Atypical neuroleptic malignant syndrome presenting as fever of unknown origin in the elderly.

Abstract: A 79-year-old nursing home resident who was taking olanzapine presented to multiple hospitals with fevers of unknown origin. After an extensive workup to rule out infection, the patient was diagnosed as having neuroleptic malignant syndrome (NMS). This patient's presentation was atypical because of the lack of rigidity or tremor at the time of admission. This case highlights the importance of including NMS in the differential diagnosis of fever of unknown origin in medical patients. A review of the medical literature is taken, focusing on diagnoses, prevalence, predisposing factors, pathophysiology, and current treatment approaches for NMS.

Key Words: adverse drug effect, central nervous system toxicity, elderly, fever of unknown origin, neuroleptic malignant syndrome

**********

Case Report

A 79-year-old male nursing home resident presented in the busy emergency room of a city community teaching hospital with a chief complaint of fever (40.8[degrees]C), weakness, and acute changes in mental status. The patient's medical history was significant for transient ischemic attacks, seizures, Alzheimer dementia, and fevers of unknown origin. He had recently been hospitalized at another institution with similar symptoms, but without a definite diagnosis having been made. The patient's medications at the time of admission included phenytoin, lorazepam, amlodipine, senna, and 10 mg of olanzapine at night, with 5 mg of olanzapine as required up to every 4 hours. The frequency of the doses of olanzapine as required was unknown. Physical examination at time of admission was within normal limits, with no significant findings noted. Therapeutically, the patient initially received acetaminophen, piperacillin/tazobactam, and intravenous fluids, which reduced his fever to 38.2[degrees]C. The diagnostic workup consisted of a lumbar puncture, chest radiography, computed tomographic scan of the head to rule out trauma, computed tomography of the abdomen to rule out infection, phenytoin level, and blood and urine cultures (Table 1). Significant laboratory tests upon admission included a white blood cell count of 8,300 cells/[mm.sup.3] (automated differential of 68.5 neutrophils, 13.2 lymphocytes, and 18.1 monocytes), blood urea nitrogen level of 19 mg/dL, creatinine of 1.5 mg/dL, and a normal urine analysis. An infectious disease consult was obtained, and it was concluded that the fever was most likely due to phenytoin, because no overt nidus for infection was found. Because of the patient's seizure history and possible phenytoin-induced fever, neurology was consulted and a diagnosis of NMS neuroleptic malignant syndrome was entertained because of the patient's use of olanzapine (Table 2). A creatine phosphokinase (CPK) level was obtained and found to be 3,822 U/L. Before the CPK results were known, the patient received a 5 mg dose of olanzapine, after which his temperature quickly rose to 39.2[degrees]C and could not be lowered by the administration of acetaminophen. Dantrolene was given as a slow intravenous push, resulting in a temperature decline from 39.2[degrees] to 37.8[degrees]C by the completion of the infusion. The patient's CPK level dropped to 2,822 U/L 12 hours after the first dose of dantrolene. The patient required an additional dose of dantrolene the next day, when his fever again began to rise. His fever lysed by the second hospital day and his CPK level dropped to 144 U/L by the time of discharge.

Discussion

The first case of NMS was reported in France in 1959 by Delay et al and was coined "syndrome malin des neuroleptics." (1-3) Historically, the prevalence of NMS has been estimated to be between 0.5% to 2.4% of patients taking neuroleptics. However, recent prospective trials estimate the prevalence to be 0.02% to 0.2%, which may reflect a decreased incidence with the newer neuroleptics. (1-7) NMS is characterized by an abnormal mental status (eg, confusion, catatonia, and coma), hyperthermia, "lead-pipe" rigidity, akinesia or dystonia, autonomic instability, rhabdomyolysis, myoclonus, coarse tremors, and, occasionally, cogwheeling. (1-5,8-11) Usually, extra-pyramidal symptoms occur before autonomic symptoms. (4) Common laboratory findings of NMS include increased CPK, leukocytosis (10,000 to 40,000 cells/[mm.sup.3] with a left shift), and myoglobinuria. (1,3,11) Diagnosis can sometimes be confounded by a forme fruste presentation, in which the classic symptoms of hyperthermia and muscular rigidity may be absent or delayed in presentation. (1,12) NMS is usually caused by drugs that antagonize the central nervous system effects of dopamine, especially in the nigrostriatal, hypothalamospinal, and mesolimbic pathways. NMS usually develops within 24 to 72 hours after the administration of a neuroleptic agent. (2-4,8,13) Cases of NMS have also occurred following withdrawal of L-dopa used for the treatment of Parkinson disease. (1,8,14-16) The duration of reported symptoms have varied from 1 to 44 days, with an expected duration of persistence of between 1 to 10 days. (4)

Although NMS is rare, it can cause lethal complications, including autonomic instability, brain damage, acute renal failure, pulmonary embolism secondary to rigidity, aspiration pneumonia secondary to obtundation or dysphagia, fatal arrhythmias, and respiratory failure. (1,2,4,17) Cases complicated by thrombocytopenia and disseminated intravascular coagulation have been reported. (1,18) The overall reported mortality rate for NMS varies between 4% to 20%, with a 50% mortality rate in cases in which renal failure develops. (1,3,4,19)

NMS is an idiosyncratic reaction that is independent of the blood levels of the offending medication. It can occur after a single dose of any neuroleptic drug (eg, phenothiazines, thioxanthenes, or the newer atypical antipsychotics). (1,2,4,20) High-potency neuroleptic drugs with strong D2 receptor antagonism, such as haloperidol and thiothixene, are more likely to cause NMS than are lower potency antipsychotic drugs, such as chlorpromazine or thioridazine. (1,2,4) NMS has also been reported to occur with other classes of medications, such as antiemetics (prochlorperazine), pro-peristaltic agents (metoclopramide), anesthetics (droperidol), and sedatives (promethazine). (2,5) Inciting or predisposing factors include dehydration, malnutrition, exhaustion, intramuscular injection of neuroleptics, advanced age, male sex, neuropsychiatric disorders, traumatic head injuries, agitation, neuroleptic dose increases, organic brain disease (eg, dementia), minor infections, HIV infection, and the concomitant use of lithium, anticholinergic agents, and some antidepressants (eg, venlafaxine). (1,2,4-6,21-24)

The pathophysiology of NMS is still not fully understood but is thought to be caused by central and peripheral dopaminergic blockade that results in muscle rigidity, core temperature elevations, and hypermetabolism. (1-3,19,23,25,26) Since central thermoregulation is mediated by interactions between the noradrenergic, dopaminergic, serotoninergic, and cholinergic pathways, it is unlikely that NMS is due solely to the antidopaminergic effects of neuroleptic drugs. (4) The major cause of hyperthermia is believed to be due to dopaminergic blockade affecting serotoninergic pathways in the hypothalamus, resulting in the establishment of a higher set-point of core temperature. (3,4) It has also been proposed that glutamatergic neurotransmitters are involved in NMS, thus explaining why a medication such as amantadine (an N-methyl-D-aspartate S-type glutamate receptor antagonist) is beneficial in NMS treatment. (1)

The rigidity that occurs with NMS has been hypothesized to be secondary to a peripheral dopaminergic blockade, which results in increased skeletal muscle contractions. (4) Patients who have had NMS have been shown to have multiple genetic defects involving calcium transport and availability in sympathetic neurons, as well as skeletal muscle sarcoplasmic reticulum. Others have been shown to have mutations of the D2 receptor. (3,4,17,22)

An often overlooked and underdiagnosed cause of NMS is the use of neuroleptic medications in medical and surgical settings for nonpsychiatric patients. In a recent review of the literature by Caroff et al, (5) 28 cases of NMS (4 of which resulted in death) were reported on surgical critical care units. Haloperidol and droperidol are among the most commonly reported causes of NMS in intensive care settings and on standard medical units, where they are used to control delirium and nausea and to provide sedation. (5,20)

The first step in the treatment of NMS is to stop the administration of any medication suspected of causing the condition. (1,4,5) Supportive therapies including fever reduction (medications, cooling blankets), hydration, aspiration precautions, deep venous thrombosis prophylaxis, and nutritional support should be routinely initiated. (1,2,4,5) Dantrolene can be used to reduce both core body temperature and peripheral muscle rigidity by inhibiting calcium release from the muscle sarcoplasmic reticulum. (1,4) Dantrolene is administered intravenously at an initial dose of 2 mg/kg, followed by doses of 1 mg/kg repeated every 6 hours as needed, to a maximum of 10 mg/kg per day. (1,4,8) Reduction of core body temperature should be evident within 10 minutes after dantrolene administration. (8) Caution should be used in administrating dantrolene because of its potential hepatotoxicity. Interactions with verapamil may result in severe hyperkalemia, myocardial depression, heart block, or death. (4,27,28)

Bromocriptine, a dopamine agonist, can also be given at a dosage of 2.5 to 10 mg every 6 hours as needed to resolve the clinical symptoms of NMS. (1,4,23) The potential side effects of bromocriptine include hypotension and aggravation of psychosis. (3,4) Dantrolene and bromocriptine can be administered at the same time without any increased adverse effect. (4,8,29) In cases of severe hypertension, sodium nitroprusside infusion reduces blood pressure and lowers core body temperature by increasing heat dissipation from the skin through vasodilatation. (2)

Alternative pharmacologic treatments include amantadine, pancuronium, and L-dopa/carbidopa. (1,4,25) Electroconvulsive therapy with nondepolarizing anesthetic agents has also been shown to be an effective treatment, especially in cases of NMS that have demonstrated symptom progression to catatonic-like stages. (1,3,10,23,30-33) Pharmacologic treatment of NMS can be continued for 1 to 3 weeks to prevent possible recurrence of symptoms. (2,8) Reinstitution of an equipotent neuroleptic results in approximately an 80% relapse rate of NMS. However, only 10% of patients will have NMS if a lower potency neuroleptic or a lower dose of the initiating neuroleptic is used. (4) A low dose of a low-potency neuroleptic can be restarted 2 to 3 weeks after an occurrence of NMS as long as the patient is carefully monitored and followed. (1,3)

The forme fruste presentation of this case makes it noteworthy. This particular patient was treated at numerous hospitals before NMS was identified as the cause of his fever. Factors predisposing him to the development of NMS included being slightly dehydrated, having an underlying dementia and seizure disorder, being male, advanced in age, and having causative medications maintained at the same dose and frequency throughout several episodes of fever of unknown origin.

This patient never had muscular rigidity or tremors, which are usually considered, along with fever, to be the cardinal features of NMS. Initial laboratory work was also misleading, since his initial urinalysis was negative for blood/myoglobin, which would have suggested the rhabdomyolysis, which was subsequently proven by significantly elevated CPK levels. The suspicion for NMS may have also been lowered by the fact the patient was being treated with olanzapine, a newer atypical antipsychotic, which is generally considered to be less likely to induce NMS than older, higher-potency neuroleptic drugs. Observation of the patient's further deterioration, after each administration of olanzapine while in the hospital, appropriately led to the correct diagnosis of NMS.

Conclusion

A diagnosis of NMS should be considered for any patient taking any form of neuroleptic agent who presents with fever of unknown origin with or without documented muscle rigidity.

Behind every great fortune there is a crime.
--Honore de Balzac

Table 1. Neuroleptic malignant syndrome workup (a)

 1. Complete blood count with differential
 2. Serum electrolyte levels
 3. Creatinine and blood urea nitrogen levels
 4. Blood cultures
 5. Urine cultures
 6. Urine analysis to look for myoglobin
 7. Serum creatine phosphokinase
 8. Chest radiograph
 9. Head CT scan to rule out other cause of confusion
10. Lumbar puncture to rule out central nervous system infection
11. Thyroid studies
12. Arterial blood gas, if pulmonary complications, to rule out
    pulmonary embolism

(a) CT, computed tomography.

Table 2. Differential diagnosis for neuroleptic malignant syndrome (a)

 1. Infection (CNS, abscess, sepsis, pulmonary)
 2. Serotonin syndrome
 3. Myocardial infarction
 4. Thyroid storm
 5. Fall or trauma
 6. Severe dehydration
 7. Heat stroke
 8. Drug toxicities (lithium toxicity, cholinergic reaction, monoamine
    oxidase inhibitor)
 9. Vasculitis (primary CNS vasculitis)
10. Malignant hyperthermia if exposed to anesthesia

(a)CNS, central nervous system.

Accepted July 7, 2004.

References

1. Carbone JR. The neuroleptic malignant and serotonin syndromes. Emerg Med Clin North Am 2000;18:317-325.

2. Tsai HC, Kuo PH, Yang PC. Fever, consciousness disturbance, and muscle rigidity in a 68-year-old man with depressive disorder. Chest 2003;124:1598-1601.

3. Bertorini TE. Myoglobinuria, malignant hyperthermia, neuroleptic malignant syndrome and serotonin syndrome. Neurol Clin 1997;15:649-671.

4. Chandran GJ, Mikler JR, Keegan DL. Neuroleptic malignant syndrome: case report and discussion. CMAJ 2003;169:439-442.

5. Caroff SN, Rosenberg H, Mann SC, et al. Neuroleptic malignant syndrome in the critical care unit. Crit Care Med 2002;30:2609-2610.

6. Caroff SN, Mann SC. Neuroleptic malignant syndrome. Med Clin North Am 1993;77:185-202.

7. Keck PE Jr, Pope HG, McElroy SL. Declining frequency of neuroleptic malignant syndrome in a hospital population. Am J Psychiatry 1991;148:880-882.

8. Rosenbaum HK, Miller JD. Malignant hyperthermia and myotonic disorders. Anesthesiol Clin North Am 2002;20:623-664.

9. Desai AK. Use of psychopharmacologic agents in the elderly. Clin Geriatr Med 2003;19:697-719.

10. Ihara M, Kohara N, Urano F, et al. Neuroleptic malignant syndrome with prolonged catatonia in a dopa-responsive dystonia patient. Neurology 2002;59:1102-1104.

11. Pelonero AL, Levenson JL, Pandurangi AK. Neuroleptic malignant syndrome: a review. Psychiatry Serv 1998;49:1163-1172.

12. Lev R, Clark RF. Neuroleptic malignant syndrome presenting without fever: case report and review of the literature. J Emerg Med 1994;12:49-55.

13. Gurrera RJ. Sympathoadrenal hyperactivity and the etiology of neuroleptic malignant syndrome. Am J Psychiatry 1999;156:169-180.

14. Keyser DL, Rodnitzky RL. Neuroleptic malignant syndrome in Parkinson's disease after withdrawal or alteration of dopaminergic therapy. Arch Intern Med 1991;151:794-796.

15. Toru M, Matsuda O, Makaguchi K, Sugano K. Neuroleptic malignant syndrome-like state following a withdrawal of antiparkinsonian drug. J Nerv Ment Dis 1981;169:324-327.

16. Ong KC, Chew EL, Ong YY. Neuroleptic malignant syndrome without neuroleptics. Singapore Med J 2001;42:85-88.

17. Gurrera RJ. Is neuroleptic malignant syndrome a neurogenic form of malignant hyperthermia? Clin Neuropharmacol 2002;25:183-193.

18. Eles GR, Songer JE, Di Pette DJ. Neuroleptic malignant syndrome complicated by disseminated intravascular coagulation. Arch Intern Med 1984;144:1296-1297.

19. Shalev A, Hermesh H, Munitz H. Mortality from neuroleptic malignant syndrome. J Clin Psychiatry 1989;50:18-25.

20. O'Grady NP, Barie PS, Bartlett J. Practice parameters for evaluating new fever in critically ill adult patients: Task Force of the American College of Critical Care Medicine of the Society of Critical Care Medicine in collaboration with the Infectious Disease Society of America. Crit Care Med 1998;26:392-408.

21. Keck PE Jr, Pope HG, Cohen BM, et al. Risk factors for neuroleptic malignant syndrome: a case-controlled study. Arch Gen Psychiatry 1989;46:914-918.

22. Keck PE, Caroff SN, McElroy SL. Neuroleptic malignant syndrome and malignant hyperthermia: end of a controversy? J Neuropsychiatry Clin Neurosci 1995;7:135-144.

23. Bond WS. Detection and management of the neuroleptic malignant syndrome. Clin Pharm 1984;3:302-307.

24. Breitbart W, Marotta RF, Call P. AIDS and neuroleptic malignant syndrome. Lancet 1988;2:1488-1489.

25. Henderson VW, Wooten GF. Neuroleptic malignant syndrome: a pathophysiologic role for the dopamine receptor blockade? Neurology 1981;31:132-137.

26. Guze BH, Baxter LR. Neuroleptic malignant syndrome. N Engl J Med 1985;313:163-166.

27. Lynch C III, Durbin CG, Fisher NA, et al. Effects of dantrolene and verapamil on atrioventricular conduction and cardiovascular performance in dogs. Anesth Analg 1986;65:252-258.

28. Saltzman LS, Kates RA, Corke BL, et al. Hyperkalemia and cardiovascular collapse after verapamil and dantrolene administration in swine. Anesth Analg 1984;63:474-478.

29. Rosenberg MR, Green M. Neuroleptic malignant syndrome: review of response to therapy. Arch Intern Med 1989;149:1927-1931.

30. Hermesh H, Aizenberg J, Weizman A. A successful electroconvulsive treatment of neuroleptic malignant syndrome. Acta Psychiatr Scand 1987;75:237-239.

31. Lazarus A. Treatment of neuroleptic malignant syndrome with electroconvulsive therapy. J Nerv Ment Dis 1986;174:47-49.

32. McKinney P, Kellner C. Multiple ECT late in the course of neuroleptic syndrome. Convuls Ther 1997;13:269-273.

33. Verwiel JM, Verwey B, Heinis C, et al. Successful electroconvulsive therapy in a pregnant woman with neuroleptic malignant syndrome. Ned Tijdschr Geneeskd 1994;138:196-199.

RELATED ARTICLE: Key Points

* Neuroleptic malignant syndrome is a rare but important condition to be included in the differential diagnosis of fever of unknown origin for medical patients, especially those with predisposing factors.

* Neuroleptic malignant syndrome may have a forme fruste presentation, with delayed onset of or the absence of rigidity.

* Neuroleptic malignant syndrome can be caused by any neuroleptic medication (phenothiazines, thioxanthenes, or the newer atypical antipsychotics) and by other classes of medication commonly used in hospitals, such as antiemetics (prochlorperazine), pro-peristaltic agents (metoclopramide), anesthetics (droperidol), and sedatives (promethazine).

* The most significant factor for predicting death in cases of neuroleptic malignant syndrome (50% correspondence rate) is the new onset or worsening of renal failure.

* Continuing a patient on an equipotent neuroleptic medication at the same dose as before an episode of neuroleptic malignant syndrome leads to a relapse rate of 80%.

Ryan C.W. Hall, MD, Brian Appleby, MD, and Richard C.W. Hall, MD

From Johns Hopkins/Sinai Hospital, Department of Medicine, Baltimore, MD; Johns Hopkins Hospital, Department of Psychiatry, Baltimore, MD; and the Department of Psychiatry, University of Florida, Gainesville, FL.

The authors received no financial support in the production of this manuscript. The authors have no commercial or proprietary interest in any drugs mentioned in this article.

Reprint requests to Dr. Ryan C. W. Hall, 4503 Lyons Run Circle, #101, Owings Mills, MD 21117-6317. E-mail: [e-mail]