Individuals with post-traumatic stress disorder (PTSD) are at significantly higher risk of developing cardiovascular disease. While there is accumulating evidence that insufficient sleep independently contributes to adverse cardiovascular conditions, the specific associations between PTSD symptoms, cardiovascular variables and sleep disturbances (i.e. disruptive nighttime behaviors) are still unclear. In a group of trauma-exposed civilian women, we assessed the predictive value of the PTSD diagnostic clusters of re-experiencing, arousal and avoidance/numbing on blood pressure (BP), heart rate (HR) and sleep disturbances. We hypothesized that BP, HR and sleep will be differentially linked to specific diagnostic clusters. PTSD status was assessed using the gold-standard clinician administered PTSD scale (CAPS) and PTSD symptom severity via the PTSD Symptom Scale (PSS). We measured baseline BP, HR and quantified sleep disturbances via the Pittsburgh Sleep Quality Index Addendum for PTSD (PSQI-A). Of the 49 women included in the analysis, a subset of 43 (age, 49±12 years old; body mass index (BMI), 37±7 kg/m2) also presented with comorbid depression as per Beck Depression Inventory (BDI). Our analysis yielded no significant correlations between resting BP and CAPS or PSS scores. However, we found a modest correlation between HR and CAPS score (r=.210, p=.074), solely driven by avoidance symptoms (r=.344, p=.009). PSQI-A was correlated with CAPS (r=.536, P<.001) and PSS total scores (r=.685, P<.001). Interestingly, the correlation with PSS appeared to be mainly driven by hyperarousal (r=.621, p<.001) and re-experiencing (r=.704, p<.001) symptoms than avoidance/numbing symptoms (r=.442, p=.001). PSQI-A was also correlated with depression (r=.589, P<.001). To further determine the predictive value of the PTSD diagnostic clusters, we conducted two hierarchical linear regression models predicting HR and PSQI-A by relevant covariates of age, BMI, and BDI total score in step 1, and re-experiencing, hyperarousal, and avoidance symptoms scores in step 2. The final model with age, BMI, depression, and all three PTSD symptom clusters was significant (R2=.607, p=.016), and those clusters explained 21.3% of the variance beyond other predictors in PSQI-A score; the re-experiencing cluster was the strongest predictor (b =.535, p<.001). On the other hand, the final model predicting HR (R2=.205, p=.035) included age, BMI, and avoidance symptoms as significant predictors (p<.05). In summary, the current results show that while avoidance/numbing symptoms might partly contribute to resting HR in PTSD, the disruptive nighttime behaviors described in PTSD are strongly predicted by symptoms of re-experiencing (i.e. intrusive thoughts, flashbacks), even when controlling for the common comorbid depression. This suggests that the level of sleep dysfunction reported in PTSD could depend on the presence of specific symptoms which can serve to identify at-risk individuals who may benefit from targeted sleep interventions.
Read full abstract