|Home | About | Journals | Submit | Contact Us | Français|
This is a multisite, 12-week, open-label trial of lamotrigine augmentation in 57 older adults (≥ 60 years; mean ± SD age = 66.5 ± 6.7 years) with either type I or type II bipolar depression.
Primary outcome measure was change from baseline on the Montgomery-Åsberg Depression Rating Scale (MADRS). Secondary outcome measures included Hamilton Depression Rating Scale (HAM-D), Clinical Global Impression-Bipolar version (CGI-BP), and the WHO-Disability Assessment Schedule II (WHO-DAS II). The Udvalg for Kliniske Undersøgelser (UKU) was used to assess side effects.
A total of 77.2% of the study subjects had bipolar I disorder. The mean (SD) lamotrigine dose was 150.9 (68.5) mg/day. There was significant improvement in the MADRS, HAM-D, CGI-BP, and in most domains on the WHO-DAS II. For patients for whom final MADRS score was available: 31 (57.4%) met remission criteria and 35 (64.8%) met response criteria. There were 19/57 (33.3%) who dropped out of the study prematurely, with 6 dropouts due to adverse events (4 cases of rash, 1 manic switch, and 1 hyponatremia). Two cases of rash were possibly drug related and were resolved with drug discontinuation. The most common UKU adverse effects were reduced sleep duration (n = 14, 24.6%), weight loss (n = 12, 21.1%), increased dream activity (n = 12, 21.1%), polyuria/polydipsia (n = 11, 19.3%), weight gain (n = 9, 15.8%), diminished sexual desire (n = 9, 15.8%), increased sleep (n = 9, 15.8%), lassitude/fatigue (n = 8, 14%), and unsteady gait (n = 8, 14%). No significant changes in electrocardiogram or laboratory tests were observed.
In bipolar depressed elders, lamotrigine was associated with improvement in depression, psychopathology, and functional status. There was a moderate number of adverse events, although relationship of adverse events (particularly falls) to study medication could not be clearly determined in this uncontrolled trial. Controlled studies are needed to further evaluate efficacy and tolerability of lamotrigine therapy in geriatric bipolar depression.
The growing population of elders is increasing the attention to bipolar disorder (BD) in late life (1, 2). The scant relevant literature highlights the challenges of treating BD in older adults, including greater medical comorbidity and lower tolerance to standard pharmacotherapies than in younger patients (3-6). Evidence specific to geriatric BD is urgently needed (7, 8). While depressive symptoms contribute to reduced quality of life among BD elders (4), there are no published prospective studies of the treatment of geriatric bipolar depression.
The challenge of managing bipolar depression in mixed-age populations has been highlighted by previous reports (9-12), and a limited number of medications have been shown to be efficacious for bipolar depression. In addition to the possible precipitation of mania or rapid cycling (13, 14), the addition of multiple psychotropic agents to stabilize mood and treat depression is of concern in older adults due to the risks associated with polypharmacy (15, 16).
Lamotrigine was approved by the U.S. Food and Drug Administration for the treatment of epilepsy in 1994, and for the maintenance treatment of BD in 2003. Meta-analysis and meta-regression of monotherapy randomized controlled trials (RCTs) suggest minimal to modest efficacy for lamotrigine in acute bipolar depression (17, 18). However, lamotrigine is widely used in clinical settings for the treatment of bipolar depression, typically in combination with other agents. A recent study of combined lamotrigine and lithium in bipolar depression demonstrated significant improvement and good tolerability in mixed-age adults (19).
A literature review and a secondary data-analysis of lamotrigine in older adults with BD (20, 21) suggest that lamotrigine is well tolerated and efficacious, with particular benefit against depressive relapse. The secondary analysis (20) focused on older adults (≥ 55 years) from two placebo-controlled, RCTs evaluating lamotrigine, lithium, and placebo in BD maintenance. There were 638 patients in the double-blind treatment phase including 98 older adults (mean age 61 years, SD = 6.0; range: 55–82 years). Lamotrigine significantly delayed time-to-intervention for depression compared with lithium, while lithium performed better than lamotrigine for time-to-intervention for mania. Side effects for both lamotrigine and lithium were generally time-limited and mild to moderate in intensity, including similar rates of skin rash (3% for lamotrigine, 5% for lithium).
Given the positive prospective findings in mixed-age patients and encouraging results in the secondary analysis with older BD patients, we conducted a 12-week, open label trial of lamotrigine in adults age 60 and older with type I or II bipolar depression, assessing its dosing, tolerability, and efficacy. We hypothesized that lamotrigine would be associated with improvement in depressive symptoms and would be well tolerated by these older adults with bipolar depression.
This study was conducted at five U.S. academic institutions between March 10, 2008 and March 11, 2010, in parallel with a National Institute of Mental Health-funded study of lithium versus divalproex for the treatment of geriatric mania (8). Non-demented outpatients age 60 and older with bipolar I or bipolar II disorder confirmed by the Structured Clinical Interview for Axis I DSM-IV Disorders-Patient Edition (22) were recruited from ambulatory services and the community. All subjects had to present with a score of 18 or higher on the 24-item Hamilton Rating Scale for Depression (HAM-D24) (23, 24) for inclusion in the study. Exclusion criteria included: chronic psychosis, contraindication or previous intolerance/resistance to lamotrigine, current substance dependence, rapid cycling, recent history of cardiovascular or cerebrovascular events, and high suicide risk. The study was approved by all local Institutional Review Boards and oversight included an external Data and Safety Monitoring Board that reviewed study progress and all adverse effects.
Depressive symptoms were evaluated with the Montgomery–Åsberg Rating Scale (MADRS) (25) and the HAM-D24; manic symptoms with the Young Mania Rating Scale (YMRS) (26); and global psychopathology with the Clinical Global Impression–Bipolar version (CGI-BP) (27, 28). General health/disability status was evaluated with the WHO-Disability Assessment Schedule II (WHO-DAS II) (29) and the Medical Outcomes Study Short Form General Health Survey (SF-12) (30). Adverse events were evaluated with the Udvalg fur Kliniske Undersøgelser (UKU) (31), the Simpson Angus Scale (SAS) (32), the Barnes Akathisia Scale (BAS) (33), and the Abnormal Involuntary Movement Scale (AIMS) (27).
Raters were centrally trained to achieve adequate reliability prior to conducting study assessments. Assessments were conducted at study entry screening, baseline (day 0), day 7, day 14, day 21 (week 3), week 4, week 6, week 9, and week 12 (termination). Subjects who discontinued active medication prematurely for any reason were requested to continue to provide study outcome assessments under naturalistic follow-up (i.e., while they received regular clinical care) for the remainder of the 12-week study period.
Body weight, vital signs, and spontaneous report of adverse events were assessed at each study visit. Safety laboratory testing (i.e., electrolytes, renal function, thyroid function, liver and lipid profile, and complete blood count with differential) was obtained at baseline, week 6, and week 12. A 12-lead electrocardiogram was performed at baseline and at week 12.
The Mini-Mental State Examination (MMSE) was used to assess global cognitive function (34) at baseline and week 12. More detailed cognitive testing, i.e., the Trail Making Test (35, 36), Stroop Test (37), Dementia Rating Scale 2 (38), and Hopkins Verbal Learning Test-Revised (HVLT-R) (39), was added to the assessment battery after the study was initiated, but this data was collected in only a minority of patients (n = 20). Medical illness burden was evaluated at baseline with the Cumulative Illness Rating Scale for Geriatrics (40).
The primary outcome was change from baseline on MADRS. Response was defined as 50% or greater reduction in MADRS score from baseline. Remission was defined as completing the 12-week trial with a final MADRS score < 10. Secondary outcomes included change from baseline on the HAM-D24, YMRS, CGI-BP, WHO-DAS II, SF-12, UKU, and body weight. Subjects who completed the study were asked to rate how satisfied they were with lamotrigine therapy (a great deal/a lot, somewhat, not at all) and if they intended to remain on lamotrigine.
During the screening period (1–14 days), study psychiatrists evaluated the feasibility of switching ineffective medications to lamotrigine (or initiating lamotrigine in patients who were not receiving any treatment). Lamotrigine (25 mg/day) was initiated and titrated as per manufacturerÕs recommendations up to a maximum of 200 mg/day (400 mg/day if the subject was on carbamazepine and 100 mg/day if on valproate). The study psychiatrist could reduce lamotrigine dosage if a subject experienced adverse effects.
Mood stabilizers (lithium, other anticonvulsants, clonazepam) that had been prescribed at stable dose for at least 90 days could be continued unchanged. Antidepressants and antipsychotics prescribed at stable dose for at least 30 days could also be continued unchanged. Antidepressants or antipsychotics that had to be discontinued were tapered and discontinued within 14 days of study baseline. Subjects could receive lorazepam for insomnia or agitation up to 3.0 mg/day for the first 21 days of the study and up to 0.5 mg/day thereafter. No additional psychotropic medications or new course of psychotherapy were permitted during the study.
Screened patients were compared to eligible subjects and the eligible subjects who refused participation were compared to subjects who initiated study medication. Descriptive statistics were calculated for baseline characteristics. Using paired t-tests, we performed a pre/post analysis of changes in the primary and secondary outcomes from baseline to last visit on lamotrigine therapy, including all subjects who received at least one dose of study medication. When 12-week data were unavailable, the final visit was based on the last date the subject followed protocol (and was known to be taking lamotrigine). Statistical analysis was performed using SAS software version 9.2 (41).
Adverse events were defined as increases of 2 points on UKU or a score of 3 on any one item (42). Mean lamotrigine final daily doses were calculated based on unadjusted doses and adjusted doses (corresponding to the last prescribed daily dose divided by 2 for subjects on carbamazepine and multiplied by 2 for those on valproate).
Figure 1 illustrates the study flow, with 332 patients screened, 57 subjects enrolled, and 38 study completers. Table 1 identifies the baseline characteristics of the 57 consented subjects. Age, gender, and race did not differ significantly between screened and eligible patients, and between eligible patients who refused or agreed to participate. Patients screened but not enrolled (n = 275) had a mean (SD) age of 65.4 (7.7) years; 57.8% were male, and 72.1% were white. Eligible patients who refused participation (n = 12) had a mean (SD) age of 72.3 (10.2) years; 58.3% were male and 100% were white.
Mean (SD) unadjusted final lamotrigine doses were 144.1 (78.0) mg/day (range: 25–400 mg/day) for all patients, 71.9 (38.8) mg/day for the 8 subjects on valproate, and 400 mg/day for the single subject on carbamazepine. After adjusting the dose for patients on valproate or carbamazepine, mean (SD) adjusted lamotrigine daily dose was 150.9 (68.5) mg/day (range: 25–200 mg/day). Five (8.8%) subjects received lamotrigine orally-dissolving formulation (ODT) when it became available (all at a single site). While the study was neither intended nor powered to compare the ODT and regular lamotrigine formations, the final daily doses were similar for the two formations.
Nineteen subjects (33%) terminated the study prematurely (i.e., they did not complete the 12 weeks) and 12 of these were followed naturalistically (i.e., ratings were collected while the patients were treated outside of the study setting). Reasons for dropout included: withdrawal of consent (n = 7, 12.3%), adverse events (n = 6, 10.5%), protocol deviation (n = 4, 7.0%), and lack of improvement or worsening in depression/anxiety (n = 2, 3.5%). Four (21.0%) of the 19 noncompleters dropped out prior to day 7, one (5.3%) prior to day 14, one (5.3%) prior to week 3, five (26.3%) prior to week 4, three (15.8%) prior to week 6, two (10.5%) prior to week 9, and three (15.8%) prior to week 12.
Mean depression scores (MADRS, HAM-D24) and mania scores (YMRS) for patients in active treatment on lamotrigine over the course of the study are presented in Figure 2. The mean (SD) [range] primary outcome measure (MADRS) decreased significantly from 25.3 (8.3) [9–48] at baseline to 9.8 (8.3) [0–38] at endpoint (t = 12.1, df = 53, p < 0.01). Similarly, the mean (SD) [range] HAM-D24 decreased significantly from 27.1 (6.7) [18–49] to 11.9 (9.8) [0–46], (t = 13.2, df = 53, p < 0.01). There was no significant change in the mean (SD) [range] YMRS score baseline: 4.8 (4.7) [0–21] or endpoint: 4.4 (4.6) [0–18], (t = 0.70, df = 48, p = 0.49). Final MADRS scores were available for 54 subjects, 31 (57.4%) of whom met remission criteria and 35 (64.8%) met response criteria. Table 2 shows changes from baseline to endpoint in global psychopathology. Higher baseline depression scores were associated with greater improvement from baseline while on lamotrigine (based on the MADRS, p = 0.0001; and based on the 17- and 24-item versions of the HAM-D, p = 0.06 and p = 0.04, respectively). However, baseline depression scores were not associated with responder or remitter status at endpoint.
Functional status, as measured by subscales of the WHO-DAS II, was significantly improved in most domains including self-care, life activities, ability to understand and communicate, participation in society, and getting along with people (see Table 2). The mental health but not physical health measure of the SF-12 was significantly improved. There was no significant change in MMSE. With respect to more extensive cognitive data obtained in a subset of subjects (n = 20), and with the exception of the HVLT-R, no significant differences pre/post were identified in the test administered. HVLT-R improvement from baseline could be attributed to practice effects.
Of the six subjects who dropped out due to adverse events, four presented with a skin rash, and one developed hyponatremia and experienced the emergence of manic symptoms. Based on review after study termination, two of these rash cases were judged not to be related to lamotrigine. Two of the other cases of rash were judged to be possibly related to lamotrigine and were resolved after discontinuation. There was no occurrence of Stevens-Johnson syndrome or toxic epidermal necrolysis.
Physical parameters were unremarkable: there were no significant changes in mean body weight, serum glucose, total cholesterol, or triglycerides. Changes in weight ranged from a decrease of 6.4 kg to a gain of 9.1 kg; one subject gained over 7% in body weight. No clinically significant EKG changes were observed. Numerical improvement on the AIMS and SAS were not statistically significant, while the improvement in BAS was statistically significant (see Table 2).
Among the 57 patients who received at least one dose of lamotrigine, the most common adverse events captured by the UKU were: reduced sleep duration (n = 14, 24.6%), weight loss (n = 12, 21.1%), increased dream activity (n = 12, 21.1%), polyuria/polydipsia (n = 11, 19.3%), weight gain (n = 9, 15.8%), dimished sexual desire (n = 9, 15.8%), increased sleep (n = 9, 15.8%), lassitude/fatigue (n = 8, 14%), and unsteady gait (n = 8, 14%). In addition, 18 (31.6%) subjects experienced a total of 28 falls. Five subjects (8.8%) had falls that required medical attention. Only one of the falls was judged to be related to lamotrigine.
Six subjects (10.5%) experienced emergence of manic symptoms (i.e., CGI-BP score ≥ 3 after the initiation of lamotrigine); three had manic symptoms of at least moderate intensity (i.e., CGI-BP score ≥ 4), including one who dropped out due to manic symptoms.
Of the 57 individuals in the trial, the overall rate of treatment satisfaction was 34/57 (59.6%). (Individuals who discontinued lamotrigine prematurely were assumed to not be satisfied with treatment). Most treatment completers (34/38 individuals, 89.5%) felt that they benefited a great deal/a lot from lamotrigine and stated that they would continue to take lamotrigine after the study, while four (10.5%) felt they either did not benefit at all or only somewhat. Thirty-four subjects stated that they would continue to take lamotrigine after the study concluded.
In this first prospective treatment study of geriatric bipolar depression, a 12-week trial of open-label lamotrigine was associated with significant improvement in bipolar depression, overall global psychopathology, and functional status. Drug tolerability in these BD elders was fair, although determination of relationship of adverse events (particularly falls) to study medication is difficult to determine without a control group.
Our findings are consistent with the limited literature on lamotrigine use in late-life BD. In a retrospective chart review, (13, 43) 20 older BD patients (mean age 63 years, range: 55–90) experienced an improvement in their mean CGI score from 4.8 to 2.0 during long-term lamotrigine treatment. In a small case series, three of five older BD patients (mean age 71.5 years) had > 50% improvement on HAM-D score from baseline (44). Our results are also consistent with a recent report by van der Loos et al. (19) on 124 mixed-age type I or II bipolar depressed patients [mean (SD) age: 46.4 (11.9)] randomized to receive either lamotrigine (n = 64) 200 mg/day or placebo (n = 60) as an adjunct to lithium for a period of eight weeks. In this study, lamotrigine was associated with significantly greater change in MADRS scores than placebo. Also, 51.6% of subjects in the mixed-age study on lamotrigine responded (defined as ≥ 50% improvement in MADRS score) versus 31.7% of those on placebo (19).
While any interpretation of relative response rates based upon our uncontrolled geriatric bipolar depression study must be considered within the obvious methodological limitations, the responder rate based upon change-from-baseline improvement in MADRS was 64.8%. Remission rate in bipolar depressed elders was 57.4%.
A particular concern with elderly populations who require pharmacologic management is reduced tolerability and increased propensity to adverse effects. In the mixed-age bipolar depression study by van der Loos et al. (19) the proportion of patients who completed the eight-week trial was 82% (52/64 of those on lamotrigine; 50/60 on placebo). Treatment related dropouts in the lamotrigine arm included 2/12 (16.7%) who dropped out because of lack of efficacy and 4/12 (33.3%) because of adverse effects. The 82% completion rate over eight weeks in the van der Loos et al. study is higher than the 66.7% rate we observed in our older subjects over 12 weeks. A secondary analysis of BD older adults (mean age 61.2 years) in a BD maintenance treatment study (20) found that 18% of lamotrigine-treated older adults prematurely terminated double-blind treatment compared to 13% of placebo-treated patients and 29% of lithium-treated patients. The most common side effects of lamotrigine in BD older adults from this secondary analysis were headaches (12%) and backache (12%).
Tolerance to lamotrigine in our elderly BD sample was fair. The overall proportion of dropouts due to adverse events was 10.5%. Among the 19 subjects who dropped out of study for any reason, treatment dropout due to lack of efficacy occurred in 10.5%, while dropouts due to adverse effects occurred in 31.5%. Two subjects appeared to have lamotrigine-related rash (3.5% of all subjects who received lamotrigine) which resolved quickly and without complications after discontinuation of lamotrigine. No Stevens-Johnson syndrome or toxic epidermal necrolysis were observed in our study; however the sample is too small to draw conclusions given the rarity of these adverse events. Emergence of hypo/mania occurred in approximately 10% of subjects but only one dropped out due to emergence of manic symptoms. In contrast to most BD medications with which substantial weight gain is likely to occur (15), we observed minimal weight gain in BD elders treated with lamotrigine.
The largest literature on safety and tolerance of lamotrigine in the elderly is with epilepsy patients (45-47). An 18-center study of 593 elderly patients (45) randomized to gabapentin (1500 mg/day), lamotrigine (150 mg/day), or carbamazepine (600 mg/day) found termination rates for adverse events of 12.1% for lamotrigine, 21.6% for gabapentin, and 31% for carbamazepine (p = 0.001). An RCT of lamotrigine versus carbamazepine in 185 elderly patients with epilepsy found adverse events leading to withdrawal occurring in 14% on lamotrigine versus 25% on carbamazepine (46). An RCT of lamotrigine versus carbamazepine in 150 elderly epilepsy patients (mean age 77 years) found dropouts due to adverse events were 18% with lamotrigine versus 42% with carbamazepine (47). The lamotrigine-treated patients also complained less frequently of somnolence (12% versus 29%, p < 0.05). Dropouts due to adverse events among lamotrigine-treated elders with epilepsy ranged from 12–18% in RCTs and lamotrigine appears relatively better tolerated than other anconvulsant drugs, particularly carbamazepine. However, elderly patients with BD may differ in tolerance to medication from elderly patients with seizures and generalizability from epilepsy trials is limited.
Falls and fall-related injuries are a special concern in all older patients. Just over 31% (n = 18) of our subjects fell during the study and five of them (8.8% of all subjects who received lamotrigine) required medical attention. However, only one fall was judged to be related to lamotrigine. In the absence of a placebo group, we cannot determine what the base rate for falls might be in elders with bipolar depression, and minimizing falls must be addressed both in research studies and when providing clinical care to these patients.
An important outcome in this study was functional impairment/disability status. Bipolar depressed elders experienced an improved ability to provide for self-care, participate in life activities, understand and communicate with others, and get along with others. Given that nearly half of the subjects lived alone, we expect these changes to promote independence and decrease the likelihood of nursing home placement that is very high in elders with serious mental illness. Functional improvement would also be expected to facilitate better self-management of comorbid medical conditions that were prevalent in these BD elders. However, in the absence of a control condition, we cannot determine whether the functional improvement we observed were due to lamotrigine or to personal interactions during the data collection and intervention visits.
As discussed above, major limitations of this study include its open-label and uncontrolled design as well as the fact that lamotrigine was used concomitantly with other psychotropic agents. The potential additional effect and tolerance of lamotrigine might well be dependent on the specific treatment to which it is added on. Also switch rate into mania might be influenced by the concomitant treatment. The low number of subjects included in this study does not allow for comparison in this respect. In addition, there may be a potential lack of generalizability from study subjects to BD elders treated in routine, noninvestigational clinical settings. However, study entry criteria were fairly broad with respect to medical comorbidity and reflect the burden of medical conditions that might be seen in many practice settings. An additional limitation of the study was the lack of detailed cognitive evaluation. While MMSE scores did not change in this well-educated group of patients it is possible that a finer-grained cognitive evaluation could have identified cognitive changes with symptom and functional improvement.
In conclusion, this study provides preliminary evidence suggesting that lamotrigine may be beneficial for some older adults with bipolar depression. However, controlled and larger trials are needed to confirm these preliminary results.
This study was supported by an investigator-initiated research grant from GlaxoSmithKline, Research Triangle Park, NC, USA.
Portions of this data have been presented at the 48th Annual Meeting of the American College of Neuropsychopharmacology, December 6-10, 2009, Hollywood, FL, USA and at the 163rd Annual Meeting of the American Psychiatric Association, May 22-27, 2010, New Orleans, LA, USA.
MS has received funding for research from GlaxoSmithKline, AstraZeneca, and Pfizer; has been a consultant for GlaxoSmithKline, Cognition Group, United Biosource, and ePharma Solutions; and has received speaker honoraria from Janssen-Cilag. AG has received funding for research from GlaxoSmithKline. RKAlJ has received funding for research from GlaxoSmithKline. LG has received funding for research from GlaxoSmithKline. RLG has received funding for research from GlaxoSmithKline, NIMH, and Janssen. MLB has received funding for research from the National Institute of Health (NIH); is a consultant for Medispin, Inc.; and has received honoraria from the University of Arkansas and Dartmouth Medical College. BHM has received funding for research from GlaxoSmithKline, the Canadian Institutes of Health Research, NIH, Bristol-Myers Squibb, and Wyeth. RCY has received funding for research from NIH, GlaxoSmithKline, AstraZeneca, and Janssen. KAC and TTH do not have any conflicts of interest to report.