Association between encephalopathy and clinical outcomes of COVID-19: Findings from the Philippine CORONA Study

Introduction: This study aimed to determine whether encephalopathy is associated with such COVID-19 outcomes as disease severity, mortality, respiratory failure, intensive care unit (ICU) admission, duration of ventilator dependence, and length of ICU and hospital stay. Methods: We performed a subgroup analysis comparing outcomes in patients with and without encephalopathy, based on data from a nationwide retrospective cohort study among adult patients hospitalized with COVID-19 at 37 hospital sites in the Philippines. The patient outcomes included for analysis were disease severity, mortality, respiratory failure, ICU admission, duration of ventilator dependence, and length of ICU and hospital stay


Introduction
][6] Encephalopathy is a diffuse brain dysfunction characterized by impaired level of consciousness, which can present as delirium, stupor, or coma. 7The most common causes of encephalopathy among patients with COVID-19 were sepsisassociated encephalopathy, uremic encephalopathy, and hypoxic ischemic encephalopathy. 8The underlying causes of encephalopathy may be attributed to multiple mechanisms, such as neurotropism of the SARS-CoV-2 virus, severe inflammatory response leading to multiple organ dysfunction or acute respiratory distress syndrome, coagulopathy, and hypoxia. 9,102][13] However, data are still limited on the association between encephalopathy and other outcome measures among patients with COVID-19.
In the Philippines, the recently concluded nationwide, multicenter, retrospective study of neurological manifestations and associated outcomes (the Philippine CORONA Study) identified encephalopathy as the most common new-onset neurological symptom associated with COVID-19. 3The present study aims to further elucidate whether encephalopathy is associated with COVID-19 outcomes including disease severity, mortality, respiratory failure, intensive care unit (ICU) admission, duration of ventilator dependence, and duration of ICU and hospital stays.

The Philippine CORONA Study design: population and setting
The Philippine CORONA Study was a nationwide, multicenter, comparative, retrospective, cohort study including patients with COVID-19 who were admitted consecutively to 37 hospital sites across the Philippines between February and December 2020. 3All data were collected from patient medical charts and coded using the Epi Info Software (version 7.2.2.16).Encephalopathy is defined and widely accepted as a disease in which brain function is altered, often with depressed or fluctuating sensorium.Due to the retrospective nature of the study, any physician assessment stating ''encephalopathy'' on the chart was counted as a case.The study obtained approval from the respective research ethics boards of the participating hospital sites and was registered on ClinicalTrials.gov(NCT04386083).Specific details about the research design, patient enrollment, and data collection procedures were discussed in the published protocol. 14We included all eligible participants from the main study and classified the patients into 2 cohort groups: those with encephalopathy and those without encephalopathy.

Outcome variables
The main outcomes were disease severity (mild: presence of mild pneumonia or absence of pneumonia; severe: presence of dyspnea, respiratory rate > 30, hypoxia or SpO 2 < 93%, or 50% lung involvement on imaging within 24-48 h), mortality, respiratory failure (patients with signs or symptoms of respiratory insufficiency: increased work of breathing/respiratory rate of ≥22, a need for supplemental oxygen, or abnormal blood gases), ICU admission, duration of ventilator dependence, duration of ICU stay, and duration of hospital stay, among patients with encephalopathy compared to the no-encephalopathy group.

Statistical analysis
The different variables of interest were summarized according to the independent variable: encephalopathy versus no encephalopathy.Numerical variables are expressed as median and interquartile range (IQR), while the categorical variables are expressed as count and percentage.Comparison of the different clinical characteristics between the 2 groups was conducted using the Mann-Whitney U test for numerical variables, to determine differences in median/ranks, and the Chi-square test of homogeneity for categorical variables, to determine differences in proportion.
Associations between the different individual dichotomous outcome variables and the independent variable were determined by binary logistic regression.The logistic regression model was adjusted for the following confounders: age, sex, hypertension, diabetes mellitus, smoking, coronary artery disease, cerebrovascular disease, dementia, epilepsy, and COVID-19 severity.The associations between the independent variable and the time to event for different outcome variables were determined by Cox proportional hazards regression.A stratified analysis of mortality between COVID-19 severity types was also performed.All statistical analyses were conducted using Stata version 17.1.

Study participants
A total of 10 999 patients consecutively admitted due to confirmed COVID-19 from the 37 participating hospital sites were gathered, with 10 881 patients included in the final analysis.The prevalence of encephalopathy among patients with COVID-19 was 5.72% (95% confidence interval [CI], 5. 29-6.17).The encephalopathy group comprised 622 patients, whereas the no-encephalopathy group comprised 10 259 patients (Fig. 1).Among patients aged 60 years or older, the prevalence of encephalopathy was higher, at 9.59% (368 of 3834 patients).

COVID-19 outcomes
Table 2 summarizes the comparison of outcomes between the encephalopathy group and the no-encephalopathy group.Overall, mortality, respiratory failure, and number of ICU admissions were higher in the encephalopathy group compared to the no-encephalopathy group (P < .001).Table 3 shows the association between encephalopathy and the out- comes of interest.Fig. 2 shows the Kaplan-Meier curves for the encephalopathy and no-encephalopathy groups for mortality, respiratory failure, and ICU admission.

Mortality and associated causes
The mortality rate was higher in the encephalopathy group (62.4%) than in the no-encephalopathy group (12.8%,P < .001).The risk of mortality was 1.72 times greater (95% CI, 1.53-1.94) in the encephalopathy group.Acute respiratory distress syndrome and septic shock were the most common causes of mortality.On stratified analysis, among patients with mild COVID-19, those with encephalopathy presented 9.26 times (95% CI, 5.75-14.92)greater mortality risk than those without encephalopathy.Among patients with severe/critical COVID-19, those with encephalopathy presented a risk of mortality 1.63 times (95% CI, 1.45-1.84)higher than those without encephalopathy.As seen in the Kaplan-Meier curve (Fig. 2), patients with COVID-19 and presenting encephalopathy had shorter survival times than those without encephalopathy (log-rank test, P < .001).

Respiratory failure and duration of ventilator dependence
There was a higher frequency of respiratory failure in the encephalopathy group (69.8% vs 11.4%, P < .001),with an adjusted hazard ratio (HR) of 5.4 (95% CI, 4.8-6.07).Patients with COVID-19 and encephalopathy presented a shorter time to respiratory failure than those without encephalopathy (log-rank test, P < .001)(Fig. 2).Meanwhile, encephalopathy was not found to be associated with

ICU admissions and duration of ICU and hospital stays
Patients with encephalopathy presented 4.26 times (95% CI, 3.77-4.80)greater risk of ICU admission.Among patients with encephalopathy, 63.2% were admitted to the ICU, compared to only 13.3% among the no-encephalopathy group (P < .001).Based on the Kaplan-Meier curve (Fig. 2), patients with COVID-19 and encephalopathy presented a shorter time to ICU admission compared to those without encephalopathy (log-rank test, P < .001).Those with encephalopathy tended to have longer hospital stays compared to the no-encephalopathy group, with 1.36 times (95% CI, 1.14-1.61)higher odds of staying 15 days or more in hospital.
However, there was no difference in the length of ICU stay between the encephalopathy and no-encephalopathy groups (P = .522).

Discussion
The results of the Philippine CORONA Study identified encephalopathy as the most common new-onset neurological disorder among patients with COVID-19. 3Overall, our findings suggested that encephalopathy was associated with COVID-19 severity, mortality, respiratory failure, ICU admission, and longer hospital stays; this is consistent with findings of previous studies. 8,15,16][18][19] The 5.7% prevalence of encephalopathy in our cohort is comparable to the rates reported in the recent prospective, multicenter study in New York (6.8%) 20 and in the TriNetX COVID-19 Research Network database consisting of 12 601 hospitalized patients (8.7%). 21Other studies with smaller sample sizes reported rates of encephalopathy and/or delirium of up to 32-33%. 15,16In the ICU setting, a prospective study of 2 ICU cohorts comprising 140 patients from France reported a 79.5% incidence rate of delirium diagnosed with the Confusion Assessment Method-ICU score. 17These discrepancies in the rate of occurrence of encephalopathy may be due to differences in study design, sample size and demographic characteristics, ICU or special unit admission, and inclusion criteria.For instance, our cohort presented a high proportion of mild COVID-19 cases, which may explain the lower prevalence rate of encephalopathy.
Encephalopathy and delirium in patients with COVID-19 may be caused by a myriad of factors, such as metabolic, respiratory, and coagulation alterations caused either by the direct effects of SARS-CoV-2 on the brain, remote effects, or systemic inflammation affecting the blood-brain barrier. 22lder age, male sex, presence of comorbidities (hypertension, diabetes mellitus, kidney disease, coronary artery disease, malignancy, heart failure), smoking, and previous neurological history (cerebrovascular disease, dementia, epilepsy, movement disorders, CNS infection) were more frequent in the encephalopathy group compared to the noencephalopathy group.These findings are congruent with those of 2 previous large-scale studies of COVID-19 patient cohorts. 8,15In the general acute medical setting, older age, male sex, and presence of cognitive disorder are among the known risk factors for encephalopathy. 23The higher incidence of encephalopathy among elderly patients and those with history of neurological disorders may be explained by decreased cognitive reserve, making the brain vulnerable to such insults as hypoxia and other metabolic alterations. 24oreover, the majority of our patients in the encephalopathy group were elderly.Polypharmacy is also common in elderly people and can easily predispose to delirium, a feature of encephalopathy. 23Furthermore, studies in elderly patients with dementia found that delirium is a common initial presentation of COVID-19 in this population, and is associated with a higher mortality rate. 18,25ur findings showed that the adjusted HR for mortality among patients with encephalopathy was higher among patients with mild COVID-19 than among those with severe/critical COVID-19.7][28] The increased risk of mortality among patients with mild COVID-19 and encephalopathy suggests that encephalopathy may play a role as a prognostic marker.

ARTICLE IN PRESS
Some results have shown that patients with encephalopathy present significantly higher rates of ICU admission than those without encephalopathy.ICU admissions, along with presentation of stupor or coma, have been shown in turn to be predictors of worse outcome at discharge, 29 with patients aged > 65 years having a markedly increased risk for persistent and new sequelae of various clinical conditions, including chronic respiratory failure, cardiac rhythm disorders, acute coronary syndrome, stroke, and kidney injury. 30hus, further studies on the follow up of elderly patients with encephalopathy may be prudent to explore long-term complications of COVID-19.
In the current study, no association was observed between length of ICU stay and encephalopathy.Several factors influence length of ICU stay, such as social and institutional factors and the presence of a specialized team at the ICU 31 ; as these factors were not taken into consideration during our data collection, the lack of association cannot be fully explained by our study.
The limitations of the present study lie in the retrospective nature of the research, due to the inherent risks of recording bias.The onset of encephalopathy, its duration, and correlation with other events such as sepsis, strokes, seizures, and the notorious cytokine storm would likewise clarify the underlying mechanism better.No additional data on cerebrospinal fluid were recorded.Nevertheless, this was the first large-scale nationwide study of COVID-19 in the Philippines, providing valuable information on the incidence of encephalopathy and its associations with clinical outcomes.A prospective study to reduce recording bias may provide more accurate data and remove confounding variables affecting the outcomes of COVID-19 patients with encephalopathy.

Conclusion
Our study showed that encephalopathy was associated with COVID-19 severity, mortality, respiratory failure, ICU admission, and longer hospital stays.Further studies to determine specific risk factors for the development of encephalopathy in COVID-19 may help in improving the outcomes of patients admitted to hospital with COVID-19 in our setting.

Informed consent statement
Informed consent was not necessary since the study design was a retrospective cohort study employing medical chart review.All data obtained were completely anonymized.

Submission declaration and verification
This manuscript has never been published in part and is not under consideration for publication elsewhere.All authors have read and agreed to the published version of the manuscript.If accepted, this manuscript will not be published elsewhere in the same form, in English or in any other language, including electronically, without the consent of the copyright holder.

Figure 1
Figure 1 Flow diagram of patients included in the study.ICU: intensive care unit.

Figure 2
Figure 2 Kaplan-Meier curves for mortality, respiratory failure, and ICU admission.ICU: intensive care unit.

Table 1
Comparison of the clinico-demographic profiles of the encephalopathy and no-encephalopathy groups.
a To include encephalitis, meningitis, meningoencephalitis. b Any acute CVD (inclusive of CVD infarction and CVD hemorrhage.c To include acute disseminated encephalomyelitis, optic neuritis.d

Table 2
Comparison of outcomes in patients with COVID-19, with and without encephalopathy.
ICU: intensive care unit.

Table 3
Association between encephalopathy and the different outcomes of interest.