Increase
No effect
Decrease
Lithium
Loop diuretics
Aspirin
Mannitol
Thiazide diuretics
Paracetamol
Aminophylline
Calcium blockers
Nefazodone
Theophylline
ACE inhibitors
Mirtazepine
AT II receptor blockers
Valproate
COX2 inhibitors
Lamotrigine
Trimethoprim
Amisulpride
Metronidazole
Ziprasidone
Spectinomycin
Risperidone
Levofloxacin
Quetiapine
Alprazolam
Carbamazepine
NSAIDs
Acetazolamide
Topiramate
Valproate
Stiripentol
Orlistat
Carbamazepine
Bupropion
Cholestyramine
Phenytoin
Colesevelam
Rifampicin
Carbapenems
Ritonavir
Chemotherapeutics (temporary)
Carbamazepine
Cimetidine (temporary)
Valproate (variable effect)
Efavirenz
Ciprofloxacin
Tramadol
Nevirapine
Claritromycin
Hypericum
Rifampicin
Danazol
Chemotherapeutics (temporary)
Diltiazem
Valproate (variable effect)
Erythromycin
Fluconazole
Fluoxetine
Fluvoxamine
Isoniazid
HIV protease inhibitor
Terbinafine
Verapamil
Stiripentol
Ketoconazole
Valproate (variable effect)
Trazodone
Lamotrigine
Valproate
Retigabine
Carbamazepine
Orlistat
Phenobarbital
Bupropion
Phenytoin
Olanzapine
Oxcarbazepine
Aripiprazole
Rifampicin
Risperidone
Lopinavir
Lithium
Primidone
4.5.2 Valproate
Valproate inhibits liver enzymes, e.g., CYP2C9, and increases the availability of several antipsychotics, benzodiazepines, moclobemide, and antidepressants. It reduces the metabolism of lamotrigine. The most probable explanation for this effect is hepatic competition between valproate and lamotrigine for glucuronidation [19].
4.5.3 Carbamazepine
Interactions are caused by induction of the liver enzymes CYP2C9 and CYP3A4. Carbamazepine increases the metabolism of haloperidol and valproate, and the neurotoxicity of lithium [20].
4.5.4 Lamotrigine
Valproate reduces the metabolism of lamotrigine and requires a 50 % dose reduction even in the start-up phase. Carbamazepine increases the metabolism of lamotrigine. Therefore, doubling the dose of lamotrigine is required when coprescribed with carbamazepine.
4.5.5 Antipsychotics
All antipsychotics can prolong the QTc interval.
4.6 Specific Pharmacotherapies and Electroconvulsive Therapy
4.6.1 General Recommendations
Research on pharmacotherapy in OABD is limited as older adults are often excluded from randomized controlled registration trials (RCTs) due to the increasing risk of medical complications with advancing age. In fact, the mean age of most RCTs investigating the efficacy and safety of atypical antipsychotic medications for the treatment of acute mania, for example, is approximately 40 years of age. Nevertheless, post hoc analyses of older adult cohorts from such trials support the use of similar first line treatments recommended for younger adults with bipolar disorder. However, due to changes in pharmacokinetic and pharmacodynamic properties, older adults are more vulnerable to side effects and drug interactions which increase with the number of medications prescribed [15]. Although a specific medication may have been tolerated for many years, new side effects may develop with aging or the onset of medical comorbidities. Clinical factors that may influence dose adjustment include age, medical comorbidity, and drug interactions. Polypharmacy is common in older adults with bipolar disorder. For example, 31.7 % of OABD are prescribed six or more medications [21]. Over the Counter (OTC) drugs and medications prescribed by other physicians may adversely interact with those medications used to treat symptoms of bipolar disorder (e.g., interaction between nonsteroidal anti-inflammatory medications (NSAIDs) and lithium) Interestingly, a prescription data study done with the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) data revealed that recovery can be achieved in the elderly, though more than one medication is often required regardless of age [22]. Here, we describe the limited evidence on pharmacotherapy for acute mania, bipolar depression, and maintenance therapy in OABD.
We will describe and discuss the available research on pharmacotherapy in older patients with bipolar disorder and conclude with specific evidence-based recommendations for each mood phase.
4.6.2 Lithium
Lithium is described in more detail in Chap. 7, and we will provide only an overview of lithium therapy in older adults in this section. Clinicians have been widely prescribing lithium in OABD for the management of acute mania but with much less frequency in recent years [23]. Lithium has been demonstrated to be effective in placebo-controlled studies of mixed-age patients [24]. There are eight studies of lithium in manic patients [25–32]. No randomized placebo-controlled studies have been published to date. Nonetheless, available evidence supports the use and efficacy of lithium in the treatment of mania in OABD. In terms of efficacy of lithium compared to other classes of medications, lithium has been shown to be equally efficacious. Although there are several case studies suggesting that elderly patients respond to lower lithium levels, most studies have required treatment at conventional lithium levels to achieve optimal results [28].
In twelve small retrospective case series and open-label studies of the effectiveness of lithium in OABD (Table 4.2), methodological differences between the studies preclude comparisons. The study samples were relatively small (2–80 patients). Four studies included patients only treated with lithium as maintenance therapy [33–36]. Two prospective, open-label studies concluded that lithium is more effective than placebo [35] and valproate [36]. There are eight studies of lithium in manic patients [25–32]. No randomized placebo-controlled studies have been published to date. All studies concluded that lithium is effective in reducing manic symptoms. Five studies concluded that lithium is also effective in treating bipolar depression [25, 27, 29, 30, 32]. From the limited available evidence, we conclude that the effectiveness of lithium in OABD is not different than in younger adults.
Table 4.2
Studies of lithium in OABD
N | Age (range) | Design | Dose-concentration | Duration (weeks) | Results (outcome measure) | |
---|---|---|---|---|---|---|
Van der Velde [26] | 12 | 67 (60–74) | R | Unknown | 2–156 | 33 % recovery of mania (global rating scale) |
Himmelhoch et al. [27] | 81 | 63.3 (55–88) | R | Unknown | 3–8 | 69 % response of depressive or manic symptoms (scale for clinical efficacy) |
Abou-Saleh and Coppen [25] | 7 | 57.1 | P | Unknown | 52 | 43 % remission of mania and depression (affective morbidity index) |
Murray et al. [33] | 25 | (60–78) | P | Unknown | 104 | Compared to younger patients clinical effect on maintenance treatment was independent of age |
Schaffer et al. [28] | 14 | 69 (65–77) | P | 900 mg–0.58 mEq/ml | >2 | 10 patients (71 %) had clinical improvement of manic symptoms |
Stone [34] | 48 | 70.3 (65–82) | R | Unknown | 26 | 40 % had no relapse after 6 months, no difference in recovery of mania between lithium users (n = 48) and non-lithium users (n = 44) |
Sharma et al. [29] | 4 | 68.5 (66–71) | P | 300–600 mg/day | 40–78 | Response in all rapid-cycling patients, 2/4 had a substantial recovery of depressive or manic symptoms |
Sanderson [30] | 41 (72) | 67.2 | R | Unknown | 5 | Duration of admission (mania and depression) was equal for lithium users (n = 41), valproate users (n = 20) and carbamazepine users (n = 11) |
Chen et al. [31] | 30 | 69.4 (>55) | R | Unknown | 2.3 | Mania improved in 67 % of lithium users (n = 30) versus 35 % of valproate users (n = 29). At therapeutic serum levels 83 % of lithium users improved (>0.8 mmol/l) versus 75 % of valproate users (65–90 micro g/l) |
Goldberg et al. [32] | 2 | 76; 71 | P | 600 mg/day–0.63 mmol/l; 900 mg/day–0.43 mmol/l | 3 | Remission of depressive and manic episodes after re-introduction of lithium following toxicity |
Sajatovic et al. [35] | 34 | 60.1 (55–82) | RCT | 750 mg/day 0.8—1.1 mmol/l | 76 | Lithium (n = 34) is more effective than placebo (n = 31) in prevention of relapse into (hypo)mania, 29 % dropped out |
Geddes et al. [36] | 27 | (>53) | P | 0.4–1.0 mmol/l | 104 | Lithium is as effective (n = 27) as the combination lithium–valproate (n = 22) and more effective than valproate alone (n = 31) in preventing relapse |
4.6.2.1 Clinical Recommendations for Lithium Therapy
Therapeutic serum levels can be achieved with a 25–50 % lower dose compared to younger adults [35]. Lithium serum levels of 0.4–0.6 mmol/l can be effective, although levels of 0.8 mmol/l or higher may be needed for therapeutic efficacy. Lithium may also be effective at lower serum levels for certain individuals. Balancing toxicity and clinical efficacy is a great challenge when using lithium for the treatment of older adults with bipolar disorder.
Due to the reduced renal function with normal aging, the elimination half-life of lithium can be prolonged to 36–48 h compared to 24 h in younger adults. Therefore, steady state of lithium serum levels may not be reached for up to 10 days [37]. Dosing should reflect the altered pharmacodynamics and pharmacokinetics of older adults resulting in the need for significantly lower doses. As always, treatment depends on effectiveness and tolerance, and lithium should be titrated accordingly. Unfortunately, we still do not have consensus or clear guidelines for the appropriate therapeutic range of lithium in older adults. As a rough guideline, half the dose used in younger adults should be adequate and much safer. Very careful monitoring of clinical status and serum levels are indicated, even more so in older adults as maintenance treatment with lithium as practiced in modern care with regular measurements of renal function and lithium serum levels is associated with less decline of renal function [38, 39].
In older patients, lithium serum levels can rapidly elevate due to dehydration (perspiration, diarrhea, or insufficient fluid intake) and drug interactions with commonly used medications in older adults (e.g., diuretics, ACE- inhibiters, and NSAIDs).
As in younger adults, lithium toxicity in older adults is diagnosed by careful assessment of patients’ clinical symptoms and not solely quantified by serum lithium levels. Neurological and cardiovascular comorbidity and polypharmacy increase the risk of side effects, drug interactions, and therefore toxicity [40–42].
4.6.3 Anticonvulsants
The available research on anticonvulsants in older adults with bipolar disorder is summarized in Table 4.3. Older patients are more susceptible to side effects of anticonvulsants than younger adults with an increased risk for delirium and falls [43].
Table 4.3
Studies of anticonvulsants in OABD
N | Age (range) | Design | Dose-concentration | Duration (weeks) | Results (outcome measure) | |
---|---|---|---|---|---|---|
Valproate | ||||||
McFarland et al. [46] | 6 | 66 (56–74) | R | 500 mg/day 50–150 microg/ml | 4 | Significant improvement of manic symptoms after valproate addition in therapy-resistant patients |
Sharma et al. [29] | 4 | 68.5 (66–71) | P | 1000–1250 mg/day | 40–78 | Combination of lithium and valproate results in response in all rapid-cycling patients, 2/4 had significant recovery |
Risinger et al. [47] | 4 | 70 (65–73) | R | 1000–1500 mg/day 50–75 microg/ml | 2–4 | Recovery of manic symptoms in all patients |
Puryear et al. [48] | 7 | 70 (63–81) | R | 1000 mg/day (100–1750) 57 nanog/ml (34–82) | >1 | Significant improvement of mainly manic symptoms |
Kando et al. [49] | 24 | 71.3 | R | 743 mg/day(250–2000) 53 mg/l (11–102) | >2 | Effective in 62 % of manic patients with adequate treatment |
Schneider and Wilcox [50] | 4 | 74.8 (65–81) | R | 52–115 mg/l | 72–156 | Remission after addition of valproate to lithium therapy in manic rapid cyclers |
Sanderson [30] | 20 | 67.2 | R | Unknown | 4 | Duration of admission was equal for lithium users (n = 41), valproate users (n = 20) and carbamazepine users (n = 11) |
Niedermier and Nasrallah [51] | 23 | 67 (60–86) | R | 1.029 mg/day (500–2250) 72 mg/l (36–111) | >1 | 87 % response (CGI) in manic, depressive, and mixed episodes |
Noaghiul et al. [52] | 21 | 71 (60–82) | R | 1.405 mg/day 72 mg/l | 1–7 | 19 patients (90 %) had significant clinical recovery of mania (CGI) |
Chen et al. [31] | 29 | 71.2 (>55) | R | Unknown | 2.3 | Mania improved in 67 % of lithium users (n = 30) versus 35 % of valproate users (n = 29). At therapeutic serum levels, 83 % of lithium users improved (>0.8 mmol/l) versus 75 % of valproate users (65–90 microg/l) |
Mordecai et al. [53] | 6 | 70.8 (64–75) | R | 250–1000 mg/day 23–51.7 meq/ml | 2–43 | 3 patients stabilized with valproate monotherapy 2 lithium users improved after valproate addition Both manic and depressive symptoms |
Geddes et al. [36] | 31 | (>53) | P | 750–1250 mg/day | 104 | Lithium is as effective (n = 27) as the combination lithium–valproate (n = 22) and more effective than valproate alone (n = 31) in preventing relapse |
Carbamazepine | ||||||
Cullen et al. [44] | 3 | 57 (48–72) | R | 200–1200 mg/day 2236 Umol/l | >1 | 2/3 patients recovered from therapy-resistant melancholic depression |
Sanderson [30] | 11 | 67.2 | R | Unknown | 4 | Duration of admission was equal for lithium users (n = 41), valproate users (n = 20) and carbamazepine users (n = 11) |
Lamotrigine | ||||||
Robillard and Conn [55] | 5 | 71.5 (65–85) | P | 25–100 mg/day | 12 | 50 % reduction of depressive symptoms (HDRS) in 3 rapid cyclers |
Sajatovic et al. [58] | 33 | 60.1 (55–82) | RCT | 100–400 mg/day | 76 | Lamotrigine (n = 34) is more effective than placebo (n = 31) in preventing relapse (mania/depression) 18 % dropped out |
Sajatovic et al. [56] | 57 | 66.5 (60–90) | P | 150.9 mg/day | 57.4 % remission (MADRS) 64.8 % response 33 % dropout | |
Gabapentin | ||||||
Sethi et al. [57] | 7 | 72.7 (59–90) | R | 1028.5 mg/day (600–1200 mg/day) | All patients experienced improvement in manic symptoms. On gabapentin in combination with antipsychotic medications, and in 1 case in combination with valproate |
4.6.3.1 Carbamazepine
In two studies, the effectiveness of carbamazepine was evaluated in OABD. In one study, carbamazepine was effective in 2 of 3 treatment-resistant patients [44]. The other study showed that duration of admission was not different in patients on lithium (n = 41), valproate (n = 20), or carbamazepine (n = 11) [30]. In a subgroup analysis of a double-blind study in younger adults with bipolar disorder assessing both carbamazepine and lithium individually as an anti-manic drug, both drugs were shown to be equally effective [45].
4.6.3.2 Valproate
Clinicians have increasingly prescribed valproate over the years for the treatment of bipolar disorder. Retrospective studies have suggested its anti-manic effects in OABD [29–31, 46–53]. There are 12 small studies (4–31 patients) on the effectiveness of valproate in treating OABD, none are randomized placebo-controlled.
In eleven studies, valproate was effective in treating mania [29–31, 46–53], and in five studies, it was effective as add-on in treatment resistance [29, 36, 46, 50, 53]. Valproate monotherapy has been reported in several studies to improve manic symptoms in almost all OABD, with the duration of treatment ranging from 1 to 7 weeks [47, 48, 52]. The National Institute of Mental Health has also funded a multisite, randomized controlled trial of lithium and valproate for the management of acute mania, with the aim of assessing the tolerability and efficacy of lithium and valproate in acute mania, hypomania, or mixed episodes in elderly bipolar patients [54].