Discussion

In our study, unilateral acute left-sided hemiparesis and/or hemisensory symptoms, compared with right-sided symptoms, were more frequently associated with negative DWI in a population of suspected minor strokes, showing an increased probability of approximately one-third. These findings were mainly driven by the presence of motor symptoms and were consistent across different ages and sexes.

In our study sample, the proportion of patients presenting with acute left-sided symptoms was slightly higher than those with right-sided symptoms (∼53% vs. ∼47%). While previous studies have shown a higher overall prevalence of right-sided symptoms among stroke patients, ^{Reference Foerch, Misselwitz, Sitzer, Berger, Steinmetz and Neumann-Haefelin16,Reference Portegies, Selwaness, Hofman, Koudstaal, Vernooij and Ikram17} an opposite trend has been observed in the distinctive subgroup of minor strokes. ^{Reference Mendelson and Prabhakaran18,Reference Coutts, Ankolekar and Appireddy19} These findings can be explained by the higher clinical eloquence of the left hemisphere, which facilitates diagnosis (high prevalence overall) but, at the same time, translates into greater clinical severity (low prevalence in the minor stroke subgroup). ^{Reference Mendelson and Prabhakaran18,Reference Menezes, Ay and Wang Zhu20,Reference Güntürkün, Ströckens and Ocklenburg21} However, in our study, the proportion of patients with positive DWI was equally balanced (∼50%), in contrast with the proportion with negative DWI (∼60% vs. ∼40%). This supports the concept that the difference we observed in symptoms laterality was mainly driven by stroke mimics rather than true acute strokes.

The underpinning biology of the laterality preference in DWI-negative patients remains uncertain. This group includes stroke mimics and possibly a small representation of true acute strokes without lesions (TIAs). ^{Reference Tanaka, Coutts and Joundi22} Among stroke mimics, there is little to no evidence for a laterality preference. The higher clinical eloquence and greater metabolic demand of the left hemisphere might suggest a lower resilience and a lower clinical threshold for detecting mimics in the left hemisphere. For instance, hypoglycemia-induced hemiparesis is twice as common with right-sided symptoms. ^{Reference Yoshino, Meguro, Soeda, Itoh, Kawai and Itoh23} In other words, the minimal evidence available supports a preference for right symptoms in stroke mimics rather than left symptoms. Conversely, functional disorders have been consistently associated with a preference for left-sided symptoms. ^{Reference Stern2,Reference Stone, Sharpe and Carson3}

There are several potential explanations for this latter phenomenon, which are not mutually exclusive. First, left-sided symptoms represent one of the heralding presentations of acute coronary syndromes due to the distinctive sensory innervation of the heart. ^{Reference Eslick24} This might result in an overrepresentation of left upper arm symptoms in anxious individuals who fear they are experiencing a myocardial infarction. Second, there might be a functional advantage in presenting symptoms on the non-dominant side, which is the left side for the vast majority of individuals. ^{Reference Stone, Sharpe and Carson3} Third, differences in attention or perception of disability in the dominant side might have influenced the outcome. Finally, the right hemisphere is believed to play a more significant role in the mediation of affective or motivationally determined somatic symptoms and has a different spatial representation of the body in comparison to the left hemisphere. ^{Reference Stern2,Reference Stone, Sharpe and Carson3}

In our cohort, in the subgroup analyses, the left-sided preference in patients with negative DWI was observed only in those with motor symptoms, but not in those with sensory symptoms. The underlying reasons for this observation remain unclear, yet it is possible that while sensory symptoms are more commonly associated with functional disorders, ^{Reference Hallett, Aybek, Dworetzky, McWhirter, Staab and Stone1} the different body laterality might be more pronounced in motor symptoms. Moreover, the smaller number of patients presenting with sensory symptoms may have limited the ability to detect a statistically significant difference.

Our findings should be interpreted as exploratory, and several limitations must be acknowledged. First, DWI negativity does not fully exclude a vascular etiology, ^{Reference Tanaka, Coutts and Joundi22} and second, we did not evaluate for FNDs. Current diagnostic criteria for FND emphasize the need for positive rule-in signs on neurological exam (e.g., incongruence with known neurological disease, inconsistency and variability, improvement with distraction or automatic movements) and not simply normal imaging. ^{Reference Aybek and Perez14} Moreover, FND can co-occur with structural disease, including stroke, in a minority of cases. ^{Reference Popkirov, Stone and Buchan25} Finally, a variety of alternative diagnoses to functional and vascular disorders are also possible, including neurological (e.g., spinal cord pathologies and neuropathies) or non-neurological (e.g., acute coronary syndrome) conditions. Third, hand dominance or a comprehensive assessment of hemisphere dominance was not collected. Fourth, ethnic differences might influence the clinical presentation of functional disorders. Fifth, multiple subgroup analyses might have increased the risk of type I error in our study. Finally, we defined suspected minor stroke on the persistence of neurological symptoms at the initial medical examination. This pragmatic definition aimed to reduce the proportion of ischemic strokes with negative DWI. However, there was no restriction on the minimum duration from symptom onset, so some TIA patients were likely included.

Notably, our study aims to explain the laterality preference of DWI-negative patients, not to inform diagnostic or therapeutic decision-making in the emergency department.