Background
For decades, oxytocin has been used for the induction of labor and prevention of postpartum hemorrhage [
1]. In recent years there has been increasing evidence that this nonapeptide, which is synthesized in the hypothalamus, might also be suitable for the treatment of psychiatric disorders such as schizophrenia, autism, major depression, anxiety disorders [
2] and post-traumatic stress disorder (PTSD) [
3]. In response to a variety of stimuli such as suckling, parturition and stress, oxytocin is released from the posterior pituitary into the bloodstream and transported to its effector organs such as the heart, the kidney and the brain [
4]. Oxytocin acts through the oxytocin receptor (OTX) [
4], which is expressed for instance in the amygdala and the anterior cingulate cortex [
5], two brain regions known to be involved in the pathobiology of stress-related psychiatric diseases such as PTSD [
3]. There is an unmet need for the development of drugs specifically tackling the core symptoms of this trauma spectrum disorder [
6], which are hyperarousal, aversive re-experiencing, emotional numbing and avoidance anxiety, because 20–30% of PTSD patients do not respond at all to treatment with the current gold standard of PTSD drug therapy, the serotonin re-uptake inhibitors [
7]. Among other neuropeptides such as neuropeptide S (NPS) [
8] and neuropeptide Y (NPY) [
8,
9], oxytocin has been repeatedly suggested to be effective in PTSD treatment [
3,
9].
In 1993, Pitman and colleagues were the first to test the efficacy of oxytocin versus placebo in PTSD [
10]. They applied either 20 IU oxytocin, vasopressin or placebo to male PTSD combat veterans before their exposure to various tasks (inter alia to combat audiovisual stimuli) and compared the heart rate (HR), electromyography, skin conductance and psychological responses between the three groups [
10]. The only differences they noted were “a significantly higher baseline SC [skin conductance] level and a trend toward a higher baseline EMG [electromyographic]” response in the oxytocin group. Seventeen years later, another study, published hitherto as a poster abstract only, revealed more promising results because the authors reported a significant effect of a single dose of 24 IU oxytocin on non-provoked PTSD symptoms, in particular on the intensity of recurrent thoughts about the traumatic event and on the desire for social interaction [
11,
12]. Accordingly, Olff and colleagues reported that oxytocin treatment decreased amygdalar reactivity toward emotional faces in PTSD patients [
13] and Acheson and colleagues demonstrated that 24 IU of intranasal oxytocin facilitated fear extinction in healthy human subjects [
14]. Currently, the efficacy of oxytocin on secondary prevention of PTSD is evaluated in a double-blind randomized placebo-controlled trial [
15]. Intranasal treatment with 40 IU oxytocin was found to intensify trauma-script-induced re-experiencing symptoms in recently traumatized healthy subjects [
16] but to attenuate stress reactivity, including the cortisol response, in patients with borderline personality disorder [
17], a psychiatric disease known to have a high comorbidity and symptom overlap with PTSD [
18]. In addition, two other studies have demonstrated that intranasal oxytocin reduces stress-induced cortisol levels [
19,
20], thereby emphasizing the interrelationship of the oxytocin system and the hypothalamic-pituitary-adrenal axis, which, in turn, has been repeatedly shown to play a central role in PTSD [
6,
21].
There are, to the best of our knowledge, only two studies thus far that have analyzed the efficacy of oxytocin on PTSD symptom intensity in PTSD patients, among them one symptom provocation study that was performed in a cohort of male veterans [
10]. This motivated us to perform the first study analyzing the efficacy of oxytocin on provoked PTSD symptoms in female PTSD patients. Testing a female PTSD patient cohort is of particular relevance because gender differences in the effects of oxytocin have been repeatedly reported [
22,
23] and because PTSD is more prevalent in women. Besides analyzing the influence of intranasal oxytocin on PTSD symptoms, we put particular emphasis on evaluating its effects on cardiac control, because, apart from oxytocin’s undisputed role in cardiovascular regulation [
24], its effects on HR remain unclear: on the one hand, oxytocin has been reported to reduce HR [
24]; on the other hand, there are numerous publications reporting that oxytocin treatment, especially oxytocin bolus application, can lead to tachycardia [
25,
26]. Accordingly, there are studies showing that oxytocin increases HR in mice [
27,
28], rats [
29] and dogs [
30]. In contrast, Pitman and colleagues found no changes in HR in male PTSD veterans, neither at baseline nor in response to combat audiovisual stimulus [
10]. In consequence of these discrepancies, Novartis reports that “cardiovascular changes including tachycardia and bradycardia can be common” (p.1115) in response to oxytocin treatment [
31]. There is one study that nicely demonstrated the influence of individual psychological factors on oxytocin-mediated cardiac control by showing that higher levels of loneliness were associated with reduced parasympathetic cardiac reactivity to intranasal oxytocin [
32]. In the latter study, oxytocin significantly increased overall autonomic cardiac control because it elevated high frequency heart rate variability (HRV) and decreased the pre-ejection period (PEP), a well-known indicator of sympathetic cardiac control. Because PTSD dramatically increases the risk for cardiovascular mortality [
33], it is of utmost importance to study the influence of potential PTSD drugs on cardiac parameters. In the study presented here, we analyzed, in addition to assessing the psychological response, the respiration rate and several cardiac parameters, namely HR, HRV and PEP, not only at baseline but also after symptom provocation.
Discussion
This is the first study assessing the effects of oxytocin on the intensity of provoked PTSD symptoms in female PTSD patients and, to the best of our knowledge, the second symptom provocation study analyzing the efficacy of oxytocin in PTSD patients ever. Taken together, we show here for the first time that intranasal oxytocin reduces the trauma script-provoked expression of PTSD symptoms, in particular avoidance (Table
2). Furthermore, the oxytocin-mediated increase in HR (Additional file
1: Table S2; placebo baseline: 75.9 bpm (SD 9.6); oxytocin challenge: 77.9 bpm (SD 8.9);
p = 0.020) gets slightly more pronounced in response to stress (Table
2; oxytocin baseline: 82.1 bpm (SD 11.4); oxytocin challenge: 87.2 bpm (SD 15.0);
p = 0.095). The here-observed immediate influence of intranasal oxytocin on psychopathology and hence on brain function is supported by primate studies showing that intranasal oxytocin, such as other intranasally administered anxiolytic neuropeptides like NPS [
45], are able to reach the cerebrospinal fluid [
46,
47] and thus, assumingly, also the brain. In addition, our experiments revealed that oxytocin significantly increased the HR in female PTSD patients both at baseline (Additional file
1: Table S2) and upon trauma-script exposure (Table
2), helping to clarify the hitherto still not fully elucidated influence of oxytocin on HR [
10,
24‐
26,
31,
32]. Our finding of a positive chronotropic effect of oxytocin is supported by the positive correlation of endogenous oxytocin levels and HR both before (Fig.
2c) and after (Fig.
2d) TSST challenge in a cohort of healthy subjects.
In contrast to our findings (Fig.
2a, b), de Jong and colleagues found a significant increase in salivary oxytocin levels of healthy subjects in response to TSST exposure [
48]. However, from the fact that a correlation of salivary and plasma oxytocin levels was absent in another study on healthy individuals [
49], we conclude that our results do not contradict those of de Jong and colleagues. Because there is some controversy on the measurement of oxytocin in the literature, in particular on the determination of oxytocin in unextracted plasma samples [
50,
51], we decided to determine oxytocin levels in serum samples and to use a hitherto uncriticized ELISA kit, although it has not yet been extensively validated. In spite of this, the oxytocin levels we report here are higher than those obtained by others with extracted samples. However, the facts that the levels were still much lower than those documented with the criticized assays and, moreover, showed a positive association with stress (Fig.
2c, d), as expected from the extrapolation of previously published data [
52], strongly suggest that a potential sample-matrix interference, if existent at all, did not corrupt our results.
Because we found intranasal oxytocin to enhance the challenge-mediated decrease in PEP but not oxytocin-mediated modulation of HRV (Table
2), a known measure for parasympathetic activity [
53], we hypothesize that the positive chronotropic effect of oxytocin observed here occurs through an oxytocin-elicited activation of the SNS rather than through an oxytocin-mediated dampening of the PNS. Our results on the influence of oxytocin on HR and PEP in PTSD patients are in full accordance with those of another research group that found intranasal oxytocin to increase the HR and to decrease PEP in healthy individuals [
32]. However, in contrast to the latter publication, we did not detect a significant increase in HRV in our cohort (Table
2). This difference might have been caused by differences in study design, but could also indicate that oxytocin-mediated activation of the PNS might be altered in PTSD patients. This supposition is supported by previous studies showing that, besides the SNS [
54], the PNS also plays a role in PTSD. For example, parasympathetic activity was found to influence basal HR in PTSD patients [
55] and to be lower in Croatian combat veterans [
56]. In contrast, it is well accepted that the SNS is overactive in PTSD [
54]. Of note, only little of the intranasally administered oxytocin reaches the brain. Accordingly, it has been suggested that the central nervous effects of oxytocin might be a consequence of its peripheral actions [
57], such as its influence on cardiac function.
Gender differences in oxytocin effects have been repeatedly described [
22,
23] and might, besides a variety of other factors, contribute to the partial differences between our study and that of Norman and colleagues [
32], as well as of Pitman and colleagues [
42], who, in contrast to us, did not find an effect of intranasal oxytocin on provoked PTSD symptoms in male veterans. In recent years, oxytocin has been repeatedly proposed as an effective novel drug treatment for PTSD [
3,
58], but, to the best of our knowledge, there have only been two prior studies that assessed the effects of oxytocin treatment on the intensity of PTSD symptoms in humans [
10,
12]. Thus, our study contributes significantly to our understanding of the therapeutic potential of oxytocin in PTSD. The findings of our clinical trial support the use of oxytocin in PTSD treatment, in particular for medication-enhanced exposure therapy. Interestingly, separate analysis of the different PTSD symptoms revealed that avoidance was the only PTSD symptom to be reduced, at least with a trend for statistical significance, by oxytocin treatment (Table
2). Similarly, previous studies found that oxytocin attenuated avoidance behavior in rodents [
59,
60]. Because our experiments revealed that oxytocin treatment trend-wise reduced avoidance behavior despite enhancing HR and sympathetic cardiac control (Table
2), we speculate that PTSD-associated avoidance behavior is not causally linked to PTSD-associated SNS overactivity [
54].
The relatively small sample size of both samples and the heterogeneity of the patient sample (e.g., the broad range of age, differences in pharmacotherapy and in trauma types) as well as the facts that we analyzed women only and did not include the ovarian cycle phase as a covariate in the analysis of the patient sample limit the generalizability of our results. In a future study we aim to address the question of whether the therapeutic effect of intranasal oxytocin on provoked PTSD symptoms observed here is gender dependent. Moreover, even though we did not observe any unintended effects of oxytocin treatment in addition to its positive chronotropic effects, there is still the possibility of unknown unintended long-term consequences. Future experiments will have to clarify whether the beneficial effects of oxytocin in PTSD treatment described here (Table
2) outweigh the fact that oxytocin stimulates both HR and sympathetic activity (Additional file
1: Tables S2, S3 and [
32]), which are both known to be already enhanced in PTSD [
54].
Acknowledgements
We thank the Horst Kübler-Foundation for their constant support and Heribert Sattel, PhD, for statistical advice.