DOI: 10.1111/j.1464-5491.2005.01769.
Effect of pregnancy on the pharmacokinetics of
Pregnancy
Short Communication
report
Oxford,
Diabetic
DME
Blackwell
0742-3071
22 UK and metformin
Medicine
Publishing, Ltd.pharmacokinetics •R. C. E. Hughes et al.
2005
metformin
R. C. E. Hughes*†, S. J. Gardiner*‡, E. J. Begg*‡ and M. Zhang‡
*Department of Clinical Pharmacology,
Christchurch Hospital, Christchurch, †Department
of Medicine, National Women’s Hospital, Auckland
and ‡Department of Medicine, Christchurch School
of Medicine, Christchurch, New Zealand
Accepted 4 May 2005
Introduction
Metformin improves insulin sensitivity and glucose metabolism
in insulin-resistant states [1,2] such as Type 2 diabetes (Type 2
DM) and polycystic ovary syndrome. Increasingly, women are
conceiving on metformin with some continuing it throughout
pregnancy [3–6]. In other cases, metformin is initiated in
pregnancy as an adjunct or alternative to insulin. While data
suggest metformin is not overtly teratogenic [7], little is known
about the effect of pregnancy on metformin pharmacokinetics.
Correspondence to: Dr Ruth Hughes, Department of Clinical Pharmacology,
Christchurch Hospital, Private Bag 4710, Christchurch, New Zealand.
E-mail: ruth.hughes@xtra.co.nz
© 2006 Diabetes UK. Diabetic Medicine, 23, 323–326
Abstract
Aims To determine the effects of pregnancy on metformin pharmacokinetics.
Methods Seven women with Type 2 diabetes mellitus taking metformin throughout
pregnancy were studied on two occasions, once at 28–36 weeks gestation and once
at least 8 weeks postpartum. Serum metformin concentrations were determined
across a dosing interval using high-performance liquid chromatography.
The areas under the serum concentration-time curve from 0 to 4 h post-dose
(AUC0 − 4) and 0 to 8 h post-dose (AUC0 − 8) where possible, were compared in
the pregnant and non-pregnant state.
Results Metformin concentrations were lower in pregnancy in six subjects,
with a mean (95% CI) AUC0 − 4 that was 69% (53.6, 84.8) of the postpartum
value. The AUC0 − 4 of one subject was higher in pregnancy at 142% of the
postpartum value. Overall, the mean (95% CI) AUC0 − 4 during pregnancy for all
seven subjects was 80% (51.3, 107.8) of the postpartum value (P = 0.053,
two-tailed t-test; P = 0.027, one-tailed t-test).
Conclusion These results are consistent with our hypothesis that the clearance
of metformin increases in pregnancy as a result of enhanced renal elimination.
A larger study is required to establish whether metformin dose adjustments are
required in late pregnancy to maintain therapeutic effect.
Diabet. Med. 23, 323–326 (2006)
Keywords metformin, pharmacokinetics, diabetes mellitus, pregnancy
Abbreviations AUC, area under the concentration time curve; GFR, glomerular
filtration rate; Type 2 DM, Type 2 diabetes mellitus
Metformin is a small basic compound (molecular weight
129 Da) that is ionized at physiological pH. Oral availability
(50 – 60%) is inversely related to dose ingested, suggesting
saturable absorption [8]. The extent of absorption is influenced
by food (which decreases metformin concentrations by ∼20%
compared with the fasting state [8,9]) and gastrointestinal transit
time (propantheline slows gut motility and increases concentra-
tions by ∼20% [10]). Metformin has negligible plasma protein
binding and a moderately large apparent volume of distribu-
tion of 63–276 l. It is eliminated unchanged in urine, the renal
clearance (27–31 l / h) exceeding glomerular filtration rate (GFR,
∼5.4 l/ h) indicating active tubular secretion [8] probably via
organic cation transporters such as OCT2 [11,12]. The physi-
ological changes of pregnancy, such as reduced gastrointestinal
323
 324 Pregnancy and metformin pharmacokinetics • R. C. E. Hughes et al.

motility [13,14] and the 40 –50% increase in GFR by the second
trimester [15,16], could significantly alter metformin
pharmacokinetics. This rise in GFR may enhance the renal
clearance of metformin as has been reported for some β-lactam
antibiotics [17].
The aim of this study was to determine the effects of
pregnancy on metformin pharmacokinetics by comparing the
area under the concentration time curves (AUC) for metformin
in pregnancy and the non-pregnant state. To our knowledge
this is the first study to describe the effect of pregnancy on the
pharmacokinetics of metformin.
Patients and methods
Subjects were recruited from the antenatal clinic at National
Womens’ Hospital, a tertiary referral centre serving the
multiethnic community of central Auckland, New Zealand.
Women eligible for inclusion had: Type 2 DM treated with
metformin prenatally; decided to continue metformin dur-
ing pregnancy independently of this study; a pre-pregnancy or
early pregnancy creatinine clearance ≥ 90 ml /min (using the
Cockcroft and Gault equation); and were not taking any
medicines known to interact pharmacokinetically with
metformin. Approval was obtained from the Auckland ethics
committee and informed written consent was obtained from
each participant.
Subjects were studied in an inpatient ward at the National
Womens’ Hospital or in their own homes. Each subject was
studied on two occasions, once at 28–36 weeks gestation (the
time of maximal pregnancy-related increase in GFR), and once
in the postpartum period at least 8 weeks after delivery (when
GFR is expected to have returned to baseline). On the study
days, the subjects ingested their usual dose of metformin with
meals (low glycaemic, low fat) prepared by the hospital or
homemade. The exact time of metformin administration was
recorded and serial blood samples (5 ml) were collected into
plain tubes via an indwelling venous cannula. Blood samples
were taken prior to the first daily dose (at 0 h) and then at 1, 2,
4, 8 and 12 h post-dose if these time points occurred within the
dose interval. Blood samples were centrifuged immediately at
2000 g for 10 min and the serum stored at −20°C until analysis.
The samples were analysed for metformin in triplicate using
high-performance liquid chromatography [18] and the mean
concentration used in subsequent analyses. The standard curve
was linear (r 2 > 0.99) over the range 20 – 4000 µg / l and intra-
and inter-day coefficients of variation were less than 9% at the
concentrations of 62.5, 250, 1000 and 4000 µg / l. The limit of
detection (signal/noise 3 : 1) was 20 µg / l.
The AUC from 0 to 4 h (AUC0 − 4) and 0 to 8 h (AUC0 − 8) were
determined using GraphPad Prism version 4.00 for Windows
[GraphPad Software, San Diego, CA, USA (www.graphpad.
com)]. Student’s t-test (paired) was performed for comparison
of the AUC0 – 4 in pregnancy and postpartum. Use of the one-
tailed test was justified on the basis of our hypothesis that the
clearance of metformin would increase in pregnancy as a result
of increased GFR. The level of significance was set at P < 0.05.
Results
Nine subjects were enrolled during the time period available to
the project. Two subjects were subsequently excluded, subject
1 had postpartum metformin intolerance and subject 6 became
pregnant a second time before completion of the study. The
remaining seven subjects ranged in age from 31 to 41 years and
were from diverse ethnic backgrounds (four Polynesian, one
Caucasian, one Asian and one Indian). Subjects were studied at
a median (range) of 34 weeks gestation (29 – 36) and at 10 weeks
postpartum (8 –11). The median (range) pregnancy and
postpartum weights were 108.4 kg (88–118.5) and 99 kg (74 –
104.6), respectively. No formal measure of GFR was under-
taken and the Cockcroft and Gault equation has not been
validated in pregnancy. However, mean (SEM) serum creatinine
concentrations were lower in pregnancy 0.05 (0.003) mmol/ l
vs. 0.07 (0.002) mmol / l postpartum (P = 0.0015, two-tailed
t-test). Concurrent drug therapy over the study period included
insulin, sulphonylureas, aspirin, vitamins/minerals, fluoxetine,
oxprenolol and depot progesterone.
Valid data points for a full AUC0−8 could not be collected in
all subjects therefore, the AUC0−4 was calculated for seven
subjects and AUC0−8 for the four subjects with sufficient data
points (Table 1). Concentration-time profiles for two subjects

Table 1 Dosing data and AUC for metformin during pregnancy (28–36 weeks gestation) and postpartum

Creatinine Ratio of AUC0 − 4

clearance† AUC0 − 4 (mg / l)h AUC0 − 8 (mg / l)h to AUC0 − 8

Subject Metformin* dose Non-pregnant Pregnant Postpartum Pregnant Postpartum Pregnant Postpartum

2 1000 mg twice daily 90 3.962 2.799 6.524 4.579 1.65 1.70

3 850 mg twice daily 93 3.502 5.838 — — — —
4 850 mg thrice daily 92 3.654 5.247 — — — —
5 500 mg twice daily 100 3.025 3.865 4.821 6.295 1.59 1.63
7 500 mg twice daily 94 3.595 3.833 — — — —
8 1000 mg twice daily 100 2.978 5.585 5.028 10.18 1.69 2.02
9 850 mg twice daily 114 3.162 5.261 5.064 8.469 1.60 1.61

*Metformin (Metomin, Pacific Pharmaceuticals, Christchurch, New Zealand) was a plain release formulation.
†Calculated in ml/min using the Cockcroft and Gault equation (note this may underestimate renal function in the Polynesian volunteers (subjects 2–4
and 7).

© 2006 Diabetes UK. Diabetic Medicine, 23, 323–326


Figure 1 Steady-state serum concentration-time curves for subjects 4 and 5 who were on a metformin dosing regimen of 850 mg thrice daily and 500 mg
twice daily, respectively. Closed squares represent the pregnant state and closed triangles represent the non-pregnant state.
are shown in Fig. 1. With the exception of subject 2, all sub-
jects had a smaller AUC0 − 4 during pregnancy compared with
the postpartum AUC0 − 4. In these six subjects, the AUC0 − 4 was
a mean of 69% (95% CI 53.6, 84.8) of the postpartum value,
while that of subject 2 was 142%. Overall, including data from
all seven subjects, the AUC0−4 during pregnancy was a mean
(95% CI) of 80% (51.3, 107.8) of the postpartum value
(P = 0.053, two-tailed t-test; P = 0.027, one-tailed t-test) (the
Mann–Whitney U-test gave identical results to the paired t-test).
The AUC0−8 was a mean (95% CI) of 1.63-fold (1.56, 1.70) and
1.74-fold (1.54, 1.84) higher than the corresponding AUC0−4
in pregnancy and postpartum, respectively. There was excellent
correlation (r 2 = 0.97) between the AUC0 − 4 and AUC0 − 8 in sub-
jects where both were calculated suggesting that the AUC0 − 4 is
a valid index of relative interdose metformin exposure.
Observed peak serum concentrations were at 1–2 h post-
dose in all subjects during pregnancy and postpartum. Peak
serum concentrations in pregnancy were 81.0% (mean; 95%
CI 47.3, 114.8%) of the postpartum value (P = 0.077, two-
tailed) (data not shown).
Discussion
In all but subject 2, a marked decrease in the AUC 0 − 4 for
metformin was seen in pregnancy, the magnitude of which was
in keeping with our hypothesis that the renal clearance of met-
formin would increase by 40–50% secondary to the pregnancy
associated increase in GFR. This is supported by the observed
decrease in serum creatinine concentrations in all subjects.
Enhanced active tubular secretion of metformin during preg-
nancy may also be a contributing factor, although this has
© 2006 Diabetes UK. Diabetic Medicine, 23, 323–326
Short report 325
undergone little research. Variable long-term compliance with
metformin may be responsible for the incongruent results of
subject 2 who, after completing the study, was admitted with
collapse and poor glycaemic control. It is possible therefore
that the serum concentrations of metformin were not at a
steady state for the second part of the study and thus a rise in
the AUC was not seen as predicted. Alternatively, the disparate
result may be secondary to factors such as genetic variation in
OCT2 and in the renal clearance of metformin [19], or variable
food ingestion on study days (although food composition may
not significantly influence absorption [20]).
The observed decrease in the AUC of metformin in preg-
nancy may have clinical implications with respect to dosing.
The renal clearance of metformin is expected to peak at the
time of maximal pregnancy induced insulin resistance. From
a pharmacokinetic point of view, this may necessitate a ≥ 20%
increase in metformin dose to maintain a given therapeutic
effect which may ultimately delay the need to commence or
increase insulin in some patients. A larger study incorporating
measures to assess patient compliance is required to confirm
our findings before specific recommendations can be made on
metformin dosing in pregnancy.
Competing interests
None declared.
References
1 Kirpichnikov D, McFarlane SI, Sowers JR. Metformin: an update.
Ann Intern Med 2002; 137: 25–33.
 326 Pregnancy and metformin pharmacokinetics • R. C. E. Hughes et al.
2 Hundal RS, Inzucchi SE. Metformin: new understandings, new uses.
Drugs 2003; 63: 1879–1894.
3 Glueck CJ, Goldenberg N, Pranikoff J, Loftspring M, Sieve L,
Wang P. Height, weight, and motor-social development during the
first 18 months of life in 126 infants born to 109 mothers with poly-
cystic ovarian syndrome who conceived on and continued metformin
through pregnancy. Hum Reprod 2004; 19: 1323–1330.
4 Glueck CJ, Wang P, Kobayashi S, Phillips H, Sieve-Smith L.
Metformin therapy throughout pregnancy reduces the development
of gestational diabetes in women with polycystic ovary syndrome.
Fertil Steril 2002; 77: 520–525.
5 Glueck CJ, Goldenberg N, Wang P, Loftspring M, Sherman A.
Metformin during pregnancy reduces insulin, insulin resistance,
insulin secretion, weight, testosterone and development of gestational
diabetes: Prospective longitudinal assessment of women with
polycystic ovary syndrome from preconception throughout pregnancy.
Hum Reprod 2004; 19: 510–521.
6 Coetzee EJ, Jackson WP. Metformin in management of pregnant
insulin-independent diabetics. Diabetologia 1979; 16: 241–245.
7 Gutzin SJ, Kozer E, Magee LA, Feig DS, Koren G. The safety of
oral hypoglycemic agents in the first trimester of pregnancy: a meta-
analysis. Can J Clin Pharmacol 2003; 10: 179–183.
8 Scheen AJ. Clinical pharmacokinetics of metformin. Clin Pharma-
cokinet 1996; 30: 359–371.
9 Sambol NC, Brookes LG, Chiang J, Goodman AM, Lin ET, Liu CY
et al. Food intake and dosage level, but not tablet vs solution dosage
form, affect the absorption of metformin hcl in man. Br J Clin
Pharmacol 1996; 42: 510–512.
10 Marathe PH, Wen Y, Norton J, Greene DS, Barbhaiya RH, Wilding
IR. Effect of altered gastric emptying and gastrointestinal motility on
metformin absorption. Br J Clin Pharmacol 2000; 50: 325–332.
11 Dressler MJ, Xiao G, Leabman MK, Gray AT, Giacomini KM.
Interactions of n-tetraalkylammonium compounds and biguanides
with a human renal organic cation transporter (hoct2). Pharm Res
2002; 19: 1244–1247.
12 Wang DS, Jonker JW, Kato Y, Kusuhara H, Schinkel AH, Sugiyama Y.
Involvement of organic cation transporter 1 in hepatic and intestinal
distribution of metformin. J Pharmacol Exp Ther 2002; 302: 510–
515.
13 Parry E, Shields R, Turnbull AC. Transit time in the small intestine in
pregnancy. J Obstet Gynaecol Br Commonw 1970; 77: 900–901.
14 Davison JS, Davison MC, Hay DM. Gastric emptying time in late
pregnancy and labour. J Obstet Gynaecol Br Commonw 1970; 77:
37–41.
15 Dunlop W, Davison JM. Renal haemodynamics and tubular function
in human pregnancy. Baillieres Clin Obstet Gynaecol 1987; 1: 769–
787.
16 Davison JM, Dunlop W. Renal hemodynamics and tubular function
in normal human pregnancy. Kidney Int 1980; 18: 152–161.
17 Heikkila A, Erkkola R. Review of beta-lactam antibiotics in preg-
nancy. The need for adjustment of dosage schedules. Clin Pharma-
cokinet 1994; 27: 49–62.
18 Zhang M, Moore GA, Lever M, Gardiner SJ, Kirkpatrick CMJ, Begg
EJ. A rapid and simple hplc assay for the determination of metformin
in human plasma and breast milk. J Chromatogr B Analyt Technol
Biomed Life Sci 2002; 766: 175–179.
19 Leabman MK, Giacomini KM. Estimating the contribution of genes
and environment to variation in renal drug clearance. Pharmacoge-
netics 2003; 13: 581–584.
20 Saffar F, Aiache JM, Andre P. Influence of food on the disposition of
the antidiabetic drug metformin in diabetic patients at steady-state.
Methods Find Exp Clin Pharmacol 1995; 17: 483–487.
© 2006 Diabetes UK. Diabetic Medicine, 23, 323–326