Strategies for prevention and management of second generation antipsychotic-induced metabolic side effects

BACKGROUND

Metabolic disturbances and metabolic syndrome (MetS) are highly prevalent in patients with severe mental disorders^1–4 and play an important role in the development of atherosclerosis.⁵ MetS includes a cluster of risk factors, such as abdominal obesity, high blood pressure, atherogenic dyslipidemia, insulin resistance and hyperglycemia that interact synergistically to increase the risk for coronary heart disease (CHD) and diabetes mellitus.⁶ Schizophrenia itself may be a risk factor for MetS and metabolic disturbances, but there is also increasing evidence that use of antipsychotics, particularly second-generation antipsychotics (SGAs), have metabolic consequences that contribute to the risk.³⁷

In current clinical practice, “atypical” or second-generation antipsychotics (SGAs) have become mainstay in the management of schizophrenia, and other psychiatric conditions in children, adolescents and adults.⁸ Currently, there are ten SGAs available in the US: aripiprazole (Abilify), asenapine (Saphris), clozapine (Clozaril), iloperidone (Fanapt), lurasidone (Latuda), olanzapine (Zyprexa), paliperidone (Invega), quetiapine (Seroquel), risperidone (Risperdal), and ziprasidone (Geodon). They vary in their efficacy, formulation, receptor-binding properties, and type and extent of side effect profiles.³ Aside from clozapine that can be associated with agranulocytosis,⁹¹⁰ and thus is reserved for treatment-resistant schizophrenia, severely-ill refractory patients or patients with schizophrenia who are at high risk for suicide, SGAs have become first-line agents for their indicated uses and have largely replaced first-generation antipsychotics (FGAs). In general, SGAs carry lower risk for tardive dyskinesia and extrapyramidal symptoms (EPS) compared to FGAs when administered at clinically therapeutic doses. However, varying degrees of weight gain, atherogenic dyslipidemia, hyperglycemia, and even MetS are clinically-important metabolic side effects of SGAs.^11–20 The use of SGAs is also associated with new onset of type 2 diabetes mellitus (T2DM), worsening of pre-existing type 1 (T1DM) or T2DM, and rare cases of diabetic ketoacidosis and hyperosmolar-hyperglycemic state.^21–23 Second generation antipsychotic induced dysmetabolic effects are usually evident within the first 12 weeks after initiation of medication.²⁴ The extent of weight gain and metabolic disturbances and consequent risk for MetS depends on the specific SGA prescribed along with patient-specific factors.^11–20 The metabolic effect appears to be an intrinsic effect of individual SGAs, and there appears to be a dose-dependent relationship between clozapine or olanzapine use and metabolic complications.²⁵²⁶ Preliminary evidence suggests a dose-response relationship between clozapine and olanzapine serum concentrations and weight gain;²⁵ however, the association between antipsychotic dose and metabolic outcomes in not clear. Yood et al. followed adult patients (mean age: 44) exposed to aripiprazole, clozapine, olanzapine, ziprasidone, risperidone, and ziprasidone for at least 45 days between January 2002 and March 2005.²⁶ Newly-treated diabetes patients were identified using pharmacy data indicators of initiation of anti-diabetic therapies. They reported the risk of T2DM for patients on olanzapine, quetiapine and risperidone appeared to be dose-dependent. Olanzapine demonstrated increased risk at the second tertile dose (5–<10 mg/day) (HR: 1.7, 95%CI: 1.0–3.1) and third tertile dose (≥10 mg/day) (HR:2.5, 95%CI: 1.4–4.5). Quetiapine (>150 mg/day) and risperidone (≥2 mg/day) exhibited elevated risk only at third dose tertile. No T2DM risk was associated with lower doses of quetiapine (≤150 mg/day) or risperidone (≤2 mg/day). Unlike the three aforementioned medications (olanzapine, quetiapine, risperidone), ziprasidone and aripiprazole did not exhibit a dose-dependent association with T2DM or an elevated rate of new-onset T2DM.

Antipsychotic-induced weight gain is a very common and serious problem in patients with psychotic disorders. It is associated with increased central obesity,²⁷ and weight gain and consequent excess of visceral fat can lead to exacerbation of co-morbid conditions, development of insulin resistance, hypertension, and lipid abnormalities as well as patient noncompliance.²⁸ This emphasizes the need for prevention and management of weight gain associated with antipsychotic treatment. In addition, weight/obesity-independent drug mechanism(s) for hyperglycemia, T2DM and dyslipidemia were suggested.^29–33

Clozapine and olanzapine represent the highest offenders causing metabolic disturbances, while aripiprazole and ziprasidone seem to have the most favorable metabolic profiles.²¹⁵¹⁶ Based on available evidence, it appears the newer SGAs: asenapine, iloperidone, lurasidone and paliperidone, possess lower propensity for significant changes in weight, glycemic and lipid profiles.^17–1923 However more long-term data needs to be compiled in order to comprehensively evaluate their metabolic safety. The relative potential for commonly-used SGAs to induce individual metabolic disturbances is summarized in Table 1.

Table 1. Relative incidence of metabolic abnormalities with selected SGAs²⁰

View Fullscreen

Prevention and Management of SGA-Induced Metabolic Disturbances

The treatment and treatment goals should be individualized and considered in the context of the patient's psychiatric condition, contraindications and precautions, metabolic profile, and other patient-specific factors. There is no current consensus on the prevention and management of weight gain and other SGA-induced metabolic disturbances. Several strategies and recommendations for limiting the possibility or extent of antipsychotic-induced metabolic disturbances have been proposed (Table 2) and can be broadly divided into monitoring, choice of antipsychotic, non-pharmacologic and pharmacologic interventions.

Table 2. Prevention/management strategies for SGA-induced metabolic disturbances ¹⁶34–3540–45⁵¹54–62^73–76

View Fullscreen

Monitoring and initial SGA Selection Strategies

In 2004, the American Psychiatric Association (APA), the American Diabetes Association (ADA), the American Association of Clinical Endocrinologists (AACE), and North American Association for the Study of Obesity (NAASO) jointly developed consensus guidelines that recommend scheduled screening for and monitoring of the metabolic problems on commencement of a SGA and periodically thereafter as shown in Table 3.¹⁶

Table 3. Adult monitoring protocol of patients on SGAs¹⁶

View Fullscreen

It should be noted that elevated post-load glucose precedes elevated fasting glucose in the development of T2DM.³⁶ Therefore, FPG testing alone might be less reliable, missing cases of impaired glucose tolerance. A 75-g oral glucose tolerance test (OGTT) is a more sensitive indicator for glucose tolerance in early diagnosis of diabetes.³⁷³⁸ However, this test is not routinely performed in patients on SGAs and is not recommended by 2004 consensus guidelines.¹⁶ This could be due to the fact that OGTT requires more cooperation and compliance from patients and thus clinicians might struggle to successfully complete the testing. Additionally, when fasting samples are difficult or impossible to obtain from a patient, glycosylated hemoglobin (A1C) can be considered as a screening test for elevated plasma glucose and diabetes since it does not requiring fasting. Current recommendations from the ADA include the following criteria for prediabetes: A1C 5.7%–6.4% or FPG 100–125 mg/dL or 2-hr plasma glucose (OGTT) 140–199 mg/dL.³⁹ Criteria for T2DM diagnosis are A1C ≥ 6.5% or FPG ≥ 126 mg/dL or 2-hr plasma glucose (OGTT) ≥200 mg/dL

Appropriate selection of an antipsychotic is one of the important steps in prevention of metabolic complications for psychiatric patients. Baseline screening information (Table 3) should be considered when making the antipsychotic choice. When initiating an antipsychotic, the patient, caregivers and family should be informed regarding drug-induced metabolic risks and symptoms of emergent diabetes and diabetic ketoacidosis so they would know what to expect and for what they should watch. It is essential that appropriate ongoing metabolic monitoring should be included in the patient's care plan (Table 3).

NON-PHARMACOLOGIC INTERVENTIONS

Non-pharmacologic interventions such as cognitive behavioral therapy (CBT) and nutritional counseling, alone or in combination with an exercise program, showed promise as they were helpful in some patients with antipsychotic-induced weight gain,^40–44 mild or moderate dyslipidemia,⁴³⁴⁵ and/or hyperglycemia.⁴³ They also lack side effects and drug interactions that can be associated with adjuvant pharmacologic treatments. It was suggested that prevention studies with individual psychoeducational programs including diet and/or physical activity seemed to have the highest impact on antipsychotic-induced weight gain.⁴² Therefore, early education and counseling on physical activity and healthy diet for weight maintenance/loss should be promoted shortly after initiation of SGA therapy and referral to a dietitian, psychologist or weight loss program should be considered when appropriate. This is even more important for those who are initiated on a SGA with high propensity for weight gain and those who are at higher risk for antipsychotic-induced weight gain including young individuals, especially children and adolescents, antipsychotic-naïve patients, underweight patients at the beginning of treatment, and those who experience early significant weight gain.^3246–50 The most recent meta-analysis expanded upon two prior publications that reported similar results indicating significant benefits of non-pharmacologic interventions for weight and BMI reduction.⁴⁰⁴¹ Recently, Caemmerer et al. expanded on these studies by including additional RCTs with non-pharmacologic interventions, and analyzing their effect on other cardiometabolic indices.² This meta-analysis of 17 RCTs with 810 participants (mean age: 38.8 years; 85.6% schizophrenia-spectrum disorders) reported that non-pharmacologic interventions led to a significant reduction in weight (−3.12 kg; CI: −4.03, −2.21, p<0.0001) and BMI (−0.94 kg/m²; CI: −1.45, −0.43, p=0.0003) compared with control groups. In addition to weight-loss benefits, patients also experienced significant decreases in waist circumference, % body fat, glucose, insulin, total cholesterol, LDL-cholesterol, and triglycerides. No benefit was observed in the values of HDL-cholesterol and systolic blood pressure. The loss of 3.12 kg in the non-pharmacologic intervention group as reported above is comparable to weight reduction observed with metformin (2.94 kg) and topiramate (2.52 kg) use as reported by Maayan et al,⁵¹ keeping in mind the limitations that exist in making a direct comparison between different studies. Currently, there is a large heterogeneity among different studies investigating an effect of various non-pharmacologic interventions on antipsychotic-induced metabolic disturbances, and more long-term data needs to be compiled in order to comprehensively evaluate their short-term and long-term benefit. There seems to be merit in promoting initiation of and adherence to healthy lifestyle interventions among patients being treated with SGAs.

PHARMACOLOGIC INTERVENTIONS

If non-pharmacologic interventions are not effective, the clinician might choose to manage metabolic issues by: 1) changing from polypharmacy to monotherapy if patient is on ≥ 2 antipsychotics, 2) changing antipsychotic medication to one with lower potential for causing weight gain or other metabolic abnormalities, or 3) treating the patient with adjuvant pharmacological agent(s). Before using adjuvant medication, a change to monotherapy or a different SGA should be considered first.

Current treatment guidelines discourage antipsychotic polypharmacy (use of ≥2 antipsychotics) unless an adequate trial with antipsychotic monotherapy proved to be unsatisfactory, although some combinations, particularly aripiprazole addition to a stable dose of olanzapine, were reported to significantly reduce weight gain, total triglycerides, very low-density lipoprotein cholesterol, and adiponectin levels.⁵²⁵³ However, more data are needed before making this a general recommendation. Simple prescribing regimens with antipsychotics should be associated with fewer side effects including metabolic disturbances, lower cost, and improved patient adherence. The benefit of changing from antipsychotic polypharmacy to a single antipsychotic agent should be carefully weighed against risks associated with psychiatric illness and goals for patient outcomes. Several studies highlighted that, in many cases, antipsychotic polypharmacy is avoidable,⁵⁴⁵⁵ and a switch to monotherapy is associated with reduction of BMI.⁵⁵ Essock et al. studied 127 adult outpatients with schizophrenia on two antipsychotics, randomized to either stay on polypharmacy or switch to monotherapy by discontinuing one antipsychotic.⁵⁵ This study found a switch to monotherapy can be accomplished successfully for the majority of patients and 69 percent of those switched to monotherapy remained on monotherapy for the entire six-month study period. On average, the switch to monotherapy resulted in a statistically significant loss of body mass (−0.5 kg, SD:2.32) compared to those remaining on polypharmacy (+0.3 kg, SD:2.31) with no worsening of symptoms and no increase in rate of hospitalization among those switching to monotherapy.

The APA/ADA/AACE/NAASO Consensus Guidelines recommend, as one possible option for a patient experiencing a weight gain >5% of initial weight, worsening hyperglycemia, and/or dyslipidemia, changing to a SGA with lower risk for metabolic side effects.¹⁶ The switch to another antipsychotic with better metabolic side effects is also endorsed in the 2010 Outcomes Research Team (PORT)³⁴ and the 2012 World Federation of Societies of Biological Psychiatry (WFSBP) guidelines.³³ This strategy was demonstrated to improve weight and lipid profile benefits with minimal risks of worsening symptoms or relapse.^162456–61 A multicenter, double-blind, twelve-week RCT by Stroup et al. with 173 participants demonstrated that changing from olanzapine to aripiprazole resulted in a statistically significant weight loss after antipsychotic switch.⁶⁰ At week 16, weight decreased significantly (p<0.001) in the aripiprazole-treated group (Δ: −1.8 kg, n=85) compared to the olanzapine-treated group (Δ: +1.41 kg, n=88). Improvements in total cholesterol, TG and HDL cholesterol levels were also reported. Recently, Chen et al. studied the effectiveness of switching antipsychotic drug treatment to aripiprazole or ziprasidone to improve metabolic profile and atherogenic dyslipidemia in a 12-month, prospective, open-label study.⁶¹ They concluded patients changing to either ziprasidone or aripiprazole when they were experiencing SGA-induced metabolic side effects had the potential to improve weight, BMI, TG and TG/HDL indices.

It needs to be stressed that before considering a change to another SGA, one needs to weigh the benefit of switching to ziprasidone, aripiprazole or another lower-metabolic risk SGA against the potential risk for loss of efficacy, destabilizing a stable patient, relapse, interaction, contraindication, non-metabolic side effects such as QTc prolongation and EPS.⁶³⁶⁴ It is recommended that a new SGA is cross-titrated rather than old agents being abruptly discontinued.¹⁶ Pre-switch SGA should be gradually tapered with a similarly gradual initiation and dose titration of ziprasidone or aripiprazole, and discontinued when the new SGA is in a clinically effective dose. To ensure a patient's clinical stability and medication tolerability, careful monitoring and follow-up should be required during antipsychotic switching.^62–64

A meta-analysis of 32 RCTs (n=1,482) with 15 pharmacologic agents (amantadine, dextroamphetamine, d-fenfluramine, famotidine, fluoxetine, fluvoxamine, metformin, nizatidine, orlistat, phenylpropolamine, reboxetine, rosiglitazone, sibutramine, topiramate, and metformin+sibutramine) found only five medications that significantly reduced risk for antipsychotic-induced weight gain compared to a placebo.⁵¹ However, out of those medications, only metformin and topiramate are currently available in the United States. Metformin is an oral antidiabetic agent that decreases hepatic gluconeogenesis and increases peripheral insulin sensitivity. The use of metformin in diabetic patients was associated with improved glycemic control and moderate weight loss.⁶⁵ The antidiabetic effect of metformin is particularly interesting because of its dual effects of decreasing body weight and improving insulin sensitivity. It has been suggested that metformin's effect on body weight is due to reduction in appetite,⁶⁶ that can be secondary due to taste disturbance, nausea or increase in anorexic glucagon-like peptide 1.⁶⁷ Multiple RCTs with metformin in doses ranging from 750 to 2500 mg/day for periods from 12–16 weeks demonstrated short-term, modest loss of olanzapine-induced weight gain in patients with schizophrenia.^68–71 It should be noted that contradictory studies exist that reported no difference between metformin and placebo groups regarding olanzapine-induced weight gain during similar trials.⁷⁰⁷² Generally metformin was well-tolerated and only limited mild gastrointestinal symptoms were reported. Recently, Maayan et al. reported from their RCT meta-analysis findings of significant weight reduction in metformin groups compared to placebo (mean:−2.93kg, p<0.003) in patients treated with olanzapine, risperidone, quetiapine, and a few SGA combinations.⁵¹ Based on existing data, metformin might be a particularly promising option for antipsychotic-induced weight gain and glucose metabolism dysregulation in the subpopulation experiencing first episode of schizophrenia.⁵³⁶⁸⁶⁹ In addition, it was reported that metformin treatment (750 mg/day) in this subpopulation may bolster weight reduction and decrease FPG, and insulin resistance index (IRI) attained through healthy lifestyle changes.⁶⁹ Hasnain et al. proposed using a ”point system” approach to facilitate decision-making about when to consider a metformin trial.⁷³ This system is based on various characteristics such as family history of diabetes or CVD, BMI, waist circumference, current hyperglycemia and/or impaired glucose tolerance. Each item is assigned a score with individual items ranging from 1–5 points. Patients scoring ≥4 points on the point system are identified as potential candidates for a metformin trial if no contraindications exist. In addition, they suggested discontinuing metformin if benefits are not noted after 6–8 months of treatment.

Another agent showing some promise in management of weight gain related to SGAs is the anticonvulsant, topiramate; however, current data are limited. Three RCTs found topiramate treatment with doses between 100–200 mg/day for 12-weeks to be effective for weight and BMI reduction in patients also treated with clozapine or olanzapine.^74–77 There was a dose-dependent relationship between topiramate in both benefit and side effects. More patients experienced a weight loss of at least five percent in the topiramate 200 mg/day-group than in other groups. In addition, there were significantly more incidents of paresthesia with topiramate 200 mg/day (~60%) compared to topiramate 100 mg/day (25%) and placebo (10%).⁷⁴ Maayan et al. found that adjunct treatment with topiramate resulted in a mean weight change of −3.95 kg (95%CI:1.77–6.12) in SGA-treated psychotic patients.⁵¹ In general, topiramate was associated with more reports of side effects including paresthesia, dizziness, psychomotor slowing, drooling and headache.⁷⁶⁷⁸ In addition, patients taking topiramate experienced worsening of psychiatric symptoms compared to placebo.⁷⁴ The mechanism by which topiramate induces weight loss is not currently known; however, it is believed that topiramate exerts this effect via central nervous system modulation.

CONCLUSIONS

Weight gain, dyslipidemia, and hyperglycemia in patients with psychotic disorders secondary to SGA use are concerning because they can affect patients' medication adherence and increase risk for atherosclerosis and coronary heart disease. In order to prevent and or manage SGA-induced metabolic side effects, recommendations include structured metabolic screening/monitoring, lifestyle and dietary modifications, antipsychotic switch from polypharmacy to monotherapy, and change to alternative SGAs with low/lower propensity for metabolic disturbances. In addition, various adjuvant pharmacologic agents have been used to counteract antipsychotic-induced side effects, particularly weight gain. Metformin and topiramate are the agents with the most promising albeit preliminary results suggesting modest effect, in some cases contradictory, though neither medications was able to entirely counter weight gain related to SGA use. Compared to topiramate, metformin has a better safety profile, larger body of clinical evidence and longer period of clinical use for weight loss and glucose metabolism dysregulation. There are a limited number of studies with issues including sample size, heterogeneous patient populations, and relatively short trial periods of metformin and topiramate; thus, more research needs to be done to generate evidence for any short- and long-term benefits associated with the use these medications in therapy. Current evidence is insufficient to support widespread use of metformin and topiramate in clinical practice. When non-pharmacologic interventions or an antipsychotic switch fails or when neither is feasible, metformin may be a viable option for some patients.