Therapeutic Controversies

Phenytoin and Enteral Feedings:

Does Evidence Support an Interaction?

Sam CS Au Yeung and Mary HH Ensom

OBJECTIVE: To systematically review the reported interaction between oral dosage forms of phenytoin and enteral feeding
formulations with respect to the evidence supporting or refuting its existence, proposed mechanism(s) underlying the interaction,
and recommended interventions.
DATA SOURCES: We conducted a MEDLINE search (1966–April 2000) for English-language articles on phenytoin–enteral feeding
interactions; the search terms used were phenytoin, enteral feeding, and/or interaction, and/or in vitro. This search was supple-
mented by a bibliographic review of all relevant articles.
STUDY SELECTION: Prospective, randomized, controlled studies; prospective, nonrandomized, controlled studies; prospective,
nonrandomized, uncontrolled studies; retrospective studies; clinical experience reports; case reports; in vitro studies; and letters
were evaluated for relevant information.
DATA EXTRACTION: Data elements abstracted from these articles were study design, type (patients or healthy volunteers) and
number of subjects involved, method of administration of phenytoin and enteral feeding, formulation of phenytoin, type of feeding
(and whether it was continuous or interrupted), major findings, and proposed mechanisms of the interaction.
DATA SYNTHESIS: Although four prospective, randomized, controlled trials in healthy human volunteers refute the existence of the
interaction, there are numerous reports and studies showing dramatic decreases of serum phenytoin concentrations in patients
when it is coadministered with enteral feeding formulations. Therefore, evidence supports the existence of this interaction in patients
and in vitro studies, but not in healthy volunteers. Unfortunately, the exact mechanisms underlying this interaction remain unknown.
Many methods have been devised to prevent and treat the interaction once it has occurred; however, a single, generally accepted,
and practical intervention strategy is still lacking.
CONCLUSIONS: The exact role of enteral feeding in this interaction is unclear due to the lack of prospective, randomized, controlled
trials performed in patients. However, decreased serum phenytoin concentrations associated with enteral feeding may increase the
risk of seizures. Clinicians should be aware of this potential drug–nutrient interaction and design a patient-specific care plan that
includes consideration of the enteral feeding formulation and method of administration, as well as the phenytoin dosage form,
schedule of administration, and monitoring.
KEY WORDS: phenytoin, enteral feedings.
Ann Pharmacother 2000;34:896-905.

ACPE UNIVERSAL PROGRAM NUMBER: 407-000-00-014-H01

1
he first report of the potential interaction between
T phenytoin and enteral feeding products was published
in 1982. Later investigations of hospitalized patients have
reported decreased serum phenytoin concentrations associ-
ated with enteral tube feeding; minimal effect has been re-
ported in most studies involving healthy volunteers. The
interaction has been associated with a notable decrease in
Author information provided at the end of the text.
serum phenytoin concentrations, thereby jeopardizing sei-
zure control in patients receiving phenytoin.2-13 Thus, clini-
cians should recognize this potential drug–nutrient interac-
tion. However, conflicting reports create a dilemma for the
clinician developing a patient-specific care plan for pa-
tients receiving enteral feeding formulations and oral dosage
forms of phenytoin. A systematic review of this therapeu-
tic controversy was performed to describe existing evi-
dence supporting or refuting this interaction and character-
ize current recommendations.

896 ■ The Annals of Pharmacotherapy ■ 2000 July/August, Volume 34 www.theannals.com

Downloaded from aop.sagepub.com at NANYANG TECH UNIV LIBRARY on May 21, 2015
 We conducted a MEDLINE search from 1966 to April
2000 for English-language articles describing phenytoin–
enteral feeding interactions using the search terms pheny-
toin and enteral feeding, and/or interaction, and/or in vitro.
This search was supplemented by a bibliographic review
of all relevant articles. We categorized the quality of evi-
dence provided by relevant articles according to the US
Preventive Services Task Force rating scale. 14 Data ele-
ments abstracted from these articles were study design,
type (patients or healthy volunteers) and number of sub-
jects involved, method of administration of phenytoin and
enteral feeding, formulation of phenytoin, type of feeding
(and whether it was continuous or interrupted), major find-
ings, and proposed mechanisms of the interaction (Tables
11-13,15-30 and 21,4,12,18,19,21,24-26,28).
Study Designs
We identified 29 pertinent citations in our literature search;
abstracts were not included. There were four prospective,
randomized, controlled studies15-18; one prospective, non-
randomized, controlled study19; four prospective, nonran-
domized, uncontrolled studies1,12,20,21; one retrospective
study2; two clinical experience reports22,23; 10 case reports3-11,13
(4 of which were letters)8-11; and seven in vitro studies.24-30
According to the US Preventive Services Task Force rating
scale, only four studies qualified as level I,15-18 one was lev-
el II-1,19 three were II-3, 1,20,21 10 were level III,2-7,12,13,22,23
and the remaining were classified as other (i.e., below level
III).8-11 The in vitro studies24-30 were classified separately.
All serum phenytoin concentrations were reported as total
concentrations.
Studies Refuting the Interaction
EVIDENCE LEVEL I
Evidence Obtained from at Least
One Properly Randomized, Controlled Trial
Only four studies15-18 showed no interaction between
phenytoin and enteral feedings. All were single-dose stud-
ies conducted in healthy volunteers.
Doak et al.17 studied 10 healthy volunteers, aged 23– 43
years. The three phases of the study were separated by at
least seven days. During phase A, subjects received pheny-
toin sodium 435 mg intravenously over 30 minutes. Dur-
ing phases B and C, subjects had a nasogastric (NG) tube
placed into the distal stomach through which phenytoin
acid suspension 400 mg and phenytoin sodium solution
435 mg, respectively, were administered. Continuous en-
teral feedings (Isocal) were given by NG tube for 14 hours
before and after the phenytoin dose. Blood samples were
collected over 72 hours after each phenytoin dose. Bio-
availability; rate of absorption; maximum concentration;
and time to maximum concentration, as determined via an
empirical quadratic function of time method, were com-
pared for two enteral doses (phenytoin acid suspension and
phenytoin sodium solution administered concomitantly
www.theannals.com The Annals of Pharmacotherapy
Downloaded from aop.sagepub.com at NANYANG TECH UNIV LIBRARY on May 21, 2015
with enteral feedings). There were no significant differ-
ences in bioavailability for phenytoin acid suspension and
phenytoin sodium solution (0.88 ± 0.15 vs. 0.91 ± 0.7; p =
0.57, 90% CI 0.14 to 0.071). However, the absorption pat-
terns were significantly different, with the sodium solution
being more rapidly absorbed (2.5 ± 3.8 vs. 14.8 ± 11.2 h; p
= 0.004).
Ten healthy volunteers (aged 35.2 ± 8.0 y) were evaluat-
ed by Marvel and Bertino.15 Phenytoin suspension was
used, and the subjects were randomly divided into three
groups: the first group took phenytoin alone, the second
group took phenytoin and Vivonex TEN, and the third
group took phenytoin and Ensure. No statistical difference
in AUC (273.2 ± 162.5 vs. 232.9 ± 76.2 vs. 256.3 ± 153.7
µg•h/mL), time to peak phenytoin concentration (tmax; 8.9
± 5.9 vs. 6.3 ± 3.3 vs. 5.9 ± 3.3 h), or peak phenytoin con-
centration (4.4 ± 1.2 vs. 5.3 ± 1.2 vs. 4.7 ± 1.1 µg/mL) was
observed during the three treatment phases. The authors
concluded that the two enteral feeding formulations did
not interfere with or enhance phenytoin absorption. Their
power analysis indicated that the study had an 80% chance
of detecting a 10% change in AUC.
In a randomized, crossover study,16 six healthy volun-
teers (aged 22–29 y) completed the following two phases:
phenytoin without enteral feedings; and concomitant enter-
al feedings before phenytoin, which were continued at 100
mL/h for 10 hours. The AUC0-48h was not significantly dif-
ferent between the two phases (198.7 ± 43.65 vs. 203.8 ±
55.66 µg•h/mL; p = 0.572). The authors concluded that en-
teral feedings do not affect the serum concentrations of
phenytoin after a single dose given in capsule form. Fol-
lowing a power analysis, the authors determined that there
was a 96% chance of recognizing a 20% difference in AUC
if one existed, whereas their study found only a 2% differ-
ence in AUC between treatments. Despite this, the authors
suggested that the difference between the AUC0-48h of the
two phases would probably not be clinically important,
even if the sample size was large enough to detect a signif-
icant difference.16
Krueger et al.18 conducted a randomized, crossover study
involving eight healthy volunteers (aged 22–30 y). Sub-
jects received phenytoin suspension 400 mg after fasting
(group A), with hourly Ensure (group B), and with Ensure
given every four hours (group C). Their results showed
that AUC values were not different among groups A, B,
and C (222.1 ± 86.9 vs. 233.9 ± 92.9 vs. 226.0 ± 95.7
µg• h/mL; p > 0.05), respectively. Interestingly, tmax was
significantly shorter (p < 0.05) when phenytoin was ad-
ministered with Ensure (group B, 8.5 ± 3.0 h; group C, 5.3
± 2.0 h) than without Ensure (group A, 18.5 ± 10.5 h). The
authors concluded that the bioavailability of phenytoin was
not decreased by the enteral feeding formula administra-
tion schedule, but that the rate of absorption of phenytoin
was enhanced by the formula. Although this study failed to
detect a difference in phenytoin bioavailability from con-
comitant ingestion of an enteral feeding formula, the au-
thors still believed that such an interaction exists. They
postulated that factors other than direct contact may be re-
■ 2000 July/August, Volume 34 ■ 897
 Table 1. Clinical Investigation of the Phenytoin–Enteral Feeding Interaction
Phenytoin
Study Route of Feeding Continuous/ Support/Refute
Reference Design Subjects Administration Formulation Formulation Interrupted Feeding the Interaction

Evidence level Ia
Doak et al. prospective 10 healthy iv, NG tube phenytoin sodium Isocal continuous refute
(1998)17 volunteers feeding injection,
Dilantin
suspension
Krueger et al. prospective 8 healthy oral ingestion Dilantin Ensure interrupted refute
(1987)18 volunteers suspension
Marvel and prospective 10 healthy oral ingestion Dilantin Vivonex TEN, interrupted refute
Bertino volunteers suspension Ensure
(1991)15
Nishimura prospective 6 healthy oral ingestion Dilantin, Kapseal Ensure interrupted refute
et al. (1988)16 volunteers
Evidence level II-1b
Guidry et al. prospective 5 healthy oral ingestion Dilantin Osmolite, Com- interrupted support
(1989)19 volunteers suspension pleat Modified
Evidence level II-3c
Bauer (1982)1 prospective 20 neurosur- pts.: NG tube, Dilantin Isocal pts.: continuous, support
gery pts., volunteers: suspension volunteers:
5 healthy po interrupted
volunteers
Ozuna and prospective 7 pts., NG tube Dilantin Ensure, Osmolite continuous (days 1– support
Friel (1984)20 review of suspension 6), interrupted
44 head- (days 7–10)
injured pts.
Randall and prospective 11 head- iv Dilantin Isocal, Ensure, continuous support
Tett (1994)21 injured pts. injection Jevity, Vital/Jevity,
Isocal/Ensure
Evidence level IIId
Bader (1993)12 prospective 2 head- iv initially, suspension, BNS Jevity, Isocal HN continuous support
case study injured pts. then J-tube
Faraji and Yu retrospective 22 brain- G-tube NA NA unclamped group: support
(1998)2 injured pts. continuous,
clamped group:
interrupted
Gilbert et al. experience NA NA NA NA NA support
(1996)22 report
Hatton and experience NA NA NA NA NA support
Magnuson report
(1996)23
Maynard et al. case report 1 pt. with feeding tube suspension, BNS Osmolite, Isocal continuous initially, support
(1987)3 cerebral into the HN, Compleat interrupted later
hemorrhage duodenum Modified
O’Hagan and case report 2 pediatric NG tube suspension, BNS Fortison NA support
Wallace pts. with
(1994) 4 seizures
Olsen et al. case report 1 pt. with NG tube suspension, BNS Isotein Hn continuous support
(1989) 13 seizure
history
Rodman et al. case report 1 pt. with J-tube injection, then Jevity interrupted support
(1995)5 seizure suspension,
history BNS
Saklad et al. case report 1 pt. with NG tube injection, then Osmolite continuous support
(1986)6 seizures suspension,
secondary BNS
to brain
metastases
Sneed and case report 1 brain- NG tube injection, then Osmolite continuous initially, support
Morgan (1988)7 injured pt. suspension, interrupted later
BNS

BNS = brand not specified; G-tube = gastrostomy tube; J-tube = jejunostomy tube; NA = not available; NG = nasogastric.
a
Evidence obtained from at least one properly randomized, controlled trial.
b
Evidence obtained from well-designed, controlled trials without randomization.
c
Evidence obtained from multiple time series with or without the intervention. Dramatic results in uncontrolled experiments could also be regarded as this
type of evidence.
d
Opinions of respected authorities, based on clinical experience; descriptive studies and case reports, or reports of expert committees.
(continued on page 899)

898 ■ The Annals of Pharmacotherapy ■ 2000 July/August, Volume 34 www.theannals.com

Downloaded from aop.sagepub.com at NANYANG TECH UNIV LIBRARY on May 21, 2015
 Phenytoin–Enteral Feeding Interaction

sponsible for the decreases in phenytoin concentrations by EVIDENCE LEVEL II-3

coadministered enteral feeding formulas that were ob-
served in previous reports.1,8,11
Studies Supporting the Interaction
EVIDENCE LEVEL II-1
Evidence Obtained from Well-Designed
Controlled Trials without Randomization
In a study conducted by Guidry et al.,19 healthy volun-
teers first received phenytoin suspension 500 mg without
concurrent enteral feeding; this dose was repeated while
they ingested protein hydrolysate enteral feedings hourly,
and again during hourly ingestions of meat-base enteral
feedings. Serum phenytoin concentrations were lowest
(approximately 2.5–3.5 µg/mL) with protein hydrolysate
feedings. Mean serum phenytoin concentrations were con-
sistently higher (approximately 4.0 –5.0 µg/mL) with the
meat-base feeding than with the protein hydrolysate for-
mula, although concentrations did not reach those obtained
during the control period (approximately 5.5–6.5 µg/mL).
The authors found that it is easier to attain therapeutic
serum phenytoin concentrations in patients receiving a
meat-base enteral feeding than in those receiving a protein
hydrolysate formula (p < 0.01).
Evidence Obtained from Multiple Time Series
with or without the Intervention. Dramatic Results in
Uncontrolled Experiments Could also be
Regarded as This Type of Evidence
Bauer1 first reported decreased serum phenytoin con-
centrations associated with enteral tube feedings. While a
formal pharmacokinetic analysis was not performed, low
serum phenytoin concentrations (2.59 ± 0.96 µg/mL) were
obtained in 10 neurosurgery patients receiving phenytoin
suspension and continuous NG feedings. When the feed-
ings were discontinued, the average phenytoin concentra-
tion rose to 10.22 ± 2.90 µg/mL in seven days. The aver-
age serum concentration decreased from 9.80 to 2.72 µg/mL
in 10 other patients receiving phenytoin suspension after
seven days of continuous enteral tube feedings. In an at-
tempt to further characterize the interaction, five healthy
volunteers took phenytoin suspension 400 mg orally and
had blood specimens collected for phenytoin serum con-
centration analysis. Three weeks later, each subject drank
the enteral formulation (Isocal) at a rate of 100 mL/h.
Fourteen hours later, while still ingesting the enteral for-
mulation hourly, they drank the phenytoin suspension. The
phenytoin concentrations were significantly lower during
the continuous feeding (p < 0.002) compared with pheny-

Table 1. Clinical Investigation of the Phenytoin–Enteral Feeding Interaction (continued)

Phenytoin
Study Route of Feeding Continuous/ Support/Refute
Reference Design Subjects Administration Formulation Formulation Interrupted Feeding the Interaction

Other (below Evidence level III)

Hatton (1984)8 case report 1 postpartum NG tube Dilantin, Kapseal Osmolite continuous support
(letter) pt. with seizures
Taylor et al. case report 1 pt. with NG tube, oral suspension, BNS Fresubin interrupted support
(1993)9 (letter) seizure disorder ingestion
Weinryb and case report 1 pt. with cere- NG tube NA Osmolite continuous support
Cogen (1989)10 (letter) bral infarct
and seizures
Worden et al. case report 1 pt. with NA capsule, BNS Isocal continuous support
(1984)11 (letter) seizure disorder
In vitro studies
Cacek et al. support
(1986)24
Cutie et al. support
(1983) 25
Fleisher et al. support
(1990) 26
Holtz et al. support
(1987) 27
Hooks et al. support
(1986)28
Smith et al. support
(1988)29
Splinter et al. support
(1990)30

BNS = brand not specified; NA = not available; NG = nasogastric.

www.theannals.com The Annals of Pharmacotherapy ■ 2000 July/August, Volume 34 ■ 899

Downloaded from aop.sagepub.com at NANYANG TECH UNIV LIBRARY on May 21, 2015
toin administration alone, respectively (0.76 ± 0.21 vs.
2.90 ± 0.53 µg/mL at 3 h). The author concluded that there
is an interaction between enteral feedings and phenytoin,
which required the administration of large phenytoin dos-
es, ranging from 800 to 1600 mg/d to achieve optimal
serum phenytoin concentrations. He suggested that stop-
ping the enteral feeding from two hours before to two
hours after phenytoin administration would help to lessen
the problem.
Serum phenytoin concentrations were compared in pa-
tients who received phenytoin suspension and enteral tube
feedings administered either by continuous infusion or in-
terrupted from two hours before to two hours after pheny-
toin suspension was administered through the tube.20 This
intervention resulted in a decrease of 0.7 µg/mL in the
mean phenytoin concentration. The authors concluded that
this procedure does not maintain therapeutic phenytoin
concentrations and is time-consuming for the nurse. They
suggested that increasing the phenytoin dosage would be
more effective.
Randall and Tett21 suggested that the clinically observed
decrease in phenytoin concentrations with enteral feedings
is the result of increased drug clearance rather than malab-
sorption of phenytoin. They found that patients who had
been receiving enteral feedings for more than five days had
higher values of maximal velocity of the reaction (Vmax;
709 vs. 394 mg/d; p = 0.008) and clearance (4.3 vs. 1.7
L/h; p = 0.003) compared with those who received enteral
feedings for less than five days. The authors suggested that
since the cause of the interaction is not due to physical loss
of oral phenytoin dosage forms as suggested by other in-
vestigators,1,3,6,8,10,11,19,20 the best intervention is to decrease
the dosing interval as well as increase the phenytoin dos-
age when the drug is given orally or intravenously.
EVIDENCE LEVEL III
Opinions of Respected Authorities, Based on
Clinical Experience; Descriptive Studies and Case
Reports; or Reports of Expert Committees
Numerous reports of clinical experience and cases relat-
ed to the phenytoin–enteral feeding interaction have been
published; all of these have suggested the existence of the
interaction. We provide a brief summary of each report
that met criteria for evidence level III.
Bader12 conducted a prospective case study (n = 2) to
examine the effects of diluting the phenytoin suspension
and irrigating the tube. Treatment I involved administering

Table 2. Proposed Mechanisms of the Phenytoin–Enteral Feeding Interaction

Reference Proposed Mechanisms

Bader et al. decrease in phenytoin concentration is due to phenytoin binding to the plastic tubing and/or an interaction between
(1993) 12 phenytoin and the dietary supplement
Bauer (1982)1 continuous NG tube feeding formulas interfere with phenytoin suspension absorption by decreasing the drug’s bioavailability
Cacek et al. binding of the phenytoin suspension to the plastic NG tube lumen; complex formation between calcium and phenytoin;
(1986)24 constant NG suctioning, concurrent administration with enteral tube feedings or antacids, alter the physiologic state of
the pt.
Cutie et al. (1983)25 physical incompatibility between phenytoin and enteral feedings
Fleisher et al phenytoin, as a weak acid (pKa 8.1), will become nonionized at low pH values; therefore, unlike the phenytoin anion that
(1990)26 binds reversibly to NG tubing, substantial amounts of nonionized drug will be irreversibly lost from solution to NG tubing if
the pH is low enough; also, less phenytoin will be available if solid phenytoin is not provided with adequate GI residence
time for complete dissolution
Guidry et al. dietary fiber in meat-base formula may aid absorption of phenytoin by complexing with and inactivating a substance in
(1989) 19 feeding that binds phenytoin, or it may alter GI transit time; protein hydrolysate formulas do not have dietary fiber content
Hooks et al. because of the high pKa of phenytoin, the acidic nature of the enteral formula may alter the solubility of phenytoin and lead
(1996)28 to reduction in phenytoin recovery
Krueger et al. longer periods of enteral nutrient formula consumption may decrease GI transit time or alter GI pH, limiting phenytoin
(1987)18 absorption
O’Hagan and phenytoin suspension is a weak acid, dissolving best in alkaline conditions as exist in the small intestine; enteral feedings
Wallace (1994)4 (pH ~6.6) may reduce absorption
Randall low concentrations of phenytoin observed in patients receiving long-term enteral feedings appear to be due to increased Cl
and Tett (1994)21 of the drug, not decreased bioavailability
Rodman et al. interaction with concurrent medications; binding of phenytoin to the lumen of the plastic tube; binding of phenytoin to
(1995)5 certain components of enteral formulas (e.g., proteins, calcium salts)
Smith et al. phenytoin binds to casein and calcium in the enteral products, thereby decreasing absorption
(1988)29
Splinter et al. phenytoin suspension/enteral formula interaction is pH-dependent; there appears to be physical entrapment of the large-
(1990)30 suspension particles when the protein in the enteral solution denatures at lower pH values, resulting in decreased
phenytoin absorption
Weinryb and binding of phenytoin to calcium caseinate or magnesium present in enteral feeding products may be responsible for the
Cogen (1989)10 lowered phenytoin concentration

GI = gastrointestinal; NG = nasogastric.

900 ■ The Annals of Pharmacotherapy ■ 2000 July/August, Volume 34 www.theannals.com

Downloaded from aop.sagepub.com at NANYANG TECH UNIV LIBRARY on May 21, 2015

phenytoin suspension in a diluted form for five days and
irrigating the feeding tube before and after medication ad-
ministration. Treatment II involved administering pheny-
toin suspension in an undiluted form without tube irriga-
tion for five days. The phenytoin suspension was adminis-
tered via a jejunostomy tube (J-tube) in the first patient.
Her serum phenytoin concentration was 9.85 µg/mL at the
end of treatment I compared with a concentration of 6.49
µg/mL after treatment II. The serum phenytoin concentra-
tion of the second patient was 1.56 µg/mL after treatment I
compared with a concentration of 2.24 µg/mL after treat-
ment II. One explanation for these differing results is that,
in addition to phenytoin and Isocal HN, the second patient
was receiving intravenous medications: Fibrad (a supple-
ment composed of dietary fibers) and lactobacillus acido-
philus. The author suggested that the low phenytoin con-
centration observed in the second patient after treatment I
was probably due to phenytoin binding to the tubing and/or
an interaction between phenytoin and Fibrad. Therefore,
based on the first patient’s results, the author concluded
that dilution of the phenytoin suspension and irrigation of
the tube produced higher phenytoin concentrations than
those obtained when the suspension was administered
undiluted.
A retrospective review2 was undertaken to examine the
effect of clamping the gastrostomy tube (G-tube) after a
phenytoin dose. Clamping the G-tube for only one hour af-
ter each phenytoin dose was associated with significantly
higher serum phenytoin concentrations compared with those
obtained in patients whose G-tubes were not clamped
(14.4 ± 4.7 vs. 9.2 ± 6.8 µg/mL; p < 0.05). The authors
concluded that the beneficial effects of tube-clamping may
occur after only one hour of interruption instead of four
hours per dose as suggested by others.1,6
Maynard et al. 3 reported a patient who was receiving
continuous Osmolite and phenytoin suspension through a
soft feeding tube placed into the duodenum. The serum
phenytoin concentration was <2.5 µg/mL, despite dosages
up to 600 mg/d. Even with higher dosages (1.0–1.2 g/d) of
phenytoin and interruption of enteral feeding, phenytoin
concentrations remained low (maximum concentration 9
µg/mL) with Osmolite. Therapeutic phenytoin concentra-
tions (maximum concentration 14 µg/mL) were obtained
by using a high dosage of phenytoin (1.6–1.8 g/d) with
Isocal HN or Osmolite. However, therapeutic concentra-
tions (maximum concentration 19 µg/mL) were maintained
with much lower dosages (600 mg/d) when phenytoin was
given with meat-base (Compleat Modified) tube-feeding
formula.
An algorithm was described by Gilbert et al.22 to deal
with subtherapeutic phenytoin concentrations when ad-
ministered by different methods (e.g., NG tube, percuta-
neous endoscopic G-tube). For example, according to the
algorithm, if the enteral feeding is given continuously via
needle catheter jejunostomy, then phenytoin should not be
given concurrently using that method. In this case, pheny-
toin should be given intravenously and serum concentra-
tions measured 72 hours later. The authors reported that
www.theannals.com The Annals of Pharmacotherapy
Downloaded from aop.sagepub.com at NANYANG TECH UNIV LIBRARY on May 21, 2015
Phenytoin–Enteral Feeding Interaction
the algorithm was very effective, reflected by fewer hospi-
tal admissions related to subtherapeutic phenytoin concen-
trations.
Hatton and Magnuson23 described a protocol used at
their institution to minimize this drug–nutrient interaction.
Continuous tube feedings are stopped from one hour be-
fore to one hour after each phenytoin dose administered
via feeding tube. Also, the feeding tubes are flushed with
≥30 mL of water before and after each dose is adminis-
tered to ensure tube patency.
Two children developed subtherapeutic serum pheny-
toin concentrations associated with enteral feeding.4 In
both cases, phenytoin concentrations fell dramatically (by
3.3 and 5.6 µg/mL) after initiation of enteral feedings.
Therapeutic phenytoin concentrations were maintained af-
ter dosage increases and the addition of other anticonvul-
sants. Alternatives that the authors suggested included us-
ing intravenous phenytoin preparations or other oral anti-
convulsants until normal feeding can be resumed.
Olsen et al.13 reported a lowered phenytoin concentra-
tion in a patient receiving concurrent phenytoin suspension
and Isotein Hn via an NG tube. Despite gradual dosage in-
creases of phenytoin over two days (from 200 to 400 mg/d),
the patient’s serum phenytoin concentrations remained
subtherapeutic (<2.5 µg/mL). As a result, a decision was
made to switch the phenytoin therapy to 100 mg intra-
venously three times daily. The phenytoin concentration
remained within the therapeutic range (11.8 µg/mL) while
he was receiving intravenous phenytoin. On day 11, ad-
ministration of the oral suspension was reinstituted, which
resulted in a rapid decrease in the serum concentration (6.8
µg/mL). Intravenous therapy was reinstituted, with a return
to therapeutic serum concentrations (10.4 µg/mL). The au-
thors suggested that if the patient is receiving enteral tube
feedings, phenytoin should be administered in capsular or
intravenous form.
Rodman et al.5 reported a case of phenytoin malabsorp-
tion via a J-tube placed distal to the ligament of Treitz. Af-
ter discontinuation of valproic acid, the patient began re-
ceiving phenytoin; a serum concentration of 19.1 µg/mL
was maintained. The enteral formula Jevity was adminis-
tered through the J-tube on day 19, with an interruption for
approximately five to 10 minutes to administer drugs, and
the tube was flushed with 30 mL of NaCl 0.9% before and
after drug delivery. On day 33, the phenytoin serum con-
centration decreased to <2.5 µg/mL. The authors suggested
that an interaction between phenytoin and valproic acid or
enteral feeding might be the cause of the decreased con-
centration; adsorption of phenytoin to the J-tube may also
play a role.
In a report by Saklad et al.,6 a patient who had brain
metastases with seizures was receiving phenytoin suspen-
sion. After Osmolite continuous feedings were initiated,
his serum phenytoin concentrations decreased substantial-
ly, so the phenytoin dosage was increased. However, after
10 days, the patient pulled out his feeding tube and the
phenytoin concentration rose dramatically (to 53 µg/mL).
The authors concluded that enteral tube feedings must not
■ 2000 July/August, Volume 34 ■ 901
be discontinued abruptly if patients are maintained on a
high dosage of phenytoin.
A retrospective chart review by Sneed and Morgan7
identified seven of 11 cases that reflected an alteration of
phenytoin concentrations by enteral formula. In the repre-
sentative case, a five-year-old girl had multiple occasions
of subtherapeutic phenytoin concentrations (lowest con-
centration <2.5 µg/mL) while she was receiving Osmolite.
When Osmolite was discontinued, her serum phenytoin
concentration increased to 31.2 µg/mL. The authors’ sug-
gestions to overcome the problem included increasing the
phenytoin dosage, administering oral phenytoin between
intermittent enteral feedings, and replacing phenytoin with
another anticonvulsant.
CASE REPORTS BELOW EVIDENCE LEVEL III
Several case reports (letters),8-11 graded below evidence
level III, support the existence of the phenytoin–enteral
feeding interaction. All reported decreases in phenytoin
concentrations after enteral feeding formula was adminis-
tered. In some cases,9-11 the lowered phenytoin concentra-
tions returned to therapeutic levels once the enteral feeding
was discontinued.
IN VITRO STUDIES
Clinical investigation of the effect of enteral tube feed-
ing on serum phenytoin concentrations is difficult because
of ethical concerns related to designing a study that would
include withholding enteral feedings in some patients. There-
fore, many investigators have tried to examine this interac-
tion using an in vitro model to predict potential mecha-
nisms and then extrapolate their results to in vivo situa-
tions. These in vitro studies have tested the hypothesis that
the interaction between phenytoin and enteral feedings is
caused by physical incompatibility. All in vitro studies we
identified support the existence of an interaction.
Cacek et al.24 used 10 different methods to deliver phen-
ytoin acid suspension through an NG tube. In the method
that used no diluent or irrigation, significantly more drug
(55.7% ± 15.3% change in initial dose concentration) was
lost (p ≤ 0.005) compared with methods that included dilu-
ents and/or irrigation. The authors concluded that phenytoin
suspension should be diluted with at least 20 mL of fluid
(e.g., NaCl 0.9%) and irrigated when administered.
Phenytoin was mixed with Ensure Plus, Ensure, and Os-
molite individually to test for compatibility in a study by
Cutie et al.25 On visual inspection, the authors noticed a
slight (extent not specified) increase in viscosity in all cas-
es when phenytoin suspension was mixed with small vol-
umes of enteral products. The authors recommended that
clinicians be aware of potential phenytoin–enteral feeding
interactions and avoid combining enteral products with
pharmaceutical formulations.
The study by Fleisher et al.26 suggested that adsorption
of phenytoin to the tubing is a function of pH. For exam-
ple, the total recovered mass over 20 minutes of phenytoin
902 ■ The Annals of Pharmacotherapy ■ 2000 July/August, Volume 34
Downloaded from aop.sagepub.com at NANYANG TECH UNIV LIBRARY on May 21, 2015
in phosphate buffer at pH 3 for glass tubing was 99.8%,
whereas with NG tubing, the recovered mass was only
38.7%. The results from this study demonstrated that drug
irreversibly lost from solution to NG tubing can be sub-
stantial if the pH is low enough for the nonionized form to
dominate. Furthermore, if the stomach empties immediate-
ly or tubing is positioned to empty into the duodenum, sol-
id phenytoin will not be provided with adequate gastroin-
testinal residence time for complete dissolution.
Unexpected decreases in phenytoin concentrations oc-
curred within the enteral product/medication mixtures fol-
lowing addition of phenytoin suspension to Ensure, Ensure
Plus, and Osmolite in an in vitro experiment.27 However,
the authors also found that phenytoin concentrations de-
creased to a lesser extent when phenytoin injection was
added to all three products. For instance, with Ensure Plus,
the phenytoin concentration change over 12 hours for the
suspension was 106%; the concentration change with the
injection was only 8%. The authors concluded that intra-
venous phenytoin might be an alternative to enteral admin-
istration.
In another in vitro study,28 phenytoin recovery was re-
duced after it was mixed with Osmolite. The mean pheny-
toin concentration in 10 Osmolite samples was 3.70 ± 0.28
µg/mL compared with 9.87 ± 0.27 µg/mL in 10 control so-
lution samples. However, the authors noted that further re-
search in humans is necessary to determine the clinical sig-
nificance of their results.
Smith et al.29 found that the in vitro recovery of pheny-
toin from solutions containing protein caseinates and calci-
um, which are both present in Osmolite, was significantly
lower than that from an aqueous control solution (8.2 ± 0.7
µg/mL vs. 10.1 ± 0.3 µg/mL, respectively) at a theoretical
phenytoin concentration of 10.0 µg/mL. Recovery of pheny-
toin was lowest from solutions containing sodium casein-
ate (5.8 ± 0.3 µg/mL).
Splinter et al.30 suggested that the phenytoin suspension–
enteral formula interaction is pH-dependent. The interac-
tion is less significant (p < 0.05) at high pH values (i.e.,
6–8). Their results agree with those obtained by Fleisher et
al.,26 which suggested that lower pH values would result in
more irreversible binding of phenytoin to the NG tube.
Mechanism of the Interaction
The exact mechanism underlying the phenytoin–enteral
feeding interaction is unknown; several have been specu-
lated (Table 2). The majority of studies1,5,10,12,19,24,25,29 sug-
gested that there is physical incompatibility between phen-
ytoin and certain components in enteral feeding formulas,
resulting in complexation of phenytoin particles and, there-
fore, decreased bioavailability. Some5,12,24,26 suggested bind-
ing of phenytoin to the tube lumen as the mechanism; oth-
ers4,18,26,28,30 proposed that the interaction is pH-related. In-
terestingly, some investigators21 commented that the lower
phenytoin concentrations observed clinically are actually
due to increased metabolism or clearance of the drug after
prolonged use of enteral feedings. One study5 suggested
www.theannals.com

that an interaction between phenytoin and concurrent med-
ications might be the cause of the decreased concentration.
Discussion
As shown in Table 1, the great majority of articles in the
literature support an interaction between phenytoin and en-
teral feeding formulas. None of those studies were prospec-
tive, controlled, randomized trials with large sample sizes.
However, the fact that almost all of the existing literature
describes decreased serum phenytoin concentrations when
it is given with enteral feeding suggests that it is highly un-
likely for such a phenomenon to occur by chance.
Notably, the four studies15-18 that refuted the existence of
the interaction were prospective, controlled, randomized
trials involving single doses of phenytoin administered to
healthy volunteers, not patients. While it is generally not
feasible to conduct such controlled trials in patients, results
obtained from trials with healthy subjects may not be ap-
plicable to clinical situations, due to numerous confound-
ing factors (e.g., presence of gastrointestinal conditions
that require tube feeding, concomitant disease states, other
drugs, different types and locations of enteral feeding tube)
and inability to isolate individual variables in patients. As
such, healthy volunteers do not simulate patients who re-
quire phenytoin and enteral feeding. Also, phenytoin was
administered orally in three of the level I studies15,16,18; this
method usually does not reflect clinical practice in acute-
care patients who often receive enteral formulations through
feeding tubes. Furthermore, due to small sample sizes and
variability of the data, some of these level I studies16-18 may
lack the power to detect a significant difference in the
bioavailability of phenytoin. These limitations make ex-
trapolation of data to clinical practice difficult. Thus, while
studies in healthy volunteers may help delineate mecha-
nisms of an interaction, data from noncontrolled studies in
patients are often more useful in the management of this
interaction.
Of the studies that supported the interaction, none was
graded as a level I trial according to the US Preventive
Services Task Force rating scale.14 Several of the stud-
ies1,2,12,20,21 supporting the interaction were performed in pa-
tients. However, ethical and logistic considerations limit
the ability to conduct a prospective, controlled, random-
ized trial in patients. Discontinuing enteral feedings to
compare the serum phenytoin concentrations with and
without feedings may be considered unethical and imprac-
tical. The remaining articles were in the form of clinical
experience reports,22,23 case reports,3-11 and in vitro stud-
ies,24-30 with limitations of lack of controls, publication
bias, and inability to directly extrapolate to in vivo condi-
tions, respectively.
The exact mechanisms responsible for the interaction
are poorly understood. Most studies suggested that the in-
teraction is likely due to physical incompatibility between
phenytoin and certain components in the enteral feeding
formula; however, none was able to identify the causative
components. Some authors5,12,24 proposed that irreversible
www.theannals.com The Annals of Pharmacotherapy
Downloaded from aop.sagepub.com at NANYANG TECH UNIV LIBRARY on May 21, 2015
Phenytoin–Enteral Feeding Interaction
binding of phenytoin to the wall of the feeding tube was
the cause; in another study,21 decreased serum phenytoin
concentrations were observed even when the drug was ad-
ministered intravenously. Other investigators21,31 suggested
that the lowered phenytoin concentrations detected were
due to increased clearance of phenytoin from enzyme in-
duction; however, the substance that is purported to cause
the enzyme induction, as well as which enzymes are in-
duced, are unknown. One study18 proposed that longer pe-
riods of enteral nutrient formula administration may de-
crease gastrointestinal transit time or alter gastrointestinal
pH, thereby limiting phenytoin absorption. Results from in
vitro studies 26,30 suggest that pH may play an important
role in this interaction. Since phenytoin is a weak acid, the
nonionized form, which is believed to have a high tenden-
cy to bind to the tubing, will predominate at low pH val-
ues. Further studies are required to determine the validity
of these results in clinical situations. In many cases, the ob-
served decreases in phenytoin concentrations are likely
due to more than a single factor, such as enteral feeding ad-
ministration. There are also other confounding factors,
such as concurrent medications and illnesses, whose con-
tributions to pH changes or other aspects of the interaction,
albeit still poorly understood, cannot be ignored.
Various methods have been developed to prevent the in-
teraction from occurring and to reestablish therapeutic
serum phenytoin concentrations once the interaction oc-
curs. For instance, the algorithm described by Gilbert et al.22
and the protocol described by Hatton and Magnuson23 are
reported to be effective means to deal with this problem.
Due to study limitations and lack of prospective, ran-
domized, controlled trials in patients, comparing serum
phenytoin concentrations with and without enteral feed-
ings, the extent and true nature of the interaction remain in-
conclusive. Unfortunately, we still do not know the an-
swers to the following questions:
1. What is the exact mechanism of the phenytoin–enter-
al feeding interaction?
2. Does this interaction occur with some enteral feeding
formulas but not others?
3. What is the best procedure to prevent this interac-
tion?
4. Once the interaction has occurred, what is the best
intervention?
5. What factor(s) other than enteral feedings contrib-
ute(s) to the interaction?
Such well-designed trials probably will never be per-
formed in patients; therefore, clinicians must be aware of
the possibility of this drug–nutrient interaction and monitor
serum phenytoin concentrations to guide therapy. To mini-
mize the occurrence of this interaction, we recommend
that phenytoin and feeding administration times be stag-
gered to avoid concurrent administration. When continu-
ous feedings are used, we recommend stopping the feed-
ings one to two hours before each phenytoin dose, flushing
the tubing with at least 20 mL of water or NaCl 0.9% be-
fore and after the dose, and withholding the feedings for
one to two hours after phenytoin administration. Other
■ 2000 July/August, Volume 34 ■ 903
measures to ensure that the patient’s seizure control is not
jeopardized include increasing the phenytoin dosage, switch-
ing to an antiepileptic agent that is not affected by enteral
feeding formulas, switching to another enteral feeding for-
mula, or using intravenous phenytoin. Which method is
the best to use depends on, but is not limited to, the nutri-
tional status of the patient, the presence of seizures, the ur-
gency to obtain therapeutic phenytoin concentrations, and
concurrent medications. While serum concentrations can
serve as a useful tool to monitor patients receiving pheny-
toin and enteral feedings, it is very important to interpret
drug concentrations only in the context of a patient’s clini-
cal condition.32
Summary
Although the underlying cause of the phenytoin–enteral
feeding interaction remains unknown, evidence supports
the existence of this interaction in patients and in vitro
studies, but not in healthy volunteers. We have provided
recommendations for minimizing this drug–nutrient inter-
action in patients, although there is no good consensus as
to the best intervention.
Sam CS Au Yeung BSc(Pharm) PhD Candidate, Division of Phar-
maceutics and Biopharmaceutics, Faculty of Pharmaceutical Sci-
ences, University of British Columbia, Vancouver, British Columbia,
Canada
Mary HH Ensom PharmD FASHP FCCP, Associate Professor, Di-
vision of Clinical Pharmacy, Faculty of Pharmaceutical Sciences,
University of British Columbia; Clinical Pharmacy Specialist, Phar-
macy Department, Children’s and Women’s Health Centre of British
Columbia, Vancouver
Reprints: Mary HH Ensom PharmD FASHP FCCP, Pharmacy De-
partment 0B7, Children’s and Women’s Health Centre of British
Columbia, 4480 Oak St., Vancouver, British Columbia V6H 3V4,
Canada, FAX 604/875-3735, E-mail ensom@interchange.ubc.ca
References
1. Bauer LA. Interference of oral phenytoin absorption by continuous naso-
gastric feedings. Neurology 1982;32:570-2.
2. Faraji B, Yu PP. Serum phenytoin levels of patients on gastrostomy tube.
J Neurosci Nurs 1998;30:55-9.
3. Maynard GA, Jones KM, Guidry JR. Phenytoin absorption from tube
feedings. Arch Intern Med 1987;147:1821.
4. O’Hagan M, Wallace SJ. Enteral formula feeds interfere with phenytoin
absorption. Brain Dev 1994;16:165-7.
5. Rodman DP, Stevenson TL, Ray TR. Phenytoin malabsorption after je-
junostomy tube delivery. Pharmacotherapy 1995;15:801-5.
6. Saklad JJ, Graves RH, Sharp WP. Interaction of oral phenytoin with en-
teral feedings. JPEN J Parenter Enteral Nutr 1986;10:322-3.
7. Sneed RC, Morgan WT. Interference of oral phenytoin absorption by en-
teral feedings. Arch Phys Med Rehab 1988;69:682-4.
8. Hatton RC. Dietary interaction with phenytoin (letter). Clin Pharm 1984;
3:110-1.
9. Taylor DM, Massey CA, Dhillon S. Lowered serum phenytoin concen-
trations during therapy with liquid food concentrates (letter). Ann Phar-
macother 1993;27:369.
10. Weinryb J, Cogen R. Interaction of nasogastric phenytoin and enteral
feeding solution (letter). J Am Geriatr Soc 1989;37:195-6.
11. Worden JP, Wood CA Jr, Workman CH. Phenytoin and nasogastric feed-
ings (letter). Neurology 1984;34:132.
12. Bader MK. Case study of two methods for enteral phenytoin administra-
tion. J Neurosci Nurs 1993;25:233- 42.
13. Olsen KM, Hiller FC, Ackerman BH, McCabe BJ. Effect of enteral
feedings on oral phenytoin absorption. Nutr Clin Pract 1989;4:176-8.
904 ■ The Annals of Pharmacotherapy ■ 2000 July/August, Volume 34
Downloaded from aop.sagepub.com at NANYANG TECH UNIV LIBRARY on May 21, 2015
14. US Preventive Services Task Force. Guide to clinical preventive ser-
vices: report of the US Preventive Services Task Force. 2nd ed. Balti-
more: Williams & Wilkins, 1996:861-2.
15. Marvel ME, Bertino JS. Comparative effects of an elemental and a com-
plex enteral feeding formulation on the absorption of phenytoin suspen-
sion. JPEN J Parenter Enteral Nutr 1991;15:316-8.
16. Nishimura LY, Armstrong EP, Plezia PM, Iacono RP. Influence of enteral
feedings on phenytoin sodium absorption from capsules. Drug Intell Clin
Pharm 1988;22:130-3.
17. Doak KK, Curtis EH, Dunnigan KJ, Reiss RA, Reiser JR, Huntress J, et
al. Bioavailability of phenytoin acid and phenytoin sodium with enteral
feedings. Pharmacotherapy 1998;18:637-45.
18. Krueger HA, Garnett WR, Comstock TJ, Fitzsimmons WE, Karnes HT,
Pellock JM. Effect of two administration schedules of an enteral nutrient
formula on phenytoin bioavailability. Epilepsia 1987;28:706-12.
19. Guidry JR, Eastwood TF, Curry SC. Phenytoin absorption on volunteers
receiving selected enteral feedings. West J Med 1989;150:659-61.
20. Ozuna J, Friel P. Effect of enteral tube feeding on serum phenytoin lev-
els. J Neurosurg Nurs 1984;16:289-91.
21. Randall CTC, Tett SE. Phenytoin pharmacokinetics after intravenous ad-
ministration to patients receiving enteral tube feeding. Pharm World Sci
1994;16:217-24.
22. Gilbert S, Hatton J, Magnuson B. How to minimize interaction between
phenytoin and enteral feedings: two approaches — a strategic approach.
Nutr Clin Pract 1996;11:28-30.
23. Hatton J, Magnuson B. How to minimize interaction between phenytoin
and enteral feedings: two approaches — therapeutic options. Nutr Clin
Pract 1996;11:30-1.
24. Cacek AT, DeVito JM, Koonce JR. In vitro evaluation of nasogastric ad-
ministration methods for phenytoin. Am J Hosp Pharm 1986;43:689-92.
25. Cutie AJ, Altman E, Lenkel L. Compatibility of enteral products with
commonly employed drug additives. JPEN J Parenter Enteral Nutr 1983;
7:186-91.
26. Fleisher D, Sheth N, Kou JH. Phenytoin interaction with enteral feedings
administered through nasogastric tubes. JPEN J Parenter Enteral Nutr
1990;14:513-6.
27. Holtz L, Milton J, Sturek JK. Compatibility of medications with enteral
feedings. JPEN J Parenter Enteral Nutr 1987;11:183-6.
28. Hooks MA, Longe RL, Taylor AT, Francisco GE. Recovery of phenytoin
from an enteral nutrient formula. Am J Hosp Pharm 1986;43:685-8.
29. Smith OB, Longe RL, Altman RE, Price JC. Recovery of phenytoin from
solutions of caseinate salts and calcium chloride. Am J Hosp Pharm
1988;45:365-8.
30. Splinter MY, Seifert CF, Bradberry JC, Tu YH, Allen LV. Effect of pH
on the equilibrium dialysis of phenytoin suspension with and without en-
teral feeding formula. JPEN J Parenter Enteral Nutr 1990;14:275-8.
31. Messahel FM, Solis GL, Aguwa CN. Does total parenteral nutrition low-
er serum phenytoin levels? Curr Ther Res 1990;47:1017-20.
32. Ensom MHH, Davis GA, Cropp CD, Ensom RJ. Clinical pharmacoki-
netics in the 21st century: does the evidence support definitive out-
comes? Clin Pharmacokinet 1998;34:265-79.
EXTRACTO
OBJETIVO: Revisión sistemática de la evidencia que apoya/refuta la
interacción entre las formas de dosificación de fenitoína y las
formulaciones de alimentación enteral y los mecanismos propuestos e
intervención recomendada para esta interacción.
FUENTES DE INFORMACIÓN: Búsqueda bibliográfica de artículos en inglés
en MEDLINE de enero de 1966 a abril de 2000. Esta búsqueda se
suplementó con una revisión bibliográfica de artículos relevantes.
SELECCIÓN DE FUENTES DE INFORMACIÓN: Se seleccionaron estudios con
los siguientes diseños: controlados, prospectivos, aleatorios; controlados,
prospectivos, no aleatorios; prospectivos, no controlados, no aleatorios;
y retrospectivos. También se evaluaron informes de experiencias
clínicas, informes de casos, análisis in vitro, y cartas.
SELECCIÓN DE DATOS: Se obtuvieron los siguientes datos: diseño del
estudio, tipo y número de sujtos (pacientes o voluntarios saludables),
método de administración de fenitoína y administración enteral,
formulación de fenitoína, tipo de administración (continua o
interrumpida), resultados, y mecanismo de la interacción.
www.theannals.com

SÍNTESIS: A pesar de que cuatro estudios prospectivos, aleatorios, y
controlados en voluntarios saludables refutan la existencia de la
interacción, se encontraron numerosos informes y estudios que
demostraron una dramática disminución en las concentrationes séricas
de fenitoína cuando esta se administra con formulaciones enterales. Por
lo tanto, la evidencia apoya la existencia de esta interacción en pacientes
y en estudios in vitro pero no en sujetos saludables.
Desafortunadamente, el mecanismo de esta interacción es desconocido.
Se han propuesto muchos métodos para prevenir o tratar la interacción
una vez haya ocurrido, sin embargo todavía no existe una estrategia
práctica, y generalmente aceptada.
CONCLUSIÓNES: Debido a la falta de estudios controlados, prospectivos y
aleatorios en pacientes, el papel de la alimentación enteral en esta
interacción no está clara. Sin embargo, una disminución en la
concentración sérica de fenitoína asociada con alimentación enteral
puede aumentar el riesgo de convulsiones. Los profesionales de la salud
deben estar al tanto del potencial de interacción fármaco-nutriente para
diseñar un plan de cuidado específico que tome en consideración la
formulación de la alimentación enteral, el método de administración, la
forma de dosificación de fenitoína, el régimen de administración y
seguimiento.
Sonia I Lugo
RÉSUMÉ
OBJECTIF: Faire une revue systématique de l’interaction entre les
formulations orales de phénytoïne et les préparations d’alimentation
entérale, en évaluer l’importance, les mécanismes proposés et
recommander un plan d’intervention.
PROVENANCE DES DONNÉES: Une recherche automatisée à partir de la
banque de données MEDLINE a été effectuée (janvier 1966 et avril
2000). Seules les publications de langue anglaise ont été retenues. À
www.theannals.com The Annals of Pharmacotherapy
Downloaded from aop.sagepub.com at NANYANG TECH UNIV LIBRARY on May 21, 2015
Phenytoin–Enteral Feeding Interaction
cette recherche, s’est ajouté la revue des bibliographies des publications
sélectionnées.
SÉLECTION DES DONNÉES: Toutes les études cliniques, qu’elles soient
prospectives ou rétrospectives, contrôlées ou non, à répartition aléatoire
ou non ont été évaluées. Les rapports de cas, les études in vitro et les
lettres à l’éditeur ont aussi été considérées. Les informations retenues
comprennent le plan d’étude, le type (patients ou volontaires sains) et le
nombre de sujets, le mode d’administration de la phénitoïne et de
l’alimentation entérale, la formulation de phénitoïne, le type
d’alimentation (continue ou interrompue), les résultats, et les
mécanismes proposés.
RÉSUMÉ: Bien que quatre études prospectives, contrôlées, à répartition
aléatoire effectuées chez des volontaires sains rejettent l’existence d’une
telle interaction, il existe de nombreux rapports et études démontrant une
réduction importante des concentrations sériques de phénitoïne lorsque
celle-ci est administrée en même temps que des préparations
d’alimentation entérale. Les évidences supportent donc l’existence d’une
telle interaction chez les patients et in vitro, mais pas chez les volontaires
sains. Malheureusement, le mécanisme exact de cette interaction est
toujours inconnu. Plusieurs méthodes ont été essayées afin de prévenir
ou traiter cette interaction lorsqu’elle se présente, cependant, aucune de
celles-ci ne semble faire l’unanimité.
CONCLUSIONS: En l’absence d’études prospectives, contrôlées à
répartition aléatoire chez des patients, le rôle de l’alimentation entérale
dans cette interaction est encore nébuleux. Cependant, comme une
réduction des concentrations sériques de phénitoïne peut entraîner une
augmentation du risque de crise épileptique, les cliniciens devraient être
au courant d’une telle possibilité. Le plan de soins devrait aussi prendre
cette possibilité en considération.
Suzanne Laplante
■ 2000 July/August, Volume 34 ■ 905
Management of phenytoin with enteral tube feeding
Nicole Phelps, PharmD, BCPP1

1
Clinical Pharmacist, Duke University Hospital, Durham, NC

ABSTRACT
Though the mechanism of the interaction is poorly understood, phenytoin absorption is reduced when given
concomitantly with enteral tube feedings. This article reviews strategies that may help maintain phenytoin levels when
enteral nutrition is required.
KEYWORDS

Downloaded from http://meridian.allenpress.com/mhc/article-pdf/2/5/108/2094882/mhc_n126907.pdf by Indonesia user on 24 May 2021

phenytoin, enteral tube feeding, drug-nutrition interaction
Though the interaction between enteral tube feeding and with bolus feeding, but clear communication with nursing
phenytoin was first described 30 years ago, the staff is important to ensure patients receive all phenytoin
mechanism of the interaction is still poorly understood.1 doses and the required amount of nutrition.
One theory is that the drug can adhere to the plastic
Even with separating phenytoin from tube feeds, the
tubing when phenytoin is given via tube.2 Another
phenytoin dose will still likely need to be substantially
possibility is that physical incompatibility with the enteral
increased to maintain therapeutic serum levels once tube
nutrition causes decreased absorption of phenytoin.2 It
feeding is started. If the patient has been stabilized on a
remains unclear which components of the enteral
therapeutic oral or intravenous dose, it is common clinical
nutrition are most problematic or if there may be fewer
practice to start with a 50% empiric increase in dose once
interactions with some formulas versus others.2 Whatever
enteral nutrition is initiated. It is also important to
the mechanism, phenytoin absorption is reduced when
remember to reduce the dose if tube feeds are held,
given concomitantly with tube feedings to such an extent
stopped or if the phenytoin administration route is
that phenytoin serum levels may be reduced as much as
changed to oral or IV. Phenytoin serum levels must be
50 to 75%, leaving the patient at risk for seizures.3
followed closely and the dose adjusted as needed to
Additionally, there is not a standard approach to
maintain therapeutic levels. It may also be useful to
managing this drug-nutrition interaction, though there
measure free phenytoin levels if possible, as many
are some strategies that may help maintain therapeutic
patients requiring enteral nutrition have some degree of
phenytoin levels when enteral nutrition is necessary.
malnutrition and altered protein binding with phenytoin
The first and most important intervention is to stagger due to hypoalbuminemia. Free phenytoin levels are a
the administration of phenytoin to separate it as much as more accurate way to determine if the level is therapeutic
possible from the tube feedings. Ideally, the feeds should in nutritionally complex patients, though they are not
be held for 1-2 hours before and after each phenytoin always available or cost effective.
dose with adequate tube flushes before and after
The final factor to consider is the type of phenytoin
phenytoin administration.2,3,4 If the feeding schedule is
formulation to use. If the patient has been taking
continuous, the tube feeding rate will have to be
extended release capsules once daily, this usually is
increased for the remaining time to ensure that the
converted to immediate release phenytoin that should be
patient receives the required calories.3,4 Since immediate
divided two to three times daily. The available immediate
release phenytoin must be administered multiple times
release formulations are phenytoin suspension or
daily, holding continuous tube feeds 2 to 3 times daily
phenytoin chewable tablets. If phenytoin suspension is
may result in a rate of feeding the patient cannot tolerate.
selected and dispensed as a bulk product, it is very
One method described in the literature for continuous
important for the nurse to shake the bottle vigorously and
feeding is to make a slurry with extended release
give the dose immediately. It is difficult to get the drug
phenytoin capsules and administer it just once daily.4 As
evenly distributed in suspension, and if the nurse neglects
long as the micro crystal particles are not crushed when
to shake well each time, the drug concentration may be
making the slurry, the extended release properties should
higher in the bottom of the bottle in comparison to the
be maintained enough to administer phenytoin once
top, which would likely result in inconsistent serum levels.
daily. Staggering doses from feeding times is much easier

Mental Health Clinician, November 2012, Vol. 2, Issue 5 108

If unit dose suspension is available, however, this problem
can be avoided.4 In general, chewable tablets are
preferable to bulk phenytoin suspension for a more
consistent and uniform dose. However, crushed tablets
are more likely than the suspension to clog small bore
tubes.
If a patient’s phenytoin levels continue to be problematic
after having tried the above interventions, intravenous
phenytoin or fosphenytoin may be reasonable
alternatives until the patient tolerates oral meds. Another
option may be discontinuing phenytoin and changing to
another antiepileptic medication. As always, it is

Downloaded from http://meridian.allenpress.com/mhc/article-pdf/2/5/108/2094882/mhc_n126907.pdf by Indonesia user on 24 May 2021

important to remember to treat the patient and not the
lab values. Therefore, consider the patient’s overall
history, seizure risk, other medications, and long-term
nutrition plan when making recommendations and
decisions about managing phenytoin and tube feeds.
REFERENCES
1. Bauer LA. Interference of oral phenytoin absorption by continuous
nasogastric feedings. Neurology. 1982;32(5):570-2. PubMed PMID:
6803191.
2. Au Yeung SC, Ensom MH. Phenytoin and enteral feedings: does evidence
support an interaction?. Ann Pharmacother. 2000;34(7-8):896-905.
PubMed PMID: 10928402.
3. Beckwith MC, Barton RG, Graves C. A guide to drug therapy in patients
with enteral feeding tubes: dosage form selection and administration
methods. Hospital Pharmacy. 2004;39(3):225-237.
4. Gilbert S, Hatton J, Magnuson B. How to minimize interaction between
phenytoin and enteral feedings: two approaches. Nutr Clin Pract.
1996;11(1):28-31. PubMed PMID: 8700059.

How to cite this editor-reviewed article

Phelps N. Management of phenytoin with enteral tube feeding. Ment Health
Clin [Internet]. 2012;2(5):108-9. Available from:
http://dx.doi.org/10.9740/mhc.n126907

Mental Health Clinician, November 2012, Vol. 2, Issue 5 109

 Enteral Feeding - Preventing Drug
Nutrient Interactions UHL Guideline Nutrition and Dietetic Service

Trust reference B62/2019

1. Introduction and Who Guideline applies to

1.1 This guideline is for Dietitians to refer to, when reviewing patients on an enteral feed
and medications. It may also be useful of other members of the Multidisciplinary
Team (MDT).
1.2 The purpose of this guideline is to highlight possible drug and enteral feed
interactions, advise when a break should be provided in an enteral feeding regimen
to allow effective drug administration and allow the MDT to work effectively together
to avoid drug nutrient interactions that may cause the medication to be ineffective.
1.3 A drug nutrient interaction has been defined as one that causes a 20% change in the
pharmacodynamics or pharmacokinetics of the medication (Lingtak-Neander, 2013).
1.4 It is the responsibility of the Dietitians along with other MDT members to highlight
these interactions. The Dietitian can then make appropriate adjustments to the
patients enteral feeding plan.
1.5 There are 4 types of reactions:
i) Chemical interaction, binding the drug and reducing its absorption.
ii) Physical interaction between the drug formulation and the feed formulation, causing a
change in the feed consistency and potentially resulting in blockage of the feeding
tube.
iii) Interaction between the drug and a specific nutrient involved in the metabolism of
that drug which can cause loss of drug effect due to impaired absorption, increased
drug clearance or blocking of pharmacological action.
iv) Interaction with the feeding tube itself which can reduce plasma concentration of the
drug compared to oral administration.

2. Guideline Standards and Procedures

2.1 General points to consider when a patient is on an enteral feed and requires
medications that may interact with it.
2.2 Will Nurses be able to realistically implement the number of recommended breaks?
2.3 Will the number of breaks limit your ability to meet the patient’s full nutritional needs?
2.4 Will the number of breaks lead to higher feed rates and will that impact feed
tolerance?
2.5 Liaise with pharmacy and medical team to review drug necessity (de-prescribe where
possible), medication preparations, administration routes, timings or correct
monitoring of the therapeutic effect.
2.6 Drugs with enteric coating should not be crushed to go down a tube as the enteric
coating is designed to protect the medication. Slow release drugs should not be crushed as
these are intended to be released over a long period of time. Cytotoxic and hormone
medications should not be crushed due to risk to the administrator. Please refer to policy
Administration of Medicines to Adult Patients who cannot Swallow Solid Dosage Forms (e.g:
Tablets or Capsules) Guidelines for Practice (B31/2008) for further guidance.

Enteral Feeding - Preventing Drug Nutrient Interactions UHL Guideline

V2 approved by Policy and Guideline Committee on 20 March 2020 Trust ref: B62/2019 Next Review: April 2023
NB: Paper copies of this document may not be most recent version. The definitive version is held on INsite Documents
2.7 Ensure that nurses are flushing the feeding tubes adequately before, between and
after medications to minimise tube blocking. Please refer to Enteral Feeding Tube
Administration in Adults Policy and Procedures (B30/2019)
2.8 Potential drug and enteral feed interactions will be documented on the enteral
feeding regimen, and in the dietetic medical note entry. Verbal handover will be given
to nursing staff and added to nerve centre where appropriate. Long term plans are to
include a flagging system on eMeds.

2.8 Table one contains a list of possible drugs that may interact with enteral feeds. This
list is not exhaustive and enteral feed and drug interactions should be considered if a
medication is not providing expected results. Liaise with Pharmacist for more
information. The information has been taken from 3 main sources and where there is
conflicting information, the longest break period has been recommended.

Table one. Medications that require an enteral feed break

Name of Interaction Administration Additional
Medication information and
action required
Anti-retroviral Dolutegravir interacts Dolutegravir
with feed and Should be
(HIV) medications containing administered 2 hours
Dolutegravir polyvalent cations, such before or 6 hours after
Elvitegravir as calcium, iron, zinc enteral feeds
Raltegravir magnesium
supplements.

Elvitegravir and
Raltegravir interact with Elvitegravir and
medications and feeds Raltegravir
containing polyvalent
cations, such as calcium, Should be separated
iron, zinc magnesium by 4 hours from
supplements. enteral feeds

Bisphosphonates Feed/food reduces 2 hour break in Careful when

: absorption feeding before and 2 crushing as drug
Risedronate hour break in feeding powder can cause
Alendronic acid after administration of irritation
drug
(osteoporosis)

Carbamazepine Interacts with enteral 2 hour break in Medical team to

feeding tube (polyvinyl feeding before and 2 check bloods levels
(seizures) tube) which in some hour break in feeding of the drug
studies has been shown after administration of regularly.
to reduce plasma drug if therapeutic If short term
concentration of the drug levels of carbamazepine is
when compared with oral carbamazepine are required a
administration. not reached. suppository is
Enteral Feeding - Preventing Drug Nutrient Interactions UHL Guideline
V2 approved by Policy and Guideline Committee on 20 March 2020 Trust ref: B62/2019 Next Review: April 2023
NB: Paper copies of this document may not be most recent version. The definitive version is held on INsite Documents

Digoxin This is affected by high
fibre intake. Changes in
(Atrial fibrillation) enteral fibre content can
have an effect on drug’s
narrow therapeutic
window
Fluconazole Some reports indicate
that enteral fluconazole
(Antifungal) interacts with feeds
particularly Jevity.
Itraconazole Requires stomach pH to
be as acidic as possible
(Antifungal) for absorption.(NB - this
does not apply to
commercial liquid
preparation)
Levothroxine Interacts with calcium
and iron in enteral feeds
(hypothyroidism) and interacts with the
feeding tube. Most likely
to have reduced
therapeutic levels with
jejunostomy feeds.
Enteral Feeding - Preventing Drug Nutrient Interactions UHL Guideline
V2 approved by Policy and Guideline Committee on 20 March 2020 Trust ref: B62/2019
NB: Paper copies of this document may not be most recent version. The definitive version is held on INsite Documents
Appropriate dilution of
drug with sterile water
to minimise the
interactions.
If patient requires a
fibre feed stop feed 2
hours before
administration and for
1 hour afterwards. No
break required in a
non fibre containing
feed.
If feeding allows give
a feeding break but if
on continuous not
necessary to
compromise feeding
unless on Jevity in
which case the risk of
sub therapeutic
treatment should be
weighed against
benefits of feeding.
If withheld – 1hr
before and 1 hour
after
2 hour break in
feeding before and 2
hour break in feeding
after administration of
drug.
Usually prescribed in
the morning at 06.00
therefore consider if
need break 1 hour
before and 1 after.
recommended.
Dietitians to raise
awareness of
impact of starting or
stopping an enteral
feed may have on
carbamazepine
levels.
Ensure the medical
team are made
aware – blood
levels need to be
checked regularly.
Changing to a lower
fibre feed can could
cause a rise in the
therapeutic level
This is not time
critical and will be
applicable to
patients receiving
feeds for more than
5-7 days. Suggest
to medical team
that thyroid function
may require
monitoring in long
term patients on
enteral feeds.
Next Review: April 2023
Penicillin The peak plasma levels 2 hour break in No problems with
Flucloxacillin of feeding before and 1 amoxicillin.
phenoxymethylpenicillin hour break in feeding Penicillin and
(antimicrobial) and flucloxacillin are after administration of flucloxacillin are
affected by the presence drug. dosed QDS and so
of food and therefore the number of
best given on an empty breaks may be
stomach. Amoxicillin is impractical. Speak
unaffected. with medical team
or pharmacist to
consider different
antibiotic or
different
preparation.
Phenytoin Phenytoin reacts with the 2 hour break in Not applicable if
enteral feeding tube itself feeding before and 2 prescribed
(seizures) and the protein in the hour break in feeding intravenously (IV).
feed. This reduces drug after administration of
absorption and phenytoin (oral or Discuss with
consequent plasma enteral dose) is Pharmacist the
concentration required. ability to reduce
frequency or alter
timings of
administration

NB – Caution -
despite using
feeding break there
have been reports
of unexpected low
phenytoin levels
Quinolones: Bioavailability can be 2 hour break in Advise medical
decreased by up to 33 feeding before and 2 team to consider
Ciprofloxacin with some quinolones hour break in feeding doses at the higher
Levofloxacin when administered via after administration of end of the dosage
Ofloxacin jejunostomy and drug range if on enteral
Moxifloxacin absorption is reduced a feed and medical
further 25% if Ensure Nurses are team to liaise with
(antibiotics) administered with flushing feeding tube microbiology about
continuous enteral feed thoroughly before and this.
as it binds with ions in after drug
the feed to produce administration.
insoluble chelates.
Rifampicin Interact with Rifampicin:
Isoniazid carbohydrate in enteral 2 hour break in
feed and reduce plasma feeding before and 1
(Antibacterial) levels. hour break in feeding
after administration.

Isoniazid:
1 hour break in
feeding before and 1
hour break in feeding
Enteral Feeding - Preventing Drug Nutrient Interactions UHL Guideline
V2 approved by Policy and Guideline Committee on 20 March 2020 Trust ref: B62/2019 Next Review: April 2023
NB: Paper copies of this document may not be most recent version. The definitive version is held on INsite Documents
 after drug
administration
Tetracycline Absorption is decreased Provide enteral feed Oxytetracycline is
by 80% with enteral break 2 hours before the same as
(antibacterial) feeding as they bind to and after or medical tetracycline as
calcium, iron and team and pharmacist it chelates calcium,
magnesium. to consider a different iron, magnesium
These ions are present medication. and zinc so can be
in higher concentrations affected.
in enteral feed than Doxycycline also
regular diet. binds to calcium
ions and therefore
recommend a break

Trimethoprim Reduced absorption Administer during a Break length not

when administered break to maximise specified in
(antimicrobial) alongside enteral feed. absorption – 1 hour references.
before and 1 hour
after.
Warfarin High levels of vitamin K Consider holding Highlight to medical
(vitamin K in enteral feeds which enteral feed for 1 hour team so that INR
antagonist) may cause variation in before and 1 hour can be closely
the INR. after administration. monitored.
Warfarin also binds to
protein, reducing drug
absorption.

Table Two. Enteral medications that should require therapeutic levels to be

checked, or may block the tube..
Name of Interaction
Medication
Digoxin This is affected by
high fibre intake.
(Atrial fibrillation) Changes in enteral
fibre content can
have an effect on
drug’s narrow
therapeutic window
Administration Additional
information and
action required
If patient requires a Ensure the medical
fibre feed stop feed team are made
2 hours before aware – blood levels
administration and need to be checked
for 1 hour regularly. Changing
afterwards. No break to a lower fibre feed
required in a non can could cause a
fibre containing feed. rise in the
therapeutic level

Phenytoin Phenytoin reacts 2 hour break in Not applicable if

with the enteral feeding before and 2 prescribed
(seizures) feeding tube itself hour break in intravenously (IV).
and the protein in the feeding after
feed. This reduces administration of Discuss with
drug absorption and phenytoin (oral or Pharmacist the
consequent plasma enteral dose) is ability to reduce
concentration required. frequency or alter
Enteral Feeding - Preventing Drug Nutrient Interactions UHL Guideline
V2 approved by Policy and Guideline Committee on 20 March 2020 Trust ref: B62/2019 Next Review: April 2023
NB: Paper copies of this document may not be most recent version. The definitive version is held on INsite Documents

Levodopa Protein in the feed
(sinemet/madopar) and blood stream
competes with the
(Parkinson’s drug for absorption
Disease) and transport into the
brain
Potassium Potassium can
supplementation physically react with
(sandoK, Kay-c- enteral feed or with
Lee) tube itself, thus
leading to tube
(Hypokalaemia) blockage if the
supplement has not
been sufficiently
diluted before
administration.
Theophylline Narrow therapeutic
(bronchodilator) index. A diet high in
protein and low in
carbohydrate can
induce hepatic
metabolism and
reduce bioavailability
of the drug
Enteral Feeding - Preventing Drug Nutrient Interactions UHL Guideline
V2 approved by Policy and Guideline Committee on 20 March 2020 Trust ref: B62/2019
NB: Paper copies of this document may not be most recent version. The definitive version is held on INsite Documents
timings of
administration
NB – Caution -
despite using
feeding break there
have been reports of
unexpected low
phenytoin levels
Higher drug dose or Dietitian to raise
break in feed may awareness to
be required. medical team if
Timing of the breaks protein levels alter
should be consistent within enteral feed.
to reduce variations
in plasma levels set
time in liaison with
pharmacist/ medical
team.
Ensure adequate
flushing by Nurses
Close therapeutic Highlight any
monitoring, changes to enteral
especially if changes feed protein content
made to enteral to prescriber
regimen
Next Review: April 2023
Table 3 Drugs not recommended for administration via enteral tube

Name of medication Reaction Action required

Mucogel Metal ions in the antacids Highlight to pharmacist and
Gaviscon bind protein in the enteral medical teams that patient
Peptac feed which can block the on enteral feed, so that an
Maalox tube. alternative can be
(Antacid) considered.

Sucralfate Combines with enteral feed Administration not

(gastroprotective complex) and forms an insoluble recommended.
complex which can cause
tube blockage.

3. Education and Training

All staff groups highlighted above will need to be made aware of the guideline through team meetings
and department news letter. No formal training will be required.

4. Monitoring Compliance
What will be measured to How will compliance be Monitoring Reporting
Frequency
monitor compliance monitored Lead arrangements
Datix Number of datixes related Dietetic Monthly Via team
to drug and feed Managers meetings and
interactions and newsletters
Medication
Safety
Pharmacist

5. Supporting References
Boullata J.I (2019) Guidebook on Enteral Medication Administration. ASPEN
The Newt Guidelines for administration of medication to patients with enteral feeding tubes or swallowing
difficulties. Access via pharmacy on www.newtguidelines.com (password required).
Lingtak-Neander, C., (2013). Drug-Nutrient Interactions. Journal of Parenteral and Enteral Nutrition, pp.
450-459.
White, R. & Bradnam, V., 2018. Drug, enteral feeding, tube feeding, drug administration Handbook of Drug
Nutrient Interactions. www.medicinescomplete.com ed. London: Pharmacetical Press.

CONTACT AND REVIEW DETAILS

Guideline Lead (Name and Title) Executive Lead: Medical Director
Moira Dawson
Details of Changes made during review:
Pharmacy changes to the version approved in 2019

Enteral Feeding - Preventing Drug Nutrient Interactions UHL Guideline

V2 approved by Policy and Guideline Committee on 20 March 2020 Trust ref: B62/2019 Next Review: April 2023
NB: Paper copies of this document may not be most recent version. The definitive version is held on INsite Documents