REVIEWS

Reproductive health and pregnancy

in women with chronic kidney disease
Kate S. Wiles1,2, Catherine Nelson-Piercy2,3 and Kate Bramham4
Abstract | Chronic kidney disease (CKD) is associated with reduced fertility and an increased risk
of adverse pregnancy outcomes. Rates of pre-eclampsia, fetal growth restriction and preterm
delivery increase incrementally with the severity of CKD and proteinuria. Pre-pregnancy
counselling can facilitate informed decision-making. Safe and effective contraception is required
for women who wish to delay or avoid pregnancy. Pregnancy planning for women who wish to
conceive involves appropriate substitution of known teratogens — including mycophenolate
mofetil, angiotensin blockers and cyclophosphamide — and can aid optimization of disease
control. However, pregnancy, which can occur in women with any stage of CKD, can exacerbate
comorbidities such as anaemia, vitamin D deficiency and hypertension. Increased haemodialysis
provision is associated with improved pregnancy outcomes for women on dialysis. Diagnosis of
pre-eclampsia in women with CKD is complicated in patients with pre-existing hypertension and
proteinuria but can be improved by the use of vasoactive biomarkers as well as placental and fetal
Doppler ultrasound. Pregnancy data for newer drugs used in CKD are limited as pregnancy and
CKD are common exclusion criteria for drug and intervention trials. Although prospective data
may be available for older drugs, the use of most drugs in pregnancy is based on retrospective
data and expert consensus.

1
Women’s Health Academic
Centre, St Thomas’ Hospital,
Westminster Bridge Road,
London SE1 7EH, UK.
2
Guy’s and St Thomas’ NHS
Foundation Trust,
Westminster Bridge Road,
London SE1 7EH, UK.
3
Imperial Healthcare NHS
Trust, Du Cane Road,
London W12 0HS, UK.
4
King’s College Hospital NHS
Foundation Trust, Denmark
Hill, London SE5 9RS, UK.
Correspondence to K.W.  ­pre-eclampsia in the context of CKD.
and C.N.-P.
kate.wiles@kcl.ac.uk;
Catherine.Nelson-Piercy@
gstt.nhs.uk
doi:10.1038/nrneph.2017.187
Published online 22 Jan 2018
Chronic kidney disease (CKD) is estimated to affect 3%
of pregnant women in high-income countries1, and the
prevalence of CKD in this population is predicted to rise
owing to increasing maternal age and obesity, highlight-
ing the importance of reproductive health in nephrol-
ogy. Although CKD is not a barrier to reproduction in
most patients, the risk of adverse pregnancy outcomes
— including pre-eclampsia, fetal growth restriction,
preterm delivery and accelerated loss of maternal renal
function — is increased in these individuals.
In this Review, we outline fertility issues and contra-
ceptive options for women with CKD, which are impor-
tant components of women’s health that are frequently
neglected in the CKD literature. We also discuss com-
mon comorbidities (anaemia, vitamin D deficiency
and hypertension), maternal and neonatal complica-
tions and therapeutic management of pregnancy in
CKD, including drug safety and the use of dialysis.
Last, we summarize advances in the identification of
Of note, the data on which the information pre-
sented in this Review is based are limited by several
factors, including changes in CKD definition over time,
small study cohorts, disease heterogeneity and insuf-
ficient data regarding pre-pregnancy renal function.
Furthermore, a complex interplay with overlapping
clinical phenotypes exists between CKD and pre-­
eclampsia, with no standardized diagnostic criteria for
pre-eclampsia super­imposed on CKD. Variation in rates
of adverse outcomes is due in part to threshold differ-
ences between studies in the diagnosis and management
of superimposed pre-­eclampsia, including iatrogenic
preterm delivery. Of note, reproductive health issues in
relation to renal transplantation are not covered in this
Review and are addressed elsewhere2–5.
Fertility
Pregnancy rates for women with CKD
Fertility of women with CKD is difficult to precisely
evaluate on a population level as pregnancy rates for
women with CKD can be measured only when both
CKD prevalence and the total number of pregnancies
are known. CKD prevalence rates remain an approx-
imation at best for most women of childbearing age
because routine screening for CKD is not under-
taken. Pregnancy rates can, however, be calculated
within specific CKD populations. The UK Obstetric
Surveillance System6 identified all pregnancies in renal
transplant recipients (January 2007 to January 2010)2
and in women on dialysis (February 2012 to January

NATURE REVIEWS | NEPHROLOGY ADVANCE ONLINE PUBLICATION | 1

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

Key points indicate that amenorrhoea is present in 37–59% of

• Although chronic kidney disease (CKD) adversely affects fertility, pregnancies can
occur at all stages of CKD severity
• Safe and effective contraception should be made available for all women with CKD
who do not wish to conceive and those who take teratogenic medications
• CKD increases the risk of adverse pregnancy outcomes, including pre-eclampsia, fetal
growth restriction, preterm delivery and post-partum loss of maternal renal function
• Pre-pregnancy hypertension and proteinuria in CKD complicate the diagnosis of
superimposed pre-eclampsia, which could be improved through vasoactive
biomarkers as well as placental and fetal Doppler ultrasound
• Although data on the use of many drugs in pregnancy are limited, low-dose aspirin,
low‑molecular-weight heparin, labetalol, nifedipine, prednisolone,
hydroxychloroquine, azathioprine, ciclosporin and tacrolimus are considered safe
during pregnancy and breastfeeding
2014) in the UK. Combining these data with national
renal registry data indicated pregnancy rates of 7.6 per
1,000 patients per year for women with a functioning
renal transplant and 1.4 per 1,000 patients per year
for women on dialysis compared with national con-
ception rates of 79.1–79.5 per 1,000 women per year 7.
Comparable data from Italy reveal a live birth rate of
5.5–8.3 per 1,000 women with a functioning transplant
and 0.7–1.1 per 1,000 women on dialysis (compared
with 72.5 per 1,000 women nationally)8. Thus, preg-
nancy or birth rates among renal transplant recipients
and dialysis patients compared with rates in the general
population are 1:10 and 1:100, respectively. Whether
these data represent a true assessment of fertility or
indicate the impact of CKD on the decision to conceive
remains unknown.
Sexual dysfunction in CKD
Although only limited prevalence data exist, sexual
dysfunction — an umbrella term used to describe any
problem experienced during sexual activity, including
reduced libido, difficulty with arousal, dyspareunia and
anorgasmia — was identified by one study in 81% of 21
female predialysis patients9. In addition, sexual absti-
nence or decreased frequency of intercourse as well as
a negative body image have been described in women
with CKD and are apparent before the need for dialy­
sis10. The aetiology of sexual dysfunction in CKD is
likely to be multifactorial, including both biological and
psychological factors. Although depression may be both
a cause and consequence of sexual dysfunction10–12, the
complex interplay between CKD, sexual dysfunction,
fertility and the perceived risks of pregnancy remains
poorly understood.
Mechanistic effects of CKD on fertility
In women with CKD, oligomenorrhoea progresses
to amenorrhoea as glomerular filtration rate (GFR)
declines. However, the threshold GFR at which this
progression becomes clinically significant for repro-
ductive health is unknown owing to a lack of data. In
a cohort of 76 women on dialysis aged ≤55 years, 42%
Amenorrhoea reported a regular menstrual cycle compared with
The absence of menstruation. 75% before the start of dialysis13. Small cohort studies
women on dialysis13,14. Although the effect of CKD on
menstruation and fertility is hypothesized to be pro-
portional to the degree of renal dysfunction, the effect
of mild renal dysfunction remains unclear. However,
the effects of all stages of CKD on the hypothalamic–­
pituitary–ovarian axis (FIG. 1) are thought to be revers-
ible, and normalization of levels of gonadotropins
(luteinizing hormone (LH) and follicle-stimulating
hormone (FSH)), prolactin and oestrogen is seen
­following renal transplantation15.
LH and FSH are produced in the anterior pituitary
gland and act synergistically to regulate gonad function,
including sex steroid production and gameto­genesis
(FIG. 1c). A surge of LH levels at the midpoint of the
menstrual cycle, which is caused by oestrogen-mediated
positive feedback to the hypothalamus and pituitary
gland, is responsible for ovulation. Progesterone —
released from the corpus luteum and later the placenta
if pregnancy occurs — maintains the endometrium and
prevents menstruation. Prolactin, which is secreted
by the anterior pituitary gland during pregnancy and
lactation, stimulates breast development and milk
­production and inhibits ovulation during lactation.
Although LH levels are overall increased in women
receiving dialysis compared with levels in age-matched
controls with regular menstrual cycles16, LH levels do
not fluctuate in women on dialysis as they do in healthy
individuals17 (FIG. 1a,b). Dysfunctional oestrogen feed-
back to the hypothalamus and pituitary in patients with
renal dysfunction results in an absence of pre-ovulatory
surges in both oestrogen and LH, which leads to ovula-
tion failure14,16. However, the exact stage and/or degree
of CKD disease severity resulting in dysfunctional LH
release — and, consequently, suppression of ovulation
— is currently unknown (FIG. 1d). In addition, renal clear-
ance of prolactin in women on dialysis is reduced17,18.
Reduced prolactin secretion, in conjunction with
autono­mous secretion that appears resistant to stimu-
latory factors (for example, hypoglycaemia or arginine)
and suppressive factors (dopamine)16, also contributes to
the ­suppression of ovulation.
Dialysis provision and fertility
Daily haemodialysis is associated with increased patient
survival19, reduced cardiovascular-related hospitaliza-
tion19, improved blood pressure20 and better quality of
life21 compared with the standard dialysis provision
(three times per week). The effects of dialysis on fertility
are less clear. One study of seven women showed that
increasing dialysis provision from 16.5 to 28.5 hours
per week was associated with return of menses in two
out of three previously amenorrhoeic women <40 years
of age22, and increasing dialysis provision from 12 to
36 hours per week (intensified haemodialysis) has
been used as a tool to maximize fertility in women
with end-stage renal disease (ESRD), resulting in con-
ception rates of 15.6%23,24. However, functional assess-
ment of the hypothalamic–pituitary–ovarian axis in
women r­ eceiving intensified haemodialysis has not
been conducted22.

2 | ADVANCE ONLINE PUBLICATION www.nature.com/nrneph

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

a Physiological menstrual cycle b Menstrual cycle in end-stage CKD

Menstrual Proliferative Menstrual Proliferative

phase phase Midcycle Secretory phase phase phase Midcycle Secretory phase

Ovulation FSH FSH

LH LH
Hormone concentration

Hormone concentration
OE OE
PROG PROG

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28
Day of cycle Day of cycle

c Ovulation and lactation d CKD

Brain
Hypothalamus GnRH
GnRH

• Breast development Anterior

• Milk production pituitary FSH
gland

FSH
↑ Prolactin ↑ LH
Prolactin LH
↓ Renal clearance

Lack of LH
midcycle surge

Uterus Fallopian Anovulation

tube Ovulation

↓ Oestrogen
↑ Oestrogen
Ovary Cyclophosphamide

Figure 1 | Endocrine effects of chronic kidney disease on the hypothalamic–pituitary–ovarian axis. a | During the
Naturefeedback,
midcycle (late follicular) phase of the menstrual cycle, high levels of oestrogen (OE) confer positive Reviews |sensitizing
Nephrology
the pituitary to the effects of gonadotropin-releasing hormone (GnRH) from the hypothalamus. As a result, a surge of
luteinizing hormone (LH) in midcycle occurs, which stimulates ovulation. Following ovulation, progesterone (PROG) is
secreted from the corpus luteum, which acts to prepare the endometrium for implantation. If implantation does not occur,
PROG levels fall and menstruation follows. b | Hypothesized hormone profile during the menstrual cycle in chronic kidney
disease (CKD). Serum LH levels are increased in CKD, but the absence of an LH surge in the late follicular phase of the
menstrual cycle means that ovulation does not occur. Without ovulation and the development of a corpus luteum, PROG
levels do not rise. OE levels are low throughout the cycle. c | The anterior pituitary gland secretes follicle-stimulating
hormone (FSH), LH and prolactin under hypothalamic control. FSH and LH act synergistically to regulate gonad function,
including sex steroid production and gametogenesis in the ovary. At the midpoint of the menstrual cycle, positive OE
feedback leads to a surge in LH, which triggers ovulation. Prolactin production is increased only during pregnancy and
lactation, when it stimulates breast development and lactation while inhibiting ovulation. d | Low OE levels in CKD confer
negative feedback to the hypothalamus–pituitary axis. The absence of an LH surge leads to anovulation. Impaired renal
clearance of prolactin causes inhibition of GnRH secretion from the hypothalamus, also suppressing ovulation.
Cyclophosphamide is gonadotoxic, causing age and dose-dependent premature ovarian failure. The level of CKD at
which these changes occur remains unknown.

NATURE REVIEWS | NEPHROLOGY ADVANCE ONLINE PUBLICATION | 3

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

Ovarian stimulation
The use of drugs to stimulate
oocyte development in the
ovary before retrieval for
artificial reproductive
techniques.
Natural-cycle oocyte
retrieval
Oocyte retrieval from the ovary
following a normal menstrual
cycle, without the use of
stimulatory drugs.
‘Mini pill’
An oral contraceptive pill that
contains a synthetic
progestogen (no oestrogen).
Intrauterine device
(IUD). A small birth control
device that is inserted into the
uterus to prevent pregnancy.
May contain a slow-releasing
progestogen (for example,
Mirena) or offer contraception
without hormonal release
(copper coil).
Subdermal implant
A small device inserted under
the skin. The contraceptive
implant delivers an effective
dose of a synthetic
progestogen, providing
long-acting, reversible
contraception.
Progestogen
A synthetic form of
progesterone.
Table 1 | Contraceptive options for women with chronic kidney disease
Contraceptive
Safe and effective methods
Progesterone-only pill
Progesterone intrauterine device
Progesterone-only subdermal implant
Female sterilization
Unsafe and/or ineffective methods
Oestrogen-containing methods
(pill, patch or ring)
Male condom
Female condom
No method
*% of couples experiencing an unplanned pregnancy in the first year of use. Data from REF. 36.
Cyclophosphamide and fertility
The alkylating agent cyclophosphamide has a therapeu-
tic role in the immunosuppression of rapidly progres-
sive glomerular disease. However, cyclophosphamide
has several adverse effects, including dose-dependent
gonadotoxicity in both men25 and women26, causing
oligo­spermia, amenorrhoea and ovarian failure. Toxicity
is more likely to be irreversible in women, given that
the oocyte pool is determined in fetal life and cannot be
regenerated. In addition, older age at initiation of treat-
ment increases the risk of cyclophosphamide-induced
ovarian failure26,27.
Hence, fertility preservation should be considered
for all women of childbearing age receiving cyclophos-
phamide. Cryopreservation of oocytes or embryos can
be undertaken before cyclophosphamide treatment
is initiated; however, concerns exist regarding ovarian
stimulation in patients with active systemic lupus ery-
thematosus (SLE) owing to the risks of disease exacer-
bation and thrombosis, which are hypothesized to be
due to increased serum oestrogen concentrations28, and
treatment delay. Although natural-cycle oocyte retrieval29
and cryopreservation of ovarian tissue negate the need
for ovarian stimulation, long-term fertility outcomes
remain unclear in women with CKD. Alternatively,
LH‑releasing hormone (LHRH) analogues can be used
to inhibit ovarian function for the duration of cyclo-
phosphamide treatment in order to try to preserve
future fertility. Hypothetical mechanisms by which
LHRH analogues act include the protective inhibition
of the hypothalamic–pituitary–ovarian axis and/or a
reduction in ovarian blood flow and therefore expo-
sure to circulating cyclophosphamide. A trial of 20
women receiving cyclophosphamide for lupus nephri-
tis showed a reduction in the incidence of premature
ovarian failure in patients receiving an LHRH analogue
compared with that of age-matched and cyclophos-
phamide-dose-matched controls (5% versus 30%)30.
Findings from randomized controlled trials (RCTs)
with >250 participants 31,32 and a meta-analysis of
12 RCTs comprising >1,200 patients treated for breast
Perfect-use Typical-use
failure rate (%)* failure rate (%)*
0.3 9 sistent inhibition of ovulation (in 102 out of 103 treated
0.2 0.2
0.05 0.05
0.5 0.5
0.3 9 (Bayer) and the subdermal implant Nexplanon (Merck,
2 18
5 21
85 85
cancer 33 show that LHRH analogues are safe and effec-
tive in the prevention of cyclophosphamide-induced
ovarian failure.
Contraception
Pregnancy planning for women with CKD is impor-
tant because of the increased risk of adverse pregnancy
outcomes and the common use of teratogenic medica-
tions. Although CKD is associated with reduced fer-
tility (see above), unplanned pregnancies — which are
associated with an increased risk of obstetric compli-
cations, even in the absence of comorbidities34 — can
occur at all CKD stages and in renal transplant recip-
ients. For example, around 33–50% of all pregnancies
among renal transplant recipients are unplanned2,35.
Hence, contraceptive counselling should be provided
for all women on dialysis who do not want to have chil-
dren, particularly those receiving intensified dialysis,
as this method is associated with an increased concep-
tion rate compared with standard dialysis regimens23.
Unfortunately, evidence suggests that few nephrologists
discuss fertility issues and contraception with their
patients. Although no published contemporary data
exist, questionnaire data from a cohort of 76 women
with CKD showed that despite 50% being sexually
active, only 36% used contraception and only 13% had
discussed reproductive health issues with their neph-
rologist 13. Therefore, contraceptive counselling and the
provision of safe and effective contraception remain
inadequate for many women with CKD. The risks and
acceptability of different contraceptive methods must be
balanced against the risks of an unplanned pregnancy.
In addition, assessment of contraceptive effectiveness
should be based on ‘typical use’ rather than presuming
‘perfect use’, as discrepancies exist in the failure rate of
some contraceptive methods36 (TABLE 1).
Safe and effective methods in chronic kidney
disease
Several contraceptive methods exist, but not all are safe
for women with CKD. Progesterone-only contraceptives
— for example, in the form of the ‘mini pill’, an intrauterine
device (IUD) or a subdermal implant — can be used safely
in women with CKD, including patients on dialysis and
transplant recipients.
The ability of progesterone-only therapy to inhibit
ovulation varies, depending on the synthetic progestogen
used. One study showed that desogestrel provides con-
subjects) and that this inhibition is maintained even
after 12‑hour delays before re-dosing 37. Therefore, this
preparation likely confers improved typical-use effec-
tiveness over other progestogens that require re‑dosing
within a 3‑hour window each day. The IUD Mirena
Sharp & Dohme) — which are progesterone-only, long-­
acting, reversible contraceptives — provide effective
contraception for 5 and 3 years, respectively. Moreover,
these contraceptives do not rely on daily compliance
and have a typical-use failure rate comparable to that of
­sterilization36 (TABLE 1).

4 | ADVANCE ONLINE PUBLICATION www.nature.com/nrneph

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

Hypothalamus Systemic changes methods are therefore likely to be contraindicated

• ↓ Sensitivity to RAAS for many women with CKD, particularly given the
• ↑ Water retention ­availability of safer, more effective alternatives.
Renal haemodynamics
• ↓ Plasma osmolality
• 50–85% ↑ RBF Condoms are effective against the transmission of
• 50% ↑ GFR HIV and sexually transmitted diseases. However, these
• Afferent and efferent
arteriole vasodilatation barrier methods are unforgiving of imperfect use —
approximately 1 in 5 women conceive within a year 36
↓ Threshold for
? ? (TABLE 1) — and should therefore not be considered as
vasopressin release
Relaxin Nitric oxide an effective long-term contraceptive option.
Progesterone
Endocrine function Contraceptive failure
• ↑ Erythropoietin
• ↑ RAAS activity The most widely used emergency contraceptives (for
Corpus example, levonorgestrel) do not contain oestrogen
luteum and can therefore be used safely in patients with CKD
to prevent pregnancy within 72 hours of unprotected
Tubular function intercourse. For the medical termination of pregnancy,
• ↑ Glycosuria Structural changes a combination of mifepristone and misoprostol is most
• ↑ Calcium excretion • Pelvicalyceal dilatation
• ↑ Glycosaminoglycans • Hydronephrosis commonly used. However, only limited data are available
• ↓ Reabsorption of • ↑ GBM pore size contributes on the use of both drugs in the context of renal impair-
uric acid to physiological proteinuria ment. Whereas mifepristone is largely metabolized in
the liver, misoprostol is predominantly renally excreted.
Figure 2 | Physiological changes in the renal system in pregnancy. The kidney The peak concentration, half-life and bioavailability of
undergoes structural, haemodynamic, tubular and endocrine changes during pregnancy.
Natureto
Reviews | Nephrology misoprostol are all potentially increased in patients with
In addition, systemic effects in pregnancy are mediated by changes the renin–
angiotensin–aldosterone system (RAAS) and vasopressin. Relaxin, progesterone and nitric
CKD. The clinical relevance of potentially reduced clear-
oxide are hypothesized to play a part in mediating the haemodynamic changes in the ance of misoprostol is unknown, and no dose reduction
kidney. GBM, glomerular basement membrane; GFR, glomerular filtration rate; RBF, renal is recommended, although increased clinical surveillance
blood flow. Adapted by permission from BMJ Publishing Group Limited from REF. 185. should be considered for women with a p ­ re-pregnancy
estimated GFR (eGFR) of <30 ml/min/1.73 m2.

‘Combined pill’
Contraceptive pill containing
a synthetic oestrogen and
progestogen.
Transdermal patch
Contraceptive patch that
delivers synthetic oestrogen
and progestogen through
the skin.
Vaginal ring
A soft plastic ring worn inside
the vagina that provides
contraception via the release
of synthetic oestrogen and
progestogen.
Corpus luteum
The remnants of the ovarian
follicle after ovulation.
Theoretical concerns exist regarding the use of IUDs
in immunosuppressed women, as immunosuppres-
sants might reduce their effectiveness owing to inhib­
ition of the uterine inflammation that forms part of
their contraceptive mechanism. However, the uterine
milieu is predominantly populated by macrophages,
whereas immunosuppression used in the management
of immunological renal disease and following transplan-
tation acts predominantly via lymphocyte inhibition (see
below), suggesting that different pathways are involved.
No evidence of an excess of IUD failures following trans-
plantation has been reported38,39. Furthermore, concern
regarding pelvic infection in the context of immunosup-
pression also seems to be unfounded as no correlation
was found between infective complications and numbers
of CD4+ T cells in women with HIV-mediated immuno­
suppression40. Similarly, a retrospective study of 11
women with renal transplants and a total of 484 months
of progesterone–IUD use reported no cases of pelvic
infection or unplanned pregnancy 41.
Other contraceptive methods
Oestrogen-containing contraceptives include the
oestrogen-­containing ‘combined pill’, the transdermal
patch and the vaginal ring. All of these methods con-
fer a risk of hypertension, venous thromboembolism
(VTE), arterial thrombosis and cervical cancer 42. These
risks are particularly relevant for women with CKD who
are at increased risk of chronic hypertension, coagulo­
pathy (due to antiphospholipid antibodies or nephrotic
syndrome), vascular diseases, and/or neoplasia (in the
context of immunosuppression). Oestrogen-containing
Renal physiology during pregnancy
The physiological changes that occur during preg-
nancy include haemodynamic, tubular, endocrine and
­structural alterations in the kidney (FIG. 2).
Renal blood flow
During pregnancy, renal blood flow (RBF) increases by
50–85% with a concurrent increment in GFR43. This
physiological hyperfiltration — mediated by vasodilata-
tion of both afferent and efferent arterioles — occurs in
the absence of glomerular hypertension44 and is there-
fore different from pathological hyperfiltration associ-
ated with glomerular injury and CKD. Pregnancy and
multiparity are therefore not associated with a decline in
maternal renal function in the absence of CKD. Dextran
clearance modelling suggests that a physio­logical
increase in glomerular membrane pore size occurs with
pregnancy, which underlies the finding of gestational
proteinuria in the absence of a change in hydrostatic
force45,46. Amino acid loading during pregnancy can fur-
ther increase RBF and GFR, which suggests that extra
filtration capacity — in addition to that already seen in
adaptation to pregnancy — exists46.
As the above described renal physiological changes are
measurable from as early as 8 weeks of gestation, they are
not thought to be mediated by the fetoplacental unit,
which is still underdeveloped at this stage. Instead, the
observed rise in GFR in both the luteal phase of the men-
strual cycle and pregnancy may be mediated by relaxin,
which is released by the corpus luteum47. In rats, both ova-
riectomy and administration of neutralizing relaxin anti-
bodies impaired the renal adaptation to pregnancy48, with

NATURE REVIEWS | NEPHROLOGY ADVANCE ONLINE PUBLICATION | 5

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

evidence from other animal studies that relaxin mediates
its effects via matrix metalloproteinase 2 and endothelin
receptor type B49. However, other unknown factors —
possibly progesterone and nitric oxide50 — also have a
role, as evidenced by the physiological renal adaptation
to assisted-conception pregnancies in which the corpus
luteum is absent and relaxin is not detectable51.
Measuring GFR in pregnancy
As a result of the dynamic changes in RBF and filtra-
tion fraction that lead to a sustained, augmented GFR
until term52, the serum creatinine concentration falls
with physiological renal adaptation to pregnancy.
Modification of Diet in Renal Disease (MDRD)53 and
CKD-epidemiology (CKD-EPI)54 formulae tend to
underestimate GFR and cannot be used in pregnancy.
Cystatin C provides an alternative surrogate measure
of GFR as it is freely filtered at the glomerulus and, in
pregnancy, synthesis is not thought to be affected by
the utero-placental unit 55,56. However, serum concen-
trations of maternal cystatin C fail to correlate with
other measures of GFR in pregnancy, including iohexol
clearance57, serum creatinine58 and eGFR59. This failure
to correlate may be caused by an increase in glomerular
negative charge, which reduces the excretion of anionic
cystatin C57. Assessment of renal function in pregnancy
is therefore limited to serial monitoring of serum cre-
atinine level, and caution should be taken in interpret-
ing data from publications that use eGFR equations in
pregnancy 60, which might underestimate pre-pregnancy
CKD severity.
Tubular changes
During pregnancy, several changes occur that affect
renal tubular function. Glycosuria frequently occurs
during pregnancy owing to an increase in filtration
in conjunction with reduced tubular reabsorption.
Although calcium excretion is increased, leading to
supersaturation of the urine, simultaneous increases in
urinary glycoproteins are thought to protect the kid-
neys against the formation of renal calculi and nephro­
lithiasis61,62. Furthermore, reabsorption of uric acid is
reduced in pregnancy, resulting in increased renal excre-
tion. By contrast, serum uric acid is increased in pre-­
eclampsia but is insufficiently sensitive for diagnosis63,
with diagnostic specificity reduced still further if there
is coexisting CKD.
Endocrine and systemic changes
Several changes occur during pregnancy, which relate
to the renal synthesis of or renal response to hormones.
Pregnancy is a state of profound physiological vasodilata-
tion with mediators (including progesterone, nitric oxide
and prostaglandins) leading to a 5–­10 mmHg reduction
in blood pressure during pregnancy. This change occurs
despite evidence of increased renal renin production and
activation of the systemic renin–angiotensin–­aldosterone
system (RAAS), therefore suggesting that pregnancy is
a state of relative RAAS resistance. By contrast, despite
high blood pressure, the RAAS is suppressed in patients
with pre-eclampsia compared with levels in normal
pregnancy, suggesting an increased sensitivity to RAAS
in pre-eclampsia, which has been attributed to activat-
ing angiotensin receptor autoantibodies and angio­
genic proteins, including soluble fms-like tyrosine
kinase 1 (sFLT1)64.
Vasopressin is an antidiuretic hormone that is released
by the posterior pituitary gland and stimulates water
reabsorption in the kidney in response to an increase in
osmolality or a reduction in extracellular fluid volume.
The systemic arterial vasodilatation that occurs in preg-
nancy is thought to lead to non-osmotic stimulation of
vasopressin, effectively lowering the osmotic threshold
for vasopressin release, contributing to water retention
and a reduction in plasma osmolality in pregnancy 65.
Outcomes for pregnant women with CKD
Historically, clinicians advised avoidance of pregnancy
for women with CKD66, a recommendation that — on
the basis of qualitative assessment — is associated with
frustration, anger and regret 67,68. Over the past decades,
the focus of advice has changed owing to advances in
obstetric surveillance and neonatal care. Nonetheless, we
find that many clinicians still desire a general ‘thresh-
old’ of renal disease severity above which pregnancy is
ill-­advised. Such a threshold cannot be determined by
a single creatinine measurement and does not reflect
the complex effects of renal dysfunction on pregnancy.
Instead, effectively communicated, individualized
pre-pregnancy counselling — based on disease aetiol-
ogy and severity, rate of GFR decline, obstetric history
and the presence of additional risk factors (see below)
— should enable women with any stage of CKD and
their partners to make an informed choice regarding
pregnancy. Pre-pregnancy counselling is recommended
for all women with pre-existing medical conditions69.
In  a recent survey of women with CKD, 66 of 72
(92%) respondents reported that expert pre-pregnancy
­counselling assisted with decision-making 70.
Risk factors
Data from population studies (778 women with CKD)71,
large cohort studies (504 women with CKD)60 and a
meta-analysis (23 studies, including 1,514 pregnant
women with CKD)72 show that women with CKD have
worse maternal and neonatal outcomes — ­including
higher rates of pre-eclampsia, preterm delivery,
small-for-gestational-age infants, admission to neonatal
intensive care units (NICUs) and perinatal death — than
women without CKD. Absolute rates of adverse out-
comes vary between studies and are affected by cohort
size, the measure of GFR60 and variance in the thresholds
for both iatrogenic preterm delivery and admission to
NICUs. However, an increment in all complications with
worsening renal function exists60 (FIG. 3).
Additional factors associated with adverse preg-
nancy outcomes that are independent of renal disease
severity include pre-existing hypertension60, transplan-
tation status3, lupus nephritis73, proteinuria74 and super­
imposed pre-eclampsia75. However, one study reported an
increased risk of adverse pregnancy outcomes in women
with stage 1 CKD (OR 1.88; 95% CI 1.27–2.79), even in

6 | ADVANCE ONLINE PUBLICATION www.nature.com/nrneph

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

a Bramham et al.78 b Piccoli et al.60

Pre-eclampsia

SGA <5% SGA <5%

Preterm delivery Preterm delivery

<34/40 <34/40

Preterm delivery Preterm delivery

<37/40 <37/40

NICU admission NICU admission

25% reduction in CKD stage shift

eGFR at 6 months or RRT start
post-partum

0 20 40 60 80 100 0 20 40 60 80 100
Rate of complications (%) Rate of complications (%)

Controls CKD stage 2 CKD stage 4–5 Controls CKD stage 2 CKD stage 4–5
CKD stage 1 CKD stage 3 CKD stage 1 CKD stage 3

Figure 3 | Adverse pregnancy outcomes according to chronic kidney disease stage. Two cohort studies60,78 show that
worsening of renal function (as assessed by chronic kidney disease (CKD) stage) correlates with Nature Reviews
a consistent | Nephrology
increment in
several pregnancy complications: preterm delivery; pre-eclampsia; neonatal intensive care unit (NICU) admission; loss of
residual maternal renal function; and small-for-gestational-age (SGA) infants (parts a and b).  Variation of absolute rates
of adverse pregnancy outcomes between the studies could be due to differences in cohort size, the use of estimated
glomerular filtration rate (eGFR) in pregnancy and different thresholds for both iatrogenic preterm delivery and NICU
admission. In addition, pre-eclampsia was excluded from the control group in one study78. RRT, renal replacement therapy.

the absence of systemic disease, hypertension and pro-
teinuria, leading to the hypothesis that CKD per se confers
risk in pregnancy 60.
Progression of maternal renal disease
Until the past decade, our understanding of CKD progres-
sion in pregnancy was limited to findings from two small
cohort studies76,77. A retrospective study of 70 pregnancies
in 58 women (mean serum creatinine level  168 μmol/l dur-
ing early pregnancy) reported a 25% reduction in kidney
function in 43% of women in the 6 weeks after delivery,
with 11% of the cohort progressing to ESRD by 6 months
post-partum76. Similarly, a prospective Italian study of 49
pregnancies reported that women with a pre-pregnancy
GFR <40 ml/min/1.73 m2 and proteinuria >1 g per day had
an accelerated decline in renal function (1.17 ± 1.23 ml/min
per month) at 6 months post-partum compared with
pre-pregnancy rates (0.21 ± 0.20 ml/min per month)77.
Comparable rates of loss of maternal renal function
were also found in studies from 2015 (REF. 60) and 2016
(REF. 78), but the sample sizes of pregnancies in women
with advanced disease (CKD stages 4–5) remain small.
Interestingly, one study 60 reported that even women with
CKD stage 1 are at risk, with 7.6% (28 out of 370) demon-
strating a post-partum decline in renal function. However,
this finding could relate to the use of GFR estimating equa-
tions during pregnancy that might have underestimated
the severity of renal disease and misclassified women with
a higher pre-pregnancy stage of disease as CKD stage 1 (see
above). A meta-analysis from 2015 that compared renal
disease progression in 552 pregnant women and 716 non-
pregnant women with CKD reported no significant differ-
ence (OR 0.96; 95% CI 0.69–1.35); however, in that study,
progression was crudely defined as a doubling of serum
creatinine level or a >50% reduction in eGFR, and only
women with CKD stages 1–3 were included72.
Specific disease conditions
Most cohort studies of CKD in pregnancy are hetero­
geneous and include a variety of renal disease aetiologies.
Maternal outcomes for specific disease aetiologies are
discussed below.

NATURE REVIEWS | NEPHROLOGY ADVANCE ONLINE PUBLICATION | 7

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

Pre-implantation genetic
diagnosis
(PGD). Examination of the
genetic profile of a gamete or
embryo before implantation.
Ribonucleoproteins
Protein–RNA complexes.
Endocardial fibroelastosis
A disease of the endocardium
characterized by collagen
deposition, endocardial
thickening and ventricular
hypertrophy.
Autosomal dominant polycystic kidney disease.
Autosomal dominant polycystic kidney disease
(ADPKD) has an estimated prevalence of 4 in 10,000
individuals79. Although normotensive women with
ADPKD reportedly have comparable pregnancy out-
comes to those of unaffected family members, pre-­
existing hypertension is a recognized risk factor for
adverse pregnancy outcomes in ADPKD75. Another
study reported no difference in neonatal risk between
54 women with ADPKD compared with 92 women with
simple cysts, although maternal complications, includ-
ing hypertension and pre-eclampsia, were increased in
patients with ADPKD80. Isolated case studies support
the theory that kidney volume is unaffected by preg-
nancy, and that the condition does not compromise
uterine growth81.
Through the autosomal dominant nature of ADPKD,
offspring have a 50% chance of inheriting the condi-
tion; this probability is markedly reduced through the
use of pre-implantation genetic diagnosis (PGD) to iden-
tify relevant mutations in embryos before implantation.
However, the ethics of genetic testing for an adult-onset
disorder — which is asymptomatic for many years —
need to be considered. In a single-centre study of 96
women with ADPKD, 50% of patients with CKD and
63% of those with ESRD would have opted for PGD if it
were available82. Single-embryo transfer is recommended
in women with CKD owing to the additive risks associ-
ated with multifetal pregnancy, including p ­ re-eclampsia
and preterm delivery.
Primary glomerulonephritis. Only limited data on
pregnancy outcomes in women with primary glomer-
ulonephritis are available to guide counselling and
management, as highlighted in a systematic review 83.
For example, to our knowledge only two studies of
pregnancy in women with minimal change disease have
been performed84,85. However, the association of pre-­
existing hypertension, proteinuria and renal impairment
with worse pregnancy outcomes is consistent between
­studies of different CKD aetiologies. Proteinuria has
been found to be an important determinant of outcome
in immunoglobulin A (IgA) nephropathy, which is the
most common primary glomerulonephritis in pregnant
women83. Proteinuria and birthweight are negatively
correlated86, and the presence of >1 g per day of uri-
nary protein has been shown to be associated with loss
of residual renal function independent of pregnancy 87.
A comparison of pregnancy outcomes between women
with IgA nephropathy and other glomerular diseases has
not been carried out.
Lupus nephritis. SLE predominantly affects women of
childbearing age, with lupus nephritis occurring in ~40%
of cases88. Although the risk of adverse pregnancy out-
comes is increased in women with active lupus nephritis89,
even in the context of preserved renal function90, preg-
nancy outcomes are improved when disease remission
is achieved before pregnancy. However, pregnancy can
trigger disease relapse, as evidenced by findings from a
meta-analysis of 37 studies comprising 2,751 pregnancies
which showed that lupus flares and lupus nephritis occur
in 26% and 16% of pregnancies, respectively 91. A pro-
spective cohort study comprising 61 pregnant women
demonstrated that an increase in lupus disease severity
score was associated with a proportional and significant
increase in the risk of preterm delivery 73. Similarly, the
presence of lupus nephritis increases the risk of preterm
delivery among women with SLE90. Other risk factors for
adverse pregnancy outcomes in lupus nephritis include
black92 and Hispanic93 ethnicity, pre-existing hyper­
tension89 and severity of proteinuria94. Therefore, delayed
conception until 6 months after lupus disease activity is
recommended, and pre-pregnancy renal biopsy might be
indicated to exclude active disease if proteinuria is per-
sistent. Given the risks associated with lupus nephritis, in
conjunction with the clinical difficulties in distinguishing
active lupus nephritis from pre-eclampsia (TABLE 2), all
women with current or previous lupus nephritis should
be managed by a specialist team that includes both physi-
cians and obstetricians with expertise in SLE. For women
with lupus nephritis and antiphospholipid antibodies,
active disease or proteinuria, prophylaxis against VTE
using low-molecular-weight heparin should be consid-
ered; however, the threshold urinary protein:creatinine
ratio at which the risk of this complication becomes
clinically relevant remains unknown95. Unless contra­
indicated, women with lupus nephritis should be offered
low-dose aspirin for pre-eclampsia prophylaxis and
hydroxychloroquine (see below).
Fetal complications in lupus nephritis may arise
through exposure to maternal autoantibodies against
the intracellular ribonucleoproteins Sjogren syndrome
type A antigen (SSA; also known as 52 kDa Ro protein
or TRIM21) and SSB (also known as lupus La protein),
which undergo placental transfer and confer a 2% risk of
congenital heart block (CHB) and a 16% risk of neonatal
cutaneous lupus96. Whereas neonatal cutaneous lupus is
benign and resolves with elimination of the maternal
antibody, CHB and endocardial fibroelastosis are associ-
ated with increased mortality. Therefore, surveillance
of fetal heart rate and echocardiography for suspected
cardiac involvement are recommended for women with
SSA+ and SSB+ antibodies. Weekly fetal echocardio­
graphy is recommended from 16 weeks gestation if
there is a history of a previous affected infant, although
the clinical benefits and cost-effectiveness of high-level
surveillance in women without an affected child remain
unknown97. Maternal hydroxychloroquine is associated
with a reduced risk of CHB in women with SLE and
­previously affected pregnancies98.
Diabetic kidney disease. Kidney disease is evident in
2.5–6.5% of pregnancies in women with type 1 diabe-
tes mellitus99, although rates of up to 25% are described
when disease definitions are extended to include
microalbuminuria with a normal eGFR100. Diabetes
is associated with an increased risk of complications
— including progression of retinopathy, ketoacidosis,
exacerbation of pre-existing cardiovascular disease and
fetal death — independent of whether or not there is
coexisting renal disease. A retrospective analysis of 43

8 | ADVANCE ONLINE PUBLICATION www.nature.com/nrneph

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

Macrosomia pregnancies in women with diabetic kidney disease who
A large-for-gestational-age had a median pre-pregnancy serum creatinine level
infant. of 68 μmol/l (0.77 mg/dl) reported much higher rates of
adverse pregnancy outcomes than expected for the level
of renal impairment — including pre-­eclampsia in 42%
and preterm delivery in 77% of patients99. Gestational
proteinuria can be substantial and the risk of VTE
should be considered, although only limited data regard-
ing the proteinuric threshold for heparin prophylaxis are
available95. Fetal growth surveillance in patients with dia-
betic kidney disease may be falsely reassuring, as growth
restriction due to underlying CKD can be masked by
hyperglycaemia-induced m ­ acrosomia. Congenital mal-
formation rates increase with poor glycaemic control
around the time of conception, but diabetic kidney
disease confers an additional risk of fetal malforma-
tion compared with women without nephro­p athy
(adjusted OR 2.5; 95% CI 1.1–5.3)101. Some evidence
supports intensification of RAAS blockade through the
use of either angiotensin-converting enzyme inhibi-
tors or angiotensin receptor blockers before pregnancy,
although these need to be discontinued in p ­ regnancy
because of fetotoxicity (see below). Data from single-­
arm studies with 8–24 participants show that pre­
conception proteinuria is reduced by RAAS blockade
with an associated decrease in pregnancy-induced pro-
teinuria102,103. The use of intensified targets for blood
pressure (<135/85 mmHg) and proteinuria (<300 mg
per day) — in conjunction with preconceptual RAAS
blockade — is associated with obstetric outcomes that
are comparable to those of women with diabetes but
without proteinuria104. Comprehensive guidelines for the
management of diabetic kidney disease in pregnancy are
available elsewhere105.
Important comorbidities
Anaemia, vitamin D deficiency and chronic hyper­
tension are comorbidities that are associated with
adverse pregnancy outcome and can develop secondary
to underlying CKD.
Anaemia
Gestational increases in plasma volume in pregnancy
are proportionally higher than the corresponding
increase in red blood cell mass, leading to haemo­
dilution and a relative fall in haemoglobin levels.
Although variation exists in the definition of anaemia
in pregnancy (haemoglobin concentrations <105 g/l
to <110 g/l, depending on gestation106,107), values that
are <85 g/l in the general obstetric population have
been shown to be associated with increased risks
of low birthweight (<2,500 g) and preterm delivery
(<37 weeks) by 62% and 72%, respectively, across
­multiple ethnic groups108.
Erythropoiesis is regulated by erythropoietin, which
is produced by fibroblast-like cells within the kidney.
Erythropoietin concentrations increase approximately
twofold during pregnancy 109. As women with CKD may
have insufficient capacity for a gestational increase in
erythropoietin production, supplementation with
synthetic erythropoietin may be required, even in
the context of mild renal impairment. For women
who required erythropoietin before pregnancy, an
increased dose should be anticipated during pregnancy.
As erythropoietin is a large molecule that does not
cross the placental barrier, its use is considered safe in
pregnancy and breastfeeding 2,110,111; however, a theor­
etical risk of exacerbating pre-existing or n ­ ew-onset
hypertension exists.

Table 2 | Clinical overlap between normal pregnancy, CKD, active lupus and pre-eclampsia
Clinical features Normal pregnancy CKD Lupus flare Pre-eclampsia
Hypertension No Yes Yes Yes
Proteinuria <300 mg per day Yes Yes Yes
Haematuria No Yes Yes No
Increased serum Physiological Not above pre-pregnancy Yes Yes
creatinine creatinine level
Rate of change in Third trimester None unless disease Can be rapid (hours or Can be rapid
serum creatinine progression or days) (hours or days)
superimposed acute injury
Skin Hyperpigmentation, No Malar rash, discoid lupus, No
striae acute cutaneous lupus
Joints Mechanical pain No Arthritis No
Hair loss Yes (post-partum No Yes No
telogen effluvium)
Haemoglobin Iron-deficiency Erythropoietin deficiency Anaemia of chronic Haemolysis
anaemia disease, haemolysis
Platelets Gestational Normal Thrombocytopenia Thrombocytopenia
thrombocytopenia
Other blood tests ↑ ESR ↓ eGFR prior to pregnancy • ↓ Complement • Transaminitis
• ↑dsDNA • ↓ PLGF
• ↑sFLT1
CKD, chronic kidney disease; dsDNA, double-stranded DNA; eGFR, estimated glomerular filtration rate; ESR, erythrocyte
sedimentation rate; PLGF, placenta growth factor; sFLT1, soluble fms-like tyrosine receptor kinase 1.

NATURE REVIEWS | NEPHROLOGY ADVANCE ONLINE PUBLICATION | 9

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

Hyperparathyroidism
Increase in parathyroid
hormone levels, which can be
primary due to pathology
within the parathyroid gland
or secondary due to
hypocalcaemia or
hyperphosphataemia (both of
which can be caused by CKD).
The most common cause of anaemia in pregnancy
is iron deficiency, which is estimated to affect >40% of
pregnancies112. Hence, pregnant women with CKD may
need both iron and erythropoietin supplementation.
Markers of iron deficiency in nonpregnant patients
with CKD include ferritin <100 μg/l, transferrin satura-
tion (<20%), >6% hypochromic red cells and reticulo­
cyte haemoglobin content equivalent to <25 pg 113;
however, the specificity and sensitivity of these markers
in assessing iron status in pregnancy are unknown and
should be investigated further. Oral iron is inexpen­
sive and accessible, although the intravenous route may
provide more effective replacement in CKD owing to
better bioavailability 114. Parenteral iron is generally
considered safe in pregnancy and breastfeeding 5,115–118,
although a rare risk of an allergic reaction exists.
Vitamin B12 (also known as cobalamin) deficiency
is a rare cause of anaemia during pregnancy but may
need to be considered for patients on low-protein diets.
Protein prescription in patients with CKD who are not
pregnant requires a balance between the potential for
a reduction in renal disease progression and optimal
nutritional status. However, in pregnancy, nutritional
status predominates and low-protein diets should not
be recommended. Of note, assessment of vitamin B12
status during pregnancy is inherently difficult, as
physiological lowering of plasma vitamin B12 ­levels
by up to 30% contributes to misdiagnosis of defi-
ciency when nonpregnant reference ranges are used
for interpretation119.
Hypoxia-inducible factor (HIF) activators are an
emerging class of drugs that might have a therapeutic
role in the management of renal anaemia via stimula-
tion of erythropoietin production. However, the small
size of these molecules potentially enables placen-
tal transfer, and HIF has multiple direct and indirect
effects that have the potential to modulate almost any
developmental, physiological process120. No formal rec-
ommendation has been made for the use of this class of
drug in pregnancy, but on the basis of their molecular
characteristics, we would not recommend their use at
conception or during pregnancy.
Vitamin D deficiency
Vitamin  D deficiency affects 13–64% of pregnant
women121 and is associated with increased incidences of
pre-eclampsia and gestational diabetes mellitus. Meta-
analyses report that oral vitamin D supplementation
is associated with reduced risks of pre-eclampsia, low
birthweight and preterm birth, although inconsistencies
exist in the findings of individual studies122,123.
Vitamin  D status in pregnancy is assessed by
measuring serum levels of calcifediol (also known as
25‑hydroxyvitamin D, a precursor of active vitamin D),
although optimal serum calcifediol levels (variably
defined as a >20–30 ng/ml (50–75 nmol/l)124), as well as
optimal doses of colecalciferol or ergocalciferol, remain
unknown. According to expert consensus, serum cal-
cifediol levels are to be checked in pregnancy for all
pregnant women with CKD, and replacement colecalci­
ferol can be given at a dose of 20,000 international units
(IU) per week until replete for women with a serum
calcifediol <20 ng/ml (50 nmol/l). This treatment is
followed by a maintenance daily dose of 400–1,000 IU,
depending on ethnicity and body mass index 121.
Although calcifediol is the major circulating form of
vitamin D, it has low biological activity until converted
to calcitriol (also known as 1,25‑dihydroxyvitamin D3).
Serum calcitriol levels are approximately threefold
higher in the first trimester and five to six times higher
in the third trimester than those in nonpregnant
women125. To what extent this increase is dependent on
the capacity of the kidney to augment ­1α‑hydroxylase
enzyme activity is unknown126, as 1α‑hydroxylase activ-
ity is also present in the colon, skin, macrophages and
the placenta127. In the absence of better evidence, active
vitamin D analogues (such as alphacalcidol and calcit-
riol) should be continued throughout pregnancy at the
appropriate pre-pregnancy dose.
In nonpregnant women with CKD, assessment of
bone health includes both vitamin D status and para-
thyroid hormone levels. Hyperparathyroidism can occur
in CKD secondary to both vitamin D deficiency and
impaired renal clearance of phosphate. Of note, as
pregnancy does not cause an increase in parathyroid
hormone levels, they can be interpreted as for nonpreg-
nant individuals. However, treatment of hyperpara­
thyroidism with calcimimetics is not recommended
owing to i­ nadequate safety data (see below).
Chronic hypertension
Chronic hypertension affects 20–50% of pregnant
women with CKD, depending on the severity of renal
disease60. A meta-analysis comprising >750,000 women
shows that chronic hypertension is associated with
superimposed pre-eclampsia (relative risk (RR) 7.7;
95% CI 5.7–10.1), preterm delivery (RR 2.7; 95% CI
1.9–3.6), low birthweight (RR 2.7; 95% CI 1.9–3.8),
NICU admission (RR 3.2; 95% CI 2.2–4.4) and peri­
natal death (RR 4.2; 95% CI 2.7–6.5)128. Control of
hypertension is a key component of nonpregnant CKD
management. However, a historical, theoretical concern
that aggressive blood pressure control during preg-
nancy could impair fetal growth and contribute to fetal
morbidity exists. In the 2015 Control of Hypertension
in Pregnancy Study (CHIPS) trial, in which 987
patients with hypertension were randomly assigned
to either tight (85 mmHg) or less tight (100 mmHg)
diastolic blood pressure targets129, tight blood pres-
sure control (mean 133/85 mmHg) was not associated
with an increase in adverse pregnancy outcomes but
was associated with a reduced risk of severe maternal
hypertension compared with less tight control (mean
139/90 mmHg). Although the CHIPS trial excluded
women with proteinuria, increasing expert consen-
sus is that hypertension in pregnant women with
CKD should be treated to a target blood pressure of
<140/90 mmHg (REFS 83,130). Nevertheless, no com-
prehensive studies of chronic hypertension specific
to pregnant women with CKD have been carried out to
date, and published data are insufficient to establish
evidence-based targets.

10 | ADVANCE ONLINE PUBLICATION www.nature.com/nrneph

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

Therapeutic management to the timing of intercourse and the regularity of the

Supratherapeutic dosing
Administering a drug dose that
is higher than that needed to
achieve therapeutic effects.
As many pregnant women with CKD require medi-
cation for control of hypertension, inflammatory glo-
merular disease or prevention of transplant rejection,
an understanding of drug compatibility with pregnancy
and breastfeeding is essential before any prescription
in all women of childbearing age. However, manufac-
turers and medication administrative bodies (such as
the US Food and Drug Administration (FDA)) have
understandably taken a cautionary approach to med-
ication use in pregnancy. As a result, very few drugs
have a licence for use in pregnancy, and pregnant and
lactating women are often specifically excluded from
­studies of safety and efficacy. Recommendations regard-
ing safety are usually based on data from retrospective
case series, isolated case reports and animal studies. An
attempt by the FDA to translate available drugs into
a letter-based category (A–D plus X; corresponding
to increasing levels of evidence of harm) is simplistic
and has led to false assumptions about drug safety.
For example, on the basis of isolated case reports and
supratherapeutic dosing in animals, azathioprine is clas-
sified by the FDA as category D and therefore unsafe
for use in pregnancy; however, data in humans indi-
cate that the drug is safe in clinical practice (see below).
Although FDA categorization was changed in 2015 to a
Pregnancy and Lactation Labelling Rule (PLLR) aimed
at improving the available information, the revision of
labelling for existing drugs may take up to 3 years. An
overview of medications commonly used in nephrology
practice and their safety in relation to their use during
conception, pregnancy and ­breastfeeding is provided
in TABLE 3.
Antihypertensive drugs
The safety of antihypertensive drugs in pregnancy is
based on historical, retrospective analyses and favours
older classes of drugs, including the adrenergic recep-
tor agonists labetalol and methyldopa and the calcium
channel blocker nifedipine. These drugs are considered
safe in the management of both pre-existing and gesta-
tional hypertension in pregnancy. Although the thera-
peutic effect of different classes of antihypertensives is
known to vary with ethnicity in nonpregnant patients,
no equivalent robust data are currently available to
guide drug selection in pregnancy; therefore, the choice
of drug is determined by its availability, licensing, local
protocols as well as the experience and preference of
the physician.
Angiotensin-converting enzyme inhibitors and
angiotensin receptor blockers are able to cross the pla-
centa. These classes of drugs are fetotoxic in the second
and third trimesters — causing oligohydramnios and
neonatal renal failure — but not in the first trimester,
when population data are corrected for confounding
variables, including maternal demography, and comor-
bidities such as diabetes and hypertension131,132. Thus,
RAAS blockers can be continued for nephroprotection
during the period of time taken to conceive, provided
they are discontinued at an early diagnosis of preg-
nancy. Pregnancy testing is recommended according
menstrual cycle and should be undertaken at least once
per month for those with irregular cycles.
Labetalol, nifedipine, captopril and enalapril are
compatible with lactation133,134. Methyldopa is substi-
tuted post-partum because of the risk of exacerbating
postnatal depression.
Immunosuppressant drugs
Corticosteroids. Steroids are used to inhibit immune
responses in inflammatory and autoimmune condi-
tions and to prevent transplant rejection. Prednisolone
is the preferred agent for use in pregnancy as placen-
tal metabolism of prednisolone results in only 10% of
the maternal dose reaching the fetal circulation135,136.
However, as corticosteroid use is associated with an
increased risk of gestational diabetes mellitus, hyper-
tension, urinary tract and other infections as well
as preterm delivery 137, all corticosteroids should be
administered at the lowest possible therapeutic dose.
Screening for gestational diabetes mellitus is recom-
mended at 28 weeks gestation or sooner if other risk
factors such as increased body mass index and ethnicity
are present. Although historical, retrospective studies
suggested an association between first-trimester corti-
costeroid exposure and an increase in the incidence of
cleft lip and palate138,139, this link has not been substan-
tiated in prospective case–control studies140–142, and a
cohort study with >50,000 steroid-exposed pregnancies
failed to show any risk of oral facial clefts143. Parenteral
steroids to cover the physiological stress of delivery
should be considered for women receiving >7.5 mg
prednisolone per day for >2 weeks during pregnancy.
The concentration of prednisolone in breastmilk is
estimated to be 0.015–0.074% of the maternal dose133. No
adverse effects have been reported in breastfed infants
with maternal use of any corticosteroid, and evidence
suggests that neonatal adrenal suppression ­following
in utero or lactation exposure does not occur 144.
Azathioprine. Azathioprine is a precursor to immuno­
suppressive antagonists of purine metabolism. The drug
undergoes degradation to 6‑mercaptopurine before
enzymatic conversion to active thioguanine nucleo-
tides by a number of enzymes, including thio­purine
S-methyltransferase (TPMT). The fetus is likely to be
protected to some extent from the clinical effects of
azathioprine and 6‑mercaptopurine as the immature
fetal liver is not thought to express inosine triphosphate
pyrophosphatase, which is another of the enzymes
involved in conversion to active metabolites, although
further studies of the metabolic pathways in mother,
placenta and fetus are needed. Outcome s­ tudies in
women receiving thiopurines for renal disease, inflam-
matory bowel disease145 and connective tissue disease
do not demonstrate an increased risk of preterm deliv-
ery, congenital malformations or childhood neoplasia
when data are corrected for underlying maternal dis-
ease. Azathioprine is therefore considered safe to use
throughout pregnancy 146–148. If azathioprine treatment is
started during pregnancy, it is advisable to check TPMT

NATURE REVIEWS | NEPHROLOGY ADVANCE ONLINE PUBLICATION | 11

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

Table 3 | Considerations for drugs commonly used in nephrology during conception, pregnancy and lactation
Drug Conception Pregnancy Lactation Refs
Overall Maternal Fetal considerations
considerations
Antihypertensive drugs
Labetalol Safe Safe Licenced for pregnancy. No association with Safe (also 133,166
Consider other congenital abnormalities metoprolol and
comorbidities (for atenolol)
example, asthma)
Nifedipine Safe Safe None No association with Safe 133,166
congenital abnormalities
Methyldopa Safe Safe None No association with Avoid owing to 133,166
congenital abnormalities risk of postnatal
depression
ACE inhibitors No apparent Unsafe None Fetotoxic in second Safety data 131–134,
increase in risk and third trimesters, available for 166
with first-trimester leading to fetal and captopril and
use when correct neonatal renal failure, enalapril
for underlying bone and aortic
hypertension. arch malformations,
Continue until oligohydramnios and
conception if required pulmonary hypoplasia
for nephroprotection
Angiotensin Insufficient data Unsafe None Fetotoxic in second No data 133,166
receptor blockers on exposure in and third trimesters,
early pregnancy. leading to fetal and
Discontinue in neonatal renal failure,
advance of pregnancy bone and aortic
arch malformations,
oligohydramnios and
pulmonary hypoplasia
Immunosuppressant drugs
Corticosteroids Safe Safe Potential risks: Fetus exposed to <10% Safe. Small amounts 133,
diabetes; hypertension; maternal dose owing to in breast milk. 135–144
pre-eclampsia; placental deactivation. Consider timing
infection; preterm No evidence of increased feeds to 4 hours
rupture of membranes; congenital abnormalities after administration
aim for minimum if high dose given
maintenance dose (for example,
methylprednisolone
induction) and
monitor infant
Azathioprine Safe Safe Recommend checking Placental transfer. Safe. Low 133,
TPMT status before No association with concentration in 145–149
dosing congenital abnormalities breast milk
Mycophenolate Unsafe. Effective Unsafe None Placental transfer. Excreted in breast 133,150
contraception during Teratogenic, causing milk. Avoid use
treatment and for ear, heart, eye, lip during lactation
minimum 6 weeks and/or palate, kidney
after treatment. and bone abnormalities,
Ensure disease tracheoesophageal
stability prior to fistula and congenital
conception diaphragmatic hernia.
Increased rate of
miscarriage
Ciclosporin Safe Safe Monitor pre-dose levels. Placental transfer. Safe 133,152,
May need higher dose in No association with 154
pregnancy congenital abnormalities.
Increased risk of
gestational diabetes
Tacrolimus Safe Safe Monitor pre-dose levels. Placental transfer. Safe 133,
May need a higher dose No association with 153–155
in pregnancy congenital abnormalities.
Increased risk of
gestational diabetes

12 | ADVANCE ONLINE PUBLICATION www.nature.com/nrneph

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

Table 3 (cont.) | Considerations for drugs commonly used in nephrology during conception, pregnancy and lactation
Drug Conception Pregnancy Lactation Refs
Overall Maternal Fetal considerations
considerations
Immunosuppressant drugs (cont.)
Hydroxychloroquine Safe Safe Withdrawal may Placental transfer. No Safe 73,98,133,
precipitate lupus flare. increase in miscarriage or 157–160
Indicated throughout congenital abnormality.
pregnancy if patient May reduce risk of CHB
has a history of lupus with anti-SSA+ and/or
nephritis anti-SSB+
Cyclophosphamide Unsafe. Effective Unsafe None Placental transfer. Excreted in breast 133,161
contraception during Teratogenic. Congenital milk. Discontinue
and for 3 months abnormalities of the skull, breastfeeding
after treatment. ear, face, limb and visceral during and for
Dose-related and organs. Increased risk of 36 hours after
age-related risk of miscarriage treatment
infertility
Rituximab Unclear (limited data Unclear If indicated for severe Active placental transfer Unclear (limited 133,162,
available). Treatment (limited disease, aim to give during second and third data available). 163
decision depends data dose before, or in early, trimester. Potential Possible excretion of
on indication and available) pregnancy to minimize risk of neonatal B cell trace amounts, but
alternative options the risk of neonatal depletion. Avoid unless neonatal absorption
B cell depletion potential benefit to unlikely
mother outweighs
risk. Long-term effects
unknown
Eculizumab Unclear (limited data Unclear Morbidity of underlying Active placental transfer Unclear (limited 133,164,
available). Treatment (limited condition may mean in second and third data available). 165
decision depends data treatment required in trimester. No congenital Possible excretion of
on indication and available) pregnancy. Monitor abnormality reported trace amounts, but
alternative options for increased dosage in 20 infants. Long-term neonatal absorption
requirements effects unknown unlikely
Other drugs
Aspirin (75–150 mg) Safe Safe Decreases risk of No association with Safe 133,
pre-eclampsia. No congenital abnormalities 166–170
evidence of maternal
haemorrhagic
complications
Low-molecular- Safe Safe Used for VTE No placental transfer Safe 95,133
weight heparin prophylaxis. Threshold
for use in proteinuria is
unknown
Allopurinol Unclear (limited data Unclear None Placental transfer. No Therapeutic levels in 133,173
available) (limited increase in congenital breastfed infants
data abnormality in 31
available) infants. Cautionary
use in first trimester
recommended
Colchicine Safe Safe None Associated with Infants receive an 133,174
preterm delivery and estimated 10% of
lower birthweight, maternal dose with
but contribution of no reported adverse
underlying disease is effects
unclear
Iron Safe Safe Intravenous None Safe 114,118,
preparations may offer 133
better bioavailability
in CKD
Erythropoietin Safe Safe Increased gestational No placental transfer Safe 109‑111,
requirement. Monitor 133
blood pressure
ACE, angiotensin-converting enzyme; CHB, congenital heart block; CKD, chronic kidney disease; SSA, Sjogren syndrome type A antigen; SSB, Sjogren syndrome
type B antigen; TPMT, thiopurine methyltransferase; VTE, venous thromboembolism.

NATURE REVIEWS | NEPHROLOGY ADVANCE ONLINE PUBLICATION | 13

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS
Diaphragmatic hernia
Congenital defect in the
diaphragm that allows
movement of abdominal
viscera into the chest.
Microtia
A congenital abnormality in
which the pinna (external ear)
is underdeveloped.
Micrognathia
A congenital abnormality
in which the jaw is
underdeveloped.
14 | ADVANCE ONLINE PUBLICATION www.nature.com/nrneph
levels as a dose reduction is required for those with an has been used in women with SLE during pregnancy
inherited deficiency of TPMT to prevent maternal for over 3 decades with reassuring safety data, and a
toxicity. As for nonpregnant patients with CKD, mon- meta-analysis reported no increase in the risk of con-
itoring of leukocyte count and serum transaminases is genital malformation, preterm delivery, still birth or
recommended because of possible adverse effects of low-birthweight infants158. Of note, the association of
myelotoxicity and liver injury. Monitoring is required hydroxychloroquine with a higher spontaneous abor-
fortnightly at drug initiation, reducing to once per tri- tion rate may be confounded by the earlier presentation
mester when a stable dose is achieved. Azathioprine is of women taking hydroxychloroquine to health services,
compatible with breastfeeding 146,149. which improves the documentation of early pregnancy
loss compared with controls. This association also lacks
Mycophenolate mofetil. Mycophenolate mofetil biological plausibility, as hydroxychloroquine has been
(MMF) is used in nonpregnant patients for the treat- associated with evidence of improved placentation.
ment of lupus nephritis and for ongoing immuno­ Pregnant women with SLE who stop hydroxychloro-
suppression following renal transplantation. However, quine treatment have a higher requirement for antenatal
the compound is teratogenic, causing a typical clinical corticosteroids to suppress disease activity than those
syndrome that includes hypoplastic nails, shortened fin- who continue treatment 159. In addition, hydroxychloro-
gers, diaphragmatic hernia, microtia, micrognathia, cleft lip quine use is linked to reductions in both growth restric-
and palate and congenital heart defects150,151. The risk tion and the recurrence of CHB73,98. Counselling and
of these outcomes is estimated to be 20–25%, but this screening for hydroxychloroquine-related retinopathy
approximation may be influenced by reporting bias. In is recommended regardless of pregnancy 160. Although
addition, the relative risks of withdrawal during organo- hydroxychloroquine is excreted into breast milk, meas-
genesis in the first trimester, followed by continued use ured levels are 0.25–2% of the maternal dose, with no
in the second and third trimester, are difficult to quan- adverse effects reported133.
tify. Women receiving MMF are therefore advised to
switch to a nonteratogenic immunosuppressant agent Cyclophosphamide. Cyclophosphamide, which is used
(such as azathioprine or a calcineurin inhibitor) at in the management of rapidly progressive glomerulo-
least 3 months before conception. For the rare cases in nephritis, is a potent immunosuppressant that causes
which MMF is the only drug that can achieve disease or lymphocyte cytotoxicity via inhibition of DNA replica-
transplant stability, very careful counselling regarding tion. The drug is teratogenic and fetotoxic, causing skull
the risks of continuing MMF treatment during preg- abnormalities, ear and craniofacial malformations, limb
nancy is required. There is an absence of data regarding and visceral organ abnormalities and growth restriction.
MMF and lactation, and treatment is not recommended In addition, suppression of haematopoiesis and neuro-
­during breastfeeding. logical impairment occurs with exposure after the first
trimester 161. Cyclophosphamide is excreted into breast
Calcineurin inhibitors. Calcineurin inhibitors (such as milk, with cytopenia reported in exposed infants. Thus,
ciclosporin and tacrolimus) target signalling pathways the use of cyclophosphamide is contraindicated in
that regulate T cell activation. Although neither ciclo- pregnancy and lactation. Although pregnancy should
sporin nor tacrolimus are teratogenic152,153, both drugs be avoided in the context of progressive renal disease
are diabetogenic154; therefore, screening for gestational and active vasculitis, plasma exchange and rituximab
diabetes mellitus should be undertaken at 28 weeks of (see below) can be considered as alternative treatment
gestation or earlier if other risk factors are present. options in the context of unplanned ­pregnancy or for a
Owing to pregnancy-related changes in protein bind- new diagnosis made during pregnancy.
ing, the volume of distribution, and drug clearance, ther-
apeutic drug-level monitoring of calcineurin inhibitors Rituximab. Rituximab, which is a monoclonal antibody
is essential, and higher doses may be needed. According that is increasingly used to treat primary and second-
to expert consensus, however, the lower end of the tac- ary glomerular diseases, opsonizes and destroys B cells
rolimus pre-pregnancy dosing range is recommended in by binding to the CD20 antigen. Despite its half-life of
pregnant patients because of the possibility of high free
~22 days, B cell depletion occurs for up to 6 months162;
drug levels during pregnancy, which are not detectable hence, rituximab administration, even before preg-
with the use of standard assays155. nancy, may lead to intrauterine exposure. However,
The use of tacrolimus is considered safe during continued rituximab administration during pregnancy
breastfeeding. Babies of mothers taking tacrolimus are may be n ­ ecessary for disease remission.
born with therapeutic drug levels, but these fall progres- Rituximab freely crosses the placenta, but data
sively in both breastfed and bottle-fed infants156. Fewer
on the effects of fetal exposure are limited: one study
lactation data for ciclosporin exist, but no adverse effects
reported B cell depletion in 11 of 90 exposed infants but
have been reported. no increase in congenital malformations163. Neonatal
effects are minimized with use in the first trimester,
Hydroxychloroquine. Hydroxychloroquine has diverse when placental transfer is passive, rather than during
immunomodulatory effects, including a reduction the second and third trimesters, when active placen-
in inflammatory cytokines, inhibition of antigen-­ tal transfer of antibodies occurs. Neonatal B cell count
presenting cells and antioxidant activity 157. The drug and rituximab levels can be checked after delivery
©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

if indicated (for example, in neonatal sepsis). Delay growth restriction in pregnant women with CKD60.
of neonatal vaccination with live vaccines (such as Low-dose aspirin can be continued during breastfeed-
Bacillus Calmette-Guérin (BCG), rotavirus or varicella ing if indicated, as transfer into milk is undetectable
zoster virus) is advisable until the neonate is older than (with levels of the metabolite salicylic acid probably
6 months following in utero exposure to rituximab. being subclinical to infants170).
Long-term infant outcomes following in utero exposure
to rituximab remain unknown. Cinacalcet. Cinacalcet, which is used in CKD for the
Limited data indicate that rituximab is transported treatment of refractory secondary hyperparathy-
into breast milk in trace concentrations, although roidism, is a calcimimetic that increases the sensitivity
absorption is considered unlikely due to its destruction of the calcium-­sensing receptor to extracellular calcium,
in the gastrointestinal tract of the infant 133. Therefore, thereby inhibiting secretion of parathyroid hormone.
­rituximab is not a contraindication to lactation. The few available studies that have described the use
of cinacalcet in pregnancy 171,172 have reported success-
Eculizumab. Eculizumab is a monoclonal antibody that ful outcomes. However, in the absence of sufficient
targets active complement protein C5. The antibody is safety data, current practice is to stop c­ inacalcet before
used for the treatment of paroxysmal nocturnal haemo­ ­pregnancy and during breastfeeding.
globinuria as well as the induction and maintenance
treatment of atypical haemolytic uraemic syndrome Allopurinol. Allopurinol crosses the placenta, and its
(aHUS), which can be triggered or relapse during preg- use in the first trimester is cautioned, although mal-
nancy. Although published data are limited, eculizumab formation rates were not increased in a cohort of 31
use in pregnancy may be necessary to reduce morbid- women173. A therapeutic allopurinol dose is received by
ity, including ESRD associated with aHUS164. In 75 breastfed infants, and although no adverse effects have
pregnancies from 61 women with paroxysmal noc- been reported, infants should be monitored for adverse
turnal haemoglobinuria, no increased risk of congen- effects such as rash and gastrointestinal symptoms133.
ital abnormality was reported with use of eculizumab,
although the antibody was detected in seven of 20 cord Colchicine. Colchicine is an anti-inflammatory agent
blood samples165. Long-term effects of in utero exposure used in the management of rheumatological disease and
to eculizumab remain unknown. As pregnancy affects gout. A meta-analysis of 550 pregnancies exposed to col-
the pharmacokinetics of this agent, an increase in dos- chicine doses of 1–2 mg per day showed no association
age or frequency of treatment may be required165. No with miscarriage or malformation. Although there was
antibody was detected in the breast milk of ten patients a measured association between colchicine and reduc-
with p­ aroxysmal nocturnal haemoglobinuria165. tions in gestational age and birthweight in a small num-
ber of heterogeneous studies, the clinical significance
Other drugs of these data is unclear, with possible confounding due
Aspirin. Aspirin is an antiplatelet agent used in preg- to the underlying condition174. An estimated 10% of
nancy from 12 weeks of gestation to reduce the risk the dose is received from breast milk, with no adverse
of pre-eclampsia in high-risk women 166. A meta-­ effects reported133.
analysis shows that low-dose aspirin (60–150 mg)
confers a reduction in pre-eclampsia (RR 0.76; 95% CI Dialysis in pregnancy
0.62–0.95), preterm birth (RR 0.86; 95% CI 0.76–0.98) Haemodialysis. A dramatic increase in the number of
and intrauterine growth restriction (RR 0.8; 95% CI pregnancies has been reported in women requiring renal
0.65–0.99) as well as an absence of evidence of harm167. replacement therapy. Systematic review data reported
An RCT in women diagnosed as being at high risk of 90 cases between 2000 and 2008, which increased to
pre-eclampsia on the basis of a maternal physiological 574 cases only 6 years later 175. Augmentation of hae-
and biomarker profile performed at 11–13 weeks of modialysis dose is associated with fewer fetal deaths176,
gestation showed a 62% reduction (OR 0.38; 95% CI and a meta-analysis shows a continuous inverse corre-
0.20–0.74) of preterm pre-eclampsia in women taking lation between the number of haemodialysis hours per
150 mg per day aspirin compared with those taking pla- week and rates of preterm birth (P = 0.044; r2 = 0.22) and
cebo168. Interestingly, a similar risk reduction was not small-for-gestational-­age infants (P = 0.017; r2 = 0.54)175.
seen in this study for women with chronic hypertension In a contemporary cohort of 17 women including 22
(OR 1.30; 95% CI 0.33–5.12)169, despite the hypothesis pregnancies, intensive haemodialysis (defined as an
that pre-eclampsia risk, and therefore bene­f it from average of 43 hours per week) conferred an 86% live
aspirin prophylaxis, would be proportionally greater birth rate at a mean gestational age of 36 weeks com-
than in normotensive women. However, this finding is pared with a live birth rate of 61% at a mean gestational
limited by the possibility of an underpowered subgroup age of 27 weeks in 70 women receiving haemodialysis
analysis and the difficulty in making a diagnosis of for a mean of 17 hours per week177. Provision of dialysis
superimposed pre-eclampsia in the absence of standard for 48 hours per week optimizes maternal physiology,
blood pressure parameters. Given the limitations of this with patients achieving gestational blood pressures of
single study, all women with CKD should be offered 112/72 to 122/81 mmHg with minimal use of antihy-
low-dose aspirin in pregnancy owing to the increased pertensive agents, and predialysis urea concentrations
risks of pre-eclampsia, preterm delivery and fetal of 8–14 mmol/l, although increased dialysate provision

NATURE REVIEWS | NEPHROLOGY ADVANCE ONLINE PUBLICATION | 15

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

of potassium, calcium and phosphate may be required24.
Nonetheless, complication rates remain high even in
women receiving intensive haemodialysis, and >75% of
such women require preterm delivery 175,177. Pregnancy
on haemodialysis remains a high-risk option, and the
Box 1 | Reproductive health issues for women with CKD
Pre-pregnancy
Fertility
• Reduced in chronic kidney disease (CKD), but pregnancy can occur at all CKD stages.
• Recommend pursuit of fertility preservation if cyclophosphamide is administered.
Contraception
• Risk of adverse outcomes to unplanned pregnancy at all CKD stages.
• Progesterone-only methods (including the ‘mini pill’, intrauterine devices and
subdermal implants) are safe and effective in women with CKD.
• By contrast, barrier methods have high failure rates with typical use.
Conception
• Pre-pregnancy counselling before conception is strongly recommended.
• Remove or substitute teratogens before conception.
• Consider risks and/or benefits to remaining on angiotensin-converting enzyme
inhibitors until conception.
• Ensure disease and/or transplant stability before conception.
Pregnancy
Risk management
• Incremental risk of adverse outcomes with increasing CKD severity.
• Increased obstetric surveillance (including growth scans and Doppler ultrasound)
required.
• Venous thromboembolism prophylaxis for women with substantial proteinuria, lupus
flare and antiphospholipid antibodies.
• Increased dialysis provision improves outcomes in stage 5 CKD.
• Gestational diabetes mellitus screening in women receiving steroids and/or
calcineurin inhibitors (CNIs).
• Therapeutic drug monitoring for women taking CNIs.
Hypertension
• Labetalol, nifedipine and methyldopa are safe to use during pregnancy.
• No increase in adverse fetal outcome when treated to a diastolic blood pressure
of 85 mmHg.
• Treatment target of <140/90 mmHg (according to expert consensus).
Pre-eclampsia
• Low-dose aspirin (75–150 mg) prophylaxis recommended for all pregnant women
with CKD.
• Diagnosis of pre-eclampsia is complicated by pre-existing hypertension and
proteinuria.
Anaemia
• Erythropoietin and iron supplementation may be required.
Post-partum
Lactation
• Captopril and enalapril are safe to use in breastfeeding.
• Neonatal tacrolimus clearance in breastfed infants is comparable to that of
bottle-fed infants.
• Up-to-date drug safety data in lactation are available103.
Drugs
• Review CNI dose if dose was increased during pregnancy.
Renal function
• Risk of loss of residual renal function during pregnancy and post-partum.
decision to undergo pregnancy should be weighed
carefully with options for transplantation within the
window of childbearing age. Maternal indications for
initiating haemodialysis in pregnancy mirror indica-
tions outside of pregnancy and include fluid overload,
hyperkalaemia and acidosis that are resistant to medi-
cal management. However, the most common trigger
for dialysis initiation in pregnancy is the fetotoxicity
of urea, although the maternal serum urea threshold
at which dialysis should be commenced in pregnancy
continues to be a subject of ongoing debate. Historically,
the threshold of urea for dialysis initiation in pregnancy
was set at a concentration of 17–20 mmol/l owing to
the correlation between higher urea levels with deliv-
ery <32 weeks and birthweight <1,500 g, both of which
are recognized to adversely affect neonatal prognosis178.
However, given the positive correlation between hours
of haemo­dialy­sis provision and pregnancy outcomes,
the emerging consensus is that dialysis initiation should
be considered at lower levels of urea (15 mmol/l), par-
ticularly for women with progressive renal dysfunc-
tion, with dialysis provision matched to residual
renal function23.
Peritoneal dialysis. The incidence of pregnancy in
women on peritoneal dialysis is lower than for haemo-
dialysis175, which has been attributed to factors such as
high dialysate osmolality and fallopian tube injury sec-
ondary to peritoneal infection. Systematic review data
reveal a higher rate of small-for-gestational-age infants
in women undergoing peritoneal dialysis (67%) than in
those undergoing haemodialysis (31%), which could be
due to placental development being adversely affected
by peritoneal dialysis175. However, possible limitations
of cohort size and publication bias should be taken
into account.
Biomarkers in CKD and pre-eclampsia
In the past 5 years, prospective research has sought
to improve the diagnosis of superimposed pre-­
eclampsia in women with CKD. Diagnostic criteria
for pre-­e clampsia in the absence of CKD are based
on the development of de novo hypertension and pro-
teinuria after 20 weeks of gestation. Hence, the pres-
ence of these signs before pregnancy in women with
CKD complicates the diagnosis of superimposed pre-­
eclampsia. Moreover, hypertension and proteinuria can
also occur de novo in later pregnancy in the absence
of pre-eclampsia owing to gestation-related physio-
logical changes. As a result, the distinction between
physiological gestational change, CKD, lupus disease
flare and superimposed pre-eclampsia is challenging
(TABLE 2), leading to difficult decisions with regard to
delivery and iatrogenic prematurity. Superimposed
pre-eclampsia contributes to preterm delivery and
small-for-gestational-age infants among patients with
CKD. Heightened obstetric surveillance is therefore
warranted for all women with CKD to enable early
detection of impaired fetal growth and to assess the
umbilical artery through the use of Doppler ultrasound,
a  technique by which impending fetal distress is

16 | ADVANCE ONLINE PUBLICATION www.nature.com/nrneph

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

Doppler ultrasound predicted through detection of progressive changes in Conclusions

The use of sound waves to
detect movement. This
technique is used in pregnancy
to examine the vascular
waveform in uterine and
umbilical arteries to predict or
diagnose pathology.
waveform. In addition, placental insufficiency due to
pre-eclampsia can be predicted at 20 weeks gestation
through assessment of uterine artery vascular resist-
ance. One prospective cohort comprising 15 women
showed that elevated uterine artery Doppler pres-
sure index has 100% specificity for the prediction of
superimposed pre-eclampsia in women with CKD78,
but this finding requires further validation in a larger
cohort including women with advanced CKD. Another
study demonstrated that uterine and umbilical artery
Doppler ultrasound could be used to differentiate pre-­
eclampsia from CKD in the absence of pre-­eclampsia,
with normal uterine and umbilical waveforms pre-
dicting CKD, whereas abnormal waveforms were
­associated with pre-eclampsia179.
In terms of biomarkers, an increase in the antiangi-
ogenic marker sFLT1 and a decrease in the angiogenic
marker placenta growth factor (PLGF; also known as
PGF) have been demonstrated to be useful tools for the
prediction of pre-eclampsia180 and the need for delivery
within 2 weeks181. A similar pattern of biomarkers has
been reported to discriminate between women with
CKD with and without superimposed pre-eclampsia78,
including women with lupus nephritis182 and women on
dialysis183. However, larger cohort studies are required
to confirm the diagnostic utility of these markers in
pregnant women with CKD. Research into the thera-
peutic utility of removing antiangiogenic factors such as
sFLT1 from the circulation in pre-eclampsia currently
excludes women with underlying CKD184.
Fertility, family planning and pregnancy considerations
are of increasing importance to both nephrologists and
obstetricians owing to trends in women’s health and
renal disease. CKD in women of childbearing age con-
fers a risk of reduced fertility and adverse pregnancy out-
comes. Safe and effective contraceptive methods should
be available for all women of reproductive age with CKD
to allow pregnancy to be planned, with individualized and
detailed risk counselling undertaken before conception
(BOX 1). Anaemia, vitamin D deficiency and hypertension
are complicating comorbidities in pregnancy that may
develop as a consequence of underlying CKD and require
tailored surveillance and management. The prescription
of drugs in pregnancy is always a balance of risk versus
benefit. For many drugs used in CKD, the benefit of ther-
apy during pregnancy is greater than any associated risk,
provided known teratogens and fetotoxic drugs (RAAS
blockers, MMF and cyclophosphamide) are avoided. For
women requiring dialysis in pregnancy, increasing the
number of hours of haemodialysis ­provision is associated
with better outcomes.
Future research is required into the mechanisms and
management of impaired fertility in CKD, preferential
choices and optimal dosage of therapeutic agents for
the management of hypertension, and the pathophysiol-
ogy and prevention of progressive renal deterioration in
pregnancy. Furthermore, proteinuric thresholds for VTE
prophylaxis as well as vasoactive biomarker and Doppler
validation in the prediction and diagnosis of s­ uperimposed
pre-eclampsia warrant further investigation.

1. Piccoli, G. B. et al. Pregnancy and chronic kidney
disease: a challenge in all CKD stages. Clin. J. Am. Soc.
Nephrol. 5, 844–855 (2010).
2. Bramham, K. et al. Pregnancy in renal transplant
recipients: a UK national cohort study. Clin. J. Am.
Soc. Nephrol. 8, 290–298 (2013).
3. Piccoli, G. B. et al. Outcomes of pregnancies after
kidney transplantation: lessons learned from CKD.
A comparison of transplanted, nontransplanted
chronic kidney disease patients and low-risk
pregnancies: a multicenter nationwide analysis.
Transplantation 101, 2536–2544 (2017).
4. Webster, P., Lightstone, L., McKay, D. B. &
Josephson, M. A. Pregnancy in chronic kidney disease
and kidney transplantation. Kidney Int. 91,
1047–1056 (2017).
5. Hladunewich, M. A., Melamad, N. & Bramham, K.
Pregnancy across the spectrum of chronic kidney
disease. Kidney Int. 89, 995–1007 (2016).
6. Knight, M., Kurinczuk, J. J., Tuffnell, D. &
Brocklehurst, P. The UK Obstetric Surveillance System
for rare disorders of pregnancy. BJOG 112, 263–265
(2005).
7. [No authors listed.] Statistical bulletin: conceptions
in England and Wales 2015. Office for National
Statistics https://www.ons.gov.uk/
peoplepopulationandcommunity/
birthsdeathsandmarriages/conceptionandfertilityrates/
bulletins/conceptionstatistics/2015#main-points
(2017).
8. Piccoli, G. B. et al. The children of dialysis: live-born
babies from on‑dialysis mothers in Italy — an
epidemiological perspective comparing dialysis,
kidney transplantation and the overall population.
Nephrol. Dial. Transplant. 29, 1578–1586
(2014).
9. Basok, E. K. et al. Assessment of female sexual
function and quality of life in predialysis, peritoneal
dialysis, hemodialysis, and renal transplant patients.
Int. Urol. Nephrol. 41, 473–481 (2009).
10. Finkelstein, F. O., Shirani, S., Wuerth, D. &
Finkelstein, S. H. Therapy insight: sexual dysfunction
in patients with chronic kidney disease. Nat. Clin.
Pract. Nephrol. 3, 200–207 (2007).
11. Steele, T. E. et al. Sexual experience of the chronic
peritoneal dialysis patient. J. Am. Soc. Nephrol. 7,
1165–1168 (1996).
12. Solak, Y. et al. Effects of sildenafil and vardenafil
treatments on sleep quality and depression in
hemodialysis patients with erectile dysfunction.
Int.J. Impot. Res. 23, 27–31 (2011).
13. Holley, J. L., Schmidt, R. J., Bender, F. H., Dumler, F.
& Schiff, M. Gynecologic and reproductive issues in
women on dialysis. Am. J. Kidney Dis. 29, 685–690
(1997).
14. Lim, V. S., Henriquez, C., Sievertsen, G. &
Frohman, L. A. Ovarian function in chronic renal
failure: Evidence suggesting hypothalamic anovulation.
Ann. Intern. Med. 93, 21–27 (1980).
15. Wang, G. C. et al. Measurements of serum pituitary-
gonadal hormones and investigation of sexual and
reproductive functions in kidney transplant recipients.
Int. J. Nephrol. 2010, 612126 (2010).
16. Palmer, B. F. & Clegg, D. J. Gonadal dysfunction in
chronic kidney disease. Rev. Endocr. Metab. Disord.
18, 17–30 (2017).
17. Holley, J. L. & Schmidt, R. J. Changes in fertility and
hormone replacement therapy in kidney disease.
Adv. Chron. Kidney Dis. 20, 240–245 (2013).
18. Yavuz, D., Topçu, G., Ozene, r C., Akalin, S. & Sirikçi, O.
Macroprolactin does not contribute to elevated levels
of prolactin in patients on renal replacement therapy.
Clin. Endocrinol. 63, 520–524 (2005).
19. Weinhandl, E. D., Liu, J., Gilbertson, D. T.,
Arneson, T. J. & Collins, A. J. Survival in daily home
hemodialysis and matched thrice-weekly in‑center
hemodialysis patients. J. Am. Soc. Nephrol. 23,
895–904 (2012).
20. Rocco, M. V. et al. The effects of frequent nocturnal
home hemodialysis: the frequent hemodialysis network
nocturnal trial. Kidney Int. 80, 1080–1091 (2011).
21. Finkelstein, F. O. et al. At‑home short daily
hemodialysis improves the long-term health-related
quality of life. Kidney Int. 82, 561–569 (2012).
22. van Eps, C. et al. Changes in serum prolactin, sex
hormones and thyroid function with alternate nightly
nocturnal home haemodialysis. Nephrology 17,
42–47 (2012).
23. Hladunewich, M. & Schatell, D. Intensive dialysis
and pregnancy. Hemodial. Int. 20, 339–348
(2016).
24. Barua, M. et al. Successful pregnancies on nocturnal
home hemodialysis. Clin. J. Am. Soc. Nephrol. 3,
392–396 (2008).
25. Meistrich, M. L., Wilson, G., Brown, B. W.,
da Cunha, M. F. & Lipshultz, L. I. Impact of
cyclophosphamide on long-term reduction in sperm
count in men treated with combination chemotherapy
for ewing and soft tissue sarcomas. Cancer 70,
2703–2712 (1992).
26. Boumpas, D. T. et al. Risk for sustained amenorrhea in
patients with systemic lupus erythematosus receiving
intermittent pulse cyclophosphamide therapy.
Ann. Intern. Med. 119, 366–369 (1993).
27. Ioannidis, J. P., Katsifis, G. E., Tzioufas, A. G. &
Moutsopoulos, H. M. Predictors of sustained
amenorrhea from pulsed intravenous
cyclophosphamide in premenopausal women with
systemic lupus erythematosus. J. Rheumatol. 29,
2129–2135 (2002).
28. Bellver, J. & Pellicer, A. Ovarian stimulation for
ovulation induction and in vitro fertilization in
patients with systemic lupus erythematosus and
antiphospholipid syndrome. Fertil. Steril. 92,
1803–1810 (2009).
29. Elizur, S. E. et al. Fertility preservation treatment for
young women with autoimmune diseases facing
treatment with gonadotoxic agents. Rheumatology
47, 506–1509 (2008).
30. Somers, E. C., Marder, W., Christman, G. M.,
Ognenovski, V. & McCune, W. J. Use of a
gonadotropin-releasing hormone analog for
protection against premature ovarian failure
during cyclophosphamide therapy in women with
severe lupus. Arthritis Rheum. 52, 2761–2767
(2005).

NATURE REVIEWS | NEPHROLOGY ADVANCE ONLINE PUBLICATION | 17

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

31. Moore, H. C. et al. Goserelin for ovarian protection
during breast-cancer adjuvant chemotherapy.
N. Engl. J. Med. 372, 923–932 (2015).
32. Lambertini, M. et al. Ovarian suppression with
triptorelin during adjuvant breast cancer
chemotherapy and long-term ovarian function,
pregnancies, and disease-free survival: a randomized
clinical trial. JAMA 314, 2632–2640 (2015).
33. Lambertini, M. et al. Ovarian suppression using
luteinizing hormone-releasing hormone agonists
during chemotherapy to preserve ovarian function and
fertility of breast cancer patients: a meta-analysis of
randomized studies. Ann. Oncol. 26, 2408–2419
(2015).
34. Shah, P. S. et al. Intention to become pregnant and
low birth weight and preterm birth: a systematic
review. Maternal Child Health J. 15, 205–216 (2009).
35. Yildirim, Y. & Uslu, A. Pregnancy in patients with
previous successful renal transplantation.
Int. J. Gynaecol. Obstet. 90, 198–202 (2005).
36. Trussell, J. Contraceptive failure in the United States.
Contraception 83, 397–404 (2011).
37. Korver, T. et al. Maintenance of ovulation inhibition
with the 75‑microg desogestrel-only contraceptive
pill (Cerazette) after scheduled 12‑h delays in tablet
intake. Contraception 71, 8–13 (2005).
38. Estes, C. M. & Westhoff, C. Contraception for the
transplant patient. Semin. Perinatol. 31, 372–377
(2007).
39. Krajewski, C. M., Geetha, D. & Gomez-Lobo, V.
Contraceptive options for women with a history of
solid-organ transplantation. Transplantation 95,
1183–1186 (2013).
40. Morrison, C. S. et al. Is the intrauterine device
appropriate contraception for HIV‑1 infected women?
BJOG 108, 784–790 (2001).
41. Ramhendar, T. & Byrne, P. Use of the levonorgestrel-
releasing intrauterine system in renal transplant
recipients: a retrospective case review. Contraception
86, 288–289 (2012).
42. Brynhildsen, J. Combined hormonal contraceptives:
prescribing patterns, compliance, and benefits versus
risks. Ther. Adv. Drug Saf. 5, 201–213 (2014).
43. Davison, J. M. & Dunlop, W. Renal hemodynamics
and tubular function in normal human pregnancy.
Kidney Int. 18, 152–161 (1980).
44. Helal, I., Fick-Brosnahan, G. M., Reed-Gitomer, B.
& Schrier, R. W. Glomerular hyperfiltration:
definitions, mechanisms and clinical implications.
Nat. Rev. Nephrol. 8, 293–300 (2012).
45. Roberts, M., Lindheimer, M. D. & Davison, J. M.
Altered glomerular permselectivity to neutral dextrans
and heteroporous membrane modelling in human
pregnancy. Am. J. Physiol. 270, F338–F343 (1996).
46. Milne, J. E. C., Lindheimer, M. D. & Davison, J. M.
Glomerular heteroporous membrane modeling in third
trimester and postpartum before and during amino
acid infusion. Am. J. Physiol. Renal Physiol. 282,
F170–F175 (2002).
47. Conrad, K. P. & Davison, J. M. The renal circulation in
normal pregnancy and preeclampsia: is there a place
for relaxin? Am. J. Physiol. Renal Physiol. 306,
F1121–F1135 (2014).
48. Novak, J. et al. Relaxin is essential for renal
vasodilation during pregnancy in conscious rats.
J. Clin. Invest. 107, 1469–1475 (2001).
49. Jeyabalan, A. et al. Essential role for vascular
gelatinase activity in relaxin-induced renal
vasodilation, hyperfiltration, and reduced myogenic
reactivity of small arteries. Circ. Res. 93, 1249–1257
(2003).
50. Cadnapaphornchai, M. A. et al. Chronic NOS inhibition
reverses systemic vasodilation and glomerular
hyperfiltration in pregnancy. Am. J. Physiol. Renal
Physiol. 280, F592–F598 (2001).
51. Conrad, K. P. & Baker, V. L. Corpus luteal contribution
to maternal pregnancy physiology and outcomes in
assisted reproductive technologies. Am. J. Physiol.
Regul. Integr. Comp. Physiol. 304, R69–R72 (2013).
52. Odutayo, A. &, Hladunewich, M. Obstetric
nephrology: renal hemodynamic and metabolic
physiology in normal pregnancy. Clin. J. Am. Soc.
Nephrol. 7, 2073–2080 (2012).
53. Smith, M. C., Moran, P., Ward, M. K. & Davison, J. M.
Assessment of glomerular filtration rate during
pregnancy using the MDRD formula. BJOG 115,
109–112 (2008).
54. Alper, A. B. et al. Performance of estimated
glomerular filtration rate prediction equations in
preeclamptic patients. Am. J. Perinatol. 28, 425–430
(2010).
55. Kristensen, K. et al. Temporal changes of the plasma
levels of cystatin C, beta-trace protein,
beta2‑microglobulin, urate and creatinine during
pregnancy indicate continuous alterations in the renal
filtration process. Scand. J. Clin. Lab Invest. 67,
612–618 (2007).
56. Kristensen, K., Strevens, H., Lindström, V. & Grubb, A.
Wide-Swensson, D. Increased plasma levels of
beta2‑microglobulin, cystatin C and beta-trace protein
in term pregnancy are not due to utero-placental
production. Scand. J. Clin. Lab Invest. 68, 649–653
(2008).
57. Strevens, H., Wide-Swensson, D., Torffvit, O. &
Grubb, A. Serum cystatin C for assessment of
glomerular filtration rate in pregnant and non-
pregnant women. Indications of altered filtration
process in pregnancy. Scand. J. Clin. Lab Invest. 62,
141–147 (2002).
58. Bramham, K., Makanjuola, D., Hussein, W., Cafful, D.
& Shehata, H. Serum cystatin is not a marker of
glomerular filtration rate in pregnancy. Obstet. Med.
2, 121–122 (2009).
59. Larsson, A., Palm, M., Hansson, L. O. & Axelsson, O.
Cystatin C and modification of diet in renal disease
(MDRD) estimated glomerular filtration rate differ
during normal pregnancy. Acta Obstet. Gynecol.
Scand. 89, 939–944 (2010).
60. Piccoli, G. B. et al. Risk of adverse pregnancy
outcomes in women with CKD. J. Am. Soc. Nephrol.
26, 2011–2022 (2015).
61. Gambaro, G. et al. Increased urinary excretion of
glycosaminoglycans in pregnancy and in diabetes
mellitus: a protective factor against nephrolithiasis.
Nephron 50, 62–63 (1988).
62. Butler, E. L., Cox, S. M., Eberts, E. G. &
Cunningham, F. G. Symptomatic nephrolithiasis
complicating pregnancy. Obstet. Gynecol. 96,
753–756 (2000).
63. Williams, K. P. & Galerneau, F. The role of serum uric
acid as a prognostic indicator of the severity of
maternal and fetal complications in hypertensive
pregnancies. J. Obstet. Gynaecol. Can. 24, 628–632
(2002).
64. Verdonk, K., Visser, W., Van Den Meiracker, A. H. &
Danser, A. H. The renin-angiotensin-aldosterone
system in pre-eclampsia: the delicate balance between
good and bad. Clin. Sci. 126, 537–544 (2014).
65. Davison, J. M. & Lindheimer, M. D. Volume
homeostasis and osmoregulation in human pregnancy.
Baillieres Clin. Endocrinol. Metab. 3, 451–472 (1989).
66. [No authors listed.] Pregnancy and renal disease.
Lancet 306, 801–802 (1975).
67. Tong, A., Brown, M. A., Winkelmayer, W. C.,
Craig, J. C. & Jesudason, S. Perspectives on pregnancy
in women with CKD: a semistructured interview study.
Am. J. Kidney Dis. 66, 951–961 (2015).
68. Tong, A., Jesudason, S., Craig, J. C. &
Winkelmayer, W. C. Perspectives on pregnancy in
women with chronic kidney disease: systematic review
of qualitative studies. Nephrol. Dial. Transplant. 30,
652–661 (2015).
69. Knight, M. et al. Saving Lives, Improving Mothers’
Care — Surveillance of maternal deaths in the UK
2012–2014 and lessons learned to inform maternity
care from the UK and Ireland Confidential Enquiries
into Maternal Deaths and Morbidity 2009–2014.
(National Perinatal Epidemiology Unit, University of
Oxford, 2016).
70. Wiles, K. S. et al. Pre-pregnancy counselling in chronic
kidney disease: a retrospective analysis of nine years’
experience. BMC Nephrol. 16, 28 (2015).
71. Kendrick, J. et al. Kidney disease and maternal and
fetal outcomes in pregnancy. Am. J. Kidney Dis. 66,
55–59 (2015).
72. Zhang, J. J. et al. A systematic review and meta-
analysis of outcomes of pregnancy in CKD and CKD
outcomes in pregnancy. Clin. J. Am. Soc. Nephrol. 10,
1964–1978 (2015).
73. Moroni, G. et al. Fetal outcome and recommendations
of pregnancies in lupus nephritis in the 21st century.
A prospective multicenter study. J. Autoimmun. 74,
6–12 (2016).
74. Piccoli, G. B. et al. Maternal-foetal outcomes in
pregnant women with glomerulonephritides. Are all
glomerulonephritides alike in pregnancy?
J. Autoimmun. 79, 91–98 (2017).
75. Chapman, A. B., Johnson, A. M. & Gabow, P. A.
Pregnancy outcome and its relationship to progression
of renal failure in autosomal dominant polycystic
kidney disease. J. Am. Soc. Nephrol. 5, 1178–1185
(1994).
76. Jones, D. C. & Hayslett, J. P. Outcome of pregnancy in
women with moderate or severe renal insufficiency.
N. Engl. J. Med. 335, 226–232 (1996).
77. Imbasciati, E. et al. Pregnancy in CKD stages 3 to 5:
fetal and maternal outcomes. Am. J. Kidney Diseases.
49, 753–762 (2007).
78. Bramham, K. et al. Diagnostic and predictive
biomarkers for pre-eclampsia in patients with
established hypertension and chronic kidney disease.
Kidney Int. 89, 874–885 (2016).
79. Willey, C. J. et al. Prevalence of autosomal dominant
polycystic kidney disease in the European Union.
Nephrol. Dial. Transplant. 32, 1356–1363 (2017).
80. Wu, M. et al. Pregnancy outcomes in autosomal
dominant polycystic kidney disease: a case-control
study. J. Matern. Fetal Neonatal Med. 29, 807–812
(2016).
81. Jung, J. H. et al. Successful pregnancy in a patient with
autosomal dominant polycystic kidney disease on long-
term hemodialysis. J. Kor. Med. Sci. 29, 301–304
(2014).
82. Swift, O., Vilar, E., Rahman, B., Side, L. & Gale, D. P.
Attitudes in patients with autosomal dominant
polycystic kidney disease toward prenatal diagnosis
and preimplantation genetic diagnosis. Genet. Test.
Mol. Biomarkers 20, 741–746 (2016).
83. Blom, K., Odutayo, A., Bramham, K. &
Hladunewich, M. A. Pregnancy and glomerular
disease: a systematic review of the literature with
management guidelines. Clin. J. Am. Soc. Nephrol. 12,
1862–1872 (2017).
84. Jungers, P. et al. Chronic kidney disease and
pregnancy. Adv. Nephrol. Necker Hosp. 15, 103–141
(1986).
85. Abe, S., et al. The influence of antecedent renal
disease on pregnancy. Am. J. Obstet. Gynecol. 153,
508–514 (1985).
86. Packham, D., Whitworth, J. A., Fairley, K. F. &
Kincaid-Smith, P. Histological features of IgA
glomerulonephritis as predictors of pregnancy
outcome. Clin. Nephrol. 30, 22–26 (1988).
87. Limardo, M. et al. Pregnancy and progression of IgA
nephropathy: results of an Italian multicenter study.
Am. J. Kidney Dis. 56, 506–512 (2010).
88. Hoover, P. J. & Costenbader, K. H. Insights into the
epidemiology and management of lupus nephritis
from the US rheumatologist’s perspective. Kidney Int.
90, 487–492 (2016).
89. Imbasciati, E. et al. Pregnancy in women with
pre-existing lupus nephritis: predictors of fetal and
maternal outcome. Nephrol. Dial. Transplant. 24,
519–525 (2009).
90. Bramham, K. et al. Pregnancy outcomes in systemic
lupus erythematosus with and without previous
nephritis. J. Rheumatol. 38, 1906–1913 (2011).
91. Smyth, A. et al. A systematic review and meta-
analysis of pregnancy outcomes in patients
with systemic lupus erythematosus and lupus
nephritis. Clin. J. Am. Soc. Nephrol. 5, 2060–2068
(2010).
92. Chakravarty, E. F. et al. Factors that predict
prematurity and preeclampsia in pregnancies that
are complicated by systemic lupus erythematosus.
Am. J. Obstet. Gynecol. 192, 1897–1904 (2005).
93. Buyon, J. P. et al. Predictors of pregnancy outcomes in
patients with lupus: a cohort study. Ann. Intern. Med.
163, 153–163 (2015).
94. Moroni, G. et al. Maternal outcome in pregnant
women with lupus nephritis: a prospective multicenter
study. J. Autoimmun. 74, 194–200 (2016).
95. Royal College of Obstetricians and Gynaecologists.
Reducing the risk of venous thromboembolism in
pregnancy and the puerperium. Green Top
Guideline 37a https://www.rcog.org.uk/globalassets/
documents/guidelines/gtg-37a.pdf (2015).
96. Cimaz, R., Spence, D. L., Hornberger, L. &
Silverman, E. D. Incidence and spectrum of neonatal
lupus erythematosus: a prospective study of infants
born to mothers with anti‑Ro autoantibodies.
J. Pediatr. 142, 678–683 (2003).
97. Andreoli, L. et al. EULAR recommendations for
women’s health and the management of family
planning, assisted reproduction, pregnancy and
menopause in patients with systemic lupus
erythematosus and/or antiphospholipid syndrome.
Ann. Rheum. Dis. 6, 476–485 (2017).
98. Izmirly, P. M. et al. Maternal use of
hydroxychloroquine is associated with a reduced risk
of recurrent anti-SSA/Ro‑antibody-associated cardiac
manifestations of neonatal lupus. Circulation 126,
76–82 (2012).

18 | ADVANCE ONLINE PUBLICATION www.nature.com/nrneph

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

99. Klemetti, M. M. et al. Obstetric and perinatal outcome
in type 1 diabetes patients with diabetic nephropathy
during 1988–2011. Diabetologia 58, 678–686
(2015).
100. Piccoli, G. B. et al. Type 1 diabetes, diabetic
nephropathy, and pregnancy: a systematic review and
meta-study. Rev. Diabet. Stud. 10, 6–26 (2013).
101. Bell, R., Glinianaia, S. V., Tennant, P. W., Bilous, R. W.
& Rankin, J. Peri-conception hyperglycaemia and
nephropathy are associated with risk of congenital
anomaly in women with pre-existing diabetes: a
population-based cohort study. Diabetologia 55,
936–947 (2012).
102. Hod, M. et al. Diabetic nephropathy and pregnancy:
the effect of ACE inhibitors prior to pregnancy on
fetomaternal outcome. Nephrol. Dialysis Transplant.
10, 2328–2333 (1995).
103. Bar, J. et al. Pregnancy outcome in patients with
insulin dependent diabetes mellitus and diabetic
nephropathy treated with ACE inhibitors before
pregnancy. J. Pediatr. Endocrinol. Metab. 12,
659–666 (1999).
104. Nielsen, L. R., Damm, P. & Mathiesen, E. R. Improved
pregnancy outcome in type 1 diabetic women with
microalbuminuria or diabetic nephropathy: effect of
intensified antihypertensive therapy? Diabetes Care.
32, 38–44 (2009).
105. Bramham, K. Diabetic nephropathy and pregnancy.
Semin. Nephrol. 37, 362–369 (2017).
106. World Health Organization. WHO recommendations
on antenatal care for a positive pregnancy experience.
http://apps.who.int/iris/bitstream/10665/
250796/1/9789241549912-eng.pdf?ua=1
(2016).
107. Pavord, S. et al. UK guidelines on the management of
iron deficiency in pregnancy. Br. J. Haematol. 156,
588–600 (2012).
108. Steer, P., Alam, M. A., Wadsworth, J. & Welch, A.
Relation between maternal haemoglobin
concentration and birth weight in different ethnic
groups. BMJ 310, 489–491 (1995).
109. McMullin, M. F., White, R., Lappin, T., Reeves, J. &
MacKenzie, G. Haemoglobin during pregnancy:
relationship to erythropoietin and haematinic status.
Eur. J. Haematol. 71, 44–50 (2003).
110. Sienas, L., Wong, T., Collins, R. & Smith, J.
Contemporary uses of erythropoietin in pregnancy:
a literature review. Obstet. Gynecol. Surv. 68,
594–602 (2013).
111. Sanchez-Gonzalez, L. R. et al. Efficacy and safety of
adjuvant recombinant human erythropoietin and
ferrous sulfate as treatment for iron deficiency
anemia during the third trimester of pregnancy.
Eur. J. Obstet. Gynecol. Reprod. Biol. 205, 32–36
(2016).
112. de Benoist, B., McLean, E., Egll, I. & Cogswell, M.
Worldwide prevalence of anaemia 1993–2005.
WHO Global Database on Anaemia (World Health
Organization, 2008).
113. Gaweda, A. E. Markers of iron status in chronic
kidney disease. Hemodial. Int. 21 (Suppl. 1), S21–27
(2017).
114. Albaramki, J., Hodson, E. M., Craig, J. C. &
Webster, A. C. Parenteral versus oral iron therapy for
adults and children with chronic kidney disease.
Cochrane Database Syst Rev. 1, CD007857 (2012).
115. Tariq, N., Ayub, R., Khan, W. U., Ijaz, S. & Alam, A. Y.
Parenteral iron therapy in the treatment of iron
deficiency anemia during pregnancy: a randomized
controlled trial. J. Coll. Physicians Surg. Pak. 25,
193–197 (2015).
116. Kriplani, A. et al. Intravenous iron sucrose therapy
for moderate to severe anaemia in pregnancy.
Indian J. Med. Res. 138, 78–82 (2013).
117. Al, R. A. et al. Intravenous versus oral iron for
treatment of anemia in pregnancy: a randomized trial.
Obstet. Gynecol. 106, 1335–1340 (2005).
118. al‑Momen, A. K. et al. Intravenous iron sucrose
complex in the treatment of iron deficiency anemia
during pregnancy. Eur. J. Obstet. Gynecol. Reprod.
Biol. 69, 121–124 (1996).
119. Moll, R. &. Davis, B. Iron, vitamin B12 and folate.
Clin. Sci. 45, 198–203 (2017).
120. Maxwell, P. H. & Eckardt, K. U. HIF prolyl hydroxylase
inhibitors for the treatment of renal anaemia and
beyond. Nat. Rev. Nephrol. 12, 157–168 (2016).
121. Royal College of Obstetricians and Gynaecologists.
Vitamin D in Pregnancy. Scientific Impact Paper
Number 43 https://www.rcog.org.uk/globalassets/
documents/guidelines/scientific-impact-papers/
vitamin_d_sip43_june14.pdf (2014).
122. De‑Regil, L. M., Palacios, C., Lombardo, L. K. &
Peña-Rosas, J. P. Vitamin D supplementation for
women during pregnancy. Cochrane Database Syst.
Rev. 2, CD008873 (2016).
123. Palacios, C., De‑Regil, L. M., Lombardo, L. K. &
Peña-Rosas, J. P. Vitamin D supplementation during
pregnancy: updated meta-analysis on maternal
outcomes. J. Steroid Biochem. Mol. Biol. 164,
148–155 (2016).
124. World Health Organization. Vitamin D
supplementation in pregnancy. http://www.who.int/
nutrition/publications/micronutrients/guidelines/
vit_d_supp_pregnant_women/en/ (2012).
125. Tamblyn, J. A. et al. Dysregulation of maternal and
placental vitamin D metabolism in preeclampsia.
Placenta 50, 70–77 (2017).
126. Turner, M., Barré, P. E., Benjamin, A., Goltzman, D. &
Gascon-Barré, M. Does the maternal kidney contribute
to the increased circulating 1,25‑dihydroxyvitamin D
concentrations during pregnancy? Miner. Electrolyte
Metab. 14, 246–252 (1988).
127. Hewison, M., Zehnder, D., Chakraverty, R. &
Adams, J. S. Vitamin D and barrier function: a novel
role for extra-renal 1 alpha-hydroxylase. Mol. Cell
Endocrinol. 215, 31–38 (2004).
128. Bramham, K. et al. Chronic hypertension and
pregnancy outcomes: systematic review and meta-
analysis. BMJ 348, g2301 (2014).
129. Magee, L. A. et al. Less-tight versus tight control of
hypertension in pregnancy. N. Engl. J. Med. 372,
407–417 (2015).
130. Bramham, K., Hall, M., Lightstone, L. &
Nelson-Piercy, C. Renal Disease in Pregnancy 2nd edn
(Cambridge Univ. Press, in press).
131. Li, D. K., Yang, C., Andrade, S., Tavares, V. &
Ferber, J. R. Maternal exposure to angiotensin
converting enzyme inhibitors in the first trimester and
risk of malformations in offspring: a retrospective
cohort study. BMJ 343, d5931 (2011).
132. Bateman, B. T. et al. Angiotensin-converting enzyme
inhibitors and the risk of congenital malformations.
Obstet. Gynecol. 129, 174–184 (2017).
133. US National Library of Medicine. Drugs and Lactation
Database (LactMed). TOXNET https://toxnet.nlm.nih.
gov/newtoxnet/lactmed.htm (2017).
134. Nice, F. J. et al. Medications and breast-feeding: a
guide for pharmacists, pharmacy technicians, and
other healthcare professionals part II. J. Pharmacy
Technol. 20, 85–95 (2004).
135. Benediktsson, R., Calder, A. A., Edwards, C. R. &
Seckl, J. R. Placental 11 beta-hydroxysteroid
dehydrogenase: a key regulator of fetal glucocorticoid
exposure. Clin. Endocrinol. 46, 161–166 (1997).
136. van Runnard Heimel, P. J., Schobben, A. F.,
Huisjes, A. J., Franx, A. & Bruinse, H. W. The
transplacental passage of prednisolone in pregnancies
complicated by early-onset HELLP syndrome. Placenta
26, 842–845 (2005).
137. Laskin, C. A. et al. Prednisone and aspirin in women
with autoantibodies and unexplained recurrent fetal
loss. N. Engl. J. Med. 337, 148–153 (1997).
138. Park-Wyllie, L. et al. Birth defects after maternal
exposure to corticosteroids: prospective cohort study
and meta-analysis of epidemiological studies.
Teratology 62, 385–392 (2000).
139. Carmichael, S. L. et al. Maternal corticosteroid use
and orofacial clefts. Am. J. Obstet. Gynecol. 197,
585.e1–585.e7 (2007).
140. Källén, B. Maternal drug use and infant cleft lip/palate
with special reference to corticoids. Cleft Palate
Craniofac. J. 40, 624–628 (2003).
141. Czeizel, A. E. & Rockenbauer, M. Population-based
case-control study of teratogenic potential of
corticosteroids. Teratology 56, 335–340 (1997).
142. Tata, L. J. et al. Effect of maternal asthma,
exacerbations and asthma medication use on
congenital malformations in offspring: a UK
population-based study. Thorax 63, 981–987 (2008).
143. Hviid, A. & Mølgaard-Nielsen, D. Corticosteroid use
during pregnancy and risk of orofacial clefts. CMAJ
183, 796–804 (2011).
144. Norwood, F. et al. Myasthenia in pregnancy: best
practice guidelines from a UK multispecialty working
group. J. Neurol. Neurosurg. Psychiatry 85, 538–543
(2014).
145. Francella, A. et al. The safety of 6‑mercaptopurine
for childbearing patients with inflammatory bowel
disease: a retrospective cohort study.
Gastroenterology 124, 9–17 (2003).
146. lint, J. et al. BSR and BHPR guideline on prescribing
drugs in pregnancy and breastfeeding-part I: standard
and biologic disease modifying anti-rheumatic drugs
and corticosteroids. Rheumatology 55, 1693–1697
(2016).
147. Götestam Skorpen, C. et al. The EULAR points to
consider for use of antirheumatic drugs before
pregnancy, and during pregnancy and lactation.
Ann. Rheum. Dis. 75, 795–810 (2016).
148. Ostensen, M. et al. Pregnancy and reproduction in
autoimmune rheumatic diseases. Rheumatology 50,
657–664 (2011).
149. Sau, A. et al. Azathioprine and breastfeeding: is it
safe? BJOG 114, 498–501 (2007).
150. Sifontis, N. M. et al. Pregnancy outcomes in solid
organ transplant recipients with exposure to
mycophenolate mofetil or sirolimus. Transplantation
82, 1698–1702 (2006).
151. Perez-Aytes, A. et al. In utero exposure to
mycophenolate mofetil: a characteristic phenotype?
Am. J. Med. Genet. A 146A, 1–7 (2008).
152. Bar Oz, B., Hackman, R., Einarson, T. & Koren, G.
Pregnancy outcome after cyclosporine therapy during
pregnancy: a meta-analysis. Transplantation 71,
1051–1055 (2001).
153. Kainz, A., Harabacz, I., Cowlrick, I. S., Gadgil, S. D.
& Hagiwara, D. Review of the course and outcome of
100 pregnancies in 84 women treated with tacrolimus.
Transplantation. 70, 1718–1721 (2000).
154. Chakkera, H. A., Kudva, Y. & Kaplan, B. Calcineurin
inhibitors: pharmacologic mechanisms impacting both
insulin resistance and insulin secretion leading to
glucose dysregulation and diabetes mellitus.
Clin. Pharmacol. Ther. 101, 114–120 (2017).
155. Zheng, S. et al. Pharmacokinetics of tacrolimus
during pregnancy. Ther. Drug Monit. 34, 660–670
(2012).
156. Bramham, K., Chusney, G., Lee, J., Lightstone, L. &
Nelson-Piercy, C. Breastfeeding and tacrolimus: serial
monitoring in breast-fed and bottle-fed infants.
Clin. J. Am. Soc. Nephrol. 8, 563–567 (2013).
157. Wallace, D. J., Gudsoorkar, V. S., Weisman, M. H. &
Venuturupalli, S. R. New insights into mechanisms of
therapeutic effects of antimalarial agents in SLE.
Nat. Rev. Rheumatol. 8, 522–533 (2012).
158. Kaplan, Y. C., Ozsarfati, J., Nickel, C. & Koren, G.
Reproductive outcomes following hydroxychloroquine
use for autoimmune diseases: a systematic review and
meta-analysis. Br. J. Clin. Pharmacol. 81, 835–848
(2016).
159. Clowse, M.,E., Magder, L., Witter, F. & Petri, M.
Hydroxychloroquine in lupus pregnancy. Arthritis
Rheum. 54, 3640–3647 (2006).
160. Marmor, M. F., Kellner, U., Lai, T. Y., Melles, R. B. &
Mieler, W. F. American Academy of Ophthalmology
recommendations on screening for chloroquine and
hydroxychloroquine retinopathy (2016 revision).
Ophthalmology 123, 1386–1394 (2016).
161. Zemlickis, D. et al. Fetal outcome after in utero
exposure to cancer chemotherapy. Arch. Intern. Med.
152, 573–576 (1992).
162. United States Food and Drug Administration. Rituxan
final labeling text https://www.accessdata.fda.gov/
drugsatfda_docs/label/2010/103705s5311lbl.pdf
(2010).
163. Chakravarty, E. F., Murray, E. R., Kelman, A. &
Farmer, P. Pregnancy outcomes after maternal
exposure to rituximab. Blood 117, 1499–1506
(2011).
164. Bruel, A. et al. Hemolytic uremic syndrome in
pregnancy and postpartum. Clin. J. Am. Soc. Nephrol.
http://dx.doi.org/10.2215/CJN.00280117 (2017).
165. Kelly, R. J. et al. Eculizumab in pregnant patients with
paroxysmal nocturnal hemoglobinuria. N. Engl. J. Med.
373, 1032–1039 (2015).
166. National Institute for Health and Care Excellence.
Hypertension in pregnancy: the management of
hypertensive disorders in pregnancy. (RCOG Press,
2011).
167. Henderson, J. T. et al. Low-dose aspirin for the
prevention of morbidity and mortality from
preeclampsia: a systematic evidence review for the U.
S. Preventive Services Task Force. Ann. Intern. Med.
160, 695–703 (2014).
168. Rolnik, D. L. et al. Aspirin versus placebo in
pregnancies at high risk for preterm preeclampsia.
N. Engl. J. Med. 377, 613–622 (2017).
169. Poon, L. C. et al. Aspirin for evidence-based
preeclampsia prevention trial: effect of aspirin in
prevention of preterm preeclampsia in subgroups of
women according to their characteristics and medical
and obstetrical history. Am. J. Obstet. Gynecol. 217,
585.e1–585.e5 (2017).

NATURE REVIEWS | NEPHROLOGY ADVANCE ONLINE PUBLICATION | 19

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

170. Datta, P., Rewers-Felkins, K., Kallem, R. R.,
Baker, T. & Hale, T. W. Transfer of low dose aspirin
into human milk. J. Hum. Lact. 33, 296–299
(2017).
171. Edling, K. L. et al. A pregnant dilemma: primary
hyperparathyroidism due to parathyromatosis in
pregnancy. Endocr. Pract. 20, e14–e17 (2014).
172. Nadarasa, K. et al. The use of cinacalcet in pregnancy
to treat a complex case of parathyroid carcinoma.
Endocrinol. Diabetes Metab. Case Rep. 2014,
140056 (2014).
173. Hoeltzenbein, M., Stieler, K., Panse, M., Wacker, E.
& Schaefer, C. Allopurinol use during pregnancy —
outcome of 31 prospectively ascertained cases and
a phenotype possibly indicative for teratogenicity.
PLoS ONE 8, e66637 (2013).
174. Indraratna, P. L., Virk, S., Gurram, D. & Day, R. O.
Use of colchicine in pregnancy: a systematic review
and meta-analysis. Rheumatology http://dx.doi.org/
10.1093/rheumatology/kex353 (2017).
175. Piccoli, G. B. et al. Pregnancy in dialysis patients in
the new millennium: a systematic review and meta-
regression analysis correlating dialysis schedules and
pregnancy outcomes. Nephrol. Dial. Transplant. 31,
1915–1934 (2016).
176. Hou, S. H. Frequency and outcome of pregnancy in
women on dialysis. Am. J. Kidney Dis. 23, 60–63
(1994).
177. Hladunewich, M. A. et al. Intensive hemodialysis
associates with improved pregnancy outcomes:
a Canadian and United States cohort comparison.
J. Am. Soc. Nephrol. 25, 1103–1109 (2014).
178. Asamiya, Y. et al. The importance of low blood urea
nitrogen levels in pregnant patients undergoing
hemodialysis to optimize birth weight and gestational
age. Kidney Int. 75, 1217–1222 (2009).
179. Piccoli, G. B. et al. Pre-eclampsia or chronic kidney
disease? The flow hypothesis. Nephrol. Dial. Transplant.
28, 1199–1206 (2013).
180. Agrawal, S., Cerdeira, A., S., Redman, C. & Vatish, M.
Meta-analysis and systematic review to assess the role
of soluble fms-like tyrosine recpetor kinase and
placental growth factor ratio in prediction of
preeclampsia: the SaPPhirE study. Hypertension http://
dx.doi.org/10.1161/HYPERTENSIONAHA.117.10182
(2017).
181. Chappell, L. C. et al. Diagnostic accuracy of placental
growth factor in women with suspected preeclampsia:
a prospective multicenter study. Circulation 128,
2121–2131 (2013).
182. Kim, M. Y. et al. Angiogenic factor imbalance early in
pregnancy predicts adverse outcomes in patients with
lupus and antiphospholipid antibodies: results of the
PROMISSE study. Am. J. Obstet. Gynecol. 214, 108.
e1–108.e14 (2016).
183. Akbari, A. et al. Circulating angiogenic factors in
a pregnant woman on intensive hemodialysis:
a case report. Can. J. Kidney Health Dis. 3, 7 (2016).
184. Thadhani, R. et al. Removal of soluble fms-like
tyrosine kinase‑1 by dextran sulfate apheresis
in preeclampsia. J. Am. Soc. Nephrol. 27, 903–913
(2016).
185. Williams, D. & Davison, J. Chronic kidney disease in
pregnancy. BMJ 336, 211 (2008).
Acknowledgements
The authors acknowledge the US National Institute for Health
Research (NIHR) Rare Diseases Translational Research
Collaboration as well as the Biomedical Research Centre at
Guy’s and St. Thomas’ UK National Health Service (NHS)
Foundation Trust and King’s College London for funding K.W.
under the terms of a doctoral research fellowship. The views
expressed are those of the authors and not necessarily those
of the NHS, the NIHR or the UK Department of Health.
Author contributions
K.W., C.N.P. and K.B. contributed equally to the conception,
design, drafting and revision of this article.
Competing interests statement
The authors declare no competing interests.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
Reviewer information
Nature Reviews Nephrology thanks G. Lipkin and the other,
anonymous reviewers for their contribution to the peer review
of this work.

20 | ADVANCE ONLINE PUBLICATION www.nature.com/nrneph

©
2
0
1
8
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.