Nitrogen balance in mechanically ventilated obese patients

This study aimed to evaluate if the protein intake recommendations for obese critically ill requiring mechanical ventilation are sufficient to promote

to negative group with median of -5.27g/day (-10.38 to -3.86g/day). Adults had higher ratio of negative nitrogen balance (57.1%) than elderly (44.4%), with protein intake of 2.0 versus 2.1g/day, respectively. No correlation was found between nitrogen balance and variables assessed.

Conclusion
High-protein enteral nutrition therapy contributed to positive or neutral nitrogen balance for approximately half of obese ventilated individuals. With similar protein intake, elderly showed a higher proportion of positive or neutral nitrogen balance. Nitrogen balance can be influenced by various factors, so further studies are required to identify different protein needs in obese critically.

I N T R O D U C T I O N
In the critical care setting, obesity can contribute to exacerbate ventilatory, cardiovascular, renal and endocrine dysfunctions and pharmacokinetic and pharmacodynamic alterations, leading to an increased metabolic and inflammatory state [1][2][3][4]. Due to these metabolic alterations added to physical inactivity, muscle protein synthesis is decreased and muscle mass can be utilized as an energy substrate for gluconeogenesis, leading to excessive proteolysis, a condition associated with higher mortality [3][4][5][6].
Nutritional Therapy (NT) recommendations suggest hypocaloric high-protein NT for obese, with the purpose to prevent overfeeding, promote an adequate glycemic control and a positive or neutral Nitrogen Balance (NB), in order to contribute to immune function regulation, wound healing and to decrease the loss of muscle mass [6][7][8][9][10][11][12].
Studies suggest that high-protein NT contributes to an adequate NB in obese individuals even when in a hypocaloric diet [13][14][15][16][17]. However, these studies were based on small sample sizes, heterogeneous populations, with most individuals categorized as surgical or trauma patients, most of whom received parenteral nutrition, instead of Enteral Nutrition (EN) and were not necessarily submitted to Mechanical Ventilation (MV) [7,9]. Considering the characteristics of these studies, there are concerns regarding the generalization of the recommendations to all obese individuals admitted to Intensive Care Units (ICU), including those on MV. The objective of this study is to evaluate if the protein intake recommendations for obese critically ill requiring MV and with hypocaloric EN are sufficient to promote a positive or neutral NB.

M E T H O D S
Cross-sectional study conducted at the adult ICU of a public teaching hospital. The sample was composed by patients classified as obese according to Body Mass Index (BMI) ≥30kg/m², ≥18 years old, on MV and receiving full EN. Between April and August 2018, 70 patients underwent the screening assessment, and after exclusions (n=45), 25 individuals were included on this study. Exclusion criteria were the presence of diarrhea, renal failure (considered as serum creatinine >2.0mg/dL, glomerular filtration rate <30mL/min or renal replacement therapy), hepatic failure (considered as bilirubin levels higher than three times the reference value) and pregnancy.
The following data were analyzed: (a) socio-demographic: age, gender, ethnicity; (b) clinical: diagnosis at admission classified as clinical (respiratory, neurological, cardiovascular or gastrointestinal) or surgical, comorbidities, disease severity (SAPS 3 score -Simplified Acute Physiology Score III and SOFA score -Sequential Organ Failure Assessment Score) (c) nutritional: body weight, height, BMI, time until the onset of NT, protein and caloric intake and adequacy; (d) biochemical: one time 24-hour urinary urea excretion; (e) outcomes: NB, MV time, ICU and hospital time and mortality in the ICU and in the hospital.
Anthropometric data was collected by the staff at the moment of admission, height was measured using a Luft ® ruler [18], and weight by using a Hill Rom ® in-bed. Subsequently, BMI was calculated and patients were considered class I obesity when BMI ≥30kg/m², class II when BMI ≥35kg/m² and class III when BMI ≥40kg/m². There were no changes in the routine of care at the unit.
Nutritional therapy protocol utilized stratifies the protein recommendations by BMI degrees [7]. The protein target for BMI 30 to 40kg/m² is 2.0g/kg of Ideal Body Weight (IBW) and for BMI >40kg/m² is 2.5g/kg IBW, and calorie target for BMI 30 to 50kg/m² is 11-14Kcal/kg of Actual Body Weight (ABW) and for BMI >50kg/m² is 22-25Kcal/kg IBW [7]. After defining individual calorie and protein targets, the amount of EN formula is calculated and when protein targets are not reached with enteral formulas, protein supplements are prescribed.
Nutritional therapy was provided by enteral route, using a high-protein polymeric fiber-free formula, with caloric density of 1.5kcal/mL. During the first 48 hours in ICU, patients were evaluated by ICU physicians regarding to hemodynamic and perfusion conditions to start NT. Beggining of EN was at 10 to 30mL/hour and it was increased in 10 to 15mL/hour per day until the target is reached. After EN target was reached, protein supplements were included. Protein supplements were constituted by 100% hydrolyzed whey protein and provided 10-20 grams-dose, given 3 to 5 times a day, as targets prior determined. At the unit, the diet is provided continuously, using a closed system, with nutritional targets computed for 22 hours of feeding, considering a 2-hour break every day for ICU procedures aiming to optimize nutritional adequacy.
Nitrogen Balance was measured one day after the patient achieved the prescribed nutritional target, and was calculated by the equation: NB (g/day)=nitrogen intake (g/day) -[urinary urea nitrogen (g/day)+4 (g/day)]. Considering 4g as a constant factor from integumentary, gastrointestinal and insensible losses [19]. NB was defined as positive when nitrogen intake was greater than excretion, negative when excretion was greater than intake and in equilibrium or neutral when both were equal. Nitrogen intake was computed using records from the total amount of enteral diet and protein supplements received during the NB realization day. Urinary urea nitrogen was measured in a one-time 24-hour urine sample, being the urine collected by a urinary catheter every six hours and stored in a refrigerator during the 24 hours of collection. This study only included patients that already had a urinary catheter for clinical purposes. The urine analyses were realized in the clinical analyses laboratory of the hospital using the glutamate dehydrogenase enzymatic kinetic assay method. Electronic medical records were analyzed to assess clinical, demographic, nutritional and biochemical data. The patients were included only after the informed consent was obtained from those legally responsible. This study was approved by the Institutional Ethics Committee under record number #18-0006.
The study conducted by Dickerson et al. [15] was used to estimate the sample size, considering the mean and standard deviation of daily protein ingestion of 2.3±0.2g/kg IBW, a standard error of 0.2, 95% confidence interval and 10% of losses, requiring 25 patients for inclusion. SPSS, v.21.0 (IBM ® SPSS ® Statistics Inc., Chicago, Illinois, United States) was used for statistical analysis with a statistically significant p value of <0.05. Mann-Whitney test was performed for comparisons between continuous variables. Correlation analysis between socio-demographic, clinical and nutritional variables and NB were conducted with Spearman's rank correlation. Fisher's exact test was used to association analysis between nutritional variables and outcomes.
Median overall NB was 0.3g/kg (-5.3 to 4.8g/day), with median time for NB measurement of three (2-6) days after ICU admission, the period required for NT to reach the target. Considering clinical outcomes, median length of MV was 10 (7-19) days, ICU was 14 (8-23) days and hospital was 28 (19-49) days. Mortality rates in ICU and hospital were of 20% and 32%, respectively. Thirteen (52%) individuals had positive or neutral NB, while twelve (48%) had negative NB. Table  2 describes comparisons between NB groups. Differences between NB groups for NB and urinary urea excretion were observed. Similar protein intake, age, BMI and SAPS and no differences for secondary outcomes (MV days, ICU and hospital time, mortality in ICU and in hospital) were found.
When stratified by age, negative NB was observed in 44.4% of elderly (≥60 years old) and 57.1% of adults, without statistical difference. The median NB for elderly and adults were 0.7g/kg (-6.4 to 4.5g/kg) versus -3.5g/kg (-4.7 to 5.9g/day) and median protein intakes were 2.1g/kg (2.0-2.3g/kg) versus 2.0g/kg (1.9-2.2g/kg) IBW, respectively.   Patients with class III obesity showed a higher prevalence of positive or neutral NB, being 60.0% compared to class I with 54.5%, and class II with 44.4%. The median of NB in these groups was 1.66g/day (-5.88 to 3.24g/day) for class I, -1.81g/day (-6.19 to 4.64g/day) for class II, and 4.23g/day (-5.93 to 7.22g/day) for class III obesity, without statistical difference.
No significant correlation was found between NB and socio-demographic, clinical and nutritional characteristics, as seen in Table 3. No association was found between NB and mortality in ICU and in hospital. https://doi.org/10.1590/1678-9865202134e190263

D I S C U S S I O N
All individuals received high-protein enteral feeding, with a median of 2.1g/kg IBW, providing positive or neutral NB for approximately half of obese individuals requiring MV. Analyzing by groups, positive or neutral versus negative NB, a difference was found for NB and urinary urea excretion, but no differences were found for protein and calorie intake and adequacy, and SAPS 3 score. No significant correlation was found between NB and assessed variables. No association was found between NB and the outcomes assessed, and we hypothesize that it could be explained by the restricted number of patients in our sample and maybe due to its heterogeneity. Patients had different obesity degrees and diagnoses at admission, also, even the most of them were admitted for clinical purposes, three were surgical and three had oncologic disease, situations that can influence the protein needs.
Catabolism is exacerbated by critical illness, specially protein catabolism that contributes to loss of muscle mass, with loss proportional to disease severity [20]. Considering that stressed obese commonly are insulin resistant, this group may have a higher protein turnover and greater catabolism [21]. Some studies showed that a greater adequacy in protein and calorie targets is associated with greater outcomes for ICU patients, as reported by Nicolo et al. [22], where a protein adequacy >80% independent of calorie intake had association with reduced mortality rates. In our study, protein adequacy was greater than 95% for most of the sample, but even while receiving adequate protein intake no difference was found between positive and negative NB groups in relation to mortality rate.
Recommendations suggest that obese ICU patients can achieve adequate NB receiving high-protein NT, even when receiving hypocaloric diet [7]. However, these studies were conducted with populations of obese who were mostly classified on admission in ICU as surgical or trauma, and who were predominantly administered parenteral nutrition, with no difference in NB values found when compared to patients who received hypocaloric versus eucaloric high-protein NT [13,23,24].
Recently studies demonstrated that hypocaloric high-protein NT is safe to obese critically ill, including sarcopenic obese [6,11,21,25]. A subset analysis of pre-morbid obese critically ill found that individuals with hypocaloric vs normocaloric NT, but with similar protein intake, had no statistical difference in relation to ICU and hospital time, mortality and infection rates [26].
In our sample 52% had positive or neutral NB, being NB 0.3 (-5.3 to 4.8) g/day, with median protein intake of 2.1g/kg IBW receiving enteral feeding. A study with 25 mechanically ventilated obese patients found a NB of -2.3±5.1g/day, but protein requirements met 25% of target [17]. Forty obese critically ill  [14]. However, obese critically ill with traumatic injury showed similar results to our study in relation to positive or neutral NB, being 55% and 56% for older and younger individuals, respectively [15,16]. In some specific situations, critically ill obese may need higher amounts of protein compared to nonobese, as described in two studies that evaluated patients in extracorporeal membrane oxygenation, whereas obese groups had a more pronounced negative NB [27,28].
Literature suggests that elderly individuals may have a decline in anabolic response and difficulty mobilizing body fat stores as energy sources, conditions that lead to an increase consumption of muscle mass [11,29]. Our analysis showed that adults had higher rates of negative NB compared to elderly subjects. In a similar way, a study that compared elderly obese patients to adult ones, showed that adults had greater values for NB 3.44 versus 0.2g/day, with similar protein ingestion of 1.6-1.7g/kg adjusted body weight [29]. However Dickerson et al showed that in trauma patients NB values were lower, but not statistically significant, between adults (-4.9d/day) in comparison to elderly (-3.2g/day) with similar protein intake of 2.3g/kg IBW of protein [15].
Nitrogen balance may be influenced by various factors and it has the aim to help in evaluation of protein intake adequacy to individual's needs. Although our sample consisted solely by obese patients undergoing MV, this study has some limitations. NB was measured only once, at the early stage of acute injury and sample was not stratified by severity, thus our results may reflect various metabolic states, making it difficult to interpret protein requirements and adequacy at different disease severities and stages. NB was not analyzed according to different degrees of obesity and number of comorbidities, which can influence in protein needs.

C O N C L U S I O N
Our data suggest that a daily protein intake around 2.1g/kg IBW was able to promote positive or neutral NB in 52% of the population of obese individuals undergoing MV, enteral NT and predominantly receiving clinical treatment. Studies with large sample size, serial NB measures in the first ICU week, severity stratification, and analysis according to different obesity degrees are required in order to identify different protein needs of obese critically ill patients undergoing MV.

C O N T R I B U T O R S
All authors have contributed to the conception, design, analysis and interpretation of the data and agree to be fully accountable for ensuring the integrity and accuracy of the work, and read and approved the final manuscript.