CHD is a common congenital defect that occurs in about 1% of live births, with similar prevalence throughout the world. Reference Triedman and Newburger1 A significant proportion of these children need at least one corrective surgery during their lifetime. Cardiovascular surgery has significantly improved the survival of children with CHD.
Children with CHD often have a normal birthweight but are at risk for malnutrition due to increased metabolic demand and reduced calorie intake and therefore may fail to thrive. Reference Blasquez, Clouzeau and Fayon2–Reference Varan, Tokel and Yilmaz7 Malnutrition has been observed in children with CHD in both developed and developing countries. Reference Blasquez, Clouzeau and Fayon2,Reference Hassan, Albanna and Morsy6 Corrective CHD surgery has been shown to have a positive impact on catch-up growth for weight and height. Reference Ratanachu-Ek and Pongdara8,Reference Carmona, Hatanaka and Barbieri9 However, prior poor nutritional status may impair the outcomes of corrective surgery, leading to increased post-operative morbidity and mortality. The adverse effects of malnutrition on post-operative complications and outcomes have been demonstrated in several studies. Poor nutrition, as defined by low weight-for-age, Reference Ross, Latham and Joffe4,Reference Anderson, Beekman and Border10,Reference Ross, Radman and Jacobs11 low height for age, Reference Ross, Latham and Joffe4,Reference Ross, Radman and Jacobs11,Reference Lim, Lim and Moorakonda12 or low serum visceral protein levels, Reference Henry, Borasino and Ortmann13 was associated with post-operative morbidity including longer hours of mechanical ventilation, a longer length of ICU stay, and a longer duration of inotropic support. Furthermore, micronutrients deficiency, including vitamin D Reference Ye, Dong and Deng14 and iron, Reference Puri, Price and Spinner15 has been associated with adverse outcomes in CHD children.
Recognition of specific nutritional prognostic factors might lead to interventions that would improve outcomes. Although a complete nutritional assessment includes several aspects, anthropometric measurements and their transformation to relative indices of normal or malnutrition as well as selected laboratory testing could serve as the mainstay of the nutritional assessment of the child with CHD. The purpose of this study was to determine the pre-operative nutritional status of CHD children through anthropometric and biochemical measures and to identify specific parameters that predict poorer short-term post-operative outcomes including ICU length of stay, duration of mechanical ventilation, and duration of continuous inotropic support.
Methods
This was a prospective observational study including patients with CHD who underwent corrective operations between June 2020 and October 2021 at Shahid Rajaei Cardiovascular, Medical, and Research Center, Tehran, Iran. The study was approved by the Ethics Committee of National Nutrition and Food Technology Research Institute, Shahid Beheshti University of Medical Sciences, Tehran, Iran (the ethical committee code was IR.SBMU.nnftri.Rec.1399.017). Families read and signed an informed consent form. Eligibility criteria were age 6–24 months and an indication of surgical intervention. Exclusion criteria were patients with a history of prematurity (born before 37 weeks of pregnancy), low birthweight (birthweight < 2500 g), known genetic malformations (Turner, Down), and intrauterine growth retardation which could lead to malnutrition independent of CHD.
Data collection
Pre-operative demographic and clinical information collected on all patients included gender, age, and anatomic diagnosis. The infants’ pre-operative calories and macronutrients intake were estimated with the use of a 24-hour food recall questionnaire from their parents. A digital weighing scale with a length measuring rod was used for weight and length measurement. Mid-upper arm circumference was measured with the measuring tape at the midpoint between the scapula acromion and the ulna olecranon with the relaxed arm and was recorded to the nearest 1 mm. All measurements were done by one individual to minimise the inter-observer variability. Pre-operative anthropometric assessments were entered into the World Health Organization Anthro software (version 3.2.2) to calculate weight-for-age z score, length-for-age z score, or weight-for-length z score, and mid-upper arm circumference z score. In addition to pre-operative laboratory tests as part of the surgical clinical guidelines, serum total protein, vitamin D, iron, ferritin, and B-natriuretic peptide were measured as part of the research protocol. Operative data included cardiopulmonary bypass time, aortic cross-clamp time, and operation time.
The primary outcome variable assessed was the length of ICU stay. Secondary outcome variables included duration of mechanical ventilation, vasoactive-inotropic score in the first 24 hours after cardiac surgery (VIS 24), Reference Davidson, Tong, Hancock, Hauck, da Cruz and Kaufman16 and duration of inotropes use post-operatively.
Statistics
Data were analysed using SPSS software (version 25.0, SPSS Inc., Chicago, IL, United States of America). Shapiro–Wilk test was used for data normality determination. The normally distributed continuous variable was described as mean and standard deviation, and non-normally distributed continuous variables as the median and interquartile range. The categorical data were shown as frequency and per cent. Initially, based on clinical relevance, and previously reported associations, covariates were identified as potential confounders of the association between nutrition status and outcomes. Univariate regression analysis was used to select parameters associated with outcomes, where variables significant at p < 0.2 level in univariate level were entered into the multivariate model. The final model was chosen based on significant variables at p < 0.05 level. Due to its non-parametric distribution, outcomes were log-transformed before analysis to satisfy model assumptions. Next, the ICU length of stay has dichotomised into ICU length of stay ≤ 5 days or >5 days, and the differences in perioperative characteristics between the two groups were compared using the t-test or Mann–Whitney U-test for continuous variables and Pearson’s chi-squared test for categorical variables.
Results
Study population
Between April 2020 and October 2021, among CHD patients admitted to heart surgery, 120 children with a median age of 8 months were included in this study (Table 1). There were more male than female patients. The most prevalent CHD was ventricular septal defect (30.8%) and tetralogy of Fallot (20.8%) (Table 1). Four (11.8%) of the ventricular septal defect patients, two patients with transposition of the great arteries, one of the patients with tetralogy of Fallot, and one of the patients with single ventricle and transposition of the great vessels, had undergone palliation with a pulmonary artery banding prior to corrective surgery. None of the study patients had documented neurologic injury or abnormalities and did not require tracheostomy or home ventilation prior to surgery.
General and clinical characteristics of the study population
Data expressed as median (quartile 1, 3), mean ± standard deviation, or n(%).
ACC=aortic cross-clamping; ASD=atrial septal defect, BNP=B-natriuretic peptide; CAVSD=complete atrioventricular septal defect; CPB=cardiopulmonary bypass; COA=coarctation of the aorta; Cr=creatinine; DORV=double outlet right ventricle; ESR=estimated sedimentation rate; Hb=haemoglobin; HCT=haematocrit; HLHS=hypoplastic left heart syndrome; MS=mitral stenosis; PA=pulmonary atresia; PDA=patent ductus arteriosus; PLT=platelets; PS=pulmonary stenosis; STAT=Society of Thoracic Surgeons–European Association for Cardio-Thoracic Surgery Congenital Heart Surgery; SV=single ventricle; TGA=transposition of the great arteries; TOF=tetralogy of Fallot; VSD=ventricular septal defect; WBC=white blood cells.
Pre-operative nutritional variables
The pre-operative caloric intake of infants was about 103 kcal/kg. Pre-operatively, all patients had per oral intake. The mean weight-for-age z score of the enrolled children was −1.97 ± 1.36 (Table 2). Reduced anthropometric-related indices (z score ≤ −2) of the patients were observed by 50.8% (n = 61) by weight-for-age z score, 23.3% (n = 28) by length-for-age z score, 38.3.4% (n = 46) by weight-for-length z score, and 59.2% (n = 71) by mid-upper arm circumference z score. The mean serum total protein and 25 (OH) D levels of the study population were 59.36 ± 9.16 g/L and 29.09 ± 14.83 ng/mL, respectively. The serum total protein level was <60 g/L in 50% of children. The demographics of patients according to serum total protein are presented in Table 3. Serum 25(OH) D level was below 20 ng/mL in 25.0% of children. Low serum iron (<50 µg/dL) concentration was observed by 64.2% of the enrolled children. Pre-operative data of four common CHDs in the study population (ventricular septal defect, tetralogy of Fallot, ventricular septal defect-atrial septal defect, and atrial septal defect) are presented in Table 4.
Patients’ pre-operative nutritional status data
Data expressed as median (quartiles 1, 3), mean ± standard deviation, or n (%).
LAZ=length-for-age z score; MACZ=mid-upper arm circumference for age z score; WAZ=weight-for-age z score; WLZ=weight-for-length z score.
Demographics of patients according to serum total protein
Data expressed as median (quartiles 1, 3) or n (%).
ASD=atrial septal defect; CAVSD=complete atrioventricular septal defect; COA=coarctation of the aorta; DORV=double outlet right ventricle; HLHS= hypoplastic left heart syndrome; MS=mitral stenosis; PA=pulmonary atresia; PDA=patent ductus arteriosus; PS=pulmonary stenosis; STAT=Society of Thoracic Surgeons–European Association for Cardio-Thoracic Surgery Congenital Heart Surgery; SV=single ventricle; TGA=transposition of the great arteries; TOF=tetralogy of Fallot; VSD=ventricular septal defect.
Pre-operative data of four common CHDs in the study population
Data expressed as median (quartiles 1, 3), mean ± standard deviation, or n (%).
BNP = B-natriuretic peptide; LAZ=length-for-age z score; MACZ=mid-upper arm circumference for age z score; STAT=Society of Thoracic Surgeons–European Association for Cardio-Thoracic Surgery Congenital Heart Surgery; WAZ=weight-for-age z score; WLZ=weight-for-length z score.
Post-operative outcomes
The mean length of ICU stay of the enrolled patients was 5.4 days, and the mean length of mechanical ventilation of the study patients was 47.4 hours. The mean duration of inotropes post-operation was 3.3 days.
Effects of age-adjusted anthropometric z scores, serum biochemical values as well as age, sex, cardiopulmonary bypass time, aortic cross-clamp time, and operation time on outcomes were evaluated by linear regression (Table 5). Boys were the reference category for sex, and for anthropometric z scores, the reference category was z score >−2.0. Serum iron and ferritin levels as well as weight-for-age z score were not significant predictors of ICU length of stay in univariate analysis, and these were excluded from multivariate analysis. The multivariable analysis indicated that serum total protein level inversely (p = 0.004) and STAT category 2–4 (p = 0.018), length-for-age z score ≤ −2.0 category (p = 0.028), as well as mid-upper arm circumference z score ≤ −2.0 category (p = 0.036) positively were associated with an increased length of ICU stay. The R2 of the model was 0.299 (regression ANOVA p < 0.0001). Regarding the length of mechanical ventilation as the outcome, the multivariable analysis indicated patients with a lower serum protein (p = 0.004) or those with mid-upper arm circumference z score ≤ −2.0 (p = 0.001) were more likely to have prolonged mechanical ventilation (the R2 of the model = 0.203, and regression ANOVA p = 0.001). The assessment of the relationship between nutritional as well as confounding variables and the VIS 24 (the VIS value at the first 24 hours after surgery) showed that a minor proportion of the variance is explained by the model (the R2 = 0.071 and regression ANOVA p < 0.036) indicating that longer surgery time and patients with weight-for-length z score ≤ −2.0 were likely to have higher VIS 24. However, analysis of the relationship between variables and the duration of inotropic infusion (dopamine, milrinone, epinephrine, dopamine, and dobutamine) showed that a larger proportion of the variance of variance was explained by the model (the R2 = 0.286 and regression ANOVA p < 0.0001), suggesting that boys, those who were younger, those with higher STAT category, and mid-upper arm circumference z score ≤ −2.0, were likely to receive inotropes for a longer duration.
Univariate and multivariate linear modelling for predictors of outcomes
a Multivariate linear regression analysis was performed with factors selected from the univariate analysis with p < 0.2.
b STAT category 1 was set as the reference category.
ACC=aortic cross-clamp; BNP=B-natriuretic peptide; CPB=cardiopulmonary bypass; LAZ=length-for-age z score; MACZ=mid-upper arm circumference for age z score; STAT=Society of Thoracic Surgeons–European Association for Cardio-Thoracic Surgery Congenital Heart Surgery; WLZ=weight-for-length z score.
Patients with ventricular septal defect were the largest number of patients in the study. When these patients were grouped based on serum B-natriuretic peptide level into two groups with high B-natriuretic peptide (B-natriuretic peptide ≥ 100 pg/mL, n = 19) and low B-natriuretic peptide level (B-natriuretic peptide < 100 pg/mL, n = 15) and outcomes, as well as nutritional status indicators (z score categories, serum total protein, 25 [OH] D, calorie, and protein intake), were compared between the two groups, no significant difference was observed in any of the study outcomes or nutritional variables between the two groups which may be due to the small sample size for the analysis.
Table 6 shows the comparisons of anthropometric and laboratory parameters in patients with ICU length of stay ≤ 5 days and >5 days. Patients with ICU length of stay > 5 days were mostly boys and had longer aortic cross-clamp time, cardiopulmonary bypass, and operation time, but serum total protein was lower compared to patients with ICU length of stay ≤ 5 days. Furthermore, in patients with ICU length of stay > 5 days, age, mid-upper arm circumference z score ≤ −2.0 as well as serum 25 (OH) D and B-natriuretic peptide levels tended to differ than those with ICU length of stay ≤ 5.
Perioperative data in patients according to the length of ICU stay
Data expressed as the number of patients, median (quartiles 1, 3), or mean ± standard deviation.
ACC=aortic cross-clamp; BNP=B-natriuretic peptide; Cr=creatinine; CPB=cardiopulmonary bypass; ESR=estimated sedimentation rate; Hb=haemoglobin; HCT=haematocrit; LAZ=length-for-age z score; MACZ=mid-upper arm circumference for age z score; PLT=platelets; WAZ=weight-for-age z score; WLZ=weight-for-length z score.
Discussion
The aim of this study was to determine the nutritional status of patients prior to corrective cardiac surgery and examine the impact of nutritional status on the post-operative short-term outcomes. Anthropometric measures were used as indirect measures of nutritional status. Consistent with prior studies, this investigation demonstrates that malnutrition is prevalent in CHD patients. A high incidence of malnutrition according to weight-for-age z score ≤ −2 (50.8%) was observed in children undergoing CHD surgery. However, none of weight-for-age z score, or weight-for-length z score did predict any of the study’s post-operative morbidities suggesting that the applied criteria for malnutrition may not be sensitive enough to detect short-term outcomes or were confounded by other factors, including unmeasured ones. The results are consistent with the findings of Lim et al’s study in which pre-operative weight-for-age z score was not associated with short-term outcomes, including ICU length of stay and duration of mechanical ventilation, but pre-operative weight-for-age z score was associated with 30-day mortality. Reference Lim, Lim and Moorakonda12 In contrast, weight-for-age z score was associated with longer post-operative hospital stays in CHD children undergoing Glenn procedure Reference Anderson, Beekman and Border10 or Fontan Reference Wallace, Jaggers and Li17 operation or it was associated with duration of mechanical ventilation in Indonesian CHD children. Reference Marwali, Darmaputri, Somasetia, Sastroasmoro, Haas and Portman18 Furthermore, in a retrospective multicentre cohort study using data from the Society of Thoracic Surgeons Congenital Heart Surgery Database, lower values for weight-for-age z score and length-for-age z score were significantly associated with increased infection and longer hospitalisation after paediatric heart surgery in a diverse population of infants and young children. Reference Ross, Radman and Jacobs11
The pre-operative caloric intake of infants was about 103 kcal/kg. Since more than half of the studied infants had weight-for-age z score ≤ −2, and regarding the caloric goal of 120–130 kcal or higher in infants with growth retardation, Reference Hopkins, Marino and Shaw19 it seems that the amount of pre-operative calories intake by the study patients was inadequate to compensate for elevated energy expenditure in most cases. Thus, in addition to CHD, which itself contributes to malnutrition, it appears that inadequate calorie intake has also contributed to malnutrition of the study patients.
Mid-upper arm circumference z score identified more children in the undernourished category than weight-for-age z score, length-for-age z score, and weight-for-length z score, and patients with mid-upper arm circumference z score ≤ −2 at the time of the corrective operation had a longer ICU length of stay, mechanical ventilation, and duration of inotropic infusion. We could not find any study that used mid-upper arm circumference z score to diagnose malnutrition and post-operative complications in children with CHD. However, a study comparing the body mass index for age z score with mid-upper arm circumference z score in children with cystic fibrosis reported that mid-upper arm circumference z score identified a larger number of children with malnutrition compared to body mass index for age z score. Reference Phong, Taylor, Robinson, Jhawar and Nandalike20 Mid-upper arm circumference z score has been listed as an independent indicator for paediatric malnutrition and a more sensitive prognostic indicator for mortality than weight-for-height parameters in malnourished children. Reference Becker, Nieman Carney and Corkins21,Reference Stephens, Escobar and Jennison22 The observed difference between the prevalence of malnutrition assessed with mid-upper arm circumference z score and other anthropometric indices has already been reported, and it has been proposed that the z score ranges used to define various stages of malnutrition may not be the same for all indicators of malnutrition. Reference Stephens, Escobar and Jennison22 This may be due to the fact that each anthropometric parameter reflects the nutritional status differently. Mid-upper arm circumference may have a stronger association with fat mass than fat-free mass. Reference Chomtho, Fewtrell, Jaffe, Williams and Wells23,Reference Grijalva-Eternod, Wells and Girma24 Grijalva-Eternod et al examined the associations between mid-upper arm circumference and weight-for-length in infants with fat mass and fat-free mass and found that mid-upper arm circumference was more strongly associated with variability in adiposity relative to variability in fat-free mass. They proposed that mid-upper arm circumference may act more like a composite index of poor growth and wasting, including low adiposity. Reference Grijalva-Eternod, Wells and Girma24
Serum total protein level was a better predictor of some type of morbidity, particularly ICU length of stay and mechanical ventilation duration, than many other pre-operative variables. Pre-operative total serum protein level has been a predictor/prognostic factor in adult patients. Reference Watanabe, Kinoshita and Itoh25–Reference Czempik, Czepczor and Czok27 Malnutrition can lead to lower protein reserves, which may affect normal healing, and can hinder the ability to recover after surgical intervention. Serum total protein is an indication of the circulating amount of albumin and globulin. Serum albumin concentration in this study was not available to all patients and therefore was not included in the analysis. However, albumin accounts for more than half of the body’s serum total proteins. The presence of hypoalbuminaemia may be a sign of underlying malnutrition, associated inflammatory syndrome, and some other conditions. Reference Arques28 Serum albumin has been reported to be an independent predictor of clinical outcomes in critically ill children. Reference Leite, Rodrigues da Silva, de Oliveira Iglesias and Koch Nogueira29,Reference Kittisakmontri, Reungrongrat and Lao-Araya30 Furthermore, lower serum albumin levels have been associated with increased post-operative morbidity in patients undergoing operative correction of CHD. Reference Henry, Borasino and Ortmann13,Reference Leite, Fisberg, de Carvalho and de Camargo Carvalho31
There are limitations to this investigation that must be acknowledged. Results of the study must be interpreted with caution as only a small proportion of the variance is explained by the models provided. Furthermore, the patients’ follow-up was short and the sample size was small. No power calculation was performed to determine sample size since this was an observational study. However, the observed relationship between pre-operative serum total protein and mid-upper arm circumference z score and some of the outcomes warrants further investigation in a larger study. Moreover, patients younger than 6 months were not included in the study. We assumed that in older infants who could not undergo corrective surgery until the time of the study, the effect of congenital heart lesions on anthropometric measurements would be more obvious. Additionally, except for serum total protein, the concentrations of other visceral proteins such as albumin and pre-albumin were not measured. Moreover, the group of patients was heterogeneous, and it might have been preferable to study patients with similar defects in detail. Furthermore, the infant’s food intake at the pre-operative time was estimated with the use of food recall questionnaire from their parents and it should be noted that feeding data based on parental survey is vulnerable to a recall bias.
Conclusions
In conclusion, pre-operative identification of malnutrition by performing anthropometric and biochemical measurements including mid-upper arm circumference z score and serum total protein concentrations appears to be the predictors of the post-operative prolonged ICU stay, respiratory support, and the need for any inotropic infusion in children undergoing CHD surgery. These markers could be used as a guide to highlight patients needing appropriate perioperative nutritional interventions to improve nutritional status and potentially reduce post-operative morbidity. Future prospective studies investigating which nutritional interventions could be the most effective, particularly in patients with mid-upper arm circumference z score ≤ −2 and low serum protein, are required.
Acknowledgements
The authors would like to thank the children and parents who took part in the study.
Financial support
This investigation was funded by a project grant supported by the National Nutrition and Food Technology Research Institute, Shahid Beheshti University of Medical Science, Tehran, Iran.
Conflicts of interest
None.
Ethical standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national guidelines and with the Helsinki Declaration of 1975, as revised in 2008, and have been approved by the Ethics Committee of National Nutrition and Food Technology Research Institute, Shahid Beheshti University of Medical Sciences, Tehran, Iran (the ethical committee code was IR.SBMU.nnftri.Rec.1399.017).
