Drug
D2
5HT2A
M
H1
Alpha-1
Adverse side-effect association
Reference compound
Spiperone
4
–
–
–
–
EPS, tardive dyskinesia, prolactin
Ketanserin
–
4
–
–
–
Rhabdomyolysis
3-quinuclidinyl-4-iodobenzilate (QNB)
–
–
4
–
Anticholinergic, memory impairment, ↑ narrow angle, glaucoma, sinus tachycardia, blurred vision
Pyrilamine
–
–
–
4
–
Sedation and weight gain
Prazosin
–
–
–
–
4
Cardiovascular, orthostatic hypotension
First-generation typical antipsychotics
Chlorpromazine
3
2
2
3
3
Fluphenazine
4
2
1
2
2
Haloperidol
3
3
0
1
1
Loxapine
3
2
1
2
2
Mesoridazine
3
3
1
3
2
Perphenazine
4
3
0
3
2
Thioridazine
3
3
2
2
3
Second-generation typical antipsychotics
Aripiprazole
4
3
0
2
1
Asenapine
3
3
0
2
3
Clozapine
1
4
3
4
2
Iloperidone
3
4
0
2
4
Lurasidone
3
3
0
0
4
Olanzapine
2
3
2
4
1
Paliperidone
3
3
0
3
2
Quetiapine
1
2
0
2
2
Risperidone
3
4
0
3
3
Ziprasidone
3
4
0
2
1
9.2 Specific Populations
9.2.1 Elderly
Unless the patient has an existing diagnosis for schizophrenia or bipolar disorders, antipsychotic use in the elderly is “off-label” by the regulatory agencies. Antipsychotics are commonly used in the elderly and present a different set of issues compared to the adult population as polypathology, polypharmacy, potential drug-drug interactions, and age-related pharmacokinetic and pharmacodynamics are important variables in selecting the drug dosages. Due to these multiple variables, antipsychotics doses are generally lower in the elderly versus the adult group (Gareri et al. 2003). The adverse event profile for the antipsychotic drugs in the elderly differs than the adult population with specific patient warnings such as regarding their use for dementia-related psychosis and potential for cerebrovascular disorders.
9.2.2 Children and Adolescents
Most antipsychotic drugs prescribed for children are also “off-label” that include use in aggression and irritability in patients with autism, conduct disorders, and pervasive developmental disorders with an increasing frequency (Schneider et al. 2014). Antipsychotic use in clinical practice especially in young children as early as 2 years of age presents on-going challenges to healthcare systems. The overall adverse side-effect profile of antipsychotics in children and adolescents does not substantially differ from the adult population, but due to the earlier exposure to these drugs, the long-term risks and benefits must be taken into consideration by prescribers. Database systems have been set up to monitor antipsychotic adverse events in the children and adolescent population. Compared to other medications, it was reported from the FDA Adverse Event Reporting System (FAERS) that the atypical agents aripiprazole, risperidone, and quetiapine were among the top 20 medications with adverse events (Lee et al. 2014). Some countries have established nationwide antipsychotic safety monitoring programs for children and adolescents to provide system-wide safety information for clinicians (Harrison-Woolrych et al. 2007; Rani et al. 2009; Palanca-Maresca et al. 2014).
9.2.3 Pregnancy
Antipsychotic use in pregnant women continues, and information from the Tennessee Medicaid program reported use of the typical first-generation agents (FGAs) decreased while prescribing of the atypical second-generation agents (SGAs) has increased (Epstein et al. 2013). For the FGAs, the incidence of use decreased from 7.77 to 0.99 per 1,000 pregnancies between 1995 and 2005. In contrast, SGAs use increased from 1.73 to 16.52 per 1,000 pregnancies between 2000 and 2005. The specific warnings for atypical antipsychotic use during pregnancy were possible risk of abnormal muscle movements and withdrawal symptoms described for the mother and infant.
9.2.4 Dementia-Related Psychosis Fatality
Psychosis associated with Alzheimer’s disease (PAD) forms part of the behavioral and psychological symptoms of dementia (BPSD) that include hallucinations, delusions, agitation, paranoia, combativeness, or depression (Madhusoodanan et al. 2007). Delirium was reported to be the third most common cause of psychosis in the elderly characterized by thought disturbance, poverty of thinking, irrationality, and usually visual hallucinations (Jeste and Finkel 2000). Major depression was reported to be the second most common diagnosis in the elderly accounting for most of the psychosis in this population (Kyonmen and Whitfield 2009). Depression-related psychosis is typically characterized by themes of somatic troubles, persecution, guilt, and poor self-esteem. While regulatory agencies have yet to recognize BPSD as a disease, the diagnostic criteria for the concept of PAD could be acknowledged (Madhusoodanan et al. 2007; Jeste and Finkel 2000). Yet, these patients can present the clinician with a complex, multifactorial, and fluctuating nature of psychotic symptoms (Kyonmen and Whitfield 2009; Devanand 2013).
In April 2005, the FDA issued a “black box” warning for the SGAs and for the FGAs in June 2008 indicating an increased risk for death in persons with dementia (Kalapatapu and Schimming 2009). Data reported by the FDA noted a relative risk (RR) ratio of 1.47 and 1.68 for the SGAs and FGAs, respectively (US Food and Drug Administration 2014). A large meta-analysis study with 15 placebo-controlled clinical trials 10–12 weeks of duration treated with the atypical antipsychotics in patients with dementia reported an increased odds ratio (OR) of 1.54 [95 % C.I. 1.06–2.23, p = 0.02] for death that was similar to the FDA findings with a number needed to harm (NNH) = 87 (Schneider et al. 2005). A later systematic review and meta-analysis evaluated adverse events and divided the patients into two categories: elderly patients with dementia and all other nonelderly adult patients (Simoni-Wastila et al. 2009). This study examined the combined adverse events that included cardiovascular symptoms, edema, and vasodilatation. Risperidone (2.10, 95 % C.I. 1.38–3.22) and olanzapine (2.30, 95 % C.I. 1.08–5.61) was found to have a higher OR compared to placebo for these combined symptoms. Quetiapine and aripiprazole were found not associated with cardiovascular outcomes. However, when antipsychotics were examined in an outpatient population in patients with probable Alzheimer’s disease for an extended time period (mean of 4.3 years), the presence of antipsychotic drugs was found not to increase the risk of death using time-dependent sensitive models (Lopez et al. 2013). It was reported that nursing home admission and death were more frequent in patients treated with FGAs than SGAs. But, both FGAs and SGAs were not associated with nursing home admission or time to death when including covariate factors such as diabetes, hypertension, preexisting heart disease, and cognitive and demographic variables.
The efficacy of SGAs in patients with PAD (N = 423) was assessed with olanzapine, quetiapine, and risperidone, and the main outcomes were discontinuation of treatment for any reason and the number of patients with at least a minimal improvement on the Clinical Global Impression of Change (CIBIC) at 12 weeks (Schneider et al. 2006). Significant differences were not found between the SGAs, and improvement based upon CIBIC scores was found with olanzapine 32 %, quetiapine 26 %, and risperidone 29 % compared to placebo 21 % (p = 0.22). Patients treated with SGAs had a higher discontinuation rate due to intolerability (p = 0.009). The study reported that the overall adverse events offset the efficacy from SGAs. While the risk of fatality was not an outcome, these results pose difficult questions for the use of SGAs in elderly patients with dementia that display PAD or BPSD (Maher et al. 2011). The American College of Neuropsychopharmacology (ACNP) White Paper noted that the two most common causes for deaths were due to cardiovascular disease or infections in this population (Jeste et al. 2008). Also noted that besides psychosis and agitation commonly occurring in patients with dementia, these symptoms lead to significant caregiver distress and accelerated patient placement into the nursing home environment. Finally, regulatory agencies have not yet “approved” any medications for the treatment of patients with PAD or BSPD.
Clinicians face unusual challenges in risk management when prescribing antipsychotics in this patient population. Risk management procedures recommended are to avoid unnecessary medications, balancing acute versus long-term risks, informed consent from the patient and family members, and documentation in the medical records (Jeste et al. 2008; Recupero and Rainey 2007). The ACNP White Paper recommended these 11 steps: (1) determine the etiology of psychotic symptoms, (2) general treatment considerations, (3) shared decision-making, (4) identify specific target symptoms for treatment, (5) pharmacotherapy selection, (6) dose, (7) monitor for efficacy, (8) monitoring for safety, (9) educate patient and caregivers, (10) know when to discontinue or switch pharmacotherapy, and (11) coordinate care among the treatment team and family members (Jeste et al. 2008).
9.2.4.1 Cerebrovascular Warning
The FDA black box warning for SGAs in 2003 of increased fatality in patients with dementia-related psychosis coincided with the cerebrovascular warning (US Food and Drug Administration 2003). Cerebrovascular adverse events (CVAEs) were specifically noted for olanzapine, aripiprazole, and risperidone which included strokes, transient ischemic events, and other undetermined events thought to be vascular in origin. Whether or not the increased risk of fatality and CVAEs risk have similar mechanisms or are interrelated remains unknown, but the overall death rate was not overly represented by the CVAEs (Nelson 2005). The risk for CVAEs is higher among patients with vascular dementia, vascular disease, or risk factors for stroke. Added to the pathophysiology are potential drug-induced orthostatic hypotension and oversedation as both events could result in falls and aspiration pneumonia. A cohort study compared the OR of CVAEs between FGAs and SGAs in patients with dementia aged 65 years or older (Laredo et al. 2011). The OR for FGAs was reported to be 1.16 [95 % C.I. 1.07–1.27], whereas the OR for the atypical agents was 0.62 [95 % C.I. 0.53–0.72]. The results of this study indicate that FGAs have a slightly higher risk for CVAEs but substantially and that SGAs appear to be safe. A large systematic meta-analysis reported that the SGAs were associated with an increased risk of stroke (pooled OR, 3.12 [95 % C.I. 1.32–8.21]) and determined the NNH = 53. The optimal approach to care is that each patient who needs an antipsychotic drug must be carefully assessed with both short-term and long-term risks and benefits. The ACNP White Paper included the CVAE assessment and noted significant limitations of employing randomized clinical trial data and attempting to determine long-term patient outcome with adverse events (Jeste et al. 2008). The identical 11 steps for patient assessment and treatment with SGAs in consideration of the CVAEs possibility were recommended in the patients with dementia-related psychosis.
9.3 Specific Adverse Events
9.3.1 Extrapyramidal Side Effects (EPS) and Tardive Dyskinesia (TD)
The advent of the SGAs was viewed as a major improvement in the pharmacotherapy of psychotic disorders as a lower incidence of extrapyramidal side effects (EPS, dystonia, pseudoparkinson’s, and akathisia) was found compared to the first-generation agents (Divac et al. 2014; Peluso et al. 2012). Although not entirely EPS-free, the use of adjunctive anticholinergic medications was 30-fold lower among the patients treated with the atypical antipsychotics (Peluso et al. 2012). CLZ had the reported lowest EPS odds ratio (OR) of 0.3 [95 % C.I. 0.12–0.62], and lurasidone had the highest for the SGAs with an OR of 2.46 [95 % C.I. 155–3.72] (Leucht et al. 2013). In comparison to the FGAs, the OR for chlorpromazine and haloperidol was 2.65 [95 % C.I. 1.33–4.76] and 4.76 [95 % C.I. 3.70–6.04], respectively (Leucht et al. 2013). A 17-year experience of drug-induced Parkinsonism from a regional pharmacovigilance center reported that antipsychotics had the highest incidence compared to all other drugs reported to cause this adverse event (49 % versus miscellaneous drugs 28.7 % and antidepressants 8.0 %) (Bondon-Guitton et al. 2011). Although other medications can cause EPS, antipsychotics (both SGAs and FGAs) are likeliest agents when compared to other medication classes.
A systematic review of long-term studies investigating the annual incidence of tardive dyskinesia (TD) reported a rate of 3.9 % for the SGAs and 5.5 % for the FGAs (Correll and Schenk 2008). Two risk factors were identified possibly associated with an increased risk of TD – mood disorders and age (Keck et al. 2000). The observed rate of TD from a 15-year time period of atypical use on patients with mood or anxiety disorder was 5.9 % from a retrospective chart review of 268 patients (Coplan et al. 2013). The mean length of treatment for TD occurrence was 28.7 months (range 1–83 months) and average chlorpromazine equivalent dose was 350 mg (range 67–969 mg) and symptoms resolved in all but one patient when antipsychotics were discontinued. Most patients (90.9 %) consented to a second treatment with an atypical antipsychotic and did not reexperience TD. The risk of TD with FGAs was found to be three to five times higher in patients >55 years than younger patients (Woerner et al. 1998). Two large studies in elderly patients with risperidone and olanzapine reported TD rates comparable to the adult population, but either drug had an equal or lower incidence than FGAs (Woerner et al. 2011; Kinon et al. 2014). Careful patient monitoring for EPS and early detection of TD in all patients are recommended with FGAs and SGAs.
9.3.2 Seizures
Patients with schizophrenia may be more prone to seizures than the general population (Hyde and Weinberger 1997). Psychotropic drugs and especially antipsychotics and some selected antidepressants reduce seizure threshold with a reported range of 0.1–1.5 % in patients with psychiatric disorders versus the general population of 0.07–0.09 % with therapeutic doses (Devinsky et al. 1991). Dose-dependent and rapid dose increase were factors reported with clozapine and chlorpromazine that can lower seizure threshold. The overall incidence of seizures reported with CLZ daily doses were <300 mg 1.0 %, 300–600 mg 2.7 %, and >600 mg 4.4 % (Devinsky and Pacia 1994). Tonic-clonic seizures were most often noted, but myoclonic seizures can also occur (Pacia and Devinsky 1994). The FDA analysis of seizure incidence from the atypical antipsychotic clinical trial data reported a standard incidence ratio (SIR) of 2.05 for all antipsychotics, whereas the SIR for CLZ was found to be 9.50 (Pisani et al. 2002). The Spanish Pharmacovigilance network reported a higher reporting odds ratio (ROR) with the SGAs versus the FGAs of 3.2 [95 % C.I. 2.21–4.63] with the highest incidence with CLZ (94/169 of the total number of convulsions with SGAs) (Alper et al. 2007). Whether or not atypical agents possess a higher seizure induction compared to first-generation agents should be further evaluated (Lertxundi et al. 2013).
9.3.3 Metabolic Syndrome (MetS)
The MetS associated with SGAs is well known and associated with increased weight gain, diabetes, and hyperlipidemia (Bak et al. 2014). The MetS definition includes the presence of at least three of five parameters – blood pressure >130/85, fasting blood glucose (FBS) >110 mg/dL, fasting triglycerides >150 mg/dL, HDL <40 mg/dL (men) or <50 mg/dL (women), waist circumference >102 cm (men) or >88 cm (women) (Ganguli and Strassnig 2011). The prevalence of MetS in the general population was estimated to be about 23 % compared to a higher incidence of 35–37 % in patients with psychiatric disorders treated with SGAs (Newcomer and Hennekens 2007; Kagal et al. 2012). The FDA in 2003 issued a warning regarding hyperglycemia and diabetes and recommended monitoring fasting blood glucose in patients with diabetic risk factors, symptoms of hyperglycemia, or diabetes (Mittal et al. 2013). The consensus panel a year later from the American Diabetic Association (ADA), American Psychiatric Association (APA), and the Mount Sinai Summit panel recommended monitoring weight, glucose, or glycosylated hemoglobin and lipids every 12 weeks. Blood pressure was included every 12 weeks while body weight monitoring every 4 weeks and waist circumference annually (ADA et al. 2004).
Weight gain is among the strongest factors leading to the MetS (Papanastasiou 2013). Data from the antipsychotic registration clinical trials reported the risk for clinically significant weight gain (≥7 % than baseline) was about ten times greater for olanzapine compared to placebo. Ziprasidone, aripiprazole, and paliperidone were only about twice the risk of placebo. Data reported on weight gain associated with SGAs from highest to lowest were olanzapine, quetiapine, risperidone, ziprasidone, aripiprazole, and paliperidone (Leucht et al. 2013). From the comparative study with the FGAs and SGAs, the latest SGA lurasidone had a low SMD (0.10, 95 % C.I. −0.02–0.21) for weight gain and was about equal to the SMD for ziprasidone (0.10, 95 % C.I. −0.02–0.22) and haloperidol (0.09, 95 % C.I. −0.00–0.17) in producing weight gain. Asenapine SMD (0.23, 95 % C.I. 0.07–0.39) was slightly greater than aripiprazole (0.17, 95 % C.I. 0.05–0.28) but less than paliperidone (0.38, 95 % C.I. 0.27–0.48). The iloperidone SMD (0.62, 95 % C.I. 0.49–0.74) was above chlorpromazine (0.55, 95 % C.I. 0.34–0.76) and slightly less than clozapine (0.65, 95 % C.I. 0.31–0.99). Clozapine was slightly lower than olanzapine (0.74, 95 % C.I. 0.67–0.81) in weight gain, and both agents are well known to cause MetS (Leucht et al. 2013). The underlying pathophysiology for antipsychotics to cause weight gain is complex but thought to involve genetic variables that include the 5HT2C receptor system and leptin promoter gene variants MTHFR and MC4R genes. Additional factors reported were HRH1, BDNF, NPY, CNR1, GHRL, FTO, and the AMPK gene (Kao and Muller 2013). Further research remains underway to determine biological mechanisms that influence antipsychotic-induced weight gain.
As the monitoring parameters for MetS in patients treated with SGAs were recommended, implementation remains challenging for healthcare systems. A large VA retrospective cohort study (N = 12,009) evaluated practices for monitoring MetS from antipsychotics between April 2008 and March 2009 (Kumra et al. 2008). Weight, glucose or Hb1Ac, and LDL cholesterol were monitored within 30 days of antipsychotic therapy initiation (baseline) and every 60–120 days afterward. This study reported that weight was the most frequently monitored. Patients without a schizophrenia diagnosis were less likely to be monitored. Antipsychotics with lowest MetS risk were more likely to have weight monitored versus the higher-risk agents at baseline. Whereas, patients taking antipsychotics with a moderate MetS risk were more likely to have baseline glucose and LDL compared to low- and high-risk antipsychotics. The study concluded that improvements are needed and should be applied to all patients regardless of the antipsychotic agent being prescribed.
Another factor associated with antipsychotic-induced weight gain and MetS was age (Papanastasiou 2013). This finding can have significant impact as the antipsychotics are used in the elderly population. Antipsychotics can also cause weight gain in children and adolescents (Kumra et al. 2008; Kranzler and Cohen 2013). No appreciable differences were reported between men and women (Papanastasiou 2013). Waist size was found to be a useful factor in predicting MetS with a sensitivity of 79.4 % and specificity of 78.8 % with the highest rate found with CLZ (Mitchell et al. 2013). Hospitalized patients with various psychiatric disorders were evaluated for MetS in which weight gain and BMI were strongly associated (Centorrino et al. 2012). Risk of MetS was also greater with polytherapy (defined as >2 antipsychotics or one plus adjunctive mood stabilizer) (Centorrino et al. 2012). Factors reported to be associated with MetS were sex, ethnicity, and martial and living status where patients with schizophrenia can develop quickly even within 6 weeks of antipsychotic treatment (Papanastasiou 2013).
As the landscape of MetS associated with FGAs and SGAs became evident for clinicians and monitoring parameters established, the issue of prediabetes emerged given that patients with serious mental illnesses had increased incidences of MetS compared to the general population. The ADA in January 2010 published the prediabetes criteria that includes impaired FBG (>125 mg/dL) or impaired glucose tolerance (140–199 mg/dL 2 h after 75 g of glucose) or Hb1Ac in the range of 5.7 %–6.4 % (ADA 2010). Abdominal obesity, hypertension, dyslipidemia, decreased HDL, and or increased triglycerides are other factors that are associated with prediabetes. From 2003 to 2007, patients (N = 783) without a history of type 1 or type 2 diabetes admitted to a psychiatric facility treated with antipsychotics were assessed for metabolic screening (Manu et al. 2012). Using the ADA 2010 criteria for prediabetes, 413 (52.8 %) subjects had normal glucose, while 290 (37.0 %) were identified as prediabetic. Newly diagnosed diabetes was found among 80 patients (10.2 %). Prediabetes diagnosis was established by using one criteria in 209 patients (72.1 %), 71 patients (24.5 %) had two criteria, and only 10 patients (3.4 %) fulfilled all three criteria. An abnormal Hb1Ac identified 120 of the 290 (41.4 %) of the prediabetic patients when used as the sole criteria. When FBS and/or Hb1Ac was used with the glucose tolerance test, 89.7 % of prediabetic patients were identified. FGA and SGA use was evaluated among all patients. CLZ and olanzapine had the highest incidence of prediabetes and diabetes, while the other SGAs and FGAs had about an equal incidence.
MetS associated with antipsychotic treatment is an important component when clinicians prescribe these medications when treating patients with various psychiatric conditions. The consequences of patients with prediabetes and then those who develop diabetes present significant challenges in antipsychotic drug therapy and lifestyle management in balancing the risks and benefits of short-term and long-term use. MetS can lead to other significant diseases like cardiovascular, cerebrovascular, ophthalmologic, and renal impairment with an increased morbidity and mortality resulting in significant healthcare expenditures (Riordan et al. 2011).
9.3.4 Cardiovascular
9.3.4.1 Torsades de Pointes (TdP)
A meta-analysis review of 15 antipsychotics revealed that the lowest OR occurred with lurasidone −0.10 [95 % C.I. −0.21–0.01], a modest OR with ziprasidone of 0.41 [95 % C.I. 0.31–0.51], and the highest OR with sertindole 0.90 [95 % C.I. 0.76–1.02]. Other SGA agents had OR between lurasidone and ziprasidone. Sertindole and amisulpride had the highest OR; however, these two agents are not available in the USA (Leucht et al. 2013). From the historical perspective, in 1996, sertindole was under development in the USA, but due to prolonged QTc interval and association with 12 sudden unexplained deaths, the FDA did not grant approval. TdP is a malignant ventricular arrhythmia associated with syncope and sudden death. The primary form of TdP occurs from congenital courses either familial or sporadic. The secondary TdP form is usually drug-induced (Beach et al. 2013). From an electrical cardiovascular pathophysiological approach, the heart rate is inversely related to the QT interval and can be viewed upon an ECG. The “c” in the QTc accounts for a “correction” intended to remove the heart rate as a confounding variable. The Bazett formula is the most method QTc = QT/square root of the R-R interval (Beach et al. 2013). Clinicians should be concerned when the QTc interval is between 450 and 500 ms, and >500 ms has increased risk of developing TdP.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
