Original Contribution

Evaluation of intramuscular fosphenytoin vs intravenous pHenytoin loading in the ED^?

Patrick Michael Aaronson PharmD ^a,?, Bernadette S. Belgado PharmD ^a,

Joseph P. Spillane PharmD ^a, Thomas A. Kunisaki MD ^b

^aDepartment of Pharmacy Shands Jacksonville, College of Pharmacy/University of Florida, Box C-89,

Jacksonville, FL 32209, USA

^bDepartment of Emergency Medicine, University of Florida and Shands Jacksonville, Box C-89, Jacksonville, FL 32209, USA

Received 9 March 2010; revised 6 May 2010; accepted 7 May 2010

Abstract

Objective: A comparison of length of stay in an emergency department (ED) after loading patients at risk for seizures with either intravenous (IV) phenytoin or intramuscular fosphenytoin was studied. Methods: This was a retrospective observational cohort study that was conducted over a 24-month period in an academic teaching hospital (693 beds). Patients included were 18 years or older, discharged from the ED without hospital admission, and loaded with either IV phenytoin or IM fosphenytoin. The primary end point was the comparison of length of stay in the ED until discharge after loading. Characterization of seizure etiology, cardiac risk factors, and adverse drug events were also observed. Results: A total of 51 patients were evaluated who received IV phenytoin compared with 59 for IM fosphenytoin. The median time-to-discharge difference between IV phenytoin vs IM fosphenytoin was 1:49 hours (95% confidence interval, 1:24-2:24 hours; P b .001). There was no statistical difference in cardiac risk factors and occurrence of adverse drug events between groups.

Conclusions: This study found that patients were discharged from the ED earlier with the loading of IM fosphenytoin compared to IV phenytoin.

(C) 2011

Introduction

Patients presenting with a complaints relating to seizures represent approximately 1% to 2% of all emergency department (ED) visits in the United States [1]. Approxi- mately 28% of the epilepsy populace require treatment in the ED annually [2,3]. Parenteral anticonvulsants are often crucial in the ED for patients unable to take oral formula-

^? The authors of this study have no conflict of interest.

* Corresponding author. Tel.: +1 904 393 2688 (Pager).

E-mail addresses: [email protected] (P.M. Aaronson), [email protected] (B.S. Belgado), [email protected] (J.P. Spillane), [email protected] (T.A. Kunisaki).

tions. Available since 1956, intravenous (IV) phenytoin serves as a first-line agent for non-emergent supplementation of subtherapeutic Serum phenytoin levels or “loading” in those stabilized after recent onset of an acute seizure [4].

Usefulness of IV phenytoin in the ED stems from physician experience, proven efficacy, negligible sedation, and minimal effects on Respiratory function [4]. Conversely, monitoring and managing the notorious adverse cardiac and extravasation events continue to be a challenge in the ED. These adverse effects are primarily ascribed to the excipients used to solubilize the weakly acidic phenytoin molecule: propylene glycol 40%, ethanol 10%, and sodium hydroxide (adjusted to a pH 12) [5]. The alkaline/ethanol vehicle is associated with tissue damaging effects contributing to an

0735-6757/$ - see front matter (C) 2011 doi:10.1016/j.ajem.2010.05.005

increased risk for extravasation ranging from mild local edema to severe skin sloughing and necrosis [6,7]. The incidence of local cutaneous infusion reactions occurs in approximately 25% of patients [8], whereas a characteristic Purple glove syndrome (delayed progressive distal limb edema leading to discoloration and pain) has an incidence of 3% to 7% [9]. In addition, propylene glycol has been linked to cardiac complications via sympatholytic vagal reflex and vasodilatory response associated with its rapid infusion [10,11]. Administration of IV phenytoin requires continuous cardiac telemetry and Blood pressure monitoring due to the risk of hypotension, bradycardia, myocardial depression, and cardiac arrhythmias that could potentially result in cardiac arrest [4,12]. For these reasons, the manufacturer of phenytoin revised its package insert in 1983 to reflect a maximum IV infusion rate of 50 mg/min [12].

Fosphenytoin (Cerebyx; Pfizer, New York, NY), avail- able since 1996, was developed to overcome the limitations of the parenteral form of phenytoin [13]. Fosphenytoin is a disodium phosphate ester prodrug that is hydrolyzed to phenytoin through phosphatases found in plasma and vascular tissues [14]. The half-life conversion of fospheny- toin to derived phenytoin occurs within 8 to 15 minutes in healthy subjects independent of dose or rate of administra- tion [15,16]. Counterintuitively, a delay in obtaining therapeutic phenytoin levels with IV fosphenytoin compared to IV phenytoin has not been shown to be clinically significant [17]. Of note, fosphenytoin is distinctively prescribed in phenytoin equivalence (PE) in attempts to avoid molecular-weight-based adjustments when converting from phenytoin; nonetheless, this has been a source of misinterpretation [18,19]. Advantages of fosphenytoin are attributed to improved water solubility in the absence of propylene glycol and ethanol; thus, exhibiting a much less caustic (pH 8.8) solution [20]. Intravenous fosphenytoin has an improved safety profile and can be infused 3 times faster (maximum: 150 mg PE/min); nonetheless, its utilization also requires telemetry and BP monitoring due to the negative inotropic action on the myocardium from derived phenytoin when fosphenytoin is metabolized [18,21,22]. In addition, unlike IV phenytoin, IV fosphenytoin is associated with unique transient infusion rate-related pruritis (21%) [16] and paresthesia (2.5%) [23] similar to other phosphate ester drugs.

To obviate the need for telemetry and strict hemodynamic monitoring requirements, fosphenytoin may be administered intramuscularly (IM). This route is devoid of infusion-related Cardiac effects and has decreased occurrence of pruritis (4%- 14%) and paresthesia (rare) in contrast to IV fosphenytoin infusion [15,16]. It is important to note the unique amount of fosphenytoin that can be administered IM, it is acceptable to give up to 20 mL within a single injection into a large muscle group [18]. This challenges the traditional practice of limiting IM injections to 5 mL; however, studies have shown that because of its unique physiochemical properties, large volume IM loading is well tolerated [22,24,25].

Intramuscular fosphenytoin offers significant advantages over IV phenytoin with its improved safety profile and ease in administration (less than a minute). Furthermore, IM administration of fosphenytoin precludes the need for cardiac telemetry, is 100% bioavailable, and has the ability to produce therapeutic phenytoin serum concentrations similar to IV phenytoin (within 30 minutes) [26,27]. Conversely, IM administration of phenytoin is not recommended because of erratic absorption, severe pain, and tissue damage (ranging from swelling to painful muscle necrosis) attributed to precipitation of phenytoin crystals at the injection site [28]. For these reasons, IM fosphenytoin appears to be an encouraging option for administration to patients in the ED setting that may result in a safer earlier discharge from the hospital.

Intramuscular fosphenytoin was approved through our Pharmacy and Therapeutics Committee for non-emergent loading of seizure patients in the ED. Encouraging IM fosphenytoin use over IV phenytoin may decrease ED overcrowding, wait times, and time-to-safe discharge; in addition to an increase in the availability of cardiac monitored beds for more critical patients. Length-of-stay (administration costs), telemetry (monitoring costs), and adverse events (outcome costs) are primary drivers of total expenditure for anticonvulsant loading in the ED [29]. The purpose of this study is to evaluate the use of IV phenytoin vs IM fosphenytoin in patients discharged from the ED for treatment of non-emergent seizures with respect to length of stay (LOS) and adverse outcomes.

Methods

Setting and design

This study was a retrospective observational cohort study that reviewed patient medical records over a 24-month period in the ED of an academic urban teaching hospital (693 beds). The ED serves an indigent population, contains a level I trauma center, and receives approximately 90 000 visits per year. Based on national statistics, 2 to 6 seizure patients are seen per day [1-3]. This study was approved by the institutional review board and was an exempt study with a waiver of informed consent because of its retrospective nature. The primary objective of this study was to compare the LOS in patients who received a loading dose of IV phenytoin vs IM fosphenytoin in the emergency room intensive care unit (ERICU) of the ED. The secondary objective included an evaluation of adverse drug events for both medications.

Data collection

Patients were included if they were 18 years or older and received a non-emergent loading dose of either IV phenytoin

or IM fosphenytoin in the ERICU. Patients were excluded if they were admitted to the hospital from the ERICU during the same visit, if their medical record or time of discharge were not available.

A master list was generated to identify patients who received IV phenytoin or fosphenytoin withdrawn from the automated medication dispensing cabinet in the ERICU over a 24-month period. Subsequently, this list was randomized and patients were selected until the a priori sample size was reached. Medical records were selected and reviewed by the primary investigator. Medical records were evaluated for patient demographics, medical history, ED chief complaint and reason for visit, seizure etiology, electrocardiogram readings, dose of medications, infusion rates, route of administration, time phenytoin and fosphe- nytoin was removed from the automated medication dispensing cabinet, medication order details, serum phenyt- oin concentrations, BP monitoring, Adverse drug reactions, and discharge information.

Time-to-discharge was defined as the time phenytoin or fosphenytoin was removed from the automated medication dispensing cabinet to the time the patient was discharged from the ED as documented in the medical record. Non- emergent risk for convulsions was defined as non-acute, non- status epilepticus, or stabilized after a recent acute seizure episode [4]. Cardiac risk factors in this study were defined as any one of the following: age 60 years or older, valvular heart disease, congestive heart failure, cardiomyopathy, pericardial disease, coronary artery disease, history of myocardial infarction, preexisting arrhythmias, chronic stable or unstable angina, peripheral vascular disease, pulmonary hypertension, previous transient ischemic attack, or cerebrovascular accident. Significant hypotension from IV phenytoin or IM fosphenytoin was defined as BP decrease greater than 20 mm Hg systolic and/or 10 mm Hg diastolic [4].

Statistical analysis

A priori sample size was determined to consist of 50 subjects in each group to detect a 15% difference in LOS between the IV phenytoin and IM fosphenytoin groups. A P value of less than .05 was considered the level of significance for inferential statistical analysis. An ? error was set at 0.05 with an 80% power and a standard deviation within 60%. All descriptive data are expressed as the mean +- SD unless otherwise specified. Group comparisons of demographic characteristics were evaluated using Student t test for continuous variables, and ?2 or Fisher exact test where appropriate for discrete variables. Primary outcome of time- to-discharge between IV phenytoin and IM fosphenytoin was evaluated using the Mann-Whitney U test (2-tailed) for nonparametric data. Post hoc analysis of potential confoun- ders was analyzed with 2-way analysis of variance. Standardized abstraction form was used to ensure uniform handling of data and evaluated periodically by the research committee. Data end points were entered into Microsoft

Excel 2007 version 12, and statistical analysis was computed using Analyse-it version 1.73, UK.

Results

Five hundred eighty-nine medical records were reviewed during the study period. Of these, 141 patients were excluded due to inaccessible medical record and 328 were patients admitted to the hospital from the ED. A total of 51 patients in the IV phenytoin group and 59 patients in the IM fosphenytoin group were analyzed (Fig. 1). Both groups were similar in all characteristics with the exception of seizure etiology. There was a statistical difference between groups with respect to noncompliance or lack of antiepileptic medication (47% vs 23% for IV phenytoin and IM fosphenytoin, respectively, P = .037); however, this statistical significance influencing the primary outcome was not found (P = .235). It would be difficult to formulate a conclusion due to the percentage of unknown seizure etiologies (47% vs 73% for IV phenytoin and IM fosphenytoin respectively, P = .00020 (Table 1).

There was a significant difference in time-to-discharge between IV phenytoin vs IM fosphenytoin (P b .001). The median difference of time-to-discharge between groups was 1:49 hours (95% confidence interval, 1:24-2:24 hours) (Table 2). Approximately 85% of the IM fosphenytoin group was discharged from the ED within the first 2 hours of receiving medication compared to only 28% of IV phenytoin patients (Fig. 2).

Statistically, there was no difference in adverse drug events between groups. The only adverse event documented was hypotension, which was comparable between groups. Patients with reported decrease in systolic BP of greater than 20 mm Hg was 17% and 7% (P = .918) for IV phenytoin and IM fosphenytoin, respectively. Patients with reported decrease in Diastolic BP greater than 10 mm Hg were 30% and 7% (P = .352) for IV phenytoin and IM fosphenytoin, respectively. Hemodynamic measurement was not consistent in relation to the time the anticonvulsant dose was given. In

Fig. 1 Summary of excluded patients. PTs indicates patients.

addition, IV phenytoin was administered in 45% of the patients who had underlying cardiac risk factors, however, had no document adverse effects.

Table 1 Demographic and clinical characteristics (n = 110)

IV PHT (n n (%)	=	51)	IM FOS (n n (%)	=	59)	P
Mean age (y) 41.83	45.09			.260
Sex
Men 31 (61%)	37 (63%)			.771
Women 20 (39%)	22 (37%)			.771
Race
African 25 (51%)	36 (61%)			.154
American
White 25 (49%)	21 (36%)			.063
Hispanic 1 (2%)	2 (3%)			.784
Admitting
complaint
Seizure 46 (90%)	55 (93%)			.447
Other 7 (14%)	5 (9%)			.268
More than two 5 (10%)	5 (9%)			.809
chief
complaints
Seizure type
Generalized 31 (61%)	36 (61%)			1.000
Petite mal 0 (0%)	3 (5%)			.143
Unknown 20 (39%)	20 (34%)			.463
Seizure etiology
Out of AEDs 24 (47%)	14 (23%)			.037
AED switch 1 (2%)	2 (4%)			.407
drug induced 2 (4%)	1 (2%)			.407
Unknown 24 (47%)	43 (73%)			.0002
Cardiovascular 23 (45%)	33 (56%)			.258
risk factors
IV PHT, indicates intravenous phenytoin; IM FOS, intramuscular fosphenytoin; AED, antiepileptic drug.

Discussion

There are no known studies that exclusively compare LOS in the ED between IV phenytoin and IM fosphenytoin.

Table 2 Primary outcome; time-to-discharge (hours) from the ED

Time to discharge (h)	IV PHT (n = 51)	IM FOS (n = 59)
Mean (SD)	3:27 +- 2:23	1:09 +- 1:28
95% CI	2:45-4:09	0:46-1:32
Median	2:36	0:43
95% CI	2:11-3:22	0:34-0:54
Range	0:26-11:48	0-9:36
Median difference		1:49
95% CI		1:24-2:24
		P b .001; d = 1.17
CI, indicates confidence interval; PHT, phenytoin; FOS, fosphenytoin; d, effect size.

Fig. 2 Percentage of patients in each group discharged from the ED at each time interval. FOS indicates fosphenytoin; PHT, phenytoin.

This study observed a significant decrease in time-to-safe discharge of approximately 2 hours (mean of 3:27 hours and 1:09 hours for IV phenytoin vs IM fosphenytoin, respectively) with a large effect size (d) = 1.17. More complicated patients with additional chief complaints were likely to influence the outlying times with a maximum time of 11:48 hours for IV phenytoin and 9:36 hours for IM fosphenytoin (Fig. 3).

Our results were comparable to a pharmacoeconomic study by Rudis et al [30] who found an observed time-to-safe discharge for IV phenytoin patients (n = 14) to be 1:42 hours. Rudis et al observed time-to-safe discharge from when the nurse started preparing IV phenytoin until the time therapeutic plasma concentrations of phenytoin was achieved in an absence of an adverse drug event. As with our study, the patient’s disposition was adequately pre- determined before the administration of the anticonvulsant. In addition, an open-label, randomized drug evaluation of 307 parenteral doses by Coplin et al found an observed median ED LOS to be 6:42 hours for IV phenytoin (n = 77) vs 5:42 hours (n = 202) for IV fosphenytoin (P = .6). Interestingly, Coplin et al reports a mean ED LOS for IM fosphenytoin (n = 28) at 7:42 hours [23]. This study initiated time studies from the point when the patient presented to the ED to actual discharge. Many confounders must be considered in this type of study; specifically, the

Fig. 3 Cumulative percentage of patients discharged per time. PHT indicates phenytoin; FOS, fosphenytoin.

responsibilities of the ED physician in evaluating and treating seizure patients, which may include diagnosis differential, seizure stabilization, determination of new- onset vs recurrent seizures, preventing seizure-related complications, identifying life-threatening process etiology of the seizure (laboratory studies), disposition determination, and minimizing future seizure incidences before the determination that a specific anticonvulsant is to be given. In contrast to our study, disposition was not predetermined. In the recent past, the use of fosphenytoin through any route in the ED was controversial because of its cost compared to generic parenteral phenytoin [30,31]. Fosphe- nytoin now available as a nonproprietary product and its cost has decreased significantly comparable to that of IV phenytoin [32]. Acquisition cost of a medication, however, is only one component to the Total cost of loading a seizure patient in the ED. Opportunity costs such as the availability of cardiac monitored beds, decompressing ED overcrowding and wait times, and decreased LOS in the ED are other considerations when choosing a medication for a seizure patient. Our retrospective observation of an earlier discharge may provide a resource for future cost-effectiveness studies

in the ED.

We found that obtaining phenytoin levels after loading and before discharge were not the common practice, which may be reflective of the position statement set forth by American College of Emergency Physician [26]. Although therapeutic Free phenytoin levels are between 1 to 2 mg/L, patients may remain seizure-free below the therapeutic window or may require doses greater than the therapeutic window. In addition, it is recommended that 4 hours elapse after IM fosphenytoin injection to allow for complete conversion to derived phenytoin distribution [18]. For these reasons, phenytoin levels are not routinely drawn before ED discharge [4].

The retrospective observational nature of this study is a significant limitation to this study. Blinded chart reviewers were not used, which may have enhanced the reproducibility of this study [33]. Determination of seizure etiology and precise time of drug preparation and discharge may be easier to collect in a prospective, randomized trial. The choice to use the time of removal of drug from the automated dispensing cabinet was based on the availability of these data and the imprecise documentation of the exact time the medication was administered to the patient.

Our study results showed that a significantly larger proportion of patients whose underlying seizure was felt to be due to noncompliance (out of antiepileptic medication) were administered IV phenytoin more often than IM fosphenytoin potentially driving the major effect (Table 1). Nevertheless, post hoc comparison between groups suggests this is not a statistically vital confounder influencing the primary outcome. Conversely, it is conceivable that patients who are noncompliant may require more time in the ED for reasons such as social evaluation and should be examined critically with this type of study.

Another limitation was related to the observation of adverse events with IV phenytoin and IM fosphenytoin administration. Blood pressure was not consistently moni- tored or recorded in this study; thus, it was difficult to determine if patients receiving IV phenytoin experienced any adverse effects. We did not observe documented adverse effects except for transient hypotension described in our results. serious adverse events may not have been detected because we excluded patients who were admitted to the hospital from the ED after anticonvulsant loading. As with this study, many studies that report adverse events lack standardized measurement and severity. Nonetheless, reported adverse events seem to be comparable between both IV phenytoin and IM fosphenytoin [22,23]. We found that 45% of our patients who received parenteral phenytoin had underlying cardiac compromise and may have benefited by receiving fosphenytoin.

Conclusion

This study found that patients whose disposition was predetermined and required non-emergent supplementation of phenytoin were discharged from the ED earlier with the loading of IM fosphenytoin compared to IV phenytoin. Resurrecting the question posed by Ramsay et al [34] given the risks associated with the use of IV phenytoin and the fact that phenytoin and fosphenytoin have comparable therapeutic profiles without the burden of increased cost, it would seem reasonable for an ED to replace parenteral phenytoin with fosphenytoin. Prospective studies are needed to determine if IM fosphenytoin is effectively comparable to IV phenytoin with respect to Seizure recurrence and in status epilepticus.

Acknowledgments

The authors are indebted to the pharmacy research committee of the Department of Pharmacy at University of Florida and Shands Jacksonville Medical Center; and to Thanh Hogan, PharmD, and June McAdams, PharmD, CCRC, who helped with the study design and methodology. They also acknowledge Joel Parnes, PharmD, MHA, who helped to identify patients for this study.

References

1. Huff JS, Morris DL, Kothari RU, et al. Emergency department management of patients with seizures: a multicenter study. Acad Emerg Med 2001;8:622-8.
2. Pashko S, McCord A, Sena MM. The cost of epilepsy and seizures in a cohort of Pennsylvania Medicaid patients. Med Interface 1993;1: 79-84.
3. Pallin JD, Goldstein JN, Moussally JS, et al. Seizure visits in US emergency departments: epidemiology and potential disparities in care. Int J Emerg Med 2008;1:97-105.
4. Meek PD, Davis SN, Collins MD, et al. Guidelines for nonemergency use of parenteral phenytoin products: proceedings of an expert panel consensus process. Arch Intern Med 1999;159(22):2639-44.
5. Bauman JL, Siepler JK, Fitzloff J. Phenytoin crystallization in intravenous fluids. Drug Intell Clin Pharm 1997:646-9.
6. Kilarski DJ, Buchanan C, Von Behren L. soft tissue damage associated with intravenous phenytoin. N Engl J Med 1984;311:1186-7.
7. Hagan HJ, Hastings H. Extravasation of phenytoin in the hand. J Hand

Surg 1988;13:942-73.

1. O’Brien TJ, Meara FM, Matthews H, et al. Prospective study of local cutaneous reactions in patients receiving IV phenytoin. Neurology 2001;57:1508-10.
2. O’Brien TJ, Cascino GD, So EL, et al. Incidence and clinical consequence of the purple glove syndrome in patients receiving intravenous phenytoin. Neurology 1998;51:1034-9.
3. Louis S, Kutt H, McDowell F. The cardiocirculatory changes caused by intravenous dilantin and its solvent. Am Heart J 1967;74:525-9.
4. Al-Khudhairi D, Whitwam JG. Autonomic reflexes and the cardio- vascular effects of propylene glycol. Br J Anesth 1986;58:897-902.
5. DeToledo JC, Lowe MR, Rabinstein A, et al. Cardiac arrest after fast intravenous infusion of phenytoin mistaken for fosphenytoin. Epilepsia 2001;42:288 [letter].
6. Fierro LS, Savulich DH, Benezra DA, et al. Safety of fosphenytoin sodium. Am J Health Syst Pharm 1996;53:2707-12.
7. Quon CY, Stamfli HF. In vitro hydrolysis of ACC-9653 (phosphate ester prodrug of phenytoin) by human, dog, rat blood and tissues. Pharm Res 1986(Suppl 5):134S [abstract].
8. Pryor FM, Gidal B, Ramsay RE, et al. Fosphenytoin: pharmacokinetics and tolerance of intramuscular loading doses. Epilepsia 2001;42(2): 245-50.
9. Leppik IE, Boucher BA, Wilder BJ, et al. Pharmacokinetics and safety of a phenytoin prodrug given IV or IM in patients. Neurology 1990;40: 456-60.
10. Boucher BA. Fosphenytoin: a novel phenytoin prodrug. Pharmaco- therapy 1996;16(5):771-91.
11. Product labeling. Cerebyx Parke-Davis a Division of Pfizer Inc., New York, NY. June 2002. Accessed 6 March 2010. Available from: URL: http://media.pfizer.com/files/products/uspi_cerebyx.pdf
12. U.S. Food and Drug Administration. Cerebyx (fosphenytoin sodium injection): important vial and carton label change. Accessed 23 October 2009. Available from: URL: www.fda.gov/medwatch/safety/ 1999/cereby.htm.
13. Fischer JH, Cwik MJ, Luer MS, et al. Stability of fosphenytoin sodium with intravenous solutions in glass bottles, polyvinyl chloride bags, and polypropylene syringes. Ann Pharmacother 1997;31:553-9.
14. Purin PS. The effect of diphenylhydantoin sodium (Dilantin) on myocardial contractility and hemodynamics. Am Heart J 1971;2:62-8.
15. Boucher BA, Feler CA, Dean JC, et al. The safety, tolerability, and pharmacokinetics of fosphenytoin after intramuscular and intravenous administration in neurosurgery patients. Pharmacotherapy 1996;16: 638-45.
16. Coplin WM, Rhoney DH, Rebuck JA, et al. Randomized evaluation of adverse events and length-of-stay with routine emergency department use of phenytoin or fosphenytoin. Neurol Res 2002;24:842-8.
17. Harrington L. Administer single-site 30-ml Intramuscular injection? Medsurg Nurs 2005;14(6):379-82.
18. Wilder BJ, Campbell K, Ramsay RE, Garnett WR, et al. Safety and tolerance of multiple doses of intramuscular fosphenytoin substituted for oral phenytoin in epilepsy or neurosurgery. Arch Neurol 1996;53: 764-8.
19. American College of Emergency Physicians. Clinical policy: critical issues in the evaluation and management of adult patients presenting to the emergency department with seizures. Ann Emerg Med 2004;43(5): 605-25.
20. Swandron SP, Rudis MI, Aziman K, et al. A comparison of phenytoin- loading techniques in the emergency department. Acad Emerg Med 2001;11:244-52.
21. Spengler RF, Arrowsmith JB, Kilarski DJ, et al. Severe soft tissue injury following intravenous infusion of phenytoin: patient and drug administration risk factors. Arch Intern Med 1988;148:1329-33.
22. Holliday SM, Benfield P, Plosker GL. Fosphenytion. Pharmacoeco- nomic implications of therapy. Pharmacoeconomics 1998;14:685-90.
23. Rudis MI, Touchette DR, Swandron SP, et al. Cost-effectiveness of oral phenytoin, intravenous phenytoin, and intravenous fosphenytoin in the emergency department. Ann Emerg Med 2004;43:386-98.
24. Marchetti A, Magar R, Fischer J, et al. Pharmacoeconomic evaluation of intravenous fosphenytoin (Cerebyx) vs. intravenous phenytoin (Dilantin) in hospital emergency departments. Clin Ther 1996;18(5): 953-66.
25. Thomson 2008 redbook pharmacy’s fundamental reference. Montvale (NJ): Thomson Healthcare; 2008. p. 454.
26. Gilbert EH, Lowenstein SR, Koziol-McLain J, et al. Chart reviews in emergency medicine research: where are the methods? Ann Emerg Med 1996;27:305-8.
27. Ramsay RE, DeToledo J. Intravenous administration of fosphenytoin: options for the management of seizures. Neurology 1996;46(suppl 1): S17-9.