Preoperative iron status and its association with adverse outcomes in pediatric congenital heart disease Authors: Huayan Shen, Qiyu He, Sirui Zhou, Wenke Li, Yuze Liu, Zhou Zhou, Shoujun Li Published: 2026 ABSTRACT AbstractBackground:Iron imbalance, including iron deficiency and overload, is associated with increased cardiovascular morbidity and mortality. However, the iron status and its association with postoperative adverse outcomes in pediatric patients with congenital heart disease (CHD) is unclear. This study aimed to investigate the preoperative iron status and its correlation with adverse outcomes in pediatric patients with CHD.Methods:This is a single-center retrospective study. A total of 8065 pediatric patients (aged 1 month to 5 years) who underwent surgical treatment at Fuwai Hospital’s Pediatric Cardiac Surgery Center between 2017 and 2022 were consecutively included. All patients had undergone comprehensive preoperative laboratory tests, including iron-related hematology and biochemistry examinations. The association between ferritin and adverse outcomes was examined using restricted cubic splines. A multivariable logistic regression model was used to determine the associations between preoperative iron status and postoperative adverse outcomes.Results:Iron deficiency, anemia, and iron deficiency anemia were observed in 23.0% (1856/8065), 12.0% (969/8065), and 7.0% (564/8065) of patients, respectively. Patients with cyanotic CHD had a higher prevalence of iron deficiency but a lower incidence of anemia compared with patients with acyanotic CHD. A right-skewed U-shaped relationship was observed between ferritin and death and composite adverse events. Multivariable logistic regression analysis demonstrated that patients with high ferritin levels (ferritin ≥100 µg/L) had a significantly higher risk of in-hospital death (odds ratio [OR]: 8.20, 95% confidence interval [CI]: 1.61–41.86]) and composite adverse events (OR: 4.21 [95%CI: 2.65–6.68]) compared with patients with iron repletion (ferritin levels of ≥15 and <33 µg/L). In addition, patients with iron deficiency (OR: 1.80 [95%CI: 1.18–2.73]) and intermediate ferritin levels (ferritin levels of ≥33 and <100 µg/L; OR: 1.64 [95%CI: 1.11–2.44]) had a higher incidence of composite adverse events compared with those with iron repletion.Conclusions:Iron deficiency and anemia are common concern in children with CHD. Preoperative iron imbalance, including both iron deficiency and high iron load, was significantly associated with postoperative adverse outcomes in pediatric patients with CHD. Assessment of preoperative iron status might facilitate early identification of high-risk patients, and correcting iron imbalance could potentially mitigate adverse outcomes. Introduction Iron deficiency is globally the most prevalent micronutrient deficiency in children, with a prevalence of 10–30% in children aged 6 months to 5 years from different regions.[1] It accounts for approximately 25% of anemia cases in children[2] and is linked to developmental delays, behavioral disturbances, impaired immune function, and early childhood caries.[3,4] Iron imbalance, including iron deficiency and overload, has been reported to be associated with increased cardiovascular morbidity and mortality,[5] highlighting the significant role of iron metabolic disorders in cardiovascular diseases. Several studies have elucidated the impact of iron status on prognosis in pediatric patients. Iron deficiency is associated with a greater risk of ventricular assist device implantation, heart transplantation, or death in children with heart failures.[6] Ferritin is a widely used biomarker for evaluating iron status. It is a positive acute phase protein and its levels are increased in response to inflammatory conditions,[7] and its elevation may increase the risks of critical care and subsequent death in the pediatric population.[8] Congenital heart disease (CHD) is a predominant birth defect and a substantial contributor to the cardiovascular disease burden in China.[9,10] It is one of the leading causes of death among children under 5 years old.[11] Iron status in pregnant women is associated with CHD risks in offspring.[12] Iron deficiency has been shown to independently increase the risk of postoperative blood transfusion in infants who undergo congenital cardiac surgery.[13] However, the iron status and its association with postoperative adverse outcomes in pediatric patients with CHD are unclear. In this study, we aimed to provide a comprehensive report on the iron status of pediatric patients with CHD based on data from a large Chinese cohort and investigate the correlations between preoperative iron status and early postoperative adverse outcomes. Methods Ethical approval This study was approved by the Ethics Committee of Fuwai Hospital (Approval No. 2023–2191). Informed consent from participants was waived due to the retrospective nature of the study. Study population We reviewed 16,278 patients who were hospitalized and underwent surgical treatment at the Pediatric Cardiac Surgery Center of Fuwai Hospital between January 2017 and December 2022. Patient data were extracted from electronic medical records. Data were double-checked by two authors and confirmed by a senior pediatric cardiac surgeon. Children aged between 1 month and 5 years who had undergone at least one comprehensive preoperative laboratory test, including iron-related hematology and biochemical examinations, were included. We primarily focused on patients within this age interval because: (1) Most biomarkers exhibit significant age-related fluctuations, and newborns inherently have high ferritin levels;[14] (2) most patients admitted to our pediatric cardiac surgery center were aged less than 5 years; and (3) patients with CHD have highly heterogeneous physiological conditions, and pediatric patients in different age groups may need to be analyzed separately; thus, limiting our analysis to a specific age might provide robust findings. Patients with end-stage heart failure who were eligible for heart transplantation, aortic aneurysm, cardiomyopathy, infection, acute liver dysfunction (alanine aminotransferase [ALT] >100 U/L),[15] severe renal dysfunction, and hypersensitive C-reactive protein (hs-CRP) >3 mg/L were excluded.[16] Finally, a total of 8065 patients were included. A detailed presentation of the inclusion and exclusion criteria is presented in a flowchart [Figure 1]. About 30% of patients were excluded because of incomplete preoperative laboratory tests. The Pediatric Cardiac Surgery Center at our hospital is divided into three parallel departments managed by different chief surgeons, and hence, certain postadmission clinical laboratory tests could vary. The International Classification of Disease, 10th edition (ICD-10, 745–747.9) was used to identify CHD diagnosis. The Society of Thoracic Surgeons–European Association for Cardio-Thoracic Surgery (STAT) mortality categories were used to stratify patients into five levels: STAT 1, STAT 2, STAT 3, STAT 4, and STAT 5, with the highest risk assigned to STAT 5.[17] Flowchart illustrating the process of inclusion and exclusion of the CHD patients who underwent surgical treatment at Fuwai Hospital between 2017 and 2022. CHD: Congenital heart disease; hs-CRP: Hypersensitive C-reactive protein. Clinical management Surgical procedures and postoperative management were conducted in accordance with the standard practices at our institution. For the policy on iron deficiency management, no intervention will be implemented preoperatively, while the postoperative red blood cell (RBC) transfusion was practice-dependent (the threshold was generally set at hemoglobin [Hb] <80 g/L according to patients’ physical condition). Thus, preoperative iron-related laboratory tests would not be affected by interventions. Surgery scheduling for patients with severe iron deficiency or anemia was postponed until an improvement in patients’ iron deficiency or anemia was achieved at another healthcare facility. Blood samples and laboratory tests Conducting preoperative biochemical and hematological tests is a standard procedure after admission. The laboratory test results used in this study were the first preoperative test results that were conducted upon hospital admission. The median duration from blood collection for iron profile laboratory tests to surgical treatment was 2.41 (1.27–5.03) days. Blood samples were collected through the median cubital vein (veins of hands or feet in some circumstances), the serum was used for iron-related tests and biochemical tests, and whole blood was used for hematology tests. Iron-related tests and other biochemical tests were conducted using an automatic biochemical analyzer (HITACHI, LABOSPECT 008AS, Tokyo, Japan). Hematology tests were conducted using an automated hematology analyzer (Sysmex XN-10, Kobe, Japan). Ferritin detection ranges from 0 to 1500 µg/L, and any values exceeding the upper limit were substituted with 1500 µg/L. Definition of iron deficiency and anemia We defined iron deficiency as (1) ferritin <15 μg/L or (2) low transferrin saturation (TS <15%) plus elevated red cell distribution width (RDW >16%).[1,18] Since we excluded patients with hs-CRP levels >3 mg/L, the effect of false ferritin elevation as an acute-phase reactant was effectively mitigated. Anemia was defined as Hb <110 g/L for children aged 6 months to 5 years, according to the World Health Organization (WHO) guidelines.[19] In addition, the criteria for anemia in infants aged 1–5 months were defined as Hb <95 g/L.[20] Iron deficiency anemia was defined as patients who had both iron deficiency and anemia. Outcomes The primary adverse outcome was in-hospital deaths. Due to the limited occurrence of inhospital deaths, we used series of critical conditions as composite adverse events including inhospital deaths, abandoned treatment due to critical illness, nonscheduled reintervention (pericardial fenestration and exploratory thoracotomy), prolonged mechanical ventilation (more than 2 weeks, 336 hours), prolonged pediatric intensive care unit (PICU) stay (more than 30 days), postoperative extracorporeal membrane oxygenation (ECMO) administration, and peritoneal dialysis. Statistical analysis Statistical analyses were conducted using SPSS Statistics version 26.0 (IBM, Armonk, NY, United States) and R software version 4.2.0 (R Foundation for Statistical Computing, Vienna, Austria). Categorical variables are presented as frequencies (percentages) and compared using chi-squared or Fisher’s exact tests, as appropriate. Continuous variables are presented as mean ± standard deviation or median (25th–75th percentile) depending if the variables showed normal or skewed distribution. Independent t-tests were used for normally distributed variables, and Mann–Whitney U tests were used for variables with a skewed distribution. Analysis of variance (ANOVA) or Kruskal–Wallis test was used to assess the differences in means between multiple groups. The shape of the association between ferritin and the outcomes was examined by restricted cubic splines. A multivariable logistic regression model was used to identify the associations between preoperative iron status and postoperative adverse outcomes. In addition, a sensitivity analysis (excluding patients with iron deficiency) was conducted to investigate the correlation between high iron load and adverse events based on cyanotic CHD, STAT, and age. P values of <0.05 were considered statistically significant. Results Demographics The overall clinical characteristics of 8065 patients are presented in Table 1. In this study, the median age of patients was 17.3 (7.7–34.7) months, and 4003 (49.6%) patients were male. A total of 1305 patients were diagnosed with cyanotic CHD, and patients with cyanotic CHD had a higher proportion of male (59.0% vs. 47.8%, P <0.001) and were younger in age (9.7 [5.9–24.2] months vs. 19.1 [8.7–35.9] months, P <0.001) compared with acyanotic CHD. A comprehensive overview of surgical procedures is presented in Supplementary Table 1, http://links.lww.com/CM9/C844. A total of 250 patients had postoperative adverse events including 29 inhospital deaths; 23 patients required ECMO support; 18 patients had abandoned treatment due to terminal illness; 45 patients underwent unplanned reinterventions; 31 patients required peritoneal dialysis; 78 patients had a prolonged PICU admission exceeding 30 days; and 157 patients were on mechanical ventilation for over 2 weeks. Table 1Clinical characteristics of the whole cohort, and patients with acyanotic CHD or cyanotic CHD.CharacteristicsAll (n = 8065)Acyanotic CHD (n = 6760)Cyanotic CHD (n = 1305)StatisticsP valueSex (male)4003 (49.6)3233 (47.8)770 (59.0)χ2 = 54.676<0.001BMI (kg/m2)15.6 (14.4–16.9)15.5 (14.4–16.8)16.1 (14.8–17.6)Z = –10.374<0.001Age (months)17.3 (7.7–34.7)19.1 (8.7–35.9)9.7 (5.9–24.2)Z = –14.648<0.0011–51354 (16.8)1011 (15.0)343 (26.3)6–111684 (20.9)1282 (19.0)402 (30.8)12–231824 (22.6)1592 (23.6)232 (17.8)24–603203 (39.7)2875 (42.5)328 (25.1)Iron deficiency1856 (23.0)1364 (20.2)492 (37.7)χ2 = 189.588<0.001Anemia969 (12.0)844 (12.5)125 (9.6)χ2 = 8.7420.003Iron deficiency anemia564 (7.0)465 (6.9)99 (7.6)χ2 = 0.8420.359Ferritin (µg/L)29.9 (16.9–48.1)30.4 (18.1–47.9)26.0 (9.2–49.4)Z = –6.783<0.001Serum iron (μmol/L)11.3 (8.0–15.3)11.5 (8.3–15.3)10.3 (6.0–15.2)Z = –7.163<0.001Transferrin (g/L)2.7 (2.4–3.0)2.6 (2.4–2.9)2.9 (2.6–3.4)Z = –17.952<0.001TS (%)19.9 (13.6–27.0)20.2 (14.3–27.2)16.9 (8.8–25.6)Z = –10.294<0.001TIBC (μmol/L)58.7 (53.0–65.7)58.0 (52.6–64.1)64.2 (56.0–75.5)Z = –18.298<0.001Hb (g/L)122.0 (113.0–130.0)121.0 (112.0–128.0)133.0 (118.0–152.0)Z = –23.068<0.001RBC (1012/L)4.6 (4.3–4.9)4.5 (4.3–4.8)5.2 (4.6–6.0)Z = –31.248<0.001RDW (%)13.1 (12.5–14.2)13.0 (12.4–13.9)14.1 (12.9–16.8)Z = –20.845<0.001HCT (%)36.5 (34.2–38.7)36.1 (34.0–38.0)40.6 (36.6–46.8)Z = –29.828<0.001MCV (fl)80.5 (77.5–83.3)80.5 (77.8–83.1)80.2 (75.3–84.5)Z = –1.9350.053MCH (pg)27.0 (25.7–28.1)27.0 (25.8–28.1)26.8 (24.1–28.4)Z = –4.943<0.001MCHC (g/L)334.0 (326.0–341.0)334.0 (327.0–341.0)330.0 (317.0–339.0)Z = –12.795<0.001ALT (U/L)17.0 (13.0–24.0)16.0 (13.0–23.0)19.0 (13.0–27.0)Z = –7.845<0.001AST (U/L)42.0 (35.0–51.0)41.0 (35.0–50.0)45.0 (37.0–55.0)Z = –9.283<0.001GGT (U/L)11.0 (9.0–14.0)11.0 (9.0–14.0)12.0 (9.5–17.0)Z = –9.169<0.001ALP (U/L)231.0 (192.0–279.0)229.0 (191.0–275.0)241.0 (199.0–298.0)Z = –6.779<0.001Cre (μmol/L)31.4 (26.6–36.6)31.3 (26.6–36.6)31.5 (26.9–36.7)Z = –0.7210.471BUN (mmol/L)4.1 (3.0–5.1)4.2 (3.1–5.1)3.6 (2.6–4.8)Z = –9.019<0.001Albumin (g/L)42.1 (40.2–43.9)42.3 (40.4–44.0)41.5 (39.4–43.6)Z = –7.662<0.001Sodium (mmol/L)139.3 (137.8–140.7)139.3 (137.9–140.7)139.2 (137.7–140.6)Z = –0.8910.373Potassium (mmol/L)4.8 (4.5–5.1)4.8 (4.5–5.1)4.9 (4.6–5.2)Z = –8.242<0.001Data were presented as counts (percentage) or median (25th–75th percentile). ALP: Alkaline phosphatase; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; BMI: Body mass index; BUN: Blood urea nitrogen; CHD: Congenital heart disease; Cre: Creatinine; GGT: Gamma-glutamyl transpeptidase; Hb: Hemoglobin; HCT: Hematocrit; MCH: Mean corpuscular hemoglobin; MCHC: Mean corpuscular hemoglobin concentration; MCV: Mean corpuscular volume; RBC: Red blood cell; RDW: Red cell distribution width; TIBC: Total iron-binding capacity; TS: Transferrin saturation. Clinical characteristics of the whole cohort, and patients with acyanotic CHD or cyanotic CHD. Data were presented as counts (percentage) or median (25th–75th percentile). ALP: Alkaline phosphatase; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; BMI: Body mass index; BUN: Blood urea nitrogen; CHD: Congenital heart disease; Cre: Creatinine; GGT: Gamma-glutamyl transpeptidase; Hb: Hemoglobin; HCT: Hematocrit; MCH: Mean corpuscular hemoglobin; MCHC: Mean corpuscular hemoglobin concentration; MCV: Mean corpuscular volume; RBC: Red blood cell; RDW: Red cell distribution width; TIBC: Total iron-binding capacity; TS: Transferrin saturation. Iron status and anemia Overall, 23.0% (1856/8065) of patients had iron deficiency, and 12.0% (969/8065) had anemia with more than half of these cases being attributed to iron deficiency. Significant disparities in iron-related laboratory indicators were found between patients with acyanotic and cyanotic CHD. The prevalence of iron deficiency in patients with cyanotic CHD was twice as high (37.7%, 492/1305) as in patients with acyanotic CHD (20.2%, 1364/6760). In addition, patients with cyanotic CHD had significantly higher levels of Hb, RBC, RDW, hematocrit, and lower levels of mean corpuscular volume, mean corpuscular Hb, and mean corpuscular Hb concentrations compared with patients with acyanotic CHD. Furthermore, the incidence of anemia was observed to be lower in the cyanotic CHD (9.6%, 125/1305) than in the acynotic CHD (12.5%, 844/6760) group. Detailed comparisons are presented in Table 1. Patients with adverse outcomes had a significantly higher proportion of iron deficiency and high ferritin status, elevated levels of ferritin, Hb, RBC, and RDW, and lower serum iron and TS compared with patients without adverse outcomes [Supplementary Table 2, http://links.lww.com/CM9/C844]. Association between iron profile and outcomes The association between ferritin levels and adverse outcomes, which were assessed using restricted cubic splines, showed right-skewed U-shaped curves between ferritin levels and death and composite adverse events [Figure 2 and Supplementary Figure 1, http://links.lww.com/CM9/C844]. Patients with ferritin levels between 15 and 33 µg/L had a lower risk of adverse outcomes (ferritin level of 15 µg/L was used as the threshold for identifying iron deficiency, and ferritin levels of 15 µg/L and 33 µg/L demonstrated a nearly equivalent OR). Based on the splines, we categorized patients into four groups representing different iron statuses: (1) 23.0% had iron deficiency (ferritin <15 µg/L, or TS <15% plus RDW >16%), (2) 33.3% had iron repletion (ferritin levels of ≥15 and <33 µg/L), (3) 36.3% had intermediate ferritin (ferritin levels of ≥33 and <100 µg/L), and (4) 7.4% had high ferritin (ferritin ≥100 µg/L). Restricted spline curves showing the associations between ferritin and postoperative adverse outcomes. The models were adjusted for sex, age, anemia, cyanotic CHD, BMI, ALT, AST, Cre, BUN, albumin, sodium, and potassium. (A) Restricted spline curve for the associations between ferritin and inhospital death. (B) Restricted spline curve for associations between ferritin and composite adverse events. The dotted lines represent the ferritin thresholds of 15, 33, and 100 µg/L, respectively. The results presented in this figure are limited to a ferritin range of 0–200 µg/L. The overall diagrams are illustrated in Supplementary Figure 1, http://links.lww.com/CM9/C844. ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; BMI: Body mass index; BUN: Blood urea nitrogen; CHD: Congenital heart disease; CI: Confidence interval; Cre: Creatinine. We performed a multivariable logistic regression analysis to investigate the association between iron status and adverse events with adjustment for sex, age, anemia, cyanotic CHD, body mass index (BMI), ALT, aspartate aminotransferase (AST), creatinine (Cre), blood urea nitrogen (BUN), albumin, sodium, and potassium [Table 2]. The risk of inhospital death was significantly higher in patients with high ferritin levels (OR: 8.20 [95%CI: 1.61–41.86], P = 0.011), but the increase was statistically nonsignificant in patients with iron deficiency (OR: 3.68 [95%CI: 0.77–17.59], P = 0.102) or intermediate ferritin levels (OR: 3.09 [95%CI: 0.65–14.69], P = 0.156). Compared to patients with iron repletion, those with high ferritin (OR: 4.21 [95%CI: 2.65–6.68], P <0.001), iron deficiency (OR: 1.80 [95%CI: 1.18–2.73], P = 0.006), and intermediate ferritin (OR: 1.64 [95%CI: 1.11–2.44], P = 0.014) had a significantly higher risk of composite adverse events. In addition, the patients with high ferritin had the longest duration of mechanical ventilation and PICU stay, followed by those with iron deficiency, intermediate ferritin levels, and iron repletion. Both iron deficiency and high iron load were demonstrated as independent risk factors of postoperative RBC transfusion. Table 2Association between preoperative iron status and postoperative adverse events.VariablesIron repletionIron deficiencyIntermediate ferritinHigh ferritinPatients2686 (33.3)1856 (23.0)2924 (36.3)599 (7.4)Death2 (0.1)10 (0.5)8 (0.3)9 (1.5)OR (95% CI), death1, reference3.68 (0.77–17.59)3.09 (0.65–14.69)8.20 (1.61–41.86)P value, death0.1020.1560.011Composite events39 (1.5)74 (4.0)78 (2.7)59 (9.8)OR (95% CI), composite1, reference1.80 (1.18–2.73)1.64 (1.11–2.44)4.21 (2.65–6.68)P value, composite0.0060.014<0.001RBC transfusion1534 (57.1)1488 (80.2)1749 (59.8)530 (88.5)OR (95% CI), RBC1, reference1.28 (1.08–1.52)1.16 (1.02–1.33)1.71 (1.23–2.37)P value, RBC0.0050.0250.002RBC transfusion, units1.0 (0–1.0)1.0 (1.0–2.0)1.0 (0–1.0)2.0 (1.0–2.0)Mechanical ventilation, hours5.0 (3.0–9.0)7.0 (4.0–21.0)5.0 (3.0–15.0)24.0 (9.0–88.0)PICU (days)1.0 (1.0–3.0)2.0 (1.0–4.0)2.0 (1.0–3.0)4.0 (2.0–8.0)Data were presented as counts (percentage) or median (25th-75th percentile). The multivariable logistic model was adjusted by sex, age (continuous variable), anemia, cyanotic CHD, BMI (continuous variable), ALT (continuous variable), AST (continuous variable), Cre (continuous variable), BUN (continuous variable), albumin (continuous variable), sodium (continuous variable), and potassium (continuous variable). ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; BMI: Body mass index; BUN: Blood urea nitrogen; CHD: Congenital heart disease; CI: Confidence interval; Cre: Creatinine; OR: Odds ratio; PICU: Pediatric intensive care unit; RBC: Red blood cell. Association between preoperative iron status and postoperative adverse events. Data were presented as counts (percentage) or median (25th-75th percentile). The multivariable logistic model was adjusted by sex, age (continuous variable), anemia, cyanotic CHD, BMI (continuous variable), ALT (continuous variable), AST (continuous variable), Cre (continuous variable), BUN (continuous variable), albumin (continuous variable), sodium (continuous variable), and potassium (continuous variable). ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; BMI: Body mass index; BUN: Blood urea nitrogen; CHD: Congenital heart disease; CI: Confidence interval; Cre: Creatinine; OR: Odds ratio; PICU: Pediatric intensive care unit; RBC: Red blood cell. To further investigate the impact of high iron load on the occurrence of adverse outcomes, we excluded individuals with iron deficiency and conducted sensitivity analyses using multivariable logistic regression models based on cyanotic CHD, various surgical risks (based on STAT), and different age groups [Table 3]. Although the sensitivity analyses led to a reduction in both the sample size and the number of adverse events, high ferritin levels remained a significant risk factor for inhospital death among patients with STAT 1–2 (OR: 11.90 [95%CI: 1.13–124.98], P = 0.039) and for patients aged between 1 month and 1 year (OR: 10.87 [95%CI: 1.17–100.86], P = 0.036). Additionally, it was identified as an independent predictor for composite adverse events in patients with acyanotic CHD (OR: 3.90 [95%CI: 2.09–7.29], P <0.001), cyanotic CHD (OR: 2.32 [95%CI: 1.08–4.97], P = 0.031), STAT 1–2 (OR: 4.16 [95%CI: 2.33–7.44], P <0.001), STAT 3–5 (OR: 2.65 [95%CI: 1.14–6.15], P = 0.023), and those aged between 1 month and 1 year (OR: 3.40 [95%CI: 1.98–5.84], P <0.001). Table 3Association of high iron load with postoperative adverse events stratified by cyanotic CHD, STAT, and age.VariablesIron repletionIntermediate ferritinHigh ferritinPatients with acyanotic CHD and no iron deficiency (n = 5396)Death, n (%)1 (0.04)3 (0.12)4 (0.82)OR (95% CI), death1, reference2.73 (0.28–26.93)8.54 (0.75–97.53)P value, death0.3890.084Composite events, n (%)21 (0.88)41 (1.63)41 (8.40)OR (95% CI), composite1, reference1.60 (0.94–2.74)3.90 (2.09–7.29)P value, composite0.086 <0.001Patients with cyanotic CHD and no iron deficiency (n = 813)Death, n (%)1 (0.34)5 (1.23)5 (4.50)OR (95% CI), death1, reference2.76 (0.31–24.66)4.75 (0.44–51.01)P value, death0.3630.198Composite events, n (%)18 (6.06)37 (9.14)18 (16.22)OR (95% CI), composite1, reference1.40 (0.77–2.54)2.32 (1.08–4.97)P value, composite0.2730.031Patients undergoing surgeries of STAT 1–2 with no iron deficiency (n = 5654)Death, n (%)1 (0.04)6 (0.22)4 (0.88)OR (95% CI), death1, reference5.25 (0.62–44.21)11.90 (1.13–124.98)P value, death0.1270.039Composite events, n (%)29 (1.15)46 (1.72)34 (7.44)OR (95% CI), composite1, reference1.34 (0.83–2.15)4.16 (2.33–7.44)P value, composite0.233 <0.001Patients undergoing surgeries of STAT 3–5 with no iron deficiency (n = 555)Death, n (%)1 (0.59)2 (0.82)5 (3.52)OR (95% CI), death1, reference1.37 (0.12–15.89)3.57 (0.33–39.31)P value, death0.8020.298Composite events, n (%)10 (5.88)32 (13.17)25 (17.61)OR (95% CI), composite1, reference2.52 (1.17–5.39)2.65 (1.14–6.15)P value, composite0.0180.023Patients aged ≥1 month and ≤1 year with no iron deficiency (n = 2144)Death, n (%)1 (0.14)4 (0.43)8 (1.63)OR (95% CI), death1, reference4.38 (0.46–41.54)10.87 (1.17–100.86)P value, death0.1980.036Composite events, n (%)21 (2.87)41 (4.45)53 (10.79)OR (95% CI), composite1, reference1.47 (0.87–2.48)3.40 (1.98–5.84)P value, composite0.155 <0.001Patients aged >1 year and ≤5 year with no iron deficiency (n = 4065)Death, n (%)1 (0.05)4 (0.20)1 (0.93)OR (95% CI), death1, reference3.70 (0.38–36.17)7.75 (0.40–150.80)P value, death0.2610.176Composite events, n (%)18 (0.92)37 (1.85)6 (5.56)OR (95% CI), composite1, reference1.44 (0.81–2.56)52.51 (0.91–6.88)P value, composite0.2180.074The multivariable logistic model was adjusted by sex, age (continuous variable), anemia, cyanotic CHD, BMI (continuous variable), ALT (continuous variable), AST (continuous variable), Cre (continuous variable), BUN (continuous variable), albumin (continuous variable), sodium (continuous variable), and potassium (continuous variable). Cyanotic CHD was removed from the model when stratified by cyanotic CHD, and age was removed from the model when stratified by age. ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; BMI: Body mass index; BUN: Blood urea nitrogen; CHD: Congenital heart disease; CI: Confidence interval; Cre: Creatinine; OR: Odds ratio; STAT: The Society of Thoracic Surgeons–European Association for Cardio-Thoracic Surgery. Association of high iron load with postoperative adverse events stratified by cyanotic CHD, STAT, and age. The multivariable logistic model was adjusted by sex, age (continuous variable), anemia, cyanotic CHD, BMI (continuous variable), ALT (continuous variable), AST (continuous variable), Cre (continuous variable), BUN (continuous variable), albumin (continuous variable), sodium (continuous variable), and potassium (continuous variable). Cyanotic CHD was removed from the model when stratified by cyanotic CHD, and age was removed from the model when stratified by age. ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; BMI: Body mass index; BUN: Blood urea nitrogen; CHD: Congenital heart disease; CI: Confidence interval; Cre: Creatinine; OR: Odds ratio; STAT: The Society of Thoracic Surgeons–European Association for Cardio-Thoracic Surgery. According to the WHO guidelines,[21] the cut-off value for iron deficiency for children aged <5 years is ferritin levels of 12 µg/L; however, the use of lower ferritin levels (12 µg/L) for diagnosis of iron deficiency in children is not evidence-based.[1] To investigate the effect of this lower threshold of ferritin, we conducted a series of sensitivity analyses by redefining iron deficiency as (1) ferritin <12 μg/L or (2) TS <15% plus RDW >6%. The definition of intermediate ferritin was appropriately modified to 36–100 µg/L based on restricted cubic splines [Supplementary Figure 2, http://links.lww.com/CM9/C844]. Detailed results are presented in the Supplementary Tables 3 and 4, http://links.lww.com/CM9/C844. Discussion Iron status is not only pivotal for the growth and development of children, but also plays a critical role in the prognosis of pediatric patients. In addition, the impact of iron imbalance on cardiovascular diseases has garnered increasing clinical attention. However, iron metabolism and its prognostic role in pediatric patients with CHD have not been thoroughly elucidated. In this study, we aimed to comprehensively investigate preoperative iron status and its correlation with adverse outcomes in pediatric patients with CHD. We have reported the preoperative iron status and prevalence of anemia in a large Chinese cohort comprising 8065 pediatric patients with CHD. Our findings demonstrated a significant association between preoperative iron imbalance and increased risks of postoperative mortality, composite adverse events, and postoperative RBC transfusion. This study describes the prevalence of iron deficiency and anemia in pediatric patients with CHD. Based on previously published literature, we adopted an optimized approach to define iron deficiency using multiple indicators such as ferritin, TS, and RDW, to overcome the limitation of using a single test for the diagnosis of iron deficiency.[22] Furthermore, all patients included in this study had preoperative hs-CRP tests, which, combined with clinical assessment of infection, effectively mitigated the confounding influence of acute phase ferritin elevation on iron status evaluation. Of the pediatric patients with CHD, iron deficiency, anemia, and iron deficiency anemia were observed in 23.0%, 12.0%, and 7.0% of patients in this study, respectively. The prevalence observed in this study was substantially higher than that reported in a previous study from the United States that used data from the 2007–2010 National Health and Nutrition Examination Survey. That study reported a prevalence of 7.1% for iron deficiency, 3.2% for anemia, and 1.1% for iron deficiency anemia in children aged 1–5 years.[3] Our center has previously reported that the prevalence of iron deficiency (ferritin <12 ng/mL) in infants (aged 1–12 months) who had cardiac surgery with cardiopulmonary bypass was 58.4% in patients with cyanotic CHD and 24.1% in patients with acyanotic CHD.[13] Camile et al[23] reported that approximately 63.6% of patients aged 6–48 months with cyanotic CHD exhibited iron deficiency (ferritin <12 µg/L and TS <12%). In East Africa, the prevalence of iron deficiency (ferritin <12 ng/mL) was 26.9% in children (aged <18 years) with CHD.[24] The rates of iron deficiency reported in these studies were significantly higher than those observed in our data. According to the Statistical Monitoring Report of the Program for the Development of Chinese Children (2011–2020) released by the National Bureau of Statistics,[25] the prevalence of anemia in children younger than 5 years of age was 5.38%, indicating a relatively high burden of anemia (12.0%) among pediatric patients with CHD in China. We observed that approximately 50% of anemia cases were attributable to iron deficiency, which was slightly higher than the estimates provided by the WHO.[26] The prevalence of iron deficiency anemia was much lower in our study (7.0% for all and 7.6% for cyanotic CHD) compared to the findings reported in Tanzania[24] (20.2% of patients with CHD, aged <18 years) and India[27] (47.06% of patients with cyanotic CHD, aged between 0 and 12 years). In addition, we found that the prevalence of anemia in patients with cyanotic CHD was significantly lower compared to those with acyanotic CHD, which could be attributed to the compensatory increase in Hb and abnormal erythropoiesis observed in our study and other relevant research.[13,28] Taken together, iron deficiency and anemia continue to pose a common challenge in children with CHD, and the prevalence of iron deficiency, anemia, and iron deficiency anemia varies significantly across different studies. These differences in results could be due to the age of patients, ethnicity, type of CHD, diagnosis criteria, and region. Iron metabolism disorders, including deficiency or overload, are associated with a higher risk of cardiovascular morbidity and mortality.[5] Iron deficiency may cause cardiac dysfunction as iron is a crucial microelement for oxidative metabolism in cardiomyocytes and mitochondrial energy metabolism.[29] In addition, patients with iron deficiency exhibit more significant microcytic erythrocyte deformation, leading to a higher incidence of thromboembolic complications and cerebrovascular accidents.[30] On the other hand, excess iron leads to the formation of nontransferrin-bound iron, which is implicated in progressive liver and heart damage.[31] High iron load may also interact with concomitant conditions (such as myocardial ischemia) to induce damage on cardiomyocyte,[5] which is likely to occur during pediatric cardiac surgery involving cardiopulmonary bypass and aortic crossclamping. Iron overload has been shown to induce arrhythmias, including bradycardia, altered electrical conduction, and atrial fibrillation in animal models and in human studies.[32] Elevated ferritin could also be a sign of systemic inflammation, which can potentially impact the cardiomyocytes and cardiac function (particularly after cardiac interventions). In conclusion, iron status has a dual impact and requires careful consideration for patients with cardiovascular diseases. Iron deficiency has been associated with poor prognosis in adult patients with various cardiac conditions such as heart failure, coronary artery disease, aortic stenosis, atrial fibrillation, and pulmonary hypertension.[33] In pediatric cardiovascular diseases, iron deficiency primarily poses a concern in children with heart failure. A retrospective study of 28 patients with heart failure attributed to dilated cardiomyopathy (patients <21 years old) demonstrated that iron deficiency was related to severe heart failure and mortality.[34] However, although Phillips et al[35] reported a high prevalence of iron deficiency in pediatric patients with dilated cardiomyopathy-induced heart failure, they found no significant difference in the risk of cardiac adverse events between iron-deficient patients and iron-sufficient patients. In a larger cohort of patients with various etiologies of pediatric heart failure, Puri et al[6] demonstrated a significant association between iron deficiency and a higher risk of ventricular assist device implantation, heart transplantation, or death. Our findings that patients with iron deficiency were more likely to require RBC transfusion aligned with previous research, which demonstrated that preoperative iron deficiency was an independent risk factor for postoperative transfusion in infants (1–12 months) who underwent cardiac surgeries with cardiopulmonary bypass.[13] Pediatric patients with iron overload had risks of delayed sexual maturation, retarded growth, progressive liver and heart disease, and reduced life expectancy.[36] A meta-analysis demonstrated that high ferritin was a significant predictor of poor outcomes in several pediatric critical illnesses, including sepsis, multisystem inflammatory syndrome, coronavirus disease 2019 (COVID-19), and hemophagocytic lymphohistiocytosis.[37] Considerable variability in serum ferritin threshold for risk stratification was observed across different conditions or diseases of interest, and a concentration of 500 µg/L was the most frequently used threshold value. Unfortunately, the association between high-iron status and the outcomes of pediatric patients with cardiac disease is lesser investigated. In this study, we classified patients with ferritin levels between 33 and 100 µg/L as the intermediate ferritin group and those with ferritin levels exceeding 100 µg/L as the high ferritin group, both of which were considered at higher risk of postoperative adverse outcomes compared to the iron repletion group (ferritin levels between 15 and 33 µg/L). The threshold of intermediate ferritin was determined based on the criteria that the OR of ferritin at 33 ng/mL was comparable to that of ferritin at 15 ng/mL (used as one of the definitions of iron deficiency). The threshold for high ferritin was set at 100 ng/mL (the 95th percentile of serum ferritin concentrations in healthy children in the United States was found to be approximately 70 ng/mL),[38] as this value was identified as a turning point by the restricted cubic spline analysis, and the OR increased more rapidly for ferritin levels between approximately 25 and 100 ng/mL than for those exceeding 100 ng/mL. Serum ferritin is widely recognized as the predominant indicator for assessing iron overload; however, it should be acknowledged that ferritin levels may not reliably reflect total body iron stores.[36] We excluded patients with infections, elevated hs-CRP (>3 mg/L), and severe liver or renal dysfunction to ensure accurate iron burden assessment based on ferritin measurements. Therefore, the definition of high iron load in this study was relatively reasonable and accurate. Considering the association between preoperative anemia and poor outcomes in children undergoing cardiac surgery,[39,40] some researchers have focused on the effectiveness of oral or intravenous iron supplementation.[41,42] However, based on our findings, a preoperative ferritin of approximately 25 ng/mL indicated minimal risk for postoperative complications. Therefore, it might be necessary to establish a safety threshold for preoperative iron supplementation. Further, the necessity of employing iron chelation in pediatric patients with CHD exhibiting excessively high ferritin levels warrants further investigation. The present study has certain limitations. First, this is a single-center study and hence, the results cannot be generalized and should be interpreted cautiously. Further validation studies are needed. Second, the retrospective study design might introduce bias in data collection, confounding factors, and other related aspects. For example, we did not comprehensively consider the preadmission administration of iron supplements in all patients. Third, we did not conduct any follow-ups. Thus, any potential adverse effects of preoperative iron status on patients’ long-term prognosis could not be assessed. Fourth, although the soluble transferrin receptor is a valuable indicator of tissue iron requirements and erythropoiesis, it was not shown in the results due to its limited implementation within our center. Fifth, a single CRP test alone was insufficient for effectively assessing the inflammatory status, as WHO recommended testing for both CRP and α-1 acid glycoprotein (AGP). The lack of routine AGP testing in our hospital might not have provided an accurate assessment of inflammation in this study. In conclusion, in this study, we comprehensively reported preoperative iron and anemia status in more than 8000 pediatric patients with CHD. Our findings demonstrated a significant association between preoperative iron imbalance (both iron deficiency and high iron load) and postoperative adverse outcomes. These results highlighted the importance of monitoring preoperative iron status in pediatric patients with CHD and provided a theoretical foundation for iron-related therapeutic management. Acknowledgment We appreciate the statistical assistance of Dr. Lang Xinyue (Department of Pharmacy and Clinical Trial Unit, Beijing Anzhen Hospital, Capital Medical University). Funding This study was supported by the National Key R&D Program of China (No. 2023YFC2413005), National High-Level Hospital Clinical Research Funding (No. 2022-GSP-GG-6), and Capital Health Research and Development of Special Fund (No. 2022-1-4032). Conflicts of interest None. Supplementary Material