Repeated exposure to traumatic events such as rape, combat or natural disasters has dramatic effects on mental health [1–3]. Frequent symptoms in the aftermath of such events include repetitive intrusive recollections of the trauma, persistent avoidance of trauma reminders, emotional numbing and hyperarousal – the core symptoms of posttraumatic stress disorder PTSD .
In addition to psychiatric morbidity, numerous studies have shown that traumatic stress and especially PTSD are associated with poor self-reported physical health, increased utilization of medical services, and an elevated risk for multiple comorbid medical disorders such as respiratory, gastrointestinal, musculoskeletal, inflammatory and autoimmune diseases [5–7]. In particular, it has been found that cardiovascular disease and its risk factors are more prevalent among individuals with PTSD [5, 6, 8], with two recent studies even demonstrating a prospective relationship between PTSD symptoms and coronary heart disease  or PTSD and cardiovascular mortality, respectively .
Besides traditional behavioral cardiovascular risk factors such as smoking, alcohol abuse or low physical activity , chronic low-level inflammation has been discussed as a potential mechanism linking cardiovascular disease to PTSD . This argumentation is supported by epidemiological and clinical studies demonstrating a strong and consistent relationship between markers of inflammation and risk of future cardiovascular events [13, 14].
However, studies investigating pro-inflammatory markers in PTSD patients have yielded mixed results. Some studies reported higher levels of IL-1β [15, 16], IL-6 , and TNF-α  in the plasma or elevated levels of IL-6 in the cerebrospinal fluid  of PTSD patients compared to control subjects. Furthermore, Spitzer et al.  found that, in a sample of 3049 adults, PTSD positive participants had significantly higher odds for elevated levels of C-reactive protein (CRP) than those without PTSD. In contrast, other studies did not find significant group differences with respect to circulating levels of IL-1β , IL-6 [12, 20], and CRP  or even reported lower levels of CRP  and IL-8  in PTSD patients.
Likewise, ambiguous results have been obtained by studies investigating the production of pro-inflammatory cytokines in whole blood following ex vivo stimulation with lipopolysaccharide (LPS) and/or phytohemagglutinin (PHA). Whereas some studies reported higher LPS-stimulated IL-6 production  or increased LPS/PHA-induced production of IL-6 and TNF-α  in individuals with PTSD, other studies showed no group differences with respect to PHA-induced interferon (IFN)-γ  or LPS-induced TNF-α production . One study by de Kloet et al.  even reported decreased TNF-α production in LPS-stimulated whole blood of PTSD patients compared to controls.
These inconsistent findings might partly be explained by variations in sample characteristics, such as differences with respect to the type of traumata experienced (e.g. childhood vs. adulthood trauma), time elapsed since the traumata (e.g. populations studied shortly after traumatic experiences or during ongoing threat vs. patients with chronic symptoms), comorbidities, and differences with respect to PTSD symptom severity (mild symptoms after a single trauma in comparison to severe symptoms after experiencing multiple traumata), as well as varying sample sizes. In particular studies investigating plasma cytokine levels consisted mostly of small patient groups, making the results error-prone.
Here we investigated plasma levels of pro- and anti-inflammatory cytokines in a well-defined group of 35 severely affected PTSD patients with war and torture experiences mainly experienced during late adolescence and adulthood and a chronic disease pattern, compared to 25 healthy ethnically matched control subjects. For a subsample of 16 PTSD patients and 18 control subjects, we additionally analyzed 1) LPS-stimulated production of IL-1β, IL-6 and TNF-α by peripheral blood mononuclear cells (PBMCs), as well as 2.) spontaneous cytokine production by these cells, as this has not been reported previously.