Two important research lines have been structured for vascular vertigo and dizziness (VVD). One is related to vertigo or dizziness due to stroke or transient ischemic attack (TIA) with the primary focus on their diagnosis in the emergency department (ED). The other line is dedicated to the prognosis of vertigo and dizziness due to vertebrobasilar TIAs (VBTIAs).

Diagnosing TIA is challenging because it is mostly based on the patient's symptoms, the physician's ability to interpret them correctly, and the exclusion of other diagnoses (Tuna and Rothwell, 2021; Yao and Zu, 2023). Many studies still refer to the National Institute of Neurological Disorders and Stroke (NINDS) criteria for diagnosing TIA (Paul et al., 2013). However, those do not properly embrace vertigo and dizziness as symptoms of TIA. Thus, the studies on vertigo and dizziness due to VBTIA have been sparse (Markus et al., 2013; Pollak, 2012; Fife et al., 1994; Jenkins et al., 2001).

In 2021, our group established a cohort of patients with VVD and treated them according to the Guidelines for Prevention of Stroke in Patients with Stroke and Transient Ischemic Attack published by the American Heart Association and American Stroke Association (AHA/ASA) (Kleindorfer et al., 2021). The majority of patients suffered from recurrent VBTIAs. In 2022, the diagnostic criteria for VVD were launched by the Bárány Society, which summarized and facilitated the characterization of this clinical entity and injected fresh air into our cohort (Kim et al., 2022).

This study aimed to characterize the clinical features, response to treatment, and prognostic factors of VVD due to VBTIA that could interfere with the outcomes after treatments.

This cohort study started in January 2021 at the Neuro-otology Service of the Clinics Hospital of the Medical School of São Paulo University. In this article, we present the data collected until January 2024. All patients were first evaluated in an ED or an outpatient clinic.

Out of 1,957 patients, 103 were finally included in the study. Among these patients, 58 (56.3%) were female individuals, with a mean age of 70.9 ± 9.3 years (ranging from 37 to 85 years). In all, 69 patients (67.0%) were recruited from the outpatient clinic, and 34 patients (33.0%) were recruited from the ED. All data are summarized in Supplementary Table 1. Figure 2 shows the selection process.

Up to 25% of strokes are preceded by a TIA (Giles and Rothwell, 2007; Halmagyi, 2017). The early risk of stroke after a TIA is high, requiring urgent investigation and treatment (Johnston et al., 2000; Rothwell et al., 2007; Lavallée et al., 2007). In addition, considering the evidence of high stroke risk in transient VVD, active secondary prevention measures, including antithrombotic agents, statin, and risk factor control, deserve special attention (Yao and Zu, 2023).

In this sample, we observed a balanced distribution of sex, age, and prevalence of comorbidities, similar to the distribution found in VVD studies conducted in ED settings (Filippopulos et al., 2022; Kuroda et al., 2017) and studies including both VB and anterior circulation strokes (Andersen et al., 2010).

Other authors have presented the elapsed time from symptom onset to the diagnostic evaluation and treatment (Tuna and Rothwell, 2021; Pollak, 2012; Fife et al., 1994). The mean interval from symptom onset to recruitment was 11.8 months. However, four patients presented with symptoms that lasted longer than 5 years (Supplementary Table 1). Many factors may favor the infratentorial region for TIA rather than stroke. One is the high proportion of the white matter and the vascular distribution in the cerebellum and brainstem because collaterals can be formed by the rich web from the main trunks (Supplementary Figure 2) (Mattle et al., 2011). This anatomical background probably explains why TIAs and isolated transient neurological attacks (TNAs) are more common in the 90 days preceding a vertebrobasilar stroke compared to their occurrence before a carotid stroke [noted in 59 out of 275 patients before a vertebrobasilar stroke vs. 65 out of 759 patients before a carotid stroke; odds ratio (OR) 2.92, 95% CI (1.99, 4.28), p < 0.0001] (Paul et al., 2013). This same rationale supports a longer time window for the treatment of VB strokes, such as those caused by occlusion of the basilar artery (Mattle et al., 2011; Merwick and Werring, 2014; Räty et al., 2024; Alemseged et al., 2023). Another possible protective factor found in the participants in our study was the prior use of antiplatelet therapy, even with a regimen insufficient to prevent VVD attacks (Tuna and Rothwell, 2021). However, both physicians and patients probably neglected this condition, which contributed to the long interval between the onset of symptoms and the first evaluation in patients with recurrent VBTIA. It is estimated that only 22% of patients with TNA from the brainstem sought medical care before stroke. In only one case (2%), a vascular cause was suspected by their physician (Paul et al., 2013).

Vertigo/dizziness was the only symptom in 16 of our patients. Some studies have reported that isolated episodic vertigo was the only manifestation in 21%−62% of patients with a presumptive diagnosis of VBTIA and that 29% of patients with VB infarction had a history of isolated episodic vertigo (Kim et al., 2022; Gomez et al., 1996; Grad and Baloh, 1989). Even in VB strokes, obvious focal neurological signs were absent in more than half of the individuals presenting with acute vestibular syndrome (Tarnutzer et al., 2011).

Another diagnostic challenge was that nearly 10% of patients had no comorbidities other than age. Conducting a thorough search for other associated symptoms of hypoflow, undiagnosed risk factors, and other possible etiologies, such as postural hypotension and medication side effects, among others, is imperative before considering prophylactic treatments.

Findings on MRA or CTA scans may also contribute to the suspicion of VBTIA. In one study, patients with TNA were more likely to show arterial stenosis ≥50% in the posterior circulation than those with “classic” TIA [according to the NINDS; 84 of 467 (18%) vs. 80 of 896 (9%); odds ratio (OR) 2.21, 95% CI (1.59, 3.08), p < 0.0001] (Tuna and Rothwell, 2021). The correlation between the luminal irregularities in the posterior circulation and VVD is well established (Choi et al., 2017; Lima Neto et al., 2017). In our sample, 39 (37.8%) patients presented with vertebrobasilar artery abnormalities, with vertebral artery stenosis and severe hypoplasia being the most frequent. There is substantial evidence for vertebral artery hypoplasia as a risk factor for posterior circulation ischemia, especially when other vascular risk factors coexist (Kleindorfer et al., 2021; Katsanos et al., 2013; Park et al., 2007). Nevertheless, in patients with a high suspicion of VBTIA and without findings in MRA or CTA (64 patients in our sample, 62.2%), abnormalities of the secondary arteries and focal atherosclerosis should be considered in the pathogenesis of symptoms (Park et al., 2014; Kim et al., 2011; Lima Neto et al., 2020).

Even though we achieved an acceptable average follow-up time, a longer follow-up period would be beneficial for monitoring vascular and chronic disorders. Throughout this period, the use of aspirin, clopidogrel, DAPT, and anticoagulants appear to be an effective approach in treating VVD patients, as only seven patients (6.8%) had a new attack. Considering that two of these cases had their new attack the day after the prescription, it is likely that they were not yet under the effect of aspirin (Figure 4). Another three patients experienced a new attack as soon as they were switched from DAPT to a single therapy. Using anticoagulants and stent placement in patients with VVD who have not responded to antiplatelet therapy has been reported in some case series; however, there is no established guidance on how to implement this approach (Fife et al., 1994; Jenkins et al., 2001). Our results suggest that this management of the treatment strategies must be individualized for each patient. This will certainly be the focus of future studies.

However, a reasonable criticism would be that we excluded patients whose first crisis was a stroke, patients with recurrent strokes, and patients who were already on an antithrombotic regimen that did not need to be changed according to the AHA/ASA guidelines. These excluded cases may have had a worse prognosis with more recurrences. We decided not to include them because we focused our method on VBTIA and on evaluating the possible reduction in the number of VVD attacks with treatment. For example, we would not know whether the dizziness was related to recurrent ischemic events or sequelae of stroke. Similarly, the response of patients whose regimens were neither initiated nor changed by us could not be evaluated. As this cohort study is ongoing, we aim to have a larger sample of these individuals whose treatment regimen remained unchanged to form a comparison group.

This sample did not allow us to identify factors that could increase the risk of developing a new attack in patients receiving treatments. Even in other prospective study with 17,009 years of follow-up, the number of cases of vertigo/dizziness (considered “non-consensual TIA”) was too small to reliably estimate the risks (Tuna and Rothwell, 2021). Only one variable, “VB large artery disease,” showed significance (p = 0.04) but favored the absence of events. Even though the mechanism is unknown, we hypothesize that this protective effect against a new attack may have been due to neovascularization facilitated by the chronic hypoperfusion in these patients because most of them presented vertebral artery hypoplasia. However, given the low statistical power of this result, a confounding factor must be considered due to the small number of patients with a recurrence. Our data will be reviewed again when we have a larger sample size and longer follow-up.

We emphasize that this cohort study was conducted in a Latin American country with a predominantly mixed ethnicity and a tropical climate. Future comparisons must consider these characteristics.

Some implications of our findings for clinical practice are as follows: (a) we demonstrated that secondary prophylaxis is the correct treatment for vertigo/dizziness attacks due to VBTIA; (b) we also confirmed that “non-consensual VBTIA” is “VBTIA indeed” in terms of treatment efficacy in preventing stroke, aligning with authors who previously used this term (Tuna and Rothwell, 2021); (c) patients with VVD due to VBTIA should not be routinely excluded from clinical trials and other research studies; (d) medical education is needed to provide physicians with the knowledge and tools to identify and treat these patients; and (e) public education is also essential to encourage people to seek medical attention after an episode of dizziness or vertigo.

VBTIA has a broad clinical spectrum. In this study, the predominant symptom was weekly imbalance attacks lasting up to 10 min accompanied by transient visual loss, syncope, diplopia, and balance-related falls.

Adhering to AHA/ASA guidelines, secondary prevention treatments using aspirin, clopidogrel, or DAPT were quite effective, with 93.2% of patients showing no recurrences during the follow-up period. However, this sample size was insufficient to identify factors that might increase the risk of a new attack in treated patients.