Decrease in peripheral natural killer cell level during early pregnancy predicts live birth among women with...

Previous studies have suggested that trophoblast cells inhibit the proliferation of peripheral natural killer cells and that the level of peripheral natural killer cells decrease in the middle and late pregnancy stage among healthy women. The change in peripheral natural killer cell level during early pregnancy and the relationship between the change in peripheral natural killer cell level and pregnancy outcomes among women with unexplained recurrent pregnancy loss have not been sufficiently explored.

This study aimed to characterize the level of prepregnancy peripheral natural killer cells in comparison with those in early pregnancy among women with unexplained recurrent pregnancy loss and to determine if the change in the level of peripheral natural killer cells from prepregnancy to early pregnancy can predict pregnancy outcomes.

In this prospective cohort study, 1758 women with recurrent pregnancy loss were recruited between January 2017 and December 2021 among whom 252 women with unexplained recurrent pregnancy loss had prepregnancy and early pregnancy (4-6 weeks gestation) peripheral natural killer cell measurements. These 252 women were divided into 2 groups, namely those with a lower gestational peripheral natural killer cell level (group 1) when compared with prepregnancy levels and those who did not (group 2). The respective outcomes of these groups in terms of live birth and pregnancy loss were comparatively analyzed using chi-square and Student's t tests. Candidate factors that could influence live birth were selected using the Akaike information criterion. The participates were then randomly divided into training and testing groups. A multivariable logistic regression analysis was performed and a nomogram was created to assess the possibility of live birth. The predictive accuracy was determined by the area under the receiver operating characteristic curve and validated by plotting the predicted probabilities and the observed probabilities. A Hosmer-Lemeshow test was used to assess the goodness of fit.

When early gestational peripheral natural killer cell levels were compared with prepregnancy peripheral natural killer cell levels, 61.5% (154) of women had a comparatively lower early-gestational peripheral natural killer cell level and 38.9% (98) of women had an increase or no change in the peripheral natural killer cell level. The live birth rate in group 1 was 89.0% (137/154), which was significantly higher than the rate of 49.0% (48/98) in group 2 (P<.001). A decrease in the peripheral natural killer cell level (odds ratio, 1.36; 95% confidence interval, 1.22-1.55; P<.001) and the anti-Muellerian hormone level (odds ratio, 1.41; 95% confidence interval, 1.14-1.81; P=.003) were important predicting factors for a higher live birth rate. Female body mass index (odds ratio, 0.97; 95% confidence interval, 0.82-1.15; P=.763) and parity (odds ratio, 1.61; 95% confidence interval, 0.71-4.12; P=.287) also were predicting factors. Furthermore, the area under the receiver operating characteristic curve of the model to diagnose of live birth was 0.853 with a sensitivity of 81.6% and a specificity of 78.0% using the training data set. And the Hosmer-Lemeshow test showed that the model was a good fit (p=6.068).

We report a comparative decrease in the peripheral natural killer cell levels in early gestation when compared with prepregnancy cell levels in more than 60% of women with unexplained recurrent pregnancy loss at 4 to 6 weeks of gestation. When compared with prepregnancy peripheral natural killer cell levels, a decrease in the peripheral natural killer cell level during early pregnancy might be a useful predictor of the live birth rate among women with unexplained recurrent pregnancy loss.