ED Dizziness and Stroke Risk: What The Meta-Analysis Found

Key Takeaways:

• Across the included studies, stroke was reported in a minority of overall emergency department dizziness presentations and at a higher pooled rate in isolated dizziness cohorts.
• Prevalence estimates varied substantially according to diagnostic method, imaging coverage, and hospital setting.
• HINTS and STANDING were summarized favorably overall, TriAGe+ was framed as more flexible, and ABCD2 was reported as having limited value in this context.

The authors pooled data from 29 studies involving 161,013 patients, offering an international evidence base across multiple care environments. The pooled analysis aimed to quantify stroke prevalence across varied emergency settings and study designs. The synthesis focused on ED patients aged ≥16 years presenting with dizziness, vertigo, or unsteadiness and assessed how often stroke was ultimately identified. The authors also separated patients with isolated dizziness from broader emergency department dizziness cohorts to examine whether the reported burden differed between groups.

In the isolated-dizziness subgroup, 9 studies including 2,559 patients yielded a pooled stroke prevalence of 13.9%. Investigators treated isolated dizziness as a distinct cohort rather than merging it with all dizziness presentations. Diagnostic method and hospital level were described as the main contributors to between-study heterogeneity in this analysis. The evidence base for this subgroup was much smaller than the broader cohort despite the higher pooled estimate. Across studies, stroke-prevalence estimates varied substantially, with very high heterogeneity reported for both isolated-dizziness and all-dizziness pooled analyses.

The authors reported higher stroke-prevalence estimates in studies using MRI-based confirmation and in cohorts with complete (100%) imaging coverage, although statistical significance varied by subgroup. Studies using less intensive confirmation approaches generally reported lower estimates. Hospital setting also shaped the range of reported findings, with tertiary centers, teaching hospitals, and stroke centers differing from community sites. These contrasts were described across subgroup analyses rather than as a single uniform pattern for every cohort. Overall, the reported stroke burden shifted with diagnostic intensity and study setting.

The review characterizes bedside diagnostic tools from a limited set of studies rather than as a pooled accuracy analysis. Because only four studies addressed these measures, the synthesis presented the findings qualitatively. HINTS and STANDING were described as having high diagnostic accuracy overall across the included reports. The authors summarize the TriAGe+ score as being applied flexibly for screening versus confirmation based on different cut-off points. ABCD2, by contrast, was reported to have limited diagnostic value in this emergency dizziness setting.

Across analyses, the investigators emphasized substantial between-study variation rather than a single stable estimate. Cohort definition, confirmation strategy, and hospital setting all shaped how often stroke was reported. The compiled evidence therefore portrays emergency department dizziness literature as having setting-dependent stroke estimates.