for overt aggression (verbal or physical).
The Placebo group ended in the same range as the Moderate-dosage group, yet it started much lower. Hence the Placebo group did not progress nearly as much as either the Low-dosage or Moderate-dosage groups.
A similar series of effects were observed on the Young Mania Rating Scale (YMRS) for irritability, aggression and verbal outbursts, which uses total scores on a different scale. Yet the euthymic (normal) level for YMRS is also viewed as 0 to 2.
The initial mean YMRS scales were:
- Placebo : 3.6
- Low dosage : 3.3
- Moderate dosage : 4.4
A strong decrease in verbal & physical aggression (YMRS) was noted. The following mean changes per week were noted in Table 4:
- – 0.88 for Placebo
- – 2.08 for Low dosage
- – 2.40 for Moderate dosage
Hence the final mean MOAS scores were:
- – Placebo : 2.72
- – Low dosage : 1.22
- – Moderate dosage : 2.0
On this fine scale, it is clear the YMRS score of Low-dosage group reached the normal range, the Moderate-dosage group was just on the border, while the Placebo group remained in a disorder range.
Notes For Extra Care:
- These were mean scores. If we look at individual patients, one could surely see variations across individuals. We do not have this data to discuss here.
- The plateau reached at week 6 is interesting, and the fact that the Low dosage group continued going down at week 9 and 10. Is it medically significant, or too small an effect to mention?
- We do not know whether the effects of the medication were long lasting. Did the patients level on the scales change after stopping to take the medication? Within how many days or weeks after stopping the medication did their levels change. Taking the analogy of a medication that reduces the fever of a patient, the fever was closer to normal at the end of the trial. Did the fever remain low after they stopped taking medication, or did it go back up later on?
ZYPREXA (OLANZAPINE) STUDY WITH (DBT) TALK THERAPY FOR BPD, 2000-2002
“Olanzapine Plus Dialectical Behavior Therapy for Women With High Irritability Who Meet Criteria for Borderline Personality Disorder: A Double-Blind, Placebo-Controlled Pilot Study”, by Marsha Linehan, Joshua McDavid, Milton Brown, Jennifer Sayrs, and Robert Gallop. J. Clin. Psychiatry (June 2008) Vol. 69, No. 6, pp. 999-1005.
The purpose of the study was to find out if the medication Olanzapine (commercial name: Zyprexa) reduced anger and hostility. The thought was that if patients’ emotions were less intense, they would benefit more from talk therapy.
This study ran from September 2000 to December 2002, and was performed with assistance from the University of Washington. Individuals were recruited from mental health clinics and by newspaper advertisements in the Seattle area.
For inclusion in the study, patients were required to be women aged 18 to 60 who met the following criteria:
- diagnosis of borderline personality disorder according to 2 structured interviews, the Personality Disorder Examination and the Structured Clinical Interview for DSM-IV Axis II Personality Disorders (SCID-II);
- met borderline personality disorder criterion for inappropriate anger on the SCID II; and
- scored 6 or higher on the irritability subscale of the Overt Aggression Scale Modified for Outpatients (OAS-M).
Individuals were excluded if they:
- had a present diagnosis of schizophrenia, bipolar I disorder, schizoaffective disorder, major depressive disorder with psychotic features or other psychotic disorder, mental retardation or seizure disorder, or a diagnosis of substance dependence in the last 6 months according to the DSM-IV ;
- had an episode of self-inflicted injury (including a suicide attempt) in the 8 weeks prior to the screening interview; or
- were pregnant, breastfeeding, or planning to become pregnant.
Researchers found 24 borderline women satisfying those criteria of BPD with high levels of irritability and anger. They provided 21 weeks of Dialectical Behavior Therapy to these patients and split them randomly into two groups of 12 women. Individuals in one group received a low-dose Olanzapine (Zyprexa). The other group received a placebo.
At treatment start, all patients were given 1 tablet per day of study medication. Each tablet contained either 5 mg of olanzapine or placebo. After the first week of treatment, the daily dose was adjusted upward or downward in response to perceived response and side effects by 2.5 to 5 mg with an allowed dosage range of 2.5 to 15 mg per day.
Patients were also provided with Dialectical Behavioral Therapy (DBT) for the 21-week duration of the study.
At weeks 7, 14 and 21, the researchers measured symptoms of patients with standardized psychological measures (i.e. tests).
24 people is a relatively small group, yet it’s a start. 21 weeks is a bit less than 6 months which is a relatively long time, but that’s what it usually takes for DBT therapy to take effect.
This study experienced the following attrition. Eight patients (33%: 4 olanzapine, 4 placebo) dropped out of DBT and consequently were not continued on medication. Overall attrition rate in the study was 33%, leaving 67% of the subjects in the study till completion.
Five of those (21% of total) also dropped out of the assessment sequence before the final assessment; 4 (17%) missed the time-2 and time-3 assessments and 1 missed only the time-3 assessment.
One patient, assigned to the olanzapine condition, dropped out due to pregnancy (at week 10). In addition, 1 patient assigned to the olanzapine condition was removed from the study at week 7 due to psychotic symptoms (she was not counted as a drop- out).
Researchers concluded that “Olanzapine may promote more rapid reduction of irritability and aggression than placebo for highly irritable women with borderline personality disorder.”
DETAILED RESULTS OF THE 2000-2002 OLANZAPINE STUDY
Researchers found that both the people who received the medicine and the sugar pill both felt less irritable and hostile, and were less likely to self harm. However, the group that received the actual medication felt better more quickly. A notable reduction of several measurements of physical and verbal aggression was reported after 7 weeks and 14 weeks.
Overt Aggression Scale Modified for Outpatients (OAS-M) and Hamilton Rating Scale for Depression (HAM-D) were given at pretreatment (time 0) and at weeks 7 (time 1), 14 (time 2), and 21 (time 3) by clinical interviewers, naive to the subjects’ treatment conditions.
As can be seen on the chart below, the Olanzapine group decreased notably in physical and verbal aggression after just 7 weeks, while the Placebo group either remained about the same or went up in scores.
For physical aggression: after 7 weeks, the Olanzapine group decreased from 5.5 before treatment to 2.0 at week 7, while the Placebo group online increased in physical aggression from 6.0 to 7.0. At week 14, the Placebook group decreased to 1.0 and caught up with the Olanzapine group still at 2.0. Then at week 21, the Placebo group went up to 2.0 while the Olanzapine group decreased to 0.0.
For verbal aggression: after 7 weeks, the Olanzapine group decreased from 20.0 before treatment to 9.0 at week 7, while the Placebo group increased in physical aggression from 23.0 to 27.5. Then at week 14, the Olanzapine group further decreased to 4.0, while the Placebo group progressed to 15.5. Then at week 21, both groups were at level 9.0, which was a higher level for the Olanzapine group than at week 14.
For irritability and overt aggression: after 7 weeks, the Olanzapine group decreased from 7.0 before treatment to 5.0 at week 7, while the Placebo group did not change in score and stayed at 6.5. At week 14, the Olanzapine group stayed at level 5.0, while the Placebo group progressed to level 5.5. Then at week 21, both groups progressed, Olanzapine to level 3.5 and Placebo to level 4.5, hence Olanzapine ended up progressing slightly more (3 levels, from 6.5 down to 3.5) then Placebo (2.5 levels, from 7.0 to 4.5).
Scores for intentional self-injury fluctuated for both groups, toward a notable reduction. Both groups started at level 33.3. After 7 weeks, the Olanzapine group decreased to 16.7, while the Placebo group decreased to level 25.0. At week 14, the Olanzapine group decreased to level 12.5, while the Placebo group remarkably was at level 0.0. Then at week 21, both groups went back up 12.5 points, to reach level 25.0 for the Olanzapine group and level 12.5 for Placebo.
High suicidality was the one criteria where therapy did better than medication. Placebo started at level 50.0, then dropped to 9.3 at week 7, then 0 at week 14 and ended at level 12. 5 at week 21, while the Olanzapine group went from 41.7 at the start, down to 33.4 at week 7, up to 37.5 at week 14, and finally 25.0 at week 21.
Depression was more aided by medication than by therapy. On the Hamilton Rating Scale for Depression, the score of the Olanzapine group went 20.4 down to 12.6 at week 21, while the Placebo group went from 19.3 to 15.4 at week 21. The reduction in depression score also happened a bit faster at week 7 for the Olanzapine group.
Researchers concluded that “Analyses indicated that both treatment conditions resulted in significant improvement in irritability, aggression, depression, and self-inflicted injury. Irritability and aggression scores tended to decrease more quickly for the Olanzapine group than for the placebo group.
To sum up, the benefit of the Zyprexa medication compared to the Placebo was that people felt better more quickly. Both groups (real medicine vs. sugar pill) eventually experienced less anger and irritability.
Feeling better more quickly and lesser over aggression were the main benefit of medication seen here.
Considering the small size of this study, one could not say for sure that this finding was significant until we do more studies with a larger number of subjects and controls. The small sample size may have limited the ability to detect significant results.
The above study on Quetiapine (Seroquel XR) was a follow-on study focusing entirely on effects of medication for BPD.
All comments welcome. Please let us know what you think, especially if you have direct experience with medication and BPD
— by Frederic Bien, Ph.D.
President, PDAN (Personality Disorder Awareness Network)