PATIENTS in the intensive care unit (ICU) are at high risk of developing delirium. Delirium is an independent determinant of hospital length of stay and may be associated with increased morbidity and mortality. Although estimates of the incidence of delirium in the ICU vary considerably, it is widely recognized as difficult to diagnose and treat. Standard therapy for delirium and agitation in the ICU is intravenous haloperidol. However, haloperidol is sometimes ineffective in managing delirious patients and large doses pose an increased risk of side effects. Newer “atypical” antipsychotic agents are equivalent to traditional agents in controlling psychomotor agitation in acutely psychotic patients.1It is not currently known whether they have a role in treating delirium in ICU patients. Ziprasidone (Geodon®; Pfizer Pharmaceuticals, New York, NY), one of the new atypical agents, is the only one that is currently available as a parenteral formulation. In this report, we describe the first successful use of ziprasidone via the intravenous route in an ICU patient with severe, life-threatening delirium that failed to respond to standard haloperidol therapy.
Case Report
A 47-yr-old, 112-kg man was admitted to the trauma/surgical ICU after placement of a free fibular graft to the left clavicle. His medical history was significant for hypertension treated with a β-blocker, elevated cholesterol, anxiety, and depression. Home medications included clonazepam, mirtazepine, risperidone, and acetaminophen/oxycodone. The patient drank alcohol, occasionally heavily, and had a 30-pack/yr cigarette smoking history.
On the first postoperative day, cisatracurium, propofol, and fentanyl infusions were discontinued, and the patient was extubated. Atenolol, clonazepam, and transdermal clonidine were prescribed. Several hours later the patient became confused and agitated. He was unable to maintain attention and demonstrated disorientation and perceptual disturbances. Continuous intravenous hydromorphone, up to 2.5 mg/h and lorazepam 2 mg intravenously every 6 h and every 1 h as needed were given for pain and agitation. Despite escalating doses of intravenous lorazepam (4 mg bolus + 12 mg/h continuous infusion), the patient remained delirious and dangerously agitated. Intravenous haloperidol 2 mg, 5 mg, 10 mg, and 20 mg had no effect on the agitation. The viability of the fibular graft was at risk because of his delirium; he was therefore sedated and reintubated. Dexmedetomidine loading dose and continuous infusion were initiated while haloperidol and lorazepam were continued. Despite multiple combinations of sedative agents, the patient was either well-sedated and hypotensive (further worsening graft blood flow) or delirious and agitated, risking disruption of the vascular pedicle of the free fibular graft.
Because of the severity of delirium and absence of response to haloperidol, ziprasidone therapy was considered. Corrected QT interval (QTc) was found to be 470 ms; serum potassium and magnesium levels were determined to be normal. Ziprasidone 20 mg was given intravenously. QTc was 476 ms 5 min later and the patient’s restlessness dramatically improved. Haloperidol, dexmedetomidine, and hydromorphone were gradually discontinued. QTc 4 h after ziprasidone administration decreased to 436 ms. The patient was extubated and ziprasidone and lorazepam were tapered via the oral route. Ziprasidone was discontinued 1 week later. The patient was transferred from the ICU on postoperative day 8 receiving oral methadone and lorazepam.
Discussion
Delirium is characterized by an acute change in cognition and disturbance in consciousness, usually resulting from an underlying medical condition or from medication or drug withdrawal. Estimates of the incidence of delirium in the ICU vary from 20–80%.2Patients suffering from delirium while hospitalized in the ICU have prolonged lengths of ICU and overall hospital stay, more frequent medical complications, and increased mortality.3,4
The mainstay of therapy for agitation and delirium in the ICU is intravenous haloperidol, a potent antipsychotic agent that blocks dopaminergic receptors in the central nervous system.5,6A recent clinical practice guideline published by the American College of Critical Care Medicine recommends haloperidol for treatment of agitation and delirium, although the level of recommendation is graded C.5
Haloperidol is associated with antidopaminergic side effects, including extrapyramidal effects and tardive dyskinesia, and can cause prolongation of the QT-interval on electrocardiogram. QT-interval prolongation can lead to an increased risk of ventricular arrhythmias, including torsades de pointes.7,8Sharma et al. found the incidence of torsades de pointes associated with intravenous haloperidol was significantly greater in patients receiving 35 mg or more over 24 h and in those with a QTc of >500 ms.9
Newer “atypical” antipsychotic agents have been recently introduced. Because these agents exert their effect mainly through modulation of nondopaminergic pathways, their use may result in diminished incidence of tardive dyskinesia. Ziprasidone can prolong the QT interval and thereby increase the risk of ventricular arrhythmias, including torsades de pointes.10For this reason, ziprasidone is contraindicated in patients with corrected QT interval (QTc) prolongation >500 ms, recent myocardial infarction, or uncompensated heart failure. It can produce additive effects with other agents known to prolong QT interval and therefore, close hemodynamic monitoring, including five-lead electrocardiogram, is warranted. Ziprasidone should also be avoided in patients with congenital long QT syndrome and in patients with a history of cardiac arrhythmias. Food and Drug Administration-mandated studies of QT prolongation associated with neuroleptic administration demonstrated a mean increase of QTc of 20.3 ms for ziprasidone versus 4.7 ms with haloperidol.11In clinical trials of ziprasidone, the electrocardiograms of two of 2,988 patients (0.06%) who received ziprasidone and one of 440 (0.23%) who received placebo revealed QTc intervals exceeding the potentially clinical relevant threshold of 500 ms.12
Ziprasidone is the only atypical antipsychotic agent that is currently formulated for parenteral administration. Usual adult dosing is 10–20 mg intramuscularly as required to a maximum dose of 40 mg a day. Doses of 10 mg may be administered every 2 h; doses of 20 mg may be administered every 4 h. There appear to be no anticholinergic effects associated with its use. Half-life is 6–7 h.13Use of the intramuscular preparation for more than 3 days has not been studied; every effort should be made to convert to oral dosing as early as possible. Intramuscular administration of ziprasidone in healthy volunteers demonstrated time to peak plasma level of 60 min.12More rapid control of symptoms, however, is often necessary in the treatment of a patient with severe agitation necessitating intravenous administration. Phase I clinical trials of intravenous ziprasidone administered as a 60-min infusion demonstrated pharmacokinetics similar to intramuscular administration. Further experience with intravenous administration is lacking.
Published studies support the use of atypical antipsychotics in the treatment of delirium in hospitalized, medically ill patients. Olanzapine, quetiapine, and risperidone have all been reported to improve the symptoms of delirium in small, uncontrolled trials.14The only published case report describing ziprasidone treatment in delirium utilized oral dosing in a patient with cryptococcal meningitis as a result of human immunodeficiency virus.15
This report is the first of the use of intravenous ziprasidone to control delirium in an agitated, critically ill patient in the ICU. Delirium remains a daunting clinical challenge in the practice of intensive care medicine. It can be difficult to diagnose and treat and is associated with worsened patient outcomes. The recent availability of newer atypical antipsychotic agents offers an alternative approach to the treatment of delirium. Because this is only the first report of the use of intravenous ziprasidone to treat ICU-related delirium refractory to standard haloperidol treatment, much further investigation remains. The use of such therapy is anecdotal and the risks—including QT prolongation and ventricular arrhythmia—must be considered. Only appropriately powered, prospective, randomized trials will be able to satisfactorily answer questions of safety and efficacy that this novel therapy raises.
Fig. 1. Strain differences in MAC levels for isoflurane. Significantly smaller mean (± SD) end-tidal isoflurane concentrations were required to eliminate movement response to tail clamp in Dahl Salt Sensitive rats (SS) and SS.BN.16 consomics compared with Brown Norway (BN) and SS.BN.13 animals. No differences exist between SS versus SS.BN.16 or between BN versus SS.BN.13. *P ≤ 0.05 versus BN and SS.BN.13, pooled (spontaneous and controlled ventilation); n = 20–22 for each group
Fig. 1. Strain differences in MAC levels for isoflurane. Significantly smaller mean (± SD) end-tidal isoflurane concentrations were required to eliminate movement response to tail clamp in Dahl Salt Sensitive rats (SS) and SS.BN.16 consomics compared with Brown Norway (BN) and SS.BN.13 animals. No differences exist between SS versus SS.BN.16 or between BN versus SS.BN.13. *P ≤ 0.05 versus BN and SS.BN.13, pooled (spontaneous and controlled ventilation); n = 20–22 for each group
Fig. 2. Strain differences in MAC levels for sevoflurane. Significantly smaller mean (± SD) end-tidal sevoflurane were required to eliminate movement response to tail clamp in Dahl Salt Sensitive rats (SS) and SS.BN.16 consomics compared with Brown Norway (BN) and SS.BN.13 animals. There were no differences between SS versus SS.BN.16 or between BN versus SS.BN.13. *P ≤ 0.05 versus BN and SS.BN.13, pooled (spontaneous and controlled ventilation;) n = 20–22 for each group.
Fig. 2. Strain differences in MAC levels for sevoflurane. Significantly smaller mean (± SD) end-tidal sevoflurane were required to eliminate movement response to tail clamp in Dahl Salt Sensitive rats (SS) and SS.BN.16 consomics compared with Brown Norway (BN) and SS.BN.13 animals. There were no differences between SS versus SS.BN.16 or between BN versus SS.BN.13. *P ≤ 0.05 versus BN and SS.BN.13, pooled (spontaneous and controlled ventilation;) n = 20–22 for each group.
Fig. 3. Strain differences in MAC levels for halothane. Significantly smaller mean (± SD) end-tidal halothane were required to eliminate movement response to tail clamp in Dahl Salt Sensitive rats (SS) and SS.BN.16 consomics compared with Brown Norway (BN) and SS.BN.13 animals. There were no differences between SS versus SS.BN.16 or between BN versus SS.BN.13. *P ≤ 0.05 versus BN and SS.BN.13, pooled (spontaneous and controlled ventilation); n = 20–22 for each group.
Fig. 3. Strain differences in MAC levels for halothane. Significantly smaller mean (± SD) end-tidal halothane were required to eliminate movement response to tail clamp in Dahl Salt Sensitive rats (SS) and SS.BN.16 consomics compared with Brown Norway (BN) and SS.BN.13 animals. There were no differences between SS versus SS.BN.16 or between BN versus SS.BN.13. *P ≤ 0.05 versus BN and SS.BN.13, pooled (spontaneous and controlled ventilation); n = 20–22 for each group.
