Abstract Objective The study objective was to determine whether adequately delivered bilateral remote ischemic preconditioning is cardioprotective in young children undergoing surgery for 2 common congenital heart defects with or without cyanosis. Methods We performed a prospective, double-blind, randomized controlled trial at 2 centers in the United Kingdom. Children aged 3 to 36 months undergoing tetralogy of Fallot repair or ventricular septal defect closure were randomized 1:1 to receive bilateral preconditioning or sham intervention. Participants were followed up until hospital discharge or 30 days. The primary outcome was area under the curve for high-sensitivity troponin-T in the first 24 hours after surgery, analyzed by intention-to-treat. Right atrial biopsies were obtained in selected participants. Results Between October 2016 and December 2020, 120 eligible children were randomized to receive bilateral preconditioning (n = 60) or sham intervention (n = 60). The primary outcome, area under the curve for high-sensitivity troponin-T, was higher in the preconditioning group (mean: 70.0 ± 50.9 μg/L/h, n = 56) than in controls (mean: 55.6 ± 30.1 μg/L/h, n = 58) (mean difference, 13.2 μg/L/h; 95% CI, 0.5-25.8; P = .04). Subgroup analyses did not show a differential treatment effect by oxygen saturations (p[interaction =] .25), but there was evidence of a differential effect by underlying defect (p[interaction =] .04). Secondary outcomes and myocardial metabolism, quantified in atrial biopsies, were not different between randomized groups. Conclusions Bilateral remote ischemic preconditioning does not attenuate myocardial injury in children undergoing surgical repair for congenital heart defects, and there was evidence of potential harm in unstented tetralogy of Fallot. The routine use of remote ischemic preconditioning cannot be recommended for myocardial protection during pediatric cardiac surgery. Key Words: clinical trial, cyanosis, myocardial protection, pediatric cardiac surgery, remote ischemic preconditioning, tetralogy of Fallot Graphical abstract [77]graphic file with name ga1.jpg [78]Open in a new tab __________________________________________________________________ graphic file with name fx1.jpg [79]Open in a new tab Mean hs-troponin-T release in the first 24 hours after surgery by treatment group. Central Message. RIPC does not improve myocardial protection in young children undergoing surgical repair of common congenital heart defects. It may cause harm and should not be used routinely. Perspective. Previous trials of RIPC in children undergoing cardiac surgery have shown mixed results, and their designs are criticized. We found that adequately delivered, bilateral preconditioning did not attenuate myocardial injury and may be harmful in those with unstented TOF. Its routine use cannot be recommended, and alternative protective strategies should be sought. Myocardial protection against ischemic-reperfusion injury is a key determinant of heart function and outcome after cardiac surgery in children.[80]^1 With current strategies, myocardial injury occurs routinely after aortic crossclamping, as quantified by an increase in circulating troponin in the first 24 hours.[81]^2 This myocardial damage frequently impairs ventricular function, which may manifest as low cardiac output requiring inotropic support. This is a major cause of morbidity and death in the early postoperative period,[82]^1^,[83]^3 and children with preoperative cyanosis may be more vulnerable than acyanotic children.[84]^4^,[85]^5 Remote ischemic preconditioning (RIPC), the application of brief, nonlethal cycles of ischemia and reperfusion to a distant organ or tissue, is a simple, low-risk, and readily available technique that when delivered immediately before surgery may improve myocardial protection. Previous trials of RIPC in children undergoing cardiac surgery have shown mixed results[86]6, [87]7, [88]8, [89]9, [90]10, [91]11, [92]12, [93]13, [94]14 and have been criticized for a potentially inadequate stimulus; a manual sphygmomanometer may allow subclinical reperfusion during the ischemic phase,[95]^12 and propofol anesthesia has been suggested to interfere with the preconditioning pathway.[96]^15 In addition, they have not evaluated the effects of preoperative cyanosis on RIPC[97]^16 and have only applied the cuff to a single limb, potentially delivering a subtherapeutic stimulus in young children with a lower skeletal muscle mass compared with adults. To provide a robust answer to the question of whether RIPC can attenuate perioperative myocardial injury in young children undergoing surgery, we conducted a randomized, prospective, double-blind trial comparing state-of-the-art bilateral preconditioning with a sham control in children with the 2 most common congenital heart defects requiring surgery and investigated the impact of RIPC on myocardial metabolism during cardioplegic arrest. Materials and Methods Study Design The Bilateral Remote Ischemic Conditioning in Children trial was a double-blind, prospective, parallel group, randomized controlled trial in young children undergoing elective cardiac surgery at 2 centers in the United Kingdom: Birmingham Children's Hospital and Leeds Children's Hospital ([98]Figure 1). The study was approved by the West Midlands-Solihull NHS Research Ethics Committee (16/WM/0309, August 5, 2016). The published trial protocol provides a detailed description of the trial and methods, including patient and public involvement.[99]^17 Figure 1. [100]Figure 1 [101]Open in a new tab The Bilateral Remote Ischemic Conditioning in Children trial. A 2-center, double-blind, randomized controlled trial in which 120 young children with the 2 most common congenital heart defects requiring surgery were randomized to bilateral RIPC or sham intervention. AUC for hs-troponin-T in the first 24 hours was higher in the preconditioning group than in controls (P = .04), and subgroup analysis suggested a differential effect by underlying defect (p[interaction =] .04). Bilateral RIPC does not attenuate myocardial injury during surgery in young children, with evidence of potential harm in unstented TOF, and its routine use cannot be recommended. AUC, Area under the curve; CI, confidence interval; RVOT, right ventricular outflow tract; RIPC, remote ischemic preconditioning; ToF, tetralogy of Fallot; VSD, ventricular septal defect; hs, high sensitivity. Patients All infants and young children, aged 3 months to 3 years at the time of surgery, undergoing complete repair of tetralogy of Fallot (TOF) or surgical closure of a ventricular septal defect (VSD), with or without concomitant atrial septal defect closure or pulmonary artery repair/augmentation, were eligible. Children were excluded if they required an additional procedure (other than atrial septal defect closure or pulmonary artery repair/augmentation); had significant airway or parenchymal lung disease, bleeding disorder, or a recent ischemic event; had undergone previous cardiac surgery with cardioplegic arrest; required emergency surgery; or their parents declined to give consent. As in previous trials,[102]^6^,[103]^12 those with a known major chromosomal defect were also initially excluded, but this was amended during the trial because there was no biological reason for exclusion. Eligible patients were identified from multidisciplinary team meetings, clinics, or surgical waiting lists. A parent information sheet was provided, in person or via post, and parental written informed consent for publication obtained by a consultant surgeon before enrollment. Randomization and Blinding On the day of surgery, participants were randomized (1:1) to receive RIPC or a sham procedure (control) using a secure online randomization system with a minimization algorithm incorporating (1) congenital heart defect, (2) presence of a right ventricular outflow tract (RVOT) stent in those with TOF, and (3) surgical center. The trial intervention was delivered by an independent healthcare professional trained and competent in its delivery according to a standard operating procedure, who also performed the randomization and was not involved in postoperative care. Blinding was maintained by covering the child with a surgical drape throughout cuff application, intervention, and removal. The research nurse and clinical teams involved in the child's care were blinded to group allocation throughout the trial. Procedures After induction of anesthesia but before sternotomy, the treatment group received RIPC in both lower limbs simultaneously using the PTSii digital tourniquet system (Delfi Medical Innovations) inflated to 50 mm Hg or greater above systolic pressure for 3 cycles of 5 minutes ischemia and 5 minutes reperfusion[104]^18; if 1 lower limb was unavailable, the cuff was placed on the upper arm. Continual loss of arterial flow during ischemia was confirmed by concealed distal pulse oximetry.[105]^10 Once the intervention had begun, each cuff was kept on the same limb to ensure repeated doses of ischemia-reperfusion to the same muscle mass. In the control group, cuffs were applied to a plastic tubing dummy limb placed between the participant's legs and 3 sham inflation-deflation cycles performed, covered by a drape before, during, and after the sham intervention to maintain blinding. Adherence to intervention was defined as receiving the allocated intervention, with documented loss of arterial flow during each ischemic phase in the RIPC group. All other aspects of anesthesia, surgery, perfusion, and postoperative care were at the sole discretion of the blinded clinical team, except for propofol, which was not used for induction or maintenance of anesthesia, with isoflurane as the preferred volatile anesthetic.[106]^19 St Thomas' cardioplegia was used at both sites and delivered according to local practice. Myocardial reperfusion on first release of the aortic crossclamp was considered as time zero for postoperative events. Blood was drawn before sternotomy and at 3, 6, 12, and 24 hours after reperfusion, and plasma high-sensitivity (hs) troponin-T concentrations were quantified in batches using the fifth-generation Elecsys Tn-T HS assay (Roche) at an approved core laboratory. Right atrial biopsies were obtained soon after aortic crossclamping (onset ischemia) and just before its release (late ischemia) for metabolic phenotyping. Briefly, tissue extracts were analyzed using ultra high-performance liquid chromatography-mass spectrometry. Two complementary assays were applied: HILIC assay for water-soluble metabolites and C[18] reversed-phase assay for lipids. The impact of RIPC on myocardial metabolism was assessed through robust statistical analysis using correction for multiple testing and pathway enrichment analysis ([107]Supplementary Methods). Outcomes The primary outcome was area under the curve (AUC) for plasma hs-troponin-T in the first 24 hours after aortic crossclamp release (reperfusion) as a biomarker of myocardial injury. Secondary outcomes were peak hs-troponin-T in the first 12 hours; total vasoactive inotrope score in the first 12 hours[108]^20; arterial lactate and central venous oxygen saturations in the first 12 hours; length of postoperative stay in intensive care unit (ICU); and length of postoperative stay in hospital. Cardiac index was measured using ICON (Osypka Medical) as an exploratory outcome in Birmingham only ([109]Supplemental Methods). The following serious adverse events were reported to the sponsor within 48 hours of identification: death; extracorporeal life support; major neurological event; and further surgery or catheter intervention in the early postoperative period. Follow-up was until discharge from hospital or 30 days, whichever was sooner. Sample Size We hypothesized that RIPC would reduce AUC for hs-troponin-T in the first 24 hours compared with controls, but that exposure to chronic hypoxemia may impact this reduction. Based on limited published data using a standard (fourth-generation) troponin assay, the proposed sample size was sufficient to detect a 35% reduction in postoperative AUC troponin, assuming a mean release equivalent to 350 μg/L/h in the control group compared with 228 μg/L/h in the RIPC group (extrapolated from the similarly mixed cohort of cyanotic and acyanotic children[110]^6), with a variability of 220 μg/L/h[111]^9 A sample size of at least 52 children per treatment group was needed to have a power of 80% and a significance level of .05 (2-sided). We aimed to recruit up to 120 children to allow for dropouts, randomized in a 1:1 ratio between RIPC and control. Statistical Analysis Primary analysis of the primary and secondary outcome measures was performed according to the intention-to-treat principle. Analyses were undertaken using SAS v9.4. The primary comparison compared the RIPC group with the control group, and all estimates of differences are presented with 95% CIs. To calculate the primary outcome, AUC for hs-troponin-T in the first 24 hours, data were collected at baseline (presternotomy) and at 3, 6, 12, and 24 hours after reperfusion. The AUC was calculated for each participant using the trapezoidal method and compared between groups using a linear regression model, adjusting for the minimization variables (congenital heart defect, center) and baseline troponin. Missing baseline troponin values were imputed using the median value of the participant's randomized group and type of defect, whereas any missing postoperative value led to exclusion from the primary analysis. Participants who received their randomized treatment, including those with incomplete postoperative troponin data, were included in the per-protocol analysis. Subgroup analyses were performed for the primary outcome only to assess whether the treatment effect differed by preoperative oxygen saturations (cyanotic < 90% or acyanotic ≥ 90%); congenital heart defect (TOF with RVOT stent, TOF without RVOT stent, or VSD); and age (<1 year or ≥1 year). For the secondary outcomes, continuous data items were analyzed using a linear regression model. Continuous outcomes measured across more than 3 time points were analyzed using mixed-effect repeated-measures models. Time to event outcomes were analyzed using a Cox regression model. P values are reported from 2-sided tests. The statistical analysis plan was agreed and signed off before database lock, and the Chief Investigator and trial statisticians had access to the final dataset. The trial was prospectively registered (ISRCTN12923441) before recruiting the first patient. An independent Data Monitoring Committee (see Acknowledgments) met at regular intervals during recruitment to