Major complications associated with unfavorable outcome in right‐sided large hemisphere infarctions: A single‐center study

Abstract Objectives To identify the major complications independently associated with unfavorable outcomes in right‐sided large hemisphere infarction (RLHI) patients. Methods We retrospectively enrolled consecutive patients admitted within 24 h with the diagnosis of RLHI. The unfavorable outcome was defined as a modified Rankin Scale score of 4–6 at 3 months. Univariate and multivariate analyses were performed to identify the major complications independently associated with 3‐month unfavorable outcomes. Results Of the 171 cases with RLHI included, 126 (73.7%) had unfavorable outcomes at 3 months: A total of 64 (37.4%) cases died, and 62 (36.3%) lived with severe disability. Stroke‐related complications occurred in 168 (98.2%) patients during hospitalization. The five most common stroke‐related complications were pulmonary infection (75.4%), electrolyte disorder (61.4%), hypoalbuminemia (49.1%), malignant brain edema (MBE) (48.5%), and hemorrhagic transformation (48.0%). RLHI patients with unfavorable outcomes had more frequent MBE (58.7% vs. 21.4%, p < .001), pulmonary infection (86.5% vs. 42.9%, p < .001), gastrointestinal bleeding (46.8% vs. 28.6%, p = .038), electrolyte disorder (68.3% vs. 40.5%, p = .001), acute renal failure (32.5% vs. 4.8%, p < .001), and hypoalbuminemia (61.1% vs. 11.9%, p < .001) than patients with favorable outcome. Multivariate analyses suggested that only MBE (adjusted OR 4.06, 95% confidence interval [CI] 1.14–14.48, p = .031), pulmonary infection (adjusted OR 4.69, 95%CI 1.48–14.85, p = .009), and hypoalbuminemia (adjusted OR 6.58, 95%CI 1.74–24.86, p = .005) were independently associated with 3‐month unfavorable outcome in patients with RLHI. Conclusions Most of the RLHI patients have at least one stroke‐related complication during hospitalization, and nearly three‐quarters suffered unfavorable outcomes. Only MBE, pulmonary infection, and hypoalbuminemia are independently associated with 3‐month unfavorable outcome.


INTRODUCTION
Large hemispheric infarction is one of the most devastating conditions with high mortality and disability rate among acute ischemic stroke (AIS) patients (Huttner & Schwab, 2009;Uhl et al., 2004). The laterality of the affected hemisphere is an important determinant of the natural history of large hemisphere infarction (LHI) (Rastogi et al., 2015). The right hemisphere is often considered nondominant hemisphere, so that right hemisphere stroke may be less clinically evident, and the stroke severities assessed by the National Institutes of Health Stroke Scale (NIHSS) score are also weighted toward left hemisphere lesions (Fink et al., 2002). As a result, right hemisphere stroke can be deemed to be less severe than left-sided, such that physicians and surgeons might be less aggressive in treatment (Fink et al., 2002).
Although it is reasonable to suspect that right hemisphere stroke can drive more favorable outcomes, strokes in the right hemisphere can cause more autonomic and cardiovascular dysfunction that affect mortality than left hemisphere stroke (Krause et al., 2017;Naver et al., 1996;Tokgözoglu et al., 1999;Yoon et al., 1997). Additionally, right hemisphere strokes might lead to lower rehabilitation potential due to visuospatial neglect or emotional indifference (Aszalós et al., 2002;Ween et al., 1996). We recently reported more than two thirds of rightsided LHIs (RLHIs) suffered unfavorable outcomes and suggested that stroke lateralization was not an independent predictor of death and unfavorable outcomes in patients with LHIs (Li et al., 2021), which was in line with previous published systematic review and meta-analysis (Almekhlafi et al., 2019;Rastogi et al., 2015). Despite this evidence, right hemisphere stroke has usually been considered to be less severe and has a higher risk of being excluded from clinical trials or established treatments (Fink et al., 2002).
It has been demonstrated that stroke-related complications have a great influence on the death and unfavorable outcomes of AIS (Johnston et al., 1998;Weimar et al., 2002). Meanwhile, neurological impairment level is the most substantial factor in predicting the rate of complications (Roth et al., 2001). It is reasonable to suspect that strokerelated complications might less frequently occur in right hemisphere stroke than in left hemisphere stroke. However, our previous study suggested that RLHI was independently associated with an increased risk of malignant brain edema (MBE) and a composite of cardiovascular events during hospitalization (Li et al., 2021). Several studies have also reported a higher incidence of pneumonia (Kemmling et al., 2013), and a higher risk of hemorrhagic transformation following intravenous thrombolysis in right hemisphere stroke patients (Audebert et al., 2011). Physicians need to be aware of the incidence of common stroke-related complications of RLHI and their importance to patient outcome. Although our previous study has assessed the hemispheric differences in clinical characteristics, the incidence rate of stroke-related complications, and outcomes of large hemisphere infraction patients, until recently, which complication might play a significant role in the poor outcomes of RLHI patients remains unclear.
Therefore, we conducted a retrospective cohort study using the prospectively acquired data of the Deyang stroke registry, to identify the major acute complications independently associated with 3-month unfavorable outcomes in RLHI patients.

Study design and subjects
AIS patients who were admitted to the People's Hospital of Deyang City within 24 h from symptoms onset were prospectively and consecutively registered from January 1, 2016 to March 31, 2019. We enrolled RLHI patients with computed tomography (CT) and/or magnetic resonance imaging (MRI) evidence of supratentorial cerebral infarction involving more than 50% of the right middle cerebral artery (MCA) region within 7 days after stroke onset, no matter the involvement of the right anterior cerebral artery (ACA) or posterior cerebral artery (PCA) (Uhl et al., 2004). Bilateral hemispheric strokes were excluded. All enrolled patients had a noncontract CT (NCCT) before initial treatment. A routine follow-up NCCT or MRI was performed during the first 7 days after stroke onset. Other CT scans were performed when suffering neurological deterioration, to identify brain edema or hemorrhagic transformation. We excluded cases with incomplete hospital records or missing imaging that would prevent complete data collection. We also excluded cases with a preexisting score of more than 2 on the modified Rankin scale (mRS) and lived dependently (de Haan et al., 1995). The mRS score is a scale of 0-6, which categorized the magnitude of neurologic disability into one of the seven exclusive categories (de Haan et al., 1995): 0 (no symptoms), 1 (no significant disability), 2 (slight disability), 3 (moderate disability), 4 (moderately severe disability), 5 (severe disability), 6 (death

Data collection and outcome
Baseline data on age, gender, admission delay, baseline NIHSS score, baseline systolic and diastolic blood pressure, serum glucose on admission, and vascular risk factors were collected, which has been described in our previous study (Li et al., 2018). The potential etiology of LHI was classified as cardio-embolism or not according to the Trial of Org 10172 in Acute Stroke Treatment (TOAST) criteria (Adams et al., 1993).
Two experienced neurologists who were blinded to clinical information independently evaluated the brain imaging. Disagreement was resolved through discussion or further consulting with a third neurologist. The hyperdense MCA sign (HDMCAS) and Alberta Stroke Program Early CT Score (ASPECTS) were assessed on the pretreatment NCCT (Pexman et al., 2001;Topcuoglu et al., 2014 (Hacke et al., 1998). MBE was defined as the development of clinical signs of herniation (including decrease in consciousness and/or anisocoria), accompanied by a midline shift of ≥5 mm at the septum pellucidum or pineal gland with effacement of the basal cisterns on follow-up imaging (Kimberly et al., 2018), excluding patients with parenchymal hematoma (defined as hemorrhage of brain parenchyma with mass effect according to the ECASS II criteria (Hacke et al., 1998)).
In-hospital treatments (defined as acute treatment administered during hospitalization) analyzed in our study included thrombolysis, endovascular interventions, decompressive hemicraniectomy (DHC), mechanical ventilation, osmotic agents (such as mannitol), antiplatelet agents, and statins. Thrombolysis or endovascular interventions were performed according to the Chinese guidelines, which had the similar inclusion and exclusion criteria compared with the American guideline (Jauch et al., 2013;Powers et al., 2015).  (Li et al., , 2021. Composite of cardiovascular events in our study was defined as a composite of myocardial infarction, acute heart failure, or any sudden cardiac death (Park et al., 2020). Hypoalbuminemia in the present study was defined as serum albumin level <35 g/L during hospitalization (Dziedzic et al., 2006). Independent outcome assessments were performed by researchers who were blind to the clinical information, at 3 months after stroke by telephone interview or by mail. The primary outcome measures in our study were 3-month unfavorable outcome (defined as an mRS score of 4-6, including death or living with a severe disability (de Haan et al., 1995)).

Statistical analyses
Baseline characteristics, in-hospital treatment, and stroke-related complications were compared between RLHI patients with 3-month unfavorable and favorable outcomes. The χ (Huttner & Schwab, 2009) tests or Fisher's exact tests were used for intergroup differences in categorical variables, whereas Student's t-tests or the Mann-Whitney U test were used for intergroup differences in continuous variables.
Univariate analysis was performed to test variables that might affect the outcomes. The included variables were (1) age, (2) baseline NIHSS score, (3) vascular risk factors, (4) imaging characteristics, (5) inhospital treatment, and (6) stroke-related complications. Multivariate analysis was performed by using the forced entry method, including variables with p < .1 in univariate analyses, to identify the factors independently associated with 3-month unfavorable outcome in RLHI patients (excluding stroke-related complications). Then, we identified the major complications via adjusting for confounders, which had a significant association with 3-month unfavorable outcome in multivariate analysis. The 95% confidence intervals (CIs) were calculated to describe the precision of the estimates. All statistical analyses were performed using SPSS v21.0 (IBM, Chicago, IL, USA). Two-sided p value <.05 was considered to be statistically significant.

RESULTS
During the study period, 3551 AIS patients were consecutively regis- (1.7%) cases were lost to follow-up at 90 days. Baseline characteristics, in-hospital treatment, and clinical outcomes of patients with RLHI have been described in detail in our previous published study (Li et al., 2021).
There was no difference in the gender, median admission delay, diastolic blood pressure, and other vascular risk factors between the two groups (all p > .05). When stroke etiology of RLHI is concerned, patients with unfavorable outcome showed a higher proportion of cardio-embolism (77.8% vs. 61.9%, p = .043) than the favorable outcome group. The unfavorable outcome group had a lower ASPECTS (IQR 3-8 vs. 5-9, p = .033) and a higher proportion of infarction involving the ACA territory (MCA + ACA) (20.6% vs. 4.8%, p = .019); however, the presence of HDMCAS and basal ganglia involvement were com-parable between the two groups (both p > .05). For the in-hospital treatments of RLHI, patients with unfavorable outcomes more frequently received mechanical ventilation (43.7% vs. 11.9%, p < .001) and osmotic agents (99.2% vs. 90.5%, p = .018), nevertheless, less frequently received antiplatelets (64.3% vs. 81.0%, p < .001) and statins (52.4% vs. 76.2%, p = .007) in the acute phase of stroke. There was no difference in the administration rates of thrombolysis, endovascular interventions, and DHC (all p > .05).

Factors associated with 3-month unfavorable outcome in RLHI patients (excluding stroke-related complications)
Variables that may have potential effects on the 3-month unfavorable outcome in univariate analysis (p < .1) were included in multivariate logistic regression, and the results are shown in Table 3
The incidence rate of stroke-related complications in the present study was similar to those previous findings. It has been demonstrated that poststroke neurological complications mainly occur in the early course of neurological deterioration-within 48-72 h of stroke onset (Balami et al., 2011), whereas medical complications usually develop in the acute or subacute phase of stroke (within the first few weeks after stroke) (Kumar et al., 2010). In addition, the greater neurological deficit assessed by the NIHSS score is related to the higher incidence rate of stroke-related complications (Roth et al., 2001).

TA B L E 2
Stroke-related complication during hospitalization of right-sided large hemisphere infarction (RLHI) patients with 3-month unfavorable and favorable outcomes. It is known that stroke-related complications have a great influence on the death and functional outcome of AIS patients (Johnston et al., 1998;Weimar et al., 2002). Although it is common for RLHI patients to experience multiple medical and neurological complications (Audebert et al., 2011;Kemmling et al., 2013;Li et al., 2021) (Rahme et al., 2012). Furthermore, because of its invasive nature and the need for multidisciplinary cooperation, DHC remains underutilized worldwide (Bar et al., 2011;Neugebauer et al., 2014). In our cohort, 48.5% of RLHI patients developed MBE, but only 14.0% underwent DHC, which was similar to the surgery rates reported in another real-world observational study conducted in China (Hao et al., 2015). Even though RLHI patients have lower baseline NIHSS scores compared with left hemisphere strokes of equivalent volume, previous studies and systematic reviews have demonstrated that MBE appears to be more common in the right hemisphere infarction (Li et al., 2021;Rastogi et al., 2015). Further study is warranted to determine the early warning signs of MBE in patients with right hemisphere stroke, especially for those with lower NIHSS scores.

Unfavorable (mRS score 4-6) (n = 126) Favorable (mRS score 0-3) (n = 42) p Value
The most common complication in our cohort was pulmonary infection, with an incidence rate of 75.4%, which was similar to our previous study . The incidence rate of pulmonary infection in our cohort is significantly higher than that of several studies conducted in stroke patients Smith et al., 2015;Teh et al., 2018;Yu et al., 2016). The higher incidence of pulmonary infection could be explained by immunodepression, impaired TA B L E 3 Multivariate analysis for factors associated with 3-month unfavorable outcome in right-sided large hemisphere infarction (RLHI) patients (excluding stroke-related complications).

Variables Unfavorable outcomes OR (95%CI) p Value
Age ( Note: Variables that had a potential association with 3-month unfavorable outcome in univariate analysis (p < .1) were listed. p Values of <.05 are shown in bold.
swallowing mechanism, impaired level of consciousness, and bedridden due to severe brain injury caused by RLHI (Yu et al., 2016). Several studies have suggested that pulmonary infection is closely associated with the prolonged length of hospital stay, increased medical cost, and poor functional outcomes in stroke patients (Hannawi et al., 2013;Teh et al., 2018;Yu et al., 2016). In the present study, after adjusting for age, stroke severity, and other confounders, the risk of 3-month unfavorable outcome in RLHI patients complicated with pulmonary infection was 4.69 times higher than those without pulmonary infection, which was in line with previous studies (Hannawi et al., 2013;Teh et al., 2018;Yu et al., 2016). An observational study found that poststroke pneumonia is closely associated with the development of several nonpneumonia complications, and pneumonia might be a risk marker for the development of several non-pneumonia complications after stroke . Because most pulmonary infection is potentially preventable or treatable, we should pay more attention to the prevention and control of pulmonary infection in RLHI patients, because of the higher incidence rate and concomitant unfavorable outcome.
The study found that the incidence rate of hypoproteinemia in RLHI patients was as high as 49.1%, which was similar to the results of previous studies (Dziedzic et al., 2007;Vahedi et al., 2011). Multivariate analysis suggested that the risk of 3-month unfavorable outcome in RLHI patients complicated with hypoproteinemia during hospital-ization was 6.58 times higher than those without hypoproteinemia.
Serum albumin, which is the most abundant protein in human blood plasma, is synthesized in the liver. It has many physiological functions, such as regulating colloid osmotic pressure, transporting endogenous and exogenous ligands, anti-inflammation, anti-oxidation, and antiapoptosis (Boldt, 2010). Experimental studies have demonstrated that albumin therapy had a variety of neuroprotective effects, such as improving neurological function and reducing the volume of cerebral infarction and brain edema in animals with AIS (Belayev et al., 2001(Belayev et al., , 2005. Observational studies suggested that higher serum albumin levels measured within 36 h after stroke onset were associated with a reduced risk of in-hospital death and 3-month unfavorable outcomes (Dziedzic et al., 2004;Vahedi et al., 2011). It is also reported that lower baseline serum albumin levels in AIS patients was associated with a higher risk of hemorrhagic transformation (Che et al., 2017;Wang et al., 2019), and higher incidence of pulmonary infection in observational studies (Dziedzic et al., 2006;Yang et al., 2020). However, the ALIAS (albumin in AIS) part 1 and 2 trials and the combined data analyses showed that intravenous infusion of 25% human serum albumin (2 g/kg) was not associated with improved outcome at 90 days and was associated with the increased incidence of intracerebral hemorrhage and pulmonary edema (Martin et al., 2016). Theoretically, the neuroprotective effects of albumin such as increasing colloidal osmotic TA B L E 4 Multivariate analysis to identify the major complications associated with 3-month unfavorable outcome in right-sided large hemisphere infarction (RLHI) patients. pressure to reduce brain edema, as well as anti-inflammation, antioxidation, and anti-apoptosis might be beneficial to LHI, but there is still a lack of evidence of clinical trials for the treatment of human serum albumin in patients with LHIs. It is worth noting that RCTs conducted in intensive care unit patients have demonstrated the safety of resuscitation with 20% albumin and its effectiveness in the treatment of patients with severe sepsis or septic shock (Caironi et al., 2014;Mårtensson et al., 2018). Further study is warranted to determine whether 20% albumin is safe and effective in improving the clinical outcome of patients with LHIs. In our study, hypoalbuminemia was defined as serum albumin level <35 g/L during hospitalization. So early nutritional assessment and nutritional support, monitoring of dynamic changes of serum albumin levels, and maintaining serum albumin level ≥35 g/L might be appropriate in improving the clinical outcomes in RLHI patients.

Variables
It is worth noting that LHI patients who were current smokers had more favorable outcomes than nonsmokers. A systematic review also found that a history of smoking is one of the protective factors for MBE in patients with AIS (Miao et al., 2020). The protective mechanism of smoking on LHI remains unclear, and it is speculated that nicotine, one of the main components of cigarettes, might have a neuroprotective effect on ischemic brain injury (Pacher & Haskó, 2008). Previous experimental studies have demonstrated that the activation of the endocannabinoid system may play an important role in the neuroprotective effects of nicotine, which increases the release of endocannabinoid and leads to lower body temperature, thereby inhibiting the inflammation and reducing brain edema (Chen et al., 2013;Panikashvili et al., 2001).

Limitations
The results of the present study should be interpreted with caution given its limitations. whereas medical complications usually develop within the first 1-2 weeks after stroke (Balami et al., 2011;Kumar et al., 2010). As we included RLHI patients admitted within 24 h from symptoms onset with a median length of hospital stay of 14 days, the occurrence of most acute complications could be observed during hospitalization.
Fourth, in the present study, hemorrhagic transformation was classified into two types according to the ECASS II criteria. However, it might be more useful to report clinically relevant cerebral hemorrhage and classify the hemorrhagic transformation in more detail according to the SITS-MOST criteria (Rao et al., 2014). Besides, because of the poor adherence to long-term follow-up in RLHI patients, we only conduct a 3-month follow-up so that the long-term effect of the major complications remains unclear. Moreover, we performed the followup by telephone interview or a mailed questionnaire instead of a clinic visit, which may result in reporting bias. Finally, the sample size of our study was relatively small, and the favorable outcome group had only 42 cases. Well-designed studies with large sample sizes are needed to construct and validate a new prognostic predictive model for RLHI patients.

CONCLUSIONS
Most of the RLHI patients have at least one stroke-related complication during hospitalization, and nearly three-quarters suffered unfavorable outcomes. Only MBE, pulmonary infection, and hypoalbuminemia are independently associated with 3-month unfavorable outcome. Physicians need to pay more attention to early prevention, identification, and treatment of the major complications to improve the clinical outcomes of patients with RLHI.

AUTHOR CONTRIBUTIONS
Jie Li and Ping Zhang conceived and designed the study, acquired the funding, collected, analyzed, and interpreted the data, as well as drafted the manuscript. Jie Li, Ping Zhang, Xingrong Luo, and Ju Zhou participated in study administration, investigation, and data collection. Jie Li, Hong Chen, and Yingying Liu contributed to study design, funding acquisition, and study administration. Hong Chen, Yingying Liu, and Chun Wang contributed to study design, administration, and supervision. All authors critically revised the manuscript for important intellectual content and approved the final manuscript.

CONFLICT OF INTEREST STATEMENT
The authors declare that there is no conflict of interest.

DATA AVAILABILITY STATEMENT
The data and materials that support the findings of this study are available from the corresponding author upon reasonable request.