435-437
Should pyridoxine be given to breastfed infants whose mothers are on isoniazid?
Authors: Khuen Foong Ng, Srini Bandi
Number of views: 238
A mother was recently started on isoniazid and rifampicin for
latent tuberculosis infection. She was breastfeeding her 1-monthold
infant. There was no indication to treat the child with antituberculous
therapy. As isoniazid can be present in breast milk,
question was raised whether the baby should receive pyridoxine
supplementation to prevent peripheral neuropathy or seizures.
There were variable views in the management approach due to
uncertainties of evidence in this topic. The authors discovered
vast differences in clinical practice even among the paediatric
infectious diseases experts.
Tuberculosis continues to be a threat to the global health, being
one of the most important cause of mortality and morbidity among
adults[1]. About 10 million people are newly diagnosed to have
tuberculosis on a yearly basis[1]. The mainstay of tuberculosis
treatment is with isoniazid, rifampicin, pyrazinamide and
ethambutol. Isoniazid and rifampicin form the backbone of antituberculous
therapy and they are usually continued for minimum
6 months depending on the site and severity of infection.
One of the recognised adverse effects of isoniazid is peripheral
neuropathy because it induces pyridoxine deficiency or by
competitive inhibition of pyridoxal action[2,3]. Pyridoxine
or vitamin B6 plays an important role in neuronal cell
survival because of its coenzyme activity in neurotransmitter
biosynthesis[4]. Deficiency of pyridoxine has long been attributed
as a cause of peripheral neuropathy although the evidence is
scant[2]. Supplementation of pyridoxine is therefore perceived
to be crucial in the prevention of peripheral neuropathy
especially among high risk groups i.e. pregnant/breastfeeding
mothers, people living with HIV infection, alcohol dependency,
malnutrition, diabetes, chronic liver disease or renal failure[5].
The occurrence of peripheral neuropathy due to pyridoxine
deficiency in association with isoniazid in children is rare, up
to 0.7%[6-8]. It is related to the dose of isoniazid, i.e. the higher
the dose of isoniazid, the higher is the incidence of peripheral
neuropathy[2]. There were 2 case reports indicating that isoniazid,
when given to a young infant, can induce pyridoxine responsive
seizures[9,10]. Multiple case series have also implicated isoniazid
overdose as a cause of neurotoxicity[11].
If a mother is prescribed with isoniazid (e.g. for the treatment
of latent tuberculosis infection) and breastfeeding her child
simultaneously, whether or not the infant (who is not treated
with isoniazid) should be supplemented with pyridoxine remains
a controversial question[12]. In this review article we asked a
structured clinical question: does giving pyridoxine [intervention]
prevent peripheral neuropathy or seizures [outcome] in a breastfed
infant (who is not started on isoniazid) [patient] whose mother is
on isoniazid?
A literature search was performed using Medline, Pubmed,
Cochrane Library, Embase, Google Scholar and Toxnet. The search
period employed was between January 1950 and August 2019. These
databases were searched for synonyms of “isoniazid”, “pyridoxine”,
“lactation”, “breastfeeding”, “breast milk”, “neuropathy”, “seizure”
and “tuberculosis”. Boolean operators AND, OR and NOT were used
within the search fields. Truncation (*) was used in order to find
out variants with one or more letters e.g. breast*. Only articles
in English and publications on human were included. We also
manually scanned reference lists of pertinent studies, reviews,
editorials and websites to retrieve any other relevant articles.
A total of 25 articles were identified initially and they were
independently analysed by two authors (KFN and SB). Nine
out of the 25 journal articles were excluded as there were no
systematic review, experimental studies (randomised and nonrandomised
controlled trial), cohort, case-control, cross-sectional
studies, case series or case reports pertinent the outcome(peripheral neuropathy and/or seizure) and population group of
interest (breastfed infants who were not treated with isoniazid
but their mothers were on isoniazid). The remaining 16 articles
were 7 literature review papers and 9 national and international
guidelines. They only discussed the topic of interest briefly and did
not address our clinical question directly.
Various guidelines were identified and their recommendations
were scrutinised. American Thoracic Society guideline
recommends supplementation of pyridoxine in breastfeeding
infants whose mothers are taking isoniazid but there is no
recommendation from World Health Organization (WHO),
National Institute for Health and Clinical Excellence (NICE),
British National Formulary (BNF) and Specialist Pharmacy
Service[5,13-18]. BNF states that breastfed infants of mothers
on isoniazid should be monitored for toxicity[17]. Electronic
Medicines Compendium (eMC) meanwhile suggests that
administration of pyridoxine may be considered in breastfed
infants when isoniazid is given to nursing mothers[19]. The
Italian Pediatric Tuberculosis Study Group mentioned in their
guidance on neonatal tuberculosis management that newborns
who were breastfed by mothers on isoniazid do not require
pyridoxine supplementation[20]. Inconsistencies among national
and international guidelines recommendations imply that there is
paucity of evidence in this area.
Isoniazid present in breast milk is estimated to be 6.4%-25.0%
of an infant dose[21]. Peak isoniazid level in the breast milk
occurs within 3 hours from maternal oral intake of isoniazid[22-
25]. The half-life of isoniazid in breast milk was between 4 hours
and 6 hours[24,25]. Only 1.2% of isoniazid is transferred to the
exclusively breastfed infant, estimated to be 89.9 mg/kg/day in
comparison with recommended treatment dose of isoniazid which
is 10 mg/kg/day[25-27].
The rate of isoniazid acetylation is determined by genetic
expression of hepatic N-acetyltransferase 2 (NAT2), rendering the
population into fast and slow metabolisers[28]. A physiologic-based
pharmacokinetic study calculated that oral exposure of isoniazid
due to breastfeeding was 0.15 mg/kg/day among infants of fast
metabolising mother (relative infant dose 0.2%) and 0.37 mg/kg/day
in infants of slow metabolising mothers (relative infant dose 0.5%)
when maternal isoniazid dose was 300 mg daily[28]. If the mothers
were on 900 mg of isoniazid every 3 days, the oral exposure
dose in infants of fast metabolising mothers was 0.44 mg/kg/day
(relative infant dose 0.6%) and 1.12 mg/kg/day in infants of slow
metabolising mothers (relative infant dose 0.5%)[28]. Relative
infant dose in this study was defined as the ratio of total oral dose
of the child to the oral dose of the mother[28].
According to most guidelines, mothers who are breastfeeding their
children whilst on isoniazid should be taking pyridoxine[5,13-15,17].
Instead, eMC stated that pyridoxine supplementation for
breastfeeding mothers was optional[19]. The newborn usually has
higher concentration of pyridoxine (22-87 ng/mL) than his or her
mother (13-51 ng/mL) at a ratio of 2:1 and the pyridoxine store in
the infant at birth is dependent on maternal intake[29]. Pyridoxine is
readily excreted into breast milk and the amount found in the breast
milk is directly proportional to maternal intake[30-37]. Pyridoxine
concentration in the breast milk peaks at 3-8 hours post-ingestion[31-
33]. Breastfed infants aged below 3 months are expected to achieve
current recommended dietary allowance of 0.2 mg/day if the
mothers were given 20 mg/day of pyridoxine[34,38].
There is no evidence to support pyridoxine supplementation in
breastfed infant (who is not treated with isoniazid) whose mother
is on isoniazid therapy. The amount of isoniazid transferred to an
infant via breast milk from mother is minimal, well below infant
treatment dose and unlikely to cause clinically significant adverse
effects. Breastfeeding mothers treated with isoniazid should
receive pyridoxine but their infants who are not on isoniazid
therapy do not require pyridoxine supplementation. However,
there should be regular monitoring of these infants for short- and
long-term side effects when breastfeeding mothers are treated with
isoniazid.
Conflict of interest statement
No potential conflicts of interests with respect to authorship and/
or publication of this article.
Acknowledgement
The authors would like to express their special thanks to Coral
Pepper (Clinical Librarian on University Hospitals of Leicester)
who assisted in literature search.
Authors’ contribution
KFN and SB contributed to design, analysis and interpretation,
drafted, critically revised, gave final approval and agrees to be
accountable for all aspects of work ensuring integrity and accuracy.
SB conceptualised the work.