Introduction

Gestational diabetes mellitus (GDM) affects approximately 2% of all pregnant women in southern Sweden [1]. Pregnancies complicated by GDM reveal an increased risk of hypertensive disorders and future manifest diabetes for the mother. GDM may also have adverse effects on foetal development, increasing the risk of macrosomia, perinatal complications and the risk for future obesity in the offspring [2].

Increased levels of free fatty acids and advanced glycation end products, which lead to the production of pro-inflammatory cytokines from adipose tissue and subsequent chronic inflammation, are typical for GDM [3]. Inflamed adipose tissue causes monocyte infiltration and activation of tissue resident macrophages. The endocytic scavenger receptor CD163 is expressed on macrophages active in adipose tissue [4]. A large prospective study showed that increased serum levels of soluble CD163 (sCD163) could predict future development of type 2 diabetes independent of age and body mass index (BMI) [5]. Studies on sCD163 in GDM are few and inconclusive with one study reporting increased levels in GDM [6], while another found no difference compared to in pregnant women with normal glucose tolerance (NGT) [7].

Soluble tumour necrosis factor-like weak inducer of apoptosis (sTWEAK) exert multiple different cellular activities on a wide array of cells, which are important in regulating the anti-inflammatory/inflammatory balance, and has also been suggested as a ligand for CD163 [8, 9]. Studies made on circulating sTWEAK in diabetes report significantly decreased serum levels in both type 1 diabetes and type 2 diabetes [10, 11]. One study reported lower levels of sTWEAK in women with GDM compared to women with NGT [7].

GDM is today universally diagnosed during late pregnancy. Treating hyperglycaemia during pregnancy is known to reduce the risk of complications [12], but the effect of intervention is often limited due to the late diagnosis. Finding early pregnancy biomarkers capable of identifying women at increased risk for GDM could motivate earlier and more efficient interventions, minimising the risk for adverse outcomes [13]. Identifying GDM patients at increased risk of later glucose intolerance development is of importance and could act as an incentive for necessary lifestyle changes and pharmacological interventions in these women [14].

The aims of this study was to investigate plasma levels of sCD163 and sTWEAK in pregnant women with and without GDM in early pregnancy and their association to later glucose intolerance.

Participants and methods

Results

No difference in plasma levels of sCD163 could be observed between women with or without GDM diagnosed in early pregnancy (318 [246–395] ng/ml and 321 [243–432] ng/ml respectively; p = 0.86). Plasma levels of sTWEAK were however decreased in women with early pregnancy GDM (0.71 [0.4–1.75] ng/ml) compared to controls (1.38 [0.63–4.86] ng/ml; p = 0.003). Plasma sTWEAK levels were not associated to participant age (p = 0.61) or BMI (p = 0.55), and no correlation between sCD163 and sTWEAK could be observed (p = 0.86). There was no association between sCD163 (p = 0.16) or sTWEAK (p = 0.27) and glucose intolerance development following GDM in this study. Women with GDM had more often a family history of diabetes (57% (36/63)) compared to controls (31% (22/70; p = 0.003). Neither sCD163 nor sTWEAK levels were found to differ in women with or without a family history of diabetes (p = 0.48 and p = 0.27 respectively). Some of the patients have had previous GDM (10% (6/59)). No association could be observed between previous GDM and sCD163 (p = 0.76) or sTWEAK (p = 0.74) levels.

In order to increase reproducibility through different modes of analysis, sTWEAK levels were arranged into tertiles (T₁-T₃). Plasma sTWEAK levels in the lowest tertile (T₁ (<0.63ng/ml)) were associated with an increased risk of GDM (OR: 2.48, CI⁹⁵ (1.20–5.10); p = 0.014), while sTWEAK levels in the highest tertile T₃ (≥1.75ng/ml) showed a tendency for a decreased risk of GDM (OR: 0.49, CI⁹⁵ (0.24–1.00; p = 0.05) (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Pregnant women in early pregnancy with plasma levels of sTWEAK in the lowest tertile had increased risk of being diagnosed with GDM.

A tendency of decreased risk of GDM could be seen in the highest sTWEAK tertile.

https://doi.org/10.1371/journal.pone.0216728.t001

Discussion

This study has investigated sCD163 and sTWEAK levels in women diagnosed with GDM in early pregnancy. While no difference in sCD163 levels could be observed, decreased levels of sTWEAK were associated with the development of GDM. A previous study has found increased levels of sCD163 at delivery in women diagnosed with GDM in late pregnancy [6], while another study where sCD163 was measured in late pregnancy did not observe a difference when compared to women with NGT [7] which is in accordance with our findings in early pregnancy. The study by Bari et al. [6] investigated sCD163 levels at parturition in women with elective caesarean section in a small group of GDM patients (n = 18) and BMI matched controls (n = 20). They suggested sCD163 to be a significant predictor of GDM, although the precision was not presented. The late sampling and small sample size could be a reason for the discrepancy among the studies. Adipose tissue explants and placenta showed increased sCD163 release in GDM and the placenta was suggested to be a significant source of sCD163 in GDM. Simon-Muela et al. measured sCD163 levels in late pregnancy GDM (n = 66) and NGT (n = 71) [7]. They found sCD163 levels to be associated to a poor metabolic profile, but could not find any significant difference in sCD163 levels between the groups. In contrast to the study by Bari et al, sCD163 release from adipose tissue and placenta was not studied by Simon-Muela et al. nor by this study.

GDM is a condition characterised by chronic inflammation and decreased sTWEAK levels have been observed in diseases associated with chronic inflammation and insulin resistance [10, 11]. Circulating sTWEAK was in this study decreased in early pregnancy GDM independent of participant age and BMI and family history of diabetes. Our results are in agreement with a Spanish study made in later pregnancy [7]. Membrane bound CD163 has been reported to bind sTWEAK in circulation [8]. We could however, in contrast to the Spanish study [7], not observe any correlation between circulating sCD163 and sTWEAK. By arranging the sTWEAK levels into tertiles, we facilitate the possibility to reproduce the findings in this study. Women with sTWEAK levels in the lowest tertile had significantly increased risk of GDM, while there was a tendency for a decreased risk of GDM with sTWEAK levels in the highest tertile. In this study we also sought to determine any associations between sCD163, sTWEAK and the development of glucose intolerance following a pregnancy complicated by GDM. Neither sCD163 nor sTWEAK were however found to be associated to such development, possibly explained by the low number of patients who developed later glucose intolerance.

A strength with this study is that the GDM patients included are part of a unique screening program that enables all pregnant women in Lund to undergo an OGTT, resulting in a heterogeneous group of participants. The women included as controls are matched in age and BMI, and they are recruited from the same geographical region as the patients, further minimising confounding effects. Limitations include a limited number of participants and lacking the possibility for a follow-up during late pregnancy. A power calculation suggests 145 participants per group for sTWEAK and 759 for sCD163 with an alpha value of 0.05 and beta value of 0.8 for later glucose intolerance. It is important to notice however that the distribution for these molecules in non-parametric which may skew the power calculation.

Conclusions

This study reports decreased levels of sTWEAK in women with early pregnancy GDM, independent of age and BMI. No difference could be observed with regard to sCD163 levels, and neither sTWEAK nor sCD163 was associated with later glucose intolerance. A future prospective cohort study should focus on elucidating the potential for sTWEAK to be used as an early pregnancy pre-screening biomarker for GDM development as this could possibly alleviate the need for onerous screening procedures in women with low risk for developing hyperglycaemia. More focus should also lie in earlier identification of women at risk of developing GDM as this could motivate more efficient treatment interventions at an earlier time point, minimizing adverse outcomes.

Supporting information

Acknowledgments

The authors would like to thank Ms. Birgitte Ekholm, the Diabetes Research Laboratory, Lund, for her excellent technical assistance. The authors also want to thank the midwives at the maternal healthcare centres in Lindängen in Malmö, Dalby and Paletten in Staffanstorp.