Preeclampsia is a pregnancy-specific disease defined by new-onset hypertension (≥140 mmHg systolic or ≥90 mmHg diastolic blood pressure) together with one of more new-onset conditions, i.e. proteinuria (≥300 mg/24 h, protein/creatinine ≥30 mg/mmol or dipstick 2+), other maternal organ dysfunction (renal insufficiency, liver involvement, neurological or haematological complications) and/or fetal growth restriction, after 20 weeks of gestation . Although the pathogenesis is not yet fully clarified, defective placental function in early pregnancy is thought to be essential for the development of preeclampsia . Potential biomarkers for the detection of developing preeclampsia, before the above criteria are met, are found in substances that are released by the placenta and/or influence placental function, including angiogenic, renin angiotensin system related, immunological, metabolic and endocrine factors . As the clinical manifestation of preeclampsia is predominantly cardiovascular and exclusive cardiovascular changes can already be found in the preclinical phase , , we focus on two markers, vasopressin and adrenomedullin. Vasopressin maintains electrolyte homeostasis, by regulating water balance, and acts as a vasoconstrictor. Infusion of vasopressin induces features of preeclampsia, including hypertension, proteinuria and intrauterine growth restriction, in pregnant mice . Adrenomedullin (“adrenal medulla”) is a peptide originally extracted from a pheochromocytoma, but issynthesised in numerous tissues. It acts as a vasodilator and affects cell growth and differentiation, although its exact mechanisms of function remain largely unknown. Adrenomedullin has been related to several physiological and pathological conditions . Both peptides are, however, unstable and have short half-lifes in vivo. The C-terminal part of the preprovasopressin (CT-proAVP, copeptin) and the midregional part of proadrenomedullin (MR-proADM) do not seem to exert biological functionality, but are stable by-products and are cosecreted in equimolar amounts with their mature peptides , . Recent studies described both raised copeptin , , , , , ,  and MR-proADM ,  concentrations in preeclampsia compared with uncomplicated pregnancy. However, the predictive values are influenced by size and definition of both the case and control groups and the match in gestational age between cases and controls. Moreover the clinical use is hampered by the lack of trimester-specific reference intervals.
We estimated trimester-specific reference intervals for copeptin and MR-proADM in disease-free Dutch pregnant women. Blood samples were collected during regular antenatal visits, which are generally at 9–13 weeks and 27–29 weeks of gestation, and at towards the end of pregnancy (>36 weeks). Post-partum (PP) blood was collected within 3 months of delivery.
Materials and methods
One hundred and seventy-five disease-free pregnant Dutch women (98% Caucasian) were recruited via primary midwife practices in the southern part of the Netherlands between 2012 and 2014. Exclusion criteria included current or past history of endocrine disease, multiple pregnancy, use of medication known to influence thyroid function and current pregnancy as a result of hormonal stimulation. None of the women reported use of medication for hypertension or cardiovascular disease. Smoking status, baseline body mass index and parity were not assessed. The Medical Research Ethics Committee (MREC) Brabant and the local MREC at Amphia Hospital exempted this study from the Medical Research Involving Human Subjects Act (WMO) and hence approval by the MREC as it did not meet the criteria for medical-scientific research. Written informed consent was obtained from each participant.
Non-fasting blood samples were collected in EDTA tubes at 9–13 weeks, 27–29 weeks and 36–39 weeks of gestation (trimesters [Tr] 1, 2 and 3 respectively) and at 4–13 weeks PP. Plasma specimens were aliquoted and frozen at −80 °C. Aliquots were thawed once before analysis.
Copeptin and MR-proADM plasma concentrations were analyzed using immunofluorescent sandwich assays (BRAHMS CT-proAVP Ref 804.050, BRAHMS MR-proADM Kryptor Ref 829.050; ThermoFisher Scientific, Hennigsdorf, Germany). Analyses were performed on the B.R.A.H.M.S. Kryptor compact (MR-proADM; ThermoFisher Scientific) and the Kryptor compact plus (copeptin; ThermoFisher Scientific). Limits of detection were 0.9 pmol/L for copeptin and 0.05 nmol/L for MR-proADM. Limits of quantitation were 1.9 pmol/L for copeptin and 0.23 nmol/L for MR-proADM. The upper reference limits (URLs) are of clinical significance only. Non-pregnant 97.5th percentiles provided by the manufacturer are 16.4 pmol/L for copeptinand 0.55 nmol/L for MR-proADM.
Copeptin and MR-proADM concentrations were not normally distributed. URLs are defined as the 97.5th percentiles estimated using the bootstrap method described by Solberg . Outliers were identified using 0.33 (“one-third”) as a cut-off for the ratio D/R, where D is the absolute difference between the extreme value and the next largest or smallest value and R is the range of all observations, including extreme values. In case of multiple outliers, the ratio D/R was calculated on the least extreme outlier . Longitudinal changes during pregnancy and PP were assessed using the Friedman non-parametric test for paired observations. A value of p<0.05 was considered statistically significant. Post hoc, intra-individual changes in copeptin and MR-proADM between two timepoints were assessed using the Wilcoxon signed rank test fort two paired observations. p-Value of <0.01 was considered statistically significant, considering multiple testing. SPSS for MacIntosh version 22 (Chicago, IL, USA) was used for statistical analysis.
Women who miscarried after the first blood collection at Tr1 (n=4), developed significant disease (hyperemesis gravidarum [n=1], pregnancy-induced hypertension, preeclampsia or hemolysis elevated liver enzymes and low platelets [n=17], diabetes [n=4] or other disease [n=2]), delivered prematurely (n=2) or had a small for gestational age neonate (n=4) were excluded from statistical analyses. Nine women were included but failed to show up for any of the blood collections. All included women had normal blood pressure during and after pregnancy. Sixty-two women had complete data for Tr1, 2 and 3 and PP. Seventy-four women had complete data for Tr1, 2 and 3.
Mean (±SD) age at Tr1 was 30 (±4) years. Mean gestational period was 40 (±1) weeks.
Trimester-specific distributions of copeptin (pmol/L) and MR-proADM (nmol/L) are shown in Table 1. Both copeptin and MR-proADM concentrations increased during pregnancy and decreased again in the PP period. Copeptin concentrations remained below the non-pregnant URL provided by the manufacturer during pregnancy and PP. The URL of MR-proADM in Tr2 and Tr3 are over three times the non-pregnant URL provided by the manufacturer (Table 1; Figure 1).
Overall, copeptin concentrations increased during pregnancy (Tr2 vs. Tr1 and Tr3 vs. Tr2, both p<0.0001, n=74). PP copeptin decreased again (Tr3 vs. PP, p<0.0001, n=62). MR-proADM concentrations showed a similar pattern, with an increase during pregnancy (Tr2 vs. Tr1 and Tr3 vs. Tr2, both p<0.0001, n=74) and a subsequent decrease in the PP period (Tr3 vs. PP, p<0.0001, n=62) (Figure 1).
In absolute terms, the intra-individual increase in copeptin concentrations were relatively stable as individuals tracked within the tertile assigned in trimester 1 throughout pregnancy (Figure 2A). This was also the case for the intra-individual increase in MR-proADM concentrations, though tertiles 1 and 2 are overlapping each other.
Copeptin increases during pregnancy, but the 97.5th percentiles remain below the non-pregnant URL provided by the manufacturer (16.4 pmol/L, females and males together although median copeptin in males [7.4 pmol/L] were statistically significantly higher than in females [3.6 pmol/L]). Trimester-specific reference intervals should be used as using the manufacturer’s URL, the increase in copeptin and potential deviations from normal physiology during pregnancy would be neglected. In addition, the 97.5th percentile (10.8 pmol/L) copeptin concentrations in the PP period indicate that the URL provided by the manufacturer may be too high for our population even in the non-pregnant state and should at least be separately specified for males and females. This is also shown by Morgenthaler et al. , who developed the assay on which our commercial assay is based. In 359 healthy individuals (153 men, 206 women) the 97.5th percentile was 11.25 pmol/L, similar to our URL in the PP period, but median copeptin concentrations were significantly higher in men (5.2 pmol/L) than in women (3.7 pmol/L) . The median non-pregnant copeptin concentrations are comparable with our study.
The relative increase of copeptin in preeclampsia , , , , , ,  approximately ranges between 1.2 to 3 times, depending on the trimester in which the first measurement was made vs. the trimester in which the preeclampsia was diagnosed and the division into mild and severe preeclampsia (Table 2). These estimates exclude studies in which the relative increase in preeclampsia is likely to be underestimated, as controls had a more advanced gestational age, and hence naturally higher copeptin levels, than cases , .
The estimated relative increases in copeptin in preeclampsia fall within the trimester-specific reference intervals (Table 1) and are close to the normal increase between trimesters (from Tr1 to Tr2 and from Tr2 to Tr3) observed in disease-free population, which range from 1.3 times in the group that has the highest copeptin concentrations in Tr1 (tertile 3, Figure 2A) to 1.7 times in the group that has the lowest copeptin concentrations in Tr1 (tertile 1, Figure 2A). In addition, there seems to be a reasonable difference in absolute concentrations measured by different assays (Table 2).
Midregional part of proadrenomedullin
MR-proADM concentrations increase during pregnancy and decline again in the PP period. In Tr1 and PP the 97.5th percentiles (Tr1 0.64 nmol/L, PP 0.72 nmol/L) are close to the non-pregnant URL provided by the manufacturer (0.55 nmol/L, males and females together). However, in Tr2 (97.5th percentile 1.92 nmol/L) and Tr3 (97.5th percentile 2.69 nmol/L) the 97.5th percentiles are over three times the non-pregnant URL provided by the manufacturer indicating that these can not be used during pregnancy and trimester-specific reference intervals should be used.
A limited number of studies have investigated MR-proADM as a biomarker for preeclampsia. Both Sugulle et al.  and Wellmann et al.  showed increased MR-proADM concentrations in preeclampsia (1.31 [95% CI 1.23–1.49] nmol/L, n=105 ; 1.4 [10th–90th percentile 0.7–2.3] nmol/L, n=27 ) compared with controls (1.15 [95% CI 1.07–1.29] nmol/L, n=71 ; 1.1 [10th–90th percentile 0.5–1.9] nmol/L, n=120 ). However, mean gestational age at blood sampling in both studies was lower in cases than in controls and ranged from 24 to 42 weeks (trimesters 2 and 3). Although the natural increase in MR-proADM from Tr2 to Tr3 is less than the increase from Tr1 to Tr2 (Figure 2B), the difference between the two groups is probably underestimated as the controls would probably have had lower concentrations should they have been measured at the same gestational age as the cases. Still, this is in line with a study showing increased levels of the active peptide, adrenomedullin, in preeclampsia .
Matson et al. , in contrast, showed blunted concentrations of MR-proADM in severe preeclampsia (1.2 [SD 0.5] nmol/L, n=15) compared with controls (1.6 [SD 0.6] nmol/L, n=18). However, bloods were collected at time of delivery, therefore the observed results should be interpreted with caution. Again, gestational age at delivery was statistically significantly lower in cases (33.9 [SD 4.2] weeks, which also included second trimester deliveries) than controls (38.7 [SD 0.8] week) and MR-proADM concentrations can thus be expected to be lower in cases than in controls.
MR-proADM is present at low levels, even after the normal increase in pregnancy. The observed differences between preeclampsia and controls are small, although probably underestimated. Similar to copeptin, the absolute levels of MR-proADM are similar to the median concentrations measured in our study. Pre-eclampsia related differences in literature stay well below the 97.5th percentiles and normal increase between the trimesters in our study. All studies, including ours, used the same assay.
In summary, it seems wise to not only use trimester- and assay-specific reference intervals for copeptin and MR-proADM, but also to consider consecutive measurements and the time frame between measurements since the differences seen with or in advance of preeclampsia can be expected to be relatively small compared to the reference intervals.
We gratefully acknowledge the primary midwife practices of Verloskundigen Etten-Leur, Trivia, De Moriaen, Doortje Uil, Verloskundig Centrum Breda, Het Zomerhuis, Prinsenbeemden, Het Klavertje, Luna, Vita and De Ooievaar and the Origine birth centre for recruitment of the volunteers. We thank ThermoFisher Scientific for providing the test kits.
Tranquilli AL, Dekker G, Magee L, Roberts J, Sibai BM, Steyn W, et al. The classification, diagnosis and management of the hypertensive disorders of pregnancy: a revised statement from the ISSHP. Pregnancy Hypertens 2014;4:97–104. PubMedWeb of ScienceCrossrefGoogle Scholar
Steegers EA, von Dadelszen P, Duvekot JJ, Pijnenborg R. Preeclampsia. Lancet 2010;376:631–44. Google Scholar
Santillan MK, Santillan DA, Scroggins SM, Min JY, Sandgren JA, Pearson NA, et al. Vasopressin in preeclampsia: a novel very early human pregnancy biomarker and clinically relevant mouse model. Hypertension 2014;64:852–9. CrossrefWeb of SciencePubMedGoogle Scholar
Birdir C, Janssen K, Stanescu AD, Enekwe A, Kasimir-Bauer S, Gellhaus A, et al. Maternal serum copeptin, MR-proANP and procalcitonin levels at 11-13 weeks gestation in the prediction of preeclampsia. Arch Gynecol Obstet 2015;292:1033–42. Web of SciencePubMedCrossrefGoogle Scholar
Santillan M, Santillan D, Scroggins S, Min J, Leslie K, Hunter S, et al. Vasopressin secretion in human pregnancy predicts the development of preeclampsia as early as the sixth week of gestation. Pregnancy Hypertens 2015;5:25–6. Google Scholar
Sugulle M, Herse F, Seiler M, Dechend R, Staff AC. Cardiovascular risk markers in pregnancies complicated by diabetes mellitus or preeclampsia. Pregnancy Hypertens 2012;2:403–10. Web of ScienceCrossrefPubMedGoogle Scholar
Tuten A, Oncul M, Kucur M, Imamoglu M, Ekmekci OB, Acıkgoz AS, et al. Maternal serum copeptin concentrations in early- and late-onset pre-eclampsia. Taiwan J Obstet Gynecol 2015;54:350–4. CrossrefWeb of SciencePubMedGoogle Scholar
Yeung EH, Liu A, Mills JL, Zhang C, Männistö T, Lu Z, et al. Increased levels of copeptin before clinical diagnosis of preelcampsia. Hypertension 2014;64:1362–7. PubMedWeb of ScienceCrossrefGoogle Scholar
Zulfikaroglu E, Islimye M, Tonguc EA, Payasli A, Isman F, Var T, et al. Circulating levels of copeptin, a novel biomarker in pre-eclampsia. J Obstet Gynaecol Res 2011;37:1198–202. Web of SciencePubMedCrossrefGoogle Scholar
Clinical and Laboratory Standards Institute (CLSI). Defining, establishing and verifying reference intervals in the clinical laboratory. Approved guideline. 3rd ed. 2008, C28-A3c. Google Scholar
Morgenthaler NG, Struck J, Alonso C, Bergmann A. Assay for the measurement of copeptin, a stable peptide derived from the precursor of vasopressin. Clin Chem 2006;52:112–9. CrossrefPubMedGoogle Scholar
Bülbül GA, Kumru S, Erol O, Isenlik BS, Ozdemir O, Cağlar M, et al. Maternal and umbilical cord copeptin levels in pregnancies complicated by fetal growth restriction. J Matern Fetal Neonatal Med 2014;18:1–7. Web of ScienceGoogle Scholar
Matson BC, Corty RW, Karpinich NO, Murtha AP, Valdar W, Grotegut CA, et al. Midregional pro-adrenomedullin plasma concentrations are blunted in severe preeclampsia. Placenta 2014;35:780–3. PubMedWeb of ScienceCrossrefGoogle Scholar
About the article
Published Online: 2017-06-16
Published in Print: 2017-11-27
Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.
Research funding: None declared.
Employment or leadership: None declared.
Honorarium: None declared.
Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.