Treatment of Diabetic Ketoacidosis Associated With Antipsychotic Medication

Supplemental digital content is available in the text.

Abstract

Background

The second-generation antipsychotics (SGAs) are associated with metabolic disturbances. Diabetic ketoacidosis (DKA) is a rare, but potentially fatal sign of acute glucose metabolism dysregulation, which may be associated with the use of SGAs. This study aims to review published reports of patients with schizophrenia and antipsychotic drug–associated DKA, focusing on the effective management of both conditions.

Methods

Using a predefined search strategy, we searched PubMed and EMBASE from their inception to July 2016. The search terms were related to “diabetic ketoacidosis” and “antipsychotic medication.” Case reports, case series, and reviews of case series written in English language were included in the review.

Results

Sixty-five reports were analyzed. In most patients who developed antipsychotic-associated DKA, 1 or more suspected antipsychotic medications were discontinued. In 5 cases, a rechallenge test was trialed, and in only 1 case, it resulted in the elevation of blood glucose. The majority was subsequently treated with a different SGA in combination with insulin/oral hypoglycemic agents; although approximately a third of patients had a complete resolution of symptoms or could control diabetes with diet only at the point of discharge.

Conclusions

Patients taking antipsychotic medications should be regularly screened for insulin resistance and educated about potential complications of antipsychotic medications. This will allow clinicians to individualize treatment decisions and reduce iatrogenic contribution to morbidity and mortality. To achieve best treatment outcomes, antipsychotic-induced DKA should be treated jointly by psychiatry and endocrinology teams.

The second-generation antipsychotics (SGAs) are considered as the first line treatment for schizophrenia spectrum disorders. In addition to their effectiveness in treating psychotic symptoms, most SGAs are associated with significant weight gain and the development of glucose intolerance, thus causing diabetes mellitus, metabolic syndrome, and overall increased risk of cardiovascular morbidity and mortality.^1–3

Diabetic ketoacidosis (DKA) is an acute complication of diabetes that mainly occurs in patients with type 1 diabetes, but it is not uncommon in patients with type 2 diabetes. This condition is a complex disordered metabolic state characterized by hyperglycemia, ketoacidosis, and ketonuria. With prompt treatment, DKA can be corrected without any complications developing. If left untreated, the condition can be life threatening. Diabetic ketoacidosis was once considered a hallmark feature that would differentiate individuals with diabetes mellitus type 1 from those with diabetes mellitus type 2, although reports of DKA occurring in patients with diabetes mellitus type 2 were described as well.⁴ Epidemiologic studies have estimated that DKA-related hospitalizations have increased over the past 2 decades, reaching the rates of 32.4 per 1000 among people younger than 45 years, 3.3 among people aged 45 to 64 years, and 1.4 among people 65 years or older.⁵

Interestingly, in schizophrenia, DKA is often the first sign of antipsychotic (AP) drug–associated diabetes mellitus among patients. For example, Henderson et al⁶ showed that the incidence of diabetes presenting as DKA in this population is more than 10-fold higher than that in the general population. Among those with preexisting diabetes mellitus, the rate is 30-fold higher.⁷ Other risk factors include polypharmacy, olanzapine monotherapy, clozapine monotherapy, male sex, and middle age (45–60 years).⁸ Antipsychotic-associated DKA has been associated with 13% mortality rate.^9,10

The management of DKA (irrespective of whether it is part of diabetes mellitus type 1 or 2) consists of intravenous insulin therapy with concomitant glucose administration when plasma glucose levels fall below 300 mg/dL (16.7 mmol/L), sufficient volume and electrolyte replacement, and careful attention to identification and management of associated problems, such as acute pancreatitis and infection.⁴ Despite all this, no recommendations exist for management of DKA in psychiatric patients. Given the long-term nature of psychiatric treatments, as well as high mortality rates of AP drug–associated DKA, this condition represents a compelling treatment challenge throughout the course of illness.

This study aims to review published reports of patients with schizophrenia and AP drug–associated DKA, focusing on the management of SGA-induced DKA.

METHODS

Using a predefined search strategy, we searched PubMed and EMBASE from their inception to July 2016. The search terms included “diabetic ketoacidosis,” “antipsychotics,” “atypical antipsychotics,” “first generation antipsychotics,” “second generation antipsychotics,” “clozapine,” “olanzapine,” “risperidone,” “quetiapine,” “ziprasidone,” “aripiprazole,” “paliperidone,” “amisulpride,” and “haloperidol.” We also conducted hand searches of key journals and publications of experts in the field for potentially relevant articles. Articles reporting clinical cases of DKA associated with the use of AP agents were included. Only articles published in English language were included.

All potential studies were exported into reference management software. Next, duplicates were removed, and we conducted the initial screening of titles and abstracts for inclusion against the eligibility criteria. The full texts of articles meeting the criteria were obtained and reviewed. Data were extracted on patients characteristics (sociodemographic and clinical status) and psychiatric pharmacotherapy associated with DKA before and after the occurrence of DKA.

Using these key words, we selected 168 titles. After reading the article titles, we selected 140 abstracts in English-language articles from the period 1994 to 2015. Of these abstracts, we selected 83 case reports and reviews of case series and included them into the review. A total of 18 cases were excluded because of lack of provided information: in 11 cases, there was no record of AP therapy after DKA, and in 7 cases, patients were deceased. After applying the inclusion and exclusion criteria, 65 reports were included in this review. The final list included 44 case reports and 6 case series (Table 1, Fig. 1).

TABLE 1.

Treatment Strategies Including AP Medication After AP-Associated DKA

FIGURE 1.

CONSORT (Consolidated Standards of Reporting Trials) diagram to report results of treatment options after AP-associated DKA. FGA indicates first-generation AP; R, risperidone; Z, ziprasidone; Q, quetiapine; A, aripiprazole; O, olanzapine; Cl, clozapine.

RESULTS

Sociodemographic and clinical characteristics of the patients are summarized in the Supplemental Digital Content 1, Supplementary Table (http://links.lww.com/JCP/A474).^10–59

Antipsychotic Medication

Results of treatment options after AP drug–associated DKA are summarized in Figure 1. In the majority of reported cases (n = 58), DKA was associated with SGA monotherapy: olanzapine (n = 26), clozapine (n = 17), risperidone (n = 6), aripiprazole (n = 6), and quetiapine (n = 3). In most cases, the suspected SGA was withdrawn from treatment after detecting DKA. Subsequently, patients were put on different AP medications, as shown in Table 2.

TABLE 2.

Recommended Treatment Strategies for Patients at Risk of AP-Associated DKA

In 5 cases, a rechallenge test was trialed. In 2 cases, clozapine was restarted 24 days and 2 months after the DKA, respectively. However, in both cases, clozapine was again discontinued because of elevation of serum glucose levels. In 1 case, clozapine was restarted and continued because of treatment resistance to other APs. Every subsequent discontinuation of clozapine due to the patient's noncompliance resulted in glucose level normalization, followed by the elevation of glucose levels after clozapine reintroduction. In 1 case, olanzapine was restarted after 1 month and discontinued after 3 days because of elevation of serum glucose levels. Blood glucose levels stabilized with oral hypoglycemic therapy (OHT). The patient was prescribed with clozapine, which again led to elevated glucose serum levels, regardless of the increased dose of OHT. One successful rechallenge was described in the patient treated with risperidone where DKA developed shortly after perospirone hydrochloride (a novel serotonin-dopamine antagonist) was added to risperidone. Risperidone was restarted in monotherapy 2 weeks after the DKA was detected, while insulin therapy was gradually replaced with diet.

In 7 cases, DKA was associated with polypharmacy, and authors reported the discontinuation of 1 or even both suspected agent (Fig. 1).

Antidiabetic Treatment

After the acute treatment of DKA, 19 patients completely recovered, and/or 10 patients were able to control diabetes mellitus with diet only. Among the patients who were discharged without antidiabetic treatment, in 1 case, hyperglycemia was reported after 6 months, but the circumstances in which this occurred were not reported. However, the majority of patients needed antidiabetic treatment with insulin (n = 30), oral hypoglycemic agent (n = 9), or both⁴ at the time of hospital discharge regardless of the change in AP treatment. Among those who were discharged from the hospital with antidiabetic medication, the majority still needed antidiabetic treatment (n = 24) in the follow-up period; 5 patients had a complete resolution of diabetes mellitus, and the rest were lost in the follow-up (n = 14).

The hypoglycemic therapy after DKA is presented in Figure 2. In cases where APs were discontinued after DKA, the majority of patients were discharged without any antidiabetic medication. On the contrary, in cases where the same DKA-inducing AP was continued, most patients required insulin antidiabetic therapy at the time of discharge.

FIGURE 2.

Hypoglycemic therapies after AP-induced DKA. HT indicates hypoglycemic therapy; ins, insulin; N/R, no record; AP, AP therapy; FGA, first-generation AP.

DISCUSSION

This literature review reveals that clinicians tend to use several strategies when treating patients with AP-associated DKA: (1) acute management includes discontinuation of the suspected SGA agent and introduction of antidiabetic agents (insulin, OHT) and lifestyle changes (eg, diabetic diet); (2) post-DKA management includes switching to a different AP medication with lower propensity or reintroduction of the same AP medication. Most patients require continuous antidiabetic treatment. In several cases, both AP and antidiabetic treatment were discontinued.

The reintroduction of the same AP medication seems the least optimal because it is associated with a significant risk of relapse of diabetes mellitus or DKA. In the few cases where rechallenge was tried with the same therapeutic agent or the one from the similar pharmacological group (olanzapine, clozapine), glucose dysregulation was observed despite the use of insulin/or oral antidiabetic agent. Furthermore, in most cases, the maintenance of the same AP therapy required insulin therapy in multiple daily injection therapy. Apart from being very demanding on patients, there are numerous risks associated with such therapy (eg, hypoglycemia, ketoacidosis, chronic complications, etc), especially given the low compliance of patients with major psychiatric disorders.⁶⁰ Therefore, the decision to reintroduce the same AP together with insulin or oral antidiabetic should require a careful risk assessment taking into account the patient's psychiatric status, possibility to achieve remission with another AP, risk of new DKA, and the possibility to maintain normoglycemia. Given the periodic nature of schizophrenia and bipolar disorder, the option to discontinue both AP and antidiabetic treatment is not suggested either, as the risk of relapse of psychiatric symptoms is significant.⁶¹

Our results show that emergency treatment with insulin and fluid resuscitation led to complete recovery of AP-induced DKA in most cases.^62–66 However, this was found only when the suspected drug was discontinued. According to recently published reports, DKA does not seem to be a time-limited medical event. According to Guenette et al,¹⁰ after acute treatment of DKA, 36% of patients had complete resolution of symptoms and required no treatment, 14% controlled diabetes mellitus with diet, and 50% relied on insulin, oral hypoglycemic agents, or both.⁸ Similarly, we found that most reported patients needed continuation of antidiabetic therapy at the time of discharge. This might reflect a certain degree of β-cell failure, which possibly persists after the AP-associated DKA, although to the best of our knowledge no data were published on this topic.

Our findings indicate that patients with glucose intolerance needing AP treatment require interdisciplinary management provided by psychiatry and diabetology teams. In line with the American Diabetes Association/American Psychiatric Association consensus conference (2016)⁶⁷ recommendations for metabolic monitoring of patients receiving atypical APs, psychiatrists should perform the following: obtain data on personal and family history of diabetes mellitus, screen at baseline and regularly monitor physical and laboratory indicators including waist circumference and glucose and lipids in blood, regularly assess adverse effects, and educate patients and their caregivers about AP-associated adverse effects. Diabetologists should diagnose and treat acute DKA following the treatment protocol (intravenous and later subcutaneous application of insulin) and make recommendations based on the type of diabetes. After DKA is diagnosed, the preferred treatment options include (1) exclusion of suspected agent in acute DKA; (2) application of the standard therapy for DKA; (2) given the morbidity and mortality associated with diabetes mellitus, as well as the need for long-term treatment, switching to another SGA with lower propensity to DKA, unless it is already proven that the change in the basic AP will lead to significant deterioration of the patient's psychiatric status; (3) slow reduction of insulin dose and possibly omission of insulin based on glycemic values (first reduce prandial and then basal insulin); (4) if there are no contraindications for metformin, the maximal dose of 2 g per day bid divided should be introduced until glycated hemoglobin (HbA_1c) reaches less than 6%. At that point, we recommend reducing the daily dose of metformin by half and monitoring HbA_1c in the next 3 to 6 months. If HbA_1c is lower than 6%, metformin should be excluded from therapy. Recommendations are summarized in Table 2.

CONCLUSIONS

The majority of the patients who developed AP-associated DKA were subsequently treated with a different SGA in combination with insulin/oral hypoglycemic. The field lacks prospective studies with longer follow-up of these patients. Therefore, the possible implications of these treatments over the longer course of illness remain unclear. Increased awareness of this problem among clinicians and patients is needed, as well as primary and secondary preventive actions such as systematic and periodic screening for insulin resistance and educating patients about potential complications of AP treatment. Patients with AP-induced DKA should be treated by psychiatry and diabetology teams and followed up to ensure somatic and mental health of these patients.