Abstract
Background
Acute gastrointestinal injury (AGI) is an underrecognized complication after cardiac surgery in children with congenital heart disease (CHD), particularly in those with low body weight. This study aimed to identify independent risk factors for postoperative AGI in underweight children (≤10 kg) undergoing CHD surgery.
Methods
This single‑center retrospective study enrolled 386 underweight children who underwent CHD surgery between October 2021 and July 2022. Patients were divided into AGI and non‑AGI groups based on postoperative occurrence. Multivariate binary logistic regression was performed to identify independent risk factors.
Results
Of the 386 enrolled patients, 91 cases (23.6%) developed clinically significant postoperative AGI. Multivariate analysis demonstrated that the minimum intraoperative rectal temperature (P=0.002) was an independent risk factor. In addition, patients with AGI had significantly longer duration of mechanical ventilation, longer intensive care unit (ICU) stay and longer overall hospital stay (all P<0.001). They also presented with higher transfusion requirements and a higher rate of extubation failure (P<0.001).
Conclusions
The minimum intraoperative rectal temperature is an independent risk factor for AGI in underweight children with CHD. Early identification of this factor facilitates risk stratification and guides targeted perioperative gastrointestinal protection strategies, thereby improving therapeutic outcomes in this high‑risk population.
Keywords: Congenital heart disease (CHD), underweight children, acute gastrointestinal injury (AGI), cardiopulmonary bypass (CPB), retrospective study
Highlight box.
Key findings
• Among underweight children (≤10 kg) undergoing surgery for congenital heart disease, clinically significant postoperative acute gastrointestinal injury (AGI) occurred in 23.6% of patients. Lower minimum intraoperative rectal temperature was independently associated with postoperative AGI.
What is known and what is new?
• AGI is an important but underrecognized complication after pediatric cardiac surgery, whereas evidence specifically concerning underweight children remains limited.
• This study identified lower minimum intraoperative rectal temperature as an independent factor associated with clinically significant postoperative AGI after adjustment for relevant perioperative variables.
What is the implication, and what should change now?
• Underweight children exposed to greater intraoperative hypothermia may warrant closer postoperative gastrointestinal monitoring. Careful perioperative temperature management should be considered, while prospective multicenter studies are needed to determine optimal temperature targets and whether temperature-directed interventions can reduce AGI.
Introduction
Congenital heart disease (CHD) remains one of the most common birth defects, and surgical intervention is often essential for survival and improved quality of life (1). However, postoperative complications, particularly those affecting the gastrointestinal system, can significantly increase morbidity and prolong hospital stay (2). Acute gastrointestinal injury (AGI) has emerged as a notable but underrecognized complication in children undergoing cardiac surgery, especially in those with low body weight, who are inherently more vulnerable due to immature organ systems and limited physiological reserves (3-7).
Despite advances in perioperative management, the incidence and risk profile of AGI in this specific population remain poorly characterized (8). Most existing studies have focused on adults or mixed-weight pediatric cohorts, leaving a gap in evidence regarding low-weight children with CHD (9,10). Identifying patient-specific and procedure-related risk factors is crucial for early risk stratification, timely intervention, and improved outcomes.
In this single-center retrospective study, we aimed to investigate the independent risk factors for postoperative AGI in underweight children with CHD. By analyzing perioperative clinical data, we sought to provide evidence to guide clinicians in identifying high-risk patients and implementing targeted preventive strategies. Our findings may contribute to better perioperative care and ultimately reduce the burden of gastrointestinal complications in this vulnerable population. We present this article in accordance with the STROBE reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0682/rc).
Methods
Study population and data collection
This single-center, retrospective study enrolled low-weight children (≤10 kg) who underwent CHD surgery at Fuwai Hospital, Chinese Academy of Medical Sciences, between October 1, 2021, and July 31, 2022. Children with severe preoperative comorbidities or systemic diseases (such as hematological diseases, immunological diseases, gastrointestinal diseases, etc.), as well as those receiving preoperative total parenteral nutrition (TPN) or with prolonged preoperative fasting, were excluded. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Fuwai Hospital of the Chinese Academy of Medical Sciences (No. 2023–2265), and the requirement for informed consent was waived due to its retrospective nature.
This study consists of four parts. The first part describes the baseline characteristics and laboratory parameters of the included children, including sex, age, body weight, and laboratory test results. The second part reviews intraoperative data in low‑weight children, including cardiopulmonary bypass (CPB) time, ultrafiltration volume, and nasopharyngeal temperature. The third part summarizes postoperative indicators and clinical outcomes, such as duration of mechanical ventilation and length of intensive care unit (ICU) stay. The fourth part uses regression analysis to test and adjust for variables that showed significant differences between groups.
AGI assessment
The definition and grading system for AGI were based on the criteria established by the European Society of Intensive Care Medicine in 2012 (4). All children were free of AGI prior to CPB-assisted cardiac surgery, and AGI was defined as a new-onset postoperative event. This grading system stratifies AGI into four grades: Grade I represents transient gastrointestinal dysfunction with identifiable and reversible etiology without impairment of organ function; Grade II is characterized by impaired digestion and absorption that fails to meet the body’s nutritional demands; Grade III refers to persistent gastrointestinal failure despite clinical interventions; Grade IV manifests as gastrointestinal failure complicated by dysfunction of distant organs, potentially accompanied by shock and multiple organ dysfunction.
According to the above criteria, patients were divided into two groups for comparison using 72 hours after cardiac surgery as the cutoff: the significant AGI group (Grade II or higher, n=91) and the non-significant AGI group (no AGI or only Grade I AGI, n=295). This grouping strategy aimed to identify risk factors associated with clinically meaningful adverse outcomes. Patients were categorized based on the time to first postoperative defecation, time to initiation of enteral nutrition, presence of gastrointestinal symptoms (e.g., gastrointestinal bleeding), and whether iatrogenic interventions targeting gastrointestinal manifestations were administered. Grade II or higher AGI indicates severe gastrointestinal dysfunction requiring medical intervention; in contrast, no AGI or Grade I AGI signifies transient, self-limiting dysfunction without the need for active intervention. This classification approach is consistent with that adopted in previous studies.
Cyanosis
Cyanosis was defined as a resting pulse oxygen saturation ≤90% upon patients’ admission to the operating room (11).
End-diastolic volume index
The end-diastolic volume index (EDVI) was calculated as the following formula: EDVI = End-diastolic volume/body surface area (BSA) (12).
Statistical analysis
Continuous variables are presented as mean ± standard deviation (SD) for normally distributed data or median with interquartile range (IQR) for non-normally distributed data, and were compared using the independent samples Student’s t-test or Mann-Whitney U test, respectively. Categorical variables are expressed as frequency (percentage) and were compared using the Pearson’s Chi-squared test or Fisher’s exact test, as appropriate.
Multivariable analysis was performed using binary logistic regression to adjust for potential confounding effects caused by demographic characteristics. Variables included in the multivariable model were selected in two ways: some were predefined factors such as age and body weight to control for confounders; others were derived from univariate analysis to identify independent risk factors. A two‑tailed P value <0.05 was considered statistically significant. All statistical analyses were conducted using SPSS software (version 25.0; SPSS Inc., Chicago, IL, USA).
Results
Baseline characteristics and preoperative information
A total of 386 children were included in this study (Table 1). Baseline characteristics were compared between patients who developed clinically significant postoperative AGI and those who did not. Among the 386 children, 91 (23.6%) developed AGI (AGI group), while 295 (76.4%) did not (non‑AGI group). The AGI group comprised 51 male patients (56.0%), compared with 147 male patients (49.8%) in the non‑AGI group. Baseline comparisons (Table 1) showed that patients who developed AGI had a younger age (median: 5.8 vs. 6.57 months, P=0.04), lower body weight (6.5 vs. 7.0 kg, P=0.002), shorter body length (65 vs. 67 cm, P=0.007) and smaller body surface area (0.33 vs. 0.35 m2, P=0.006). Children with cyanotic heart disease were overrepresented in the AGI group (36.26% vs. 21.02%, P=0.003). The Society of Thoracic Surgeons-European Association for Cardiothoracic Surgery Congenital Heart Surgery Mortality Score (STAT) congenital heart surgery mortality score was significantly higher in the AGI group (median: 2 vs. 1, P<0.001).
Table 1. Patient basic characteristics and baseline data.
| Variables | Non-AGI (n=295) | AGI (n=91) | P value |
|---|---|---|---|
| Sex (male) | 147 (49.83) | 51 (56.04) | 0.30 |
| Age (months) | 6.57 (4.30–10.07) | 5.8 (3.97–7.73) | 0.04 |
| Weight (kg) | 7 (5.8–8.4) | 6.5 (5.3–7.5) | 0.002 |
| Body length (cm) | 67 (62–73) | 65 (61–68) | 0.007 |
| BSA (m2) | 0.35 (0.30–0.40) | 0.33 (0.29–0.36) | 0.006 |
| Cyanotic | 62 (21.02) | 33 (36.26) | 0.003 |
| Pregnancy duration (weeks) | 39 (38–40) | 39 (38–40) | 0.47 |
| Premature birth | 33 (11.19) | 12 (13.19) | 0.60 |
| Weight (kg) | 3.2 (2.9–3.5) | 3.1 (2.7–3.4) | 0.11 |
| STAT | 1 (1–2) | 2 (1–2) | <0.001 |
| Predominant diagnosis | |||
| TGA | 40 (13.56) | 17 (18.68) | 0.23 |
| APVD | 28 (9.49) | 8 (8.79) | 0.84 |
| COA | 10 (3.39) | 0 (0) | <0.001 |
| ASD | 36 (12.20) | 11 (12.09) | 0.71 |
| VSD | 153 (51.86) | 37 (40.66) | 0.06 |
| PFO | 8 (2.71) | 2 (2.20) | 0.79 |
| PDA | 11 (3.73) | 3 (3.30) | 0.85 |
| Other CHD | 9 (3.05) | 13 (14.29) | <0.001 |
Data are presented as n (%) or median (interquartile range). AGI, acute gastrointestinal injury; APVD, anomalous pulmonary venous drainage; ASD, atrial septal defect; BSA, body surface area; CHD, congenital heart disease; COA, coarctation of aorta; PDA, patent ductus arteriosus; PFO, patent foramen ovale; STAT, Society of Thoracic Surgeons-European Association for Cardiothoracic Surgery Congenital Heart Surgery Mortality Score; TGA, transposition of the great arteries; VSD, ventricular septal defect.
In terms of lesion subtypes, no patients with coarctation of the aorta were found in the AGI group, whereas COA accounted for 3.39% of the non-AGI group (P<0.001). The proportion of other congenital heart defects was significantly higher in children with AGI (14.29% vs. 3.05%, P<0.001). Sex distribution, gestational duration, prematurity, birth weight, and the prevalence of transposition of the great arteries (TGA), anomalous pulmonary venous drainage (APVD), atrial septal defect (ASD), ventricular septal defect (VSD), patent foramen ovale (PFO) and patent ductus arteriosus (PDA) were comparable between groups (all P>0.05).
Preoperative laboratory and echocardiographic data are summarized in Table 2. Patients in the AGI group exhibited lower preoperative albumin (41.4 vs. 42.4 g/L, P=0.03), higher cystatin C (1.16 vs. 1.09 mg/L, P=0.008) and higher high-sensitivity C-reactive protein (hs-CRP) (0.24 vs. 0.18 mg/L, P=0.03). Serum iron was significantly reduced in the AGI group (8.62 vs. 10.92 μmol/L, P=0.04), and left ventricular end-diastolic diameter was smaller (26 vs. 28 mm, P=0.01). Other hematological, hepatic, renal, iron metabolism parameters and cardiac functional indices including LVEF and EDVI did not differ significantly between the two groups (all P>0.05).
Table 2. Patient perioperative information.
| Variables | Non-AGI (n=295) | AGI (n=91) | P value |
|---|---|---|---|
| Pre-Hb (g/L) | 113 (105–122) | 113 (102–124) | 0.74 |
| Pre-WBC (×109/L) | 8.81 (7.2–10.45) | 8.91 (6.68–10.43) | 0.55 |
| Pre-PLT (×109/L) | 329 (279–391) | 325 (281–390) | 0.96 |
| Pre-HCT (%) | 34 (31.5–36.8) | 34.3 (31.2–37.3) | 0.59 |
| Pre-PA (mg/L) | 136.04 (112.51–159.13) | 129.55 (108.9–153.97) | 0.18 |
| Pre-TP (g/L) | 59.6 (56.9–63.4) | 59.4 (56.2–63.4) | 0.24 |
| Pre-albumin (g/L) | 42.4 (40.5–44.4) | 41.4 (39.8–43.3) | 0.03 |
| Pre-ALT (IU/L) | 23 (17–35) | 24 (17–35) | 0.99 |
| Pre-AST (IU/L) | 50 (40–61) | 49 (42–61) | 0.82 |
| Pre-TBIL (μmol/L) | 6.11 (3.99–9.01) | 6.19 (4.29–10.35) | 0.37 |
| Pre-DBIL (μmol/L) | 1.67 (1.07–2.69) | 1.67 (0.9–2.64) | 0.73 |
| Pre-Cr (μmol/L) | 26.9 (21.09–31.52) | 28.03 (22.15–33.64) | 0.13 |
| Pre-BUN (μmol/L) | 3.05 (2.25–4.53) | 3.51 (2.25–4.73) | 0.30 |
| Pre-CysC (mg/L) | 1.09 (0.95–1.24) | 1.16 (1.01–1.31) | 0.008 |
| Pre-hs-CRP (mg/dL) | 0.18 (0.10–0.36) | 0.24 (0.15–0.54) | 0.03 |
| SI (μmol/L) | 10.92 (7.85–13.82) | 8.62 (6.87–12.41) | 0.04 |
| TIBC (μmol/L) | 56.33 (49.91–65.29) | 57.32 (48.60–64.33) | 0.71 |
| TSAT (%) | 18.26 (13.56–25.84) | 17.71 (12.26–22.36) | 0.13 |
| SF (μg/L) | 37.71 (19.19–79.96) | 39.22 (19.51–96.01) | 0.42 |
| TF (g/L) | 2.61 (2.33–2.93) | 2.58 (2.30–2.89) | 0.76 |
| TfR (nmol/L) | 14.41 (12.36–18.92) | 14.94 (12.54–21.01) | 0.52 |
| LVEF (%) | 70 (67–74) | 71 (67–75) | 0.43 |
| LVEDD (mm) | 28 (23–32) | 26 (19–31) | 0.01 |
| EDVI (mL/m2) | 80.31 (64.69–93.48) | 78.75 (60.02–91.29) | 0.48 |
Data are presented as median (interquartile range). AGI, acute gastrointestinal injury; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BUN, blood urea nitrogen; Cr, creatinine; CysC, cystatin C; DBIL, direct bilirubin; EDVI, end-diastolic volume index; Hb, hemoglobin; HCT, hematocrit; hs-CRP, high-sensitivity C-reactive protein; LVEDD, left ventricular end-diastolic diameter; LVEF, left ventricular ejection fraction; PA, prealbumin; PLT, platelet; SF, serum ferritin; SI, serum iron; TBIL, total bilirubin; TF, transferrin; TfR, transferrin receptor; TIBC, total iron binding capacity; TP, total protein; TSAT, transferrin saturation; WBC, white blood cell.
Intraoperative information
Intraoperative data analysis revealed that patients in the AGI group experienced significantly longer CPB time (94 vs. 82 min, P<0.001) and longer aortic cross-clamp duration (63 vs. 53 min, P=0.002). The minimum nasopharyngeal temperature during surgery was lower in the AGI group (P=0.01), and the minimum rectal temperature was substantially lower (32.5 vs. 33.5 ℃, P<0.001). Intraoperative infusion volume was higher among children with subsequent AGI (51.95 vs. 27.78 mL, P<0.001). Red blood cell volume used for CPB priming was larger (200 vs. 0 mL, P<0.001), and total blood consumption during CPB was also greater (400 vs. 200 mL, P=0.03). There were no meaningful differences in maximum Vasoactive-Inotropic Score (VIS), urine output before/during/after CPB, ultrafiltration volume, intraoperative blood loss or priming albumin dosage between the two groups (all P>0.05) (Table 3).
Table 3. Patient intraoperative information.
| Variables | Non-AGI (n=295) | AGI (n=91) | P value |
|---|---|---|---|
| Maximum VIS | 9 (9–11) | 10 (9–13) | 0.07 |
| Urine volume before CPB (mL) | 0 (0–5) | 0 (0–0) | 0.057 |
| Urine volume during CPB (mL) | 30 (5–90) | 35 (5–120) | 0.62 |
| Urine volume after CPB (mL) | 20 (10–40) | 20 (10–40) | 0.97 |
| Standard ultrafiltration (mL) | 200 (150–300) | 200 (150–300) | 0.81 |
| Modified ultrafiltration (mL) | 100 (100–150) | 100 (100–150) | 0.11 |
| Total ultrafiltration volume (mL) | 350 (300–450) | 350 (300–420) | 0.51 |
| Cardiopulmonary bypass time (min) | 82 (62–100) | 94 (72–119) | <0.001 |
| Aortic cross-clamp time (min) | 53 (41–73) | 63 (48–83) | 0.002 |
| Minimum nasopharyngeal temperature (℃) | 31 (30.6–32) | 31 (30–31.6) | 0.01 |
| Minimum rectal temperature (℃) | 33.5 (33.3–34.5) | 32.5 (31.7–33.3) | <0.001 |
| Blood loss (mL) | 10 (10–15) | 10 (10–15) | 0.50 |
| Infusion volume (mL) | 27.78 (16.67–51.28) | 51.95 (29.85–74.07) | <0.001 |
| Albumin used for priming (g/L) | 0 (0–0) | 0 (0–0) | 0.95 |
| Priming RBCs (mL) | 0 (0–200) | 200 (0–200) | <0.001 |
| CPB blood usage (mL) | 200 (200–200) | 400 (200–400) | 0.03 |
Data are presented as median (interquartile range). AGI, acute gastrointestinal injury; CPB, cardiopulmonary bypass; RBC, red blood cell; VIS, Vasoactive-Inotropic Score.
Postoperative information and outcomes
Postoperative laboratory testing showed that the AGI group had lower postoperative total protein (55.93 vs. 57.75 g/L, P<0.001), lower albumin (41.3 vs. 42.5 g/L, P=0.03), higher aspartate aminotransferase (157 vs. 133 U/L, P=0.01), lower total iron binding capacity (35.54 vs. 56.33 μmol/L, P=0.03), lower transferrin (1.68 g vs. 1.81 g/L, P=0.01) and lower transferrin receptor (11.03 vs. 12.58 nmol/L, P=0.03). Daily postoperative urine output was lower in the AGI group (P=0.02), and chest tube drainage volume on postoperative day 2 was higher (P=0.02). Crucially, time to first defecation after surgery was markedly prolonged in the AGI group (81.3 vs. 47 h, P<0.001), and the initiation time of enteral nutrition was also delayed (30 vs. 16 h, P<0.001). Other postoperative hematological, inflammatory and renal indices were similar between groups (all P>0.05) (Tables 4,5).
Table 4. Patient postoperative information.
| Variables | Non-AGI (n=295) | AGI (n=91) | P value |
|---|---|---|---|
| WBC (×109/L) | 9.32 (7.11–11.97) | 8.78 (6.97–11.19) | 0.32 |
| Hb (g/L) | 112 (103–121) | 113 (106–121) | 0.22 |
| HCT (%) | 32.74±4.03 | 33.35±3.46 | 0.20 |
| PLT (×109/L) | 173 (144–213) | 167 (123–200) | 0.10 |
| NT-proBNP (ng/L) | 581 (234–1433) | 742 (325–1537) | 0.08 |
| PCT (ng/mL) | 0.08 (0.06–0.12) | 0.09 (0.06–0.15) | 0.19 |
| PA (mg/L) | 127.11 (110.27–144.45) | 122.07 (105.75–135.71) | 0.12 |
| TP (g/L) | 57.75±4.72 | 55.93±4.83 | <0.001 |
| ALB (g/L) | 42.5 (40.2–44.7) | 41.3 (38.6–44.4) | 0.03 |
| ALT (U/L) | 21 (16–29) | 21 (15–28) | 0.81 |
| AST (U/L) | 133 (99–176) | 157 (107–205) | 0.01 |
| TBIL (μmol/L) | 17.605 (11.31–28.03) | 17.02 (9.81–25.75) | 0.43 |
| DBIL (μmol/L) | 6.17 (4.20–8.75) | 5.51 (3.59–8.12) | 0.49 |
| Scr (μmol/L) | 38.57 (32.49–45.37) | 39.66 (31.76–46.56) | 0.62 |
| BUN (mmol/L) | 8.13 (6.47–9.73) | 8.58 (6.6–10.52) | 0.15 |
| CysC (mg/L) | 1.19 (1.05–1.37) | 1.2 (1.06–1.41) | 0.42 |
| hs-CRP (mg/L) | 11.71 (11.2–12.27) | 11.88 (11.14–12.44) | 0.46 |
| SI (μmol/L) | 3.9 (2.92–6.07) | 3.82 (3.07–6.59) | 0.42 |
| TIBC (μmol/L) | 37.85±8.48 | 35.54±9.10 | 0.03 |
| TSAT (%) | 10.5 (7.425–17.475) | 12.17 (8.89–17.6) | 0.07 |
| SF (ng/mL) | 107.8 (76.54–170.59) | 124.49 (86.49–222.82) | 0.052 |
| TF (g/L) | 1.81±0.43 | 1.68±0.49 | 0.01 |
| TfR (nmol/L) | 12.58 (9.505–16.04) | 11.03 (8.61–14.42) | 0.03 |
| Urine output (mL) | |||
| Postoperative | 640 (520–750) | 600 (508–670) | 0.02 |
| Day 1 | 695 (615–825) | 765 (605–855) | 0.08 |
| Day 2 | 610 (490–720) | 620 (529–725) | 0.23 |
| Chest tube drainage volume (mL) | |||
| Postoperative | 51 (37–67) | 52 (35–93) | 0.27 |
| Day 1 | 20 (13–27) | 20 (12–31) | 0.77 |
| Day 2 | 9 (6–13) | 11 (7–17) | 0.02 |
| Time to first defecation (hours) | 47 (25–63.5) | 81.3 (74–96) | <0.001 |
| Time to initiation of enteral nutrition (hours) | 16 (11–30) | 30 (23–55) | <0.001 |
Data are presented as median (interquartile range) or mean ± standard deviation. AGI, acute gastrointestinal injury; ALB, albumin; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BUN, blood urea nitrogen; CysC, cystatin C; DBIL, direct bilirubin; Hb, hemoglobin; HCT, hematocrit; hs-CRP, high-sensitivity C-reactive protein; NT-proBNP, N-terminal pro-brain natriuretic peptide; PA, prealbumin; PCT, procalcitonin; Scr, serum creatinine; SF, serum ferritin; SI, serum iron; TBIL, total bilirubin; TF, transferrin; TfR, transferrin receptor; TIBC, total iron binding capacity; TP, total protein; TSAT, transferrin saturation; WBC, white blood cell.
Table 5. Patient outcomes.
| Variables | Non-AGI (n=295) | AGI (n=91) | P value |
|---|---|---|---|
| Duration of mechanical ventilation (hours) | 10 (6–24) | 23 (15–49) | <0.001 |
| Length of ICU stay (days) | 3 (1–5) | 4 (3–7) | <0.001 |
| Length of hospital stay (days) | 12 (9–15) | 16 (14–21) | <0.001 |
| Blood transfusion in ICU (mL/kg) | 0 (0–25) | 22.7 (0–39.2) | <0.001 |
| Failed extubation | 4 (1.35%) | 8 (8.79%) | <0.001 |
| Reoperation | 3 (1.02%) | 3 (3.30%) | 0.12 |
| ECMO | 0 | 1 (1.10%) | 0.24 |
| In-hospital mortality | 0 | 1 (1.10%) | 0.24 |
Data are presented as median (interquartile range) unless otherwise specified. AGI, acute gastrointestinal injury; ECMO, extracorporeal membrane oxygenation; ICU, intensive care unit.
Independent predictors of AGI
Multivariate binary logistic regression analysis was performed to screen independent risk factors for postoperative AGI. Variables that achieved statistical significance in univariate comparisons as well as clinically relevant confounding factors (age and body weight) were incorporated into the regression model. The candidate variables included age, body weight, STAT score, cyanotic heart disease, baseline pre-albumin, baseline hs-CRP, CPB duration and intraoperative minimum rectal temperature.
After multivariate adjustment, lower intraoperative minimum rectal temperature was the sole independent predictor for postoperative AGI [odds ratio (OR) =0.668, 95% confidence interval (CI): 0.518–0.861, P=0.002]. A one-degree Celsius rise in the minimal rectal temperature was associated with a 33.2% reduction in the odds of developing AGI, implying that deeper intraoperative hypothermia significantly increases the risk of AGI.
Other variables including age (P=0.41), body weight (P=0.26), STAT score (P=0.88), cyanosis (P=0.59), baseline albumin (P=0.88), baseline hs-CRP (P=0.12) and CPB time (P=0.73) failed to show independent predictive value for postoperative AGI (Table 6).
Table 6. Multiple logistic regression results.
| Variables | B | S.E. | Wald | df | P value | OR | 95% CI for OR | |
|---|---|---|---|---|---|---|---|---|
| Lower | Upper | |||||||
| Age | −0.048 | 0.058 | 0.689 | 1 | 0.41 | 0.953 | 0.850 | 1.068 |
| Weight | −0.156 | 0.138 | 1.277 | 1 | 0.26 | 0.856 | 0.653 | 1.121 |
| STAT | −0.046 | 0.313 | 0.022 | 1 | 0.88 | 0.955 | 0.517 | 1.764 |
| Cyanotic | −0.269 | 0.498 | 0.293 | 1 | 0.59 | 0.764 | 0.288 | 2.026 |
| Baseline albumin | −0.008 | 0.055 | 0.022 | 1 | 0.88 | 0.992 | 0.891 | 1.105 |
| Baseline hs-CRP | 0.157 | 0.102 | 2.370 | 1 | 0.12 | 1.170 | 0.958 | 1.428 |
| CPB time | −0.002 | 0.006 | 0.115 | 1 | 0.73 | 0.998 | 0.986 | 1.010 |
| Minimum rectal temperature | −0.403 | 0.130 | 9.685 | 1 | 0.002 | 0.668 | 0.518 | 0.861 |
| Constant | 14.433 | 5.094 | 8.027 | 1 | 0.005 | 1,854,969.886 | – | – |
CI, confidence interval; CPB, cardiopulmonary bypass; df, degrees of freedom; hs-CRP, high-sensitivity C-reactive protein; OR, odds ratio; S.E., standard error; STAT, Society of Thoracic Surgeons-European Association for Cardiothoracic Surgery Congenital Heart Surgery Mortality Score.
Discussion
This single-center retrospective analysis enrolled 386 children with CHD weighing ≤10 kg, and the overall incidence of clinically significant postoperative AGI was found to be 23.6%. Compared with previous studies focusing on mixed cohorts of pediatric cardiac patients, our data further demonstrated that AGI represents a prevalent yet underrecognized postoperative complication in underweight children with CHD. Patients complicated with AGI exhibited markedly prolonged duration of mechanical ventilation, longer ICU and total hospital stay, as well as increased transfusion requirements and a higher rate of extubation failure, indicating that gastrointestinal dysfunction substantially delays postoperative recovery and escalates the burden of critical care. Notably, no significant intergroup differences were observed in the rates of reoperation, extracorporeal membrane oxygenation (ECMO) implantation or in-hospital mortality. Against this clinical backdrop, we systematically screened risk factors for postoperative AGI in underweight children with CHD via retrospective analysis, so as to formulate targeted gut-protective strategies for this high-risk population.
Univariate analysis revealed significant intergroup discrepancies in multiple clinical variables at baseline, preoperatively, intraoperatively and postoperatively. At admission, patients with AGI were younger, presented inferior somatic growth indices, a higher proportion of cyanosis, and elevated STAT surgical risk scores, which reflected fragile physical development, chronic hypoxic injury and complex cardiac malformations. All patients with coarctation of the aorta fell into the non-AGI group, which could be attributed to small sample size of this lesion subtype and baseline selection bias. Preoperative laboratory tests demonstrated lower albumin and serum iron levels, alongside higher cystatin C and hs-CRP concentrations in the AGI cohort, suggesting insufficient nutritional reserve and persistent subclinical inflammation that predispose the intestinal tract to secondary injury after surgical trauma. During operation, prolonged CPB time, longer aortic cross-clamping duration, lower minimal core temperature, larger intraoperative infusion volume and increased red blood cell consumption further aggravated intestinal ischemia-reperfusion injury, visceral vasoconstriction and mucosal barrier disruption. Postoperatively, patients with AGI suffered from hypoproteinemia, deranged iron metabolism and elevated liver enzymes, accompanied by marked delays in the first defecation and initiation of enteral nutrition, manifesting comprehensive disorders of nutritional metabolism, hepatic function and intestinal peristalsis.
Multivariate binary logistic regression was subsequently performed to eliminate confounding factors and identify independent predictors of postoperative AGI. After adjusting for age, body weight, STAT score, cyanosis, baseline albumin, hs-CRP and CPB duration, only the minimal intraoperative rectal temperature was independently correlated with the development of AGI (OR =0.668, 95% CI: 0.518–0.861, P=0.002). Each 1 ℃ rise in the minimal intraoperative rectal temperature reduced the odds of postoperative AGI by 33.2%, which implied that excessively low intraoperative rectal temperature serves as a critical contributor to AGI in underweight children with CHD. Remarkably, all other adjusted variables lost statistical significance in the multivariate model, demonstrating that the detrimental intestinal effects of young age, malnutrition, preoperative inflammation and prolonged CPB could be offset or masked by intraoperative core temperature regulation.
Although induced hypothermia exerts vital neuroprotective effects during CPB, its adverse impacts on other visceral organs in underweight children with CHD deserve close attention. Children with low body weight possess immature thermoregulatory systems and a relatively high body surface area-to-weight ratio, rendering them prone to persistent hypothermia during and after CPB; hypothermia triggers systemic multi-organ damage through diverse pathophysiological pathways and specifically exacerbates structural and functional impairment of the gastrointestinal tract (6,13,14). Systemically, profound intraoperative hypothermia induces extensive vasoconstriction of peripheral and visceral microvessels, suppresses myocardial contractility and provokes arrhythmias, while impairing platelet aggregation to increase intraoperative blood loss and allogeneic transfusion volume. Sustained postoperative hypothermia inhibits hepatic and renal metabolic enzyme activity, leading to elevated liver enzymes and transient renal dysfunction, as well as markedly prolonged mechanical ventilation and ICU stay, thus raising the risk of composite complications (13-15). Large-scale cohort studies have confirmed that neonates with persistent hypothermia within the first 48 post-operative hours carry significantly higher risks of composite complications and 30-day mortality compared with normothermic counterparts (14). Cascade reactions including coagulopathy, hemodynamic instability and reduced visceral perfusion induced by perioperative hypothermia compound the inherent insufficient physiological reserve of underweight children and amplify systemic injury triggered by surgical trauma (16,17).
Underweight infants may have impaired intestinal development and a relatively vulnerable intestinal barrier. Profound intraoperative hypothermia further triggers persistent mesenteric vasospasm and directly reduces intestinal mucosal microcirculatory blood flow, resulting in sustained intestinal hypoperfusion (18). Clinical monitoring of rectal mucosal blood flow in infants undergoing CPB has verified that hypothermia leads to a marked decline in local intestinal perfusion, acting as a direct trigger for intestinal ischemia (19). Piglet CPB models have also demonstrated that compared with moderate hypothermia, deep hypothermic circulatory arrest significantly suppresses the expression of intestinal epithelial tight junction proteins and aggravates ischemia-reperfusion injury and intestinal hyperpermeability (20). Severe oxygen supply-demand imbalance during the rewarming phase robustly activates intestinal mast cells, which release pro-inflammatory mediators including TNF-α and tryptase to continuously disrupt the intestinal epithelial barrier and induce bacterial and endotoxin translocation, ultimately leading to clinical manifestations of AGI such as abdominal distension, feeding intolerance and gastrointestinal bleeding (6,20,21). Meanwhile, perioperative hypothermia disturbs the homeostasis of intestinal commensal flora by decreasing the abundance of beneficial bacteria and enriching opportunistic pathogens, which further amplifies systemic inflammatory response and forms a vicious cycle (22). Overall, perioperative hypothermia during cardiac surgery acts as a double-edged sword: while providing cerebral protection, it may exert adverse effects on other organs. Therefore, standardized perioperative temperature management is particularly essential for this special population of underweight children with CHD.
Advantages and limitations
First, the present study focused on underweight children with CHD, a special population characterized by immature gastrointestinal development and a high incidence of postoperative AGI. This overcomes the limitation of previous studies that generally enrolled all children with CHD without refined subgroup analysis. Second, we conducted both univariate analyses and multivariate logistic regression to adjust for multiple clinical confounding factors. For the first time, our findings suggest that intraoperative hypothermia may partly contribute to the association between low body weight and postoperative AGI; however, this potential mediating role requires confirmation by formal mediation analysis. Third, this study compared the value of body weight versus chronological age in reflecting gastrointestinal developmental maturity in children, and confirmed that body weight better indicates intestinal nutritional reserve and barrier function, which offers new insights for preoperative risk assessment. Fourth, the minimum intraoperative rectal temperature is a routinely monitored perioperative parameter that is easy to obtain at low cost and can be directly applied for clinical risk prediction of AGI, showing strong practical translational value. Fifth, our results highlighted the crucial effect of intraoperative temperature management on intestinal injury, providing preliminary clinical evidence for future targeted perioperative intervention research.
However, several limitations should be acknowledged. First, the single center retrospective design limits causal inference, and our findings may not be generalizable to institutions with different perioperative protocols. Second, AGI grading was based on clinical criteria, which may introduce subjectivity; nevertheless, we used standardized definitions to minimize this bias. Third, we did not assess postoperative feeding practices or the use of gastrointestinal prophylactic agents, both of which could influence AGI development. Fourth, residual confounding from unmeasured variables, such as preoperative nutritional status or intraoperative hemodynamic fluctuations, cannot be excluded. Fifth, the AGI grading system employed in this study was originally developed for adults, as no validated pediatric specific criteria currently exist for underweight children, which may introduce discrepancies between our conclusions and the true clinical picture. Sixth, our study excluded children with preoperative TPN or prolonged preoperative fasting, and the conclusions were only applicable to underweight pediatric patients who received routine cardiac surgery.
In terms of future research, multicenter prospective cohort studies can be carried out to verify the conclusions of this study. Mediation analysis may quantitatively assess the mediating effect of intraoperative hypothermia between low body weight and AGI and explore the safe intraoperative core temperature threshold for children of different weights. Meanwhile, a combined predictive model can be constructed using intestinal injury biomarkers, and further randomized controlled intervention trials can be performed to confirm the efficacy of individualized warming strategies in lowering the incidence of postoperative AGI. For clinical intervention, preoperative gastrointestinal risk stratification should be conducted according to body weight and growth status, with early nutritional intervention for growth‑retarded children. Intraoperative temperature management should be optimized to avoid unnecessary profound hypothermia in underweight children, and active warming measures should be applied to maintain intestinal microcirculatory perfusion. After surgery, intensified monitoring of gastrointestinal symptoms should be performed in children subjected to deep intraoperative hypothermia, alongside individualized early enteral nutrition and gut protection strategies, so as to reduce the risk of AGI across the whole perioperative course.
Conclusions
This study confirmed that the minimum intraoperative rectal temperature is an independent risk factor for postoperative AGI in underweight children with CHD, and AGI was associated with less favorable postoperative outcomes in these children. Clinicians should closely monitor and regulate body temperature during surgery and CPB. Targeted temperature management is expected to reduce the incidence of postoperative AGI and improve long‑term outcomes in this high‑risk population.
Supplementary
The article’s supplementary files as
Acknowledgments
None.
Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This work was approved by the Ethics Committee of Fuwai Hospital of the Chinese Academy of Medical Sciences (Beijing, China) (No. 2023–2265). Written informed consent was waived due to the retrospective nature of the study.
Footnotes
Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0682/rc
Funding: This work was supported by CAMS Innovation Fund for Medical Sciences (CIFMS) (No. 2024-12M-C&T-B-045) and Fundamental Research Funds for the Central Universities, Peking Union Medical College (No. 3332025040).
Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0682/coif). All authors report that this work was supported by CAMS Innovation Fund for Medical Sciences (CIFMS) (No. 2024-12M-C&T-B-045) and Fundamental Research Funds for the Central Universities, Peking Union Medical College (No. 3332025040). The authors have no other conflicts of interest to declare.
Data Sharing Statement
Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0682/dss
References
- 1. van der Linde D, Konings EE, Slager MA, et al. Birth prevalence of congenital heart disease worldwide: a systematic review and meta-analysis. J Am Coll Cardiol 2011;58:2241-7.. 10.1016/j.jacc.2011.08.025 [DOI] [PubMed] [Google Scholar]
- 2. Sacks D, Baxter B, Campbell BCV, et al. Multisociety Consensus Quality Improvement Revised Consensus Statement for Endovascular Therapy of Acute Ischemic Stroke: From the American Association of Neurological Surgeons (AANS), American Society of Neuroradiology (ASNR), Cardiovascular and Interventional Radiology Society of Europe (CIRSE), Canadian Interventional Radiology Association (CIRA), Congress of Neurological Surgeons (CNS), European Society of Minimally Invasive Neurological Therapy (ESMINT), European Society of Neuroradiology (ESNR), European Stroke Organization (ESO), Society for Cardiovascular Angiography and Interventions (SCAI), Society of Interventional Radiology (SIR), Society of NeuroInterventional Surgery (SNIS), and World Stroke Organization (WSO). J Vasc Interv Radiol 2018;29:441-53.. 10.1016/j.jvir.2017.11.026 [DOI] [PubMed] [Google Scholar]
- 3. Lin Y, Wang X, Zhang K, et al. Trajectories of acute gastrointestinal injury grade in critically Ill children. BMC Pediatr 2024;24:470.. 10.1186/s12887-024-04947-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Reintam Blaser A, Malbrain ML, Starkopf J, et al. Gastrointestinal function in intensive care patients: terminology, definitions and management. Recommendations of the ESICM Working Group on Abdominal Problems. Intensive Care Med 2012;38:384-94.. 10.1007/s00134-011-2459-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Zhou J, Ba Y, Du Y, et al. The Etiology of Neonatal Intensive Care Unit Death in Extremely Low Birth Weight Infants: A Multicenter Survey in China. Am J Perinatol 2021;38:1048-56.. 10.1055/s-0040-1701611 [DOI] [PubMed] [Google Scholar]
- 6. Mao QL, Yu ZH, Nie L, et al. Gastrointestinal injury in cardiopulmonary bypass: current insights and future directions. Front Pharmacol 2025;16:1542995.. 10.3389/fphar.2025.1542995 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Blakely ML, Lally KP, McDonald S, et al. Postoperative outcomes of extremely low birth-weight infants with necrotizing enterocolitis or isolated intestinal perforation: a prospective cohort study by the NICHD Neonatal Research Network. Ann Surg 2005;241:984-9; discussion 989-94.. 10.1097/01.sla.0000164181.67862.7f [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Bateman RM, Sharpe MD, Jagger JE, et al. 36th International Symposium on Intensive Care and Emergency Medicine : Brussels, Belgium. 15-18 March 2016. Crit Care 2016;20:94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Lu R, Yang B. Incidence and influencing factors of acute gastrointestinal injury after cardiac surgery. BMC Cardiovasc Disord 2023;23:437.. 10.1186/s12872-023-03475-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Hou L, Li J, Yan B, et al. Acute gastrointestinal injury in children on ECMO: an analysis of risk factors. Front Med (Lausanne) 2026;13:1786101.. 10.3389/fmed.2026.1786101 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Stout KK, Daniels CJ, Aboulhosn JA, et al. 2018 AHA/ACC Guideline for the Management of Adults With Congenital Heart Disease: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation 2019;139:e698-800.. 10.1161/CIR.0000000000000603 [DOI] [PubMed] [Google Scholar]
- 12. Foppa M, Duncan BB, Rohde LE. Echocardiography-based left ventricular mass estimation. How should we define hypertrophy?. Cardiovasc Ultrasound 2005;3:17.. 10.1186/1476-7120-3-17 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Nemeth M, Miller C, Bräuer A. Perioperative Hypothermia in Children. Int J Environ Res Public Health 2021;18:7541.. 10.3390/ijerph18147541 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Helman SM, Sereika S, Hravnak M, et al. Association Between Persistent Hypothermia After Cardiopulmonary Bypass in Neonates and Odds of Serious Complications. Crit Care Explor 2024;6:e1137.. 10.1097/CCE.0000000000001137 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Linardi D, Faggian G, Rungatscher A. Temperature Management During Circulatory Arrest in Cardiac Surgery. Ther Hypothermia Temp Manag 2016;6:9-16.. 10.1089/ther.2015.0026 [DOI] [PubMed] [Google Scholar]
- 16. Sessler DI. Perioperative thermoregulation and heat balance. Lancet 2016;387:2655-64.. 10.1016/S0140-6736(15)00981-2 [DOI] [PubMed] [Google Scholar]
- 17. Gupta B, Dodge-Khatami A, Tucker J, et al. Antegrade cerebral perfusion at 25 °C for arch reconstruction in newborns and children preserves perioperative cerebral oxygenation and serum creatinine. Transl Pediatr 2016;5:114-24.. 10.21037/tp.2016.06.03 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Booker PD, Prosser DP, Franks R. Effect of hypothermia on rectal mucosal perfusion in infants undergoing cardiopulmonary bypass. Br J Anaesth 1996;77:591-6.. 10.1093/bja/77.5.591 [DOI] [PubMed] [Google Scholar]
- 19. Booker PD, Davis AJ, Franks R. Gut mucosal perfusion in infants undergoing cardiopulmonary bypass: effect of preoperative captopril. Br J Anaesth 1997;79:14-8.. 10.1093/bja/79.1.14 [DOI] [PubMed] [Google Scholar]
- 20. Liang M, Feng K, Yang X, et al. Moderate Hypothermia Provides Better Protection of the Intestinal Barrier than Deep Hypothermia during Circulatory Arrest in a Piglet Model: A Microdialysis Study. PLoS One 2016;11:e0163684.. 10.1371/journal.pone.0163684 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Karhausen J, Qing M, Gibson A, et al. Intestinal mast cells mediate gut injury and systemic inflammation in a rat model of deep hypothermic circulatory arrest. Crit Care Med 2013;41:e200-10.. 10.1097/CCM.0b013e31827cac7a [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Salomon JD, Qiu H, Feng D, et al. Piglet cardiopulmonary bypass induces intestinal dysbiosis and barrier dysfunction associated with systemic inflammation. Dis Model Mech 2023;16:dmm049742.. 10.1242/dmm.049742 [DOI] [PMC free article] [PubMed] [Google Scholar]