Original Investigation

Kidney Function After a Hypertensive Disorder

of Pregnancy: A Longitudinal Study
Nina D. Paauw, Anne Marijn van der Graaf, Rita Bozoglan, David P. van der Ham, Gerjan Navis,
Ron T. Gansevoort, Henk Groen, and A. Titia Lely

Background: Registry-based studies report an
increased risk for end-stage kidney disease after
hypertensive disorders of pregnancy (HDPs). It
is unclear whether HDPs lead to an increased
incidence of chronic kidney disease (CKD) and/
or progression of kidney function decline.
Study Design: Subanalysis of the Prevention of
Renal and Vascular Endstage Disease (PREVEND)
Study, a Dutch population-based cohort with
follow-up of 5 visits approximately 3 years apart.
Setting & Participants: Women without and with
patient-reported HDPs (non-HDP, n = 1,805; HDP,
n = 977) were identified. Mean age was 50 years at
baseline and median follow-up was 11 years.
Factor: An HDP.
Outcomes: (1) The incidence of CKD using Cox
regression and (2) the course of kidney function
(estimated glomerular filtration rate [eGFR] and
24-hour albuminuria) over 5 visits using generalized
estimating equation analysis adjusted for age,
mean arterial pressure, and renin-angiotensin
system (RAS) blockade. CKD was defined as
eGFR < 60 mL/min/1.73 m2 and/or 24-hour
albuminuria with albumin excretion > 30 mg, and
end-stage kidney disease was defined as
receiving dialysis or kidney transplantation.
Results: During follow-up, none of the women Complete author and article
developed end-stage renal disease and the information provided before
references.
incidence of CKD during follow-up was
similar across HDP groups (HR, 1.04; 95% CI, Correspondence to
0.79-1.37; P = 0.8). Use of RAS blockade was A.T. Lely (a.t.lely@
umcutrecht.nl)
higher after HDP at all visits. During a median
of 11 years, we observed a decrease in eGFR Am J Kidney Dis. 71(5):
619-626. Published online
in both groups, with a slightly steeper decline in
December 28, 2017.
the HDP group (98 ± 15 to 88 ± 16 vs 99 ± 17
to 91 ± 15 mL/min/1.73 m2; Pgroup < 0.01, doi: 10.1053/
Pgroup*visit < 0.05). The group effect remained j.ajkd.2017.10.014
significant after adjusting for mean arterial © 2017 Published by
pressure, but disappeared after adjusting for Elsevier Inc. on behalf of the
RAS blockade. The 24-hour albuminuria did not National Kidney Foundation,
Inc.
differ between groups.
Limitations: No obstetric records available.
HDPs defined by patient report rather than health
records.
Conclusions: HDPs did not detectably increase
the incidence of CKD. During follow-up, we
observed no differences in albuminuria, but
observed a marginally lower eGFR after HDP
that was no longer statistically significant after
adjusting for the use of RAS blockers. In this
population, we were unable to identify a
significant risk for kidney function decline after
patient-reported HDP.

end-stage kidney disease after preeclampsia.5,6 One of the

H ypertensive disorders of pregnancy (HDPs) occur in
up to 10% of all pregnancies and are a major cause of
maternal morbidity and mortality worldwide.1,2 The
spectrum of HDP ranges from the mild form of gestational
hypertension to the more severe form of preeclampsia,
Editorial, p. 601
which is clinically characterized by the presence of hy-
pertension in combination with proteinuria, other organ
disturbances, and/or intrauterine growth restriction in the
second half of pregnancy.3
In past decades, it has been shown that women with a
history of HDP have an increased risk for developing end-
stage kidney disease. Vikse et al4 showed that women who
had preeclampsia have a relative risk of 4.7 for developing
end-stage kidney disease during a mean follow-up of 17
years. They also observed that risk tripled in women with a
history of recurrent preeclampsia and in women after a
pregnancy complicated by low birth weight or premature
delivery.4 Later, 2 Taiwanese cohort studies (with overlap)
found hazard ratios (HRs) of 10.6 to 14 for developing
Taiwanese cohort studies found that not only women with
a history of preeclampsia, but also women with a history
of gestational hypertension have increased risk for devel-
oping end-stage kidney disease (HR 5.8).5
Due to the limited number of longitudinal follow-up
studies with kidney assessment in women with a history
of HDP, it is currently unknown whether the increased
incidence of end-stage kidney disease represents a pro-
gressive kidney function decline after HDP. Only 2 studies
have assessed whether a history of HDP increases the risk for
developing chronic kidney disease (CKD) later in life, of
which one found a lower incidence of CKD7 whereas the
other estimated increased risk for CKD after HDP.6 In addi-
tion, only a few studies have assessed kidney function,
mostly shortly after HDP, and reported either subtle alter-
ations or no differences in kidney function and kidney
hemodynamics.7-11 With regard to moderately increased
albuminuria after HDP, conflicting results are reported in the
literature; a meta-analysis of a few small studies reports
increased risk for albuminuria after preeclampsia,8 whereas
other smaller studies did not observe an increase.7,9,12

AJKD Vol 71 | Iss 5 | May 2018 619


To evaluate the incidence of CKD and end-stage kidney
disease and the course of kidney function after HDP in a
longitudinal setting, we used data from the Prevention of
Renal and Vascular End-stage Disease (PREVEND) Study,
which is a population-based observational cohort study of
the Dutch population with prospective data for kidney
function and albuminuria.13
Methods
PREVEND Study
The PREVEND Study is a prospective investigation of kid-
ney function, albuminuria, kidney disease, and cardio-
vascular disease in a large cohort drawn from the general
population. Details of the cohort are described else-
where.14,15 In summary, 6,000 individuals with urinary
albumin excretion ≥ 10 mg/L and 2,592 individuals with
urinary albumin excretion < 10 mg/L from the population
of Groningen, the Netherlands, aged 28 to 75 years were
enrolled. From the period 1997 to 2012, five screening
assessments took place about every 3 years. The PREVEND
Study was approved by the Medical Ethics Committee of
the University Medical Center Groningen (#96/01/22 on
February 22, 1996). Written informed consent was
obtained from all participants.
Selection of Participants
In this study we included 2,782 of 4,301 women
participating in the PREVEND cohort who answered the
question regarding hypertension during pregnancy at the
first visit (n = 4,267; Fig 1, flow chart). First, we excluded
women who answered “never been pregnant” (n = 1,096)
Female PREVEND parƟcipants
n=4301
Answer on quesƟon on HDP at first visit
n=4267
Excluded:
Never been pregnant (n=1096)
Don’t know (n=389)
Yes, had to keep bed rest (n=280)
No (n=1805)
Yes, allowed to do everything (n=697)
non-HDP HDP
n=1805 n=977
Figure 1. Flow chart of patient selection for the Prevention
of Renal and Vascular Endstage Disease (PREVEND) cohort.
Abbreviations: HDP, hypertensive disorder of pregnancy; n,
number of participants.
620
Original Investigation
and “don’t know” (n = 389). Then we classified women
who answered “no” on the question of whether they
experienced hypertension during their pregnancy as
“women with no patient-reported hypertensive pregnancy
disorder” (non-HDP; n = 1,805) and women who
answered “yes, allowed to do anything” or “yes, had to
keep bed rest” as “women with a patient-reported
hypertensive pregnancy disorder” (HDP; n = 977).
Data Collection
Data collection, entry, and validation of the PREVEND Study
was coordinated by the Trial Coordination Center of the
University Medical Center Groningen. The date of the first
PREVEND visit was taken as time point 0, after which me-
dian follow-up time at each visit was calculated (4, 6, 9, and
12 years). The questionnaire at the first visit provided in-
formation for birth date, race, employment, cardiovascular
risk factors, cardiovascular comorbid conditions, and end-
stage kidney disease at baseline. Cardiovascular comorbid
conditions were defined as chronic heart disease or history
of cerebrovascular accident. Alcohol use was defined as 2 to
7 glasses a day or more. At all visits, body mass index, sys-
tolic blood pressure, and diastolic blood pressure were
measured and blood and urine were collected. The 24-hour
urine was collected on 2 consecutive days and the mean of 2
albuminuria levels was calculated after using the formula:
urinary albumin concentration (mg/L) × urinary volume
(L) over 24 hours. Estimated glomerular filtration rate
(eGFR) was calculated with the CKD-EPI (CKD Epidemi-
ology Collaboration) creatinine–cystatin C equations, using
serum creatinine and serum cystatin C values.16 The pres-
ence of CKD was calculated for each visit and defined as
eGFR < 60 mL/min/1.73 m2 and/or 24-hour albuminuria
with albumin excretion > 30 mg.17 End-stage kidney dis-
ease during follow-up was assessed by linkage of the PRE-
VEND cohort to our national database for dialysis and kidney
transplantation. Measurements of urinary albumin, serum
creatinine, serum cystatin C, cholesterol, uric acid, tri-
glycerides, glucose, and insulin were determined from
blood samples as described.17-19 Homeostatic model
assessment index was calculated using the formula
(glucose × insulin)/22.5. To calculate the percentage of
women using antihypertensive medication, angiotensin-
converting enzyme (ACE) inhibitors, angiotensin receptor
blockers (ARBs), lipid-lowering medication, and antidia-
betic medication during follow-up, prescription data
from pharmacies were collected. Oral contraceptive use was
based on patient reporting.
Assessment Validity of Questionnaire
To assess the validity of patient-reported hypertension
during pregnancy, we obtained medical records of 204
women by linking the PREVEND database to the registry
of obstetric departments from 2 hospitals in the city of
Groningen (the University Medical Centre Groningen
and the Martini Hospital Groningen). We classified those
women into 3 groups based on criteria for gestational
AJKD Vol 71 | Iss 5 | May 2018
Original Investigation
hypertension and preeclampsia of the International Soci-
ety for the Study of Hypertension in Pregnancy3: a history
of normal pregnancy (control, n = 143), gestational hy-
pertension (n = 38), and preeclampsia (n = 23). Sensi-
tivity was calculated as the percentage of true hypertensive
women (gestational hypertension and preeclampsia) re-
ported as hypertensive in the questionnaire. Specificity
was calculated as percentage of true nonhypertensive
women (control) reported as no hypertension during
pregnancy.
Statistical Analysis
Data were analyzed using SPSS, version 22.0 (SPSS Inc),
and GraphPad prism, version 5.01 (GraphPad Software
Inc). Normally distributed data are presented as mean ±
standard deviation in text, tables, and figures and were
compared between the HDP and non-HDP groups using
t test. Skewed data are presented as median with inter-
quartile range, and the HDP and non-HDP groups were
compared using Mann-Whitney U test. For categorical
variables, we used χ 2 test. Women who reported kidney
disease requiring dialysis were excluded from the longi-
tudinal analysis. To examine the association between risk
for CKD and history of HDP, a Cox proportional hazard
regression model was performed to calculate HR and 95%
confidence interval (CI), with non-HDP as the reference
group. In this proportional hazard regression model,
person time was counted from the date of the first visit
until the date that CKD was diagnosed, or the date of the
last visit, with censoring in case of death or loss to
follow-up. Generalized estimating equation (GEE) anal-
ysis was performed for studying the effect of a history of
HDP or non-HDP (factor group) on mean arterial pres-
sure (MAP), eGFR, albuminuria, and use of antihyper-
tensive medication and renin-angiotensin system (RAS)
blockade (use of ACE inhibitor/ARB) during follow-up
(factor visit, categorical) with the interaction term
group*visit. GEE analysis was adjusted for age for all
variables. Identity link function was used for MAP, eGFR,
and albuminuria, and logit link function, for medication
use. The effect of adjustment for race, employment, and
smoking on the group effect was examined. Other vari-
ables that differed at baseline (body mass index and
laboratory results) were not adjusted for in GEE analysis
because these are assumed to be part of the causal
pathway between HDP and kidney function. Adjustment
for MAP and antihypertensive and ACE inhibitor/ARB use
was added to the GEE analysis for eGFR to see whether
blood pressure affected the group effect and group*visit
interaction. To investigate whether interaction between
group and ACE inhibitor/ARB use had an effect on eGFR
decline between groups, an interaction term group*ACE
inhibitor/ARB was added in the GEE analysis. An effect of
this interaction might be expected because hyper-
filtration, which is reported in formerly preeclamptic
women,10 influences the effect of ACE inhibitor/ARB
on eGFR.20 All analyses were performed using an
AJKD Vol 71 | Iss 5 | May 2018
exchangeable correlation matrix structure, assuming a
fixed correlation between measurements. In all analyses,
P < 0.05 indicated statistical significance.
Results
Baseline Characteristics
Table 1 shows characteristics of the selected non-HDP
(n = 1,805) and HDP (n = 977) women at the first PRE-
VEND visit. At the baseline visit, median age of women
with HDP was 50 years, which was slightly higher
compared to the non-HDP group (median age, 48 years;
P = 0.04). In both groups, median follow-up was 11 years.
Blood pressure was within the normotensive range in both
groups, but systolic and diastolic blood pressure were
significantly higher in the HDP group than in the non-HDP
group at baseline. Ethnicity, body mass index, working
status, smoking, oral contraceptive use, and cardiovascular
comorbid conditions were significantly different between
the non-HDP and HDP groups. Laboratory cardiovascular
risk markers, including glucose, insulin, cholesterol, uric
acid, and triglycerides, were all significantly higher in the
HDP group, and use of lipid-lowering medication, but not
glucose-lowering medication, was more frequent in the
HDP group.
Longitudinal Blood Pressure After HDP
At baseline and follow-up, blood pressure was significantly
higher in the HDP group compared to the non-HDP group
(Pgroup < 0.001) and remained stable during follow-up in
both groups (MAP data in Fig 2A; systolic and diastolic
blood pressure data in Fig S1). The percentage of antihy-
pertensive drug and ACE inhibitor/ARB use was higher at
baseline in the HDP group, and during follow-up, use
increased significantly in both groups (Pgroup < 0.001 for
both drug groups; Fig 2B and C). A steeper increase in
both antihypertensive drug and specifically ACE inhibitor/
ARB use over time was found in the HDP group
compared to the non-HDP group (Pgroup*visit = 0.03 and
Pgroup*visit < 0.001, respectively).
Longitudinal Kidney Function After HDP
We found that none of the women within our cohort
developed end-stage kidney disease during follow-up. At
the first visit, the prevalence of CKD was comparable, with
12.3% and 14.9% in the non-HDP and HDP groups,
respectively (P = 0.1). In addition, Cox regression analysis
showed that the percentage of women developing CKD
during follow-up was comparable in the groups (HR,
1.04; 95% CI, 0.79-1.37; P = 0.8). Incidence rates of CKD
after the first visit in the non-HDP and HDP groups were
13.2 and 13.6 per 1,000 person-years, respectively. For
kidney function, we found that at baseline and follow-up,
average eGFRs were within the normal range in
both groups. GEE analyses showed a group effect
(Pgroup = 0.007), with eGFR being slightly, but signifi-
cantly, lower in the HDP group at baseline and throughout
621
 Original Investigation

Table 1. Baseline Characteristics at First Visit of Non-HDP and HDP Groups

Non-HDP (n = 1,805a) HDP (n = 977a) P
General characteristics
Age, y 48 [39-59] 50 [42-59] 0.04b
Follow-up, y 10.9 [4.3-12.8] 10.8 [4.3-11.6] 0.3
White 1,700 (94.9%) 940 (97.2%) 0.004b
Employed 882 (49.8%) 435 (45.0%) 0.03b
Cardiovascular risk profile
BMI, kg/m2 25.7 ± 4.5 27.4 ± 5 <0.001b
SBP, mm Hg 122 ± 20.3 130.4 ± 21.6 <0.001b
DBP, mm Hg 70.2 ± 8.8 73.7 ± 9.0 <0.001b
Albuminuria > 10 mg/L 738 (41%) 437 (46%) 0.04b
Current smoker 641 (35.5%) 301 (30.8%) 0.012b
Alcohol use 802 (44.6%) 449 (46.1%) 0.4
Cardiovascular comorbid condition 42 (2.4%) 36 (3.8%) 0.028b
Kidney disease requiring dialysis 7 (0.4%) 1 (0.1%) 0.2
Laboratory results
Glucose, mmol/L 4.6 [4.2-5.0] 4.7 [4.3-5.1] <0.001b
Insulin, mmol/L 7.4 [5.2-10.7] 8.3 [5.9-12.5] <0.001b
HOMA 1.5 [1.0-2.3] 1.7 [1.2-2.8] <0.001b
Cholesterol, mmol/L 5.5 [4.8-6.4] 5.7 [5.0-6.5] 0.005b
Uric acid, mmol/L 0.25 [0.22-0.30] 0.27 [0.23-0.31] <0.001b
Triglycerides, mmol/L 1.06 [0.8-1.5] 1.13 [0.9-1.6] <0.001b
Medication use
Any antihypertensive 181 (11.5%) 209 (24.3%) <0.001b
ACE inhibitor/ARB 48 (3.2%) 68 (7.8%) <0.001b
Antidiabetic 18 (1.1%) 16 (1.8%) 0.1
Lipid-lowering 46 (2.9%) 48 (5.5%) 0.002b
Oral contraceptive 407 (24.2%) 182 (20.4%) 0.03b
Note: Values for categorical variables are given as number (percentage); values for continuous variables, as mean ± standard deviation or median [interquartile range].
Abbreviations and definitions: ACE, angiotensin-converting enzyme; alcohol use, ≥2 to 7 glasses daily; ARB, angiotensin receptor blocker; BMI, body mass index; DBP,
diastolic blood pressure; HDP, women with patient-reported hypertensive disorder of pregnancy; HOMA, Homeostatic Model Assessment Index; non-HDP, women
without patient-reported hypertensive disorder of pregnancy; SBP, systolic blood pressure.
a
Total number who participated at visit 1 of the study, not all variables were available for each participant at baseline.
b
P < 0.05

follow-up. The eGFR slopes of non-HDP and HDP diverged
during follow-up, with a significant group*visit effect
(Pgroup*visit = 0.05), indicating a significantly steeper
decline in the HDP group compared to the non-HDP group
(98 ± 15 to 88 ± 16 vs 99 ± 17 to 91 ± 15 mL/min/
1.73 m2; Fig 3A; Table S1). After adjusting for MAP, the
eGFR group and group*visit effect remained significant
(Pgroup = 0.001, Pgroup*visit = 0.04). When adding the
group*ACE inhibitor/ARB effect to the analysis, we found
that the group effect disappeared (Pgroup = 0.1). At all
visits, no significant difference in 24-hour albuminuria
was found between groups (Pgroup = 0.2; Fig 3B;
Table S1). Adjustment for other variables did not change
the interpretation of the group effect (Table S2).
Validity of Questionnaire
The overall sensitivity of patient-reported HDP in the
questionnaire of PREVEND within our case-control cohort
was 84%, with 51 of 61 women with a history of gesta-
tional hypertension and preeclampsia reporting HDP dur-
ing the visit after the index pregnancy. Sensitivity was
higher for a history of preeclampsia (96%) compared to a
history of gestational hypertension (76%). Specificity of
the questionnaire was 94%.
Discussion
In this study, we report longitudinal data for kidney
function in relation to a history of HDP. For our analysis,
we used data from the PREVEND Study, a population-
based cohort study that consisted of a maximum of 5
visits with a median of 11 years of follow-up. We confirm
that women with a history of HDP have an unfavorable
cardiovascular risk profile, with higher blood pressure at
all visits that is accompanied by increasingly higher use of
antihypertensive agents and RAS blockade. Despite these
profiles, we did not observe a higher incidence of CKD
after HDP at baseline and did not observe increased risk for
developing CKD during follow-up. In addition, none of
the women developed end-stage kidney disease. Evaluation
of the course of eGFRs showed minimal, but significant,
differences between groups with slightly lower eGFRs at all
visits and a slightly steeper eGFR decline in women with a
history of patient-reported HDP, which was not accom-
panied by higher urinary protein loss. The group effect

622 AJKD Vol 71 | Iss 5 | May 2018

Original Investigation

Figure 2. Longitudinal data for blood pressure and antihypertensive (anti-HTN) medication use in women without and with hyper-
tensive disorder of pregnancy (HDP). Mean ± standard error of the mean (SEM) of mean arterial pressure (MAP) (A) in women
without and with an HDP (non-HDP, grey circle; HDP, black square). The percentage prescription of anti-HTN (B) and
angiotensin-converting enzyme inhibitor (ACEi) and angiotensin receptor blocker (ARB) (C) in women without and with an HDP
(non-HDP, white bar; and HDP, black bar; respectively). *P < 0.05 at the respective visit, P for interaction (Pgroup*visit) and P for group
(Pgroup) in HDP versus non-HDP; all derived from the generalized estimating equation analysis adjusted for age.

disappeared after adjusting for ACE inhibitor/ARB use,
indicating that the increased use of RAS blockade in
women after HDP might explain the slightly lower eGFRs
observed after HDP. Our analyses indicate that in our
population with high RAS blockade use, we cannot
confirm a relevant decline in kidney function and kidney
risk after HDP.
By providing longitudinal eGFR measurements after
HDP, our study adds robust and valuable insights
regarding the long-term course of kidney function after
hypertensive pregnancies. Previous studies report on kid-
ney function in women with a history of HDP at different
set time points postpartum, mostly shortly after pre-
eclampsia, and show either subtle alterations or no dif-
ferences in kidney function and kidney hemodynamics.7-11
At 10 years postpartum, a high-normal eGFR is reported in
women with a history of preterm preeclampsia,9,21 which
is speculated to represent a state of hyperfiltration as
compensation for loss in kidney function leading to kidney
function loss in the long term. In the present study, we
observed a slightly lower eGFR instead of a high-normal
eGFR, which might be explained by the approximately
10-year older age at the start of follow-up, during which
hyperfiltration is not compensating for loss in kidney
function anymore. However, the minimal differences in
our groups in eGFRs over time suggest that our women in
the HDP group are not likely to have had early loss of
kidney function. Certainly, because we found no group
differences in eGFRs after adjusting for the use of ACE
inhibitor/ARB, the slightly lower eGFRs might be the
result of the increased use of RAS blockade in individuals
in the HDP group.22 To our knowledge, only one previous

AJKD Vol 71 | Iss 5 | May 2018 623

 Original Investigation

Figure 3. Longitudinal data for kidney function and albuminuria in women without and with a hypertensive disorder during pregnancy
(HDP). Mean ± standard error of the mean (SEM) of estimated glomerular filtration rate (eGFR) (A), median (25th-75th percentile) of
24-hour albuminuria levels (B) in women without and with an HDP (non-HDP, grey circle; HDP, black square, respectively). *P < 0.05
at the respective visit, P for group (Pgroup), and P for interaction (Pgroup*visit) in HDP versus non-HDP; all derived from the generalized
estimating equation analysis (eGFR unadjusted and albuminuria adjusted for age).

study investigated kidney function at 2 time points (1 and
14 years postpartum).23 That study by Spaan et al23 reports
lower kidney function shortly after preeclampsia but
similar eGFRs at 14 years postpartum, which would sug-
gest that there is no accelerated age-dependent kidney
function loss.23 Taken together, the study of Spaan et al23
and our study suggest that there is no clinically relevant
kidney function loss in the long term after HDP.
Although we found a slightly steeper decline in
eGFRs, we did not find increased risk for the develop-
ment of end-stage kidney disease and earlier stages of
CKD in this longitudinal cohort. In addition, we did not
find an increased rate of albuminuria, which is in line
with the earlier findings in small cohorts studying uri-
nary protein loss after HDP.7,9,12 Our findings might be
explained by the increased use of RAS blockade after
HDP, which may protect the kidney against accelerated
kidney loss and proteinuria and thereby prevent the
development of CKD and end-stage kidney disease.24-26
This is also suggested by the role of RAS blockade as a
significant determinant in the multivariate model for
eGFR. In addition, MAP was w8 mm Hg higher at
baseline in the HDP group compared to the non-HDP
group, which is in line with previous large
studies.27,28 High blood pressure might itself affect
kidney function. However, using GEE analysis, we could
not explain the differences in GFR slopes between
groups by adjusting for blood pressure. We speculate
that the kidney and cardiovascular system are equally
affected in women with HDP, but the disturbances in
cardiovascular system manifest earlier due to the large
reserve capacity of the kidney.29 If so, it might be that
treatment with RAS blockade ameliorates the increased
renal risk for developing CKD or end-stage kidney dis-
ease in these women in the long term.
Overall, our findings are very suggestive that there is no
relevant kidney function decline after HDP and thereby
they do not match the reported 4 to 10 times increased
end-stage kidney disease risk after HDP.4-6 It is important
to note that the actual risk for end-stage kidney disease
reported in those studies might be an effect of the setting
in which these studies were performed. One of the studies
reporting increased risk for end-stage kidney disease was
derived from an old cohort (diagnosis of end-stage kidney
disease before 1980), and their findings might be
explained by suboptimal pregnancy care with inadequate
diagnosis of preeclampsia and less adequate treatment of
hypertension and CKD.4 The other 2 studies are from
Taiwan,5,6 where the prevalence of end-stage kidney dis-
ease is the highest in the world,30 and therefore analysis is
probably confounded by a high prevalence of kidney
disease preconception, which in itself increases the risk for
developing both HDP and end-stage kidney disease.31-33
Unfortunately, we do not have data for preconception
kidney function to study the link with HDP and long-term
kidney function. Future studies should investigate whether
a subpopulation of women with a history of HDP are at
high risk for developing long-term kidney function decline
after HDP based on pre-existing factors or persistent pro-
teinuria postpartum.
A strength of our study is the size of the extensively
phenotyped cohort, which makes the PREVEND Study
uniquely appropriate for investigating kidney outcome. A
limitation of our longitudinal study is a lack of specific
obstetric data because we relied on patient reporting of
HDP. This prevented us from analyzing the effect of

624 AJKD Vol 71 | Iss 5 | May 2018

Original Investigation
pregnancy to follow-up interval, severity of the HDP, and
recurrent episodes of HDP on the course of kidney func-
tion and development of CKD. HDP represents both
women with gestational hypertension and preeclampsia,
with the incidence of gestational hypertension being
almost double that of preeclampsia, suggesting that our
results might reflect the situation after gestational hyper-
tension more than after preeclampsia. Using our case-
control cohort, we found that women with preeclampsia
were more likely to have answered “HDP with bedrest”
instead of “HDP allowed to do everything.” However,
separate analysis using “HDP with bedrest” as indicative of
preeclampsia would not be valid because almost half the
women with a history of gestational hypertension also
reported “HDP with bedrest.”
In our cohort, 35% of the women reported a history of
HDP, although HDP usually complicates only w10%
of pregnancies.1 When validating patient-reported history
of HDP, we found specificity of 94% and sensitivity of
84%, which is comparable with an earlier study34 and
suggests some influence of over-reporting because of recall
bias. In addition, slight overestimation of HDP incidence
might be the result of exclusion of the group of women
who answered “don’t know” on the question of whether
they have a history of pregnancy disorders. Even after
correction for these 2 factors, the prevalence remains still
higher than expected. We speculate that this might be
explained by our population being enriched with in-
dividuals with elevated urinary albumin excretion. This
might have resulted in over-representation of women
who were already at high risk for the development of
HDP due to preconception subclinical disturbances or
women whose kidneys have been (slightly) affected by
pregnancy complications such as HDP, although it remains
uncertain whether HDP affects the kidneys in such a way.
Concluding, this longitudinal study on kidney function
shows that women with a history of HDP have marginally
lower eGFRs in the long term, which was not accompanied
by higher urinary protein loss or higher incidence of CKD.
The increased use of RAS blockade in women after HDP
might explain the slightly lower eGFRs and steeper de-
clines in eGFRs after HDP and might additionally have
established renoprotective effects. Our results suggest that
a history of HDP has only a minor impact on kidney
function in later life and they are not in line with the
earlier reported increased risk for end-stage kidney disease
reported after HDP. The discrepancy between our findings
and earlier reports might be explained by the different
settings of the studies, in which earlier studies were per-
formed in populations with suboptimal (pre)pregnancy
care and/or a higher incidence of end-stage kidney
disease.
Supplementary Material
Figure S1: Longitudinal data on blood pressure.
Table S1: eGFR and albumin values at 5 visits.
AJKD Vol 71 | Iss 5 | May 2018
Table S2: Unadjusted and adjusted GEE analysis for MAP, eGFR,
and albuminuria.
Article Information
Authors’ Full Names and Academic Degrees: Nina D. Paauw, MD,
Anne Marijn van der Graaf, MD, PhD, Rita Bozoglan, BSc, David P.
van der Ham, MD, PhD, Gerjan Navis, MD, PhD, Ron T. Gansevoort,
MD, PhD, Henk Groen, MD, PhD, and A. Titia Lely, MD, PhD.
Authors’ Affiliations: Department of Obstetrics, Wilhelmina
Children’s Hospital Birth Center, University Medical Center
Utrecht, Utrecht (NDP, ATL); Departments of Pathology (AMvdG),
Medical Biology (AMvdG), and Obstetrics and Gynaecology
(AMvdG, RB), University Medical Centre Groningen; Department
of Obstetrics and Gynaecology, Martini Hospital (DPvdH);
Department of Nephrology, University Medical Centre Groningen
(GN, RTG); and Department of Epidemiology; University of
Groningen, Groningen, the Netherlands (HG).
Address for Correspondence: A. Titia Lely, MD, PhD, Department
of Obstetrics and Gynaecology, Home Post KE 04.123, Lundlaan
6, 3508 AB Utrecht, the Netherlands. E-mail: a.t.lely@umcutrecht.nl
Authors’ Contributions: Research idea and study design: ATL, HG,
GN; study coordination: RTG, ATL; data collection: AMvdG, RB,
DPvdH; data analysis: NDP, AMvdG; statistical support: HG. Each
author contributed important intellectual content during manuscript
drafting or revision and accepts accountability for the overall work
by ensuring that questions pertaining to the accuracy or integrity
of any portion of the work are appropriately investigated and
resolved.
Support: The PREVEND Study was supported by the Dutch Kidney
Foundation (grant E.033) and The Groningen University Medical
Center (Beleidsruimte), as well as Dade Behring, AusAm
biotechnologies, Roche, Abbott, and Gentian, which financed
laboratory equipment and reagents for various laboratory
determinations. This subanalysis was also supported by a ZonMw
Clinical Fellowship (project no. 40-000703-97-12463) to Dr Lely.
The funders had no role in interpretation and reporting of the data
or in the decision to submit the report for publication.
Financial Disclosure: The authors declare that they have no other
relevant financial interests.
Peer Review: Received May 23, 2017. Evaluated by 2 external peer
reviewers, with direct editorial input from a Statistics/Methods
Editor, an Associate Editor, and the Editor-in-Chief. Accepted in
revised form October 22, 2017.
References
1. Moussa HN, Arian SE, Sibai BM. Management of hypertensive
disorders in pregnancy. Womens Health (Lond). 2014;10(4):
385-404.
2. Magee LA, Pels A, Helewa M, Rey E, von Dadelszen P; Ca-
nadian Hypertensive Disorders of Pregnancy (HDP) Working
Group. Diagnosis, evaluation, and management of the hyper-
tensive disorders of pregnancy. Pregnancy Hypertens.
2014;4(2):105-145.
3. Tranquilli AL, Dekker G, Magee L, et al. The classification,
diagnosis and management of the hypertensive disorders of
pregnancy: a revised statement from the ISSHP. Pregnancy
Hypertens. 2014;4(2):97-104.
4. Vikse BE, Irgens LM, Leivestad T, Skjaerven R, Iversen BM.
Preeclampsia and the risk of end-stage renal disease. N Engl J
Med. 2008;359(8):800-809.
5. Wu CC, Chen SH, Ho CH, et al. End-stage renal disease after
hypertensive disorders in pregnancy. Am J Obstet Gynecol.
2014;210(2):147.e1-147.e8.
625

6. Wang IK, Muo CH, Chang YC, et al. Association between
hypertensive disorders during pregnancy and end-stage renal
disease: a population-based study. CMAJ. 2013;185(3):
207-213.
7. Mangos GJ, Spaan JJ, Pirabhahar S, Brown MA. Markers of
cardiovascular disease risk after hypertension in pregnancy.
J Hypertens. 2012;30(2):351-358.
8. McDonald SD, Han Z, Walsh MW, Gerstein HC, Devereaux PJ.
Kidney disease after preeclampsia: a systematic review and
meta-analysis. Am J Kidney Dis. 2010;55(6):1026-1039.
9. Sandvik MK, Hallan S, Svarstad E, Vikse BE. Preeclampsia and
prevalence of microalbuminuria 10 years later. Clin J Am Soc
Nephrol. 2013;8(7):1126-1134.
10. Toering TJ, van der Graaf AM, Visser FW, et al. Higher filtration
fraction in formerly early-onset preeclamptic women without
comorbidity. Am J Physiol Renal Physiol. 2015;308(8):
F824-F831.
11. van Beek E, Ekhart TH, Schiffers PM, van Eyck J, Peeters LL, de
Leeuw PW. Persistent abnormalities in plasma volume and
renal hemodynamics in patients with a history of preeclampsia.
Am J Obstet Gynecol. 1998;179(3, pt 1):690-696.
12. Berks D, Steegers EA, Molas M, Visser W. Resolution of hy-
pertension and proteinuria after preeclampsia. Obstet Gynecol.
2009;114(6):1307-1314.
13. Hillege HL, Fidler V, Diercks GF, et al. Prevention of Renal and
Vascular End Stage Disease (PREVEND) Study Group. Urinary
albumin excretion predicts cardiovascular and non-
cardiovascular mortality in general population. Circulation.
2002;106(14):1777-1782.
14. Hillege HL, Janssen WM, Bak AA, et al. Microalbuminuria is
common, also in a nondiabetic, nonhypertensive population,
and an independent indicator of cardiovascular risk factors and
cardiovascular morbidity. J Intern Med. 2001;249(6):519-526.
15. Joosten MM, Gansevoort RT, Mukamal KJ, et al. Urinary and
plasma magnesium and risk of ischemic heart disease. Am J
Clin Nutr. 2013;97(6):1299-1306.
16. Levey AS, Stevens LA, Schmid CH, et al. A new equation to
estimate glomerular filtration rate. Ann Intern Med.
2009;150(9):604-612.
17. Koning SH, Gansevoort RT, Mukamal KJ, et al. Alcohol con-
sumption is inversely associated with the risk of developing
chronic kidney disease. Kidney Int. 2015;87(5):1009-1016.
18. Scheven L, Joosten MM, de Jong PE, Bakker SJ,
Gansevoort RT; PREVEND Study Group. The association of
albuminuria with tubular reabsorption of uric acid: results from a
general population cohort. J Am Heart Assoc. 2014;3(2):
e000613.
19. Dullaart RP, Perton F, van der Klauw MM, Hillege HL,
Sluiter WJ; PREVEND Study Group. High plasma lecithin:
cholesterol acyltransferase activity does not predict low inci-
dence of cardiovascular events: possible attenuation of car-
dioprotection associated with high HDL cholesterol.
Atherosclerosis. 2010;208(2):537-542.
20. Naqvi SA, Ahsan S, Fawwad A, Basit A, Shera AS. Effect
of angiotensin converting enzyme inhibitor on glomerular
626
Original Investigation
hyperfiltration in patients with type 1 diabetes. Pak J Med Sci.
2016;32(3):559-564.
21. Paauw ND, Joles JA, Drost JT, et al. High-normal estimated
glomerular filtration rate in early-onset preeclamptic women 10
years postpartum. Hypertension. 2016;68(6):1407-1414.
22. Bakris GL, Weir MR. Angiotensin-converting enzyme inhibitor-
associated elevations in serum creatinine: is this a cause for
concern? Arch Intern Med. 2000;160(5):685-693.
23. Spaan JJ, Ekhart T, Spaanderman ME, Peeters LL. Renal
function after preeclampsia: a longitudinal pilot study. Nephron
Clin Pract. 2012;120(3):c156-c161.
24. Lewis EJ, Hunsicker LG, Bain RP, Rohde RD. The effect of
angiotensin-converting-enzyme inhibition on diabetic nephrop-
athy. The Collaborative Study Group. N Engl J Med.
1993;329(20):1456-1462.
25. Maschio G, Alberti D, Janin G, et al. Effect of the angiotensin-
converting-enzyme inhibitor benazepril on the progression of
chronic renal insufficiency. The Angiotensin-Converting-
Enzyme Inhibition in Progressive Renal Insufficiency Study
Group. N Engl J Med. 1996;334(15):939-945.
26. Randomised placebo-controlled trial of effect of ramipril on
decline in glomerular filtration rate and risk of terminal renal
failure in proteinuric, non-diabetic nephropathy. The GISEN
Group (Gruppo Italiano di Studi Epidemiologici in Nefrologia).
Lancet. 1997;349(9069):1857-1863.
27. Magnussen EB, Vatten LJ, Smith GD, Romundstad PR.
Hypertensive disorders in pregnancy and subsequently
measured cardiovascular risk factors. Obstet Gynecol.
2009;114(5):961-970.
28. Lykke JA, Langhoff-Roos J, Sibai BM, Funai EF, Triche EW,
Paidas MJ. Hypertensive pregnancy disorders and subsequent
cardiovascular morbidity and type 2 diabetes mellitus in the
mother. Hypertension. 2009;53(6):944-951.
29. Rook M, Bosma RJ, van Son WJ, et al. Nephrectomy elicits
impact of age and BMI on renal hemodynamics: lower post-
donation reserve capacity in older or overweight kidney donors.
Am J Transplant. 2008;8(10):2077-2085.
30. Hwang SJ, Tsai JC, Chen HC. Epidemiology, impact and pre-
ventive care of chronic kidney disease in Taiwan. Nephrology
(Carlton). 2010;15(suppl 2):3-9.
31. Fischer MJ, Lehnerz SD, Hebert JR, Parikh CR. Kidney disease
is an independent risk factor for adverse fetal and maternal
outcomes in pregnancy. Am J Kidney Dis. 2004;43(3):
415-423.
32. Nevis IF, Reitsma A, Dominic A, et al. Pregnancy outcomes in
women with chronic kidney disease: a systematic review. Clin J
Am Soc Nephrol. 2011;6(11):2587-2598.
33. Munkhaugen J, Lydersen S, Romundstad PR, Wideroe TE,
Vikse BE, Hallan S. Kidney function and future risk for adverse
pregnancy outcomes: a population-based study from HUNT II.
Norway. Nephrol Dial Transplant. 2009;24(120):3744-3750.
34. Coolman M, de Groot CJ, Jaddoe VW, Hofman A, Raat H,
Steegers EA. Medical record validation of maternally reported
history of preeclampsia. J Clin Epidemiol. 2010;63(8):
932-937.
AJKD Vol 71 | Iss 5 | May 2018