SAN FRANCISCO – Maternal levels of several proinflammatory cytokines are elevated shortly before the onset of spontaneous labor, and thus may help pinpoint the timing of this event, new data suggest.
In a nested, case-control study reported at the annual meeting of the Society for Maternal-Fetal Medicine, women at term whose labor began within 48 hours had significantly higher levels of interleukin-1 beta (IL-1B), interleukin 6 (IL-6), and tumor necrosis factor–alpha (TNF-alpha) than did their counterparts whose labor did not begin for at least 14 days.
"This finding supports the hypothesis that inflammatory activation is present shortly prior to clinically recognizable labor," said Dr. Elizabeth Ramsey Unal of the Medical University of South Carolina, Charleston.
The investigators were also able to determine cutoff values for the cytokines that had high negative predictive values for labor onset and, in some cases, good sensitivity.
Clinically, information from this and related studies "could be of great value in identifying maternal biomarkers to select women who are either very good vs. very poor candidates for induction of labor, or could potentially also be used for labor triage in specific cases," Dr. Unal noted.
Previous research has found elevated levels of proinflammatory markers during labor. "What is unclear, however, is the timing of this involvement: Is inflammation solely a result of the labor process, or is inflammation actually involved in the steps that initiate labor?" she said, explaining the study’s rationale.
The investigators drew their study sample from a cohort of healthy, nulliparous women with a singleton pregnancy who enrolled in a related study during a routine office visit at term (defined as greater than or equal to 37 weeks’ gestation) and gave a blood sample at that time.
They were excluded if they had any signs or symptoms of latent labor, had an autoimmune disease, had any infection in the third trimester, or used anti-inflammatory medications during pregnancy.
Analyses were based on 20 women who had spontaneous onset of labor within 48 hours (cases) and 80 women who did not have spontaneous onset of labor for at least 14 days (controls).
The investigators measured serum levels of seven cytokines (IL-1B, IL-4, IL-6, IL-8, IL-10, TNF-alpha, and interferon-gamma) that "were selected to encompass the signature cytokines for the Th1 and Th2 and Treg pathways," Dr. Unal explained.
The women studied were 22 years old, on average; 44% were Hispanic, 31% were white, and 25% were black.
Compared with their counterparts whose labor did not begin for at least 14 days, those whose labor began within 48 hours had an older mean gestational age (39.2 vs. 37.3 weeks) and greater mean cervical dilatation (2 cm vs. 1 cm) at enrollment. But the groups were similar with respect to group B streptococcus positivity and the rate of infections in the first and second trimester.
The women whose labor started within 48 hours had a younger gestational age at delivery (39.2 weeks vs. 40.4 weeks) and smaller infants (3,205 g vs. 3,485 g). However, the groups were similar with respect to mode of delivery, white blood cell count on admission, and fever during labor.
Study results showed that none of the levels of the seven cytokines was correlated either with gestational age at enrollment or maternal body mass index, Dr. Unal reported.
Compared with their counterparts whose labor did not begin for at least 14 days, the women whose labor began within 48 hours had higher levels of IL-1B (P less than .01), IL-6 (P = .03), and TNF-alpha (P = .02). Levels of the other cytokines did not differ between the two groups.
In ROC (receiver operating characteristic) curve analyses, TNF-alpha was the best predictor of labor onset within 48 hours, with an area under the curve of 0.703.
Using these curves, the investigators selected cutoff values for each cytokine to optimize the negative predictive value, according to Dr. Unal.
By this process, high levels of IL-1B (greater than 0.5 pg/mL), IL-6 (greater than 4.0 pg/mL), and TNF-alpha (greater than 0.6 pg/mL) had negative predictive values of 89%, 84%, and 92%, respectively. Or, stated another way, "in our data, labor was very unlikely to occur within 48 hours in the presence of low levels" of IL-1B, IL-6, and TNF-alpha.
Also, at these cutoff values, IL-1B and TNF-alpha had good sensitivity for predicting labor onset within 48 hours, with values of 70% and 80%, respectively.
In a regression analysis, cervical dilatation at enrollment was a better predictor of labor onset, according to Dr. Unal. But it is possible that adding cytokines to cervical dilatation could improve prediction.