Abstract
Background
Long-term outcomes of reversible cerebral vasoconstriction syndrome (RCVS) have not been systematically investigated.
Methods
The following validated questionnaires were mailed to patients recruited from the RCVS registries of two academic hospitals: headache screening form, Headache Impact Test, Migraine Disability Assessment Test, Barthel Index (BI), EuroQoL (EQ-5D-5L) and Patient Health Questionnaire (PHQ-9).
Results
Of the 191 patients in the registries, 109 could be contacted and 45 responded. Median follow-up time after symptom onset was 78 months. After RCVS resolution, 24 (53%) patients continued to have headache, but the majority (88%) reported improvement in its severity. Thirteen of the 24 patients with persistent headache had a history of migraine prior to RCVS diagnosis. The majority (97.5%) of respondents were functionally independent based on BI scores. EQ-5D-5L showed better scores in the domains of mobility, self-care and usual activities, as compared to pain and anxiety/depression. Patients with persistent headache had significantly higher levels of EQ-5D-5L pain scores. PHQ-9 scores revealed only one patient (3%) with severe depression.
Conclusion
More than half of RCVS patients will continue to have chronic headaches of mild to moderate intensity that are distinct from the “thunderclap” headaches at RCVS onset. The vast majority regain complete functional ability. However, pain and anxiety/depression are frequent, often aggravated by concomitant chronic headaches, and may be associated with lower quality of life.
Introduction
Reversible cerebral vasoconstriction syndrome (RCVS) is usually a monophasic illness characterized by the acute onset of severe headaches, which often recur over a span of days to two to three weeks. Neurological deficits from ischemic and hemorrhagic strokes, or brain edema, are common. The diagnosis rests on documenting multifocal cerebral arterial vasoconstriction, which typically reverses within three months (1). RCVS is a major cause of thunderclap headaches and includes previously described distinct entities such as Call-Fleming syndrome, drug-induced angiopathy, migraine angiitis, benign angiopathy of the central nervous system (CNS), post-partum angiopathy, and drug-induced vasospasm (2). To date more than 500 cases have been published in the literature, including three large cohort studies (3–5). In our experience, stroke-related neurological deficits usually improve (i.e. the outcome is usually benign); however, many patients continue to report chronic headaches and depression. This has implications for long-term management, including treatment with antimigraine agents and antidepressants since these agents have been implicated in precipitating RCVS (6,7). In this collaborative study we systematically assessed long-term headache, pain, disability, depression, and quality of life (QoL) after an episode of RCVS in a large cohort of patients.
Methods
The study was conducted after approval from the institutional review boards of the Cleveland Clinic Foundation (CCF) and Massachusetts General Hospital (MGH). Patients with RCVS were recruited from the RCVS registries of both these institutions. Inclusion of patients into these registries was based on guidelines proposed by Calabrese et al. for the diagnosis of RCVS (2,3). At CCF, patients were contacted by phone to verify their mailing address and to introduce the study. Consent forms and validated questionnaires were then mailed to the patient’s home address on a one-time basis. Written consent was obtained from each patient prior to participation. At MGH, permission was obtained from the patient’s primary care provider to mail them informed consent forms and questionnaires. Patients were given the option of declining participation, or returning completed questionnaires (implied consent).
The questionnaires included the i) Headache Impact Test-6 (HIT-6) (8), ii) Migraine Disability Assessment Test (MIDAS) (9), iii) Barthel Index (BI) (10), iv) Patient Health Questionnaire (PHQ-9) (11,12) and v) European Quality of Life (QoL) Questionnaire (EQ-5D-5L) (13). These are available as supplementary material. We also included a headache screening questionnaire created specifically for this study in order to capture any persistent headache, and if so the characteristics, of headaches occurring after the resolution of the incident episode of RCVS. It included the following questions:
Since the diagnosis of RCVS, are you still having headaches (Yes/No). If yes, then the following questions were asked: Since your diagnosis of RCVS, how would you characterize your headache? (The headache has partially improved, but not completely resolved/The headache remained the same/The headache has gotten worse) If your headache has improved, how would you describe it? (It has improved in frequency only/It has improved in intensity only/Both). If your headache has gotten worse, how would you describe it? (It has increased in frequency only/It has increased in intensity only/Both). Is the characteristic of the headache you are experiencing now similar from the time you were diagnosed of RCVS? (Yes, it is similar/No, it is not similar).
For the patients who could not be contacted, mortality data were checked in hospital inpatient/outpatient clinic records and at two online resources (http://www.ancestor.com and http://www.archives.com) using social security numbers.
Continuous variables were described using means and standard deviations. Categorical variables were described using counts and percentages. Continuous variables were compared using Welch’s two-sample t-test. Categorical variables were compared using either Fisher’s exact test or Pearson's chi-squared test, as appropriate. All analyses were performed using R software (version 3.0.2, Vienna, Austria). Testing was two-sided and considered significant at the 5% level. P values from analyses comparing subgroups are reported without adjustment for multiplicity.
Results
A total of 191 patients were present in the RCVS registries (57 CCF, 134 MGH). We excluded 82 patients (26 CCF, 56 MGH) because of death, loss to follow-up (mailer returned because of address change, or no response to telephone calls at prior phone numbers), or inability to obtain permission from the caregiver to contact their patient for research. Among the CCF patients, 31 received the mailed questionnaires, of whom 20 consented and returned the questionnaires, three declined to participate and eight agreed to participate but never returned the forms. At MGH, 78 patients were sent questionnaires; 25 returned the questionnaires, five declined to participate, and 48 did not return forms in time for the analysis (Figure 1).
Flowchart demonstrating patients enrolled in the study.
Clinical and neuroimaging features.
For each variable, the number of patients for whom data was available is noted. ICH: intracerebral hemorrhage; SAH: subarachnoid hemorrhage; TCH: thunderclap headache.
Headaches
Twenty-four (53%) patients (21 women and three men) continued to have headache, while 21 (47%) had complete resolution of acute RCVS-associated headaches and no headaches thereafter. Median follow-up time from RCVS diagnosis was 78 and 73 months in those with and without persistent headache, respectively. Of the two patients who did not present with headache at RCVS onset, only one developed persistent headache on follow-up. Thirteen of the 24 patients with persistent headaches had a past medical history of migraine prior to the diagnosis of RCVS. Therefore, 11/24(46%) patients developed new headaches after RCVS. Most patients (21/24 or 88%) with persistent headache reported improvement in the character of the headache, while one (4%) remained the same, and two (8%) reported worsening headache compared to the initial RCVS headache. Among the patients with improved headaches, 18/21(86%) reported improvement both in headache frequency and intensity, while two of 21(10%) patients reported improvement in headache frequency only. The two patients who had worsening of headache included one who reported worsening frequency, and the other worsening intensity. Eight patients responded that the persistent headache was similar in character to the initial RCVS headache. We were able to contact five of eight by telephone and they clarified that their current headaches were not thunderclap headache.
Headache impact on life was assessed using the HIT-6 scale, and the mean score among 35 respondents was 49.1 ± 10.2 (median 50). Very severe impact (HIT-6 score ≥60) on life was present in five (14%) patients, while substantial, some, and little-to-no impact was present in eight (23%), five (14%) and 17 (49%) patients. Headache severity and disability was assessed using the MIDAS scale among 36 respondents, and mean score was 8.75 ± 17.3 (median 1.5). Severe disability (MIDAS score ≥21) was present in five (14%) patients, while moderate, mild and little/no disability was present in three (8%), two (6%), and 26 (72%) patients (Figure 2). As expected, patients with persistent headache had significantly higher scores on the HIT-6 (53.1 ± 9.3 vs. 40.4 ± 2.5, p = 0.002) and MIDAS (13.0 ± 20 vs. 0.33 ± 1.2, p = 0.037) scales compared to those without persistent headache. They also had a significantly higher number of headache days in three months (15.7 ± 19.2 vs. 0.33 ± 1.2, p = 0.009) and rated the headache intensity higher on a scale of 0 to 10 (5.2 ± 3.1 vs. 0.3 ± 0.9, p < 0.001).
Distribution of HIT-6 (N = 35, Figure 2(a)) and MIDAS (N = 36, Figure 2(b)) scores.
There was a significant difference in the rate of developing persistent headache depending on whether the patients suffered an initial stroke. Those patients who did not suffer an initial ischemic stroke, SAH or ICH had a higher rate of developing persistent headache compared to those who suffered an initial stroke (80% vs. 40%, p = 0.011).
Disability and QoL
Functional independence with activities of daily living was assessed using the BI. Of 40 patients who completed the BI scale, 39 (97.5%) patients scored ≥85 and were functionally independent. One (2%) patient scored 25 on the BI, and was the only participant who was functionally dependent. The BI score in 25/26 (96%) patients who suffered an initial ischemic stroke, SAH or ICH was ≥85, while all 14/14 patients without an initial stroke who responded to the BI scored ≥85.
The EQ-5D-5L measured QoL across five domains that included mobility, self-care, usual activities, pain and anxiety (Figure 3) and was completed by 40 patients. Results revealed that 29 (64%), 36 (80%) and 27 (60%) patients had no problems with mobility, self-care and usual activities, respectively. The same patient who was functionally dependent by BI scale was the only one who was unable to perform in any of these three domains. Fewer patients reported no pain (38%), or not being anxious or depressed (42%). Twenty-three (51%) patients reported slight, moderate or severe pain while 21 (46%) patients reported being slightly, moderately, severely or extremely anxious or depressed. Patients with persistent headache were more likely to have higher levels of pain as measured by the EQ-5D-5L (p = 0.0083), but anxiety/depression was not significantly different between patients with and without persistent headache (p = 0.760). The visual analog scale (VAS) in EQ-5D-5L reports a numerical score of the patient’s health ranging from 0 to 100 (worst health to best health). From a total of 40 respondents, the mean score was 75.3 ± 19.1 (median 80, range 10–95). Patients without persistent headache had a score of 78 ± 21.7 on the VAS compared to 73.4 ± 17.4 in those with persistent headache (p = 0.465).
(a) Distribution of patient (N = 40) EQ-5D-5L scores across the five domains. (b) and (c) Pain and anxiety domains, with percentages within each severity grade of patients with and without persistent headache.
Depression assessment using the PHQ-9 was available from 40 respondents. It revealed that one (3%) patient had severe depression. Moderately severe, moderate, mild and minimal depression was present in three (8%), six (15%), 13 (33%) and one (3%) patients. There was no difference in the severity of depression between those patients with and without persistent headache (p = 0.217) (Figure 4).
(a) Distribution of patient (N = 40) PHQ-9 scores. (b) Breakdown of patients with and without persistent headache, and the percentages within each grade of depression severity.
Mortality data.
F: female; M: male; RCVS: reversible cerebral vasoconstriction syndrome.
Discussion
This is the first study investigating the long-term outcomes of headache and stroke in patients with RCVS along with health-related QoL. The acute presentation of RCVS is well characterized through large case series. RCVS presents with acute thunderclap headache, which can be recurrent for one to two weeks, with moderate headache persisting between attacks. Presentation without headache is exceptional and the headache usually subsides about three weeks after symptom onset (4,5). Sparse information is available regarding the presence and characteristics of headache long term. Our data reveal that more than half (53%) of the patients will continue to have headache long term. However, the majority of them reported improvement in headache, with only 8% reporting worsening as compared to their headache at RCVS diagnosis. Severe headache and severe disability from headache were present in only 14% of patients. Although we do not know the character of the headache in patients with persistent symptoms, none of the patients had thunderclap headaches. Limitations of the survey methodology prevent us from characterizing the headache. The persistent headache could represent migraine, and it must be noted that 13/24 patients with persistent headaches had a past medical history of migraine prior to the diagnosis of RCVS. These data are subject to the retrospective data capture of the migraine history and the number of migraineurs may be underestimated.
Although close to two-thirds of patients had an initial ischemic stroke, ICH and/or SAH, almost all were independent at the time of responding to the questionnaires. This was also reflected in EQ-5D-5L scores, which showed a majority of the patients had no problems with mobility, self-care or usual activities. However, there were fewer patients who reported that they experienced no pain, or were not anxious/depressed, and those patients with persistent headache had significantly higher pain levels on the EQ-5D-5L. It must be noted that pain in this scale refers to pain in general and need not be a reflection of headache alone. However, analysis of the headache scores shows that patients with persistent headache had significantly higher scores on the HIT-6 and MIDAS scales. They also had significantly higher number of headache days in three months and rated the headache intensity higher. This suggests that headache may potentially be the main driver of pain. Severe depression was present in only 3% of patients.
The reasons for long-term persistent headache in RCVS patients could be multiple. Serotonergic and adrenergic drugs are commonly implicated precipitants for RCVS (3,4). Up to 45% of total respondents had use of a vasoconstrictive medication as a trigger for RCVS, and 54% of the patients with persistent headache had a previous history of migraine. As such, removal of antimigraine vasoconstrictive agents such as triptans after the initial presentation of RCVS could be one reason for persistent headache. The sensory divisions of the trigeminal nerve innervate cerebral arteries affected in RCVS (2). After the initial phase of vasoconstriction, nociceptors in this field of the trigeminal ganglion could become sensitized, which could subsequently lead to sensitization of the trigeminal nucleus caudalis, and then the brainstem nuclei/thalamus. Strokes could also induce abnormalities in the brainstem pain modulatory nuclei, such as the periaqueductal gray (PAG), the serotonergic dorsal raphe nucleus and the noradrenergic locus ceruleus nucleus (14).
The advantages of this study include the long follow-up period and the various outcomes measured. We acknowledge that the study has multiple limitations, the foremost of which is a small sample size of 45 patients who completed questionnaires out of the 191 in our cohorts. The high number of patients lost to follow-up is significant, and may affect the findings of this study secondary to response bias susceptibility. There could also be potential for selection bias from patients who refused to participate. The duration of follow-up among respondents was also variable. We do not have details of the headache characteristics and it is not clear if the reported persistent headaches represent primary headaches such as migraine. Further, no data were available on additional neurological events from the time of diagnosis till the time of the study, since many patients did not continue to follow up. As such, our data do not determine causality between RCVS and the measured outcomes. The study design also does not allow us to evaluate the recurrence rate of RCVS. However, we know from previous series that most patients who have strokes improve gradually over several weeks and few have residual deficits (2,4,5). This is reflected in this study also with up to 98% of patients being functionally independent with activities of daily living. However, it is should be noted that only survivors of RCVS and those who were not severely disabled likely replied to the questionnaire. Although headache could be a potential factor associated with worse pain, anxiety and depression in patients with persistent headache, this could be unrelated. It should be noted that other systemic conditions that could influence these domains were not accounted for.
Conclusion
More than half of patients with RCVS will continue to have headache years after the episode of illness. They are not similar to the RCVS onset headache and are markedly improved from the initial headache. Although close to two-thirds of patients in this study suffered from an initial ischemic or hemorrhagic stroke, almost all were independent with little functional disability. However, pain and anxiety/depression might be associated with lower QoL. Headache may be a potential factor aggravating pain in those patients with persistent headache.
Footnotes
Clinical implications
More than half of patients with reversible cerebral vasoconstriction syndrome (RCVS) will continue to have chronic headaches. These are of mild to moderate intensity and distinct from the “thunderclap” headaches observed at RCVS onset. Despite the frequent occurrence of ischemic or hemorrhagic strokes, the vast majority of patients regain complete functional ability. Pain and anxiety/depression are frequent, often aggravated by concomitant chronic headaches, and may be associated with lower quality of life.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Author contributions are as follows
Dr Seby John: Study concept and design, acquisition of data, analysis and interpretation, critical revision of the manuscript for important intellectual content.
Dr Aneesh B. Singhal: Study concept and design, acquisition of data, analysis and interpretation, critical revision of the manuscript for important intellectual content.
Dr Leonard Calabrese: Study concept and design, analysis and interpretation, critical revision of the manuscript for important intellectual content.
Dr Ken Uchino: Study concept and design, analysis and interpretation, critical revision of the manuscript for important intellectual content.
Dr Tariq Hammad: Study concept and design, acquisition of data, analysis and interpretation, critical revision of the manuscript for important intellectual content.
Dr Stewart Tepper: Study concept and design, analysis and interpretation, critical revision of the manuscript for important intellectual content.
Dr Mark Stillman: Study concept and design, analysis and interpretation, critical revision of the manuscript for important intellectual content.
Ms Brittany Mills: Acquisition of data, critical revision of the manuscript for important intellectual content.
Ms Tijy Thankachan: Acquisition of data, critical revision of the manuscript for important intellectual content.
Dr Rula A. Hajj-Ali: Study concept and design, analysis and interpretation, critical revision of the manuscript for important intellectual content, study supervision.
Conflict of interest
Dr Seby John has nothing to declare. Dr Aneesh B. Singhal has served as chair of the American Academy of Neurology Taskforce on Raising Awareness of Stroke in Young Adults, and has served as a medicolegal expert witness. Dr Leonard Calabrese, Dr Ken Uchino, Dr Tariq Hammad, Dr Mark Stillman, Dr Stewart Tepper, Ms Brittany Mills, Ms Tijy Thankachan and Dr Rula A. Hajj-Ali have nothing to declare.
Acknowledgements
Permission to collect data using the HIT-6 survey was obtained by QualityMetric Incorporated.
Permission to use MIDAS was provided by Dr Richard Lipton through personal correspondence.
For the EQ-5D-5L: UK (English) v.2 © 2009 EuroQol Group. EQ-5D™ is a trade mark of the EuroQol Group.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
