اکسی توسین پلاسما مربوط به کاهش واکنش پذیری قلبی و عروقی و دلسوز به استرس

Abstract In addition to known reproductive and social affiliation functions, oxytocin (OT) has been identified as a cardiovascular hormone. OT synthesis and receptors are found in cardiac and vascular tissue. Animal studies suggest that OT activates an ‘anti-stress’ response that reduces cardiovascular and neuroendocrine stress reactivity. We tested 28 early postpartum mothers, obtaining multiple blood samples for OT, the sympathetic marker, norepinephrine (NE), and the lactation hormone, prolactin, while monitoring their cardiovascular responses to two stressors: public speaking and forehead cold pressor. Although plasma OT did not increase reliably from pre-stress levels during stressors, greater overall OT level was related to greater vasodilation and cardiac stroke volume responses to both tasks, to reduction in heart rate to the cold pressor, as well as to lower plasma NE and higher prolactin levels. In contrast, higher NE was linked to increases in heart rate and decreases in stroke volume. These data support a cardioprotective role for OT, which may influence the magnitude and hemodynamic determinants of cardiovascular stress responses.

. Introduction 1.1. Oxytocin is a cardiovascular hormone In the past decade substantial attention has been focused on the role of the neuropeptide, oxytocin (OT), in social behaviors. However, OT has also been identified as a cardiovascular hormone (Petersson, 2002). OT synthesis and receptors are reported in both cardiac and vascular tissue in non-human mammalian species (Gutkowska et al., 2000, Jankowski et al., 1998 and Jankowski et al., 2000), and OT acts both centrally and peripherally at multiple sites including brainstem, heart, and vessels to exert acute and long-term inhibitory effects on cardiovascular activity (Meisenbert and Simmons, 1983, Nakamura et al., 2000, Petersson et al., 1996, Petersson et al., 1997, Petersson et al., 1999a, Petersson et al., 1999b, Sofroniew, 1983, Stock and Uvnas-Moberg, 1988, Thibonnier et al., 1999a and Thibonnier et al., 1999b). In animal models, daily peripheral administration of exogenous OT for 5 days leads to blood pressure (BP) decreases lasting 2 months or more (Holst et al., 2002 and Petersson and Uvnas-Moberg, 2008). Similarly, OT administered in vitro reduces the rate and force of cardiac cells’ intrinsic contractions causing them to ‘beat’ more slowly and contract less forcefully ( Mukaddam-Daher et al., 2001). In humans, studies reporting on BP effects of chronic OT administration are lacking. However short term intravenous (IV) administration of OT to women to enhance uterine contractions or decrease blood loss during labor or caesarean delivery confirm its effect in decreasing blood pressure ( Sartain et al., 2008, Simpson and Knox, 2009 and Thomas et al., 2007). This hypotensive response to OT is due to decreases in total vascular resistance despite compensatory increases in heart rate, stroke volume and cardiac output. In fact, even though specific organs and tissues may show local vasoconstriction with OT administration, decreases in total vascular resistance and BP that are potentially life threatening can occur ( Archer et al., 2008). 1.2. Oxytocin and social interactions both influence cardiovascular function Social behaviors are related to both OT and cardiovascular activity, and OT may serve as an important physiological mediator of the cardioprotective benefit of social bonding (Knox and Uvnas-Moberg, 1998). Animal studies reveal that affiliative social interactions elicit increases in OT activity, which then activate and integrate an ‘anti-stress’ response that promotes bonding, relaxation and growth, while reducing cardiovascular and neuroendocrine stress responsivity (Callahan et al., 1989, Petersson and Uvnas-Moberg, 2007, Petersson and Uvnas-Moberg, 2008, Uvnas-Moberg, 1998, Uvnas-Moberg et al., 2001, Uvnas-Moberg and Petersson, 2004 and Wsol et al., 2008). In rats, daily ventral stroking for 5 days is linked to long-lasting BP decreases very much like those induced by exogenous OT administration, and it is presumed that this effect is due to increases in endogenous OT activity (Holst et al., 2002). In female prairie voles, social isolation (considered a stressor), results in higher basal heart rate (HR) levels and decreased HR variability, and these effects are reversed with subcutaneous OT administration (Grippo et al., 2009). In humans, a number of studies have linked higher plasma OT to lower BP. We observed this in early postpartum mothers both on days when they had recent infant contact and when they did not, and in married women studied after 10 min of structured warm contact with their husbands (Grewen et al., 2005, Light et al., 2000, Light et al., 2005a and Light et al., 2005b). In addition to lower BP in laboratory studies, higher OT responses to structured affectionate interaction with infants were associated with lower maternal 24-h ambulatory BP levels at home (Light et al., 2000 and Light et al., 2004). We also reported consistent inverse associations between plasma OT and the sympathetic biomarker, norepinephrine (NE), in women during rest and in response to pleasant partner contact (Grewen et al., 2005), but we have not measured these hormones simultaneously during stress. Similarly, a month-long intervention in couples, involving caring support through Rosen listening touch, a therapeutic method using light touch to communicate acceptance, empathy and caring and to sense subtle physical and emotional responses felt by another (Rosen, 2003), resulted in increases in salivary OT and decreases in the indirect sympathetic marker, salivary alpha-amylase, in both husbands and wives. In addition, in husbands but not wives, 24-h BP was also reduced. However, since the wives had lower BP prior to the intervention, this may have been due to a ‘floor effect’ (Holt-Lunstad et al., 2008). 1.3. Oxytocin may attenuate physiological response to stress Although less studied than the OT-social affiliation links, findings from both animal and human studies suggest that plasma OT may reduce physiologic stress responses. OT knockout mice exhibit greater BP and corticosterone responses to acute and chronic stress (Bernatova et al., 2004 and Michelini et al., 2003). In rats, blocking central OT with intracerebral infusion of OT antagonist enhances HR and BP responses to acute stress but does not effect resting levels (Wsol et al., 2008). Findings from human studies are mixed, however. This may be due to differences in the kinds of stressors used and/or in the subject populations being studied. Light et al., 2005a and Light et al., 2005b reported that in postmenopausal women participating in a hormone replacement therapy (HRT) trial, greater treatment-induced increases in oxytocinergic activity (indexed by plasma levels of an OT precursor) were associated with greater post-treatment reductions in BP and vascular resistance reactivity to a battery of experimental stressors including speech, stroop and cold pressor tasks. More recently, Taylor et al. (2006) reported that although postmenopausal women self-selected for HRT had higher plasma OT compared with non-treated women, higher OT was not related to blood pressure or heart rate reactivity to the Trier Social Stress Test (TSST). Similarly, Altemus et al. (2001) reported that in postpartum lactating and non-lactating women and reproductive-aged control women, OT levels were not associated with blood pressure or heart rate reactivity to the TSST, but lactation was related to greater parasympathetic cardiac control.

Results 4.1. Subject characteristics Subject characteristics are provided in Table 1. The majority of mothers were white (4 African-American, 22 Caucasian, 2 Hispanic), with at least a 4-year college degree (8 = less than 4-year degree, 16 = 4-year degree, 8 = post-graduate education). Infant postnatal age ranged from 2.3 to 6.3 months at time of testing, while maternal age ranged from 24 to 41 years. All subjects were normotensive. The sample included 18 mothers who exclusively breast-fed and 10 mothers who used formula feeding in whole (n = 5) or in combination with breast feeding (n = 5). In addition to mean totals across the 28 subjects, Table 1 also shows mean values separately for exclusive breasting feeding and formula-feeding (exclusive and partial combined) subjects. Exclusive breast-feeders had lower BMI, greater stress ratings, and marginally lower plasma norepinephrine calculated as area under the curve across the session. When Exclusive Formula-feeding mothers (n = 5) were examined separately, they had significantly higher plasma norepinephrine than exclusive breast-feeders (NE AUCG means: 966.5 vs. 634.7 ng/mL; F = 6.01, p = 0.042) but did not differ on oxytocin or any other variable listed in Table 1. Prolactin data was available for only one exclusive formula feeding mother therefore comparisons could not be made between exclusive breast vs. exclusive formula for prolactin levels. Table 1. Subject characteristics for 18 exclusive breast-feeding, and 10 formula-feeding (5 exclusive and 5 formula/breast combination) mothers. All subjects n = 28 Breast feeding n = 18 Formula feeding n = 10 Mean Std. err. Mean Std. err. Mean Std. err. Maternal age (years) 29.96 0.78 30.39 0.98 29.20 1.31 Body mass index (kg/m2) 27.57 1.53 25.26 1.79 31.72a* 2.40 Infant age (months) 4.28 0.21 4.03 0.27 4.79 .36 Resting SBP (mm Hg) 107.75 1.85 106.93 2.33 109.23 3.13 Resting DBP (mm Hg) 67.46 1.48 67.56 1.90 67.30 2.48 Resting HR (beats/min) 74.33 2.16 74.08 2.74 73.23 3.67 Resting cardiac index (L/min) 4.28 1.12 4.34 0.28 4.20 0.36 Resting VRI (dyn-s-cm−5) 1604.71 465.59 1556.94 110.73 1690.71 148.55 Oxytocin AUCG (pg/mL) 8.83 0.64 8.58 0.81 9.29 1.09 Norepinephrine AUCG (ng/mL) 706.54 53.50 634.70 65.48 821.50a# 82.82 Prolactin AUCG (ng/mL)e 126.93 23.28 196.26 36.51 111.57 55.76 Baseline stressc 2.32 0.38 3.00b* 0.43 1.10 0.57 Speech stressc 4.25d*** 0.46 5.22b* 0.45 2.50 0.73 Cold pressor Stressc 3.82d* 0.42 4.78b*** 0.12 1.89 0.54 Baseline forehead painc 0.25 0.13 0.28 0.17 0.20 0.23 Cold pressor painc 6.29d*** 0.37 6.72 0.33 5.5 0.81 Postnatal depression 4.50 4.05 4.67 0.97 4.20 1.03 Spielberger Trait Anxiety 37.37 8.59 38.12 2.11 36.10 2.75 Bolded text indicates significant differences. VRI: vascular resistance index; std. err.: standard error of the mean; *** p < 0.0001; ** p < 0.01; * p < 0.05; #p < 0.09; Postnatal Depression: Edinburgh Postnatal Depression Scale Score. a Formula feeding > breast feeding. b Breast feeding > formula feeding. c Visual Analog Scales (0–10). d Post-task > baseline value for total sample. e Prolactin data available for 1 exclusive formula, 5 formula/breast combination feeders, 14 exclusive breast feeders. Table options 4.2. Cardiovascular responses to stressors Fig. 1 depicts subjects’ cardiovascular values measured at each of the 16 sampling times during the protocol. Speech: The speech task induced significant increases in SBP, DBP and HR compared with mean baseline levels (paired t-test p values < 0.05). Cold pressor induced significant increases in SBP, DBP, and VRI, and significant decreases in HR. Cardiac output (CI) and contractility, quantified as heather index, were immediately but transiently reduced below baseline levels at Minute 0 (p < 0.01). Significant increases in both SBP and DBP were seen at Minute 1 during each of the 2 stressors while peak vascular resistance levels occurred only during the forehead cold pressor task. Stroke volume and pre-ejection period were not significantly different during stress compared with baseline. Reactivity of stroke volume, pre-ejection period and vascular resistance was inconsistent across subjects, as reflected in higher variability in these measures compared with lower variation in blood pressure and heart rate across subjects at each time point. Table 2 lists delta scores (change from baseline to Minute 1) for cardiovascular values (mean ± standard error) during each of the two stressors, speech and cold pressor. Table 2. Pearson correlations (r) between log-transformed oxytocin and norepinephrine Area Under the Curve with cardiovascular reactivity calculated as delta scores (mean task − mean baseline levels). Delta Scores Mean (Std. Err.) OT AUCG OT AUCI NE AUCG NE AUCI Speech VRI (dyn-s-cm−5) +83.94 (45.65) −0.51** −0.46* +0.22 +0.21 HR (bpm) +2.86 (1.10) −0.25 −0.24 +0.44* +0.42* SVI (mL) +0.04 (6.48) +0.43* +0.38# −0.07 −0.05 CI (L/min) +0.15 (0.10) +0.34 +0.31 +0.17 +0.20 HI (Ω/s2) −0.42 (0.33) −0.04 +0.07 +0.01 +0.02 SBP (mm Hg) +9.46 (2.53) −0.25 −0.22 +0.24 +0.23 DBP (mm Hg) +5.96 (1.60) −0.27 −0.18 +0.25 +0.26 PEP (ms) −0.96 (1.30) +0.25 +0.24 −0.04 −0.03 Cold pressor VRI (dyn-s-cm−5) +355.85 (79.90) −0.38* −0.39* +0.28 +0.29 HR (bpm) −3.82 (1.02) −0.54** −0.53** +0.51** +0.48** SVI (mL) +1.69 (8.02) +0.54** +0.54** −0.45* −0.41# CI (L/min) −0.16 (0.07) +0.34 +0.35 −0.12 −0.09 HI (Ω/s2) −1.00 (0.44) +0.30 +0.37# −0.52* −0.51* SBP (mm Hg) +10.75 (2.42) −0.12 −0.17 −0.04 −0.02 DBP (mm Hg) +7.23 (1.82) +0.08 +0.02 −0.06 +0.01 PEP (ms) −0.45 (1.32) −0.13 −0.10 +0.21 +0.17 #p < 0.10, *p < 0.05, **p < 0.01; OT: oxytocin; NE: norepinephrine; AUCG: area under the curve with respect to ground; AUCI: area under the curve with respect to increase (a reactivity measure); VRI: vascular resistance index; HR: heart rate; SV: stroke volume; CI: cardiac index; HI: heather index; SBP: systolic blood pressure; DBP: diastolic blood pressure; bpm: beats per minute; mm Hg: millimeters of mercury; PEP = pre-ejection period; ms = milliseconds. Table options 4.3. Cardiovascular response to infant interaction Movement-related artifacts or disconnection of impedance sensors during mother–infant interaction resulted in lost data for impedance-derived values in 11 subjects (PEP, SVI, CI, VRI, HI) at the single reading following this period. However SBP, DBP and HR were available for all 28 subjects, and change from baseline levels in SBP, DBP and HR did not differ in those with (n = 17) and without (n = 11) available impedance data. Compared with baseline, SBP (t = 3.95, p = 0.0005), HR (t = 2.07, p = 0.045), stroke volume (SVI: t = 2.60, p = 0.018) and cardiac output (CI: t = 3.09, p = 0.0071) were significantly higher following infant interaction. Pre-ejection period (PEP: t = −3.55, p = 0.0023) was significantly reduced from baseline levels, consistent with increased HR. Vascular resistance was marginally decreased (VRI: t = −2.01, p = 0.06). This pattern of increased blood pressure, heart rate, cardiac output and mild vasodilation may have been due to the recent physical exertion of holding, talking to and playing with the infant. 4.4. Oxytocin, norepinephrine, and prolactin responses and associations Mean plasma OT levels at all 9 sampling time points, and mean plasma norepinephrine (NE) and serum prolactin (PRL) levels, sampled 5 times during the protocol, are depicted in Fig. 2A–C respectively. OT data were available for all 28 subjects. NE data for 26, and PRL data for 20 subjects were also available. Plasma NE increased significantly over baseline levels during the cold pressor task (paired t-test t = 2.81, p = 0.0096). There was no significant increase in OT or PRL in response to stress, and substantial variability in OT and PRL at each time point. Total plasma OT was inversely correlated with norepinephrine when each was calculated as AUCG (r = −0.40 p = 0.05, depicted in Fig. 2D), or when each was calculated as AUCI (r = −0.41, p = 0.04). Total serum prolactin AUCG, sampled at the same five time points as plasma NE, was also correlated with OT AUCG (r = +0.50, p = 0.02) and inversely correlated with NE AUCG (r = −0.49, p = 0.03). In response to speech stressor, 11 women demonstrated oxytocin increases above baseline, while 9 showed no change and 8 had decreased levels. There were 6 exclusive breast feeders in each of these groups. Similar differences in OT response to cold pressor were observed (12 increased, 6 no change, 10 decreased). Values for plasma oxytocin (A), norepinephrine (B), and serum prolactin (C), ... Fig. 2. Values for plasma oxytocin (A), norepinephrine (B), and serum prolactin (C), displayed as mean ± standard error of the mean at each measurement. (D) Scatterplot and regression line reflecting Pearson correlation (r) of log Norepinephrine AUGG and log Oxytocin AUCG. Values in plasma based on infant feeding method (18 breast, 10 formula in whole or in combination with breast) for: (E) oxytocin, (F) norepinephrine, and (G) serum prolactin. Figure options 4.4.1. Oxytocin, Norepinephrine and Prolactin based on infant feeding status When OT was examined separately for breast and formula feeding groups, there was no significant difference in OT level or change in OT at any time point (Fig. 2E). Fig. 2F shows that plasma NE was marginally greater in formula feeders compared with exclusive breast feeders during speech task (232.1 ± 23.29 vs. 177.9 ± 16.5 pg/mL; t = 1.93, p < 0.07). Serum prolactin levels, shown in Fig. 2G, did not change significantly across the protocol in breast or formula feeders. 4.5. Behavioral responses Self-reported stress and pain, rated on visual analog scales, are listed in Table 1. Stress ratings increased above baseline when rated immediately following speech (t = 4.84, p < 0.0001) and cold pressor (t = 2.67, p = 0.013), indicating that these tasks were experienced as stressors. Breast feeding women reported greater stress at all 3 time points compared with formula feeding women, and had greater increases in stress ratings when regression models were adjusted for baseline differences (not shown). Across the full sample, greater stress during the speech was accompanied by greater vasodilation and increases in cardiac output, consistent with a ‘fight-or-flight’ beta-adrenergically mediated stress response. Greater stress ratings of speech task correlated with reductions in vascular resistance (r = −0.58, p < .01), and increases in stroke volume (r = +0.53, p < .01) and cardiac output (r = +0.69, p < .0005) during the task. Self-reported forehead pain ratings increased after cold pressor (t = 15.87, p < 0.0001), and were related to greater stress reported at the same time (Spearman r = +0.64, p = 0.0004). Neither stress nor pain ratings were correlated with any individual or composite AUC hormone values. 4.6. OT, NE and PRL are associated with cardiovascular responses to stress Plasma oxytocin and norepinephrine were each related to stress-induced changes in some, but not all, cardiovascular variables, and had opposing effects on cardiac responses. Table 2 summarizes Pearson correlations found between cardiovascular stress reactivity values (delta scores) and OT and NE calculated as both total hormone (AUCG) and reactivity (AUCI) values. 4.6.1. OT and vascular resistance Greater oxytocin levels were correlated with smaller increases (or greater decreases) in vascular resistance (delta VRI) compared to baseline in response to both the speech (r = −0.51, p = 0.007) and the cold pressor tasks (r = −0.38, p = 0.05). Fig. 3 depicts these negative correlations of OT AUCG with changes in VRI during Minute 1 of the speech ( Fig. 3A) and Minute 1 of cold pressor ( Fig. 3B). Because greater OT and stress rating were both related to lower VRI responses to speech, we regressed VRI response on both post-speech stress rating and OT AUCG. Both OT and stress rating remained independent predictors of decreases in vascular resistance (overall F2,26 = 10.18, p = 0.0022, R-squared = 0.44; OT AUCG: beta: −0.47, p < 0.004, stress: beta: −0.46, p = 0.005). When additional covariates were added (feeding status, baseline vascular resistance, smoking status) oxytocin remained an independent predictor of vascular resistance reactivity to speech (OT AUCG beta = −0.34, t = −2.09, p = .048) and cold pressor (OT AUCG beta = −0.41, t = −2.39, p = 0.026). (A) Scatterplot and regression lines reflecting Pearson correlations (r) between ... Fig. 3. (A) Scatterplot and regression lines reflecting Pearson correlations (r) between delta VRI (dyn-s-cm5) during speech (speech Minute 1 − mean baseline) with log Oxytocin AUCG. (B) Scatterplot and regression lines reflecting Pearson correlation between delta VRI during cold pressor (ICE) with log oxytocin AUCG. Figure options 4.6.2. OT, NE and cardiac responses Table 2 shows that OT was not significantly correlated with stress-induced changes in blood pressure (BP) or cardiac output (CI). OT was, however, related to the determinants of cardiac output: heart rate and stroke volume. Greater OT correlated with greater increases in stroke volume (SVI) during both tasks, and to greater decreases in heart rate (HR) response to cold pressor. Models were then adjusted for relevant covariates. Stroke volume reactivity to speech: When stress rating of the speech task was added to the model predicting SVI response, both OT and stress remained independent predictors of greater SVI response (overall F2,26 = 7.098, p = 0.0038, R-squared = 0.37; OT AUCG: beta: +0.46, p < 0.01, stress: beta: +0.38, p < 0.05). OT remained marginally predictive (beta = + 30, p = .087) when baseline SVI, feeding and smoking status were added to the model. In contrast to OT, greater NE was related to greater increases in HR during both stressors, and to decreases in stroke volume (SVI) and contractility (HI) during the cold pressor. HR reactivity to cold pressor: OT and NE independently accounted for significant amounts of variance in HR reactivity to cold pressor when both were entered into the model simultaneously (overall F2,26 = 8.12, p = 0.0022, R-squared = 0.41; OT AUCG: beta = −0.42, p = 0.016; NE AUCG: beta = +0.53, p = 0.003) and OT remained a significant predictor in the full model (OT beta = −0.47, p = .014). Similarly, OT independently accounted for significant amounts of variance in SVI reactivity to cold pressor when both OT and NE were entered into linear regression, however NE was no longer a significant predictor of SVI change when OT was present in the model (F2,26 = 5.64, p = 0.01, R-squared = 0.36; OT beta = +0.43, p = 0.037; NE beta = −0.29, p = ns). OT remained a significant predictor in the full model (OT beta = +0.47, p = 0.037) in which baseline SVI, feeding and smoking status were added. 4.6.3. Prolactin (PRL) Although prolactin data was available for fewer subjects (n = 20), correlations with cardiovascular reactivity values were similar to those seen with oxytocin. Total prolactin (PRL AUCG) was significantly correlated with reductions in vascular resistance and increases in stroke volume during the speech task (delta VRI: r = −0.59; delta SVI: r = +0.43; p < 0.05). Similarly, PRL AUCG was correlated with reductions in vascular resistance and heart rate during the cold pressor (delta VRI: r = −0.51; delta HR: r = −0.52; p < 0.05). In this smaller subset of women, oxytocin was correlated with cardiovascular reactivity values in the same or greater magnitudes seen in the complete sample of 28 women (speech: delta VRI r = −0.52, delta SVI r = +0.046, cold pressor: delta VRI r = −0.45, p < 0.05; cold pressor delta HR: r = −0.68, delta SVI: r = −0.65, p < 0.001). PRL was also related to reduced DBP at the end of infant interaction compared with resting baseline (delta DBP: r = −0.54; p < 0.05).