4 PATIENTS AND METHODS
4.1 Study population
Helsinki Young Stroke Registry (I)
The HYSR constitutes all consecutive first-ever early-onset ischemic stroke patients (15-49 years, n=1008) treated in the Helsinki University Hospital (HUH) area between January 1994 and May 2007. The catchment area of HUH is approximately 1.5 million inhabitants, with the only neurological emergency unit in that area, and is a referral center especially for young stroke patients. The proportion of inhabitants aged between 15 and 49 years ranged between 46% and 47%. Study population and methods have been described in detail previously.3 Briefly, all patients underwent either brain CT or MRI, and ischemic stroke was defined as an episode of focal neurological deficits with an acute onset and lasting for at least 24 hours. If acute symptoms lasted less than 24 hours, imaging-positive lesions with corresponding symptoms were required for ischemic stroke diagnosis. Thus, patients with transient ischemic attack were excluded. Medical records were used to obtain data on comorbidities and baseline characteristics. For this work, a modified TOAST classification was used to compare the baseline characteristics and long-term outcome between early-onset ESUS and other well-defined etiologies. ESUS was well-defined as presented by Hart and colleagues in 2014.19,43 ESUS patients were further classified as definite ESUS (if meeting all criteria for ESUS) and probable ESUS (if intracranial arteries were not imaged). Furthermore, patients with incomplete diagnostic work-up or with multiple potential etiologies were reclassified as undetermined non-ESUS.
Searching for Explanations for Cryptogenic Stroke in the Young: Revealing the Etiology, Triggers, and Outcome (II-IV)
Based on the high frequency of cryptogenic ischemic strokes in the young observed in our studies as well as those of others (Study I and prior studies utilizing the same material), we decided to design a case-control study focusing specifically on these patients, Searching for Explanations for Cryptogenic Stroke in the Young: Revealing the Etiology, Triggers, and Outcome (SECRETO; NCT01934725) is an international prospective multicenter case-control study of young adults presenting with first-ever cryptogenic ischemic stroke. It has been
42 approved by the Ethics Committee of the Helsinki and Uusimaa Hospital District (362/13/03/00/2012) and the local Ethics Committees at each recruiting center. Written informed consent from all participants is mandatory.
In SECRETO, we planned to include all consecutive patients aged 18-49 years with first-ever cryptogenic ischemic stroke. A detailed protocol was drafted and a pilot study recruiting 50 patients was scheduled to be performed at Helsinki University Hospital in 2013-2015.
According to the study protocol, the minimum diagnostic work-up for each patient included brain MRI as well as MRA or CTA of intracranial and extracranial vessels, at least 24 hours of prolonged continuous ECG monitoring, routine laboratory testing, and both TTE and TEE.252 In selected centers, the presence of right-to-left shunt was additionally examined with TCD-BS as described above. In SECRETO, the magnitude of the shunt was examined in a unilateral TCD monitoring using agitated 10 mL saline-blood or saline-only solution as a contrast as described above. The test was performed at rest and with the Valsalva maneuver.166
We defined cryptogenic ischemic stroke using the A-S-C-O phenotyping as the absence of any disease of grade I (“definitely a potential cause of the index stroke”).29 Each patient was sex- and age-matched (±5 years) with one stroke-free control. Controls with a similar ethnic background were recruited from the same region. From the Population Registry, 20 potential controls were randomly identified for each patients and invitations were sent them. If one stroke-free control was not found for each patient with this method, a fit, willing control person was recruited from the community. Methods to find community controls were not pre-specified, as standardized strategies might not be feasible in all sites and settings; sources may include e.g. patients’ non-related proxies or proxies of the study personnel.
Baseline characteristics, cardiovascular risk factors, and comorbidities were obtained from medical records and with a structured interview of both patients and stroke-free controls.
Relevant factors included level of education, history of hypertension, dyslipidemia, and diabetes, smoking, diet, prior medications, physical activity, and measurement of waist-to-hip ratio. Validated questionnaires used to assess lifestyle factors included the short version of the International Physical Activity Questionnaire253 and an adaptation of the World Health Organization Alcohol, Smoking and Substance Involvement Screening Test.254
In Study III, a migraine screener was developed and validated against an experienced senior headache-neurologist. The main criteria of the ICHD, 3rd edition were applied in the development of the screener, which comprised nine questions (see Table 5 and Figure 2). The
43 first three questions focused on headache attacks and presence of visual aura and the remaining six questions on headache characteristics more specifically. The participant was judged as a non-migraineur if all of the first three questions were negative. With the following six questions, each participant was diagnosed with either MA or MO.
44 Table 5. Questions used to construct an algorithm to diagnose migraine and classify its subtypes in this study. If the response to any of the first three screening questions was something other than ‘Never’, the following six questions characterizing headaches were asked.
Have you had repeated headache attacks anytime during your
lifetime? Never
Have you ever had a headache attack that was preceded by a disturbance in your vision (such as zigzag lines, shimmering or vibrating patches, blind spots, bright lights, or blobs for several minutes)?
Have you ever had one or more attacks with disturbance in your vision (such as zigzag lines, shimmering or vibrating patches, blind spots, bright lights, or blobs) lasting 10-30 minutes without headache?
Duration Do your headache attacks often last more than 4 hours (if you
do not treat them)? No
Yes Nausea/vomiting,
photophobia/
phonophobia
Do you become sick (feel nausea or vomit) during your headache attacks OR do lights and sounds bother you when you have a headache?
No Yes Severity Do your headache attacks limit your ability to work or take part
in hobbies or household work?
No Yes Aggravation by
physical activity Are your headaches usually aggravated by physical activity? No Yes Pulsatility Are your headaches usually pulsatile? No Yes Unilaterality Are your headaches usually unilateral? No
Yes
45 Figure 2. Decision algorithm with three screening and six ancillary questions.
In Study IV, endothelial function was measured using the EndoPAT 2000 device (Itamar Medical Inc., Caesarea, Israel) both in patients and controls. This device is a non-invasive method measuring arterial tone change. The endothelium regulates the homeostasis of e.g.
oxidative stress and inflammation, which can be affected by several lifestyle factors and early atherosclerosis. In EndoPAT, endothelial function is measured with specifically designed biosensors that are placed on the tip of each index finger. Consumption of caffeine or alcohol prior to the test can affect the results, as can smoking or taking such medicines as alpha blockers, calcium channel blockers, and statins on the same day. Patients should also be in a calming environment without distractions.237 During the measurement the temporary occlusion of 5 minutes in the test arm is applied using air cuffs. Before and after occlusion, finger-tip arterial blood volume is registered also for 5 minutes. The signal after cuff release is compared with baseline recording in the test arm and further indexed to the contralateral control arm.
Results are reported as Reactive Hyperemia Index (RHI), and natural log transformation of RHI (LnRHI) is used to estimate endothelial function.255
46