Effect of Dose Formulation on Isoniazid Absorption in Two Young Children

(1)

850 PEDIATRICS Vol. 77 No. 6 June 1986

Effect

of Dose

Formulation

on Isoniazid

Absorption

in Two Young

Children

Daniel

A. Notterman,

MD, Michael

Nardi,

MS, and Judy

G. Saslow,

MD

From the Department of Pediatrics, New York University School of Medicine, New York

ABSTRACT.

In an 8-month-old infant with tuberculous

meningitis treatment with isoniazid was unsuccessful and was associated with lower than expected plasma concen-trations of isoniazid (measured concentration 0.1 jzg/

mL). The infant had received isoniazid as a crushed tablet

admixed with apple sauce. Oral administration of the

parenteral solution of isoniazid (Nydrazid, Squibb) mixed

in apple juice produced a higher isoniazid concentration

(2.9 .tg/mL) and the child improved clinically. Pharma-cokinetic studies in two subjects were performed follow-ing intramuscular injection of isoniazid and oral admin-istration of (1) an isoniazid tablet crushed and mixed

with apple sauce, (2) parenteral isoniazid solution mixed

with apple juice, and (3) a commercially available syrup

containing isoniazid and pyridoxine (P-I-N Forte, Lan-nett). Of the three oral preparations, the syrup produced the highest peak concentrations (8.3 and 6.9 ig/mL). The

crushed tablet in apple sauce produced the lowest peak

concentrations (1.4 and 2.4 zg/mL). Administration of

crushed isoniazid tablets with food may be associated

with impaired gastrointestinal absorption, lower than expected isoniazid concentrations, and treatment failure. Pediatrics 1986;77:850-852; isoniazid, dose formulation,

tuberculosis.

Isoniazid is a frequently prescribed drug and is well absorbed in children old enough to ingest the intact tablet.’ Liquid oral preparations of isoniazid for infants and young children who are unable to

ingest tablets are not commonly available. Such

children are customarily treated with ad hoc

prep-arations of isoniazid; however, the absorption of the drug in these forms is unknown. Isoniazid has

been orally administered after mixing crushed

tab-lets with foodstuffs such as apple sauce2’3 or mixing

the panenteral solution in apple juice. We

encoun-tened an infant in whom use of the crushed tablet

Received for publication May 7, 1985; accepted Sept 30, 1985. Reprint requests to (D.A.N.) Department of Pediatrics, Division

of Critical Care Pediatrics, The New York Hospital, 525 E 68 St, New York, NY 10021

in apple sauce was associated with low plasma isoniazid levels and treatment failure. The phar-macokinetic studies stimulated by this experience indicate that the common methods of administering isoniazid to young children may limit absorption of the drug and contribute to therapeutic failure.

CASE REPORT

An 8-month-old Korean girl was diagnosed as having

tuberculous meningitis with an organism susceptible to isoniazid and nifampin. The child was well until 6 months of age when continuous fever occurred. At 8 months of

age, she was admitted to the hospital with signs of

men-ingeal infection and left cranial nerve VI palsy. The CSF

protein and glucose were 162 and 9 mg/dL, respectively,

and the fluid contained 78 WBCs per milliliter (26%

polymorphonuclear cells). The CSF culture was positive

for Mycobacterium tuberculosis. Growth of the organism

was inhibited by concentrations of isoniazid and nifampin of 0.2 and 1.0 tg/mL, respectively.

The child was treated with isoniazid, 20 mg/kg/d;

nifampin, 20 mg/kg/d; streptomycin, 40 mg/kg/d; and methylpresnisolone, 2 mg/kg/d. Isoniazid was adminis-tered orally in the form of a crushed tablet mixed with

varying amounts of apple sauce. Although the child

ap-peared to respond favorably, several efforts to discontinue

the streptomycin were associated with recrudescence of

fever. Methylpresnisolone was changed to hydrocortisone

to facilitate dosage adjustment. After a period of apparent improvement, streptomycin and the corticosteroid were discontinued. Fever recurred and was associated with right cranial nerve III and VI palsies and CSF pleocytosis.

Serum isoniazid concentrations immediately before and two hours after an oral dose of the crushed isoniazid tablet (10 mg/kg) mixed with apple sauce were both 0.1

tg/mL, the lower limit of detection by the assay used. The concentration of nifampin two hours following a dose

was 11 g/mL. When a parenteral form of isoniazid

(Nydrazid, Squibb) was mixed with apple juice (8 oz) and given by mouth, the serum isoniazid concentration two

hours after administration was 2.9 g/mL. This

formu-lation was continued, and a third, 5-month course of streptomycin was prescribed. The child’s mental status and findings on neurologic examination improved, and

at Viet Nam:AAP Sponsored on September 7, 2020 www.aappublications.org/news

(2)

Subject I

]

Subject 2

E

a,

NJ 0

0

U,

I 2345678

Hours Following Test

Dose

(10mg

per kg)

ARTICLES

851

the CSF became normal within 2 months (protein and

glucose 45 and 50 mg/dL, respectively; WBC count, three

lymphocytes per milliliter).

PHARMACOKINETIC STUDIES

Method

Plasma isoniazid concentrations were measured

in two children, 12 and 18 months of age. The 12-month-old child (patient 1) is described in the case

report. On different days, the patients were given

test doses (10 mg/kg) of three different oral prep-arations of isoniazid: (1) the crushed tablet mixed with a household tablespoon of apple sauce, (2) the

panenteral solution (Nydrazid, Squibb) mixed with

8 oz of apple juice, and (3) a commercially available

syrup that contains isoniazid (10 mg/mL) and

pyr-idoxine (0.5 mg/mL) in a methylcellulose vehicle (P-I-N Forte, Lannett). Each patient also received an intramuscular injection (10 mg/kg) of the par-enteral solution. Because both children were

receiv-ing therapy for tuberculosus, no drug-free period

between tests occurred. However, the dosing inter-val, 12 hours, exceeded the elimination half-life of

isoniazid by at least a factor of three, even in individuals with the slow acetylaton phenotype. Al-though the study plan was reviewed with the

chil-dren’s parents, institutional review and formal

writ-ten informed consent were not deemed necessary

because the absorption data were obtained to

opti-mize therapy in these individual children.

The test dose was administered in the morning, following an overnight fast. Immediately before and at intervals up to eight hours after the test dose

(Figure), blood specimens were collected. The sam-ples were centrifuged within 30 minutes of collec-tion, and the plasma was separated and frozen at -20#{176}Cuntil analyzed. The plasma samples were

assayed for isoniazid concentration using the

spec-trophotofluorometnic method of Olson et al.4 The lower limit of detection with this assay is less than

0.1 g/mL.

RESULTS

The time concentration curves for each patient are shown in the Figure. The respective peak con-centrations after the crushed tablet in apple sauce (1.4 and 2.5 jg/mL) or orally administered paren-teral solution (3.3 and 2.5 zg/mL) were much lower than after the syrup (8.3 and 6.9 g/mL) or the intramuscular injection (11.4 and 6.3 sg/mL). The time required to achieve a peak concentration was longer after the crushed tablet and oral solution

(120 minutes) than after the syrup (45 and 60

minutes) and intramuscular injection (15 and 30 minutes).

DISCUSSION

The present report indicates that a crushed

iso-E

a,

NJ

D

0

U,

I 2345678

Hours Following

Test Dose

(10mg per kg)

Figure. Plasma concentration of isoniazid in two patients as a function of time and type

of isoniazid preparation. Subject 1, 12 months of age; subject 2, 18 months of age. Each curve represents the concentration of isoniazid following a different preparation.

#{149},

intramuscular injection; A, syrup (P-I-N Forte, Lannett) orally; 0, parenteral solution (Nydrazid, Squibb) mixed with apple juice, orally;

,

crushed tablet mixed with apple

sauce, orally.

(3)

852 ISONIAZID

ABSORPTION

niazid tablet mixed with apple sauce is not well absorbed from the gastrointestinal tract. In pre-vious work in children, Olson et a!’ found that the isoniazid concentrations recorded two hours after a dose were similar following oral administration of the intact tablet and intramuscular injection. This reflects the common finding that gastrointestinal absorption of isoniazid is usually rapid and nearly complete.5 For example, after an average oral dose of 11.6 mg/kg ofthe intact tablet, Olson et aP noted a mean peak isoniazid concentration of 10.4 tg/mL (estimated from published time concentration curves), and Gelber et a15 has reported that in six adults given 10 mg/kg of isoniazid tablets peak concentrations of 10.3 to 11.4 xg/mL developed.

Rapid and complete absorption of isoniazid clearly did not occur in our two patients, in whom the peak concentrations were substantially lower following administration of a crushed tablet than following intramuscular injection.

Further evidence of poor absorption of the crushed tablet is the relatively long delay in reach-ing peak concentrations (120 minutes) after this dosage form as compared with the syrup (45 and 60 minutes). Taken together, these findings indicate that crushing the isoniazid tablet and mixing it with apple sauce slows absorption of the drug and

may limit the total amount of dose absorbed. More

detailed kinetic studies would be required to

estab-lish whether a reduction in the total amount of a

dose absorbed had occurred. However, support for invoking incomplete, and not merely delayed, ab-sorption of the crushed tablet is provided by the data of Morse and Tull,6 who noted a 36% reduction in absorption when the drug was administered with a meal. The extent of reduction was related to the amount of food consumed. This experience has been reported by others.7 The mechanism of this effect

has not been described.

We were able to demonstrate impaired and prob-ably incomplete absorption of the crushed isoniazid tablet under optimal conditions of administration, using small and controlled amounts of apple sauce. Under actual clinical conditions, in which the iso-niazid would be admixed with varying amounts and kinds of food, there might be even greater impair-ment of absorption. This appears to have been the case during treatment of our patient 1, in whom peak and trough plasma concentrations were both 0.1 jg/mL before the dosage form was changed.

IMPLICATIONS

Occurrence of lower than expected plasma iso-niazid concentrations could lead to treatment

fail-ure, because the frequency of this adverse outcome

has been related to isoniazid concentrations in the

blood.8’9 Certainly, when extremely low concentra-tions are present, as in our patient during therapy, therapeutic efficacy cannot be assured. Low plasma concentrations might be particularly dangerous when treating tuberculous meningitis, because the concentration developed in the CSF may be lower than that achieved in the plasma.’#{176}

SUMMARY

The common practice of administering isoniazid to children in the form of a crushed tablet admixed with apple sauce can evidently result in lower than expected plasma levels of the drug and treatment failure. Children should not be treated with un-tested, ad hoc preparations of drugs. Common prac-tice and textbook recommendations based on such practice may be seriously misleading.

ACKNOWLEDGMENT

The authors thank Drs J. Dancis and K. Krasinski for their encouragement and review of the manuscript.

ADDENDUM

Since preparation of the manuscript, P-I-N Forte has

become unavailable. Other oral liquid preparations are now available.

REFERENCES

1. Olson WA, Pruit AW, Dayton PG: Plasma concentrations

of isoniazid in children with tuberculous infections.

Pediat-rics1981;67:876-878

2. Smith MHD, Marquis JR: Tuberculosis and other

mycobac-terial infections, in Feigan RD, Cherry JD (eds): Textbook ofPediatric Infectious Diseases. Philadelphia, WB Saunders

Co,1981, 1045

3. Lincoln EM, Sewall EM: Tuberculosis in Children. New

York, McGraw-Hill, 1963, p 61

4. Olson WA, Dayton PG, Israii ZH: Spectrophotofluorome-tric assay for isoniazid and acetyl isoniazid in plasma adapted to pediatric studies. Clin Chem 1977;23:745

5. Gelber R, Jacobsen BA, Levy L: A study of the availability

of six commercial preparations of isoniazid. Clin Pharmacol Ther 1969;10:841

6. Morse WC, Tull AH: Competition between food and drugs.

Drug Ther, May 1976, p40

7. Melander A, Danielson A, Hanson L: Reduction of isoniazid bioavailability in normal men by concommitant intake of

food. Acta Med Scand 1976;200:93

8. Mitchell RS, Be11JC: Clinical implications ofisoniazid blood

levels in pulmonary tuberculosis. N Engl J Med

1957;257:1066

9. Mandel W, Heaton AD, Russell WF: Combined drug

treat-ment of tuberculosis. J Clin Invest 1959;38:1356

10. Mandell GL, Sande MA: Drugs used in the chemotherapy

of tuberculosis and leprosy, in Gilman AG, Goodman LS,

Gilman A (eds): The Pharmacological Basis of Therapeutics, ed 6. New York, Macmillan, 1980, pp 1201-1203

(4)

1986;77;850

Pediatrics

Daniel A. Notterman, Michael Nardi and Judy G. Saslow