Depressive symptoms and serum inflammatory markers

We examined psychosocial factors and serum proinflammatory cytokines among 60 pregnant women who were recruited from the Ohio State University Prenatal Clinic, which serves a diverse and largely disadvantaged population. Most participants were African American (57%), had completed high school or less education (82%), and reported a total annual family income of <$15,000 per year (63%). Women were assessed at one time point, primarily in the late first or early second trimester (15 ± 7.8 weeks’ gestation). Those with greater depressive symptoms, as measured by the Center for Epidemiological Studies Depression scale, had higher levels of circulating interleukin (IL)-6 (β = .23; P = .05) and marginally higher tumor necrosis factor–α (β = .24; P = .06). The magnitude of this effect was similar to that reported in nonpregnant adults. An effect for racial differences in IL-6 approached statistical significance (t(49) = –1.6; P = .12), with African American women exhibiting nonsignificantly higher levels. African American women did not differ significantly from white women in depressive symptoms, education, income, or number of previous pregnancies. These initial findings indicate that, as is well-documented in nonpregnancy, depressive symptoms are associated with elevations in circulating inflammatory markers during pregnancy. The translation of these findings to the prenatal period is notable because pregnancy is a time of significant immune adaptation and occurs in relatively young women.

Depressive symptoms and inflammatory responses to an in vivo biologic challenge

In addition to associations with serum or circulating levels of inflammatory markers, stress can also alter immune responses to biologic challenges. Moreover, because such challenges elicit a response, these models may have more predictive power than descriptive measures of circulating markers because they induce greater variability between subjects.

For clear ethical reasons, human studies of the inflammatory response system in pregnant women to-date have relied almost exclusively on in vitro models. Although highly useful, in vitro techniques involve isolation of specific cells, removal of cells from the complex in vivo environment, and exposure to higher levels of antigen than normally occurs in vivo. By providing insight into immune function in the complex, multifaceted, naturally occurring environment, in vivo models may provide data with clearer clinical relevance.

Vaccines have been used as a model to examine in vivo inflammatory responses in nonpregnant adults. Greater inflammatory responses to vaccines have been reported among older adults with greater depressive symptoms and men with carotid artery disease, which suggests that responses to vaccination differ among those who experience conditions with an inflammatory component. Seasonal influenza virus vaccination provides a novel model for the examination of inflammatory responses to an in vivo immune challenge among pregnant women, because this vaccination currently is recommended by the Centers for Disease Control and American College of Obstetricians and Gynecologists for all women without contraindications who are pregnant or will be pregnant during flu season.

Using flu vaccine as an in vivo challenge model, we have demonstrated that psychosocial factors are associated with differential inflammatory responses in pregnant women. Twenty-two pregnant women were assessed before and approximately 1 week after vaccination. Compared with those in the lowest tertile of Center for Epidemiological Studies Depression scores (n = 8), those in the highest tertile (n = 6) had significantly higher levels of macrophage migration inhibitory factor at 1 week after vaccination. Groups did not differ in demographics (eg, age, body mass index, race, income) or health behaviors (eg, sleep, smoking, regular exercise).

The absence of inflammatory response at 1 week after vaccination among women with lower depressive symptoms is consistent with previous evidence that seasonal influenza virus vaccination generally does not cause an extended inflammatory response. Thus, the extended inflammatory responses that are seen among the more depressed women are indicative of dysregulation of normal inflammatory processes. This study provides evidence that psychological stress predicts sensitization of inflammatory responses to an in vivo immune trigger during human pregnancy. If this represents a stable response tendency, women with this predisposition may show similarly exaggerated responses to everyday immune insults that result in a cumulative exposure to inflammatory mediators that are clinically meaningful with regard to perinatal health outcomes.

Racial differences in inflammatory responses to acute psychological stress

Differential physiological reactivity to acute stress is an important predictor of health outcomes in nonpregnant populations. More than 12 studies have examined cardiovascular and neuroendocrine reactivity to acute stress in pregnancy. Overall these data suggest that stress responses are attenuated during healthy pregnancy. Similar attenuation of responsivity has been reported in animal models. These adaptations may be critical from protecting the mother and fetus from excessive exposure to physiological activation. However, data on inflammatory responses to stress during pregnancy are lacking.

We examined 39 women in the second trimester of pregnancy (19 African American women; 20 white women) and 39 demographically similar nonpregnant women who completed an acute stressor (Trier Social Stress Test). Psychosocial characteristics, health behaviors, and affective responses were assessed. Serum IL-6 was measured with high sensitivity enzyme-linked immunosorbent assay at baseline, 45 minutes, and 120 minutes after the stressor. Our results showed that IL-6 responses at 120 minutes after the stressor were 46% higher in African American women vs white women (95% confidence interval, 8–81%; t(72) = 3.51; P = .001). This effect was present in pregnancy and nonpregnancy. IL-6 responses at 120 minutes after the stressor tended to be lower (15%) in pregnant vs nonpregnant women (95% confidence interval, –5 to 32%; P = .14). Racial differences in inflammatory responses were not accounted for by demographics, psychological characteristics, health behaviors, or differences in salivary cortisol across the study session. Pregnant white women also showed lower negative affective responses than nonpregnant women of either race ( P ≤ .007).

This study provided novel evidence that stress-induced inflammatory responses are more robust among African American women vs white women during pregnancy and nonpregnancy. This could be attributable to chronic stress that is associated with racial minority status. The ultimate impact of stress on health is a function of stressor exposure and physiologic responses. Again, women who experience repeated and extended exposure to high levels of inflammatory mediators in response to psychological stressors may experience cumulatively a physiologic burden that impacts perinatal health. Thus, individual differences in stress-induced inflammatory responses represent a clear target for continued research efforts in racial disparities in health during pregnancy and nonpregnancy.

Depressive symptoms and serum inflammatory markers

We examined psychosocial factors and serum proinflammatory cytokines among 60 pregnant women who were recruited from the Ohio State University Prenatal Clinic, which serves a diverse and largely disadvantaged population. Most participants were African American (57%), had completed high school or less education (82%), and reported a total annual family income of <$15,000 per year (63%). Women were assessed at one time point, primarily in the late first or early second trimester (15 ± 7.8 weeks’ gestation). Those with greater depressive symptoms, as measured by the Center for Epidemiological Studies Depression scale, had higher levels of circulating interleukin (IL)-6 (β = .23; P = .05) and marginally higher tumor necrosis factor–α (β = .24; P = .06). The magnitude of this effect was similar to that reported in nonpregnant adults. An effect for racial differences in IL-6 approached statistical significance (t(49) = –1.6; P = .12), with African American women exhibiting nonsignificantly higher levels. African American women did not differ significantly from white women in depressive symptoms, education, income, or number of previous pregnancies. These initial findings indicate that, as is well-documented in nonpregnancy, depressive symptoms are associated with elevations in circulating inflammatory markers during pregnancy. The translation of these findings to the prenatal period is notable because pregnancy is a time of significant immune adaptation and occurs in relatively young women.

Depressive symptoms and inflammatory responses to an in vivo biologic challenge

In addition to associations with serum or circulating levels of inflammatory markers, stress can also alter immune responses to biologic challenges. Moreover, because such challenges elicit a response, these models may have more predictive power than descriptive measures of circulating markers because they induce greater variability between subjects.

For clear ethical reasons, human studies of the inflammatory response system in pregnant women to-date have relied almost exclusively on in vitro models. Although highly useful, in vitro techniques involve isolation of specific cells, removal of cells from the complex in vivo environment, and exposure to higher levels of antigen than normally occurs in vivo. By providing insight into immune function in the complex, multifaceted, naturally occurring environment, in vivo models may provide data with clearer clinical relevance.

Vaccines have been used as a model to examine in vivo inflammatory responses in nonpregnant adults. Greater inflammatory responses to vaccines have been reported among older adults with greater depressive symptoms and men with carotid artery disease, which suggests that responses to vaccination differ among those who experience conditions with an inflammatory component. Seasonal influenza virus vaccination provides a novel model for the examination of inflammatory responses to an in vivo immune challenge among pregnant women, because this vaccination currently is recommended by the Centers for Disease Control and American College of Obstetricians and Gynecologists for all women without contraindications who are pregnant or will be pregnant during flu season.

Using flu vaccine as an in vivo challenge model, we have demonstrated that psychosocial factors are associated with differential inflammatory responses in pregnant women. Twenty-two pregnant women were assessed before and approximately 1 week after vaccination. Compared with those in the lowest tertile of Center for Epidemiological Studies Depression scores (n = 8), those in the highest tertile (n = 6) had significantly higher levels of macrophage migration inhibitory factor at 1 week after vaccination. Groups did not differ in demographics (eg, age, body mass index, race, income) or health behaviors (eg, sleep, smoking, regular exercise).

The absence of inflammatory response at 1 week after vaccination among women with lower depressive symptoms is consistent with previous evidence that seasonal influenza virus vaccination generally does not cause an extended inflammatory response. Thus, the extended inflammatory responses that are seen among the more depressed women are indicative of dysregulation of normal inflammatory processes. This study provides evidence that psychological stress predicts sensitization of inflammatory responses to an in vivo immune trigger during human pregnancy. If this represents a stable response tendency, women with this predisposition may show similarly exaggerated responses to everyday immune insults that result in a cumulative exposure to inflammatory mediators that are clinically meaningful with regard to perinatal health outcomes.

Racial differences in inflammatory responses to acute psychological stress

Differential physiological reactivity to acute stress is an important predictor of health outcomes in nonpregnant populations. More than 12 studies have examined cardiovascular and neuroendocrine reactivity to acute stress in pregnancy. Overall these data suggest that stress responses are attenuated during healthy pregnancy. Similar attenuation of responsivity has been reported in animal models. These adaptations may be critical from protecting the mother and fetus from excessive exposure to physiological activation. However, data on inflammatory responses to stress during pregnancy are lacking.

We examined 39 women in the second trimester of pregnancy (19 African American women; 20 white women) and 39 demographically similar nonpregnant women who completed an acute stressor (Trier Social Stress Test). Psychosocial characteristics, health behaviors, and affective responses were assessed. Serum IL-6 was measured with high sensitivity enzyme-linked immunosorbent assay at baseline, 45 minutes, and 120 minutes after the stressor. Our results showed that IL-6 responses at 120 minutes after the stressor were 46% higher in African American women vs white women (95% confidence interval, 8–81%; t(72) = 3.51; P = .001). This effect was present in pregnancy and nonpregnancy. IL-6 responses at 120 minutes after the stressor tended to be lower (15%) in pregnant vs nonpregnant women (95% confidence interval, –5 to 32%; P = .14). Racial differences in inflammatory responses were not accounted for by demographics, psychological characteristics, health behaviors, or differences in salivary cortisol across the study session. Pregnant white women also showed lower negative affective responses than nonpregnant women of either race ( P ≤ .007).

This study provided novel evidence that stress-induced inflammatory responses are more robust among African American women vs white women during pregnancy and nonpregnancy. This could be attributable to chronic stress that is associated with racial minority status. The ultimate impact of stress on health is a function of stressor exposure and physiologic responses. Again, women who experience repeated and extended exposure to high levels of inflammatory mediators in response to psychological stressors may experience cumulatively a physiologic burden that impacts perinatal health. Thus, individual differences in stress-induced inflammatory responses represent a clear target for continued research efforts in racial disparities in health during pregnancy and nonpregnancy.