Risk Factors and Microbiological Characterization of Healthcare Associated Infection Post Cardiac Surgery Among Children in Beijing, China Authors: Yan Xu, Lei Shen, Nan Ding, Yaobin Zhu, Yige Chen, Gejingwa Zhao, Zhiqiang Li Published: 2026 ABSTRACT ObjectiveHealthcare associated infection (HAI) in children after open-heart surgery contributes substantially to morbidity and mortality. We aimed to analyze the clinical and microbiological characteristics of HAI after pediatric cardiac surgery.MethodsThis single-center retrospective study was conducted in Beijing Children’s Hospital from July 1, 2022 to June 30, 2025 including children with congenital heart disease (CHD) who underwent open-heart surgeries.ResultsA total of 1206 patients were included in this study. The overall HAI rate was 13.6%. Multivariate logistic regression analysis found underlying diseases (odds ratio [OR] = 12.342; 95% confidence interval [CI]: 5.744–26.518; P < 0.001), moderate and severe preoperative pulmonary artery hypertension (pre-PAH) (OR = 3.141; 95% CI: 1.311–7.527; P = 0.010 and OR = 3.312; 95% CI: 1.407–7.797, P = 0.006), and aortic cross-clamping (ACC) time (OR = 1.026; 95% CI: 1.008–1.045; P = 0.004) were significantly associated with postoperative HAI. Additionally, no significant association was observed between glucocorticoid use and postoperative HAI in our study. Of the 238 organisms detected, bacteria predominated (55.9%, 133/238). Gram-negative bacilli accounted for 89.5% (119/133) of the bacterial isolates, and Klebsiella pneumoniae was the only multi-drug resistant organism (MDRO) identified.ConclusionOur study suggested underlying diseases, moderate and severe pre-PAH, and ACC time were independently associated with postoperative HAI. Klebsiella pneumoniae was the only MDRO among the most common gram-negative bacilli. Introduction Congenital heart disease (CHD) is the most common and severe birth defect in China, with a comprehensive incidence of 1%.1 Surgical treatment remains the primary therapeutic strategy, with approximately 70,000 patients with CHD undergoing surgery in China annually.2 Recently, a multicenter study in China revealed that the postoperative infection rates obtained from 15 intensive care units (ICUs) varied from 1.3% to 15%.3 Similarly, in children from high-income countries, healthcare associated infection (HAI) prevalence after CHD repair ranged from 2 to 15%.4,5 Postoperative infections have an adverse effect on clinical outcomes in pediatric patients after thoracotomy, resulting in prolonged mechanical ventilation (MV) time, extended hospital and ICU stays, increased morbidity, mortality and medical costs.4–7 Hence, recognizing and mitigating the factors that predispose patients to postoperative HAIs is imperative. Several factors may contribute to postoperative infection in children undergoing surgery for CHD. Previous studies identified younger age, surgical risk grade, prolonged cardiopulmonary bypass (CPB) time and aortic cross-clamping (ACC) time as risk factors.8,9 Most infections were caused by bacteria, predominantly gram-negative bacilli, and a few can be caused by fungi or viruses.10 However, data remain limited regarding the occurrence and impact of HAI in patients with CHD from low- and middle-income countries. Furthermore, to our knowledge, data on the microbiological features and the association between postoperative glucocorticoid use and HAI risk in pediatric CHD surgical patients remain limited, particularly in China. In the present study, we aimed to investigate the prevalence, pre- and intraoperative risk factors, as well as postoperative glucocorticoid administration and outcomes of HAIs following cardiac surgery in children with CHD from our hospital, one of the largest national centers for children’s health in China. In addition, the microbiological features of the causative pathogens were analyzed. Materials and Methods Study Design and Participants This single-center retrospective observational study was conducted at Beijing Children’s Hospital. The inclusion criteria were: (1) pediatric patients with an age limit of 18 years, (2) patients who underwent cardiac surgery between July 1, 2022 and June 30, 2025. The exclusion criteria were: (1) surgery without CPB, (2) any infection within 30 days before surgery, (3) postoperative infection occurred within 2 days of admission. Diagnostic Criteria for Specific Types of Infections Postoperative HAIs were defined as cases of HAI diagnosed after surgery, with no evidence that the infection was present or incubating at the time of surgery. In the present study, HAIs were retrospectively diagnosed independently by two doctors, according to the definitions of the Center for Disease Control and Prevention.11 The specific diagnostic criteria were as follows. Pneumonia The presence of new or progressive pulmonary infiltration accompanied by infectious symptoms and signs (such as newly developed fever, purulent sputum, increased white blood cell count, decreased oxygenation, and changes in respiratory mechanics) with positive respiratory tract etiological results (bacteria and fungi were identified by culture, while viruses and atypical pathogens were detected by PCR). Ventilator-associated pneumonia: pneumonia that occurred 48 h after mechanical ventilation or within 48 h after extubation. Bloodstream Infection The presence of at least 1 of the following signs or symptoms: fever (T > 38°C), chills, or hypotension and signs and symptoms and accompanied by positive blood culture results (2 or more blood cultures drawn on separate occasions). Gastrointestinal Infection The presence of at least 2 of the following signs or symptoms with no other recognized cause: nausea, vomiting, abdominal pain, fever (T > 38°C) with positive enteric pathogen results. Catheter-Related Bloodstream Infection Patients with vascular catheters in place or those who had catheters removed within 48 h of bacteremia or fungemia, accompanied by fever (T > 38°C), chills, hypotension, and at least one positive result from peripheral venous blood culture, with no other clear source of infection except the catheter. Central Nervous System Infection The presence of at least 1 of the following signs or symptoms with no other recognized cause: fever (T > 38°C), headache, stiff neck, meningeal signs, cranial nerve signs, or irritability accompanied by organisms cultured from cerebrospinal fluid. Urinary Tract Infection Patients exhibiting signs and symptoms consistent with urinary tract infection or systemic infection and not explainable by other causes, with positive urine culture. Differentiate Contamination from Colonization Whether in sterile or non-sterile sites, if the culture result is positive without accompanying infection-related manifestations, contamination was considered and not included in the infection statistics. If common colonizing organisms are cultured in non-sterile sites without infection-related manifestations, colonization is considered and not included in the infection statistics. Data Collection For all patients, the following data were retrospectively obtained from the hospital’s electronic medical record. The demographic characteristics: gender, age at surgery, and weight at surgery; Preoperative data: underlying diseases (eg, chromosomal abnormalities, developmental retardation, malnutrition, previous surgery and hospitalization history), preoperative pulmonary artery hypertension (Pre-PAH), and preoperative low cardiac output syndrome (Pre-LCOS); Intraoperative parameters: risk adjustment in congenital heart surgery-1 (RACHS-1) score, CPB time, and ACC time; Postoperative variables: postoperative glucocorticoids (Post-GCs) administration (intravenous dexamethasone for patients deemed at risk for postextubation upper airway obstruction 0.5 mg/kg every 24 hours until extubation or methylprednisolone for hemodynamic instability 2 mg/kg every 12 hours until circulatory stabilization), postoperative complications (eg, delayed chest closure, diaphragmatic folding, multiple tracheal intubations, and peritoneal dialysis), hospital stay time, ICU stay time, MV time and in-hospital mortality. Postoperative microbiological findings and antimicrobial susceptibility results were retrieved from the medical records. For surgical infection prophylaxis, cefuroxime sodium or cefmetazole sodium was given within 0.5 h before surgery to prevent infection and the course should be limited to within 48 h if there was no infection occurred.12 When patients present with clinical symptoms such as fever, elevated infection indicators, and unstable circulation, pathogens were identified through culture or PCR. The antimicrobial susceptibility pattern was interpreted according to Clinical Laboratory Standards Institute (CLSI 2025) guidelines.13 Ethical Review This study was approved by the Ethics Committee of Beijing Children’s Hospital (No. 2026‑E‑004‑R) and was conducted in accordance with the Declaration of Helsinki. Due to the retrospective nature of this study, the committee waived the requirement for informed consent. Patient data were anonymized before analysis, and confidentiality was rigorously maintained throughout. Statistical Analysis Statistical analysis was performed with SPSS v.25.0 (SPSS Inc., Chicago, IL, USA). All data in this study were non-normally distributed, continuous variables were described by the median (interquartile range [IQR]: P25, P75) and categorical variables were presented as frequency (%). For between group comparison, Mann–Whitney U-test (for continuous variables) and Chi-squared (χ2) test or Fisher’s exact test (for categorical variables) were used. Univariate logistic regression analysis was used to analyze the influence of various factors on the occurrence of HAI. In the above univariate analysis, each parameter with P ≤ 0.1 was further included in the multivariate logistic regression analysis to identify the independent predictors of infection after cardiac surgery. Receiver operating characteristic (ROC) curve was used to examine the variables, and the area under the curve (AUC) was calculated from the ROC curve. The results of the predictors are presented with the sensitivity and specificity. A P value < 0.05 was considered to be statistically significant. Results Clinical Characteristics and Outcomes Among 1206 CHD children (599 girls and 607 boys), there were 164 (13.6%) HAIs and 1042 (86.4%) controls. Their median age and weight at surgery were 25.0 (IQR: 9.0–50.0) months and 11.5 (IQR: 8.0–16.5) kg, respectively. A total of 111 (9.2%) patients had underlying diseases and 17 (1.4%) had postoperative complications. The number of patients had mild, moderate and severe Pre-PAH were 86 (7.1%), 129 (10.7%) and 141 (11.7%), respectively. A total of 9 (0.7%) patients had Pre-LCOS. Patients classified into RACHS-1 score 1, 2, 3, 4 and 5 were 256 (21.2%), 808 (67.0%), 127 (10.5%), 10 (0.8%) and 5 (0.4%), respectively. The total CPB and ACC time were 80 (63–98) min and 51 (36–69) min, respectively. In total, 555 (46.0%) patients were administered glucocorticoids after cardiac surgery (Table 1). The total length of hospital stay time, ICU stay time and MV time were 10 (7.0–14.0) d, 1 (1.0–4.0) d and 6 (4.0–29.5) h, respectively. A total of 3 (0.2%) deaths occurred in this study (Table 2). Table 1Comparison of Patient Characteristics Between HAI and Control GroupsCharacteristicsTotal (n = 1206)HAI (n = 164)Control (n = 1042)P ValueGender0.278& Male607 (50.3)89 (54.3)518 (49.7) Female599 (49.7)75 (45.7)524 (50.3)Age at surgery (months)25.0 (9.0–50.0)6 (3.0–13.0)30 (10.0–53.0)< 0.001^Weight at surgery (kg)11.5 (8.0–16.5)6.4 (4.9–8.3)12 (8.3–17.2)< 0.001^Underlying diseases111 (9.2)59 (36.0)52 (5.0)< 0.001&Postoperative Complications17 (1.4)11 (6.7)6 (0.6)< 0.001*Pre-PAH< 0.001& No850 (70.5)47 (28.7)803 (77.1) Mild86 (7.1)16 (9.8)70 (6.7) Moderate129 (10.7)47 (28.7)82 (7.9) Severe141 (11.7)54 (32.9)87 (8.3)Pre-LCOS< 0.001* No1197 (99.3)157 (95.7)1040 (99.8) Yes9 (0.7)7 (4.3)2 (0.2)RACHS-1< 0.001# Score-1256 (21.2)13 (7.9)243 (23.3) Score-2808 (67.0)102 (62.2)706 (67.8) Score-3127 (10.5)40 (24.4)87 (8.3) Score-410 (0.8)6 (3.7)4 (0.4) Score-55 (0.4)3 (1.8)2 (0.2)CPB time (min)80 (63.0–98.0)90 (79.0–115.0)78 (61.0–95.0)< 0.001^ACC time (min)51 (36.0–69.0)62 (51.0–84.0)50 (33.0–67.0)< 0.001^Post-GCs555 (46.0)67 (40.9)488 (46.8)0.153&Notes: and chi-square test; * adjusted chi-square test; ^ Mann–Whitney U-test; # Fisher exact method.Abbreviations: HAI, healthcare associated infection; HA, hospital acquired; PAH, pulmonary arterial hypertension; LCOS, low cardiac output syndrome; CPB, cardiopulmonary bypass; ACC, aortic cross-clamping; RACHS, risk adjustment in congenital heart surgery; GC, glucocorticoid. Table 2Comparison of Clinical Outcomes Between HAI and Control GroupsOutcomesTotal (n = 1206)HAI (n = 164)Control (n = 1042)P ValueHospital stay time (d)10 (7.0–14.0)21 (16.0–29.0)9 (7.0–12.0)< 0.001^ICU stay time (d)1 (1.0–4.0)6 (4.0–9.0)2 (1.0–3.0)< 0.001^MV time (h)6 (4.0–29.5)95 (46.0–166.0)6 (4.0–23.0)< 0.001^In-hospital mortality3 (0.2)2 (1.2)1 (0.1)0.05#Notes: ^ Mann–Whitney U-test; # Fisher exact method.Abbreviations: HAI, healthcare associated infection; ICU, Intensive Care Unit; MV, mechanical ventilation. Table 1Comparison of Patient Characteristics Between HAI and Control GroupsCharacteristicsTotal (n = 1206)HAI (n = 164)Control (n = 1042)P ValueGender0.278& Male607 (50.3)89 (54.3)518 (49.7) Female599 (49.7)75 (45.7)524 (50.3)Age at surgery (months)25.0 (9.0–50.0)6 (3.0–13.0)30 (10.0–53.0)< 0.001^Weight at surgery (kg)11.5 (8.0–16.5)6.4 (4.9–8.3)12 (8.3–17.2)< 0.001^Underlying diseases111 (9.2)59 (36.0)52 (5.0)< 0.001&Postoperative Complications17 (1.4)11 (6.7)6 (0.6)< 0.001*Pre-PAH< 0.001& No850 (70.5)47 (28.7)803 (77.1) Mild86 (7.1)16 (9.8)70 (6.7) Moderate129 (10.7)47 (28.7)82 (7.9) Severe141 (11.7)54 (32.9)87 (8.3)Pre-LCOS< 0.001* No1197 (99.3)157 (95.7)1040 (99.8) Yes9 (0.7)7 (4.3)2 (0.2)RACHS-1< 0.001# Score-1256 (21.2)13 (7.9)243 (23.3) Score-2808 (67.0)102 (62.2)706 (67.8) Score-3127 (10.5)40 (24.4)87 (8.3) Score-410 (0.8)6 (3.7)4 (0.4) Score-55 (0.4)3 (1.8)2 (0.2)CPB time (min)80 (63.0–98.0)90 (79.0–115.0)78 (61.0–95.0)< 0.001^ACC time (min)51 (36.0–69.0)62 (51.0–84.0)50 (33.0–67.0)< 0.001^Post-GCs555 (46.0)67 (40.9)488 (46.8)0.153&Notes: and chi-square test; * adjusted chi-square test; ^ Mann–Whitney U-test; # Fisher exact method.Abbreviations: HAI, healthcare associated infection; HA, hospital acquired; PAH, pulmonary arterial hypertension; LCOS, low cardiac output syndrome; CPB, cardiopulmonary bypass; ACC, aortic cross-clamping; RACHS, risk adjustment in congenital heart surgery; GC, glucocorticoid. Comparison of Patient Characteristics Between HAI and Control Groups Notes: and chi-square test; * adjusted chi-square test; ^ Mann–Whitney U-test; # Fisher exact method. Abbreviations: HAI, healthcare associated infection; HA, hospital acquired; PAH, pulmonary arterial hypertension; LCOS, low cardiac output syndrome; CPB, cardiopulmonary bypass; ACC, aortic cross-clamping; RACHS, risk adjustment in congenital heart surgery; GC, glucocorticoid. Table 2Comparison of Clinical Outcomes Between HAI and Control GroupsOutcomesTotal (n = 1206)HAI (n = 164)Control (n = 1042)P ValueHospital stay time (d)10 (7.0–14.0)21 (16.0–29.0)9 (7.0–12.0)< 0.001^ICU stay time (d)1 (1.0–4.0)6 (4.0–9.0)2 (1.0–3.0)< 0.001^MV time (h)6 (4.0–29.5)95 (46.0–166.0)6 (4.0–23.0)< 0.001^In-hospital mortality3 (0.2)2 (1.2)1 (0.1)0.05#Notes: ^ Mann–Whitney U-test; # Fisher exact method.Abbreviations: HAI, healthcare associated infection; ICU, Intensive Care Unit; MV, mechanical ventilation. Comparison of Clinical Outcomes Between HAI and Control Groups Notes: ^ Mann–Whitney U-test; # Fisher exact method. Abbreviations: HAI, healthcare associated infection; ICU, Intensive Care Unit; MV, mechanical ventilation. Patients in the HAI group were significantly younger than that in the control group (6.0 [3.0–13.0] months vs 30.0 [10.0–53.0] months, P < 0.001), and the weight was significantly lower in the HAI group than that in the control group (6.4 [4.9–8.3] kg vs 12.0 [8.3–17.2] kg, P < 0.001). The proportion of patients with underlying diseases (59 [36.0%] vs 52 [5.0%], P < 0.001), postoperative complications (11 [6.7%] vs 6 [0.6%], P < 0.001), moderate and severe Pre-PAH (47 [28.7%] vs 82 [7.9%], P < 0.001 and 54 [32.9%] vs 87 [8.3%], P < 0.001, respectively), Pre-LCOS (7 [4.3%] vs 2 [0.2%], P < 0.001), RACHS-1 score-3/4/5 (49 [29.9%] vs 93 [8.9%], P < 0.001) were all significantly different between the HAI and the control group. Also, the median length of CPB time (90.0 [79.0–115.0] vs 78.0 [61.0–95.0] min, P < 0.001), ACC time (62.0 [51.0–84.0] vs 50.0 [33.0–67.0] min, P < 0.001), hospital stay time (21.0 [16.0–29.0] vs 9.0 [7.0–12.0] d, P < 0.001), ICU stay time (6.0 [4.0–9.0] vs 2.0 [1.0–3.0] d, P < 0.001) and that of postoperative MV duration (95.0 [46.0–166.0] vs 6.0 [4.0–23.0] h, P < 0.001) were all significantly longer in the HAI group. However, no significant differences were observed between the HAI group and the control group in the rates of post-GCs (67 [40.9%] vs 488 [46.8%], P = 0.153) and deaths (2 [1.2%] vs 1 [0.1%], P = 0.05) (Table 1 and Table 2). Significant risk factors in the univariate analysis associated with HAI were age at surgery (OR = 0.948; 95% CI: 0.936–0.959; P < 0.001), weight at surgery (OR = 0.775; 95% CI: 0.737–0.816; P < 0.001), underlying diseases (OR = 10.698; 95% CI: 7.004–16.341; P < 0.001), postoperative complications (OR = 12.414; 95% CI: 4.525–34.053; P < 0.001), pre-PAH (OR = 2.302; 95% CI: 2.004–2.645; P < 0.001), pre-LCOS (OR = 23.185; 95% CI: 4.774–112.606; P < 0.001), CPB time (OR = 1.013; 95% CI: 1.007–1.019; P < 0.001), ACC time (OR = 1.025; 95% CI: 1.016–1.033; P < 0.001) and RACHS-1 score (OR = 2.872; 95% CI: 2.191–3.764; P < 0.001) (Table 3). Table 3Univariate and Multivariate Logistic Regression Analysis of Risk Factors for HAICharacteristicsUnivariate AnalysisMultivariate AnalysisOR (95% CI)P ValueOR (95% CI)P ValueGender1.2 (0.863–1.67)0.278 Male FemaleAge at surgery (months)0.948 (0.936–0.959)< 0.001Weight at surgery (kg)0.775 (0.737–0.816)< 0.001Underlying diseases10.698 (7.004–16.341)< 0.00112.342 (5.744–26.518)< 0.001Postoperative complications12.414 (4.525–34.053)< 0.001Pre-PAH2.302 (2.004–2.645)< 0.001 No Mild Moderate3.141 (1.311–7.527)0.010 Severe3.312 (1.407–7.797)0.006Pre-LCOS23.185 (4.774–112.606)< 0.001 No YesCPB time (min)1.013 (1.007–1.019)< 0.001ACC time (min)1.025 (1.016–1.033)< 0.0011.026 (1.008–1.045)0.004RACHS-12.872 (2.191–3.764)< 0.001 Score-1 Score-2 Score-3 Score-4 Score-5Post-GCs0.831 (0.605–1.207)0.153Abbreviations: HAI, healthcare associated infection; PAH, pulmonary arterial hypertension; LCOS, low cardiac output syndrome; CPB, cardiopulmonary bypass; ACC, aortic cross-clamping; RACHS, risk adjustment in congenital heart surgery; GC, glucocorticoid. Table 3Univariate and Multivariate Logistic Regression Analysis of Risk Factors for HAICharacteristicsUnivariate AnalysisMultivariate AnalysisOR (95% CI)P ValueOR (95% CI)P ValueGender1.2 (0.863–1.67)0.278 Male FemaleAge at surgery (months)0.948 (0.936–0.959)< 0.001Weight at surgery (kg)0.775 (0.737–0.816)< 0.001Underlying diseases10.698 (7.004–16.341)< 0.00112.342 (5.744–26.518)< 0.001Postoperative complications12.414 (4.525–34.053)< 0.001Pre-PAH2.302 (2.004–2.645)< 0.001 No Mild Moderate3.141 (1.311–7.527)0.010 Severe3.312 (1.407–7.797)0.006Pre-LCOS23.185 (4.774–112.606)< 0.001 No YesCPB time (min)1.013 (1.007–1.019)< 0.001ACC time (min)1.025 (1.016–1.033)< 0.0011.026 (1.008–1.045)0.004RACHS-12.872 (2.191–3.764)< 0.001 Score-1 Score-2 Score-3 Score-4 Score-5Post-GCs0.831 (0.605–1.207)0.153Abbreviations: HAI, healthcare associated infection; PAH, pulmonary arterial hypertension; LCOS, low cardiac output syndrome; CPB, cardiopulmonary bypass; ACC, aortic cross-clamping; RACHS, risk adjustment in congenital heart surgery; GC, glucocorticoid. Univariate and Multivariate Logistic Regression Analysis of Risk Factors for HAI Abbreviations: HAI, healthcare associated infection; PAH, pulmonary arterial hypertension; LCOS, low cardiac output syndrome; CPB, cardiopulmonary bypass; ACC, aortic cross-clamping; RACHS, risk adjustment in congenital heart surgery; GC, glucocorticoid. After adjusted confounding factors, our study found that underlying diseases (OR = 12.342; 95% CI: 5.744–26.518; P < 0.001), moderate and severe pre-PAH (OR = 3.141; 95% CI: 1.311–7.527; P = 0.010 and OR = 3.312; 95% CI: 1.407–7.797; P = 0.006), and ACC time (OR = 1.026; 95% CI: 1.008–1.045; P = 0.004) were significantly associated with HAI in CHD children (Table 3). ROC curves were constructed to examine the performance of pre-PAH, ACC time, and underlying diseases as predictors of HAI (Figure 1). The AUC was calculated for each significant variable: 0.775 for pre-PAH (95% CI 0.719 to 0.831), 0.693 for ACC time (95% CI 0.639 to 0.747), and 0.692 for underlying diseases (95% CI 0.623 to 0.762). The ROC curve of ACC time showed a cut-off point of 46.5 min, which had 87.6% sensitivity and 54.9% specificity in predicting infection (Table 4). Figure 1Receiver operating characteristic curves for predicting healthcare associated infection risk in children with CHD.A line graph titled, ROC Curve. Legend entries: Pre-PAH, ACC time, Underlying diseases, All, Reference line. Horizontal axis label, 1 minus Specificity, unit not shown, range 0.0 to 1.0. Vertical axis label, Sensitivity, unit not shown, range 0.0 to 1.0. Reference line series: (0.0, 0.0), (1.0, 1.0). Text printed inside the plot area: Pre-PAH: 0.775 (0.719-0.831); ACC time: 0.693 (0.639-0.747); Underlying diseases: 0.692 (0.623-0.762); All: 0.870 (0.832-0.909).A line graph showing receiver operating characteristic curves for multiple predictors and a reference line. Table 4AUC of Variables and Critical Values for Predicting HAIVariablesROC-AUC (95% CI)P ValueCut-Off ValueSensitivity (%)Specificity (%)Pre-PAH0.775 (0.719–0.831)< 0.001ACC time (min)0.693 (0.639–0.747)< 0.001≥ 46.587.654.9Underlying diseases0.692 (0.623–0.762)< 0.001All0.870 (0.832–0.909)< 0.001Abbreviations: AUC, area under the curve; HAI, healthcare associated infection; ROC, receiver operating characteristic; PAH, pulmonary arterial hypertension; ACC, aortic cross-clamping. Receiver operating characteristic curves for predicting healthcare associated infection risk in children with CHD. Table 4AUC of Variables and Critical Values for Predicting HAIVariablesROC-AUC (95% CI)P ValueCut-Off ValueSensitivity (%)Specificity (%)Pre-PAH0.775 (0.719–0.831)< 0.001ACC time (min)0.693 (0.639–0.747)< 0.001≥ 46.587.654.9Underlying diseases0.692 (0.623–0.762)< 0.001All0.870 (0.832–0.909)< 0.001Abbreviations: AUC, area under the curve; HAI, healthcare associated infection; ROC, receiver operating characteristic; PAH, pulmonary arterial hypertension; ACC, aortic cross-clamping. AUC of Variables and Critical Values for Predicting HAI Abbreviations: AUC, area under the curve; HAI, healthcare associated infection; ROC, receiver operating characteristic; PAH, pulmonary arterial hypertension; ACC, aortic cross-clamping. Distribution of Pathogens and Antimicrobial Susceptibility Results Among the 164 patients in the HAI group, a total of 170 postoperative HAIs were identified: one patient experienced three HAIs during a single ICU stay, and four patients experienced two HAIs. The incidence rates of infection by site were as follows: pulmonary infection 83.5% (142/170; comprising 90 ventilator-associated pneumonia and 52 clinically diagnosed pneumonia), bloodstream infection 11.2% (19/170), gastrointestinal infection 2.9% (5/170), catheter-related bloodstream infection 1.2% (2/170), central nervous system infection 0.6% (1/170), and urinary tract infection 0.6% (1/170) (Figure 2A). Figure 2Pathogen distribution in postoperative healthcare associated infection of CHD children. (A) The incidence rates of infection by site; (B) Rates of single and mixed pathogen infections; (C) Pathogen type distribution.Image A: Pie chart with slices labeled 83.5%, 11.2%, 2.9%, 1.2%, 0.6% and 0.6%. Legend: pulmonary, bloodstream, gastrointestinal, catheter-related bloodstream, central nervous system and urinary tract infections. Image B: Two pie charts linked by a callout. Left chart slices: 34.7%, 28.8%, 27.6%, 5.9%, 2.9%. Legend: Single bacteria, virus, fungus, atypical pathogen, mixed infection. Right chart slices: 11.8%, 7.6%, 3.5%, 2.4%, 1.8%, 1.2%, 0.6%. Legend: Mixed bacteria, bacteria and fungus, bacteria and virus, bacteria, virus and atypical pathogen, virus, virus and fungus, bacteria, virus and fungus. Image C: Two pie charts linked by a callout. Left chart slices: 55.9%, 29.0%, 11.3%, 3.8%. Legend: Bacteria, viruses, fungi, atypical pathogen. Right chart slices: 89.5%, 10.5%. Legend: Gram-negative and Gram-positive bacteria.Different types of data visualizations such as three pie charts labeled A to C, with percentages from 0.6 to 89.5. Pathogen distribution in postoperative healthcare associated infection of CHD children. (A) The incidence rates of infection by site; (B) Rates of single and mixed pathogen infections; (C) Pathogen type distribution. Of these 170 episodes, the infection patterns were as follows: 34.7% (59/170) single bacterial infection, 27.6% (47/170) single viral infection, 5.9% (10/170) single fungal infection, 2.9% (5/170) single atypical pathogen infection, 11.8% (20/170) mixed bacterial infection, 7.6% (13/170) mixed bacterial and fungal infection, 3.5% (6/170) mixed bacterial and viral infection, 2.4% (4/170) mixed bacterial, viral, and atypical pathogen infection, 1.8% (3/170) mixed viral infection, 1.2% (2/170) mixed viral and fungal infection, and 0.6% (1/170) mixed bacterial, viral, and fungal infection (Figure 2B). A total of 238 pathogens were detected, including 55.9% (133/238) bacteria, 29.0% (69/238) viruses, 11.3% (27/238) fungi, and 3.8% (9/238) atypical pathogens. Of the 133 bacterial isolates, 89.5% (119/133) were gram-negative bacilli. The five most frequently isolated gram-negative bacteria were Pseudomonas aeruginosa (24.4%, 29/119), Acinetobacter baumannii (17.6%, 21/119), Klebsiella pneumoniae (15.1%, 18/119), Enterobacter cloacae complex (11.8%, 14/119), and Stenotrophomonas maltophilia (9.2%, 11/119). Gram-positive bacteria accounted for 10.5% (14/133) of all bacterial isolates, and Streptococcus pneumoniae was the most common species, representing 50.0% (7/14) of the gram-positive strains. Among the 69 viruses detected, the three most frequently identified were respiratory syncytial virus (RSV) (29.0%, 20/69), human parainfluenza virus (26.1%, 18/69), and influenza A virus (10.1%, 7/69). Of the 27 fungi recovered, Candida albicans predominated (74.1%, 20/27). All nine atypical pathogens were identified as Mycoplasma pneumoniae (Figure 2C and Supplementary Table 1). Among the five predominant gram-negative bacteria detected, only Klebsiella pneumoniae was identified as a multidrug-resistant organism (MDRO). It exhibited a relatively high overall resistance rate to β-lactam antibiotics (over 50%), with a resistance rate of 38.9% to piperacillin-tazobactam, while remaining relatively susceptible to carbapenems and tetracyclines. Streptococcus pneumoniae, the most common gram-positive isolate, showed relatively high resistance to carbapenems and sulfamethoxazole, yet it was susceptible to ceftriaxone, levofloxacin, vancomycin and linezolid (Supplementary Table 2). In this study, Candida albicans, the predominant fungal isolate, was susceptible to all commonly used antifungal agents, including 5‑fluorocytosine, amphotericin B, itraconazole, and fluconazole. Discussion China has the largest number of individuals with congenital heart disease (CHD) in the world and a heavy burden of CHD.2 In-hospital infections in children with open-heart surgery contribute substantially to postoperative mortality, morbidity, and prolonged hospitalization.7 As far as we know, literature on postoperative HAI and their specific microbiological features in pediatric CHD patients from low- and middle-income countries, including China, remains limited. Our study showed that the incidence of HAI in our center was 13.6%, which was basically consistent with recent studies from China. In 2026, Jiao et al14 reported that the postoperative infection rate was 11.6% from China. Furthermore, a multicenter study conducted in China reported that postoperative infection rates ranged from 1.3% to 15%,3 which is lower than previous result,15 and suggesting that the HAI management system implemented across ICUs in China after the COVID-19 pandemic may have led to measurable progress. There was a significant association between HAIs and prolonged MV, as well as extended ICU and hospital length of stay, similar to previous studies.4,5,9 The relationship between the occurrence of HAIs and patient outcomes remains challenging to assess, as sicker patients often require prolonged ICU stays as well as MV time and are exposed to extrinsic factors, such as the presence of MDRO in the environment, during which they may acquire an HAI. However, HAIs themselves can directly prolong the duration of MV, ICU stay and overall hospital length of stay. Previous analyses have shown that postoperative infection after congenital heart surgery was associated with increased mortality.7,16,17 However, in the present analysis, we did not find a relationship between postoperative infection and increased mortality. The world symposium on Pulmonary Hypertension classified CHD-PAH into the first category, which has been proven to be closely related to the impairment in functional class, quality of life, and survival.18 In this study, we found that moderate to severe pre-PAH was associated with HAIs. A previous study found that patients were younger in CHD-PAH group. Also, CHD-PAH could significantly increase the length of ICU stay, ventilator days and central venous catheterization days, thus increasing the prevalence of HAIs.19 On the other hand, local inflammatory reaction in the lung could induce spasm and contraction of pulmonary arterioles, aggravating pulmonary hypertension.20 Therefore, it’s usually thought that the presence of HAI in CHD-PAH is closely associated with poor prognosis. Further research are warranted to compare the clinical characteristics and prognosis of children with or without PAH and with different severities of PAH. In this study, univariate analysis showed that both prolonged CPB and ACC times were associated with an increased risk of HAI, while only ACC time remained an independent predictor in the multivariate model. CPB can induce a systemic inflammatory response syndrome, leading to tissue and organ damage as well as impaired immune function, which may increase susceptibility to HAIs.21 Moreover, a previous study22 also identified prolonged ACC time as an independent predictor of postoperative complications. One possible explanation is that prolonged CPB time can often be well managed, whereas prolonged ACC time may reflect difficulties in weaning from CPB, making it a more reliable marker of complication risk. We found that the coexistence of underlying diseases was related to postoperative infection. Previous research also demonstrated that underlying diseases genetic abnormality is a risk factor for postoperative HAIs.5,23 As expected, having a syndrome was found to be an independent risk factor in these patients owing to immunodeficiencies and developmental disorders.21 Malnutrition is also an important risk factor for HAI after cardiac surgery. Yu et al9 reported that BMI < 5th percentile was significantly associated with increased odds of nosocomial infection after cardiac surgery in CHD children. Meanwhile, a history of surgery and hospitalization increases the risk of exposure to HAI pathogens, and thus raises the probability of HAI. Several centers have described the use of hydrocortisone postoperatively to improve hemodynamics in infants and children with low cardiac output and hypotension.24,25 Other studies have reported a benefit from the use of periextubation dexamethasone to prevent postextubation upper airway obstruction.26 Previous works have demonstrated contrasting evidence for overall infections with intraoperative steroid prophylaxis. A randomized controlled trial reported that the risk of developing a postoperative infection was lower in the dexamethasone group than in the placebo group. This protective effect was primarily related to improved pulmonary condition, manifested as a lower risk of postoperative respiratory failure, shorter times to weaning from the ventilator, and reduced risk of pneumonia during postoperative hospitalization in the dexamethasone group.27 Whereas another study reported that patients who had postoperative infections received a median of 7 days of corticosteroid exposure consisting primarily of intraoperative methylprednisolone and multiple doses of postoperative hydrocortisone or dexamethasone or both.28 In this study, we found no significant association between postoperative glucocorticoid use and HAI risk. However, given the likely influence of confounding by indication, prospective randomized studies are needed to further clarify this issue. We found gram-negative bacteria (89.5%) were the most common cause in HAI, with Pseudomonas aeruginosa (24.4%), Acinetobacter baumannii (17.6%) and Klebsiella pneumoniae (15.1%) being the most prevalent. Extended-Spectrum β-Lactamase-producing Klebsiella pneumoniae ranked first among MDRO. Our results consistent with a recent multicenter study, Klebsiella pneumoniae exhibiting the highest frequency of MDRO among the total of 16 ICUs in china,3 which is also consistent with the current literature on MDRO in the worldwide.29 Also, Murni et al12 reported that gram-positive infections were still the most common cause of postoperative sepsis in high-income countries, while gram-negative sepsis was prevalent in children with postoperative bloodstream infections in low- to middle-income countries. In our center, Klebsiella pneumoniae isolates showed high susceptibility to carbapenems, while resistance rates to most β-lactam antibiotics exceeded 50%, with the exception of piperacillin-tazobactam. Given the limited number of isolates and the single-center nature of our data, these susceptibility patterns should be interpreted as reflecting local epidemiology. Consistent with previous multicenter research reports,30,31 Streptococcus pneumoniae was the most prevalent gram-positive bacteria among children with CHD. Lack of routine immunization against Streptococcus pneumonia in China is a significant contributor to its frequency, particularly in children with CHD. In China, PCV13 is self-paid, not covered by China’s national immunization program, and the immunization rates are generally low.32 In our center, Streptococcus pneumoniae isolates exhibited high resistance rates (>85%) to sulfamethoxazole, meropenem, and ceftazidime, while remaining highly susceptible to ceftriaxone, piperacillin-tazobactam, cefoperazone-sulbactam, vancomycin and linezolid. However, these findings should also be interpreted cautiously, and further multicenter studies are needed. This study showed that the detection rate of viruses in children after CHD surgery is, with RSV (29.0%), Human parainfluenza virus (26.1%) and Influenza (11.5%) being the most prevalent viruses. Also, other multicenter study conducted in Egypt detected that RSV 63.8%, Rhinovirus 10%, and Influenza 9% were the most common viruses causing lower respiratory tract infections in children.33 Different results were detected in Senegal where Adenovirus (50%), Influenza (45.7%), Rhinovirus (40%), Enterovirus (25.3%) and RSV (16%) were the most common.34 Studies have shown that the detection rate of acute respiratory viral infections in children is related to the demographics, the intensity of virus prevalence, the studied groups, and the seasonal variability in different regions. Worldwide RSV is an important cause of lower respiratory tract infections among children and those with CHD are more vulnerable to severe infection.35 In many countries, children with CHD receive RSV prophylaxis (eg, palivizumab) as part of national guidelines.36,37 In China, nisevir received approval in July 2024 and launched in the Beijing market in September 2024. However, high cost with limited government funding makes access difficult for most families. This study has some limitations. First, it was an observational analysis whose results support an association and not necessarily causation. Second, retrospective data were collected from a single center and might represent bias that may affect the present results. Third, some biomarkers of infection and organ function, such as CRP, PCT, alanine transaminase, as well as serum creatinine were not measured and analyzed. Forth, the lack of external validation is an important limitation of the study. Multicenter external validation is needed to confirm the predictive utility of the factors identified in our study. Conclusions In this single-center cohort, HAI was common after pediatric cardiac surgery, with respiratory infections predominating. Gram-negative organisms were the main pathogens, and Klebsiella pneumoniae was the most common MDRO. We identified moderate-to-severe Pre-PAH, prolonged ACC time, and underlying diseases as independent risk factors for HAI. Further prospective multicenter studies are warranted to validate these findings and guide prevention strategies.