Secondary Logo

Journal Logo


Frequency and risk factors for malnutrition in children undergoing general anaesthesia in a French university hospital

A cross-sectional observational study

Gerbaud-Morlaes, Louis; Frison, Eric; Babre, Florence; De Luca, Arnaud; Didier, Anne; Borde, Maryline; Zaghet, Brigitte; Batoz, Hélène; Semjen, François; Nouette-Gaulain, Karine; Enaud, Raphael; Hankard, Régis; Lamireau, Thierry

Author Information
European Journal of Anaesthesiology: August 2017 - Volume 34 - Issue 8 - p 544-549
doi: 10.1097/EJA.0000000000000618
  • Free



Malnutrition is a medical condition resulting from a mismatch between protein-energy needs and intake and is mostly associated with famine in many developing countries. Malnutrition is also encountered in developed countries where its prevalence is between 10 and 15% of hospitalised children.1–6 The severity of malnutrition can range from mild undernutrition to failure to thrive7 and is related to an increased length of hospital stay, increased frequency of gastrointestinal and infectious complications, weight loss during hospital stay and quality of life.8

In adults undergoing surgery, preexisting malnutrition is associated with postoperative complications,9 longer hospital lengths of stay and increased healthcare-related costs.10,11 Clinical guidelines for adults suggest that preoperative malnutrition should be evaluated to enable optimisation of perioperative care with oral supplements or enteral nutrition.12 In children, malnutrition leads to impaired growth and increased morbidity and mortality.1,9,10,13–15 Malnutrition increases the incidence of postoperative complications, and these are responsible for increasing the length of hospital stay,4,13,16,17 worsening of the underlying disease, worsening of malnutrition itself2,18–20 and increasing the hospitalisation costs.8,11,17 However, malnutrition is often underdiagnosed in children and rarely leads to specific treatment.1 The European Council considered this situation as unacceptable,21 and screening for malnutrition in children in hospital was recommended. Mostly adopted for adults, these measures are considered as a marker of improvement of quality of care. Although it has been known for a number of years that nutritional assessment is an essential part of preoperative evaluation,22 routine screening of malnutrition in children undergoing general anaesthesia is still lacking.1

The aim of the study was to estimate the frequency of malnutrition in children undergoing anaesthesia and to identify factors associated with malnutrition.


The current cross-sectional observational study was conducted in the paediatric anaesthesia department of a University Children's Hospital (Bordeaux, France) during a 10-month period (1 February 2013 to 30 November 2013). In this tertiary hospital, all procedures under general anaesthesia are performed in a multi-purpose operating wing. Patients examined at a pre-anaesthesia evaluation were prospectively included in the study. The following parameters were recorded during the pre-anaesthesia visit: duration of gestation in weeks (GW), weight (kg) and height (cm) at birth, current weight (kg), height (cm) and BMI (kg m−2), Waterlow index (weight/expected weight for height),23 presence of clinical signs of malnutrition noted by the anaesthetist (loss of muscle mass, loss of subcutaneous fat and skin or hair abnormalities), underlying chronic disease, whether surgery was urgent or scheduled, the type of surgery, the American Society of Anesthesiologists (ASA) physical status score24 and the Pediatric Nutritional Risk Score (PNRS). PNRS is a simple score that can be easily calculated from three parameters: the underlying medical condition divided into three categories (grade 1: a mild stress factor such as a minor infection or minor surgery; grade 2: a moderate stress factor such as major surgery, inflammatory bowel disease, non-decompensated cardiopathy, severe infection, cystic fibrosis; and grade 3: a severe stress factor such as cardiac or major visceral surgery, severe sepsis, multiple injuries, severe burns, haematological disease and cancer, and deterioration of chronic disease), nutritional intake (above or below 50% of daily dietary allowance, according to sex and age) and the presence or absence of pain. Based on the PNRS, patients can be classified into three nutritional risk groups – low, moderate or high – with a good prediction of weight loss during hospitalisation.18

Data were anonymously entered in the e-PINUT website (, an online nutritional assessment tool developed by RH and certified by the French Society of Parenteral and Enteral Nutrition in 2012. Nutritional indices, such as weight for age, height for age and weight for height, were automatically calculated and expressed as z-scores, percentiles and as a percentage of the median, following the French National Center for Health Statistics reference curves.25 Medical conditions were encoded according to the International Classification of Diseases, 10th edition, and therapeutic procedures were encoded according to the French Classification of Medical Acts.26 The Waterlow index was considered indicative of moderate or severe malnutrition when below 80%. BMI was considered indicative of malnutrition when below the third percentile,27 according to the French references provided by Rolland-Cachera et al.28

Statistical analysis

Characteristics were described as median and interquartile range for quantitative parameters, and number and percentage for qualitative parameters. The agreement between BMI, Waterlow index and clinical examination for the diagnosis of malnutrition was studied in pairs and described as a percentage of discordant patients and its 95% binomial proportion confidence interval and as the Cohen's kappa coefficient and its 95% confidence interval (95% CI). Strength of agreement was categorised according to the Landis and Koch scale.29 The association between PNRS and malnutrition defined by BMI was estimated using Fisher's exact test. A multivariate logistic regression model was used to estimate the association between patient clinical characteristics/scores and the presence of malnutrition defined by BMI as the dependent variable. Parameters associated with malnutrition in the univariate analysis at a threshold of P less than 0.20 were included in the multivariate logistic model. Associations were expressed as odds ratios with their 95% CI. The overall adequacy of the logistic regression models was verified by the Hosmer and Lemeshow test. Statistical analyses were performed using the software R v3.1.0. [(R Core Team (2016). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL] This study was approved by the ethics committee of the southwest of France (15/27).


During the study period, 1101 patients were admitted to our hospital for a preoperative evaluation by an anaesthetist. One hundred and sixteen (10.5%) patients were excluded because of incomplete data, leaving 985 patients in the study. Characteristics of the population are described in Table 1.

Table 1
Table 1:
Characteristics of children undergoing general anaesthesia (n = 985) – Bordeaux, France

Frequency of malnutrition

Clinical signs of malnutrition were present in 8.1% of children. Using anthropometric data, malnutrition was present in 11% of children when defined by BMI and 7.6% when defined by the Waterlow index. This latter index was indicative of acute malnutrition (wasting) in 6.1% and chronic malnutrition (stunting) in 1.5% of children.

Concordance between the three methods of assessing malnutrition

A hundred and six patients [10.8%; 95% CI (8.9 to 12.9%)] had a discordant evaluation of malnutrition between BMI and clinical examination, with a poor agreement between these two variables [κ = 0.38 (0.28 to 0.47)]. Thirty-nine patients with clinical signs of malnutrition had a BMI more than the third percentile; and 67 patients having a BMI less than the third percentile had no clinical signs of malnutrition.

In 95 patients [9.6%; 95% CI (7.9 to 11.7%)], there was disagreement in the evaluation of malnutrition between the Waterlow index and clinical examination, with a poor agreement between these two variables [κ = 0.33 (0.23 to 0.44)]. Fifty patients with clinical signs of malnutrition had a Waterlow index more than 80%; and 45 patients having malnutrition according to the Waterlow index had no clinical signs of malnutrition.

Although there was good agreement between the BMI and the Waterlow Score [κ = 0.7; 95% CI (0.64 to 0.79)], there were 47 patients in whom there was no agreement between these two indices [4.8%; (7.9 to 11.7%)]. Seven patients with malnutrition according to the Waterlow index had a BMI more than the third percentile, whereas 40 patients with a BMI less than the third percentile had a Waterlow index more than 80%.

A higher PNRS was associated with a higher proportion of malnutrition defined by BMI (Table 2).

Table 2
Table 2:
Frequency of malnutrition (as defined by a BMI < third percentile or the Pediatric Nutritional Risk Score) in children undergoing general anaesthesia (n = 979) – Bordeaux, France

Association between patient characteristics and malnutrition

In univariate analysis, compared with children with a BMI more than the third percentile, malnourished children were more often born prematurely (<37 GW) (22.4 vs 10.4%; P = 0.0008), were small for gestational age at birth (W < 10th percentile) (18.4 vs 4.5%; P < 0.0001), were admitted from the emergency department (12.0 vs 5.6%; P = 0.02), had a high ASA score (P < 0.0001) or had a high PNRS (P < 0.0001). Presence (P = 0.01) and type (P = 0.002) of chronic disease were also associated with malnutrition. Anthropometric data were similar in children with or without malnutrition (Table 3).

Table 3
Table 3:
Association between anthropometric data and presence of malnutrition (as defined by a BMI < third percentile) assessed by univariate analysis in children undergoing general anaesthesia

In the multivariate analysis, premature birth, lower birth weight and a higher PNRS were significantly associated with a higher odds of malnutrition as defined by BMI (Table 4).

Table 4
Table 4:
Association between patients’ clinical characteristics and presence of malnutrition (as defined by a BMI < third percentile) assessed by multivariate logistic regression in children undergoing general anaesthesia (n = 979 – Bordeaux, France)


The current study, conducted in a tertiary French hospital, shows that 11% of children attending a pre-anaesthesia assessment were malnourished when defined by a BMI below the third percentile, which is consistent with the prevalence of malnutrition in hospitalised children.1–6 A literature review conducted by Hankard et al.1 noted a prevalence ranging from 6 to 18%. Some studies showed that malnutrition was present in 25 to 30% of children in intensive care,12,30 but data on nutritional assessment in children undergoing general anaesthesia are scarce.

The prevalence of malnutrition is different depending on the methodology used to assign nutritional status.31 According to the Waterlow index, 6.1% of children in this population had acute malnutrition, which is similar to the prevalence found by Pawellek et al.3 in children hospitalised in general paediatric wards or paediatric surgery wards. Chronic malnutrition was present in 1.5% of children in our study, which is lower than the 7.7% of chronically malnourished hospitalised children reported in a Belgian multi-centre study published in which patients in intensive care were excluded.32 A Dutch national study conducted in 201033 showed that chronic malnutrition was more prevalent among children hospitalised in University Hospitals (14%) than in non-academics hospitals (6%), reflecting the difference in severity of diseases between different categories of paediatric care. Patients with severe diseases with a high risk of malnutrition, such as cancer and congenital heart diseases needing surgery16 or transplantation,34,35 are exclusively managed in University Hospitals. In a previous study, we showed that malnutrition was present in 15% of children with congenital heart disease, especially in cases of pulmonary hypertension.36 It should be noted that cardiac surgery is not performed in our hospital, and such children are not included in the current study. We found a fair correlation between BMI and Waterlow index with less than 5% discordance between these two parameters. On the contrary, the presence of clinical signs of malnutrition, which are only visible in cases of more advanced malnutrition, was poorly correlated with BMI and Waterlow index, leading to a nearly 10% discordance with each parameter. This indicates that clinical examination alone is not sufficient to detect malnutrition and highlights the importance of measuring weight and height, assessing the child's position on growth curves and calculating indices like the BMI and the Waterlow index. We found that the prevalence of malnutrition is higher when assessed by BMI as compared with the Waterlow index, a finding in agreement with the literature.37 Thus, BMI should be used to screen for malnutrition.1 The BMI is a simple score well known by hospital doctors and general practitioners, as it is also used to assess the overweight. In addition, standards for interpreting the BMI are present in child health books. Mid-upper arm circumference and triceps skinfold have been used in some studies,13 although these are usually used to assess the percentage of body fat. Using the triceps skinfold, Pawellek et al.3 found 17.2% of children to be malnourished compared with 24.1% using the Waterlow index. Among children with cancer, mid-upper arm circumference and/or triceps skinfold detect malnutrition more frequently than indices based on weight and height35; however, these measurements would be difficult to collect during the pre-anaesthesia assessment.

The current study confirms the value of the PNRS. Although this score was created to detect patients at risk of malnutrition during their hospital stay and theoretically requires 48 h of hospitalisation to be calculated,18 it is validated in the French population, it is reliable and it is quick to complete, unlike other scores such as the Subjective Global Nutritional Assessment,32 the Pediatric Yorkhill Malnutrition Score (PYMS)38 or the Screening Tool for Risk On Nutritional status and Growth (STRONG).33 During the pre-anaesthetic assessment, even in the absence of malnutrition, all children should be screened using the PNRS, a score that can be computed easily and rapidly during the outpatient visit. In this study, PNRS was computed in 99.4% of cases, whereas in a Belgian study, STRONG was recorded in 97.1% of cases32 and PYMS only in 72.3% of children.38

The current study found that prematurity and low birth weight were the only parameters independently associated with the presence of malnutrition. This has also been observed in critically ill children30 and in children with congenital heart diseases awaiting surgery.39 Surprisingly, the presence and type of underlying chronic diseases that were associated with the presence of malnutrition in the univariate analysis did not remain as significant factors in the multivariate analysis. This may, in part, be due to the number of categories for this factor and the low number of cases within some of the categories.


The current study shows that 11% of children undergoing anaesthesia are malnourished. In addition to the routine recording of height, weight and ASA score by anaesthesiologists, we suggest that screening for malnutrition or the risk of malnutrition using the BMI or PNRS should be routine. Identifying malnourished children, or children at risk of malnourishment, would enable the initiation of a programme of pre-operative and/or postoperative nutritional support or even that major surgery be postponed where possible to allow nutritional rehabilitation.

Acknowledgements relating to this article

Assistance with the study: none.

Financial support and sponsorship: none.

Conflicts of interest: none.

Presentation: preliminary data were presented at a poster session at the European Society for Pediatric Hepatology, Gastroenterology and Nutrition, 9 to 12 may 2015, Amsterdam.


1. Hankard R, Colomb V, Piloquet H, et al. Malnutrition screening in clinical practice. Arch Pediatr 2012; 19:1110–1117.
2. Campanozzi A, Russo M, Catucci A, et al. Hospital-acquired malnutrition in children with mild clinical conditions. Nutrition 2009; 25:540–547.
3. Pawellek I, Dokoupil K, Koletzko B, et al. Prevalence of malnutrition in paediatric hospital patients. Clin Nutr 2008; 27:72–76.
4. Joosten KF, Zwart H, Hop WC, et al. National malnutrition screening days in hospitalized children in the Netherlands. Arch Dis Child 2010; 95:141–145.
5. De Luca A, Piloquet H, Mansilla M, et al. Multicenter nutritional screening in hospitalized children. Arch Pediatr 2012; 19:545–546.
6. Marteletti O, Caldari D, Guimber D, et al. Malnutrition screening in hospitalized children: influence of the hospital unit on its management. Arch Pediatr 2005; 12:1226–1231.
7. Leite HP, Isatugo MK, Sawaki L, et al. Anthropometric nutritional assessment of critically ill hospitalized children. Rev Paul Med 1993; 111:309–313.
8. Hecht C, Weber M, Grote V, et al. Disease associated malnutrition correlates with length of hospital stay in children. Clin Nutr 2015; 34:53–59.
9. Sungurtekin H, Sungurtekin U, Balci C, et al. The influence of nutritional status on complications after major intraabdominal surgery. J Am Coll Nutr 2004; 23:227–232.
10. Kuzu MA, Terzioğlu H, Genç V, et al. Preoperative nutritional risk assessment in predicting postoperative outcome in patients undergoing major surgery. World J Surg 2006; 30:378–390.
11. Correia MITD, Waitzberg DL. The impact of malnutrition on morbidity, mortality, length of hospital stay and costs evaluated through a multivariate model analysis. Clin Nutr 2003; 22:235–239.
12. Chambrier C, Sztark F. French clinical guidelines on perioperative nutrition. Update of the 1994 consensus conference on perioperative artificial nutrition for elective surgery in adults. J Visc Surg 2012; 149:e325–e336.
13. Wessner S, Burjonrappa S. Review of nutritional assessment and clinical outcomes in pediatric surgical patients: does preoperative nutritional assessment impact clinical outcomes? J Pediatr Surg 2014; 49:823–830.
14. Forchielli ML, McColl R, Walker WA, et al. Children with congenital heart disease: a nutrition challenge. Nutr Rev 1994; 52:348–353.
15. Jadcherla SR, Vijayapal AS, Leuthner S. Feeding abilities in neonates with congenital heart disease: a retrospective study. J Perinatol 2009; 29:112–118.
16. Toole BJ, Toole LE, Kyle UG, et al. Perioperative nutritional support and malnutrition in infants and children with congenital heart disease. Congenit Heart Dis 2014; 9:15–25.
17. Secker DJ, Jeejeebhoy KN. Subjective global nutritional assessment for children. Am J Clin Nutr 2007; 85:1083–1089.
18. Sermet-Gaudelus I, Poisson-Salomon A-S, Colomb V, et al. Simple pediatric nutritional risk score to identify children at risk of malnutrition. Am J Clin Nutr 2000; 72:64–70.
19. Medoff-Cooper B, Irving SY, Marino BS, et al. Weight change in infants with a functionally univentricular heart: from surgical intervention to hospital discharge. Cardiol Young 2011; 21:136–144.
20. Rocha GA, Rocha EJM, Martins CV. The effects of hospitalization on the nutritional status of children. J Pediatr (Rio J) 2006; 82:70–74.
21. Ljungqvist O, van Gossum A, Sanz ML, et al. The European fight against malnutrition. Clin Nutr 2010; 29:149–150.
22. Cooper A, Jakobowski D, Spiker J, et al. Nutritional assessment: an integral part of the preoperative pediatric surgical evaluation. J Pediatr Surg 1981; 16 (Suppl 1):554–561.
23. Waterlow JC. Classification and definition of protein-calorie malnutrition. Br Med J 1972; 3:566–569.
24. American Society of Anesthesiologists. New classification of physical status. Anesthesiology 1963; 24:111.
25. Sempé M, Pédron G, Roy-Pernot MP. Auxologie: méthodes et sequences. Paris: Théraplix; 1979.
26. Classification commune des actes médicaux. 2016;
27. Cole TJ, Flegal KM, Nicholls D, et al. Body mass index cut offs to define thinness in children and adolescents: international survey. BMJ 2007; 335:194.
28. Rolland-Cachera MF, Cole TJ, Sempé M, et al. Body mass index variations: centiles from birth to 87 years. Eur J Clin Nutr 1991; 45:13–21.
29. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977; 33:159–174.
30. Hulst J, Joosten K, Zimmermann L, et al. Malnutrition in critically ill children: from admission to 6 months after discharge. Clin Nutr 2004; 23:223–232.
31. Joosten K, Hulst J. Prevalence of malnutrition in pediatric hospital patients. Curr Opin Pediatr 2008; 20:590–596.
32. Huysentruyt K, Alliet P, Muyshont L, et al. The STRONG(kids) nutritional screening tool in hospitalized children: a validation study. Nutrition 2013; 29:1356–1361.
33. Hulst JM, Zwart H, Hop WC, et al. Dutch national survey to test the STRONGkids nutritional risk screening tool in hospitalized children. Clin Nutr 2010; 29:106–111.
34. Chemaitilly W, Boulad F, Heller G, et al. Final height in pediatric patients after hyperfractionated total body irradiation and stem cell transplantation. Bone Marrow Transplant 2007; 40:29–35.
35. Smith DE, Stevens MC, Booth IW. Malnutrition at diagnosis of malignancy in childhood: common but mostly missed. Eur J Pediatr 1991; 150:318–322.
36. Blasquez A, Clouzeau H, Fayon M, et al. Evaluation of nutritional status and support in children with congenital heart disease. Eur J Clin Nutr 2016; 70:528–531.
37. Mei Z, Grummer-Strawn LM, Pietrobelli A, et al. Validity of body mass index compared with other body-composition screening indexes for the assessment of body fatness in children and adolescents. Am J Clin Nutr 2002; 75:978–985.
38. Gerasimidis K, Keane O, Macleod I, et al. A four-stage evaluation of the Paediatric Yorkhill Malnutrition Score in a tertiary paediatric hospital and a district general hospital. Br J Nutr 2010; 104:751–756.
39. Vaidyanathan B, Nair SB, Sundaram KR, et al. Malnutrition in children with congenital heart disease (CHD): determinants and short-term impact of corrective intervention. Indian Pediatr 2008; 45:541–546.
© 2017 European Society of Anaesthesiology