IL-17A and Chronic Stress in Pregnant Women at 24-28 Weeks Gestation

Abstract

Background:

Allostatic Load (AL) is a biopsychosocial model that suggests chronic psychosocial stress leads to physiologic dysregulation and poor outcomes. The purpose of this study was to examine AL in pregnant women operationalized using pro-inflammatory cytokines and psychosocial indicators and perinatal outcomes.

Objectives:

In pregnant women, identify relationships between circulating cytokines/chemokines and the Prenatal Distress Questionnaire (PDQ), Maternal Antenatal Attachment Scale (MAAS), Emotional and Quotient Inventory (EQi), Life Experiences Scale (LES), and demographics.

Methods:

A cross-sectional design was used to recruit pregnant women between 24–28 weeks gestation. Blood and stress/emotional indicators were obtained after informed consent. Plasma was abstracted to simultaneously measure 29 cytokines/chemokines using a multiplex array. Cytokine/chemokine levels were compared with continuous variables using Spearman’s Rho and with categorical variables using Mann-Whitney U.

Results:

Twenty-five women with medically high-risk (n = 16) and low-risk (n = 9) pregnancies consented. The majority of women were White (68%) with a mean age of 29 years (5.9). While several cytokines and chemokines showed significant correlations with the stress/emotional indicators, only interleukin-17 (IL-17A) was significantly associated with all of the indicators [PDQ (r_s = 0.528; p = 0.012); MAAS (r_s = −0.439; p = 0.036); EQi total (r_s = −0.545; p = 0.007)]; LES (r_s = 0.458, p = 0.032); and birthweight (r_s = −0.499; p = 0.013), and race (p = 0.01).

Discussion:

Increased levels of IL-17A, a known cytokine associated with chronic stress and with poor perinatal outcomes, was associated with high prenatal distress, low maternal attachment, and lower emotional intelligence in pregnant women. Increased levels of IL-17A also were associated with lower birthweight and non-White race. Results support the model of AL in pregnant women and highlight IL-17A as a potential biomarker of AL during pregnancy.

Chronic psychosocial stress is believed to affect physiological processes and exacerbate multifactorial phenomena as described by McEwen’s allostatic load (AL) theory (Juster, McEwen, & Lupien, 2010; McEwen, 1998). AL is a biopsychosocial theory suggesting psychosocial stress affects physiologic dysregulation which eventually leads to “wear and tear” on body systems and results in poor outcomes (McEwen, 1998). Relevant complex phenomena linked to multiple biological mechanisms as well as chronic psychosocial profiles in the perinatal population include preterm birth (PTB) and low birthweight (LBW) (National Center for Health Statistics, 2017). PTB, defined as parturition prior to 37 weeks gestation, and LBW, defined as weighing < 2500g at birth, are perinatal outcomes prioritized as public health concerns (United States Department of Health and Human Services, Office of Disease Prevention and Health Promotion, 2011).

Allostatic load explains how chronic psychosocial stress affects physiologic dysregulation leading to poor health outcomes. Although chronic stress is a challenging concept to quantify, we identified four stress/emotional indicators to measure in pregnant women that may be related to chronic stress: pregnancy distress, maternal attachment, major life events, and emotional intelligence. Pregnancy distress refers to the emotional distress specific to pregnancy and is consistently reported as a negative stress/emotional indicator relating to PTB (Brunton, Dryer, Saliba, & Kohlhoff, 2015; Dunkel Schetter, Niles, Guardino, Khaled, & Kramer, 2016; Guardino & Schetter, 2014; Haas et al., 2015; Liou, Wang, & Cheng, 2016; Rose, Pana, & Premji, 2016; Shapiro, Fraser, Frasch, & Séguin, 2013). Pregnancy distress was used to identify state responses specific to pregnancy but may provide a more global understanding of trait responses to situations. Maternal antenatal attachment also is believed to play an important role in psychosocial stress, albeit as a buffer of stress, affecting infant growth and development in utero and early childhood (Alhusen, Hayat, & Gross, 2013; Branjerdporn, Meredith, Strong, & Garcia, 2017; Pisoni et al., 2016). Attachment to the fetus may provide understanding of a more in-depth trait response of the mother and provides opportunity for future interventional therapies.

Major life events, especially negative events, are believed to affect perinatal outcomes, specifically increasing the risk of PTB (Barrios, Sanchez, Qiu, Gelaye, & Williams, 2014; Witt et al., 2014). Quantifying major life events may provide insight for an objective measure of stress. Emotional intelligence is an emerging concept that measures the essential emotional skills and self-awareness used in personal, professional, and public settings and interactions (Shonkoff, 2012). The impact of emotional intelligence on perinatal outcomes is unknown because this concept has received little attention in the perinatal population or as a risk factor of PTB. Theoretically, however, pregnant women with lower emotional intelligence would have decreased self-regulatory mechanisms to confront daily, or even chronic, stressors which may result in more physiologic dysregulation from stress. We chose the concept of emotional intelligence because of the emerging literature of relationships between adverse childhood events and emotional intelligence (Shonkoff, 2012).

Inflammation has been used to identify dysregulation in previous AL studies and is relevant to PTB and LBW. Inadequate responses of the immune system measured at single time points (e.g., exaggerated inflammatory response measured as cytokines) in pregnant women with PTB have been previously identified (Lucaroni et al., 2018). Specifically, a systematic review authored by Lucaroni and colleagues (2018) reported increased interleukin-6 (IL-6) levels measured in maternal serum are associated with increased odds ratio of 2 (95%CI 1–3) and increased likelihood value of 12. As noted by the authors, however, more prospective studies are needed in various subsets of pregnant women to identify if specific cytokine levels are sufficient for the prediction of PTB. Furthermore, inadequate immune responses also have been associated with LBW during the preconception period (Pearce et al., 2016), but no studies to date were identified that investigated prenatal cytokine levels associated with LBW.

The emerging concept of AL is a relevant theoretical framework in the perinatal population, yet, no studies to date have used the AL model to measure prospective prenatal cytokine levels, sociodemographics, and specific stress/emotional indicators. The specific aim of this study was to identify correlations between circulating cytokine and chemokine levels at 24–28 weeks gestation with maternal antenatal attachment, prenatal distress, life events, emotional intelligence, maternal race, gestational age at birth, and birthweight.

Methods

Pregnant women already enrolled in a larger study were recruited during a clinic visit between 24–28 weeks for this pilot study. The clinic is located in an academic perinatal center in the Midwest region. Maternal patients seen by the maternal-fetal-medicine and general obstetrics practice groups were included in the larger study. This study was approved by the institutional review board from the primary investigator’s university. Inclusion criteria were women previously enrolled in the larger study and scheduled for routine metabolic/glucose challenge between 24–28 weeks gestation. Although not an exclusion criteria, pregnant women not participating in the metabolic/glucose challenge for clinical reasons (e.g., previous diagnosis were not approached. A member of the research team with ethical access approached eligible pregnant women, obtained informed consent using the teach-back method to verify understanding. After consent, whole blood was obtained, samples were centrifuged, and the plasma was extracted and frozen at −80°C until analysis. Cytokines and chemokines were assessed using the 30 V-plex^™ assay kit and Discovery Workbench^® 4.0 from Meso Scale Discovery^® (Rockville, Maryland). At time of analysis, the 30 V-plex^™ assay kit only included 29 cytokines as MCP-4 was not available from the manufacturer. On the same day as blood collection, stress/emotional indicators also were obtained using validated and reliable tools to measure the four stress/emotional indicators: (a) Revised Prenatal Distress Questionnaire (NuPDQ); (b) Maternal Antenatal Attachment Scale (MAAS); (c) Life Experiences Survey (LES); and (d) Emotional Quotient Inventory 2.0 (EQi 2.0). Demographic data (i.e., birthweight; gestational age, race/ethnicity, maternal age; insurance coverage) were extracted through the electronic health records. Cytokine and chemokine levels were correlated with stress/emotional indicator scores using Spearman’s Rho and SPSS v25 (SPSS Inc, Chicago, IL).

The Maternal Antenatal Attachment Scale (MAAS; Condon, 1993, Condon & Corkindale, 1997) measures antenatal global maternal attachment using 19 items scored on a 5-point Likert-scale (higher levels indicate higher levels of attachment) with Cronbach’s alpha ranging between .69 and .82 (Condon, 1993; Schwerdtfeger & Goff, 2007; Thomas, 2014). The Revised Prenatal Distress Questionnaire (NuPDQ; Lobel et al., 2008) includes 17 questions about pregnancy-specific stress using a 3-point Likert scale (higher levels indicate higher distress). Cronbach’s alpha ranges from .59 −.79 (Alderice, Lynn, & Lobel, 2012). The Life Experiences Survey (LES; Sarason, Johnson, & Siegel, 1978) is a 57-item tool that measures the occurrence of specific events (y/n) experienced in the past year; higher numbers means more major life events occurred. Test-retest reliability ranges from .63 −.64 (Saranson, Johnson, & Seigel, 1978). The Emotional Quotient Inventory 2.0^® (North Tonawanda, NY) measures social-emotional intelligence using 133 questions and a 5-point Likert scale (higher scores indicate higher levels of emotional intelligence). Test-retest reliability was .92 at 2–4 weeks; Cronbach alpha coefficients range from .77 to .91 (Multi Health Systems, 2011).

Because this was a pilot study, our goal was to recruit 30 pregnant women consistent with sample sizes appropriate for pilot studies (Hertzog, 2008). Birthweight and gestational age were measured as continuous variables to identify significant correlations between birthweight, age, cytokines, and stress/emotional scores in this small sample. Descriptive and correlational statistics were completed using SPSS version 25 (SPSS Inc, Chicago, IL). Nonparametric tests were used because the assumptions of normality were not met; p-values were not adjusted for multiple comparisons because of the small sample size and exploratory aim of the study.

Results

Twenty-five women with medically high-risk (n = 16) and low-risk (n = 9) pregnancies were enrolled in the study. Missing data include blood (n = 1), EQi (n = 1), MAAS (n = 1), PDQ (n = 2), and LES (n = 2). Demographic data identifies the majority of women were White (n = 17; 68%) and all 25 women were self-identified as non-Hispanic. Perinatal outcomes include preeclampsia (n = 4), gestational diabetes (n = 4), and diagnosis of morbidly obese (n = 5). Newborn data include median birthweight of 3240 grams (ranging from 2,010–4,051 grams) and median gestational age at birth of 38.9 weeks (ranging from 34 41.3 weeks).

Table 1 shows the correlations between cytokine/chemokine levels, stress/emotional indicators, and birthweight. Several of the cytokines and chemokines were significantly correlated with stress/emotional questionnaires and birthweight. Notably, IL-17A levels were significantly correlated with all stress scales and birthweight suggesting that pregnant women with high levels of IL-17A were more likely to have lower emotional intelligence scores EQi total (r_s = −0.545; p = 0.007), lower antenatal attachment scores MAAS (r_s = −0.439; p = 0.036), higher prenatal distress scores PDQ (r_s = 0.528; p = 0.012), more major life events in the past year LES (r_s = 0.458, p = 0.032) and deliver a newborn with a lower birthweight (r_s = −0.499; p = 0.013). Table 2 identifies significant differences in cytokine/chemokine levels based on race (White vs. non-White). Several cytokines and chemokines, including IL-17A, were significantly different based on race suggesting that pregnant women with higher levels of these specific proteins were more likely to be non-White (p = 0.01). No significant differences were found with gestational age.

Table 1.

Correlations Between Maternal Circulating Cytokine and Chemokine Levels at 24-28 Weeks Gestation with Stress/Emotional Indicators

	EQi						MAAS			PDQ	LES	BW
	Total	I. Self- Perception	II. Inter- personal	III. Decision Making	IV. Self- Expression	V. Stress Management	Total	I. Quality of Attachment	II. Time in Attachment	Total	Total	Newborn Birthweight
IL-17A	rs= −.545 p= .007	rs= −.467 p= .025	rs= −.519 p= .011	rs= −.523 p= .010	rs= −.336 p= .117	rs= −.664 p= .001	rs= −.439 p= .036	rs= −.417 p= .048	rs= −.355 p= .096	rs= .528 p= .012	rs= .458 p= .032	rs= −.499 p= .013
IFN-y	rs= −.350 p= .101	rs= −.328 p= .127	rs= −.534 p= .009	rs= −.277 p= .201	rs= −.137 p= .533	rs= −.408 p= .054	rs= −.341 p= .111	rs= −.282 p= .193	rs= −.288 p= .183	rs= .521 p= .013	rs= .130 p= .563	rs= −.200 p= .349
IL-10	rs= −.220 p= .314	rs= −.276, p= .203	rs= −.190 p= .386	rs= −.278 p= .199	rs= −.192 p= .380	rs= −.401 p= .058	rs= −.233 p= .284	rs= −.038 p= .862	rs= −.228 p= .295	rs= .037 p= .870	rs= .184 p= .401	rs= −.253 p= .233
IL-1b	rs= −.204 p= .351	rs= −.077 p= .728	rs= −.086 p= .697	rs= −.325 p= .130	rs= −.074 p= .736	rs= −.396 p= .061	rs= −.141 p= .522	rs= −.248 p= .255	rs= .014 p= .950	rs= .353 p= .107	rs= .100 p= .657	rs= .030 p= .889
IL-6	rs= −.079 p=.721	rs= −.016 p= .943	rs= −.240 p= .271	rs= −.043 p= .845	rs= .139 p= .526	rs= −.154 p= .484	rs= .004 p= .986	rs= .210 p= .336	rs= −.099 p= .652	rs= .280 p= .207	rs= −.119 p= .599	rs= .388 p= .061
IL-8	rs= −.536 p= .008	rs= −.368, p= .084	rs= −.455 p= .029	rs= −.537 p= .008	rs= −.383 p= .071	rs= −.582 p= .004	rs= −.229 p= .292	rs= −.368 p= .084	rs= −.207 p= .343	rs= .348 p= .112	rs= .293 p= .185	rs= −.053, p= .806
TNFa	rs= −.439 p= .036	rs= −.242 p= .266	rs= −.460 p= .027	rs>= −.497 p= .016	rs= −.167 p= .446	rs= −.407 p= .054	rs= −.120 p= .584	rs= −.246 p= .259	rs= −.118 p= .591	rs= .388 p= .074	rs= .284 p= .201	rs= −.097 p= .654
Eotaxin	rs= −.565 p= .005	rs= −.433, p= .039	rs= −.272 p= .209	rs= −.572 p= .004	rs= −.464 p= .026	rs= −.553 p= .006	rs= −.211 p= .335	rs= −.182 p= .405	rs= −.150 p= .496	rs= .457 p= .033	rs= .132 p= .559	rs= .035 p= .872
Eotaxin-3	rs= −.266 p= .220	rs= −.093 p= .672	rs= −.020 p= .929	rs= −.240, p= .270	rs= −.265 p= .222	rs= −.285 p= .187	rs= −.109 p= .619	rs= −.118 p= .592	rs= −.059 p= .790	rs= .231 p= .301	rs= .235 p= .292	rs= −.199 p= .351
IL-8-ha	rs= −.148 p= .501	rs= .135 p= .539	rs= −.088 p= .689	rs= −.287 p= .184	rs= −.006 p= .977	rs= −.136 p= .535	rs= .075 p= .734	rs= −.129 p= .558	rs= .214 p= .326	rs= .428 p= .047	rs= .201 p= .369	rs= −.155 p= .469
IP-10	rs= −.506 p= .014	rs= −.372 p= .081	rs= −.343 p= .109	rs= −.513 p= .012	rs= −.399 p= .059	rs= −.516 p= .012	rs= −.189 p= .389	rs= −.110 p= .617	rs= −.150 p= .496	rs= .418 p= .053	rs= .135 p= .549	rs= −.102 p= .636
MCP-1	rs= −.522 p= .011	rs= −.425 p= .043	rs= −.203 p= .353	rs= −.496 p= .016	rs= −.469 p= .024	rs= −.501 p= .015	rs = −.175 p= .425	rs = −.132 p= .550	rs= −.148 p= .500	rs= .387 p= .075	rs= .017 p= .940	rs= .021 p= .923
MDC	rs= −.575 p= .004	rs= −.414 p= .050	rs=− .394 p= .063	rs= −.549 p= .007	rs= −.457 p= .028	rs= −.542 p= .008	rs= −.294 p= .174	rs= −.270 p= .213	rs= −.198 p= .364	rs= .342 p= .120	rs= .155 p= .492	rs= −.006 p= .977
MIP-1a	rs= −.357 p= .095	rs= −.258, p= .235	rs= −.217 p= .320	rs= −.407 p= .054	rs= −.187 p= .392	rs= −.357 p= .095	rs= −.085 p= .701	rs= −.023 p= .916	rs= −.102 p= .642	rs= .253 p= .257	rs= .019 p= .932	rs= −.208 p= .330
MIP-1b	rs= −.514 p= .012	rs= −.455 p= .029	rs= −.310 p= .050	rs= −.484 p= .019	rs= −.415 p= .049	rs= −.540 p= .008	rs= −.304 p= .159	rs= −.299 p= .166	rs= −.163 p= .459	rs= .484 p= .022	rs= .107 p= .637	rs= .077 p= .719
TARC	rs= −.520, p= .011	rs= −.344 p= .108	rs= −.225 p= .301	rs= −.525 p= .010	rs= −.468 p= .024	rs= −.527 p= .010	rs= −.217 p= .319	rs= −.187 p= .392	rs= −.173 p= .430	rs= .271 p= .222	rs= .009 p= .968	rs= .003 p= .987
IL-12p40	rs= −.283 p= .191	rs= −.064 p= .772	rs= −.223 p= .306	rs= −.356, p= .096	rs= −.314 p= .145	rs= −.193 p= .377	rs= −.021 p= .923	rs= −.007 p= .975	rs= −.014 p= .950	rs= −.096 p= .671	rs= .171 p= .448	rs= .141 p= .511
IL-15	rs= −.161 p= .463	rs= −.148 p= .500	rs= −.253 p= .244	rs= −.217 p= .319	rs= .020, p= .927	rs= −.165 p= .452	rs= .060 p= .785	rs= .105 p= .633	rs= −.005, p= .980	rs= .470 p= .027	rs= .209 p= .351	rs= −.249 p= .241
IL-16	rs= −.151 p= .492	rs= .004 p= .984	rs= .086 p= .696	rs= −.268 p= .216	rs= −.154 p= .483	rs= −.111 p= .614	rs= .129 p= .556	rs= −.191 p= .382	rs= .222 p= .308	rs= .107 p= .634	rs= .208 p= .353	rs= .132 p= .538
IL-1a	rs= .182 p= .405	rs= .098 p= .656	rs= −.125 p= .571	rs= .353 p= .098	rs= .178 p= .418	rs= .111 p= .613	rs= −.216 p= .321	rs= −.051 p= .816	rs= −.153 p= .485	rs= −.022 p= .922	rs= −.064 p= .779	rs= .079 p= .713
IL-7	rs= −.452 p= .030	rs= −.279 p= .198	rs= −.357 p= .095	rs= −.448 p= .032	rs= −.362 p= .090	rs= −.541 p= .008	rs= −.194 p= .375	rs= −.116 p= .598	rs= −.173 p= .431	rs= .442 p= .040	rs= .120, p= .596	rs= .109 p= .613
TNF-b	rs= −.103 p= .640	rs= .103 p= .640	rs= −.107 p= .626	rs= −.207 p= .343	rs= −.081 p= .713	rs= −.198 p= .366	rs= .085 p= .699	rs= .122 p= .579	rs= .126 p= .566	rs= .242 p= .278	rs= −.026 p= .910	rs= −.110 p= .610
VEGF	rs= .048 p= .830	rs= −.023 p= .918	rs= −.186 p= .395	rs= .177 p= .418	rs= .152 p= .488	rs= .039 p= .861	−rs= −.311 p= .149	rs= −.164 p= .453	rs= −.289 p= .181	rs= .044 p= .846	rs= −.054 p= .812	rs= .231 p= .277

Note: significant correlations (p < .05) for Spearman’s Correlations (r_s) are bolded. IFN=interferon; IL=interleukin; TNF= tumor necrosis factor; IP= inducible protein; MCP= monocyte chemoattractant protein; MDC= macrophage-derived chemokine; MIP= macrophage inflammatory protein; TARC= thymus and activation-regulated chemokine; GM-CSF=granulocyte-macrophage colony-stimulating factor; VEGF=vascular endothelial growth factor ; EQi=Emotional and Quotient Inventory for social-emotional maturity; MAAS=Maternal Antenatal Attachment Score; PDQ=Prenatal Distress Questionnaire; LES=Life Events Scale; BW=Birthweight;

^*n = 22-24 from incomplete study questionnaires for 2 of the subjects. IL-12p70, IL-13, IL-2, IL-4, IL-5, and GM-CSF had median values of 0.1 or less from many sample below the lower level of detection and thus were not included in the table.

Table 2.

Means and Standard Deviations (SD) for Maternal Circulating Cytokine and Chemokine Levels By Race

	White (n = 17) Mean(SD)	Non-White (n = 7) Mean(SD)
IL-17A*	.8 (.6)	2.1 (2.1)
IFN-y	5.7 (5.1)	6.9 (3.3)
IL-10	.3 (.1)	.3 (.1)
IL-1b*	.04 (.1)	1.1 (2.7)
IL-6	.9 (.7)	1.3 (1.3)
IL-8	2.3 (1.2)	3.6 (2.2)
TNFa	1.8 (.5)	2.2 (.7)
Eotaxin***	31.8 (17.3)	94.7 (52.8)
Eotaxin-3**	12.8 (12.9)	37.4 (23.7)
IL-8-ha**	10.0 (23.9)	82.8 (55.3)
IP-10***	67.6 (62.5)	216.9 (89.6)
MCP-1**	19.7 (15.9)	72.5 (43.6)
MDC**	485.6 (287.0)	1138.1 (470.7)
MIP-1a**	7.29 (4.58)	15.87 (7.77)
MIP-1b**	21.9 (14.7)	67.4 (15.0)
TARC**	11.7 (13.3)	40.5 (43.0)
IL-12p40	83.9 (46.9)	87.0 (29.6)
IL-15	2.4 (.4)	2.8 (.9)
IL-16	350.4 (558.2)	1439.2 (2960.2)
IL-1a	4.3 (11.7)	.8 (.6)
IL-7*	1.6 (1.1)	4.1 (2.6)
TNF-b	.3 (.1)	.3 (.1)
VEGF	16.4 (24.7)	8.0 (7.3)

Note. IFN=interferon; IL=interleukin; TNF= tumor necrosis factor; IP= inducible protein; MCP= monocyte chemoattractant protein; MDC= macrophage-derived chemokine; MIP= macrophage inflammatory protein; TARC= thymus and activation-regulated chemokine; GM-CSF=granulocyte-macrophage colony-stimulating factor; VEGF=vascular endothelial growth factor.

IL-12p70, IL-13, IL-2, IL-4, IL-5, and GM-CSF had median values of 0.1 or less from many sample below the lower level of detection and thus were not included in the table.

^*p < .05 between White vs Non-White using Mann Whitney-U tests

^**p < .01 between White vs Non-White using Mann Whitney-U tests

^***p < .001 between White vs Non-White using Mann Whitney-U tests

Pregnant women diagnosed with gestational diabetes (n = 4) had significantly higher IL-10 (p = 0.025) and IL-6 (p = 0.036) levels and the newborn infants had significantly higher birthweights (p = 0.009). Pregnant women diagnosed with morbid obesity had significantly higher TNF-α (p = 0.02) and IL-6 (p = 0.001) levels. No other significant differences were found between the stress/emotional questionnaires, cytokine levels, race, or birthweight in pregnant women diagnosed with preeclampsia, gestational diabetes, or morbid obesity.

Discussion

Herein, we used the AL model to examine stress/emotional indicators on a subset of pregnant women and identify associations between inflammatory biomarkers and psychosocial stress. These stress/emotional indicators were administered on the same day as a clinical blood draw obtained between 24–28 weeks gestation. Several correlations were identified between cytokines/chemokines and scores for the stress/emotional indicators. The most interesting finding was IL-17A and the significant relationships with birthweight, maternal race, and all of the stress/emotional indicators administered for the current study.

Briefly, IL-17A is a member of a subfamily of pro-inflammatory cytokines that is composed of six (i.e. A-F) unique IL-17 ligands (Onishi & Gaffen, 2010). Recently, elevated IL-17A has been shown to be associated with various psychological diseases, suggesting a possible mechanistic role in driving behavioral changes. For example, in a cohort of patients with rheumatoid arthritis, IL-17A levels were positively correlated with the severity of comorbid anxiety (Liu, Ho, & Mak, 2012). Additionally, IL-17A was associated with poorer cognitive status in patients with depressive symptoms following stroke (Swardfager et al., 2014), and IL-17A has been demonstrated to directly cause depressive-like behavior in rodent models (Nadeem et al., 2017). Furthermore, patients with posttraumatic stress disorder (PTSD) have been demonstrated to have increased IL-17A (Wang, Mandel, Levingston, & Young, 2016; Zhou et al., 2014). One hallmark of many psychological disorders is dysregulation of the autonomic nervous system, and our group was the first to demonstrate that elevated levels of norepinephrine directly lead to increased production of IL-17A from T-lymphocytes (Case, Roessner, Tian, & Zimmerman, 2016). Elevations in sympathetic tone could suggest a possible mechanism in these pregnant mothers that may explain the observed increases in IL-17A, but further studies are needed to examine this hypothesis. Herein, we demonstrate that IL-17A showed significant correlations with all stress evaluations as well as with birthweight and race, suggesting a possible new pathogenic role for IL-17A in the etiology of LBW.

Indeed, IL-17A also has been implicated in the etiology of PTB and other pregnancy complications (Barnie, Lin, Liu, Xu, & Su, 2015; Gargano et al., 2008; Ito et al., 2010; Marquardt et al., 2016). In a study authored by Darmochwal-Kolarz (2017), pregnant women with placental insufficiency (n = 34), specifically fetal growth restriction and preeclampsia, had higher levels of circulating IL-17 compared to healthy pregnant women (n = 35). Ito et al. (2010) measured IL-17 in the amniotic fluid of pregnant women undergoing amniocentesis between 22–34 weeks gestation (n = 150). Their results identified increased levels of IL-17 in the amniotic fluid, primarily produced by T-lymphocytes, was associated with PTB. Methods used in these studies have different specimen sources and timing of measurements compared to our study, results warrant further investigation of IL-17A as a specific cytokine of interest in PTB and LBW.

Previous literature on cytokine activity related to PTB and LBW have reported significant relationships between (a) IL-6 and PTB (Lucaroni et al., 2018); and (b) IL-13 and IFNγ with LBW (Pearce et al., 2016). In our study, IL-6 showed a trend for a significant relationship with birthweight (r_s = 0.388, p = 0.062) but did not show any significant correlations with the stress/emotional indicators. In contrast, IL-13 and IFNγ did not show any relationship with birthweight but did show several significant correlations with the stress/emotional indicators. These results highlight the complexity of PTB, LBW, and chronic stress and emphasize the need to measure multiple factors for these perinatal phenomena.

Chronic stress is a difficult concept to operationalize and objectively measure. AL is one theoretical framework to quantify the effects of chronic stress by identifying dysregulation of physiologic mechanisms, including inflammation. Our results warrant further investigation into using an AL index tool composed of multiple cytokines and psychosocial variables, to quantify AL and identify those at risk for a disease or phenomenon such as PTB.

Limitations of the study include small sample size and the accessible population used for the study. Type I error may be affected in this pilot study and future studies will use the false rate discovery method to account for multiple comparisons in a fully powered study. This pilot study included both medically high-risk and low-risk pregnant women which may be a confounding factor when interpreting data. Obtaining blood consistent with the clinical metabolic/glucose challenge measured between 24–28 weeks gestation also may have affected cytokine levels. However, this was a prospective design with stress/emotional indicators obtained on the same day as the blood used to measure inflammation. Previous literature and our current results support further investigation into IL-17A as a potential biomarker to quantify chronic psychosocial stress and identify pregnant women at-risk for poor perinatal outcomes. Measuring several cytokines increases probability of a significant correlation but it is the significant correlations of IL-17A with birthweight and all of the stress/emotional indicators, as well as implications of IL-17A in previous literature, that emphasizes the relevance and importance of IL-17A in future research in this population. This study also demonstrates the necessity of incorporating biological, psychological, and sociodemographic factors together in perinatal research. Future studies should incorporate a biopsychosocial theoretical approach, such as AL.

Conclusion

Increased levels of IL-17A, a known cytokine associated with chronic stress and with poor perinatal outcomes, measured at 24–28 weeks gestation was associated with lower birthweight, non-White race, lower emotional intelligence, lower antenatal maternal attachment, higher prenatal distress, and higher number of life events in pregnant women. This exploratory study supports the biopsychosocial model of AL suggesting individualized responses to stress/emotions may lead to physiologic dysregulation of inflammatory processes associated with poor perinatal outcomes.