Phenytoin loading in chronic alcoholic patients

Original Contributions

Phenytoin Loading in Chronic Alcoholic Patients ROBERT D. WELCH, MD,* DONALD B. SMITH, Jr, MD,* STEVE F. DANOSI, MD,t MARK L. ZWANGER, MD,+ BROOKS F. BOCK, MDt$ Controversy exists concerning the appropriate loading dose of phenytoin in chronic alcoholic patients. Chronic alcoholics are frequently assumed to have low albumin levels secondaryto malnutrition and liver disease. Phenytoin is bound to albumin, and therefore the usual loading dose of phenytoin might result In a higher percentage of unbound drug and increased toxicity in these patients. Thirty-six chronic alcoholic patients were given a 15-mg/kg loading dose of phenytoin by constant intravenous infusion. After the infusion, patients were evaluated for clinical signs of phenytoin toxicity. At 1 hour after infusion, blood was sent for determination of total phenytoin, free phenytoin, and albumin levels. Fifteen patients were hypoalbuminemic (mean, 3.4 g/dL); 21 patients had albumin levels within the normal range (mean, 4.3 g/dL). In the hypoalbuminemic group, the mean free phenytoin level was 1.1 pg/mL, and the mean total phenytoin level was 13.6 pg/mL. In patients with normal albumin levels, the mean free phenytoin level was 1.3 pg/mL, and the mean total phenytoin level was 15.7 P,g/mL. There were no statistically sig nificant differences in total phenytoin or free phenytoin levels between either groups. No patient had a postinfusion phenytoin level (either free or total) within the toxic range. Although our sample size was small, our results suggest that a lCmg/kg loading dose of phenytoin does not produce toxic levels in chronic alcoholics. (Am J Emerg Med 1988;6:89-92)

The intravenous infusion of a loading dose of phenytoin has been shown to be a safe and effective way From the *Department of Emergency Medicine, Detroit Receiving Hospital; and the tDepartment of Surgery and *Section of Emergency Medicine, Wayne State University School of Medicine, Detroit, Michigan. Manuscript received 23 December 1986; revision accepted 8 June 1987. Presented at the Annual Meeting of the University Association for Emergency Medicine, Portland, Oregon, May 1986. Address reprint requests to Dr. Welch: Section of Emergency Medicine, Detroit Receiving Hospital, 4201 St. Antoine, Detroit, Ml 48201. Key Words: Albumin, chronic alcoholism, drug toxicity, intravenous loading, phenytoin.

01988 W.B. Saunders Company 07356757188 $0.00 + .25

of rapidly achieving therapeutic concentrations of phenytoin.14 Phenytoin is protein bound in the range of 90 to 95% with only a small percentage remaining free in plasma.5 The free phenytoin fraction correlates better with clinical signs of toxicity than total phenytoin levels.6 Chronic alcoholic patients often have low albumin levels because of poor nutrition and impaired liver synthesis. In patients with low albumin levels, there is the potential for a larger percentage of unbound (free) phenytoin. A standard loading dose of phenytoin in patients with low albumin levels might therefore result in a higher, and possibly toxic, free phenytoin level. Because of this potential for phenytoin toxicity, it has been suggested that this patient population might require a reduced loading dose of phenytoin.’ Tozer and Winter,* in contrast, suggest that little or no change in the loading dose of phenytoin is required, because as the albumin concentration is reduced, the volume of distribution increases. This increased volume of distribution will result in a lower total phenytoin level with a free phenytoin fraction within the therapeutic range.’ No studies have looked at intravenous phenytoin loading, free phenytoin levels, and albumin levels in the chronic alcoholic patient. We prospectively evaluated whether a 15-mg/kg intravenous loading dose of phenytoin would result in increased toxicity in the chronic alcoholic. MATERIALS AND METHODS Patients presenting to emergency department of the Detroit Receiving Hospital with a history of chronic alcoholism and requiring a loading dose of phenytoin for control of a major motor seizure were considered for this study. Chronic alcoholism was determined by the attending physician from the patient’s history of alcohol use and frequency of alcohol-related disease. Alcohol-related disease includes prior emergency department or hospital admissions for alcohol intoxication, delirium tremens, pancreatitis, alcohol withdraw89

AMERICAN

JOURNAL

OF EMERGENCY

MEDICINE

n Volume 6, Number 2 I March 1988

al, and cirrhosis. The criteria used to exclude patients from the study are listed in Table 1. This study was approved by the Human Investigation Committee of Wayne State University. Each patient was weighed and examined for confusion, ataxia, dizziness, and horizontal and vertical nystagmus. An intravenous line delivering 0.9% normal saline was inserted, and blood was assayed for albumin, blood urea nitrogen, creatinine , glucose, and total phenytoin levels. The patient then received a loading dose of 15 mg/kg of phenytoin dissolved in 50 to 100 mL of normal saline by IVAC infusion at a rate of 25 mg/min. Vital signs were recorded at the beginning and end of the infusion. All patients had continuous electrocardiographic monitoring during the infusion. Immediately after the infusion, patients were reexamined (by the same physician) for any signs of phenytoin toxicity (confusion, ataxia, nystagmus, or dizziness). If the patient had any of these signs or symptoms prior to the infusion, they were examined closely for worsening of these signs after infusion. Patients were reported as having signs of toxicity if current signs worsened or if new signs developed. At 1 hour postinfusion, blood was sent for determination of total and free phenytoin levels. Patient disposition was then determined by the attending physician. The total phenytoin level was obtained with the DuPont ACA analyzer (Du Pont, Wilmington, DE), which is an adaptation of Syva’s EMIT system (Syva, Palo Alto, CA). The free phenytoin level was determined by separating the free phenytoin fraction from the protein-bound fraction by ultrafiltration using the Centrifree micropartition system (Amicon Corp., Danvers, MA). The free fraction was then analyzed by fluorescent polarization immunoassay on the Abbott TDx analyzer (Abbott Laboratories, Abbott Park, IL). In our laboratory, therapeutic total phenytoin levels range from 10 to 20 pg/mL, with toxic levels greater TABLE 1.

Exclusion

Criteria’

Status epilepticus Marked bradycardia of sinus, nodal, or ventricular origin Conduction disturbances such as third-degree heart block atrial fibrillation or flutter Measurable phenytoin level in the first blood sample Uremia or azotemia (blood urea nitrogen >40 mg/dL or creatinine >2.5 mg/dL) Hypotension not immediately responsive to fluid challenge Inability to weigh the patient Any drug use that may alter phenytoin binding (salicylates, phenylbutazone, valproic acid) Hypoglycemia Clinical jaundice New-onset seizures Patients who exhibited any of these signs or symptoms excluded from the study.

l

or

were

TABLE 2.

Causes of Seizures

in Chronic

Alcoholics No. (%)

Alcohol related Head trauma Cerebrovascular disease Unknown Idiopathic epilepsy

20 (55.8) 8 (22.2) 3 (8.3) 3 (8.3) 2 (56)

than 30 pg/mL. Therapeutic free phenytoin levels range from 1 to 2 @mL, and toxic levels are greater than 4 &mL. Normal albumin levels are 3.8 to 4.8 g/dL. Statistical analysis was done using the two-tailed Student’s t-test. RESULTS Thirty-six patients were entered into the study (31 men, 5 women). The causes of the seizures were determined by patient history and the medical records chart, and these are listed in Table 2. No patients with new-onset seizures were included in the study. Fourteen patients had greater than 10 admissions to Detroit Receiving Hospital for ethanol intoxication or complications of ethanol abuse; 5 patients had between 5 and 10 admissions. The remaining 17 patients had from 0 to 4 prior admissions to our facility for alcohol abuse. Fifteen patients had albumin levels of less than the normal range. The mean value in this group was 3.4 g/dL (range, 3.1 to 3.7 g/dL). Twenty-one patients had albumin levels greater than 3.8 g/dL (mean, 4.3). The mean albumin values in both groups were significantly different (p < 0.001). The mean albumin concentration in all patients was 3.9 g/dL. In the hypoalbuminemic group, the mean postinfusion free phenytoin level was I. 1 pg/mL, and the average total phenytoin level was 13.6 kg/mL. In patients with normal albumin levels, the mean postinfusion free phenytoin level was 1.3 pg/mL, and the mean total phenytoin level was 15.7 ug/mL. In all patients, the mean postinfusion free phenytoin level was 1.2 ).&rnL, and the mean total phenytoin level was 14.8 t&mL. There was no difference in total and free phenytoin levels between the hypoalbuminemic and normal albumin level groups. In the hypoalbuminemic patient, 8.3% of the phenytoin was unbound. In the normal albumin group, 8.2% of drug was unbound. These results are summarized in Table 3. No patient had a postinfusion phenytoin level (either free or total) in the toxic range. Four patients developed clinical signs of toxicity. One hypoalbuminemic patient complained of dizziness. In the normal albumin group, two patients developed nystagmus and one pa-

WELCH ET Ai n PHENYTOIN LOADING IN ALCOHOLICS

TABLE3.

Variables in 36 Chronic Alcoholics Given Phenytoin

Mean age (yrs) Sex (M/F) Mean albumin (g/dL) Mean total phenytoin (pg/mL) Mean free phenytoin (pg/mL) Unbound drug (range) (%)

Hypoalbuminemic

Normal Albumin

All Patients

41.7 (24-64) 1213 3.4 2 0.20 13.6 f 3.9 1.1 f 0.36 6.3 (5.4-10.6)

39.1 (25-56) 1912 4.3 * 0.46’ 15.7 2 3.1 1.3 2 0.40 6.2 (4.6-15.9)

40.2 (24-64) 3115 3.9 + 0.56 14.6 f 3.6 1.2 f 0.39 6.3 (4.6-15.9)

The only statistically significant difference was between the mean albumin levels in the hypoalbuminemic normal albumin values (p < 0.001). Data are given as mean +- 1 SD where appropriate.

l

tient became ataxic. Two of these three patients with normal albumin levels had plasma alcohol levels of 333 and 296 mg/dL, respectively (Table 4). Four patients complained of burning at the intravenous site. Two patients had a drop in blood pressure of 20 mm Hg, which immediately responded to fluid challenge. No changes in cardiac rhythm were noted during the infusion. Twelve patients had postinfusion free phenytoin levels of less than 1 pg/mL (subtherapeutic). Five of these patients were hypoalbuminemic, whereas the other seven patients had normal albumin levels. Four patients had total phenytoin levels in the subtherapeutic range (< 10 pg/mL). One patient, who had a total phenytoin level of 12.9 pg/mL (therapeutic) and a free phenytoin level of 0.9 pg/mL (subtherapeutic), had a seizure. That patient received an additional 300 mg of phenytoin intravenously without any further complications. No patient had a free or total phenytoin concentration in the toxic range. Patient disposition was not altered or dealyed because of phenytoin loading. DISCUSSION Alcohol abuse plays an important role in many patients with seizures. Earnest and Yarnell’ found that 41% of 472 patients admitted to a city hospital with seizures had a history of alcohol abuse. Of those patients with a history of ethanol abuse, 80% were male. Our cohort of alcoholic patients had a similarly high male preponderance (86%). This preponderance is in contrast to an even sex distribution seen in a study population with nonalcoholic seizures.’ Treatment of the alcoholic patient with a seizure disorder depends on the cause of the seizure. The treatment of alcohol withdrawal seizures remains controversial, but phenytoin is advocated by some authors.iO~” The recommended intravenous loading dose of phenytoin varies from 15 to 18 mg/kg.2,3,‘2-15 It has been suggested that the intravenous loading dose of phenytoin be reduced in chronic alcoholic patients.’ No studies, however, have examined this method of phenytoin loading exclusively in chronic alcoholics. We

patients and patients with

chose the lower loading dose of 15 mg/kg to reduce the potential of excessive phenytoin toxicity. Phenytoin is highly protein bound, usually in the range of 90 to 95%.’ Our results (range, 84.1 to 95.4%) agree with these data. Booker and Darcey6 have shown that clinical toxicity correlates better with the unbound fraction of phenytoin rather than the total phenytoin concentration. The therapeutic importance of monitoring free phenytoin levels has been well documented. DeMonaco and Lawlessi found a marked variability of phenytoin binding among epileptic patients and emphasized the value of routine free phenytoin levels. Porter and Layzer” as well as others’B-21 have shown that conditions such as uremia and hypoalbuminemia result in an increased percentage of unbound drug. In these patients, the total level may be within the therapeutic range while the free level is in the toxic range. In our study, we measured both free and total phenytoin levels, because total phenytoin levels may not accurately reflect clinical signs of toxicity in patients with hypoalbuminemia. Ramsay et al.** have shown that phenytoin rapidly crosses the blood-brain barrier, reaching peak levels in about 6 minutes, with brain concentrations remaining stable for 60 minutes. More than 50 minutes are required to reach a steady-state brain-to-plasma ratio, however.** Wilder and coworkers23 also found rapid entry of phenytoin into the brain but discovered that brain tissue concentrations increased slightly over a 60-minute period. Blood samples were drawn at 1 hour after the end of infusion to guarantee that this steadystate had been reached.

TABLE4. Toxicity in Chronic Alcoholics Given Phenytoin

Patient 5 19 24 36

Sign of Toxicity

Albumin Level (g/dL)

Free Phenytoin Level (cLg/mL)

Total Phenytoin Level (pg/mL)

Dizzy Nystagmus Ataxia Nystagmus

3.4 3.9 4.1 6.0

1.4 1.6 1.6 1.8

15.6 19.9 17.8 19.2 91

AMERICAN

JOURNAL

OF EMERGENCY

MEDICINE

n Volume 6, Number 2 n March 1988

Levels obtained 24 hours after acute loading are likely to drop considerably with tissue redistribution and metabolism. Therefore the loading dose chosen may not have produced persistently therapeutic levels prior to the next phenytoin dose. Although a small number of patients were involved in this study, no patient had a total or free phenytoin level in the toxic range after infusion of 15 mg/kg of phenytoin. In the patients who developed clinical signs of toxicity, three of four had normal albumin levels. Two of these three patients had markedly elevated plasma alcohol levels, which may have contributed to their symptoms. Twelve of our patients had free phenytoin levels in the subtherapeutic range. Of these 12 patients, only 4 had subtherapeutic total phenytoin levels. Thus, 8 patients had subtherapeutic free levels that would not have been detected if only total phenytoin levels had been measured. This demonstrates the importance of monitoring free phenytoin levels. These data also suggest that many patients require a larger loading dose. Forty-one percent of our patients were hypoalbuminemic. This degree of hypoalbuminemia does not appear to be clinically significant because none of these patients had phenytoin levels in the toxic range after receiving a lSmg/kg loading dose. CONCLUSION There was no evidence of major toxicity clinically and no evidence of toxicity by serum phenytoin levels after alcoholic patients were given a lSmg/kg loading dose of phenytoin. Although a few patients developed minor clinical signs of toxicity such as nystagmus, these did not interfere with the final disposition of these patients. This study contains a small number of patients, but it also suggests that many alcoholic patients may actually require a higher loading dose of phenytoin. REFERENCES 1. Boike SC, Rybak MJ, Tintinalli JE, et al: Evaluation of a method for intravenous phenytoin infusion. Clin Pharm 1983;2:444466 2. Carducci B, Hedges JR, Beal JC, et al: Emergency phenytoin loading by constant intravenous infusion. Ann Emerg Med 1984;13:1027-1031 3. Salem RB, Wilder BJ, Yost RL: Rapid infusion of phenytoin sodium loading dose. Am J Hosp Pharm 1981;38:355-357

92

4. Earnest MP, Marx JA, Drury LR: Complications of intravenous phenytoin for acute treatment of seizures: recommendations for usage. JAMA 1983;249:762-765 5. Lunde PKM, Rane A, Yaffe SJ, et al: Plasma protein binding of diphenylhydantoin in man: interaction with other drugs and the effect of temperature and plasma dilution. Clin Pharmacol Ther 1970;11:844-855 6. Booker HE, Darcey B: Serum concentrations of free diphenylhydantoin and their relationship to clinical intoxication. Epilepsia 1973;14:177-184 7. Daily R: Grand mal seizures in an alcoholic. Hosp Ther 1985;10(2):53-61 8. Tozer TN, Winter ME: Phenytoin. In Evans WE, Schentag JJ, Jusko WJ (eds): Applied Pharmacokinetics. San Francisco, Applied Therapeutics, 1980, pp 275-318 9. Earnest MP, Yarnell PR: Seizure admissions to a city hospital: the role of alcohol. Epilepsia 1976;17:387-393 10. McMicken D: Seizures in the alcohol-dependent patient: a diagnostic and therapeutic dilemma. J Emerg Med 1984;1:311-316 11. Sampliner R, lber FL: Diphenylhydantoin control of alcohol withdrawal seizures. JAMA 1974;230:1430-1432 12. Gal P, McCue JD, Tate M, et al: The beneficial effects of phenytoin loading dose on seizure recurrence in patients with acute repetitive seizure: a preliminary report. NC Med J 1984;45:153-154 13. Cranford RE, Leppik IE, Patrick 8, et al: Intravenous phenytoin: clinical and pharmacokinetic aspects. Neurology 1978;28:874-880 14. Cranford RE, Leppik IE, Patrick B, et al: Intravenous phenytoin in acute treatment of seizures. Neurology 1979;29:1474-1479 15. Delgado-Escueta AV, Wasterlain C, Tremain DM, et al: Management of status epilepticus. N Engl J Med 1982;306:1337-1340 16. DeMonaco HJ, Lawless LM: Variability of phenytoin protein binding in epileptic patients. Arch Neurol 1983;40:481483 17. Porter RJ, Layzer RB: Plasma albumin concentrations and diphenylhydantoin binding in man. Arch Neurol 1975;32:298-303 18. Odar-Cederlof I, Borga 0: Kinetics of diphenylhydantoin in uraemic patients: consequences of decreased plasma protein binding. Eur J Clin Pharmacol 1974;7:31-37 19. Boston Collaborative Drug Serveillance Program: Diphenylhydantoin side effects and serum albumin levels. Clin Pharmacol Ther 1973;14:529-532 20. Bauer LA, Edwards WAD, Dellinger EP, et al: Importance of unbound phenytoin serum levels in head trauma patients. J Trauma 1983;23:1058-1060 21. Kiloatrick CJ. Wanwimolruk S. Wina LMH: Plasma concentrations of unbound phenytoin in-the management of epilepsy. Br J Clin Pharmacol 1984;17:539-546 22. Ramsay E, Hammond EJ, Perchalski RJ, et al: Brain uptake of phenytoin, phenobarbital, and diazepam. Arch Neurol 1979;36:535-539 23. Wilder BJ, Ramsay RE, Willmore LJ, et al: Efficacy of intravenous phenytoin in the treatment of status epilepticus: kinetics of central nervous system penetration. Ann Neurol 1977;1:511-518