Malnutrition is frequent in the hospitalized pediatric patients. Critically ill children are even at higher risk of developing malnutrition and the majority have malnutrition at the time of hospitalization or develop acute and/or chronic malnutrition during the disease course (1–3). Malnutrition in the pediatric intensive care unit (PICU) patients is associated with increased mortality and morbidity such as predisposition to infections, impairment in gastrointestinal functions, prolonged hospitalization and prolonged mechanical ventilation (4).

Nutritional history of the patients must be gathered, appropriate energy requirements must be calculated and anthropometric data must be obtained to provide appropriate nutritional support. Nutrition guidelines recommend all children admitted to the intensive care unit undergo screening for nutritional status, monitorization of children for possible risk of malnutrition and specify a support plan in the presence of malnutrition. Metabolic demands must be determined by calculating energy consumption, feeding protocols must be followed by creating specialized feeding support teams and early enteral feeding must be provided as long as contraindications do not exist (5). The aim of the present study is to analyze anthropometric data, nutritional status, enteral and parenteral feeding support, and calorie and protein intake of critically ill pediatric patients in Turkey and evaluate the relationship between nutritional support and clinical outcomes.

In this countrywide prospective study that was conducted in Turkey, the data of 614 patients, who were admitted to the intensive care unit during a one-month study period and who met the study inclusion criteria, was examined. Of these patients, 54.7% (n = 336) were male and the mean age was 80.17 ± 68.35 months. The reason for admission to the intensive care unit was medical causes in 76.2% (n = 468) and emergency surgery in 12.2% (n = 75) of the patients. Of the patients, 62.7% (n = 385) had an underlying chronic condition. Descriptive statistics according to the reason for admission, source of admission, diagnoses and the underlying conditions are presented in Table 1.

The study found that 45.4% (n = 279) of the patients admitted to the pediatric intensive care unit had malnutrition according to weight for age classification. Of the patients, 23.0% (n = 141) had chronic malnutrition. The classification of nutritional status according to the duration of malnutrition, height-for-age, weight-for-age, weight-for-height, body mass index (BMI) Z-score, mid-upper arm circumference Z-score and triceps skin fold thickness Z-score is presented in Table 2. The body weight Z-scores upon admission to the PICU were compared with the Z-score on discharge in the groups created according to the BMI Z-score. There was a significant difference between median body weight Z-score upon admission to the PICU and on discharge in severely underweight patients (p = 0.019); the differences in the other groups were not statistically significant (p > 0.050) (Table 3).

Enteral feeding was initiated in 40.6% (n = 249) of the patients at day one upon admission to the intensive care unit, and 91.5% (n = 562) received enteral feeding at the end of one week. The rate of parenteral nutrition was 6.5% (n = 40) at day one upon admission to the intensive care unit and 8% (n = 30)at the end of one week. The barriers to the delivery of enteral nutrition was intubation/extubation procedures in 36.5% (n = 135) and intolerance to enteral nutrition in 30% (n = 111). The initiation of enteral nutrition within 48 h upon admission to the intensive care unit is regarded as early enteral nutrition that was achieved in 72.3% (n = 444) of the study patients. A comparison of nutritional status between patients initiated on early versus late enteral nutrition is presented in Table 4. No statistically significant difference was found in terms of nutritional status and feeding status according to the initiation of early enteral nutrition (p > 0.050).

The mean calorie intake at 48 h after admission to the intensive care unit was 43.94 ± 32.26 Kcal/kg/day and the mean protein intake was 1.24 ± 0.94 gr/kg/day. The mean calorie intake at the end of one week after admission to the intensive care unit was 57.39 ± 32.97 Kcal/kg/day and the mean protein intake was 1.61 ± 0.93 gr/kg/day. It was found that the target calorie intake was achieved in 86.82% (n = 533) of the patients in the first 48 h after admission. Of the remaining 376 patients, 81.65% (n = 307) achieved the target calorie intake at the end one week after admission to the intensive care unit (Table 5). The analysis of diagnoses among critically ill patients who did not achieve the target calorie intake at first 48 h and day seven after admission revealed that trauma and respiratory failure were the most common diagnoses. Intolerance to enteral nutrition occurred in 19.9% (n = 122) of the patients. These included vomiting (52%, n = 64), abdominal distension (46.3%, n = 57), electrolyte imbalance (17.9%, n = 22), diarrhea (16.3%, n = 20), gastrointestinal hemorrhage (11.4%, n = 14) and constipation (8.9%, n = 11). The factors affecting the achievement of target calorie intake at the end of 48 h and at day 7 were examined in univariate models using binary logistic regression analysis. Patients who did not receive mechanical ventilation in the first 24 h, who did not have nosocomial infections, who did not receive inotropic therapy and who received peroral nutrition than those receiving tube feeding were more likely to achieve the target calorie intake at 48 h after admission. The likelihood of achieving the target calorie intake at 48 h decreased with increasing VIS, PIM2 score, PRISM III score, patient age, duration of mechanical ventilation and length of stay in the intensive care unit and the hospital. Patients who did not undergo mechanical ventilation in the first 24 h, who did not develop feeding complications, who did not undergo plasmapheresis, who did not receive inotropic therapy and those who did not receive tube feeding were more likely to achieve the target calorie intake at day seven. The likelihood of achieving the target calorie intake at day seven decreased with increasing VIS, PIM2 score, PRISM III score, duration of mechanical ventilation and length of stay in the intensive care unit and the hospital (Tables 6, 7).

Mortality occurred in 7.2% (n = 44) of the 614 patients included in the study and the mean time to mortality was 19.02 ± 21.90 days. The factors affecting mortality were examined in univariate and multivariate models using binary logistic regression analysis. In univariate model, the risk of mortality decreased with increasing MUAC Z-score on discharge (OR = 0.871; p = 0.027). In univariate model, the risk of mortality decreased with increasing triceps skin fold thickness Z-score on discharge (OR = 0.894; p = 0.024). In univariate model, the risk of mortality was 2.723 times higher in patients who did not achieve the target calorie intake at 48 h than those who achieved the target calorie intake (p = 0.006). In univariate model, the risk of mortality was 3.829 times higher in patients who did not achieve the target calorie intake at day 7 than those who achieved the target calorie intake (p = 0.006). This rate was 8.036 in multivariate model (p = 0.043). In univariate model, the risk of mortality was 2.004 times higher in patients with malnutrition according to the weight-for-age than those who had normal nutritional status (p = 0.030). The risk of mortality was lower in patients who received early enteral nutrition (OR = 0.321; p = 0.001) (Table 8).

Nosocomial infections occurred in 13.7%of the patients (n = 84). The most common infections were bloodstream infections occurring in 9.4% (n = 58) followed by ventilator-related pneumonia and urinary tract infections occurring in 6.2% (n = 38) and 3.7% (n = 23) of the patients, respectively.

The mean length of stay in the intensive care unit was 11.14 ± 15.92 days, the mean length of hospital stay was 18.58 ± 19.07 days, and the mean duration of mechanical ventilation was 11.11 ± 17.77 days. The independent variables affecting the duration of stay in the PICU were examined using linear regression analysis. The length of stay in the PICU increased by 0.062 with increasing triceps skin fold thickness Z-score (p = 0.044). The length of stay in the PICU was 0.891 times higher in patients who did not achieve the target calorie intake at day seven (p < 0.001).

The present study found that the risk of mortality was lower, the length of stay in the intensive care unit and the hospital was shorter, and the duration of mechanical ventilation was shorter in patients who started on early enteral nutrition, who achieved the target calorie intake at the end of 48 h and 7 days, who did not have malnutrition and those who had high triceps skinfold thickness Z-score. The present study is the first prospective, multicenter study involving pediatric intensive care units in Turkey evaluating nutritional status in critically ill patients and the effect of nutritional status on clinical outcomes. There is an increased risk of malnutrition in critically ill patients who have limited calorie intake throughout their stay in the hospital. In our study patients, malnutrition before or upon admission to the intensive care unit may have developed in relation to various factors such as impairment in nutritional status, inability to take necessary nutrients and vomiting that develop secondary to medical condition. Malnutrition can also develop due to inability to feed during the hospital stay, loss of nutrients, delays in or the inability to deliver necessary nutritional therapy. In order to maintain optimal nutritional status during a critical illness, the initiation of appropriate nutritional therapies involving the administration of energy, proteins, lipids, micronutrients and vitamins at optimal amounts and through an appropriate route is required (3).

In a multicenter study conducted in Brazil, a total of 363 critically ill patients were evaluated. In their study, 62.3% of the patients were male; the mean age was 11.3 months; the reason for admission was medical causes in 75.2%; respiratory conditions were the most common diagnoses occurring in 42.2%; 45.7% had an underlying chronic condition, 76.4% initiated on feeding in the first 48 h; and the mortality rate was 5.5%(16). In the present study, 54.7% of the 614 patients were male with a mean in the entire study group of 80.17 months; the reason for admission was medical causes in 76.2%; 23.3% of the patients were admitted due to respiratory failure; and 62.7% of the patients had an underlying chronic condition. Of the present patients, 72.3% initiated on feeding in the first 48 h and the mortality rate was 7.2%. The rate of underlying chronic conditions was higher among the intensive care unit patients in Turkey, while the other parameters were similar to those reported by the study conducted in Brazil (16).

The rate of malnutrition is critically ill patients has been reported to be ranging between 40% and 70% (17–21). In a study conducted in four intensive care units in Adana province of Turkey, the rate of malnutrition was reported to be 41%, while no significant difference was reported between anthropometric measurements upon admission and on discharge from the intensive care unit (22). The rate of malnutrition classification which weight for age in the present study according to the data collected from the entire country was 45.4% (n = 279), and no statistically significant difference was noted between anthropometric measurements obtained upon admission to and on discharge from the intensive care unit. In a study of 951 pediatric patients involving females at a rate of 51.6%, the rate of acute malnutrition was 21.3% and the rate of chronic malnutrition was 41.3% (23). In a study evaluating the data of a total of 293 pediatric patients in the neonatal and pediatric intensive care unit in the Netherlands, the rate of malnutrition was found to be 24%, and when the patients were evaluated according to the duration of malnutrition, 15% had acute, 20% had chronic and 24% had acute/chronic malnutrition. Of the reported patients, 32% were critically ill patients above the age of one month (3). When pediatric patients in Turkey were classified according to the duration of malnutrition, 15% had acute, 23% had chronic and 15.6% had acute/chronic malnutrition.

Although the term malnutrition is used to denote individuals with undernutrition in our daily practice, the term is not specific to undernutrition but also encompasses overweight and obese patients. It was reported that 13% of the patients admitted to the pediatric intensive care units were either overweight or obese according to the BMI Z-score (24, 25). This rate was 6.9% in a study conducted in Brazil (16) and 6.8% in the present study. In retrospective review of 243 pediatric intensive care unit patients in Ethiopia, this rate was reportedly around 11% (26). In a multicenter cohort study of 1,622 patients undergoing mechanical ventilation in 16 countries, nutritional status based on the BMI data was found to be related to clinical outcomes such as the number of ventilator-free days, nosocomial infections and mortality (1). In the same study, 27.9% of the patients were reported to overweight or obese (1). In the present study that used mid-upper arm circumference and triceps skin fold thickness Z-scores to evaluate nutritional status in critically ill patients, the risk of mortality decreased with increasing Z-scores. In another study, mortality remained unaffected with increasing severity of malnutrition, whereas the severity of malnutrition was found to be related to the length of stay in the intensive care unit and prolonged duration of mechanical ventilation (17).

Achieving optimal calorie and protein intake in critically ill patients by early initiating optimal enteral feeding is associated with an improvement in clinical outcomes (27). It is recommended that enteral nutrition be initiated within 24–48 h after admission to the intensive care unit in hemodynamically stable children with a functional gastrointestinal tract if there is no contraindication such as vomiting, abdominal distension and gastrointestinal hemorrhage, and this approach is referred to as early enteral nutrition (28). In our country, 40.6% of critically ill patients initiated on enteral nutrition in the first 24 h after admission and this rate was 72.3% at the end of 48 h. At the end of one week, 91.5% of our patients started on enteral nutrition. The target calorie intake at the end of 48 h was achieved in 86.8% of our patients, whereas 81.6% of the patients continuing their stay in the intensive care unit at the end of one week achieved the target calorie intake. The analysis of diagnoses among critically ill patients who did not achieve the target calorie intake at both time points revealed that trauma and respiratory failure were the most common diagnoses. The rate of complications associated with enteral feeding was 19.9% in our patients, the most complications being vomiting followed by abdominal distension, electrolyte imbalance, diarrhea, gastrointestinal hemorrhage and constipation. It was observed that patients receiving mechanical ventilation support, inotropic therapy and those with high mortality scores were less likely to achieve the target calorie intake. Intubation/extubation procedures and intolerance to enteral nutrition appeared to be the most common reasons when the factors associated with a delay in enteral nutrition were evaluated. In a multicenter, cross-sectional study, mechanical ventilation, disease severity, interventional procedures and gastrointestinal diseases were identified as clinical risk factors associated with a delay in enteral nutrition (29). Other studies have reported that enteral nutrition had been interrupted in 30%–42% of pediatric intensive care unit patients due to intolerance to enteral nutrition, extubation and intubation procedures, bedside interventions, and radiological and surgical procedures (30, 31).

It was found in the present study that 6.5% of the patients started on parenteral feeding within the first 24 h after admission to the intensive care unit and this rate was 8% at the end of one week. However, available guidelines on nutrition do not recommend initiation of parenteral feeding within 24 h after admission to the intensive care unit (5, 28). In the PEPaNIC study, initiation of parenteral feeding within 24 h after admission to the intensive care unit was defined as early parenteral nutrition and initiation of parenteral feeding after seven days was defined as late parenteral nutrition, and the outcomes of early and later parenteral nutrition were compared in critically ill patients. The rate of infections, duration of mechanical ventilation and the length of stay in the intensive care unit were significantly lower in the late parenteral nutrition cohort (32, 33). In a study evaluating nutritional status in 95 critically ill patients, the rate of parenteral nutrition was 7.7% and early parenteral nutrition was initiated in 33% of the patients in a mean duration of four days (1–6 days) (34).

There is an increased risk of malnutrition in critically ill patients who have limited calorie intake throughout their stay in the hospital. It is recommended that critically ill patients should receive a minimum of 54–58 kcal/kg/day nutritional support in order to avoid a catabolic state and maintain protein and energy support as required (28). In the present study, the patients had a mean calorie intake of 57.39 ± 32.97 kcal/kg/day as recommended by the guidelines. Protein intake must be at least 1.5 gr/kg/day in pediatric intensive care unit patients to avoid negative protein balance (5, 28). Our patients received 1.6 gr/kg/day protein support. A study investigated the nitrogen and energy requirement using indirect calorimeters and by performing more than 400 measurements in patients undergoing mechanical ventilation. At the end of this study involving 74 critically ill patients, the authors concluded that minimum protein intake must be 1.5 gr/kg/day and minimum calorie intake must be 58 kcal/kg/day (35).

The present study found that the risk of mortality was lower, the length of stay in the intensive care unit and the hospital was shorter, and the duration of mechanical ventilation was shorter in patients who started on early enteral nutrition, who achieved the target calorie intake at the end of 48 h and 7 days, who did not have malnutrition and those who had high triceps skinfold thickness Z-score. In a study involving 59 pediatric intensive care units from 15 different countries and evaluating optimal protein intake and clinical outcomes in 1,245 critically ill patients undergoing mechanical ventilation, the delivery of daily protein support as required was associated with decreased risk of mortality (36). In another study examining 385 pediatric intensive care unit patients, the rate of malnutrition was 45.5%, and malnutrition was found to be associated with prolonged mechanical ventilation but no association with mortality and the length of stay in the intensive care unit was reported (37). In a multicenter, retrospective study, reaching 25% of the target calorie intake at the end of the first 48 h after admission to the intensive care unit and the initiation of early enteral nutrition were found to be associated with an increased probability of survival compared to late enteral feeders (38). In a study conducted in Turkey evaluating the initiation of early enteral nutrition and achievement of the target calorie intake in the early period, the data of 95 critically ill patients from nine pediatric intensive care units was examined. Early initiated feeding (EIF) was defined as the initiation of enteral nutrition within 24 h after admission to the intensive care unit, and early reached target enteral nutrition (ERTEN) was defined as the receipt of 25% of the total energy requirement at the end of the first 48 h. In the scope of the above-mentioned study, 42% of the patients had respiratory failure and the mortality rate was 16.8%. When the patients were evaluated for nutritional status, 47.4% were in the EIF group; the rate of enteral feeding at day two was 72.6% and 45.3% of the patients were in the ERTEN group. The authors reported that early achieving the target calorie intake decreased mortality (39).

The present study has some limitations. The sample size was small because not all pediatric intensive care units in Turkey participated in the study and there was a small number of patients included in the study. Calorie and protein intake of the patients was evaluated but the effects of other micronutrients, lipids and other elements of nutrition were not studied. In addition, there may be differences in the feeding protocols and practices of the clinics.

The present study found that the risk of mortality was lower, the length of stay in the intensive care unit and the hospital was shorter, and the duration of mechanical ventilation was shorter in patients who started on early enteral nutrition, who achieved the target calorie intake at the end of 48 h and at 7 days, who did not have malnutrition and those who had high triceps skinfold thickness Z-score. Due to the fact that inadequate and imbalanced nutrition can prolong disease course and increase the length of stay in the intensive care unit and the hospital, timely and appropriate nutritional support may reduce morbidity and mortality in critically ill children. Critically ill patients followed up in the intensive care units require evaluation of the nutrition status, early detection of malnutrition and close monitorization of the patients for the need of nutritional interventions by calculating the energy and protein requirements. The present study is the first in Turkey to evaluate nutritional status, nutritional follow-up and the effects of nutritional therapies on prognosis in pediatric intensive care unit patients, and the data deriving from the present study would be of great importance in the planning of future studies, patient follow-ups and arrangement of treatment protocols.