Crossed Leg Sign Is Associated With Severity of Unilateral Spatial Neglect After Stroke

Background The crossed leg sign in patients with right hemisphere stroke is thought to be associated with perceptual disorders, such as unilateral spatial neglect (USN). The aim of this study was to compare the crossed leg sign with the severity of USN during the acute phase of stroke. Experimental procedures This was an observational and prospective clinical study of individuals with a diagnosis of right parietal stroke, as confirmed by neuroimaging. The occurrence of the crossed leg sign, the time at which this occurred after the stroke, and a clinical diagnosis of USN were measured and recorded. The patients’ age, sex, and lesion severity, as determined by the National Institutes of Health Stroke Scale and Glasgow coma scale, were included in the analyses as confounding variables. The outcome of interest was the degree of USN, as measured by the cancellation and bisection tests. Binary logistic regression was used to analyze the effect of crossed leg syndrome on the severity of USN. In the adjusted multiple regression model, a p-value of <0.05 was considered statistically significant. Results Overall, 60 patients were included in this study. There were no associations between patient demographics and the presence of the crossed leg sign. There was, however, an association between the crossed leg sign and the absolute value of the deviation in the line bisection test (B = −0.234; p = 0.039). The crossed leg sign was not associated with other measures of USN. Conclusion Based on the results of our study, we can conclude that a crossed leg sign in the acute phase of stroke is associated with USN severity, specifically the misinterpretation of the midline.

Background: The crossed leg sign in patients with right hemisphere stroke is thought to be associated with perceptual disorders, such as unilateral spatial neglect (USN). The aim of this study was to compare the crossed leg sign with the severity of USN during the acute phase of stroke. experimental procedures: This was an observational and prospective clinical study of individuals with a diagnosis of right parietal stroke, as confirmed by neuroimaging. The occurrence of the crossed leg sign, the time at which this occurred after the stroke, and a clinical diagnosis of USN were measured and recorded. The patients' age, sex, and lesion severity, as determined by the National Institutes of Health Stroke Scale and Glasgow coma scale, were included in the analyses as confounding variables. The outcome of interest was the degree of USN, as measured by the cancellation and bisection tests. Binary logistic regression was used to analyze the effect of crossed leg syndrome on the severity of USN. In the adjusted multiple regression model, a p-value of <0.05 was considered statistically significant. results: Overall, 60 patients were included in this study. There were no associations between patient demographics and the presence of the crossed leg sign. There was, however, an association between the crossed leg sign and the absolute value of the deviation in the line bisection test (B = −0.234; p = 0.039). The crossed leg sign was not associated with other measures of USN.
conclusion: Based on the results of our study, we can conclude that a crossed leg sign in the acute phase of stroke is associated with USN severity, specifically the misinterpretation of the midline.
Keywords: crossed leg sign, unilateral spatial neglect, hemispatial neglect, stroke, cerebrovascular disease inTrODUcTiOn Unilateral spatial neglect (USN) is a perceptual disorder that is characterized by an inability to respond to people or objects that are presented contralateral to the lesioned side of the brain when these symptoms cannot be attributed to either motor or sensory deficits (1)(2)(3). USN is frequently demonstrated in the clinic as misinterpretation of the midline, which may include head and eye Crossed Leg Sign and USN Severity Frontiers in Neurology | www.frontiersin.org April 2018 | Volume 9 | Article 256 deviations on the side contralateral to hemiplegia as well as the crossed leg sign (4)(5)(6). The crossed leg sign was first described in patients with right hemisphere stroke who presented with USN, including cases in which there were associated changes in consciousness (7). It is impossible to detect USN during coma, but frequent rubbing movements of the right leg over the left observed in the first days of clinical evolution may differentiate between patients with torpor and coma. This sign is characterized by an overlap of the right leg over the left as the patient attempts to orient to the midline because there is a loss of spatial orientation of the left space. If the left leg is not perceived or felt to be one's own limb, then abnormal rubbing movements may appear, which may be of predictive value in the development of USN (7).
Our hypothesis is that patients with the crossed leg sign may have a perceptual disorder that causes severe USN after stroke. The aim of this study was to compare the crossed leg sign with the severity of USN in the acute phase of stroke.

MaTerials anD MeThODs study Design, setting, and Participants
This was a prospective clinical study in individuals with a diag nosis of stroke that had been confirmed by computed tomog raphy or magnetic resonance imaging. Patients hospitalized in the Emergency Room at the Stroke Unit at the Botucatu Medical School in Botucatu, Brazil were included in this study and were followed from January to December 2016. Stroke diag noses were established according to the routine guidelines of the hospital. In the hyperacute phase (up to 8 h following the stroke), the CT scan is performed without contrast and extra and intracranial angiotomography, in addition to perfusion CT. In patients in the acute phase (after 8 h following the stroke), the CT scan is performed without contrast and extra and intracranial angioto mography. In all subjects, unconfined CT is repeated to obtain a control. Contrast magnetic resonance imaging (T1, T2, flair, and diffusion) is only performed to exclude stroke mimics, neuroinfections, and tumors.

inclusion and exclusion criteria
Male and female subjects aged 18-85 years with right parietal ischemic stroke confirmed by a CT scan or MRI in the acute phase were included in this study. Individuals with a prior modi fied Rankin Scale >1, aphasia, preexisting dementia, previous visual changes, associated hemianopsia, mechanical orthopedic changes that impair the movement of the lower limb, or other neurological diseases were excluded.

Exposures
The occurrence of the crossed leg sign, the time at which the sign occurred after the stroke, and a clinical diagnosis of USN were measured and recorded.

Potential Confounding Variables
The main confounding variables that could potentially be associ ated with the outcome included age (older individuals have higher severity of USN after stroke), sex (women have worse outcomes of USN in some studies), severity of stroke according to the National Institutes of Health Stroke Scale (NIHSS) (the severity of stroke was associated with higher degree of USN), and the level of consciousness, as measured by the Glasgow coma scale (GCS) (greater degree of USN is expected in individuals with a lower degree of GCS).

Outcomes
The outcome of interest was the degree of USN, as measured by the cancellation and bisection tests.

USN Tests
The degree of USN in the acute phase of stroke was measured using three tests. The first test was the line cancellation test, which measures the proportion of lines omitted from a total of 40 lines that are randomly distributed on one sheet of paper (8). The second test was the star cancellation test, which measures the proportion of stars omitted from a total of 56 smaller stars that are associated with distractors (9). The final test was the line bisection test (LBT), where "the patient was asked to mark the middle of each of 18 lines arranged in three columns of six (on the right, center, and left of the page); the degree of neglect was measured as the distance (in mm) between the patient's mark and the mid dle of the lines" (10). In all USN tests, the examiner placed the exam sheet in front of the patient at a distance of 60 cm from the glabella to the center of the paper (11,12).

Follow-Up of Crossed Leg Sign
All patients were observed for their crossed leg behavior in the hyperacute phase (in the emergency room), in the acute and subacute phases (in the stroke unit) and at hospital discharge. Crossed leg behavior was defined as having occurred when the patient had the compulsion to cross one leg over the other for two or more consecutive days in three periods of the day. The presence or absence of the sign and the length of time that it remained were recorded. For a confirmed crossed leg sign, the patient must have exhibited ≥3 occurrences of crossed legs on ≥2 consecutive days, with a duration of at least 30 s. The sign was checked every 2 h. Each time, the examiners uncrossed the patient's legs to see if the legs returned to the initial pattern. To establish the time at which the sign disappeared, the absence of the crossed leg sign was considered to be the absence of the compulsion to crossed one leg over the other for two or more consecutive days after the examiners uncrossed the patient's legs.
The patient's severity and topography were blinded for and then followed by five different observers with different levels of clinical experience. Two were consultants (Eduardo de Moura Neto and Hélio Rubens de Carvalho Nunes) and two were neuro logists (Luiz Antônio de Lima Resende and Luiz Eduardo Gomes Garcia Betting). The first author (Gustavo José Luvizutto) and the last author (Rodrigo Bazan) recorded the responses from the observers.
To avoid selection bias, all patients with right hemisp here stroke were followed during hospitalization for the crossed leg sign.

resUlTs
Overall, 150 patients were screened for this study, and 90 were excluded based on the exclusion criteria. Thus, 60 patients were followed and included in the analyses (Figure 1). The demo graphic and clinical characteristics are shown in Table 1. The proportion of agreement in the observations of the crossed leg sign between the authors was 86% (74-91%). Table 2 describes the clinical and demographic differences between the patients who did and did not present the crossed leg sign. There was no statistically significant difference between the groups regarding the clinical and demographic profile, only an absolute value of midline deviation (p = 0.023).
In Figure 2, we show the crossed leg behavior in a patient (Figure 2A) with an objective diagnosis of USN by the LBT (Figure 2B), and the axial apparent diffusion tensor ( Figure 2C) and diffusion tensor imaging tractography ( Figure 2D) show a lesion in the right frontoparietal region. In the LBT, the lines to the left on the paper were not marked, being compatible with the objective diagnosis of USN.
There was no relationship between any of the patient demo graphics and the presence of the crossed leg sign. There was a relationship between the crossed leg sign and the absolute value of the deviation on the LBT (B = −0.234; p = 0.039). There were no associations between the crossed leg sign and the other USN tests ( Table 3).

DiscUssiOn
This study indicates that crossed leg behavior is a factor describ ing the severity of the development of USN after stroke. It has been shown that crosslegged individuals are twice as likely to present right midline deviations in the LBT. This was an   important factor in establishing a causal relationship between leg crossing and the severity of USN and misinterpretation of the midline. Rémi et al. described 34 patients with leg crossing behavior after stroke and reported that the median duration for this behavior was 15 days and was associated with the best functional prognosis. The authors of that study did not, however, correlate this finding with USN (13). No association was found with better outcomes in this study because the patients with USN presented higher NIHSS values, indicating a higher severity of the neuro logical condition.
Several textbooks on semiology and neurological examina tion do not describe the crossed leg sign as a clinical sign in the acute phase of stroke (14)(15)(16). In recent years, however, patients with acute stroke in the right hemisphere and changes in consciousness have been reported to present with this behavior during hospitalization, and this appears to be especially true in patients with USN and misinterpretation of the midline (7).
The authors postulate three main theories for leg crossing behavior in the acute phase of stroke in patients with USN. The first is that perceptual disorders with a misinterpretation of the midline, such as anosognosia and asomatognosia, are mainly associated with USN with parietal syndrome of the right hemisphere (9,17). As an individual's consciousness is altered and he/she perceives a foreign body in the bed, such as his/ her left foot or anything in the left hemisphere, the right foot would be expected to repeatedly and compulsively perform the movement of encountering and rejecting this member. The second theory is that when misinterpreting the midline, these patients are always searching for sensorial stimuli. These stimuli increase extrinsic feedback from sensorial input and allow a better interpretation by the parietal lobe (18,19). The final theory is the interhemispheric inhibition theory, which suggests that in the acute phase of stroke, a patient with hemiplegia associated with USN will use the compromised lower limb less frequently, which results in less information being transmitted to the injured parietal cortex (20,21). As the injured hemisphere receives less information, it will no longer inhibit the uninjured hemisphere, leading to a hyperactivation of the contralesional hemisphere and an increase in motor activity on the unaffected side.
Based on the results presented, we conclude that the crossed leg sign in the acute phase of stroke is associated with severity of USN, specifically in the misinterpretation of the midline.