Impact of Post-Stroke Post-Traumatic Stress Disorder

Abstract

Introduction

Post-Traumatic Stress Disorder (PTSD) is associated with exposure to traumatic events, especially in the military setting. However, patients who experience stroke may develop anxiety about their stroke event and may re-experience transient neurological symptoms as a result. A significant portion develop the persistent and disabling symptoms of PTSD.

Methods

At the University of South Florida, we conducted a single-center, IRB-approved, observational pilot study of 20 adult patients who were diagnosed with stroke or transient ischemic attack (TIA) in the previous 31 days to 1 year. Patients completed the post-traumatic stress disorder checklist-5 (PCL-5), Patient Health Questionnaire-9 (PHQ-9), Stroke specific Quality of Life Scale (SS-QOL-12), Modified Rankin Scale of disability (mRS), and National Institutes of Health Stroke Scale (NIHSS) and provided blood and saliva samples.

Results

All 20 subjects completed the PCL-5 and 19 subjects completed the follow up scales. Seven patients (35%) were found to have Post-Stroke Post-Traumatic Stress Disorder (PS-PTSD). Higher PCL-5 scores were significantly correlated with lower SS-QOL scores indicating worse quality of life (r = −0.709, P = .001) and higher PHQ-9 scores representing symptoms of depression (r = 0.727, P < 0.001).

Conclusion

Post-Stroke Post-Traumatic Stress Disorder (PS-PTSD) is prevalent after stroke and TIA with patients experiencing concurrent depressive symptoms, correlating with a worsened quality of life.

Introduction

Post-traumatic stress disorder (PTSD) is an under-recognized consequence of stroke. ¹ Recognition of post-stroke anxiety and depression has increased in recent years. ² Although often overlooked, a significant portion of stroke patients may develop Post-Stroke Post-Traumatic Stress Disorder (PS-PTSD) characterized by re-experiencing, avoidance, hyperarousal, and cognitive symptoms.

PTSD has been associated with worsened quality of life, decreased medication adherence and poor health care outcomes.^3,4 This translates into higher financial burden as PTSD is associated with increased health care costs; women bear a disproportionately higher burden of the accompanying health expenditures. ⁵ Given the impacts of PTSD on emotional and physical wellbeing, it is valuable to conduct research into better understanding, recognizing, and treating PS-PTSD.

Twelve-month prevalence of all PTSD in the United States is estimated at 3.5%, ⁶ but comparative studies in the post-stroke population demonstrate marked variation with prevalence of PS-PTSD ranging from 11% to over 37%.^1,7 Furthermore, PTSD is seen in patients with transient ischemic attack (TIA), ischemic, and hemorrhagic stroke as well.^8-10 Despite this, clinical research into PS-PTSD is limited. Therefore, we designed a study with the aim of identifying the associated risk factors of PS-PTSD, evaluating for co-morbid depression, and studying the impact on quality of life in this population. We hypothesize that those with PS-PTSD are more likely to have co-morbid depression and worsened quality of life compared to those who do not develop PS-PTSD.

Methods

At the University of South Florida Morsani College of Medicine, we conducted a single center prospective observational study of patients diagnosed with an ischemic stroke or TIA at our comprehensive stroke center (Tampa General Hospital). The study was approved by the university’s Institutional Review Board and was supported by the University of South Florida New Researcher Grant. Inclusion criteria consisted of: subjects 18 years of age or older, diagnosis of stroke or TIA 31 days to 1 year prior to enrollment, however exclusion criteria were: pre-existing (preceding the stroke or TIA) depressive disorder, bipolar, psychosis, or PTSD. Although those with a diagnosis of depression preceding the TIA or stroke were excluded, enrolled patients may have been on antidepressants for other indications or started post-stroke/TIA diagnosis. Informed consent was obtained from all patients prior to starting any study procedures. Data was collected from May 1, 2021 to May 31, 2022.

Patients were recruited in the outpatient post-stroke clinic; the questionnaires were conducted in-person via clinical interview after enrollment. All subjects received the post-traumatic stress disorder checklist-5 (PCL-5), a validated questionnaire designed to assess for PTSD symptoms based on the DSM-V criteria.^11,12 This has been updated from the prior PCL checklist for DSM-IV by adding additional PTSD symptoms. This 20-item questionnaire is used to self-identify PTSD symptoms. Items are grouped into clusters based on various symptoms. The clusters include intrusion symptoms such as re-experiencing (B items), avoidance symptoms (C items), negative alterations to cognition and mood (D items), and alterations in arousal and reactivity (E items). Each item is graded on a 0-4 scale with 0 corresponding to “not at all” and 4 corresponding to “extremely”. The final score ranges from 0 to 80. ¹³ A PCL-5 score of 31-33 has been correlated with PTSD diagnosis by clinician-administered structured interviews. ¹³ A score of ≥ 33 demonstrated good sensitivity and specificity. ¹² Therefore, our study on PS-PTSD used a cutoff score of 33 or greater to identify those with significant PTSD symptoms and/or endorsement of moderate symptoms in at least one B item, one C item, two D items, and two E items.

Clinical and demographic data was obtained from the patients as well as from the electronic medical record. Data collected included age, gender, time from stroke/TIA, stroke etiology (TOAST criteria), medications (including antidepressants and statins), stroke severity (NIHSS), and post-stroke neurological disability (modified Rankin Scale). ¹⁴ In addition, subjects also completed the PHQ-9 depression screen and Stroke-specific Quality of Life Scale (SS-QOL-12). ¹⁵ All 20 patients completed the PCL-5 score, but only 19 completed the follow-up scales (one patient was lost to follow up) and were therefore able to be used for analysis.

Statistical Analysis

Data analysis was done using R version 4.2.2. Descriptive statistics were produced to summarize the main demographics of the data. Median test was used to compare the medians between PS-PTSD vs non-PTSD groups.

Results

Of the entire cohort, 35% (7/20) and 32% (6/19) of those who completed the follow-up scales were found to have significant PTSD symptomology post-stroke or TIA. All subjects who completed the study were included in the analysis. Demographic characteristics are seen in Table 1. The patients were mostly middle-aged (average age of 49.3 years) with a lower average age in the PTSD group (43.8 years vs 51.8 years). Most patients were not on anti-depressants. Average PCL-5 score in the PTSD group was notably higher at 35.8 vs 13.1 in those without significant PTSD symptomology.

Table 1.

Demographic characteristics.

	Overall	PS-PTSD	No PS-PTSD
	(N = 19)	(N = 6)	(N = 13)
Age
Mean (SD)	49.3 (15.6)	43.8 (15.8)	51.8 (15.6)
Gender
Female	11 (57.9%)	3 (50.0%)	8 (61.5%)
Male	8 (42.1%)	3 (50.0%)	5 (38.5%)
Event Type
TIA	2 (10.5%)	1 (16.7%)	1 (7.7%)
Stroke	17 (89.5%)	5 (83.3%)	12 (92.3%)
Tobacco use
No	15 (78.9%)	3 (50.0%)	12 (92.3%)
Yes	4 (21.1%)	3 (50.0%)	1 (7.7%)
Stroke Etiology TOAST
Cardio embolism	4 (21.1%)	0 (0%)	4 (30.8%)
Small vessel occlusion	2 (10.5%)	1 (16.7%)	1 (7.7%)
Stroke of other determined etiology	7 (36.8%)	3 (50.0%)	4 (30.8%)
Stroke of undetermined etiology - negative evaluation	1 (5.3%)	0 (0%)	1 (7.7%)
Stroke of undetermined etiology - two or more causes	4 (21.1%)	1 (16.7%)	3 (23.1%)
Large-Artery Atherosclerosis	1 (5.3%)	1 (16.7%)	0 (0%)
Antidepressant use
No	14 (73.7%)	3 (50.0%)	11 (84.6%)
Yes	5 (26.3%)	3 (50.0%)	2 (15.4%)
PCL-5 score
Mean (SD)	20.3 (15.4)	35.8 (12.6)	13.1 (10.6)
Median [IQR]	22.0 [21]	31.0 [25]	13.0 [20]

PCL-5: post-traumatic stress disorder checklist-5.

SD: standard deviation.

IQR: interquartile range.

Total PHQ-9 and SS-QOL scores were correlated with total PCL score. A higher PHQ-9 indicating greater depressive symptoms was correlated with higher PCL scores (r = 0.727, P < .001). Lower SS-QOL total score indicating worse quality of life was correlated with higher PCL scores (r = −0.709, P = .001)

Patients with PS-PTSD had higher PHQ-9 score (Mdn 11.5 vs 6, P = .03). Lower SS-QOL scores were seen in those with PS-PTSD (Mdn 164.5 vs 198, P = .07). Additionally, lower median scores were seen in each of the SS-QOL subgroups (Table S1).

Stroke severity in ischemic stroke patients as defined by the NIHSS was similar across both groups (Mdn 0 vs 0, P = 0.91) (Table 2).

Table 2.

Stroke, disability, depression, and quality of life scores and PS-PTSD.

	PS- PTSD (N = 6)	No PS-PTSD (N = 13)	P-value (median test)
	Median [IQR]	Median [IQR]
Time from event (days)	203 [176]	132 [111]	0.88
NIHSS a	0 [4.5]	0 [1]	0.91
mRS	0.5 [1]	0 [2]	0.88
PHQ-9	11.5 [7]	6 [6.5]	0.03
SS-QOL	164.5 [49]	198 [33]	0.07

Notes: Group differences using median test.

PHQ-9: Patient Health Questionnaire-9; SS-QOL: Stroke specific Quality of Life Scale; mRS: Modified Rankin Scale of disability; NIHSS: National Institutes of Health Stroke Scale.

IQR: interquartile range.

^aNIHSS only reported for patients with ischemic stroke (n = 17).

Discussion

In this observational study, 35% of subjects post-stroke or TIA were found to have PS-PTSD. The substantial variability in the rates reported in the literature (11-37%) can be attributed to various methods of diagnosis. ¹ Our PS-PTSD incidence lies at the higher end of this reported spectrum. Among several elements that contribute to PS-PTSD, we have identified younger age as a factor in this study. This correlates with a large analysis of stroke patients in New York, in which the mean age of those with PTSD was 7 years lower than those without PTSD.^6,9,16 Our cohort had an overall post-stroke population with an average age of 49.3 years; the mean age for those with PS-PTSD was 8 years lower than those without PS- PTSD. Younger people may be more affected by PS-PTSD as they are less likely to have pre-stroke disability and may have more difficulty in returning to work and social situations than older people. This may also impact ability to complete job-related tasks and family responsibilities. Younger populations would also be expected to have more years of life living with disability, which could contribute to higher overall healthcare costs.

Other factors such as gender have been shown to be associated with the development of PS-PTSD. ¹⁶ Although women have twice the lifetime PTSD prevalence compared to men, our study did not demonstrate a similar ratio in regards to PS-PTSD, or see a difference in the stroke etiology associated with the development of PS-PTSD. This incongruency may be due to the small number of patients in this pilot study; in addition, studies show that female vulnerability to PTSD may be specific to assaultive violence, which women are more likely to experience compared to men, rather than generalized to any trauma. ¹⁷

An Erlangen study found a decreased prevalence of PTSD at 12 months compared to 3 months after the event. ¹⁸ A study involving 436 patients evaluated the effects of anatomy and coping style; they found 36 patients at 6 months after the event with probable PTSD, 36% of whom retained their PTSD symptoms 12 months later, 64% had improving symptoms, and 40% developed new symptoms between 6 and 12 months. ¹ It is likely that time from the event has a complex and variable role to play in the development of PS-PTSD but nevertheless, this possible time-dependent correlation underscores the importance of early diagnosis and treatment.

TIA patients have been shown to have PTSD at similar rates as those with stroke, a curious finding given that by definition, TIA results in no residual symptoms or disability from the neurological event. Regardless, studies have shown that one in three patients with TIA can develop PTSD. ¹⁹ A study limited to those over 50 years of age show both admission NIHSS and Modified Rankin Score (mRS) are significantly correlated with developing PTSD within 12 months (P-values 0.01 and 0.001 respectively). ¹ Our study, however, did not find an association between stroke severity on admission (measured by NIHSS) or resultant stroke disability (measured by mRS) and PTSD symptomology. The majority of patients in our study had minor stroke (both groups had mild stroke symptoms and low stroke disability). Our study suggests that even patients with mild stroke with minimal disability are at risk of developing PS-PTSD. This may be explained by difficulty managing societal expectations to reintegrate into society sooner compared to patients who require rehabilitation after severe stroke symptoms and resultant disability. With minimal to no residual symptoms and disability, the patients may face the general public not identifying them as stroke survivors and thus not appreciating that they could still have some disability that precludes them from performing at their fullest activity levels.

Our study identified a correlation between depression (PHQ-9) and PS-PTSD. Patients with PTSD scored higher on the PHQ-9 than those without PTSD. In general, patients with stroke are at increased risk of depression; about one-third of stroke survivors develop post-stroke depression (PSD), typically in the first year. ²⁰ PSD is associated with poorer functional outcomes and suicidality.^21,22 Those with PS-PTSD are also at risk for PSD and may even have increased depressive symptoms within the first year after stroke. ¹ Both conditions involve negative changes in mood and thinking; when they coexist, they can intensify each other’s effects. The combination of depression and PS-PTSD may lead to increased emotional distress, hopelessness, and a reduced ability to cope with the impact of the stroke.

Greater PTSD symptomology (higher PCL-5) was negatively correlated with quality of life (SS-QOL). Individuals with PS-PTSD following a stroke showed lower scores in SS-QOL. In a study involving prevalence of PTSD after SAH, the quality of life was far lower for patients with PTSD than without. ²³ Similarly, a cross-sectional telephone survey found that ischemic stroke patients with anxiety, depression and PTSD reported significantly worse quality of life. ²⁴ The quality of life was measured using EuroQOL EQ-5D, and not SSQOL, but nevertheless shares common domains such as mobility, self-care, usual activities, pain, and anxiety/depression. In general, PS-PTSD can significantly affect a stroke survivor’s quality of life. The presence of this condition may lead to persistent psychological distress, impairments in daily functioning, and reduced overall well-being. Symptoms such as anxiety, sadness, irritability, and intrusive thoughts can make it challenging for individuals to engage in social activities, maintain relationships, and participate in rehabilitation efforts. The resulting isolation and reduced participation in meaningful activities can further contribute to a diminished quality of life.

Limitations

Our study is limited due to its small patient size and scope of one time point in patients stroke recovery. The interview was conducted during post-stroke follow-up; therefore, the patients’ answers to the questions may have differed at various lengths after their initial diagnosis was made. The diagnosis of PTSD is made by a clinical interview with the clinically administered PTSD scale. Our study used the PTSD checklist as a validated questionnaire designed to assess for PTSD symptoms based on the DSM-V criteria. A PCL-5 score of 31-33 has been shown to correlate with PTSD diagnosis by clinician administered structured interviews. ¹¹ Nevertheless, we used a cut off of 33 and recommended criteria with good correlation to PTSD diagnosis. Despite these limitations, the study contributes pertinent information to the PS-PTSD literature.

Conclusion

This study highlights the significant presence of PS-PTSD after TIA and stroke. We found that younger patients appear to bear a larger burden of PS-PTSD. Those with low stroke severity and disability are equally affected compared to those with higher stroke severity and disability. PS-PSTD is associated with depression and worsened quality of life. Despite one in three patients expressing significant PTSD symptomology in the first year after stroke and large impacts on quality of life, there is no standard to screen for PS-PTSD. Future studies are needed to assist with identification, rapid triage, and treatment of those with PS-PTSD.

ORCID iDs

Swetha Renati https://orcid.org/0000-0001-6479-3634

Alena Makarova https://orcid.org/0000-0002-2585-6943

Kanita Beba Abadal https://orcid.org/0000-0002-5694-7725