Literature Review
By Kessing LV, Forman JL, Andersen PK, Journal of Bipolar Disorder Feb 2010
Bipolar Disord 2010 Feb; 12(1):87-94.
Introduction:
- Studies suggest that BD is associated with increased risk of dementia, and the risk increases with every new affective episode
- Lithium may have neuroprotective abilities, reduce the risk of developing dementia by inhibiting glycogen synthase kinase-3, a key enzyme in the metabolism of amyloid precursor protein and in the phosphorylation of tau protein involved in the pathogenesis of Alzheimer’s Disease
Objective
- To investigate whether continued treatment with Lithium in patients with mania or bipolar disorder is associated with reduced risk of developing dementia, and whether continued treatment with anticonvulsants, antidepressants and antipsychotics were eventually associated with a decreased rate of developing dementia
Method
- The population sample was obtained by linking Danish population-based registers, Danish Medical Register, Danish National Hospital Register and Danish Psychiatric Central Register from January 95 to December 05, with a diagnosis of manic or mixed episode or bipolar disorder (ICD-10 codes).
- Patients with a diagnosis of dementia or schizophrenia were excluded.
- Study design: non-randomized,
- Statistical analysis: Poisson regression analyses were conducted with a diagnosis of dementia as outcome and with the number of prescriptions of lithium, anticonvulsants, antidepressants and antipsychotics, respectively at the end of study period (December 05). Data were analyzed accordingly including effect of various drugs severity of bipolar disorder, multiple prescription, sex and age on rate of dementia.
Result
- A total of suitable patients of 4856 were exposed to any of the drugs of interest at least once during the study period
- 2449 (50.4%) were exposed to lithium, 1781 (36.7%) to anticonvulsants, 4280 (88.1%) to antidepressants and 3901 (80.3%) to antipsychotic.
- A total of 4856 patients (4.5%) had a diagnosis of dementia.
- Most prevalent combinations of drugs include
i) Lithium, antidepressant and antipsychotic (20.7%),
ii) Antidepressants and antipsychotics (19.7%)
iii) All 4 types (16.8%)
iv) Anticonvulsants, antidepressants and antipsychotic (12.4%)
- Purchase of 2 or more prescription of Lithium was associated with significantly decreased rate of dementia compared with the rate during the period with one purchase prescription of lithium.
- For other types of drugs, the rate of dementia during multiple prescription periods did not differ significantly from the rate during prescription period one.
Discussion
- Among patients with bipolar disorder or mania, continued use of lithium was associated with a decreased rate of dementia. However, continued treatment with other medication was not associated with a significantly decreased rate of dementia
- There is a dose-response relationship between the severity of bipolar disorder and the risk of developing dementia. It confirms that the risk of developing dementia increases with the number of prior affective episodes.
Limitations
- Diagnoses were made by clinicians and were not research based. Validity of the ICD-10 diagnoses of bipolar disorder has not been investigated, clinicians may have heterogeneous ways of handling diagnoses
- A majority of patients included in the study used drugs from or more different types of 4 drug types, which might reduce the possibility of finding effects related to individual types of drugs.
- The finding of a decreased rate of dementia in relationship to continued use of lithium could be the result of a number of potential relationship:
a) Treatment with lithium may not be given to patients who the clinician believes will develop dementia later on e.g. patients with mild cognitive impairment as this might induce confusion. Continued treatment with lithium may not be given to patients with cardiac or renal insufficiency which may be related eventually to increased risk of dementia.
b) Patients who continue taking lithium for longer periods may presumably be good compliers with less alcohol use and a healthier lifestyle associated with decreased risk of alcohol use.
c) Anticonvulsants have been associated with neuroprotective abilities in much the same way as lithium, among others, including inhibition of glycogen synthase kinase-3. It is not possible from the present study to determine whether lithium has a protective effect against dementia due to increased neurogenesis or due to mood-stabilizing abilities preventing from recurrences of affective episodes or other treatment related factor.
Conclusions
Among patient with mania or bipolar disorder, continued use of lithium was associated with a decreased rate of dementia. In contrast, no significant associations were found between continued treatment with anticonvulsants, antidepressants or antipsychotics respectively. However, methodological reasons for these findings cannot be excluded due to non-randomized nature of data.
By Jeff J. Guo, Paul E Keck, Patricia K Corey Lisle, Hong Li, Dongming Jiang, Raymond Jang, Gilbert J. L’Italien
J Clin Psychiatry. 2006 Jul; 67 (7): 1055-61
Summary and Important Issues addressed in this Article:
Brief outline of how the study was conducted
- Introduction: Atypical antipsychotic agents with different mechanism of action from conventional antipsychotic have been widely adopted in the treatment of bipolar disorder since mid 1990s, they have a different spectrum of side effects, including weight gain, alterations in glucose metabolism, increased concentrations of blood cholesterol and lipids, myocarditis and cardiomyopathy
- Evidence has shown an association between some antipsychotics (clozapine, olanzapine and risperidone) and DM in patients with schizophrenia. DM is a known and infrequent adverse effect of olanzapine and risperidone.
- This is a retrospective, population-based case control study which recruited from multi-state managed care claims database from U.S during 1st January 98- 31st December 02 (5 calender year) to examine the association of atypical antipsychotics with diabetes mellitus in bipolar patients.
- 920 of Diabetes with BAD (identified using ICD-9 codes or diabetic medication) were matched with 5258 controls ( by age, sex and bipolar index month and year). People with Diabetes had a minimum 3-month exposure to any medications (as published reports show that drug-induced diabetes usually occurs with recent or current use of antipsychotic drugs) or at least 3 prescriptions for their bipolar or co-morbidity treatment.
- Patients with a diagnosis of depression only or schizophrenia or history of DM before the prescription of antipsychotic medications during the study period were excluded from this population.
- Cox proportional hazard regression was conducted to assess the risk of diabetes associated with antipsychotic use. 2 different groups were compared. 1st group included all patients except those receiving the specific atypical antipsychotic drug of interest. 2nd group (control) included patients taking conventional antipsychotic.
- Use of different medications was not mutally exclusive for one patient (e.g. lithium, anticonvulsants, atypical antipsychotics, antidepressants and conventional antipsychotics)
- Other factors which may affect results were matched and adjusted i.e. drugs affecting DM, psychiatric comorbidities (alcohol and SA, personality disorder, anxiety disorder, impulse-control behaviour) and medical comorbidites (Hypertension, obesity, arthritis, cerebral vascular disease, COPD, dyslipidemia and coronary heart disease.)
6 Important questions addressed in this article:
1) Does antipsychotic induce DM in Bipolar Affective disorder (BAD) patients?
a) The risk of developing DM was greatest among clozapine users (Exposure Hazard Ratio = 6.9), followed by zeldox (HR=4.6), olanzapine (HR=4.0), risperidone (HR= 3.5) and quetiapine (HR=2.3), conventional (HR=1.5) with adjusted model for age, sex, duration of bipolar
follow-up, use of medication and concomitant drugs. This is not adjusted for psychiatric and medical comorbidites.
b) When compared of atypical with conventional antipsychotics, the risk of DM was greatest among clozapine (HR=7.0) > olanzapine (HR=3.2)>risperidone (HR=3.4) > quetiapine (HR=1.8) > conventional with referral of HR =1.0, with controlling covariates of age, sex, duration of follow up, use of lithium, anticonvulsants, antidepressants or concomitant drugs, psychiatric and medical co-morbidities.
c) Both patients receiving conventional and atypical antipsychotic for BAD have an increased risk of DM after controlling for personal risk factors and concomitant drug use.
2) Can DM be due to use of antipsychotic itself or underlying BAD?
a) It is unclear how much DM in the study population might be due to the use of antipsychotics, as those with BAD would have poorer overall physical health, less healthy lifestyle, poorer access to health care services.
b) Literature indicated clozapine and olanzapine are more likely to be associated with DM (indicated by Diabetic ketoacidosis and atherogenic lipid profile) than other atypical agents. Possible mechanism via impairment of insulin resistance, which may occur because of a weight gain, a change in body fat distribution, by a direct effect on insulin-sensitive target tissues.
3) How does antipsychotic affect DM in BAD compared with schizophrenia?
a) Compared with published studies of patients with schizophrenia, the findings in BAD are similar or comparable. E.g. patients with schizophrenia had risk of DM associated with clozapine (HR 7.4-8.4) > olanzapine (HR 1.2-5.8) and risperidone (HR = 1.1-2.2).
b) The results indicating the risk of developing DM is statistically significant for BAD taking clozapine, olanzapine, risperidone and quetiapine after controlling for comorbities, personal risk factors and concomitant drug use.
c) Interestingly, the hazard ratio associated with ziprasidone was large (HR=4.6) without controlling for co-morbidities; then it became smaller (HR 1.7). This indicated that co-morbidities are critical covariates for assessing the risk of drug-induced diabetes.
4) What are the other factors / co-morbidities apart from antipsychotic use would affect risk of developing DM?
a) The risk of DM is also associated with Hypertension (HR= 2.7) > dyslipidemia (HR = 2.7) > coronary heart disease (HR 2.6) > obesity (HR= 2.2) > cerebrovascular disease (HR= 1.5).
b) Some antipsychotics like olanzapine, clozapine and risperidone are associated with weight gain, hyperlipidemia and hypertriglyceridemia, which are independent risk factors for heart disease.
c) It is likely that incident DM was associated with metabolic syndrome, as indicated by higher HRs for obesity, HT, CVD, CHD and dyslipidemia.
d) From this study, patient with impulse-control disorder or anxiety disorder had higher risk for diabetes. It is postulated that they have less healthy lifestyles, less medication compliance or poorer access to health care services.
5) What are the limitations of this study?
Method-wise:
a) This is a retrospective review which does not have direct information on the severity of bipolar disorder, socioeconomic class, lipid profiles, fasting glucose or BMI related to weight gain. Other confounding factors such as ethnicity and positive family history of metabolic disease were not adjusted.
b) It did not include all atypical antipsychotics such as amisulpiride and aripripazole.
c) It is not clear that different medications prescribed before the study period might be partially limited to the increased risk of diabetes i.e. clinician might have prescribed one drug over the other based on the clinical mood state. This is a compensatory attempt to reduce this confounding bias by adjusting for known concomitant drugs and co morbidities.
d) In addition, it is possible to underestimate the prevalence of DM due to limited time window and other mental services not billed or enrolled in this managed care organization.
6) What are the clinical implications?
a) This study provides useful information for disease management strategies in terms of selection of psychotropics and consideration of relevant co-morbidities for patients with bipolar disorder.
b) Atypical antipsychotics provide great benefit to a wide variety of people with psychiatric disorders but with adverse effects related to increased risk of metabolic disease such as obesity, diabetes and dyslipidemia.
c) Atypical antipsychotic like clozapine, olanzapine, risperidone and quetiapine are consistently associated with a clinically important increased risk of diabetes in bipolar patients. Metabolic complications are a major issue for patients receiving antipsychotic therapy.
d) The choice of atypical antipsychotic should consider the risk-benefit co morbid conditions, thus the propensity of an antipsychotic to induce diabetes is a critical considerations.
By Judd LL, Akiskal HS, Schettler PJ, Endicott J, Leon AC, Solomon DA, Coryell W, Maser JD, Keller MB.
From Arch Gen Psychiatry. 2005 Dec;62(12):1322-30.
Summaries and Important Issues addressed in this Article
Brief Outline of how this study was conducted:
- This is a prospective, comparative, longitudinal, naturalistic study which recruited patients from both inpatient and outpatient at 5 US centres
- 158 of BPI and 133 BP II were recruited using the Research Diagnostic Criteria, (with no organic psychiatric disorders), whom were FU for a mean of 15 years. The control group were 1817 relatives whom did not have any psychiatric disorder
- Trained professional raters interviewed pts every 6months for the first 5 yrs and then yearly thereafter
- Psychiatric symptom rating scale employed: LIFE(Longitudinal Interval FU Evaluation)-RIFT score(Range of Impaired Functioning Tool) was obtained for every month from 25mo to 5 yrs of FU and for the final month of FU in yrs 6-20
- Random regression analysis (using SAS MIXREG software) was used to model the relationship between LIFE-RIFT ratings of psychosocial impairment and monthly severity and polarity of affective symptoms
Four important questions addressed in this article concerning about the psychosocial disability:
1) Does psychosocial disability increase significantly with each increment of symptom severity?
a) In both BPI and BPII, each increase or decrease in depressive symptom severity is associated with a highly significant stepwise increase or decrease in psychosocial disability
b) In BPI, significant change in impairment as their level of manic symptom severity changes between mild subsyndromal symptom and hypomania or between hypomania and mania. No significant change between subsyndromal hypomanic symptom and asymptomatic state
c) In BPII, no significant change in psychosocial impairment when severity change from asymptomatic to subsyndromal hypomanic to hypomanic symptom. Only slight but non-significant improvement in psychosocial functioning when going from asymptomatic to subsydromal hypomanic symptom
2) Are symptoms and signs (s/s) in manic spectrum associated with more psychosocial disability
than s/s in depressive spectrum at corresponding level of severity?
a) In BPI, minor depression is associated with significantly more psychosocial disability than hypomania
b) In BPII, subsyndromal depressive symptoms are more disabling than subsyndromal hypomanic s/s and minor depressive s/s are more disabling than hypomanic symptom
c) Therefore, at each level of depressive symptom severity, psychosocial impairment is equal to or significant greater than the corresponding level of manic symptom severity in BPI and BPII
3) Is BP II less impairing than BP I at corresponding level of severity?
a) While experiencing hypomanic symptom, patient with BPII have significant better psychosocial function than patient with BPI
b) No difference in psychosocial impairment between BPI and BPII at any level of depressive symptom severity or asymptomatic status
4) Do patients return to good psychosocial functioning when they completely free of s/s and
compare with well-control group?
a) Patient with BPI and BPII at each level of depressive symptom severity are significantly more impaired than well control group
b) In BPI, each level of manic or hypomanic symptom is associated with significant greater impairment than in well comparison group
c) In BPII, no significant impairment in patients with hypomanic or subsyndromal hypomanic symptom
Conclusions and Implication in our daily practice
1) Depressive symptom in both BP subtypes are at least as disabling as, and sometimes more disabling than corresponding levels of manic or hypomanic symptom
2) There is a need for greater attention to depressive symptom in diagnosis and Treatment since most Tx has focused on Management of more dramatic manic episodes but not on depression
3) BPII is comparable to BPI in terms of psychosocial disability at corresponding levels of affective s/s severity
4) Only during hypomania in BPI patients are more significantly impaired than BPII patients owing to non-significant improvement in psychosocial function of BPII during subsyndromal hypomania vs asymptomatic
5) This highlights an important difference in psychosocial functioning between BPI and BPII. Hypomania in BPII presents not only a diagnostic challenge but also a therapeutic challenge
6) When pts with BPI and BPII are asymptomatic, their psychosocial function is good but not as good as that of well control
Disclaimer
