First trimester biochemical markers in pregnancies developing gestational diabetes

Posted on by admin

Research

HJOG 2025, 24 (2), 101-109| doi: 10.33574/hjog.0590

Athena P. Souka1, Iliana Mengou1, Panagiotis Antsaklis1, Konstantinos Pateras2, Natasa Pithaki2, George Daskalakis1

11st Department of Obstetrics and Gynecology, Alexandra Hospital, Athens, Greece
2Department of Statistics, School of Ιnformation Sciences & Technology, Athens University of Economics and Business, Athens, Greece

Correspondence: Athena P. Souka, 1st Department of Obstetrics and Gynecology, Alexandra Hospital, Athens, Greece

Abstract

Background: Gestational diabetes mellitus (GDM) is a significant public health concern that is associated with adverse maternal and neonatal outcomes. Recent research has underscored the importance of various biomarkers in predicting the onset of GDM, with a specific focus on pregnancy-associated plasma protein A (PAPP-A). Several studies designate that low maternal PAPP-A levels are associated with an elevated risk of developing GDM.

Methods: We conducted a retrospective study on singleton pregnancies we investigate the correlation between the first trimester free β-hCG and PAPP-A concentrations and the subsequent development of gestational diabetes, explore their potential in screening for GDM and explore their potential in screening for SGA in pregnancies with and without GDM in an inner city obstetric population undergoing universal GDM screening.

Results: Overall, 1,241 singleton pregnancies were included and we observed that PAPP-A levels were significantly lower in women with GDM. PAPPA had modest but significant prognostic ability for GDM, (AUC=0.56; 95% CI: 0.52 – 0.60; p=0.005; Figure 1) and the optimal cut-off value would be <1.14 (sensitivity=58.3%, specificity=54.7%, PPV= 23.9% and NPV= 84.4%). In women without diabetes greater MoM PAPPA conferred a lower probability by about 60% of having a neonate with low BW ≤ 5th centile (OR=0.40; 95% CI: 0.19 – 0.83; p=0.014). In non-diabetic pregnancies PAPPA had a significant prognostic ability for BW ≤ 5th percentile (AUC=0.64; 95% CI: 0.56 – 0.72; p=0.006; Figure 2).

Discussion: The results of our study indicate the association of low PAPPA with subsequent development of GDM. Diagnostic accuracy analysis indicated, however, that it is unlikely to be useful as a sole predictor of GDM although it can potentially be of value in a multi-factorial predictive model to assess the risk for GDM.

Keywords: Diabetes mellitus, PAPP-A, pregnancy, diagnostic accuracy

Introduction­

Gestational diabetes mellitus (GDM) is a significant public health concern, affecting approximately 5-15% of pregnancies worldwide (1). Characterized by glucose intolerance that first manifests during pregnancy, GDM is associated with adverse maternal and neonatal outcomes, including increased risks of caesarean delivery, hypertension, and long-term metabolic disorders in both mothers and their offspring (2). Early identification and management of GDM are critical for improving health outcomes (3).

Recent research has underscored the importance of various biomarkers in predicting the onset of GDM, with a specific focus on pregnancy-associated plasma protein A (PAPP-A). PAPP-A, a metalloproteinase secreted by the placenta, plays a crucial role in regulating insulin-like growth factor (IGF) and may impact on metabolic processes during pregnancy (4). Free β-hCG and PAPPA are routinely measured in the first trimester assessment of the risk for common trisomies and thus any correlation of these biomarkers with pregnancy complications is particularly important.

The majority but not all of the existing studies have reported that low maternal PAPP-A levels are associated with an elevated risk of developing GDM, although with limited predictive ability (5-9).

The study aims to: 1. investigate the correlation between the first trimester free β-hCG and PAPP-A concentrations and the subsequent development of gestational diabetes, 2. explore their potential in screening for GDM and 3. explore their potential in screening for SGA in pregnancies with and without GDM in an inner city obstetric population undergoing universal GDM screening.

Material and methods 

This is a retrospective study on singleton pregnancies presenting for routine ultrasound examination at 11-13 weeks at the Fetal Medicine Unit of Alexandra General Hospital of Athens. The examination included fetal biometry, nuchal translucency measurement, secondary markers for trisomy, assessment of the fetal anatomy and maternal serum biochemistry. Maternal demographics and history were recorded at the first examination (Astraia Fetal Database, Gembruch, Germany). All examiners were certified by the Fetal Medicine Foundation (FMF). Blood was drawn at the time of the examination and free β h-CG and PAPP-A were measured in the maternal serum (Kryptor assay, Brahms Kryptor Compact PLUS) and converted to multiples of the median (MoM) with the FMF algorithm (uncorrected MoM).

Screening for GDM was routinely performed at 24-28 weeks with the 75gr glucose tolerance test (OGTT) according to the guidelines of the Hellenic Society of Obstetricians and Gynecologists. The test was considered positive if the fasting blood sugar was ≥ 92 mgr/dl or the 1h measurement was ≥ 180 mgr/dl or the 2h measurement was ≥ 153 mgr/dl. Obesity, previous macrosomic child, previous pregnancy with gestational diabetes or increased fasting glucose at routine antenatal check were indications for performing the OGTT earlier.

Pregnancies with GDM were managed in collaboration with the endocrinology team and were offered serial ultrasound examinations according to protocol. Cases with pre-existing diabetes and/or diabetes diagnosed before the 11-13 weeks’ examination were excluded from the analysis.

The outcome of the pregnancy was collected via the Maternity Unit records or after communication with the obstetrician or the mother for the cases that delivered elsewhere. Birth weight (BW) centiles were calculated according to local reference charts. SGA was defined as birthweight at or below the 5th centile for gestational age. The choice of the 5th centile was based on its strong correlation with adverse perinatal outcome.

Statistical analysis

Quantitative variables were expressed as mean values (SD), while qualitative variables were expressed as absolute and relative frequencies. Student’s t-tests were used for the comparison of MoM HCG and MoM PAPPA between participants with and without gestational diabetes. Μultiple linear regression analyses were used to explore the association of MoM HCG and MoM PAPPA with birthweight centiles, after adjusting for maternal weight, height, parity, assisted conception, smoking and hypertension/ preeclampsia. Multiple logistic regression analysis was used to explore the association of MoM HCG and MoM PAPPA with birthweight ≤ 5th centile, after adjusting for maternal weight, height, parity, assisted conception, smoking and hypertension/ preeclampsia. Adjusted odds ratios (OR) with 95% confidence intervals (95% CI) were computed from the results of the regression analyses.

Pearson correlations coefficients (r) were used to explore the association between two continuous variables. ROC curves (Receiver operating characteristic curves) were constructed in order to estimate the prognostic ability of MoM PAPPA for gestational diabetes and for birthweight ≤ 5th centile. Sensitivity, specificity, negative (NPV) and positive prognostic value (PPV) were calculated for optimal cut-offs. The area under the curve (AUC) was also calculated. All reported p values are two-tailed. Statistical significance was set at p<0.05 and analyses were conducted using SPSS statistical software (version 26.0).

Results 

The database was searched for singleton pregnancies that had first trimester screening for trisomies (N=4,456) and known pregnancy outcome, including information on gestational age at diagnosis of GDM and treatment, leaving a population of 1,241 singleton pregnancies. The demographic characteristics of the study sample are presented in Table 1.

PAPPA was significantly lower in women with GDM, whereas fHCG tended to be lower but the difference was not significant (Table 2).

PAPPA had modest but significant prognostic ability for GDM, (AUC=0.56; 95% CI: 0.52 – 0.60; p=0.005; Figure 1) and the optimal cut-off value would be <1.14 (sensitivity=58.3%, specificity= 54.7%, PPV=23.9% and NPV= 84.4%). Women with PAPPA<1.14 MoM had about 70% greater probability of developing GDM (OR=1.69; 95% CI: 1.27 – 2.25; p<0.001).

Figure 1. Area Under the Curve (AUC) for the prognostic value of PAPP‐A MoM for gestational diabetes. PAPP‐A=pregnancy associated placental protein A, MoM=multiples of the median.

PAPPA was significantly and positively associated with BW centile, regardless of having or not gestational diabetes but the effect was stronger in the absence of GDM (Table 3).

In women without diabetes greater MoM PAPPA conferred a lower probability by about 60% of having a neonate with low BW ≤ 5th centile (OR=0.40; 95% CI: 0.19 – 0.83; p=0.014). In non-diabetic pregnancies PAPPA had a significant prognostic ability for BW ≤ 5th percentile (AUC=0.64; 95% CI: 0.56 – 0.72; p=0.006; Figure 2). Optimal cut-off value would be <1.028 (sensitivity=60.6%, specificity=61.3%, PPV=5.2% and NPV=97.8%). Women with PAPPA<1.028 MoM had 2.44 times higher probability of delivering a small neonate with BW≤ 5th percentile (OR=2.44; 95% CI: 1.20 – 4.97; p=0.014).

Figure 2. Area Under the Curve (AUC) for the prognostic value of PAPP‐A MoM for birthweight ≤ 5th centile. PAPP‐A=pregnancy associated placental protein A, MoM=multiples of the median.

However, in GD pregnancies PAPPA was not significantly associated with low BW ≤5th centile.

MoM HCG was not significantly associated with GDM, birthweight centile nor low BW ≤ 5th centile.

In the group of women with GDM there was no significant association between gestational age at diagnosis and PAPPA (r=0.08; p=0.277). In addition, there were no significant differences in PAPPA values in pregnancies treated with diet only or diet plus insulin.

Discussion 

Principal findings

We examined a low-risk population of singleton pregnancies submitted to universal screening for GDM.

Our findings confirm the association between low first trimester PAPPA and subsequent development of GDM. Free β-hCG was not significantly associated with GDM or SGA birth. Possible confounding factors such as maternal age, weight, smoking, parity and assisted reproduction techniques were taken into account because we used multiples of the median values (MoM) for both biochemical markers.

The prognostic ability of PAPPA as sole predictor for GDM was poor (AUC=0.56, Figure 1). Nevertheless, about one in four women with PAPPA less than 1.14 MoM will develop GDM whereas having PAPPA < 1.14 MoM increases the risk for GDM by about 70%.

PAPPA was positively associated with BW centile in both diabetic and non-diabetic women (Table 3). However, in contrast with the non-diabetics, in GDM pregnancies low PAPPA could not predict fetal microsomia.

Neither the gestational age of the diagnosis of diabetes nor the type of treatment (diet/oral medication/insulin) were related to the PAPPA levels.

Results in the context of what is known

Our findings, in consistence with previous studies, confirmed the correlation of low PAPPA with GDM. Syngelaki et al examined about 31,000 pregnancies and found reduced levels of PAPPA and increased levels of PLGF at 11-13 weeks in the pregnancies that developed GDM (6). They reported that these two biochemical markers did not increase the detection rate for GDM compared to using maternal demographics/history criteria.

Donovan et al in a systematic review of about 85,000 pregnancies reported that PAPPA was decreased by about 20% and HCG by about 4% in GDM pregnancies (8). In the meta-analysis of Talasaz et al, the sensitivity and specificity of PAPP-A for predicting GDM were 55% and 90%, respectively but low precision was indicated (AUC=0.7) (10).

Although PAPP-A has low sensitivity as a sole predictor of GDM it may be useful in a multifactorial model. So far the approach of combining family history and PAPPA in order to enhance the prediction of GDM has yielded mediocre results (6-9). Other first trimester biomarkers, such as adiponectin, leptin and C-reactive protein seem promising (11,12). Recently Sweeting et al showed that the addition of parameters measured in the first trimester screening for pre-eclampsia (mean arterial pressure, uterine Doppler indices and PAPPA) marginally increased the prediction for GDM compared to history alone (model AUC =0.90 vs history AUC=0.88) (13).

PAPPA is known to be a strong predictor of the SGA neonate (14). Our study showed that this association is lost on GDM pregnancies although there remains a positive correlation with the birthweight centile. In agreement with our findings Beneventi et al compared PAPPA-MoM in 228 GDM pregnancies and an equal number of controls and reported that there was no correlation between PAPPA level and birthweight z-score (15). The loss of the predictive capacity of PAPPA in the GDM group has received little attention although it is clinically important. Presumably the drop of the PAPPA levels in this population masks the similar decline observed in the pregnancies with SGA.

The pathophysiology of reduced PAPPA in GDM is unclear. Rojas-Rodriguez et al combined clinical and experimental data and concluded that PAPPA plays a crucial role in the adipose tissue remodelling that occurs in pregnancy (4). In mice genetically modified to lack PAPPA, pregnancy induces metabolic changes similar to GDM in humans. The authors suggest that there may be a causative association between PAPPA and diabetes.

Clinical implications

There is evidence that low PAPP-A acts as a ‘red flag’ for various adverse outcomes. Chromosomal abnormalities (including common trisomies and micro-deletion syndromes) and placental-related diseases are well-known to be associated with low PAPP-A (16,17). There is strong evidence that GDM is added to this list.

Lately Fruscalzo et al reported on the long term (average seven years) follow up of the metabolic status of the mother after an index pregnancy with known first trimester biochemistry (18). They found that women at the lower quartile of PAPPA were more likely to be treated for diabetes, hypertension and dyslipidemia regardless of whether they had GDM in the index pregnancy or not. The findings of this study suggest that low PAPPA, amongst other pregnancy events such as GDM and hypertension, may be a useful marker for future metabolic and cardiovascular health.

Research implications

PAPP-A appears to be promising as an important predictor for GDM and possibly for future cardiovascular health.

Future research will aim to develop multifactorial models in the spirit of the existing models for predicting chromosomal abnormalities or pre-eclampsia that will incorporate the background risk from the medical and family history as well as biophysical and biochemical indices.

The correlation of low PAPP-A with the latter development of metabolic syndrome regardless of the presence of GDM in the index pregnancy as suggested by Fruscalzo et al is very intriguing (18). We are aware that pre-eclampsia increases the long-life risk for adverse cardio-vascular events and further research will show whether low PAPP-A is a similar marker.

Strengths and Limitations

Our study was conducted in a population that had universal screening for GDM, which was not commonly the case in previous studies. In addition, women with diabetes unmasked by routine fasting glucose test at booking and/or women with impaired glucose tolerance with or without treatment prior to the first trimester screening at 11-13 weeks, were excluded. Unfortunately, the family and obstetric history and the information on first trimester fasting glucose levels was not available and these are disadvantages of the study.

The use of local birthweight reference charts and the choice of the 5th rather than the 10th centile for the definition of SGA strengthen the clinical relevance of our results.

Conclusion 

We confirm the association of low PAPPA with subsequent development of GDM. The clinician should be aware that low PAPPA values raise suspicion not only for fetal miscrosomia but also for GDM and that if GDM develops low PAPPA has limited value in predicting SGA. PAPPA is unlikely to be useful as a sole predictor of GDM but it can potentially be of value in a multi-factorial predictive model to assess the risk for GDM.

Authors contributions

Athena P. Souka: conceptualization, data analysis and interpretation, manuscript preparation, writing original draft, review and editing, Iliana V. Mengou: data analysis and interpretation, manuscript preparation, writing original draft, review and editing, Panagiotis Antsaklis: data analysis and interpretation, writing original draft and editing, Konstantinos Pateras: data analysis , Natasa Pithaki: data analysis, Angeliki Rouvali: data analysis, writing original draft and editing, George Daskalakis: data analysis, writing original draft and editing.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Ethics Approval

The ethics committee has approved the retrospective analysis of recorded data.

Competing Interests

The authors have no relevant financial or non-financial interests to disclose.

References
  1. Moon JH, Jang HC. Gestational Diabetes Mellitus: Diagnostic Approaches and Maternal-Offspring Complications. Diabetes & Metabolism Journal. 2022 Jan 31;46(1):3–14. https://doi.org/10.4093/dmj.2021.0335
  2. HAPO Study Cooperative Research Group, Metzger BE, Lowe LP, Dyer AR, Trimble ER, Chaovarindr U, et al. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med 2008;358: 1991-2002. http://dx.doi.org/10.1056/NEJMoa0707943
  3. Parikh NI, Gonzalez JM, Anderson CA, Judd SE, Rexrode KM, Hlatky MA, et al. Adverse pregnancy outcomes and cardiovascular disease risk: unique opportunities for cardiovascular disease prevention in women: a scientific statement from the American Heart Association. Circulation 2021;143:e902-16. http://dx.doi.org/10.1161/CIR.0000000000000961
  4. Rojas-Rodriguez R, Ziegler R, DeSouza T, Majid S, Madore AS, Amir N, Pace VA, Nachreiner D, Alfego D, Mathew J, Leung K, Moore Simas TA, Corvera S., PAPPA-mediated adipose tissue remodeling mitigates insulin resistance and protects against gestational diabetes in mice and humans. Sci Transl Med. 2020; 25:12(571). http://dx.doi.org/10.1126/scitranslmed.aay4145
  5. Spencer K, Cowans NJ. The association between gestational diabetes mellitus and first trimester aneuploidy screening markers. Ann Clin Biochem. 2013 Nov;50(Pt 6):603-10. doi: 10.1177/0004563213480493. Epub 2013 Jul 29. PMID:23897108. http://dx.doi.org/10.1177/0004563213480493
  6. Syngelaki A, Kotecha R, Pastides A, Wright A, Nicolaides KH., First-trimester biochemical markers of placentation in screening for gestational diabetes mellitus, Metabolism 2015; 64 (11): 1485-1489. http://dx.doi.org/10.1016/j.metabol.2015.07.015
  7. Yu-Ting L, Chie-Pein C, Fang-Ju S, Yi-Yung C, Liang-Kai W, Chen-Yu C, Associations between first-trimester screening biomarkers and maternal characteristics with gestational diabetes mellitus in Chinese women. https://doi.org/10.3389/fendo.2024.1383706
  8. Donovan BM, Nidey NL, Jasper EA, Robinson JG, Bao W, Saftlas AF, et al. First trimester prenatal screening biomarkers and gestational diabetes mellitus: A systematic review and meta-analysis. PLoS ONE 2018; 13(7): e0201319. http://dx.doi.org/10.1371/journal.pone.0201319
  9. Tranidou, A, Tsakiridis, I, Apostolopoulou, A, Xenidis, T, Pazaras, N, Mamopoulos, A, Athanasiadis, A, Chourdakis, M, Dagklis, T. Prediction of Gestational Diabetes Mellitus in the First Trimester of Pregnancy Based on Maternal Variables and Pregnancy Biomarkers. Nutrients 2024, 16, 120. https://doi.org/10.3390 /nu16010120
  10. Talasaz, Z.H, Sadeghi, R, Askari, F, Dadgar, S, Vatanchi, A. First trimesters Pregnancy-Associated Plasma Protein-A levels value to Predict Gestational diabetes Mellitus: A systematic review and meta-analysis of the literature. Taiwan. J. Obstet. Gynecol. 2018, 57, 181–189 http://dx.doi.org/10.1016/j.tjog.2018.02.003
  11. Thagaard, I.N, Krebs, L, Holm, J.-C, Lange, T, Larsen, T, Christiansen, M. Adiponectin and leptin as first trimester markers for gestational diabetes mellitus: A cohort study. Clin. Chem. Lab. Med. (CCLM) 2017, 55, 1805–1812. http://dx.doi.org/10.1515/cclm-2017-0427
  12. Kononova, O, Pristrom, A, Vasilkova, V, Mokhort, T. C-reactive protein and gestational diabetes mellitus. Endocr. Abstr. Biosci. 2013, 32, 604. http://dx.doi.org/10.1530/endoabs.32.P604
  13. Sweeting AN, Wong J, Appelblom H, Ross GP, Kouru H, Williams PF, Sairanen M, Hyett JA. A first trimester prediction model for gestational diabetes utilizing aneuploidy and pre-eclampsia screening markers. J Matern Fetal Neonatal Med 2018 Aug;31(16):2122-2130. http://dx.doi.org/10.1080/14767058.2017.1336759
  14. Sovio U, Gaccioli F, Cook E, Charnock-Jones DS, Smith GCS. Association between adverse pregnancy outcome and placental biomarkers in the first trimester: A prospective cohort study. BJOG. 2024 May;131(6):823-831. http://dx.doi.org/10.1111/1471-0528.17691
  15. Beneventi F, Simonetta M, Lovati E, Albonico G, Tinelli C, Locatelli E, Spinillo A. First trimester pregnancy-associated plasma protein-A in pregnancies complicated by subsequent gestational diabetes. Prenat Diagn 2011; 31: 523–528. http://dx.doi.org/10.1002/pd.2733
  16. Spencer K, Souter V, Tul N, Snijders R, Nicolaides KH. A screening program for trisomy 21 at 303 10-14 weeks using fetal nuchal translucency, maternal serum free beta-human chorionic 304 gonadotropin and pregnancy-associated plasma protein-A. Ultrasound Obstet Gynecol Off J Int 305 Soc Ultrasound Obstet Gynecol. 1999;13 (4):231-237. https://doi.org/10.1046/j.1469-0705.1999.13040231.x
  17. Tan MY, Wright D, Syngelaki A, et al. Comparison of diagnostic accuracy of early screening 333 for pre-eclampsia by NICE guidelines and a method combining maternal factors and biomarkers: 334 results of SPREE. Ultrasound Obstet Gynecol Off J Int Soc Ultrasound Obstet Gynecol. 335 2018;51(6):743-750. https://doi.org/10.1002/uog.19039
  18. Fruscalzo A, Cividino A, Rossetti E, Maurigh A, Londero AP, Driul L. First trimester PAPP-A serum levels and long-term metabolic outcome of mothers and their offspring. Sci Rep. 2020 Mar 20;10(1):5131. http://dx.doi.org/10.1038/s41598-020-61830-5