Long-Term Cognitive Impairment After
Preeclampsia
A Systematic Review and Meta-analysis
Malik Elharram, MD, Natalie Dayan, MD, MSc,
Amanpreet Kaur, MSc, Tara Landry, MLIS,
and Louise Pilote, MD, PhD
OBJECTIVE: To systematically review and
summarize studies investigating an association
between a history of preeclampsia and cognitive
function later in life.
DATA SOURCES: Studies published before August
2017 were identified without any language
restriction or study design limits through
electronic searches of 10 main databases
including MEDLINE and ClinicalTrials.gov.
METHODS OF STUDY SELECTION: We considered
all observational studies that included
preeclampsia as a clearly defined prespecified
risk factor and that examined a cognition-related
outcome measure including validated cognitive
tests, magnetic resonance brain imaging, or a
clinical diagnosis of dementia. Study quality was
assessed using the New-Castle Ottawa scale. All
review stages were conducted independently by
two reviewers, and disagreement was resolved
by a third reviewer. Where possible, data were
pooled using a random-effects model
TABULATION, INTEGRATION, AND RESULTS: Of
3,126 potentially relevant studies, 13 were
included in our review (1,314 women with prior
preeclampsia and 289,080 women with prior
normotensive pregnancy); median time since
pregnancy was 6 years. A higher number of self-
reported deficits in perception, memory, and
motor functioning on the Cognitive Failure
Questionnaire was reported in women with vs
without prior preeclamptic pregnancies
(Cognitive Failure Questionnaire mean total
score 41.5 vs 36.8 out of 100, weighted mean
difference of 25.1 points [29.4 to 20.8]). Our
metaanalysis did not reveal significant
differences in studies assessing attention (Digit
Symbol Substitution or Coding); however,
women with preeclampsia performed worse on
one of two meta-analyzed tests assessing
memory (Letter Number Sequencing mean total
score: 10.6 vs 10.1 out of 21, weighted mean
difference of 0.63 points 0.06–1.2). Pooling of
cognitive outcome measures for studies
assessing brain imaging or a clinical diagnosis of
Gangguan Kognitif Jangka Panjang Setelah
Preeklampsia
Tinjauan Sistematis dan Analisis-Meta
Malik Elharram, MD, Natalie Dayan, MD, MSc,
Amanpreet Kaur, MSc, Tara Landry, MLIS,
dan Louise Pilote, MD, PhD
TUJUAN: Untuk secara sistematis meninjau dan
meringkas penelitian yang menyelidiki hubungan
antara riwayat preeklamsia dan fungsi kognitif di
kemudian hari.
SUMBER DATA: Studi yang dipublikasikan
sebelum Agustus 2017 diidentifikasi tanpa
batasan bahasa atau mempelajari batas desain
melalui penelusuran elektronik dari 10 basis data
utama termasuk MEDLINE dan ClinicalTrials.gov.
METODE PENELITIAN STUDI: Kami menganggap
semua penelitian observasional yang termasuk
preeklamsia sebagai faktor risiko yang
ditentukan dengan jelas dan yang diperiksa
mengukur hasil yang berhubungan dengan
kognisi termasuk tes kognitif yang divalidasi,
pencitraan otak resonansi magnetik, atau
diagnosis klinis demensia. Kualitas studi dinilai
menggunakan skala New-Castle Ottawa. Semua
tahap tinjauan dilakukan secara independen oleh
dua pengulas, dan perselisihan diselesaikan oleh
peninjau ketiga. Jika memungkinkan, data
dikumpulkan menggunakan model efek-acak
TABULASI, INTEGRASI, DAN HASIL: Dari 3.126
penelitian yang berpotensi relevan, 13
dimasukkan dalam tinjauan kami (1.314 wanita
dengan preeklampsia sebelumnya dan 289.080
wanita dengan kehamilan normotensif
sebelumnya); waktu median sejak kehamilan
adalah 6 tahun. Jumlah yang lebih tinggi dari
defisit yang dilaporkan sendiri dalam persepsi,
memori, dan fungsi motorik pada Cognitive
Failure Questionnaire dilaporkan pada wanita
dengan vs tanpa kehamilan preeklamsi
sebelumnya (Cognitive Failure Questionnaire
berarti total skor 41,5 vs 36,8 dari 100,
perbedaan rata-rata berbobot dari 25,1 poin
[29,4 hingga 20,8]). Metaanalisis kami tidak
mengungkapkan perbedaan signifikan dalam
penelitian yang menilai perhatian (Digit Symbol
Substitution atau Coding); namun, wanita
dengan preeklamsia memiliki kinerja yang lebih
buruk pada salah satu dari dua tes meta-analisis
yang menilai memori (Angka Surat mengurutkan
skor total rata-rata: 10,6 vs 10,1 dari 21,
perbedaan rata-rata berbobot 0,63 poin 0,06–
dementia were limited by differences in 1,2). Penggabungan ukuran hasil kognitif untuk
reporting and marked heterogeneity between studi menilai pencitraan otak atau diagnosis
studies. klinis demensia dibatasi oleh perbedaan
pelaporan dan ditandai heterogenitas antara
studi.
CONCLUSION: Although preeclampsia is KESIMPULAN: Meskipun preeklampsia
associated with subjective cognitive symptoms, berhubungan dengan gejala kognitif subjektif,
our systematic review did not demonstrate clear tinjauan sistematis kami tidak menunjukkan
evidence of impairment on standard bukti penurunan nilai yang jelas pada tes
neurocognitive tests. There is a paucity of high neurokognitif standar. Ada kekurangan
quality studies assessing cognitive outcomes penelitian berkualitas tinggi yang menilai hasil
after preeclampsia. kognitif setelah preeklamsia.
Preeclampsia is a leading cause of maternal and Preeklampsia adalah penyebab utama
fetal morbidity and mortality, complicating 3–5% morbiditas dan mortalitas ibu dan janin,
of pregnancies worldwide.1 It is characterized by menyulitkan 3-5% kehamilan di seluruh dunia.1
an abnormal vascular response to placentation, Hal ini ditandai dengan respons vaskular yang
resulting in widespread endothelial dysfunction abnormal terhadap plasentasi, yang
in multiple organ systems including the brain.2 mengakibatkan disfungsi endotel luas pada
Delivery of the placenta remains the standard of beberapa sistem organ termasuk otak. plasenta
care for patients with preeclampsia; however, tetap standar perawatan untuk pasien dengan
years after pregnancy, long term effects of preeklamsia; Namun, bertahun-tahun setelah
endothelial dysfunction may persist,3,4 Formerly kehamilan, efek jangka panjang dari disfungsi
preeclamptic women have been shown to be at endotel dapat bertahan, 3,4 Wanita yang
an increased risk of hypertension, ischemic heart sebelumnya preeklampsia telah menunjukkan
disease, stroke, and premature cardiovascular peningkatan risiko hipertensi, penyakit jantung
mortality compared with women without a iskemik, stroke, dan kematian kardiovaskular
history of preeclampsia, years after pregnancy.5 prematur dibandingkan dengan wanita tanpa
Although these risk factors are common to riwayat preeklamsia, bertahun-tahun setelah
neurovascular conditions such as cognitive kehamilan. 5 Meskipun faktor-faktor risiko ini
impairment, the long-term consequences of umum untuk kondisi neurovaskular seperti
preeclampsia on cognitive function have been gangguan kognitif, konsekuensi jangka panjang
relatively understudied. The hypothesized dari preeklamsia pada fungsi kognitif relatif telah
mechanism of preeclampsia relates to dipelajari. Mekanisme hipotesis preeklamsia
widespread endothelial dysfunction berhubungan dengan disfungsi endotel luas yang
characterized by an imbalance in proangiogenic ditandai oleh ketidakseimbangan dalam
and antiangiogenic proteins.6,7 Furthermore, proangiogenik dan protein antiangiogenik.6,7
former preeclamptic women (elapsed time 0.25– Selanjutnya, mantan wanita preeklampsia
6 years) have been shown to have dampened (waktu yang berlalu 0,25-6 tahun) telah terbukti
responses to visually evoked cerebral blood flow memiliki respons yang diredam untuk respon
responses on transcranial Doppler aliran darah otak visual yang ditimbulkan. pada
ultrasonography.8 We hypothesized that these ultrasonografi Doppler transkranial.8 Kami
changes may affect cerebral hemodynamic berhipotesis bahwa perubahan ini dapat
function, manifested clinically through cognitive mempengaruhi fungsi hemodinamik otak, yang
impairment. Formerly preeclamptic women dimanifestasikan secara klinis melalui gangguan
report cognitive slowing months to years after kognitif. Sebelumnya wanita preeklampsia
pregnancy.9,10 However, objective evidence of melaporkan bulan memperlambat kognitif
cognitive impairment from validated hingga bertahun-tahun setelah kehamilan.9,10
neurocognitive tests has shown conflicting Namun, bukti objektif dari gangguan kognitif dari
results, with some studies showing worsening of tes neurokognitif yang divalidasi telah
cognitive performance,4,11 whereas others have menunjukkan hasil yang bertentangan, dengan
shown no difference in cognitive performance beberapa penelitian menunjukkan
when compared with women with prior
normotensive pregnancies.12 Furthermore,
studies evaluating neuroimaging results after
preeclampsia have also reported conflicting
results with select studies reporting an increased
frequency and severity of white matter lesions in
formerly preeclamptic women compared with
patients with prior normotensive
pregnancy13,14 and other studies reporting no
such difference.15 Studies in elderly patients
have shown white matter lesions to be
associated with a higher risk for ischemic
strokes, cognitive decline, and dementia,16 yet
the clinical implications of these white matter
lesions in young preeclamptic women remain
largely unknown
Because there is uncertainty about the long-term
association between preeclampsia and cognitive
impairment, we carried out a systematic review
of studies that compared objective and validated
cognitive measures in women with and without
prior preeclampsia. If a clear signal of impaired
cognitive performance is found in formerly
preeclamptic women, our study may indicate a
potential role for serial screening of cognitive
function in addition to routine cardiovascular
surveillance after delivery. SOURCES The review
was conducted using a predefined protocol in
accordance with Preferred Reporting Items for
Systematic Reviews and Meta-analysis
guidelines.17 Studies published before August
2017 were identified without any language
restriction or study design limits through
electronic searches of MEDLINE via Ovid
(including Epub ahead of print), registered trials
on Clinicaltrials.gov, PubMed, EMBASE via Ovid,
Biosis via Ovid, Global health via Ovid, Popline,
the Cochrane Central Register of Controlled
Trials, the Cochrane Database of Systematic
Reviews, and Web of Science. The MEDLINE
search strategy was developed by a librarian
experienced in systematic review searching (T.L.)
and peer-reviewed by another librarian using the
peer review of electronic search strategy
standard. The Full MEDLINE strategy (Appendix
1, available online at http://links.lww.
memburuknya kinerja kognitif, 4,11 sedangkan
yang lain tidak menunjukkan perbedaan kinerja
kognitif bila dibandingkan dengan wanita dengan
kehamilan normotensif sebelumnya.
Selanjutnya, penelitian yang mengevaluasi hasil
neuroimaging setelah preeklampsia juga telah
melaporkan hasil yang bertentangan dengan
penelitian terpilih yang melaporkan peningkatan
frekuensi dan keparahan lesi materi putih pada
wanita preeklampsia sebelumnya dibandingkan
dengan pasien dengan kehamilan normotensif
sebelumnya13, 14 dan penelitian lain yang
melaporkan tidak ada perbedaan tersebut. Studi
pada pasien usia lanjut telah menunjukkan lesi
materi putih untuk dikaitkan dengan risiko yang
lebih tinggi untuk stroke iskemik, penurunan
kognitif, dan demensia, 16 namun implikasi klinis
lesi materi putih ini pada wanita preeklamsi
muda. tetap besar tidak diketahui
Karena ada ketidakpastian tentang hubungan
jangka panjang antara preeklampsia dan
gangguan kognitif, kami melakukan tinjauan
sistematis terhadap studi yang membandingkan
pengukuran kognitif objektif dan valid pada
wanita dengan dan tanpa preeklampsia
sebelumnya. Jika sinyal yang jelas dari gangguan
kinerja kognitif ditemukan pada wanita yang
sebelumnya preeklampsia, penelitian kami
mungkin menunjukkan peran potensial untuk
skrining serial fungsi kognitif selain pengawasan
kardiovaskular rutin setelah melahirkan.
SUMBER Tinjauan dilakukan menggunakan
protokol yang telah ditentukan sesuai dengan
Item Pelaporan yang Lebih Baik untuk Tinjauan
Sistematis dan Pedoman Meta-analisis.17
Penelitian yang diterbitkan sebelum Agustus
2017 diidentifikasi tanpa batasan bahasa atau
membatasi desain penelitian melalui pencarian
elektronik MEDLINE melalui Ovid (termasuk
Epub sebelum cetak), uji coba terdaftar di
Clinicaltrials.gov, PubMed, EMBASE via Ovid,
Biosis via Ovid, Kesehatan global via Ovid,
Popline, Cochrane Central Register of Controlled
Trials, Cochrane Database of Systematic Reviews,
dan Web of Science. Strategi pencarian MEDLINE
dikembangkan oleh pustakawan yang
berpengalaman dalam pencarian tinjauan
sistematis (T.L.) dan peer-review oleh
pustakawan lain menggunakan peer review
com/AOG/B105) was applied to all databases
with modifications to search terms as necessary.
Further studies (n51) were identified in Web of
Science and SCOPUS by carrying out citation
searches for studies that cited our included
studies as well as by examining their reference
lists.
STUDY SELECTION We considered all
observational studies that reported
preeclampsia as a clearly defined prespecified
risk factor and that examined any of the
following neurocognitive outcomes: brain
imaging using magnetic resonance, clinical
diagnosis of dementia, or a battery of validated
neurocognitive tests, or all of these (Appendix 2,
available online at http://links.lww.
com/AOG/B105). Women of any parity, age, or
severity of preeclampsia were included. We
excluded case reports, case series, and studies
that examined eclampsia alone or where
cognitive outcome data on patients with
preeclampsia could not be separated from
patients with eclampsia. The reason we excluded
patients with eclampsia is because of the well-
known findings of cerebral edema on magnetic
resonance imaging (MRI) of patients with
eclampsia and the association of eclampsia with
long-term subjective cognitive impairment as
reported in several studies.9,18 We also
excluded studies that did not report a
comparison group of women without prior
preeclampsia. Two independent reviewers (M.E.,
A.K.) performed the study selection using
specific inclusion criteria to ensure accuracy and
reproducibility. The first screening was based on
titles and abstracts of identified publications. All
potentially relevant studies were retrieved for
full-text evaluation. Both reviewers
independently evaluated full-text articles and
reasons for exclusion were recorded.
Disagreement was resolved by a third reviewer
(L.P.). If duplicate studies were found within the
same data source, either the most recent or
most complete publication was selected. The
study selection process is displayed in Appendix
3, available online at http://links.lww.com/
standar strategi pencarian elektronik. Strategi
MEDLINE Lengkap (Lampiran 1, tersedia online di
http: //links.lww. Com / AOG / B105) telah
diterapkan ke semua basis data dengan
modifikasi istilah pencarian seperlunya.
Penelitian lebih lanjut (n51) diidentifikasi dalam
Web of Science dan SCOPUS dengan melakukan
pencarian kutipan untuk studi yang mengutip
studi termasuk kami serta dengan memeriksa
daftar referensi mereka.
PEMILIHAN STUDI Kami menganggap semua
penelitian observasional yang melaporkan
preeklamsia sebagai faktor risiko prespecified
jelas didefinisikan dan yang memeriksa salah
satu hasil neurokognitif berikut: pencitraan otak
menggunakan resonansi magnetik, diagnosis
klinis demensia, atau baterai tes neurokognitif
divalidasi, atau semua ini (Lampiran 2, tersedia
online di http: //links.lww. Com / AOG / B105).
Wanita dengan paritas, usia, atau keparahan
preeklampsia dimasukkan. Kami mengeluarkan
laporan kasus, seri kasus, dan studi yang
memeriksa eklampsia saja atau di mana data
hasil kognitif pada pasien dengan preeklamsia
tidak dapat dipisahkan dari pasien dengan
eklampsia. Alasan kami mengeluarkan pasien
dengan eklampsia adalah karena temuan edema
otak pada pencitraan resonansi magnetik (MRI)
pasien dengan eklampsia dan hubungan
eklamsia dengan gangguan kognitif subjektif
jangka panjang seperti yang dilaporkan dalam
beberapa penelitian.9,18 Kami juga
mengeluarkan studi yang tidak melaporkan
kelompok perbandingan wanita tanpa
preeklampsia sebelumnya. Dua peninjau
independen (M.E., A.K.) melakukan pemilihan
studi menggunakan kriteria inklusi spesifik untuk
memastikan akurasi dan reproduktifitas.
Penyaringan pertama didasarkan pada judul dan
abstrak publikasi yang teridentifikasi. Semua
studi yang berpotensi relevan diambil untuk
evaluasi teks lengkap. Kedua peninjau secara
independen mengevaluasi artikel teks lengkap
dan alasan pengecualian dicatat.
Ketidaksepakatan diselesaikan oleh peninjau
ketiga (L.P.). Jika studi duplikat ditemukan dalam
sumber data yang sama, baik publikasi paling
baru atau paling lengkap dipilih. Proses
pemilihan studi ditampilkan dalam Lampiran 3,
AOG/B105.
Within the included studies, women underwent
a variety of standardized neurocognitive tests,
which are each described in detail in Appendix 2
(http:// links.lww.com/AOG/B105). Briefly, the
tests included were designed to assess
subjective cognitive symptoms (Cognitive Failure
Questionnaire)19,20; attention, processing
speed, and visuospatial skills (Digit Symbol
Substitution or Coding test,21 memory [Letter
Number Sequencing22], and Rey Auditory Verbal
Learning Test).23 Using standardized data
extraction sheets, two investigators (M.E. and
A.K.) independently extracted data on several
study characteristics for all studies that met
inclusion criteria including study design, location,
sampling strategy, population, exposure and
outcome measurements, participant
characteristics, duration of follow-up, any
adjustment in the analysis, and type of effect
measure. For studies examining cognitive
function using neurocognitive tests, the mean or
median score of the respective cognitive tests
was recorded. Authors reporting insufficient
study details or results were contacted and
allotted 3 weeks of time for a response. The
Newcastle-Ottawa Scale was used to assess the
quality of observational studies. This scale uses a
“star” system to assess the quality of a study in
three domains: selection of participants,
comparability of study groups, and
ascertainment of outcomes of interest. Case–
control and cohort studies were evaluated out of
a total score of 9, and cross-sectional studies
were evaluated out of a total score of 10 using a
modified Newcastle-Ottawa Scale designed for
crosssectional studies.24 A study with a high risk
of bias was defined as a score of 0–4, medium
risk of bias was defined as a score of 4–6, and a
low risk of bias was defined as a score from 7-9
out of 10.
No study was excluded based on quality alone.
The mean scores and standard deviation (SD) of
the neurocognitive test measures, relative risk
for developing white matter lesions on MRI, and
the relative risk of developing dementia for
normotensive and preeclamptic patients was
tersedia online di http://links.lww.com/ AOG /
B105.
Dalam studi yang termasuk, wanita menjalani
berbagai tes neurokognitif standar, yang masing-
masing dijelaskan secara rinci dalam Lampiran 2
(http: // links.lww.com/AOG/B105). Secara
singkat, tes termasuk dirancang untuk menilai
gejala kognitif subjektif (Cognitive Failure
Questionnaire) 19,20; perhatian, kecepatan
pemrosesan, dan keterampilan visuospatial
(Digit Symbol Substitution atau tes Coding, 21
memori [Nomor Surat Sequencing22], dan Rey
Auditory Verbal Learning Test) .23 Menggunakan
lembar ekstraksi data standar, dua peneliti (ME
dan AK) secara independen mengekstraksi data
pada beberapa karakteristik studi untuk semua
studi yang memenuhi kriteria inklusi termasuk
desain studi, lokasi, strategi sampling, populasi,
paparan dan pengukuran hasil, karakteristik
peserta, durasi tindak lanjut, penyesuaian dalam
analisis, dan jenis ukuran efek. Untuk penelitian
yang memeriksa fungsi kognitif menggunakan
tes neurokognitif, skor rata-rata atau median
dari tes kognitif masing-masing dicatat. Penulis
melaporkan rincian atau hasil studi yang tidak
cukup telah dihubungi dan diberikan waktu 3
minggu untuk tanggapan. Skala Newcastle-
Ottawa digunakan untuk menilai kualitas studi
observasional. Skala ini menggunakan sistem
"bintang" untuk menilai kualitas penelitian
dalam tiga domain: pemilihan peserta,
komparabilitas kelompok belajar, dan penetapan
hasil yang menarik. Studi case-control dan
kohort dievaluasi dari total skor 9, dan studi
cross-sectional dievaluasi dari total skor 10
menggunakan Skala Newcastle-Ottawa yang
dirancang untuk studi crosssectional. Sebuah
studi dengan risiko tinggi Bias didefinisikan
sebagai skor 0–4, risiko bias sedang didefinisikan
sebagai skor 4-6, dan risiko bias rendah
didefinisikan sebagai skor 7-9 dari 10.
Tidak ada penelitian yang dikeluarkan
berdasarkan kualitas saja. Nilai rata-rata dan
standar deviasi (SD) dari pengukuran tes
neurokognitif, risiko relatif untuk
mengembangkan lesi materi putih pada MRI, dan
risiko relatif mengembangkan demensia untuk
pooled using a randomeffects model if there
were at least three studies using the same
outcome measure. Results for neurocognitive
tests were pooled only if they used the same
neurocognitive test in women with the same
exposure (ie, preeclampsia). A mean score and a
SD for neurocognitive tests were estimated for
studies reporting a median, range, and sample
size using the method by Hozo et al.25 Results of
the meta-analysis were presented as a weighted
mean difference between former preeclamptic
and normotensive pregnant patients with the
corresponding 95% CI. We used the I2 statistic to
quantify the degree of heterogeneity among
trials in each meta-analysis. The choice between
fixed- and random-effects meta-analysis has
been performed on the basis of the I2 index,
where I2 greater than 25% led us to conduct
random-effects meta-analyses based on linear
(mixed-effects) models with inverse-variance
weights.26 Studies were pooled only if the I2
was below 75%. A funnel plot for publication
bias was planned for summaries of cognitive
measures with at least 10 pooled studies.
Analysis was conducted using SAS version 9.4.
RESULTS In the initial literature search, 3,126
studies were identified, of which 32 were
selected for full-text review (Appendix 3,
http://links.lww.com/AOG/ B105).
Fifteen studies were included for the final
review, after which one study was excluded
during data entry because it was not possible to
separate patients with preeclampsia from those
with eclampsia27 (author contacted), and
another study was excluded because it used the
same cohort of patients analyzed in a previously
included study28 (author contacted). The
characteristics of the 13 remaining studies,
comprising a total of 1,314 (range 4-419) women
with prior preeclampsia and 289,080 (range 4–
283,902) women with prior normotensive
pregnancy, are represented in Appendix 4,
available online at
http://links.lww.com/AOG/B105. Seven of the
included studies examined cognitive function
using neurocognitive tests,4,9,12,15,18 four
studies examined MRI,13–15,29and three
pasien normotensif dan preeklampsia
dikumpulkan menggunakan model
randomeffects jika setidaknya ada tiga penelitian
menggunakan ukuran hasil yang sama. Hasil
untuk tes neurokognitif dikumpulkan hanya jika
mereka menggunakan tes neurokognitif yang
sama pada wanita dengan paparan yang sama
(yaitu, preeklampsia). Skor rata-rata dan SD
untuk tes neurokognitif diperkirakan untuk
penelitian melaporkan median, kisaran, dan
ukuran sampel menggunakan metode oleh Hozo
et al. 25 Hasil meta-analisis disajikan sebagai
perbedaan rata-rata tertimbang antara mantan
hamil preeclamptic dan normotensive pasien
dengan CI 95% yang sesuai. Kami menggunakan
statistik I2 untuk mengukur tingkat
heterogenitas di antara uji coba dalam setiap
meta-analisis. Pilihan antara meta analisis tetap
dan efek acak telah dilakukan atas dasar indeks
I2, di mana I2 lebih besar dari 25% mendorong
kita untuk melakukan meta-analisis efek-acak
berdasarkan pada model linear (efek campuran)
dengan invers bobot -variance.26 Studi
dikumpulkan hanya jika I2 di bawah 75%. Sebuah
plot saluran untuk bias publikasi direncanakan
untuk ringkasan pengukuran kognitif dengan
setidaknya 10 penelitian gabungan. Analisis
dilakukan menggunakan SAS versi 9.4.
HASIL Dalam pencarian literatur awal, 3.126
studi diidentifikasi, 32 diantaranya dipilih untuk
peninjauan teks lengkap (Lampiran 3,
http://links.lww.com/AOG/ B105).
Lima belas penelitian dimasukkan untuk
peninjauan akhir, setelah itu satu studi
dikeluarkan selama entri data karena tidak
mungkin untuk memisahkan pasien dengan
preeklamsia dari mereka dengan eklampsia27
(penulis dihubungi), dan penelitian lain
dikeluarkan karena menggunakan kohort pasien
yang sama. dianalisis dalam studi yang
sebelumnya termasuk28 (penulis dihubungi).
Karakteristik dari 13 studi yang tersisa, yang
terdiri dari total 1.314 (kisaran 4-419) wanita
dengan preeklampsia sebelumnya dan 289.080
(kisaran 4–283,902) wanita dengan kehamilan
normotensif sebelumnya, diwakili dalam
Lampiran 4, tersedia online di http: //
links.lww.com/AOG/B105. Tujuh dari studi
termasuk memeriksa fungsi kognitif
menggunakan tes neurokognitif, 4,9,12,15,18
studies examined a clinical diagnosis of
dementia.30–32 The most commonly used
cognitive tests were the Letter Number
Sequencing, Digit Symbol Substitution or Coding,
Rey Auditory Verbal Learning Test, and Cognitive
Failure Questionnaire. Across seven studies,
there were 26 neurocognitive tests that were
evaluated in 527 (range 10–208), formerly
preeclamptic and 1,430 (range 10–959) formerly
normotensive pregnancies. The median time
between the index preeclamptic pregnancy and
the measure of cognitive function was 5 years
(range 3 months to 35 years). Eight main
cognitive domains were studied. Because several
neurocognitive tests examined multiple cognitive
domains, the tests were divided based on the
principal domain as determined by the authors,
indicated in Appendix 2 (http://links.
lww.com/AOG/B105). Of the seven studies
reporting data on neurocognitive tests, only four
studies used the same cognitive tests, which
allowed for collective pooling of the results.
The Rey Auditory Verbal Learning Test (total
score out of 75),9,11,12 and the Letter Number
Sequencing (score out of 21) (three
studies)9,11,12 were used to assess memory
across 398 patients (202 women with
preeclampsia and 196 women with
normotensive pregnancy. Women with a prior
normotensive pregnancy had better memory
function as assessed by the Rey Auditory Verbal
Learning Test and Letter Number Sequencing
compared with those with a history of
preeclampsia; however, significant
heterogeneity precluded any pooled estimate of
the results (mean Rey Auditory Verbal Learning
Test score: 53.8 vs 47.6 correct responses out of
75, I2592%; Letter Number Sequencing mean
total score: 10.6 vs 10.1 out of 21, weighted
mean difference of 0.63 points [0.063–1.2],
I250%) (Figs. 1 and 2). Other tests used to
examine memory, including the Corsi block-
tapping test,9 incidental learning task,15
California Verbal Learning Test,4 logical memory
I and II,12 visual reproduction I and II,12 and the
Location Learning Test,9 did not reveal any
significant differences between patients with
and without a history of preeclampsia. The Digit
empat studi meneliti MRI, 13-15,29 dan tiga
penelitian memeriksa diagnosis klinis
dementia.30–32 Tes kognitif yang paling sering
digunakan adalah Pengurutan Nomor Surat,
Penggantian Simbol Digit atau Coding, Uji
Pembelajaran Verbal Rey Auditory, dan
Kuesioner Kegagalan Kognitif. Di tujuh studi, ada
26 tes neurokognitif yang dievaluasi pada 527
(kisaran 10-208), sebelumnya preeklampsia dan
1.430 (kisaran 10-959) sebelumnya kehamilan
normotensif. Median waktu antara indeks
kehamilan preeklampsia dan ukuran fungsi
kognitif adalah 5 tahun (rentang 3 bulan hingga
35 tahun). Delapan domain kognitif utama
dipelajari. Karena beberapa tes neurokognitif
memeriksa beberapa domain kognitif, tes dibagi
berdasarkan domain utama sebagaimana
ditentukan oleh penulis, yang ditunjukkan dalam
Lampiran 2 (http: // links. Lww.com/AOG/B105).
Dari tujuh studi yang melaporkan data pada tes
neurokognitif, hanya empat penelitian yang
menggunakan tes kognitif yang sama, yang
memungkinkan pengumpulan kolektif hasil.
The Rey Auditory Verbal Learning Test (total
score out of 75),9,11,12 and the Letter Number
Sequencing (score out of 21) (three
studies)9,11,12 were used to assess memory
across 398 patients (202 women with
preeclampsia and 196 women with
normotensive pregnancy. Women with a prior
normotensive pregnancy had better memory
function as assessed by the Rey Auditory Verbal
Learning Test and Letter Number Sequencing
compared with those with a history of
preeclampsia; however, significant
heterogeneity precluded any pooled estimate of
the results (mean Rey Auditory Verbal Learning
Test score: 53.8 vs 47.6 correct responses out of
75, I2592%; Letter Number Sequencing mean
total score: 10.6 vs 10.1 out of 21, weighted
mean difference of 0.63 points [0.063–1.2],
I250%) (Figs. 1 and 2). Other tests used to
examine memory, including the Corsi block-
tapping test,9 incidental learning task,15
California Verbal Learning Test,4 logical memory
I and II,12 visual reproduction I and II,12 and the
Location Learning Test,9 did not reveal any
significant differences between patients with
and without a history of preeclampsia. The Digit
Symbol Substitution or Coding Test (total score
achieved in 90 seconds)4,9,12 was used as a
measure of attention across 225 patients (118
women with preeclampsia and 107 with
normotensive pregnancy). Within this group,
pooled results did not demonstrate any
significant difference between patients with and
without prior preeclampsia (mean Digit Symbol
Substitution or Coding Test score 63.0 vs 67.3
correct responses in 90 seconds, weighted mean
difference of 2.27 correct responses [20.91 to
5.47], I 2552.1%) (Fig. 3). The Stroop I and II
test,9,11 digit span task,9,11,12 and the Trail
Making Test parts 1–39 used to assess attention
demonstrated a trend toward worse
performance in women with preeclampsia;
however, the results were limited by a small
patient population. Results of neurocognitive
tests used to examine visual spatial ability
including the block design,11,12 clock design,11
picture completion,12 and incomplete figures
test9 were not significantly different between
patients with and without a history of
preeclampsia.
Similarly, the semantic fluency,9,11,12,15
phonologic fluency,9,12,15 similarities (Wechsler
Adult Intelligence Scale IV),11 and the Boston
Naming Test,11 used to examine language, were
not significantly different between patients with
and without a history of preeclampsia. Results of
neurocognitive tests used to examine motor
speed including the Grooved Pegboard Test,4,9
Trail Making Test part 5,9 and Reaction Time
Test4 were not significantly different between
patients with and without a history of
preeclampsia. Information processing as
measured using Paced Auditory Serial Addition
Test revealed a significantly lower number of
correct answers in patients with compared with
those without a history of preeclampsia in one
study (mean total correct answers: 37.6 [9.1] vs
44.0 [7.0], in 160 patients).4 The digit symbol
search was not significantly different between
patients with and without a history of
preeclampsia.9 The score on the Stroop part
III,9,11 Trail Making Test part B,9,11,12 figure
fluency test,9 and the tower test9 did not reveal
any significant difference between patients with
and without a history of preeclampsia. The Mini
Symbol Substitution or Coding Test (total score
achieved in 90 seconds)4,9,12 was used as a
measure of attention across 225 patients (118
women with preeclampsia and 107 with
normotensive pregnancy). Within this group,
pooled results did not demonstrate any
significant difference between patients with and
without prior preeclampsia (mean Digit Symbol
Substitution or Coding Test score 63.0 vs 67.3
correct responses in 90 seconds, weighted mean
difference of 2.27 correct responses [20.91 to
5.47], I 2552.1%) (Fig. 3). The Stroop I and II
test,9,11 digit span task,9,11,12 and the Trail
Making Test parts 1–39 used to assess attention
demonstrated a trend toward worse
performance in women with preeclampsia;
however, the results were limited by a small
patient population. Results of neurocognitive
tests used to examine visual spatial ability
including the block design,11,12 clock design,11
picture completion,12 and incomplete figures
test9 were not significantly different between
patients with and without a history of
preeclampsia.
Similarly, the semantic fluency,9,11,12,15
phonologic fluency,9,12,15 similarities (Wechsler
Adult Intelligence Scale IV),11 and the Boston
Naming Test,11 used to examine language, were
not significantly different between patients with
and without a history of preeclampsia. Results of
neurocognitive tests used to examine motor
speed including the Grooved Pegboard Test,4,9
Trail Making Test part 5,9 and Reaction Time
Test4 were not significantly different between
patients with and without a history of
preeclampsia. Information processing as
measured using Paced Auditory Serial Addition
Test revealed a significantly lower number of
correct answers in patients with compared with
those without a history of preeclampsia in one
study (mean total correct answers: 37.6 [9.1] vs
44.0 [7.0], in 160 patients).4 The digit symbol
search was not significantly different between
patients with and without a history of
preeclampsia.9 The score on the Stroop part
III,9,11 Trail Making Test part B,9,11,12 figure
fluency test,9 and the tower test9 did not reveal
any significant difference between patients with
and without a history of preeclampsia. The Mini
Mental State Examination15 and the national
Dutch reading test9,11 used to examine global
cognitive performance were not significantly
different between women with and without a
history of preeclampsia. Subjective cognitive
impairment was assessed using the Cognitive
Failure Questionnaire (score out of 100) across
330 patients (148 women in the case group and
182 in the control group in three studies).4,9,18
Pooled results demonstrated overall higher
subjective deficit in memory, perception, and
motor functioning in women with prior
preeclamptic pregnancies compared with
normotensive pregnancies (Cognitive Failure
Questionnaire mean total score 41.5 vs 36.8 out
of 100, weighted mean difference of 25.12
points [29.4 to 20.87], I2557.5%) (Fig. 4). In a
separate study, women with recent
preeclampsia admitted to a greater number of
“mental problems” (67% vs 29%) and loss of
concentration (67% vs between the studies
(Appendix 6, available online at
http://links.lww.com/AOG/B105). Quality
assessment is presented in Appendix 4
(http://links.lww.com/AOG/B105). Six studies
had a medium risk of bias (4–6 stars), and seven
studies had a low risk of bias (7–9 stars). None of
the studies included were at high risk of bias.
Within our review, 5 of 13 studies did not adjust
for any confounders in their analysis. Of the four
pooled studies examining neurocognitive tests,
two studies had a low risk of bias and two had a
medium risk of bias.
DISCUSSION In this systematic review, we
summarized available evidence on the
association between preeclampsia and long-
term cognitive function as measured through
neurocognitive tests, MRI, and a clinical
diagnosis of dementia. Our hypothesis was that
long-term vascular changes after preeclampsia,3
which contribute to the development of
hypertension, ischemic heart disease, and
stroke,5 might also manifest with subtle but
clinically perceptible cognitive deficits. Among
the 13 studies included, there was significant
between-study heterogeneity in participant age,
sample size, clinical definition of preeclampsia,
length of time between pregnancy and cognitive
outcome measurement, and cognitive tests
Mental State Examination15 and the national
Dutch reading test9,11 used to examine global
cognitive performance were not significantly
different between women with and without a
history of preeclampsia. Subjective cognitive
impairment was assessed using the Cognitive
Failure Questionnaire (score out of 100) across
330 patients (148 women in the case group and
182 in the control group in three studies).4,9,18
Pooled results demonstrated overall higher
subjective deficit in memory, perception, and
motor functioning in women with prior
preeclamptic pregnancies compared with
normotensive pregnancies (Cognitive Failure
Questionnaire mean total score 41.5 vs 36.8 out
of 100, weighted mean difference of 25.12
points [29.4 to 20.87], I2557.5%) (Fig. 4). In a
separate study, women with recent
preeclampsia admitted to a greater number of
“mental problems” (67% vs 29%) and loss of
concentration (67% vs between the studies
(Appendix 6, available online at
http://links.lww.com/AOG/B105). Quality
assessment is presented in Appendix 4
(http://links.lww.com/AOG/B105). Enam
penelitian memiliki risiko bias menengah (4-6
bintang), dan tujuh penelitian memiliki risiko
bias rendah (7-9 bintang). Tak satu pun dari studi
termasuk berisiko tinggi bias. Dalam ulasan kami,
5 dari 13 penelitian tidak menyesuaikan untuk
pembaur dalam analisis mereka. Dari empat
penelitian yang dikumpulkan yang menguji tes
neurokognitif, dua penelitian memiliki risiko bias
rendah dan dua studi memiliki risiko bias sedang.
DISKUSI Dalam tinjauan sistematis ini, kami
meringkas bukti yang tersedia tentang hubungan
antara preeklampsia dan fungsi kognitif jangka
panjang yang diukur melalui tes neurokognitif,
MRI, dan diagnosis klinis demensia. Hipotesis
kami adalah bahwa perubahan vaskular jangka
panjang setelah preeklampsia, 3 yang
berkontribusi pada pengembangan hipertensi,
penyakit jantung iskemik, dan stroke, 5 mungkin
juga bermanifestasi dengan defisit kognitif yang
halus tetapi secara klinis terlihat. Di antara 13
studi yang termasuk, ada signifikan
heterogenitas antar-penelitian di usia peserta,
ukuran sampel, definisi klinis preeklampsia,
lamanya waktu antara kehamilan dan
pengukuran hasil kognitif, dan tes kognitif yang
used, which precluded summary assessments in
all cognitive domains. Thus, although our meta-
analysis demonstrates a small but significant
association between preeclampsia and self-
reported cognitive impairment, there are
insufficient data to conclude about the presence
or absence of subtle objective cognitive changes
in memory, attention, or executive function.
Preeclampsia did appear to be correlated with
the presence and severity of white matter
lesions, particularly in the frontal lobe. There
was no clear association between preeclampsia
and a clinical diagnosis of dementia in the three
studies that assessed this outcome. Women with
a history of preeclampsia reported subjective
losses in perception, memory, and motor
function more often than women with
normotensive pregnancy. This significant finding
was mainly driven by the study by Postma et al,9
which had the highest sample size of women (51
with prior preeclampsia, n548 control
participants). The authors of this study also
adjusted for age in their analysis and used a
strict definition of preeclampsia. A similar trend
was found in the other three studies. Although
women performed slightly worse on the Letter
Number Sequencing test assessing memory, this
finding did not persist in other neurocognitive
tests assessing memory (Rey Auditory Verbal
Learning Test). There are a number of
explanations for the lack of a clear association in
objective cognitive outcomes in our systematic
review despite finding consistently lower
subjective cognitive scores. First, the studies
within our review had a relatively short follow-
up between preeclampsia and the measure of
cognitive function using neurocognitive tests
(median 5 years) with the majority of studies
reporting a follow-up less than 10 years.
Consequently, women in our review were
relatively young at the time of assessment. It is
possible that more clinically apparent cognitive
issues might become apparent only with time as
the neurovascular changes and white matter
disease progress. Furthermore, individual studies
may have been underpowered to detect subtle
but significant differences in neurocognitive
tests. Indeed, statistical trends observed in small
studies12,33 may well translate to significant
effects if repeated with larger sample sizes.
Alternatively, our findings of self-reported
digunakan, yang menghalangi penilaian
ringkasan di semua domain kognitif. Dengan
demikian, meskipun meta-analisis kami
menunjukkan hubungan yang kecil namun
signifikan antara preeklampsia dan gangguan
kognitif yang dilaporkan sendiri, ada data yang
tidak cukup untuk menyimpulkan tentang ada
atau tidak adanya perubahan kognitif objektif
yang objektif dalam memori, perhatian, atau
fungsi eksekutif. Preeklampsia tampaknya
berkorelasi dengan keberadaan dan keparahan
lesi materi putih, terutama di lobus frontal. Tidak
ada hubungan yang jelas antara preeklampsia
dan diagnosis klinis demensia dalam tiga
penelitian yang menilai hasil ini. Wanita dengan
riwayat preeklampsia melaporkan kehilangan
subyektif dalam persepsi, memori, dan fungsi
motorik lebih sering daripada wanita dengan
kehamilan normotensif. Temuan yang signifikan
ini terutama didorong oleh penelitian oleh
Postma et al, 9 yang memiliki ukuran sampel
tertinggi wanita (51 dengan preeklamsia
sebelumnya, peserta kontrol n548). Para penulis
studi ini juga disesuaikan untuk usia dalam
analisis mereka dan menggunakan definisi
preeklamsia yang ketat. Kecenderungan serupa
ditemukan dalam tiga studi lainnya. Meskipun
wanita melakukan sedikit lebih buruk pada tes
Number Number Sequencing yang menilai
memori, temuan ini tidak bertahan dalam tes
neurokognitif lain yang menilai memori (Rey
Auditory Verbal Learning Test). Ada sejumlah
penjelasan untuk kurangnya asosiasi yang jelas
dalam hasil kognitif obyektif dalam tinjauan
sistematis kami meskipun menemukan skor
kognitif subjektif secara konsisten lebih rendah.
Pertama, penelitian dalam kajian kami memiliki
tindak lanjut yang relatif singkat antara
preeklampsia dan ukuran fungsi kognitif
menggunakan tes neurokognitif (median 5
tahun) dengan mayoritas penelitian melaporkan
tindak lanjut kurang dari 10 tahun. Akibatnya,
wanita dalam ulasan kami relatif muda pada saat
penilaian. Ada kemungkinan bahwa masalah
kognitif yang lebih terlihat secara klinis mungkin
menjadi jelas hanya dengan waktu sebagai
perubahan neurovaskular dan perkembangan
penyakit materi putih. Selain itu, studi individu
mungkin kurang bertenaga untuk mendeteksi
perbedaan halus tetapi signifikan dalam tes
neurokognitif. Memang, tren statistik yang
cognitive decline in formerly preeclamptic
women could be limited by the confounding
effects of psychologic stressors4 and recall bias.
An interesting finding in our review is the
consistent reporting of abnormal presence or
number of white matter lesions on brain MRI
after preeclampsia, predominantly in the frontal
lobe. It is less clear whether these white matter
lesions are more severe, with one large study
showing no difference15 but smaller studies
suggesting a difference between women with
prior preeclampsia compared with those without
prior preeclampsia.13,29 We were not able to
pool the severity of white matter lesions on MRI
together in our meta-analysis as a result of
differences in reporting of white matter lesions.
The clinical relevance of white matter lesions in
young asymptomatic individuals is unknown.
Studies in elderly patients have shown white
matter lesions to be associated with a higher risk
for ischemic strokes, cognitive decline, and
dementia.16 In contrast, however, white matter
lesions have previously been reported in
relatively young and healthy cohorts34 and is
known to develop in patients with a history of
hypertensive disease.34 Within our review,
patients with preeclampsia underwent MRI soon
after preeclampsia and on average were
relatively young. These white matter lesions
could be a manifestation of hypertensive disease
often coexistent in patients with preeclampsia or
could represent an irreversible pathologic insult
during pre eclampsia. We suspect, however, that
given the compensatory capacity of the brain at
a young age,33 the clinical manifestation of
white matter lesions in patients with
preeclampsia might not be apparent until much
later in life, which necessitates the need to
follow these women for a longer period and
evaluate cognitive function objectively at later
time points. Given the evidence supporting the
increased presence and severity of white matter
lesions in former preeclamptic patients and the
diamati dalam penelitian kecil12,33 mungkin
menerjemahkan ke efek yang signifikan jika
diulang dengan ukuran sampel yang lebih besar.
Sebagai alternatif, temuan kami tentang
penurunan kognitif yang dilaporkan sendiri pada
wanita yang sebelumnya preeklamsi dapat
dibatasi oleh efek pembaur dari stressor
psikologis4 dan bias ingatan. Temuan yang
menarik dalam tinjauan kami adalah pelaporan
konsisten adanya abnormal atau jumlah lesi
materi putih pada MRI otak setelah
preeklampsia, terutama di lobus frontal. Tidak
jelas apakah lesi materi putih ini lebih parah,
dengan satu penelitian besar yang menunjukkan
tidak ada perbedaan15 tetapi penelitian yang
lebih kecil menunjukkan perbedaan antara
wanita dengan preeklampsia sebelumnya
dibandingkan dengan mereka yang tidak
mengalami preeklamsia sebelumnya.13,29 Kami
tidak dapat menyatukan keparahan lesi materi
putih pada MRI bersama-sama dalam meta-
analisis kami sebagai akibat dari perbedaan
dalam pelaporan lesi materi putih. Relevansi
klinis lesi materi putih pada individu muda tanpa
gejala tidak diketahui.
Studi pada pasien usia lanjut telah menunjukkan
lesi materi putih untuk dikaitkan dengan risiko
yang lebih tinggi untuk stroke iskemik,
penurunan kognitif, dan demensia.16
Sebaliknya, bagaimanapun, lesi materi putih
sebelumnya telah dilaporkan pada kohort yang
relatif muda dan sehat34 dan diketahui
berkembang pada pasien dengan riwayat
penyakit hipertensi. Dalam ulasan kami, pasien
dengan preeklampsia menjalani MRI segera
setelah preeklampsia dan rata-rata relatif muda.
Lesi white matter ini bisa menjadi manifestasi
dari penyakit hipertensi yang sering hidup
berdampingan pada pasien dengan preeklampsia
atau bisa mewakili penghinaan patologis yang
tidak dapat diperbaiki selama pre eklampsia.
Kami menduga, bagaimanapun, bahwa
mengingat kapasitas kompensasi otak pada usia
muda, 33 manifestasi klinis lesi materi putih
pada pasien dengan preeklampsia mungkin tidak
jelas sampai jauh di kemudian hari, yang
mengharuskan kebutuhan untuk mengikuti
wanita ini untuk periode yang lebih lama dan
mengevaluasi fungsi kognitif secara obyektif
pada titik waktu kemudian. Mengingat bukti
elevated risk for ischemic stroke, 5 we
hypothesized that preeclampsia could be linked
with an early onset of dementia. Within our
review, however, there did not appear to be any
association between preeclampsia and a clinical
diagnosis of dementia. These studies, however,
often evaluated dementia collectively without
separating neurodegenerative causes from
dementia secondary to a vascular insult. Two of
the three studies evaluating the association
between preeclampsia and dementia included
dementia that was unspecified,30,32 and one
study examined preeclampsia with Alzheimer
dementia.31 In one of the three studies, vascular
dementia was analyzed in a subanalysis of
patients with hypertension and proteinuria in
pregnancy and an elevated risk for vascular
dementia was found (hazard ratio 6.27, CI 1.64–
27.44)30; however, this was analyzed in only two
patients with vascular dementia. Further studies,
however, are needed to support this finding, and
future research should focus on examining the
association between preeclampsia and vascular
dementia. Our review has several strengths.
Although much is known about the association
between preeclampsia and cardiovascular
disease, considerably less is known about the
spectrum of cerebrovascular conditions that
might ensue after this pregnancy complication.
We used a rigorous approach following Preferred
Reporting Items for Systematic Reviews and
Meta-analysis guidelines17 and did not limit our
search based on time or language. We pooled
results only among studies with moderate or low
heterogeneity and reported results separately
for subjective and objective cognitive measures.
Despite these strengths, our review has several
limitations that should be considered.
Substantial heterogeneity in the outcome
measurements of our studies assessing
neurocognitive tests or radiologic imaging made
it challenging to pool existing results together in
a meta-analysis. Thus, although we provide a
summary of existing data, we are unable to
yang mendukung peningkatan kehadiran dan
keparahan lesi materi putih pada pasien
preeklamsi sebelumnya dan peningkatan risiko
untuk stroke iskemik, 5 kami berhipotesis bahwa
preeklampsia dapat dikaitkan dengan onset
demensia dini. Namun, dalam peninjauan kami,
tampaknya tidak ada hubungan antara
preeklamsia dan diagnosis klinis demensia. Studi-
studi ini, bagaimanapun, sering mengevaluasi
demensia secara kolektif tanpa memisahkan
penyebab neurodegeneratif dari demensia
sekunder ke penghinaan vaskular. Dua dari tiga
penelitian yang mengevaluasi hubungan antara
preeklamsia dan demensia termasuk demensia
yang tidak spesifik, 30,32 dan satu penelitian
memeriksa preeklampsia dengan demensia
Alzheimer. Pada salah satu dari tiga penelitian,
demensia vaskular dianalisis dalam subanalysis
pasien dengan hipertensi dan proteinuria pada
kehamilan dan peningkatan risiko untuk
demensia vaskular ditemukan (rasio hazard 6.27,
CI 1.64–27.44) 30; Namun, ini dianalisis hanya
pada dua pasien dengan demensia vaskular.
Studi lebih lanjut, bagaimanapun, diperlukan
untuk mendukung temuan ini, dan penelitian
masa depan harus fokus pada pemeriksaan
hubungan antara preeklampsia dan demensia
vaskular. Ulasan kami memiliki beberapa
kekuatan. Meskipun banyak yang diketahui
tentang hubungan antara preeklamsia dan
penyakit kardiovaskular, sangat kurang diketahui
tentang spektrum kondisi serebrovaskular yang
mungkin terjadi setelah komplikasi kehamilan
ini. Kami menggunakan pendekatan yang ketat
mengikuti Item Pelaporan Pilihan untuk Tinjauan
Sistematis dan panduan meta-analisis17 dan
tidak membatasi pencarian kami berdasarkan
waktu atau bahasa. Kami mengumpulkan hasil
hanya di antara penelitian dengan heterogenitas
moderat atau rendah dan melaporkan hasil
secara terpisah untuk ukuran kognitif subjektif
dan objektif. Meskipun kekuatan ini, ulasan kami
memiliki beberapa keterbatasan yang harus
dipertimbangkan.
Heterogenitas substansial dalam pengukuran
hasil dari penelitian kami menilai tes
neurokognitif atau pencitraan radiologis
membuatnya menantang untuk mengumpulkan
hasil yang ada bersama-sama dalam meta-
analisis. Dengan demikian, meskipun kami
comprehensively estimate an overall effect of
preeclampsia on cognition. A second limitation
inherent to systematic reviews of observational
studies is the potential for both known and
unknown confounders in individual studies that
may explain any observed estimates of effect.
Within our review, 5 of 13 studies did not adjust
for any additional factors in their assessment of
cognitive function in patients with preeclampsia.
Furthermore, despite a comprehensive search of
the literature, we were limited by a small
number of studies and sample size of patients in
our metaanalysis with a relatively short follow-
up for cognitive outcomes. The results of our
systematic review and metaanalysis shed light
on the gaps and limitations in existing studies
examining the association between preeclampsia
and cognitive outcomes. Future studies are
needed that follow large cohorts of women with
and without preeclampsia over a long period of
time, use well-validated and uniform
neurocognitive tests assessing each cognitive
domain, and attempt to control for confounders
and mediators including hypertension,
education, and psychologic variables. The
concomitant use of MRI with neurocognitive test
will provide a more comprehensive picture of
the presence of absence of clinically significant
cerebrovascular pathology. The use of sensitive
tests to detect subtle cognitive alterations in
young individuals should be prioritized. If
cognitive impairment does indeed persist after
preeclampsia, such tests should be added to the
recommended cardiovascular screening in these
women after pregnancy. Abundant data now
exist that preeclampsia and other pregnancy
complications increase a woman’s risk for
hypertension and premature cardiovascular
diseases.5 International consensus is needed to
establish a core outcome set and standardized
reporting recommendations after preeclampsia.
Cognitive testing, MRI, and functional brain
imaging may be additional measures to be
included in this core outcome set to gain a
comprehensive picture of the extent of both
clinical and subclinical vascular injury in these
women. Preeclampsia is associated with
subjective cognitive impairment after pregnancy
without clear evidence of deficits in attention or
memory on neurocognitive tests. Future
research efforts should focus on designing
memberikan ringkasan data yang ada, kami tidak
dapat secara komprehensif memperkirakan efek
keseluruhan preeklampsia pada kognisi.
Keterbatasan kedua yang melekat pada tinjauan
sistematis studi observasional adalah potensi
untuk pembaur yang dikenal dan tidak dikenal
dalam studi individu yang dapat menjelaskan
perkiraan efek yang diamati. Dalam ulasan kami,
5 dari 13 penelitian tidak menyesuaikan untuk
faktor tambahan dalam penilaian mereka
terhadap fungsi kognitif pada pasien dengan
preeklampsia. Selain itu, meskipun pencarian
literatur yang komprehensif, kami dibatasi oleh
sejumlah kecil studi dan ukuran sampel pasien
dalam metaanalisis kami dengan tindak lanjut
yang relatif singkat untuk hasil kognitif. Hasil
tinjauan sistematis dan metaanalisis kami
menjelaskan celah dan keterbatasan dalam
penelitian yang ada yang meneliti hubungan
antara preeklamsia dan hasil kognitif. Penelitian
selanjutnya diperlukan yang mengikuti kelompok
besar wanita dengan dan tanpa preeklamsia
selama jangka waktu yang panjang,
menggunakan tes neurokognitif yang tervalidasi
dengan baik dan seragam menilai setiap domain
kognitif, dan berusaha untuk mengontrol
pembaur dan mediator termasuk hipertensi,
pendidikan, dan variabel psikologis . Penggunaan
MRI secara bersamaan dengan uji neurokognitif
akan memberikan gambaran yang lebih
komprehensif tentang tidak adanya patologi
serebrovaskular yang signifikan secara klinis.
Penggunaan tes sensitif untuk mendeteksi
perubahan kognitif yang halus pada individu
muda harus diprioritaskan. Jika gangguan
kognitif memang tetap ada setelah
preeklampsia, tes tersebut harus ditambahkan
ke skrining kardiovaskular yang
direkomendasikan pada wanita ini setelah
kehamilan. Data yang berlimpah sekarang ada
bahwa preeklampsia dan komplikasi kehamilan
lainnya meningkatkan risiko seorang wanita
untuk hipertensi dan penyakit kardiovaskular
prematur.5 Diperlukan konsensus internasional
untuk menetapkan hasil inti dan rekomendasi
pelaporan standar setelah preeklamsia. Tes
kognitif, MRI, dan pencitraan otak fungsional
mungkin merupakan langkah tambahan untuk
dimasukkan dalam hasil inti yang ditetapkan
untuk mendapatkan gambaran yang
komprehensif tentang tingkat cedera vaskular
studies with long-term follow-up, using well- klinis dan subklinis pada wanita ini. Preeklamsia
validated and uniform neurocognitive tests, and berhubungan dengan gangguan kognitif subjektif
controlling for confounders. setelah kehamilan tanpa bukti defisit yang jelas
dalam perhatian atau memori pada tes
neurokognitif. Upaya penelitian di masa depan
harus fokus pada merancang penelitian dengan
tindak lanjut jangka panjang, menggunakan tes
neurokognitif yang divalidasi dengan baik dan
seragam, dan mengendalikan pembaur.

Long-term mortality risk and life expectancy Risiko mortalitas jangka panjang dan harapan
following recurrent hypertensive 2 disease of hidup setelah berulangnya penyakit hipertensi 2
pregnancy kehamilan

Conflict of interest/disclosure: The authors Konflik kepentingan / pengungkapan: Para

report no conflicts of interest. Presentation: Data
from this study were presented as a Fellows
Plenary oral presentation at the 2017 SMFM
Annual Meeting in Las Vegas, NV. Funding: We
thank the Pedigree and Population Resource of
the Huntsman Cancer Institute, University of
Utah (funded in part by the Huntsman Cancer
Foundation) for its role in the ongoing
collection, maintenance, and support of the Utah
Population Database. We also acknowledge
partial support for the Utah Population Database
through grant P30 CA2014 from the Huntsman
Cancer Foundation, University of Utah and from
the University of Utah’s Program in Personalized
Health and Center for Clinical and Translational
Science. Huong Meeks, Alison Fraser, and Ken
Smith were supported by R01AG022095 (Early
Life Conditions, Survival, and Health: A
Pedigree25 Based Population Study) (PI Smith).
Michael Varner is supported by NIH/NCATS
1UL1TR001067 and by the HA and Edna Benning
Presidential Endowment. Corresponding Author:
Lauren H. Theilen, M.D. | Department of
Obstetrics & Gynecology, University of Utah
Health | 30 N 1900 E, Suite 2B200, Salt Lake City,
Utah, 84132 | Phone 801 581 8425 | Fax 801 585
2594 | E-mail lauren.theilen@hsc.utah.edu
Word Count: Abstract – 317, Manuscript – 2,454
Condensation: Women with recurrent
hypertensive disease of pregnancy have
increased mortality risks compared to women
without recurrent hypertensive disease of
pregnancy. Short version of title: Mortality
following recurrent hypertensive disease of
penulis melaporkan tidak ada konflik
kepentingan. Presentasi: Data dari penelitian ini
disajikan sebagai presentasi oral Fellows Pleno di
Pertemuan Tahunan 2017 SMFM di Las Vegas,
NV. Pendanaan: Kami berterima kasih kepada
Pedigree dan Sumber Daya Penduduk Institut
Kanker Huntsman, Universitas Utah (didanai
sebagian oleh Yayasan Kanker Huntsman) untuk
perannya dalam pengumpulan, pemeliharaan,
dan dukungan berkelanjutan dari Database
Populasi Utah. Kami juga mengakui dukungan
parsial untuk Database Populasi Utah melalui
hibah P30 CA2014 dari Yayasan Kanker
Huntsman, Universitas Utah dan dari Program
Universitas Utah dalam Personalised Health dan
Pusat untuk Ilmu Klinis dan Translasional. Huong
Meeks, Alison Fraser, dan Ken Smith didukung
oleh R01AG022095 (Kondisi Kehidupan Dini,
Kelangsungan Hidup, dan Kesehatan: Studi
Populasi Berbasis Pedigree25) (PI Smith).
Michael Varner didukung oleh NIH / NCATS
1UL1TR001067 dan oleh HA dan Edna Benning
Presidential Endowment. Penulis yang sesuai:
Lauren H. Theilen, M.D. | Departemen Obstetri
& Ginekologi, Universitas Utah Health | 30 N
1900 E, Suite 2B200, Salt Lake City, Utah, 84132
| Telepon 801 581 8425 | Faks 801 585 2594 | E-
mail lauren.theilen@hsc.utah.edu Word Count:
Abstrak - 317, Manuskrip - 2.454
Kondensasi: Wanita dengan penyakit hipertensi
berulang kehamilan telah meningkatkan risiko
kematian dibandingkan dengan wanita tanpa
penyakit hipertensi berulang kehamilan. Versi
singkat dari judul: Kematian setelah penyakit
pregnancy Implications and Contributions
A. To determine whether women with recurrent
hypertensive disease of pregnancy have
increased mortality risks.
B. Recurrent hypertensive disease is strongly
associated with increased mortality from 39
diabetes, heart disease, and stroke, and women
with recurrent disease had shorter life 40
expectancy compared to women with zero or
one affected pregnancy.
C. This study shows that women with recurrent
hypertensive disease have higher risk for
adverse long-term health outcomes than women
without hypertensive disease of pregnancy or
with only one affected pregnancy.
Abstract Background: Women with a history of
hypertensive disease of pregnancy have
increased risks for early mortality from multiple
causes. The effect of recurrent hypertensive
disease of pregnancy on mortality risk and life
expectancy is unknown. Objective: To determine
whether recurrent hypertensive disease of
pregnancy is associated with increased mortality
risks. Study Design: In this retrospective cohort
study, we used birth certificate data to
determine the number of pregnancies affected
by hypertensive disease of pregnancy for each
woman delivering in Utah from 1939-2012. We
assigned women to one of three groups based
on number of affected pregnancies: 0, 1, or ≥2.
Exposed women had ≥1 affected singleton
pregnancy and lived in Utah for ≥1 year
postpartum. Exposed women were matched 1:2
to unexposed women by age, year of childbirth,
and parity. Underlying cause of death was
determined from death certificates. Mortality
risks by underlying cause of death were
compared between exposed and unexposed
women as a function of number of affected
pregnancies. Cox regressions controlled for
infant sex, gestational age, parental education,
ethnicity, and marital status. Results: We
identified 57,384 women with ≥1 affected
pregnancy (49,598 women with 1 affected
pregnancy and 7,786 women with ≥2 affected
pregnancies). These women were matched to
hipertensi berulang kehamilan Implikasi dan
Kontribusi
A. Untuk menentukan apakah wanita dengan
penyakit hipertensi berulang kehamilan telah
meningkatkan risiko kematian.
B. Penyakit hipertensi rekuren sangat terkait
dengan peningkatan mortalitas dari 39 diabetes,
penyakit jantung, dan stroke, dan wanita dengan
penyakit berulang memiliki harapan hidup yang
lebih pendek 40 dibandingkan dengan wanita
dengan nol atau satu kehamilan yang
terpengaruh.
C. Penelitian ini menunjukkan bahwa wanita
dengan penyakit hipertensi berulang memiliki
risiko lebih tinggi untuk hasil kesehatan jangka
panjang yang merugikan dibandingkan wanita
tanpa penyakit hipertensi kehamilan atau hanya
dengan satu kehamilan yang terpengaruh.
Abstrak Latar belakang: Wanita dengan riwayat
penyakit hipertensi kehamilan memiliki
peningkatan risiko untuk kematian dini dari
banyak penyebab. Efek dari penyakit hipertensi
berulang kehamilan pada risiko kematian dan
harapan hidup tidak diketahui. Tujuan: Untuk
menentukan apakah penyakit hipertensi
berulang kehamilan dikaitkan dengan
peningkatan risiko kematian. Desain Studi:
Dalam penelitian kohort retrospektif ini, kami
menggunakan data akta kelahiran untuk
menentukan jumlah kehamilan yang dipengaruhi
oleh penyakit hipertensi kehamilan untuk setiap
wanita yang melahirkan di Utah dari 1939-2012.
Kami menugaskan wanita ke salah satu dari tiga
kelompok berdasarkan jumlah kehamilan yang
terpengaruh: 0, 1, atau ≥2. Wanita yang terpajan
memiliki ≥1 mempengaruhi kehamilan tunggal
dan tinggal di Utah selama ≥1 tahun
pascapartum. Perempuan yang terpapar
dicocokkan 1: 2 dengan perempuan yang tidak
terpapar berdasarkan usia, tahun kelahiran, dan
paritas. Penyebab kematian yang mendasari
ditentukan dari sertifikat kematian. Risiko
kematian oleh penyebab kematian yang
mendasari dibandingkan antara wanita terpapar
dan tidak terpajan sebagai fungsi dari jumlah
kehamilan yang terpengaruh. Regresi Cox
dikendalikan untuk seks bayi, usia kehamilan,
pendidikan orang tua, etnis, dan status
perkawinan. Hasil: Kami mengidentifikasi 57.384
114,768 unexposed women. As of 2016, 11,894 wanita dengan ≥1 yang terkena kehamilan
(49.598 wanita dengan 1 kehamilan yang
terkena dampak dan 7.786 wanita dengan ≥2
Abstract Background: Women with a history of kehamilan yang terkena dampak). Wanita-wanita
hypertensive disease of pregnancy have ini dicocokkan dengan 114.768
increased risks for early mortality from multiple
causes.

wanita yang tidak terpapar. Pada 2016, 11.894

wanita meninggal: 4,722 (8,2%) terkena dan
The effect of recurrent hypertensive disease of 7,172 (6,3%) tidak terpapar. Wanita dengan ≥2
pregnancy on mortality risk and life expectancy mempengaruhi kehamilan mengalami
is unknown. Objective: To determine whether peningkatan mortalitas dari semua penyebab
recurrent hypertensive disease of pregnancy is [adjusted hazard ratio (aHR) = 2,04, 95% CI 1,76-
associated with increased mortality risks. Study 2,36], diabetes (AHR = 4,33, 95% CI 2,21-8,47),
Design: In this retrospective cohort study, we penyakit jantung iskemik (AHR = 3.30, 95% CI
used birth certificate data to determine the 2.02-5.40), dan stroke (aHR = 5.10, 95% CI 2.62-
number of pregnancies affected by hypertensive 9.92). Untuk wanita yang indeks kehamilannya
disease of pregnancy for each woman delivering dikirim antara 1939-1959 (n = 10,488), mereka
in Utah from 1939-2012. We assigned women to dengan ≥2 mempengaruhi kehamilan memiliki
one of three groups based on number of harapan hidup tambahan yang lebih pendek
affected pregnancies: 0, 1, or ≥2. Exposed daripada ibu yang hanya memiliki 1 atau 0
women had ≥1 affected singleton pregnancy and kehamilan hipertensi (48,92 vs 51,91 vs 55,48
lived in Utah for ≥1 year postpartum. Exposed tahun, masing-masing). Kesimpulan: Penyakit
women were matched 1:2 to unexposed women hipertensi pada kehamilan dikaitkan dengan
by age, year of childbirth, and parity. Underlying risiko yang berlebihan untuk semua penyebab
cause of death was determined from death kematian dini dan beberapa kematian spesifik
certificates. Mortality risks by underlying cause penyebab, dan risiko ini meningkat lebih lanjut
of death were compared between exposed and dengan penyakit berulang. Kata kunci:
unexposed women as a function of number of Preeclampsia berulang, kehamilan sebagai
affected pregnancies. Cox regressions controlled jendela untuk kesehatan masa depan, analisis
for infant sex, gestational age, parental survival
education, ethnicity, and marital status. Results:
We identified 57,384 women with ≥1 affected
pregnancy (49,598 women with 1 affected
pregnancy and 7,786 women with ≥2 affected
pregnancies). These women were matched to
114,768 unexposed women. As of 2016, 11,894
women were deceased: 4,722 (8.2%) exposed
and 7,172 (6.3%) unexposed. Women with ≥2
affected pregnancies had increased mortality
from all causes [adjusted hazard ratio
(aHR)=2.04, 95% CI 1.76-2.36], diabetes
(aHR=4.33, 95% CI 2.21-8.47), ischemic heart
disease (aHR=3.30, 95% CI 2.02-5.40), and stroke
(aHR=5.10, 95% CI 2.62-9.92). For women whose
index pregnancy delivered between 1939-1959
(n=10,488), those with ≥2 affected pregnancies
had shorter additional life expectancies than
mothers who had only 1 or 0 hypertensive
pregnancies (48.92 vs 51.91 vs 55.48 years,
respectively). Conclusion: Hypertensive diseases
of pregnancy are associated with excess risks for
early all-cause mortality and some cause-specific
mortality, and these risks increase further with
recurrent disease. Key words: Recurrent
preeclampsia, pregnancy as a window to future
health, survival analysiswomen were deceased:
4,722 (8.2%) exposed and 7,172 (6.3%)
unexposed. Women with ≥2 affected
pregnancies had increased mortality from all
causes [adjusted hazard ratio (aHR)=2.04, 95% CI
1.76-2.36], diabetes (aHR=4.33, 95% CI 2.21-
8.47), ischemic heart disease (aHR=3.30, 95% CI
2.02-5.40), and stroke (aHR=5.10, 95% CI 2.62-
9.92). For women whose index pregnancy
delivered between 1939-1959 (n=10,488), those
with ≥2 affected pregnancies had shorter
additional life expectancies than mothers who
had only 1 or 0 hypertensive pregnancies (48.92
vs 51.91 vs 55.48 years, respectively).
Conclusion: Hypertensive diseases of pregnancy
are associated with excess risks for early all-
cause mortality and some cause-specific
mortality, and these risks increase further with
recurrent disease. Key words: Recurrent
preeclampsia, pregnancy as a window to future
health, survival analysis

Introduction Certain medical complications of

pregnancy are associated with the later
development of chronic diseases (1-6). In
particular, hypertensive disease of pregnancy is
associated with subsequent chronic
cardiovascular disease (7-9). Several studies have
now demonstrated the association between a
history of hypertensive disease of pregnancy and
subsequent early mortality from cardiovascular
and other causes (10- 81 12). Although previous
studies have examined the effect of recurrent
preeclampsia on subsequent cardiovascular
morbidity and mortality (13, 14), there are
currently no available data regarding the effect
of recurrent pregnancies complicated by any
hypertensive disease of pregnancy on
subsequent all-cause mortality in an American
cohort. In order to address this knowledge gap,
we examined whether women with recurrent
hypertensive disease of pregnancy have
increased risk for early mortality and shorter life
spans compared to women with either zero
affected pregnancies or only one affected
pregnancy.

Materials and Methods

Data Source
We used the Utah Population Database to
identify a retrospective cohort of women who
gave birth between 1939 and 2012. The Utah
Population Database has been previously
described in detail (14, 15). It consists of linked
records pertaining to over 9 million people, and
includes genealogy records from the
Genealogical Society of Utah, official statewide
birth and death records, and hospital discharge
and ambulatory surgery records from the Utah
State Department of Health. Its population is
representative of a broad spectrum of the
Caucasian United States population (16). Study
approvals were obtained from the Resource for
Genetic and Epidemiologic Research, a special
review panel authorizing access to the Utah
Population Database, as well as the University of
Utah Institutional Review Board. Inclusion
Criteria Women were included if they had at
least one singleton pregnancy with birth
certificate data during the study period and lived
in Utah for at least one year following delivery.
Using birth certificate data, we assigned a
diagnosis of hypertensive disease of pregnancy
to each affected pregnancy. Hypertensive
disease diagnoses included gestational
hypertension, preeclampsia, HELLP syndrome,
and eclampsia, and were assigned as listed on
the birth certificate for the affected pregnancy.
When available, we used inpatient records from
the time of delivery to confirm hypertensive
diagnoses. We then determined the number of
affected pregnancies for each woman, and
assigned women to one of three categories: no
affected pregnancies, one affected pregnancy, or
two or more affected pregnancies. Exposure of
Interest Exposed women had at least one
singleton pregnancy complicated by
hypertensive disease of pregnancy. The index
exposed pregnancy was defined as the 116 most
severely affected pregnancy in the Utah
Population Database, with diagnoses ranked
from least to most severe as follows: gestational
hypertension, preeclampsia, HELLP syndrome,
eclampsia. If two pregnancies were affected
equally, then the earliest pregnancy was used.
Exposed women were excluded if they had
missing data on a key variable which would
preclude matching or if they had documented
medical

comorbidities at the time of their pregnancy

(chronic hypertension, antiphospholipid 122
syndrome, pre-gestational diabetes, and chronic
kidney disease). We also excluded 123 exposed
women who died within one year of delivery in
order to assess the risk of long124 term
mortality as opposed to maternal mortality. 125
Unexposed women had pregnancies during the
study period, but none were 126 complicated by
hypertensive disease of pregnancy. These
unexposed women were 127 excluded if they
were missing data on a key variable which would
preclude matching, if 128 they died within one
year of delivery, or if the birth or fetal death
certificate listed a 129 history of prior pregnancy
complicated by hypertensive disease. 130
Matching 131 Exposed women were matched
1:2 to unexposed women by 5-year age groups,
132 year of childbirth (within one year), and
parity (1, 2, 3, 4, 5 or more) at the time of the
133 index pregnancy. 134 Outcome of Interest
135 Our primary outcome of interest was
mortality. The date of death was determined 136
using the Utah Population Database genealogies,
Utah death certificates, or the Social 137
Security Death Index. Importantly, the Social
Security Death Index is a national 138 database,
which allowed us to identify deaths which
occurred in Utah or elsewhere. The 139
underlying cause of death was determined from
Utah death certificates based on the 140
International Classification of Diseases (ICD)
version used at the time of death and 141
summarized into broad cause of death
categories.

In addition to the matching criteria, several

confounders were included in the Cox 144
models to generate hazard ratios for mortality
among women with a history of 145
hypertensive disease of pregnancy. These
included infant sex, gestational age at 146
delivery, parental education, maternal
race/ethnicity, and maternal marital status.
Infant 147 sex was included as a potential
confounder because of the known association
between 148 male fetal sex and maternal
preeclampsia (17). Gestational age at delivery
was 149 included as a potential confounder
because of the association between a history of
150 preterm birth and adverse long-term
maternal health outcomes (18, 19), but this did
not 151 distinguish between spontaneous and
iatrogenic preterm births. 152 Statistical
Methods 153 Cause-specific mortality risks were
estimated based on comparisons between 154
the number of affected pregnancies, using
stratified Cox models to incorporate 155
matching into the analysis. The time axis was
years since index birth, and the number of 156
affected pregnancies was defined at the subject
level. Unadjusted and adjusted hazard 157 ratios
and 95% confidence intervals were calculated for
all-cause and cause-specific 158 mortality. The
broad categories of causes of death were
derived by the National Center 159 for Health
Statistics, which has implemented a standard
broad classification of causes 160 of death (14).
These categories (and their ICD9 codes) include
infectious diseases 161 (001-139), neoplasms
(140-239), endocrine/nutritional/metabolic
diseases (240-279), 162 diseases of blood and
blood-forming organs (280-289), mental
disorders (290-319), 163 nervous system
disorders (320-389), circulatory system disease
(390-459), respiratory 164 system disease (460-
519), digestive system disease (520-579),
genitourinary disease

(580-629), musculoskeletal and connective tissue

disease (710-739), ill-defined 166 diseases (780-
799), and external causes (E800-E999). 167 We
then compared mortality risk by underlying
cause of death among women 168 with one
affected pregnancy or two or more affected
pregnancies versus matched 169 women with no
affected pregnancies for deaths occurring ≤50
years of age and deaths 170 occurring >50 years
of age. Adjusted hazard ratios (aHR) were
calculated for each 171 cause of mortality, as
well as the p value for the difference in mortality
risk between the 172 two age groups. 173
Finally, we performed a companion analysis
using life table methods to 174 determine the
difference in additional life expectancy in years
for women with zero, one, 175 or two or more
affected pregnancies. Life table analysis accounts
for the number of 176 deaths occurring at each
age among persons surviving to that age. This is
the basis for 177 determining the probabilities of
survival to any age. Starting with maternal age at
the 178 time of the index pregnancy, remaining
life expectancy at each subsequent age x 179
(called ex) is also calculated. In our study, this life
table analysis was performed only for 180
women who delivered their index pregnancy
between the years 1939 to 1959 in order to 181
maximize the length of follow-up and the ability
to observe adult deaths at mid-life or 182 later.
183 We included all women identified with a
diagnosis of hypertensive disease of 184
pregnancy during the study period that met our
inclusion criteria. Hypothesis tests were 185
based on a Type I error of ____________

Results
We identified 932,788 women who had one or
more singleton births between 189 1939 and
2012. Within this group, 62,489 women had at
least one pregnancy 190 complicated by
hypertensive disease. After excluding 5,103
women, we included 57,384 191 exposed
women in our analysis (49,598 women with one
affected pregnancy and 7,786 192 women with
two or more affected pregnancies). These
affected pregnancies included 193 27,546 cases
of gestational hypertension, 27,818 cases of
preeclampsia, 884 cases of 194 HELLP syndrome,
and 1,136 cases of eclampsia. These exposed
women were 195 matched to 114,768
unexposed women with no history of
hypertensive disease in any 196 pregnancies in
the Utah Population Database. Baseline
characteristics of exposed and 197 unexposed
women are reported in Table 1. Exposed women
were significantly more 198 likely to deliver at an
earlier gestational age, to deliver a neonate with
lower birth weight, 199 to have lower maternal
and paternal Nam-Powers socioeconomic scores
[a census200 derived score relating to
occupation (20)], and to have fewer years of
maternal 201 education. 202 As of 2015, 11,894
women were deceased: 4,722 (8.2%) exposed
and 7,172 203 (6.3%) unexposed. Figure 1 is a
forest plot illustrating the aHRs and 95%
confidence 204 intervals for mortality from
several causes for women with 1 affected
pregnancy 205 (represented by blue dots) and 2
or more affected pregnancies (represented by
orange 206 dots) compared to matched women
with no affected pregnancies. In relation to
matched 207 women with no affected
pregnancies, the aHRs for mortality for women
with two or 208 more affected pregnancies were
significantly higher than those for women with
one 209 affected pregnancy for deaths from all
causes (aHR 2.04 vs 1.62, p<0.001), , infectious
210 disease (aHR 4.86 vs 1.92, p <0.001),
endocrine, nutritional, and metabolic disease
(aHR)
4.19 vs 2.76, p50 years, we found that the
association between 226 hypertensive disease of
pregnancy and mortality from each cause of
death was 227 significant only for deaths
occurring ≤50 years of age (Figure 2), with the
exception of 228 deaths from stroke among
women with two or more pregnancies
complicated by 229 hypertensive disease of
pregnancy. 230 In our analyses using life table
methods, for women in this sample whose index
231 pregnancy delivered between 1939-1959
(n=10,488), we found a trend toward shorter 232
life expectancy with increasing number of
affected pregnancies. For example, within the
233 first decade following the index pregnancy,
the life expectancy for women with two or
more pregnancies complicated by hypertensive
disease of pregnancy was 48.92 years, 235 while
mothers with only one affected pregnancy had a
life expectancy of 51.91 years 236 and mothers
with no affected pregnancies had a life
expectancy of 55.48 years. 237 Comment 238
We found that recurrent hypertensive disease of
pregnancy is associated with 239 increased risks
for early all-cause and some cause-specific
mortality. This association is 240 significant for
deaths occurring at age ≤50 years. Our findings
are consistent with those 241 published
regarding the association between recurrent
preeclampsia and subsequent 242 early-onset
cardiovascular morbidity (21). This is important
because many women with 243 recurrent
hypertensive disease of pregnancy may not
begin screening for the chronic 244 diseases
underlying many of these causes of death
(hypertension, diabetes, and 245 dyslipidemia)
until age 45-50, thereby missing an opportunity
to intervene and 246 potentially improve
outcomes. The American College of Obstetricians
and 247 Gynecologists and the American Heart
Association both acknowledge hypertensive 248
disease of pregnancy as a significant risk factor
for early-onset cardiovascular disease 249 in
women, but there are insufficient data available
at this time to determine when and 250 how
screening and intervention should be
undertaken (22, 23). 251 Strengths of our study
include our large population-based sample and
252 retrospective design, which allowed us to
examine the association between the 253
exposure of interest and mortality in many
women over many years. Our retrospective 254
design is also a limitation, as there is the
potential for selection bias and residual 255
confounding. Obesity, substance use, and the
development of chronic hypertension 256
following pregnancy are important potential
confounders for which we could not adjust due
to lack of available data during the decades of
follow-up within the Utah Population 258
Database. The association between hypertensive
disease of pregnancy and mortality 259 from
some causes which are not intuitively related
(such as gastrointestinal disease) 260 may be
representative of unmeasured confounding to
some degree; however, 261 hypertensive
disease of pregnancy is a systemic process with
well-known end-organ 262 effects on the
gastrointestinal tract (specifically, the liver), so
we would not readily 263 dismiss this
association. Additionally, because our study
period encompassed a long 264 period of time,
historical risks may not reflect current risks in
the setting of contemporary 265 medical
practices. 266 Our use of birth certificate data
constitutes another limitation of our study. In
267 order to confirm birth certificate diagnoses
and to limit inconsistencies between 268
providers over time in assigning maternal
diagnoses of hypertensive disease of 269
pregnancy, we used inpatient records when
available; however, we would expect that 270
any residual misclassification of hypertensive
diagnoses should trend our results toward 271
the null. As a result, we may have
underestimated the magnitude of our positive
272 findings, rather than erroneously rejecting
the null hypothesis. Similarly, it is possible 273
that women may have had pregnancies outside
the state of Utah, and these 274 pregnancies
may have been complicated by hypertensive
disease. We attempted to 275 minimize this
limitation by excluding women with prior
hypertensive disease listed on 276 the birth
certificate, but we cannot exclude the possibility
that women may have had 277 subsequent
affected pregnancies after leaving Utah. Such an
error would make our 278 exposure groups
more similar, increasing the potential for type II
(but not type I) error.
Additionally, our study population was limited to
women with at least one viable 280 pregnancy
while residing in Utah, a population that remains
predominantly white and 281 non-obese (24,
25). This limits the generalizability of our
findings. 282 Despite these limitations, our study
constitutes a meaningful contribution to the 283
literature on long-term health outcomes among
American women following hypertensive 284
disease of pregnancy. While obstetricians are
making great strides in the prevention of 285
immediate maternal morbidity and mortality
related to hypertensive disease of 286 pregnancy
with a focus on safety bundles and treatment
algorithms, there may be 287 additional
opportunities to improve women’s long-term
health in the months and years 288 following
pregnancy. We still do not know whether the
association between 289 hypertensive disease of
pregnancy and subsequent mortality reflects a
causal 290 relationship or an underlying
predisposition to chronic disease which is
unmasked 291 during pregnancy; regardless, we
have identified an at-risk population of women
who 292 may benefit from early screening and
intervention for chronic diseases in order to 293
prevent early mortality (7). Prospective studies
are needed to determine whether 294 adjusted
health screening schedules may be beneficial in
this population.

Acknowledgments 296 The authors thank

Jennifer West, who is employed as a Utah
Population Database 297 navigator through the
Huntsman Cancer Institute, for creating the
figures for this 298 manuscript. The individual
named in this section has consented to such 299
acknowledgment.