Scandinavian Journal of Clinical and Laboratory

Investigation

ISSN: 0036-5513 (Print) 1502-7686 (Online) Journal homepage: http://www.tandfonline.com/loi/iclb20

D-dimer during pregnancy: establishing trimester-

specific reference intervals

Irene Gutiérrez García, Pablo Pérez Cañadas, Juan Martínez Uriarte, Olivia
García Izquierdo, María Angeles Jódar Pérez & Luis García de Guadiana
Romualdo

To cite this article: Irene Gutiérrez García, Pablo Pérez Cañadas, Juan Martínez Uriarte, Olivia
García Izquierdo, María Angeles Jódar Pérez & Luis García de Guadiana Romualdo (2018): D-
dimer during pregnancy: establishing trimester-specific reference intervals, Scandinavian Journal of
Clinical and Laboratory Investigation, DOI: 10.1080/00365513.2018.1488177

To link to this article: https://doi.org/10.1080/00365513.2018.1488177

Published online: 05 Jul 2018.

Submit your article to this journal

Article views: 5

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=iclb20
SCANDINAVIAN JOURNAL OF CLINICAL AND LABORATORY INVESTIGATION
https://doi.org/10.1080/00365513.2018.1488177

ORIGINAL ARTICLE

D-dimer during pregnancy: establishing trimester-specific reference intervals

Irene Guti

errez Garc
ıaa, Pablo Pe
rez Can
~adasa, Juan Mart
ınez Uriarteb, Olivia Garc
ıa Izquierdob,

dar Pe
rez and Luis Garc
ıa de Guadiana Romualdoa
Mar
ıa Angeles Jo b

a
Biochemistry Department, Hospital General Universitario Santa Luc
ıa, Cartagena, Spain; bGynecology and Obstetrics Department, Hospital
Universitario Santa Luc
ıa, Cartagena, Spain

ABSTRACT ARTICLE HISTORY

Pregnancy is associated with an increased risk of venous thromboembolism (VTE). D-dimer is a Received 16 December 2017
biomarker used as an exclusion criterion of VTE disease, but its usefulness during pregnancy shows Revised 7 June 2018
limitations because D-dimer levels physiologically increase through pregnancy. The aim of our study Accepted 9 June 2018
was to follow the changes of D-dimer levels and to establish trimester-specific reference intervals
KEYWORDS
during normal pregnancy. This is a longitudinal prospective study in which the reference population D-dimer; pregnancy;
finally included 102 healthy pregnant women. Plasma D-dimer levels were measured during the three venous thromboembolism;
trimesters of pregnancy, using a latex-based immunoturbidimetric assay. Reference intervals were gestational
calculated according to the Clinical and Laboratory Standards Institute recommendations. D-dimer age-specific reference
levels increased progressively and significantly through pregnancy and peaked in the third trimester, intervals; trimesters
in which D-dimer levels were above the conventional cut-off point (500 mg/L) in 99% of pregnant
women. The following reference intervals were defined: first trimester: 169–1202 mg/L, second trimes-
ter: 393–3258 mg/L and third trimester: 551–3333 mg/L. The study provides reference intervals of
D-dimer during the pregnancy using latex-based immunoturbidimetry on the ACL 300 TOP automated
coagulation analyser. Further prospective studies of pregnant women with clinical suspicion of VTE are
needed to validate these results.

Introduction biochemical gold standard in the diagnostic approach to

patients with suspected VTE. A negative D-dimer in
Normal pregnancy is a hypercoagulable state because of
patients with low clinical probability safely rules out VTE
increased concentration of procoagulants, decreased levels of
[4]; however, this biomarker can be increased in patients
anticoagulants and decreased fibrinolytic activity. This hyper-
with a variety of conditions other than VTE, including preg-
coagulability may have evolved to protect women from bleed-
nancy and the puerperium [4]. During uncomplicated preg-
ing at the time of miscarriage or childbirth [1]. However, as a
nancy, D-dimer levels rise steadily, peaking in the early
consequence of these physiological changes, pregnant women
puerperium [5] and showing a high intra-individual bio-
are at higher risk of venous thromboembolism (VTE) [2].
logical variation [6]. Therefore, the value of D-dimer in
Diagnosis of VTE in pregnancy is difficult because many
pregnancy is controversial and further studies of pregnant
of the common VTE symptoms such as leg swelling, tachy-
women with regular D-dimer measurements are necessary
cardia, tachypnea and shortness of breath are also associated
to assess the natural course of D-dimer levels throughout
with normal pregnancy [3]. Furthermore, clinical prediction
pregnancy and to establish the safety of ruling out acute PE
scores, such as the Wells and Geneva scores, to assess the
based on a normal D-dimer level [2].
risk of VTE in non-pregnant women, have not been pro-
In this study, we aimed to describe the changes in
spectively validated in pregnancy or during the postpartum
D-dimer levels during pregnancy and to establish stratified
period. Finally, imaging studies, which in the absence of
reference intervals according to gestational age.
other diagnostic tests, are the current cornerstone of the diag-
nostic management of suspected PE in pregnant patients, are
associated with fetal and maternal radiation exposure. Material and methods
Therefore, the ability to safely rule out acute PE in pregnant
Design and study population
patients without radiological tests is of paramount import-
ance. Rapid, inexpensive, non-invasive and harmless tests to We conducted a longitudinal prospective study in the
the fetus are required to diagnose VTE in pregnancy. Biochemistry Department of the Santa Lucia General
D-dimer, a degradation product of fibrin, increases dur- University Hospital (Cartagena), which provides healthcare
ing an acute thrombotic process due to secondary activation activities in the Health Area 2 of Murcia (Spain). Only
of the fibrinolytic system. It is currently considered the healthy pregnant women in their first trimester, identified

CONTACT Irene Guti
errez Garc
ıa [email protected] Calle Mezquita s/n, Paraje Los Arcos 30202, Cartagena, Murcia, Spain
ß 2018 Medisinsk Fysiologisk Forenings Forlag (MFFF)
2 I. GUTI
ERREZ GARC
IA ET AL.

by a health survey, were recruited by the Department of Results

Gynaecology and Obstetrics at the time of their first tri-
Pregnant women were recruited for study at first trimester
mester ultrasound fetus scan, between February 2015 and
of gestation via a health survey. We initially selected 253
March 2016. The gestational age of the pregnant women
healthy pregnant women; 129 of them did not reappear for
was estimated using crown-rump length by ultrasound
taking blood at second or third trimester and in nine of
scan. Informed consent was obtained from all participating
them, blood samples were not properly collected in the pre-
women. Exclusion criteria were multiple gestation, women
viously defined period for the second or third trimester.
with previous VTE disease or family history of VTE,
After reviewing the medical records of each pregnant
clinical history of neoplasia, chronic disease (high blood
woman, we excluded those who had developed preeclampsia
pressure, pregestational diabetes mellitus, chronic kidney
or gestational hypertension (n ¼ 11). Furthermore, we also
disease, hepatitis, etc.), autoimmune diseases, acquired
excluded those women for which the blood samples pre-
immune deficiency syndrome (AIDS), coagulopathies,
sented any preanalytical errors that could interfere with the
admission to hospital or surgery in the previous four
measurement of D-dimer (n ¼ 2). The reference population
weeks, fever
38  C in the last week or complications dur-
finally comprised 102 pregnant women with a median age
ing pregnancy, such as gestational hypertension, including
of 30.9 (IQR: 6) years.
preeclampsia, severe intra-uterine growth restriction and D-dimer levels increased progressively and significantly
placental abruption, and development of VTE during puer- during gestation (p < .001) (Table 1 and Figure 1), reaching
perium (until 30 days after delivery). Maternal blood sam- the highest level in the third trimester. D-dimer levels were
ples were collected from each woman in the first trimester significantly higher in the second trimester than in the first
(13þ6 weeks), second trimester (14þ0–27þ6 weeks) and trimester (median (IQR): 786 mg/L (414) vs. 405 mg/L (257);
third trimester (28þ0–40þ6 weeks) to evaluate the changes p < .001). D-dimer levels in the third trimester were also
in D-dimer levels throughout pregnancy. found to be significantly increased compared to the second
trimester (median (IQR): 1244 mg/L (724) vs. 786 mg/L
Laboratory testing (414); p < .001).

Peripheral venous blood was collected in a vacuum tube Table 1. Medians and gestational age-specific reference intervals for D-dimer.
(Becton Dickinson Medical System, Franklin Lakes, NJ) con- D-dimer (mg/L)
taining 1/10 volume of liquid 0.129 M trisodium citrate, cen- Reference
trifuged at 2260 g at room temperature for 10 min. D-dimer GA mean Median interval
concentration was measured with a quantitative latex micro- GA (range) (IQR) p2.5th–p97.5th 90% CI
particle enhanced turbidimetric immunoassay on the ACL- First trimester 12þ4 (11þ3–13þ6) 405 (257) 169–1202 146–209
984–1394
TOP 300 automated coagulation analyser (Instrumentation Second trimester 24þ5 (20þ5–26þ6) 786 (414) 393–3258 261–478
Laboratory, Lexington, MA). Plasma was analysed within 4 h 1784–3999
of collection. D-dimer concentrations (mg/L) were expressed Third trimester 34þ2 (27þ0–40þ2) 1244 (724) 551–3333 395–573
2401–4306
in fibrinogen-equivalent units (FEU). The cut-off point used
GA: gestational age (weeks); IQR: interquartile range; p2.5th: percentile 2.5; p97.5th:
in our hospital as the exclusion criterion of VTE in non- percentile 97.5; CI: confidence interval.
pregnant women is 500 mg/L according to reagent man-
ufacturer’s recommendations.

Statistical analysis
The normality of continuous variables was tested by
Kolmogorov–Smirnov and the results, including D-dimer
levels, were reported as median (interquartile range [IQR]).
D-dimer levels were compared using a Related Sample
Friedman test two-way analysis of variance, and the subse-
quent comparison between pairs of trimesters used the
Wilcoxon test with Bonferroni’s correction. Statistical analy-
ses were performed using SPSS Statistical Package for the
Social Sciences (SPSS, Chicago, IL) Version 21.0. A p < .05
was considered to be statistically significant.
Reference intervals (2.5th and 97.5th percentiles with 90%
confidence intervals [CI]) for D-dimer were calculated for
each trimester using the non-parametric bootstrap method Figure 1. D-dimer levels stratified by trimesters. The upper edge and the lower
with RefVal 4.11 according to the recommendations of edge of the box indicate the 75th percentile and 25th percentile, respectively.
The horizontal bar in the middle of each box represents the median value. ap
the Clinical and Laboratory Standards Institute (CLSI) [7]. Value for the first trimester vs. the second trimester; bp value for the second tri-
Horn’s algorithm was used to remove the outliers. mester vs. the third trimester.
 SCANDINAVIAN JOURNAL OF CLINICAL AND LABORATORY INVESTIGATION 3

Table 2. Gestational age-specific values and references intervals of D-dimer.

D-dimer (mg/L)
Reference Cut-offa Methodology analyser First trimester Second trimester Third trimester
Present study (n ¼ 102)b 500 Latex-enhanced immunotur- Md (IQR): 405 (257) Md (IQR): 786 (414) Md (IQR): 1244 (724)
bidimetry ACL-TOP 300 (IL) Mean (SD): 492 (320) Mean (SD): 967 (586) Mean (SD): 1393 (839)
RI: 169–1202 RI: 393–3258 RI: 551–3333
Kline et al. [8] (n ¼ 50)c 500 Immunoturbidimetric assay Mean (SD): 579 (363) Mean (SD): 832 (456) Mean (SD): 1159 (573)
Organon Teknica
Wang et al. [11] (n ¼ 1343)d 500 Latex-enhanced immunotur- Md (IQR): 200 (220) Md (IQR): 680 (560) Md (IQR): 1330 (960)
bidimetry STA-R Evolution RI: <660 RI: <2290 RI: <3120
(Diagnostica Stago)
Hansen et al. [14]b (n ¼ 55) 500 Latex-enhanced immunotur- Mean: 400e Mean: 710e Mean: 1220e
bidimetry STA-R Evolution RI: 180–870 RI: 300–1690 RI: 500–3010
(Diagnostica Stago)
Md: median; IQR: interquartile range; SD: standard deviation; RI: reference interval.
Only studies using the same methodology and cutoff for the non-pregnant population (500 mg/L) were included in this table. Studies using a different method-
ology or cutoff, including those in which the cutoff was not reported, have been not included in this table.
a
Non-pregnant population (expressed as mg/L).
b
Reference intervals are given as 2.5th and 97.5th percentiles.
c
Reference intervals not reported.
d
Upper limit of reference interval is given as 95th percentile.
e
SD or IQR not shown.

D-dimer values of 36 (37.1%) for the pregnant women in Nevertheless, we found more variability in the first trimes-
their first trimester, 91 (93%) in the second trimester and ter, with D-dimer levels higher than those described by
101 (99%) in the third trimester exceeded the upper limit of Wang et al. [11], probably due to differences in time of
D-dimer in our laboratory (500 mg/L). blood sampling for measurement of D-dimer.
The median increase of D-dimer between the first and In concordance with previous studies [9,11,12,14,15], D-
second trimesters was 96.7%, ranging from 71% to 550%. dimer concentrations above 500 mg/L were found in the
Between the second and third trimesters it was 44.1%, rang- majority of pregnant women in the second (93%) and third
ing from 41% to 157%. (99%) trimesters. For this reason, this decision limit should
The 2.5th to 97.5th percentile reference intervals of not be used for pregnant women and it is necessary to
D-dimer at first trimester, second trimester and third search for new strategies to use D-dimer in the pregnant
trimester are listed in Table 2. population with suspected VTE. Establishing trimester-spe-
cific reference intervals in healthy pregnant women may be
Discussion one of these new strategies [18].
Our study has several limitations that should be consid-
D-dimer, a coagulation and fibrinolysis marker, is widely ered. Firstly, we did not calculate the reference interval for
used in clinical decisions to rule out VTE with low to mod- the early puerperium, which is a high-risk stage of VTE.
erate clinical suspicion [4] and ultrasound testing can be Secondly, we have reported the changes of D-dimer levels
safely omitted. However, the use of this biomarker in according to the gestational trimester, similar to other
pregnant women shows limitations due to its physiological previous studies [6,10–12]; however, some authors
and progressive increase during non-complicated gestation. have reported these variations stratified according to
Thus, using the established cut-off points for a non-preg-
gestational weeks [6,10,11]. Finally, our study did not
nant population implies a risk of false positive results in
include pregnant women with suspected VTE. For this rea-
pregnant women without VTE. Although many authors
son, it was not possible to assess the value of the proposed
have studied the evolution of D-dimer during pregnancy
reference intervals or calculate a new cut-off based on the
and puerperium [5,6,8–16], the comparison of results
discrimination capacity of D-dimer, a strategy carried out
obtained with different methods is not possible, because
D-dimer concentrations are instrument-dependent. The assays in other studies [2,19]. However, our research has an
vary because the D-dimer does not have a simple structure important strength because it was designed as a longitu-
with a uniform composition [5,17]. dinal study, as Murphy et al. have previously recom-
As expected, we found an increase of D-dimer during mended [10], reducing the inter-individual variation of
pregnancy, with the highest concentration during the third D-dimer levels.
trimester, similar to the results previously reported in other In conclusion, our study confirms that D-dimer concen-
studies [5,6,8–13]. trations increase significantly during pregnancy. Therefore, its
In Table 2, trimester-specific D-dimer values obtained in usefulness is limited because a high number of healthy preg-
different studies carried out with the same design are listed. nant women have a D-dimer level above the cut-off to rule
When our results were compared with those reported in out VTE in a non-pregnant population, particularly in the
studies using the same methodology (immunoturbidimetry) second and third trimesters. The reference intervals calculated
and the cut-off for the non-pregnant population (500 mg/L) in our study may play an important role in the management
[8,11,14], D-dimer levels in the second and third trimester of pregnant women with suspected VTE, but further studies
were comparable with those obtained in these studies. are needed to validate them in this population.
4 I. GUTI
ERREZ GARC
IA ET AL.

Disclosure statement [10] Murphy N, Broadhurst DI, Khashan AS, et al. Gestation-spe-
cific D-dimer reference ranges: a cross-sectional study. BJOG:
No potential conflict of interest was reported by the authors. Int J Obstet Gy. 2015;122:395–400.
[11] Wang M, Lu S, Li S, et al. Reference intervals of D-dimer
during the pregnancy and puerperium period on the STA-R
References evolution coagulation analyser. Clin Chim Acta.
2013;425:176–180.
[1] Bremme KA. Haemostatic changes in pregnancy. Best Pract Res [12] Kovac M, Mikovic Z, Rakicevic L, et al. The use of D-dimer
Clin Haematol. 2003;16:153–168. with new cutoff can be useful in diagnosis of venous thrombo-
[2] Van der Pol LM, Mairuhu AT, Tromeur C, et al. Use of clinical embolism in pregnancy. Eur J Obstet Gynecol Reprod Biol.
prediction rules and D-dimer tests in the diagnostic manage- 2010;148:27–30.
ment of pregnant patients with suspected acute pulmonary [13] Kovac MK, Lalic-Cosic SZ, Dmitrovic JM, et al. Thrombin gen-
embolism. Blood Rev. 2017;31:31–36. eration, D-dimer and protein S in uncomplicated pregnancy.
[3] Nijkeuter M, Ginsberg JS, Huisman MV. Diagnosis of deep Clin Chem Lab Med. 2015;53:1975–1979.
vein thrombosis and pulmonary embolism in pregnancy: a sys- [14] Hansen AT, Andreasen BH, Salvig JD, et al. Changes in fibrin
tematic review. J Thromb Haemost. 2006;4:496–500. D-dimer, fibrinogen, and protein S during pregnancy. Scand J
[4] Lippi G, Cervellin G, Casagranda I, et al. D-dimer testing for Clin Lab Invest. 2011;71:173–176.
suspected venous thromboembolism in the emergency depart- [15] R
eger B, P
eterfalvi A, Litter I, et al. Challenges in the evalu-
ment. Consensus document of AcEMC, CISMEL, SIBioC, and ation of D-dimer and fibrinogen levels in pregnant women.
SIMeL. Clin Chem Lab Med. 2014;52:621–628. Thromb Res. 2013;131:e183–e187.
[5] Szecsi PB, Jorgensen M, Klajnbard A, et al. Haemostatic reference [16] Cui C, Yang S, Zhang J, et al. Trimester-specific coagulation
intervals in pregnancy. Thromb Haemost. 2010;103:718–727. and anticoagulation reference intervals for healthy pregnancy.
[6] Hedengran KK, Andersen MR, Stender S, et al. Large D-dimer Thromb Res. 2017;156:82–86.
fluctuation in normal pregnancy: a longitudinal cohort study of [17] Jennings I, Woods TA, Kitchen DP, et al. Laboratory D-dimer
4117 samples from 714 healthy Danish women. Obstet Gynecol measurement: improved agreement between methods through
Int. 2016;2016:1. calibration. Thromb Haemost. 2007;98:1127–1135.
[7] Boyd J. Defining, establishing, and verifying reference intervals [18] Linnemann B, Bauersachs R, Rott H, et al. Working Group in
in the clinical laboratory; approved guidelines, CLSI document Women’s Health of Society of Thrombosis and Haemostasis.
C28-A3. Vol. 28, No. 3. Diagnosis of pregnancy-associated venous thromboembolism –
[8] Kline JA, Williams GW, Hernandez-Nino J. D-dimer concen- position paper of the Working Group in Women’s Health of
trations in normal pregnancy: new diagnostic thresholds are the Society of Thrombosis and Haemostasis (GTH). Vasa.
needed. Clin Chem. 2005;51:825–829. 2016;45: 87–101.
[9] Ercan S, Ozkan S, Y€ ucel N, et al. Establishing reference inter- [19] Chan WS, Lee A, Spencer FA, et al. D-dimer testing in pregnant
vals for D-dimer to trimesters. J Matern Fetal Neonatal Med. patients: towards determining the next ’level’ in the diagnosis of
2015;28:983–987. deep vein thrombosis. J Thromb Haemost. 2010;8:1004–1011.