Severe underweight status predicts adverse postoperative outcomes after congenital heart surgery in children

ABSTRACT

Background:

Malnutrition significantly affects children with congenital heart disease (CHD). Severe underweight status may prolong postoperative recovery and increase complications after cardiac surgery. Identifying modifiable risk factors may help improve outcomes.

Patients and Methods:

Between January 2022 and December 2024, we included all children with a weight-for-age Z score (WAZ) <−3 who underwent palliative or corrective cardiac surgery (group 1). Demographic, laboratory, physiological, and clinical parameters were collected. Outcomes were compared with age- and risk-adjustment for congenital heart surgery-matched controls (group 2) with a normal WAZ (>−2). The severely underweight group was subdivided into two subgroups, complicated (1a) and noncomplicated (1b), to analyze potential risk factors.

Results:

A total of 170 children were included; 86 were severely underweight. Means WAZ and weight-for-height Z scores for group 1 were −4.4 ± 0.16 and −3.75 ± 0.43, respectively. The total complication rate was more in group (1), although it was not statistically different; cardiac and respiratory complications were higher in the severely underweight group. There were no differences in major cardiac events, unplanned interventions, or mortality. Severely underweight children had longer mechanical ventilation (MV), longer ICU and hospital stays, and more feeding difficulties.

Conclusions:

Severe underweight significantly increases postoperative morbidity after congenital heart surgery. Factors that elevate oxidative stress in this fragile population may further amplify surgical risk. Large-scale studies are essential to better identify and understand the contributing risk factors.

INTRODUCTION

Congenital heart disease (CHD) is one of the most common congenital anomalies in the pediatric population, accounting for 8–11 per 1000 live births.[1] Approximately 27%–90% of children with CHD are underweight.[1] Impaired growth in these children is multifactorial and may result from insufficient intake (fatigue and dyspnea), increased energy expenditure and hypermetabolism (tachypnea and tachycardia), malabsorption (due to elevated right-sided heart pressures, lower cardiac output, and altered gastrointestinal function), and difficulties coordinating breathing with swallowing.[1,2] Despite increased awareness of the importance of preoperative nutritional support, many children still undergo major cardiac surgery with suboptimal nutritional status.

According to the World Health Organization, severe malnutrition in children is defined as a weight-for-age Z score (WAZ) below −3.[3] Studies focusing on severely malnourished children in intensive care units (ICUs) have shown that they carry a higher risk of mortality compared with children who are mildly or moderately malnourished.[4]

Malnutrition imposes a significant burden in the intensive care setting. These children often demonstrate impaired humoral and cell-mediated immunity due to reductive adaptation. Reduced nutrient intake compromises protein synthesis, impairing tissue repair and immune response, which increases susceptibility to infection, prolongs hospitalization, and raises the risk of complications.[5] Malnutrition can also disrupt hormonal balance, resulting in increased inflammation and oxidative stress.[4]

In this study, we focus on postoperative outcomes of severely underweight children with CHD across different age groups. We aim to determine the incidence of major complications after cardiac surgery, identify possible risk factors, and suggest strategies that may support enhanced recovery after surgery in this patient population.

PATIENTS AND METHODS

This retrospective observational cohort study was conducted in the pediatric cardiac surgical ICU (CSICU) at Prince Sultan Cardiac Centre, Qassim, Saudi Arabia, between January 2022 and December 2024, after Institutional Review Board approval.

Children aged 1 month to 14 years who were admitted for cardiac surgery at the study period were included. Patients were divided into two groups: group (1) severely underweight children with WAZs <−3, and group (2) children with normal WAZs >−2. Both groups had comparable age distribution and risk adjustment for congenital heart surgery (RACHS) scores. All patients underwent palliative or corrective cardiac surgery. Neonates, preterm infants, children with other congenital anomalies, and those with specific syndromes were excluded.

Data were extracted from the hospital electronic database. Anthropometric measurements (weight, height/length) were used to calculate WAZ, height-for-age (HAZ), and weight-for-height Z scores using the PediTools electronic calculator. The prognostic nutritional index (PNI) was calculated as:

PNI = (10 × serum albumin [g/dL] + [0.005 × total lymphocyte count (10⁹/L)]).

All patients received a preoperative dietitian assessment with target calorie planning. Postoperatively, early enteral feeding was initiated when feasible; otherwise, parenteral nutrition was started on postoperative day 1. The target was full enteral feeding by postoperative days 4–5.

Additional collected variables included type of CHD, RACHS category, elective or emergency surgery, presurgery admission, type of CHD (cyanotic versus noncyanotic), and biventricular or single ventricle pathway, vasoactive-inotrope score, CPB time, and aortic cross-clamp time.

Postoperative complications (POCs) were classified according to Brown et al.[6] and included major cardiac events (cardiopulmonary resuscitation, chest reopening, and major hemorrhage requiring reexploration), unplanned interventions within 30 days, cardiac complications (arrhythmias, low cardiac output and extracorporeal membrane oxygenation), respiratory complications (reintubation, atelectasis, airway injury/stridor, and chylothorax, diaphragmatic paralysis), renal impairment requiring peritoneal dialysis, gastrointestinal complications (necrotizing enterocolitis and hepatic dysfunction), neurologic complications, feeding intolerance (NG tube requirement or large gastric residuals), nosocomial infection, and mortality.

The incidence of each complication was calculated as a percentage of total cases. Secondary outcomes included MV duration, ICU length of stay, and hospital length of stay. Results were expressed as mean ± standard error. Statistical analysis was performed using GraphPad software. P < 0.05 was considered significant.

Group 1 (severely underweight) was further subdivided into a complicated cases group (1a) and a noncomplicated (1b) subgroup. These subgroups were compared for potential risk factors (age, RACHS score, and CPB time) and secondary outcomes (MV hours, ICU, and hospital stay).

Relative risk (RR) was calculated for selected variables that were assumed to be associated with POCs, such as age below 1 year, off or on CPB, presurgery admission to intensive care, cCHD or noncyanotic CHD, single ventricle vs. two ventricle pathway, residual lesion postsurgery, and late presentation at the time of surgery. Variables on univariate analysis (P < 0.20) were included in the multivariate logistic regression model.

RESULTS

A total of 170 subjects were included. Group (1) consisted of 86 severely underweight children, and group (2) consisted of 84 children with WAZs >−2. The mean age of the two groups was 19 ± 3.17 and 24.9 + 2.9 (P = 0.27), with no statistically significant difference between the two groups.

Growth parameters (WAZ, HAZ, and weight-for-height Z scores) were significantly lower in group 1 [Table 1]. The mean WAZ in group 1 was −4.4 ± 0.16, and the mean weight-for-height Z score was −3.75 ± 0.43. The mean PNI in group 1 was 40.73 ± 0.52 and <50 in all cases within group 1, which was significantly lower compared with group 2. Both groups have similar RACHS categories, and there was no difference in CPB or cross-clamp times [Table 1].

Table 1.

Demographics, surgical data, and postoperative complications

Preoperative variable	Group 1 (n=86) mean±SEM	Control Group 2 (n=84) mean±SEM	P
Age (month)	19±3.17	24.9±2.9 (84)	0.27
weight (kg)	6.13±0.388	10.5±0.59	0.0001
Weight-for-age Z score	−4.4±0.16	−0.73±0.12	0.0001
Height (cm)	68.9±1.8	82.4±2.5 (78)	0.0001
Height-for-age Z score	−2.5±0.35	−0.14±0.35 (72)	0.0001
Weight-for-height Z score	−3.75±0.43	−0.88±0.32 (72)	0.0001
PNI	40.73±0.52	65.37±1.93 (43)	0.0001
Intraoperative variable
Inotropic score	10.21±0.71	6.6±0.5 (76)	0.0001
RACHS	2.2±0.07	2.05±0.05	0.08
RACHS-1	9	6
RACHS-2	52	70
RACHS-3	21	7
RACHS-4	4	0
CPB (min)	113±8	106±7.2	0.45
Cross clamp (min)	79±6.5	76±6.6	0.75
Off pump cases (%)	21/86 (24)	18/84 (21)	0.7
Postoperative variable
ICU stay (day)	10.33±1.8	6.2±0.57 (82)	0.03
Hospital length stay (day)	15.43±2.08	7.7±0.76 (82)	0.0007
MV duration (hour)	62.3±13.6	27±7.8	0.03

SEM: Standard error of mean, RACHS: Risk adjustment for congenital heart surgery, CPB: Cardiopulmonary bypass, ICU: Intensive care unit, PNI: Prognostic nutritional index, MV: Mechanical ventilation

Postcardiac surgery complications were reported more in severely underweight children compared with group (2), 49/86 (57%) versus 29/84 (34.5%) in group (2), although it is not statistically significant (P = 0.07). Major events occurred in 9.3% of group 1 and 6% of group 2. Unplanned interventions occurred in 3.5% of group (1) and in none of group (2), but still not statistically significant (P = 0.08).

Cardiac and respiratory complications were statistically significant in group (1) than in group (2), while other complication rates were similar. Infection rate and mortality did not differ between groups. Children in group 1 required higher inotropic support, had longer MV duration, and more feeding difficulties, and had longer ICU and hospital stays compared with group 2 [Tables 1 and 2].

Table 2.

Rate and percentage of complications in Groups 1 and 2

Outcome variable	Group 1 (n=86), n (%)	Group 2 (n=84), n (%)	P
Total complications	49/86 (57)	29/84 (34.5)	0.07
Cardiac complication	26 (30)	11 (30)	0.03
Arrythmia	16 (18)	8 (9.5)
LCOS	10 (11.6)	3 (3.5)
Major events	9 (9.3)	5 (6)	0.1
CPR	7 (8.1)	1 (1.2)
Reexploration	2 (2.3)	4 (4.7)
Unplanned intervention	3 (3.5)	0	0.08
Respiratory complication	55 (64)	20 (24)	0.0004
Chylothorax	8 (9)	7 (5)
Reintubation	11 (13)	2 (2.4)
Stridor	10 (12)	1 (1)
Lung collapse	6 (7)	3 (4)
Pulmonary hypertension	18 (21)	7 (8)
Diaphragm paralysis	2 (2.3)	0
Gastrointestinal complication	6 (7)	2 (2.4)	0.18
Renal complication	6 (7)	5 (6)	0.8
Feeding difficulties	6 (6.9)	0	0.02
Infection	11 (9.5)	8 (10)	0.5
Death	3 (3.5)	1 (1.2)	0.33

LCOS: Low cardiac output state, CPR: Cardiopulmonary resuscitation

Within group (1), 7 complicated cases were reported as off-pump cases and 42 on-pump cases, with P = 0.16, with odds ratio 0.59, 95% confidence interval (CI) (0.19–1.9). Subgroup (1a) (complicated cases) had significantly longer CPB time (P = 0.016), higher vasoactive-inotropic support (P = 0.0003), longer MV (P = 0.0012), and longer ICU and hospital stay (P = 0.005) than subgroup 1b [Table 3].

Table 3.

Comparison between complicated and noncomplicated subgroups 1a and 1b

Variables	Group 1a Severe underweight with complications (n=48)	Group 1b Severe underweight with no complications (n=37)	P
Age (months)	16.04±3.24	22.41±5.43	0.29
Weight (kg)	5.76±0.44	6.62±0.69	0.27
Z-score of weight	−4.45±0.25	−4.35±0.19	0.1
Height (cm)	68.04±2.2	70±3.23	0.6
Z score Ht	−2.39±0.55	−2.7±0.43	0.67
Wt/Ht Z score	−4.02±0.62	−3.45±0.62	0.06
PNI	40±0.73	41.9±0.63	0.37
RACHS	2.35±0.1	2.1±0.1	0.085
Off bypass	8 (49)	13 (37)	0.17
CPB duration (min)	128.22±9.8	89.9±12.01	0.014
Inotropic score	12.43±1.1	7.4±0.55	0.0003
MV duration (hour)	100.14±22.8	12.8±2.85	0.0012
ICU length of stay (days)	12.58±2.46	4.37±0.22	0.0045
Hospital length of stay (days)	18.04±2.81	8.17±1.32	0.0048

RACHS: Risk adjustment for congenital heart surgery, CPB: Cardiopulmonary bypass, ICU: Intensive care unit, PNI: Prognostic nutritional index, MV: Mechanical ventilation, Ht: Height, Wt: Weight

Univariate analysis and multivariate logistic regression of multiple potential risk factors showed that Children with cCHD and patients admitted to the ICU preoperatively had increased risk of complications (RR = 1.94, 95% CI 1.03–3.65 and RR = 1.64, 95% CI 1.16–2.33), respectively. CPB cases had an increased point estimate for risk (RR = 1.39), but this was not statistically significant (95% CI 0.94–2.05).

Multivariate logistic regression analysis demonstrated the same result that CPB was associated with an increased risk of complications; however, this association did not reach statistical significance. In contrast, cCHD and preoperative ICU admission were independently associated with a statistically significant increased risk of complications.

DISCUSSION

Cardiac surgery has significantly improved outcomes for children with CHD. However, despite advances in surgical techniques and perioperative care, postoperative morbidity and mortality remain a concern. Several risk factors contribute to POCs in pediatric cardiac surgery, including procedure complexity, CPB time, and patient-related factors such as age and nutritional status. While CHD complexity is not modifiable, some factors can represent possible modifiable contributing factors that may influence outcomes.

Malnutrition remains highly prevalent among children with CHD. A recent meta-analysis reported that 27.4% of children with CHD are underweight and 24.4% are stunted.[7] Ross et al.[8] reported even higher rates. Multiple studies have identified underweight status as an independent risk factor for increased mortality, longer MV, and prolonged ICU stay. Ross et al.[8] calculated that each unit decrease in WAZ score increased ICU stay by 0.22 days. Yilmaz Ferhatoglu et al.[1] also reported significantly longer MV duration in underweight patients. In our cohort, severely underweight children required significantly longer MV, ICU stays, and hospital stays.

POCs are known to drive secondary outcomes such as MV duration and length of stay. Brown et al.[9] showed that ICU length of stay was directly linked to complications. In our cohort, total complications were more frequent in Group 1 than in Group 2, although the difference was not statistically significant. Cardiac and respiratory complications, however, occurred significantly more often in severely underweight children and likely contributed to the longer MV and hospital stay.

Feeding difficulties were also more common in the underweight group. Although no patient required long-term parenteral nutrition, some required nasogastric feeding and were unable to achieve the target caloric intake by days 4–5. Consequently, this problem contributes to a prolonged hospital stay. Malnutrition can alter gastrointestinal function through reduced perfusion, exocrine insufficiency, villus atrophy, and increased intestinal permeability.[10] Postoperative CPB effects, such as reduced mesenteric perfusion, may further impair feeding tolerance.[11]

Mortality in our cohort (3%) was not statistically different between groups and aligns with reported congenital heart surgery mortality benchmarks (<4%).[12,13] Infection occurred exclusively among complicated cases, likely reflecting impaired immunity in undernourished children compounded by prolonged MV and ICU stay.

Some authors reported that when early adequate calories were achieved, malnourished children did not necessarily have worse ICU outcomes.[14] This supports the concept that pre- and intraoperative factors and postoperative organ dysfunction may be stronger determinants of outcome than preoperative weight alone.

In our study, subgroup analysis demonstrated that children with severe underweight who required preoperative ICU admission, had cCHD, or underwent CPB were more likely to experience postoperative cardiac surgery complications.

Severely malnourished children have limited metabolic reserves, reduced antioxidants, and impaired inflammatory control.[5,15] We hypothesize that factors compromising antioxidant defense mechanisms may predispose patients to increased POCs related to oxidative stress. A recently published meta-analysis on oxidative stress in patients with CHD revealed higher levels of oxidative stress in patients with cCHD compared to patients with acyanotic CHD (aCHD).[16] Considering that, severely underweight children with cCHD in our cohort exhibit poorer clinical outcomes compared to those with aCHD. In addition, children who are admitted to the ICU presurgery may have metabolic impairment and endocrine disturbances presurgery which may worsen postoperative oxidative stress, increasing the risk of POCs. Finally, insult of surgical trauma plus CPB-related inflammation may overwhelm these limited reserves in severely underweight children, leading to organ impairment and prolonged recovery [Figure 1]. Studies have shown that the severity of surgical stress may predict outcomes better than anthropometric indices alone.[17]

Figure 1.

Factors contributing to oxidative stress and postoperative complications in severely underweight children after cardiac surgery

This single-center study has limitations, including a relatively small sample size, limited control of confounders, and reliance on subgroup analysis which may limit identification of definitive risk factors. There is also a possibility of residual confounders that were not addressed in the Nonetheless, it highlights the increased burden of POCs in severely underweight children and identifies factors that may contribute to poorer outcomes after cardiac surgery, providing a basis for future research.

CONCLUSION

Cardiac surgery in severely underweight children poses a significant burden in the intensive care setting. This vulnerable population requires special attention. Larger, multicenter studies are needed to identify potentially modifiable risk factors in order to improve outcomes and enhance recovery.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.