Determination of Cutoff Point of an Early Energy Target to Maximize Better Clinical Outcomes in SCU Patients with Cerebral Hemorrhage

Stroke is the third leading cause of death [1]. Among stroke subtypes, cerebral ischemia (CI), cerebral hemorrhage (CH) have directly insult to cerebral neurons, whereas subarachnoid hemorrhage does not. Comparing outcome of CH with CI, patients with CH have been reported to have higher in-hospital mortality [Odds ratio = 2.9] [2]. Under these circumstances, we hypothesized that CH patients have poorer outcomes compared with patients with CI and that an energy management set at 75% of target was associated with better outcome. Here, target of energy intake (TEI) was set at 25 kcal / kg of actual body weight / day according to the statement of guidelines provided by American Society for Parenter al and Enteral Nutrition [3], and European Society for Parenteral and Enteral Nutrition [4]. Then if our hypothesis was correct, then examine that the cutoff point of TEI exists to maximize better outcomes. To test our hypotheses, all data was collected and analyzed in patients with CI and CH admitted to stroke care unit (SCU) in a single tertiary institute.


Background
Stroke is the third leading cause of death [1]. Among stroke subtypes, cerebral ischemia (CI), cerebral hemorrhage (CH) have directly insult to cerebral neurons, whereas subarachnoid hemorrhage does not. Comparing outcome of CH with CI, patients with CH have been reported to have higher in-hospital mortality [Odds ratio = 2.9] [2]. Under these circumstances, we hypothesized that CH patients have poorer outcomes compared with patients with CI and that an energy management set at 75% of target was associated with better outcome. Here, target of energy intake (TEI) was set at 25 kcal / kg of actual body weight / day according to the statement of guidelines provided by American Society for Parenter al and Enteral Nutrition [3], and European Society for Parenteral and Enteral Nutrition [4]. Then if our hypothesis was correct, then examine that the cutoff point of TEI exists to maximize better outcomes. To test our hypotheses, all data was collected and analyzed in patients with CI and CH admitted to stroke care unit (SCU) in a single tertiary institute.

Methods
All consecutive acute stroke patients, admitted to a single institution between January and June 2015, were enrolled as subjects of the study. They were divided into two groups, diagnosed with CH or CI as primary diagnosis. The exclusion criteria were: (3) Nutritional parameters, including daily energy intake (EI: kcal / kg of actual body weight / day) for the first seven days staying in SCU. Here, when subject has discharged within six days from SCU, an average daily EI was calculated until the day before discharge from SCU. In addition, nutritional route, including oral or the others (enteral or parenteral) was also classified, (4) Outcome parameters, including the length of stay in SCU as the primary outcome, and the highest CRP during stay in SCU, the presence of CRP ≥ 6.0 mg/dL, both as the second outcomes.
Then, the following analyses were conducted in analysis 1, and 2 ( Figure 1). Given the nature of this study, the requirement for informed patient consent was considered unnecessary.

Analysis 1
To examine our hypothesis that CH patients have worse outcomes compared with that of CI patients, all included subjects was divided into two groups according to their primary stroke subtype, CI or CH. Then, all collected data was compared between two groups to know which group show significantly poorer outcome. Here, group that showed poorer outcome were proceeded to analysis 2.

Analysis 2
To examine our hypothesis that patients fed with %EI of the target is associated with maximizing clinical outcome in SCU, all CH subjects were divided into two subgroups by EI of % target energy (TEI), < vs. ≥ 75%, and followed by added the same analyses with the TEI set at 50% and 25%. Then, comparisons of all collected data was conducted between two subgroups to which target %EI, 75%, 50%, or 25% of %of target EI, is strongest to show significance in outcomes.

Statistical Analysis
The outcome parameters in the two groups were divided into different categories and compared using the Mann-Whitney U test for continuous variables and the chi-square test or Fisher's exact test for categorical variables. All analyses were performed using SPSS Statistics version 24 (IBM Corp., Armonk, NY, USA), and significance was examined at P < 0.05.

Results
Three hundred and sixty patients were enrolled and a hundred and thirty were excluded. Two hundred and thirty patients were included to analyze in further analyses ( Figure 1).

Result 1-Results for analysis1
Compared with patients among all patients with CH and CI, CH patients showed higher NIHSS (10 vs.3, p < 0.001), and their outcome parameters showed that longer length of stay in SCU (7 vs. 5, p < 0.001) and higher CRP during stay in SCU significantly (1.6 vs. 0.2, p < 0.001) ( Table 1).

An Impact of an Early 75% of the Target Energy Intake is Associated with Maximizing Clinical Outcomes in Critically Ill Stroke Patients
In the present study, the length of stay in SCU and CRP values were applied as outcome parameters. Among them, CRP ≥ 6.0 mg/ dL was considered as a biomarker of infectious events in the older patients [5]. In analysis 1, as CH patients showed poor outcome compared with patients with CI, next question was raised whether an energy management have an impact on clinical outcome or not.
As the results showed in analysis 2, CH patients managed with smaller energy ≥ 75% of TEI seems associated with significantly lower frequent of CRP < 6.0 mg/dL compared with patients fed with larger energy ≥ 75% of the target. This result could be interpreted that more frequent lower CRP in SCU and shorten LOS in SCU could be achieved by %EI ≥ 75% of the target. In these contexts, the cutoff point of %EI might exist at 75%. Abbreviations CRP: C-reactive protein, E-L: Energy intake < of target, E-H: Energy intake ≥ of target, SCU： stroke care unit.

Maximize Better Clinical Outcomes in Stroke Patients
To consider how much energy intake must be associated with outcomes, we would discuss three scenarios, such as %EI set at 75%, > 100%, and meta-analysis, separately.
%EI ≥ 75% of the Target: As A.S.P.E.N. [6] and ESPEN [7] guidelines has similarly recommended to enhance an early enteral nutrition to achieve better outcome for critically ill patients, addressing attentions to the question how much energy is associated with an improvement of outcomes in critically ill settings, a study entitled the tight calorie control study (TICACOS) seems to be informative. The authors reported that critically ill patients were divided into two groups by the method to calculate EI: one was calculated by indirect calorimetry (study group) and another was set by weight-based calculation, 25 kcal/kg/day as the target (control group In the other studies, underfeeding is also reported associated with adverse events, such as hypoglycemia, hypothermia, infectious complications, delayed wound healing, and an impaired immunity [9,10]. Similarly to this results, another studies also reported that < 70% energy provision was associated with significantly higher 30day mortality compared with > 70 % of the target (31.5 vs. 11.1 %, p=0.01) [11], nutritional guideline also showed that permissive,

%EI > 100 % is Associated with Adverse Events: Moving to
overfeeding with > 100% TEI, a cohort study also confirmed that both overfeeding, defined > 110 % and underfeeding, defined < 70% of the target [8], were reported harmful for critically ill patients [14]. Here, an overfeeding nutritional support has been reported associated with hyperglycemia, hyperlipidemia, and hypercapnia is associated with impairments of autophagy [15]. From these observations, energy management > 100 % of the target seems harmful associated with overfeeding of nutrients mentioned above.
However, it must be examined by well-designed studies whether the ceiling of TEI exists or not.
on comparing effects of hypocaloric with standard energy support in length of hospital and SCU stay, infectious complications and the length of mechanical ventilation, because of very low evidence [16].
To extract a conclusive energy management in critically ill settings, although our study showed EI ≥ 75% of the target was associated with better outcomes, further analyses must be studied to confirm definitive target %EI.

Prioritizing an Enteral Route for Maintenance of Integral Gut Immunity in Critically Ill Patients
The nutritional route, parenteral vs. enteral, or oral, must be considered to achieve better outcomes. In general, an enteral nutritional route has been reported to enhance immunity to prevent infectious events [17]. A nutritional management through enteral route decreases infectious complications and LOS in SCU [18]. One of the reasons to explain why enteral route did not show advantage might be that enteral hypo-perfusion might interrupt enteral digestion and absorption of the nutrients. In addition, intestinal alkaline phosphatase (iAP) has been proved to be expressed in intestinal epithelial cells to exclude endotoxin and maintain integral gut immunity [19,20]. However, iAP has never been proved its clinical roles I critically ill settings. At the present, roles of iAP regarding an effectiveness of nutritional route must be studied in the further analyses.

Limitations
First, it is a retrospective study. To draw the definite conclusion of energy intake for CH patients, prospective, randomized study must be necessary. Second, the cutoff point of %EI was resulted from a single facility and validity has not yet been fully verified.
Third, the sample size was also too small to draw definitive conclusion. The multi-institutional studies are warranted to resolve these limitations.

Conclusion
This study could draw the following two conclusions: 1) CH patients have worse outcomes compared with CI patients.
2) A cutoff point of %EI was determined at 75% of the target to maximize better clinical outcomes, such as shorter stay in SCU and lower frequency CRP ≥ 6.0 mg/dL, in CH patients