Glucose Disturbances and Atypical Antipsychotic Use in the Intensive Care Unit

Introduction

In the intensive care unit (ICU), antipsychotics are used as a treatment for delirium, a brain dysfunction involving acute and fluctuating disturbances of cognition. ¹ In addition to an increasing duration of mechanical ventilation, lengths of stay in the ICU, and costs, delirium is now recognized as a significant contributor to morbidity and mortality in the critically ill patient. ^{2 –10} The most recent clinical practice guidelines on pain, analgesia, and delirium report that delirium may occur in up to 80% of patients admitted to the ICU requiring mechanical ventilation. ¹¹ Although the etiology of delirium in the ICU is complex and poorly understood, it may be related to neurotransmission dysfunction and physiological stress from iatrogenic causes, acute illness, or systemic inflammation. ¹²

Recommended strategies to prevent and manage delirium and delirium-related symptoms include both pharmacologic and nonpharmacologic therapies, such as sleep promotion, early mobilization, and the avoidance of benzodiazepines. ¹¹ Previously, haloperidol, a typical antipsychotic, was recommended as the agent of choice for treating ICU delirium by the Society of Critical Care Medicine (SCCM) and the American Psychiatric Association (APA). ^13,14 Currently, the 2013 SCCM guidelines do not have a recommendation for the use of haloperidol in the management of ICU delirium due to limited evidence supporting its efficacy to reduce the duration of delirium. ¹¹ Additionally, the 2013 SCCM guidelines suggest that atypical antipsychotics may reduce the duration of ICU delirium, and these agents have gained popularity in the ICU due to a lower incidence of extrapyramidal symptoms (EPS), hypotension, QT (QT_c) interval prolongation, and neuroleptic malignant syndrome (NMS) compared with typical antipsychotics. ^{15 –17} It is also important to note that adverse effects associated with haloperidol, including EPS, may occur with the use of atypical antipsychotics as well but most likely to a lesser degree. ^1,18,19 Studies involving the use of antipsychotics for delirium in the ICU have shown conflicting results regarding efficacy, so it is unclear at this point in time whether atypical antipsychotics are effective in reducing morbidity, mortality, or adverse outcomes associated with delirium. ^1,18,19

The chronic use of atypical antipsychotics and the association with metabolic disturbances, such as weight gain, dyslipidemia, and diabetes mellitus, is widely recognized. ^20,21 Although antipsychotics are typically used on a short-term basis in the ICU setting, the belief that these adverse effects are only a result of long-term antipsychotic use may be inaccurate. Case reports of patients receiving atypical antipsychotics reveal that glucose dysregulation may occur within days of treatment initiation. ^{20,22 –24} Although a wide range of metabolic disturbances have been associated with atypical antipsychotic medications, glucose dysregulation is particularly important, as hyperglycemia from any cause is associated with increased mortality and morbidity in the critically ill population. ^21,25,26 The exact prevalence of hyperglycemia in the hospital is unknown due to varying consensus on glycemic targets, thus resulting in different definitions of hyperglycemia across existing literature. ^21,27,28 Researchers estimate that up to 31% of patients in the ICU may experience at least 1 blood glucose reading of ≥200 mg/dL. ^27,29,30 Current practice guidelines recommend targeting glucose levels between 100 and 180 mg/dL for patients in the ICU. ^21,28 The objectives of this study were (1) to assess the occurrence of hyperglycemia among ICU patients who were prescribed atypical antipsychotic medications and (2) to examine patient, medication, and clinical factors associated with the occurrence of hyperglycemia.

Methods

A retrospective chart review was conducted at a 600-bed regional hospital in Southwest Georgia. Patients at least 18 years of age and prescribed an antipsychotic during ICU admission between January 2007 to November 2012 were included in the study. Exclusion criteria included patients initiated on typical antipsychotics, patients receiving antipsychotics as needed only, and patients receiving antipsychotics prior to ICU admission. All patient charts with the last names beginning with letters A to E were reviewed for inclusion in this study.

The primary objective was to identify the number of hyperglycemic episodes in patients who received an atypical antipsychotic medication during the ICU stay. Hyperglycemia was defined as the occurrence of 1 blood glucose (BG) level >180 mg/dL. Analyses included an assessment of the type and dose of atypical antipsychotics used, the frequency of patients discharged from the ICU on the atypical antipsychotic, the frequency of ICU mortality, and the treatment of the hyperglycemia. Medical charts were also reviewed to assess patient factors (age, gender, ethnicity, and preexisting diabetes), medication-related factors (drugs associated with hyperglycemia and corticosteroid use), and clinical factors (ICU type, reason for ICU admission, total parenteral or enteral nutrition, and Acute Physiology and Chronic Health Evaluation II [APACHE II] score indicating disease severity). ³¹

Descriptive statistics were used to characterize all the study variables. Bivariate and multivariate logistic regression analyses were conducted to examine the relationships between the occurrence of multiple hyperglycemic episodes (yes/no) and patient, medication, and clinical factors. To construct the most parsimonious model, independent variables with associations of P < .20 with the occurrence of multiple hyperglycemic episodes were included in the multivariate analysis. ³² A P < .05 was used to indicated statistical significance for all tests. This study was approved by the institutional review boards at the study site and academic setting.

Results

A total of 273 patients’ medical charts were reviewed. Fifty patients met the inclusion criteria; their charts were reviewed for this study (Figure 1). Baseline characteristics of the participants are listed in Table 1. Forty-six percent of patients (n = 23) experienced at least 1 hyperglycemic episode (BG >180 mg/dL) after the initiation of an atypical antipsychotic in the ICU. Of the 23 patients experiencing at least 1 episode of hyperglycemia, 14 patients (60%) experienced multiple (>1) distinct hyperglycemia episodes.

OPEN IN VIEWER

Figure 1.

Patients included and excluded.

OPEN IN VIEWER

Table 1.

Patient Demographics.

Characteristic	N = 50
Male:Female, n	36:14
Age (mean ± SD), years	59.7 ± 16.7
Ethnicity, n (%)
White	34 (68)
Black	16 (32)
ICU type, n (%)
Surgical	20 (40)
Cardiac	16 (32)
Medical	14 (28)
ICU admission indication, n (%)
Neurological	14 (28)
Respiratory	12 (24)
Cardiovascular	11 (22)
Gastrointestinal/genitourinary	7 (14)
Other	6 (12)
APACHE II score (mean ± SD)	15 ± 8
Use of corticosteroids, n (%)	20 (40)
Use of total parenteral nutrition/enteral nutrition, n (%)	12 (24)
Concurrent use of medications causing hyperglycemia (protease inhibitors, octreotide, lithium, hydrochlorothiazide), n (%)	3 (6)
Use of maintenance fluids containing dextrose, n (%)	16 (32)
Preexisting diabetes mellitus, n (%)	13 (26)

Abbreviations: APACHE II, Acute Physiology and Chronic Health Evaluation II; ICU, intensive care unit; SD, standard deviation.

Results from bivariate analyses indicated that women (χ² = 4.67, P = .03), a preexisting diagnosis of diabetes mellitus (χ² = 5.82, P = .02), and a higher disease severity (t = −2.50, P = .02) were significantly associated with the occurrence of multiple hyperglycemic episodes in ICU patients who were prescribed atypical antipsychotics. No additional independent variables had an association of P < .20 with the occurrence of multiple hyperglycemic episodes, and thus only sex, diabetes status, and disease severity were included in the multivariate model. The logistic regression analysis showed that women (odds ratio [OR] = 6.14, P = .03) and ICU patients with worse disease severity (OR = 1.13, P = .04) were significantly more likely to have multiple hyperglycemic episodes.

The types and average daily doses of atypical antipsychotics used are listed in Table 2. Quetiapine was the most commonly used atypical antipsychotic (n = 22 [44%]) with an average daily dose of approximately 90 mg. Overall, 19 patients were discharged from the ICU on the atypical antipsychotic, 6 patients died in the ICU, and 31 patients were treated with an antihyperglycemic agent (oral or injectable medication).

OPEN IN VIEWER

Table 2.

Secondary Outcomes: Atypical Antipsychotics Used.

Antipsychotic	%	Antipsychotic daily dose, mg (mean ± SD)
Quetiapine	44	89.8 ± 62.5
Risperidone	20	1.8 ± 1.3
Olanzapine	20	15.5 ± 8.6
Ziprasidone	14	34.3 ± 14.0
Aripiprazole	2	4 ± 0.0

Abbreviation: SD, standard deviation.

Discussion

In this study, almost half of the patients admitted to the ICU and initiated on an atypical antipsychotic experienced at least 1 episode of hyperglycemia. Furthermore, over half of these patients experience multiple hyperglycemic episodes. The results of this study identified factors that were associated with an increase in the likelihood of multiple hyperglycemic episodes, such as female sex, a preexisting diagnosis of diabetes, and a higher disease severity. However, study findings failed to show any significant associations between the use of corticosteroids, fluids containing dextrose, total parenteral nutrition/enteral nutrition, and hyperglycemic episodes.

To our knowledge, no trials have been conducted to evaluate acute hyperglycemia stemming from the use of atypical antipsychotics in critically ill patients. ³³ However, based on case reports, atypical antipsychotics have been shown to affect short-term glucose regulation and have shown an occurrence of hyperglycemic events even after 1 dose of atypical antipsychotic therapy. ^{34 –39} It is unclear whether metabolic abnormalities are due to risk factors including obesity, reduced physical activity, and abdominal adiposity, which are closely linked to the development of type 2 diabetes mellitus or other acute factors. ⁴⁰ Obesity, physical activity, and abdominal adiposity were not included in this study, since evidence shows that glucose abnormalities may be independent of weight gain and therefore directly linked to atypical antipsychotic use. ⁴¹ Many cases of rapid or new-onset glycemic disturbances have been shown to be reversible and resolved upon antipsychotic discontinuation. ⁴¹ This is further evidence that destabilization of glucose control may be a drug-related effect and not due to patient-specific risk factors. ⁴¹

This study found that, of patients initiated on an atypical antipsychotic, over a third (38%) of patients remained on an atypical antipsychotic during transitions of care when they were discharged from the ICU. Although this was not the primary objective of the study, these results are consistent with existing literature and the percentage of patients continued on antipsychotics upon ICU discharge. ⁴² These findings also highlight an important role for pharmacists during transitions of care, especially to minimize the risk of the development of metabolic effects from chronic use and potentially to reduce inappropriate medication use. The clinical impact of continuing these agents after the initial period of delirium and discharge from the ICU is an area that should be further investigated. It is important to discuss that clinical outcomes associated with hyperglycemia were not evaluated in this study. Therefore, clinical outcomes such as length of ICU stay and its association with the incidence of hyperglycemic episodes or increasing morbidity and mortality are an area for future study.

In this study, potential confounders affecting the occurrence of hyperglycemia were considered. These factors included the use of corticosteroids, intravenous fluids containing dextrose, total parenteral nutrition/enteral nutrition, and medications commonly associated with hyperglycemia, and the presence of baseline disease states involving diabetes mellitus. The list of medications linked to hyperglycemia included in the study was based on previous literature evaluating steroid use and hyperglycemia. ⁴³ However, it is important to recognize that this study may be underpowered to detect the effects of potential confounders on the occurrence of hyperglycemia (ie, small sample size). Future studies should be adequately powered to evaluate the potential impact of factors on the occurrence of hyperglycemia in the ICU. Another study limitation is related to the setting. The study site has an active inpatient psychiatry unit, which may have contributed to the large number of patients who were excluded due to the use of an atypical antipsychotic medication prior to admission. Future research should conduct such evaluations in settings with diverse characteristics to allow for greater generalizability of findings.

Conclusion

Patients admitted to the ICU and initiated on an atypical antipsychotic may develop hyperglycemia independent of other glucose-elevating factors. Further studies are needed to investigate the link between atypical antipsychotics and acute hyperglycemia, and the clinical significance of the impact on patient outcomes.