Abstract
Objectives
Neonates undergoing complex congenital heart surgery have a significant incidence of neurological problems. Erythropoietin has anti-apoptotic, anti-excitatory, and anti-inflammatory properties to prevent neuronal cell death in animal models, and improves neurodevelopmental outcomes in full term neonates with hypoxic ischemic encephalopathy. We designed a prospective phase I/II trial of erythropoietin neuroprotection in neonatal cardiac surgery to assess safety, and indicate efficacy.
Methods
Neonates undergoing surgery for D-transposition of the great vessels, hypoplastic left heart syndrome, or aortic arch reconstruction were randomized to 3 perioperative doses of erythropoietin, or placebo. Neurodevelopmental testing with Bayley Scales of Infant and Toddler Development III was performed at age 12 months.
Results
59 patients received study drug. Safety profile, including MRI brain injury, clinical events, and death, was not different between groups. 3 patients in each group died. 42 patients (22 erythropoietin, 20 placebo, 79% of survivors) returned for 12-month follow-up. The mean Cognitive Scores were erythropoietin, 101.1 ± 13.6, placebo, 106.3 ± 10.8 (p=0.19); Language Scores were erythropoietin 88.5 ± 12.8, placebo 92.4 ± 12.4 (p=0.33); and Motor Scores were erythropoietin 89.9 ± 12.3, placebo 92.6 ± 14.1, (p=0.51).
Conclusions
Safety profile for erythropoietin administration was not different than placebo. Neurodevelopmental outcomes were not different between groups, however this pilot study was not powered to definitively address this outcome. Lessons learned from the current study suggest optimized study design features for a larger prospective trial to definitively address the utility of erythropoietin for neuroprotection in this population.
Keywords: Cardiopulmonary bypass, brain, magnetic resonance imaging, pediatrics
Introduction
Despite major advances in reducing mortality for neonatal cardiac surgery, acute neurological morbidity after cardiac surgery (seizures, coma, movement disorders) occurs in 1–25% of patients in recent reports.[1,2] Long term neurodevelopmental impairment is reported in about half of children undergoing cardiac surgery as newborns or young infants.[3] Data from long term follow-up studies in preschool and school aged children who underwent surgery for congenital heart disease reveal a spectrum of neurodevelopmental problems that are similar to those observed in low birthweight premature infants, including problems with attention, language, memory, and sensorimotor functioning.[3]
Perioperative causes of neurological injury include cerebral hypoxia/ischemia due to cardiovascular pathophysiology, surgical or cardiopulmonary bypass techniques, cerebral emboli, low cardiac output, and intercurrent events such as cardiac arrests.[2,4,5] Pre- and postoperative magnetic resonance imaging (MRI) studies document a 36–73% incidence of new or worsening white matter or other ischemic brain lesions after neonatal cardiac surgery.[6,7]
Erythropoietin (EPO) is a 30.4 kilodalton glycoprotein which has anti-apoptotic, anti-inflammatory, and anti-excitatory cell death effects, protecting the brain against a variety of cerebral insults in animal and in vitro models.[8,9] Human neonatal data demonstrates improved longer term neurodevelopmental outcome with EPO after hypoxic-ischemic encephalopathy.[10] Because EPO is already utilized in the neonatal population for prevention and treatment of anemia and has a long record of safety, it is potentially a very useful agent for neuroprotection in neonatal cardiac surgery. The predictable period of known risk of potential neurological injury in neonatal cardiac surgery makes pre-emptive intervention with EPO to protect the brain an attractive concept. However, the effects of perioperative administration of EPO, in the higher doses needed to provide neuroprotection than those routinely utilized for anemia, are not known. This phase I/II trial was designed to assess safety, and give preliminary indication of efficacy, with EPO treatment for neuroprotection in the perioperative period for neonatal cardiac surgery, in order to help determine whether larger trials are warranted.
Methods
This was a prospective, randomized, blinded, placebo-controlled trial of EPO (Epogen®, Amgen, Inc., Thousand Oaks, CA) vs. normal saline control (FDA IND 100011, NCT00513240, www.clinicaltrials.gov). The Baylor College of Medicine IRB approved the protocol, and patients were enrolled after informed signed parental consent.
Inclusion criteria were neonates (<30 days) scheduled for cardiac surgery with hypothermic cardiopulmonary bypass (CPB) for greater than 60 minutes. Three anatomic groups were studied: 1. hypoplastic left heart syndrome or variant undergoing Norwood Stage I palliation; 2. D-transposition of the great vessels undergoing arterial switch operation, and 3. interrupted aortic arch with ventricular septal defect, or other complete 2-ventricle anatomic repair including truncus arteriosus, tetralogy of Fallot, or total anomalous pulmonary venous return. Randomization was performed by computer generated random number assignment to EPO or placebo, and was stratified within each of these 3 anatomic groups.
Exclusion criteria were gestational age less than 35 weeks at birth, weight less than 2.0 kg, known recognizable dysmorphic syndrome, surgery not requiring cardiopulmonary bypass, preoperative cardiac arrest, or inability to enroll the patient greater than 12 hours preoperatively. Also excluded were cases where aortic crossclamping was not used, CPB times were anticipated to be less than 60 minutes, and a nadir temperature on bypass greater than 30° C was planned. In addition, patients with contraindications to EPO administration were excluded: Hypertension (sustained systolic blood pressure >100 mm Hg), polycythemia (Hemoglobin >20 g/dL), thrombocytosis (platelet count >600,000 per dL), or evidence of hypercoagulability: INR < 1.0; and patient or maternal history of major thrombosis.[9]
Patients had a detailed neurological examination by a pediatric neurologist, unless precluded by neuromuscular blockade or heavy sedation. For a 12–24 hour period before the scheduled surgery, patients had a near-infrared spectroscopy sensor (Somanetics 5100B, Inc. Troy MI) placed on the right forehead, to measure regional cerebral oxygen saturation (rSO2).
Surgical, anesthetic, and CPB techniques were standardized and have been described in detail previously.[7] Bilateral rSO2 was measured intraoperatively and maintained >90% during CPB. If the rSO2 was less than 50% before or after bypass, attempts were made to increase oxygen delivery to the brain, or decrease oxygen consumption, using a protocol described previously.[7]
Brain magnetic resonance imaging (MRI) under general endotracheal anesthesia was obtained immediately before surgery. MRI was performed on a 1.5 Tesla Intera scanner (Philips Medical Systems, Best, the Netherlands), including standard T1, T2, diffusion weighted imaging, and susceptibility weighted imaging. [7] Postoperative MRI was obtained when the patient was clinically stable, at 7–10 days postoperatively. All MRI were evaluated by pediatric neuroradiologists unaware of diagnosis or surgery. Abnormalities were classified as: white matter injury, intraparenchymal infarction, or intraparenchymal or intraventricular hemorrhage. [7] Sinovenous thrombosis was diagnosed separately, and confirmed by 2 dimensional time of flight MR venography with maximum intensity projection reconstructions. MRI injury definitions and grading scale have been described in detail previously; all injuries were classified as mild, moderate, or severe. [7]
The initial protocol EPO doses were 1000 units/kg intravenously over 60 minutes (or placebo equivalent) in 3 doses: 1. 12–24 hours preoperatively; 2. Immediately after CPB; 3. 24 hours after dose 2. This dose was chosen based on animal model data to provide sufficient brain tissue EPO concentrations to afford neuroprotection.[11] This schema was used in the first 33 patients. The trial was placed on full clinical hold by the U.S. Food and Drug Administration (FDA) because of adverse events in an adult EPO neuroprotection trial for stroke in 50–80 year old patients, not related to this neonatal cardiac surgery trial. After evaluation of safety data, the FDA mandated a dosing regimen change in the current trial. For the last 26 patients, EPO dose was 500 units/kg IV preoperatively, and on postoperative days 1 and 3. Aprotinin was administered to the first 21 patients for antifibrinolysis. Aprotinin marketing was suspended in December 2007, and ε-aminocaproic acid was administered to the last 38 patients in the study.
Clinical data collection preoperatively, intraoperatively, and for the first 72 hours postoperatively included rSO2, hemodynamic and respiratory data, details of CPB technique, and clinical events such as cardiac arrest, extracorporeal membrane oxygenation (ECMO) cannulation, and clinical seizures. The potential adverse effects of EPO include major thrombosis, intracranial hemorrhage, hypertension, thrombocytosis, and polycythemia.[9] These events were defined as above under Exclusion Criteria, and assessed for each patient. Clinical events during the hospital stay were recorded until discharge.
Chromosome analysis was performed by chromosomal microarray, or fluorescence in situ hybridization analysis, when a genetic syndrome was suspected. Anesthetic and sedative drug doses in the operating room, and the first 72 postoperative hours in the intensive care unit were recorded. Records were also reviewed for cardiac arrest, ECMO cannulation, and deaths.
Neurodevelopmental testing was performed with the Bayley Scales of Infant and Toddler Development, Third Edition (PsychCorp.-Harcourt, Brace, & Co., San Antonio TX, 2006) at 12 months of age. The Bayley III consists of three primary composite standard scores, the Cognitive, Motor, and Language Composite scores, measured by performance of specified tasks, scored against a normative population, and scaled to have a mean score of 100 with standard deviation of 15. In addition, a parental questionnaire is administered, and Social-Emotional and Adaptive Behavior Composite Scores are derived. These tests were administered by a single developmental psychologist unaware of diagnosis or surgery performed, or whether EPO or placebo had been administered. Maternal intelligence was evaluated using the Weschler Abbreviated Scale of Intelligence. (PsychCorp.-Harcourt, Brace, & Co., San Antonio TX, 1999).
A Data and Safety Monitoring Board (DSMB), consisting of pediatric specialists in Neurology, Cardiology/Cardiac Intensive Care, Congenital Heart Surgery, Hematology, Neonatology, and Cardiac Anesthesiology, reviewed all safety data and adverse events. Blinding of groups was maintained until the last patient had undergone 12 month Bayley III assessment. The DSMB had access to unblinded data at all times.
Erythropoietin Pharmacokinetic Data
Blood sampling for measurement of EPO pharmacokinetics was performed in consenting patients with indwelling arterial or central venous catheters. Sampling (0.5 ml) was performed immediately prior to the first dose of EPO/placebo, and at 0, 1, 2, 4, 6, 12, and 24 hours after the 1- hour infusion. Immediately following collection, blood was centrifuged at 760 × g for 15 minutes at room temperature and plasma transferred to plastic vials and frozen at −80 C until analyzed. EPO concentrations were determined using the commercially available Quantikine® IVD® (R&D Systems, Inc., Minneapolis, MN) EPO enzyme-linked immunosorbent assay kit. Because the assay does not discriminate between endogenous and recombinant EPO, the baseline EPO level was subtracted from the post EPO administration concentrations.
Sample Size Calculation and Statistical Analysis
This was a pilot phase I/II trial, and the original plan was to enroll 80 patients, to allow for approximately 60 patients to undergo 12 month neurodevelopmental assessments, accounting for deaths and withdrawal from the study. This sample size would be expected to detect a difference of 7.5 points between groups on the Cognitive Score of the Bayley III, with a standard deviation of 15, power of 80% and alpha level of 0.05, using 2-sided T test for comparison. Because the time period to enroll patients was significantly curtailed by the full clinical hold imposed by the FDA, the enrollment goal was reduced to 60, intended to yield 45 patients to evaluate at 12 months. This sample size could only detect a difference of 9 points between groups. Thus this study could only give a preliminary indication of efficacy of EPO treatment on 12 month neurodevelopmental outcomes.
A descriptive analysis of the pharmacokinetic data was planned due to the study size, and elective participation in the pharmacokinetic analysis component of the study, restricted to infants with indwelling catheters capable of blood sampling in the preoperative period. We anticipated that 10–15 patients would meet the inclusion criteria of parental consent, indwelling catheter for blood samples, and ability to enroll patient >24h before scheduled surgery.
Analysis was performed on the intention to treat basis. The primary analysis was the safety profile for the EPO vs. placebo groups, with sinovenous thrombosis, other major thromboses, hypertension, thrombocytosis, and polycythemia as the primary outcomes. Also analyzed were MRI brain injury pre- and postoperatively. Finally, neurodevelopmental outcomes were analyzed, including Cognitive, Language, and Motor Composite Standard Scores of the Bayley Scales III. Data were analyzed using two sided T test, Mann-Whitney Signed Rank Test, Fisher Exact Test, ANOVA, or χ-square analysis as appropriate. Normally distributed data are reported as mean ± SD; non-normally distributed data (Shapiro-Wilk test p<0.05) reported as median (25th–75th percentile). (Stata 12, Stata Corporation LLP, College Station TX).
Results
The study enrolled patients starting in September 2006, and finished enrollment in February 2011. The enrollment flow diagram is displayed in Figure 1. A total of 104 patients met inclusion criteria. Sixty-two patients (60% of those eligible) were enrolled and randomized; 3 patients did not receive intended surgery, and did not have further data collection, leaving 59 patients receiving study drug and perioperative data collection. Patient clinical and perioperative data are presented in Table 1. All patients survived the immediate postoperative period.
Figure 1.
Enrollment and Data Analysis Flow Diagram
Table 1.
Patient and Operative Characteristics
| Parameter | EPO group, n = 32 | Placebo Group, n = 27 | P Value |
|---|---|---|---|
| HLHS (no., %) | 16 (50%) | 10 (37%) | 0.420 |
| D-TGA (no., %) | 9 (28%) | 12 (44%) | - |
| AA+VSD/other 2V (no., %) | 7 (22%) | 5 (19%) | - |
| CPB time (min) † | 191 (163–236) | 194 (169–270) | 0.382 |
| Aortic crossclamp time (min) † | 98 (84–129) | 104 (89–148) | 0.334 |
| DHCA time (min) † | 9 (8–14) (n=19) | 11 (8–14) (n=10) | 0.562 |
| RCP time (min) | 67.0 ± 30.7 (n=23) | 67.9 ± 31.2 (n=14) | 0.930 |
| Lowest CPB temp (° C) † | 17.9 (17.4–22.8) | 18.1 (17.6–24.9) | 0.266 |
| OR fentanyl dose (mcg/kg) † | 182 (154–291) | 198 (158–295) | 0.755 |
| OR midazolam dose (mg/kg) | 0.96 ± 0.46 | 1.21 ± 0.49 | 0.044* |
| Aprotinin use (no., %) | 11 (34%) | 10 (37%) | 0.952 |
| Isoflurane MAC-hours | 1.78 (0.96–2.29) | 1.58 (1.14–2.68) | 0.879 |
| Mean rSO2 (%) | 69.6 ± 7.1 | 71.2 ± 7.4 | 0.394 |
Abbreviations: HLHS, hypoplastic left heart syndrome; D-TGA, dextrotransposition of the great arteries; AA+VSD, hypoplastic aortic arch/ventricular septal defect; CPB, cardio-pulmonary bypass; OR, operating room; DHCA, deep hypothermic circulatory arrest; RCP, regional cerebral perfusion; MAC, minimum alveolar concentration; rSO2, regional cerebral oxygen saturation. Data are mean ± standard deviation if normally distributed;
, data not normally distributed: median (25th–75th percentiles).
p<0.05.
The safety profile was similar between groups. Six patients had postoperative dural sinovenous thrombosis (3 EPO, 3 placebo); and one of these, an EPO patient, had clinical seizures, and a new cerebral infarction in the area adjacent to the thrombosis on postoperative MRI. In the other 5 patients, there were no clinical signs and sinovenous thrombosis was only discovered on the postoperative MRI scan. Laboratory workup for hypercoagulability was negative in all cases. One patient with hypoplastic left heart syndrome who received EPO suffered a cardiac arrest and required ECMO cannulation in the immediate postoperative period. There were no other acute adverse events, including no clinical seizures, coma, major thromboses, hypertension, thrombocytosis, or polycythemia in the immediate postoperative period until hospital discharge. Clinical neurological examination was performed preoperatively by a pediatric neurologist in 44 of 59 patients, and before discharge in 45 of 59 patients. The only abnormality noted was a slight symmetrical increase in lower extremity deep tendon reflexes on the postoperative examination in one placebo patient with transposition of the great arteries who had a moderate new white matter injury, and a new germinal matrix hemorrhage on postoperative MRI.
Six patients died before age 12 months (3 EPO, 3 placebo); all patients had hypoplastic left heart syndrome, and 5 of 6 died in the interstage period before a planned bidirectional cavopulmonary anastomosis, and one after. Eleven patients declined 12 month followup (7 EPO, 4 placebo, p=0.48), leaving 42 patients with 12 month BSID III (79% of survivors).
Brain MRI findings are presented in Table 2. Both preoperative baseline rate of injury, and postoperative outcomes were not different in the EPO patients. There were no differences between groups in the classification of MRI injury severity.
Table 2.
Brain Magnetic Resonance Imaging Findings
| Time Period | Parameter | EPO group, n = 32 | Placebo Group, n = 27 | P Value | |
|---|---|---|---|---|---|
| Preoperative MRI injury (no.,%) | Total with WMI, infarction, hemorrhage | 13 (41 %) | 9 (33%) | 0.759 | |
| WMI | 9 (28%) | 8 (30%) | 0.872 | ||
| WMI severity: mild/moderate/severe (n) | 6/2/1 | 6/2/0 | -- | ||
| Infarction | 6 (19%) | 2 (7%) | 0.269 | ||
| Infarction severity: mild/moderate/severe | 4/1/1 | 2/0/0 | -- | ||
| Hemorrhage | 3 (9%) | 3 (11 %) | 0.652 | ||
| Hemorrhage severity: mild/moderate/severe (n) | 3/0/0 | 3/0/0 | -- | ||
| DSVT | 0 | 0 | -- | ||
| NEW postoperative MRI injury (no., %) | Total with NEWWMI, infarction, hemorrhage | 13(41%) | 13 (48%) | 0.751 | |
| WMI | 8 (25%) | 9 (33%) | 0.678 | ||
| WMI severity: mild/moderate/severe (n) | 6/2/0 | 6/3/0 | -- | ||
| Infarction | 3 (9%) | 5 (19%) | 0.450 | ||
| Infarction severity: mild/moderate/severe (n) | 3/0/0 | 5/0/0 | -- | ||
| Hemorrhage | 7 (22%) | 3 (11 %) | 0.319 | ||
| Hemorrhage severity: mild/moderate/severe (n) | 6/0/1 | 3/0/0 | -- | ||
| DSVT | 3 (9%) | 3 (11 %) | 0.997 | ||
| DSVT severity: mild/moderate/severe (n) | 2/1/0 | 2/1/0 | -- | ||
Abbreviations: WMI, white matter injury; DSVT, dural sinovenous thrombosis. Some patients had more than one type of injury, resulting in totals of WMI, infarction, hemorrhage in each group being greater than the total number of brain injured patients.
The neurodevelopmental outcomes are presented in Table 3. The outcomes of Bayley Scales of Infant Development III Cognitive, Language, and Motor Composite scaled scores were not different between groups. We elected to analyze separately the effect of cardiac diagnosis, the change of EPO dose status, and change in aprotinin administration; results are presented in Tables 4a–4c. The only difference found in these analyses was a higher Cognitive score in the placebo group patients who did not receive aprotinin. (Table 4b) Maternal intelligence quotient for the infants tested at 12 months was 104.5 ± 14.6 in the EPO group, and 103.8 ± 17.2 in the placebo group (p=0.872). The anatomic group of other complete two ventricle repairs (VSD with aortic arch repair, n=5, truncus repair, n=2; tetralogy of Fallot and total anomalous pulmonary venous return, n=1 each) had the lowest overall scores. Of note, 6 of 9 patients in this group had chromosome anomalies, all were microdeletions at the chromosome 22q11.2 region.
Table 3.
Neurodevelopmental Outcomes: 12 month Bayley Scales of Infant and Toddler Development III Composite Scaled Scores
| Parameter | EPO group, n = 22 | Placebo Group, n = 20 | PValue | |
|---|---|---|---|---|
| Cardiac Diagnosis | HLHS (no., %) | 11 (50%) | 6 (30%) | 0.298 |
| D-TGA (no., %) | 6 (27%) | 10 (50%) | - | |
| AA+VSD/other 2V (no., %) | 5 (23%) | 4 (20%) | - | |
| Directly Measured Bayley III Composite Scores | Cognitive | 101.1 ± 13.6 | 106.3 ± 10.8 | 0.187 |
| Language | 88.5 ± 12.8 | 92.4 ± 12.4 | 0.329 | |
| Motor | 89.9 ± 12.3 | 92.6 ± 14.1 | 0.506 | |
| Adaptive Behavior: Bayley III Questionnaire Scores | Social-Emotional† | 95.0 (92.5–105.0) | 100.0 (96.3–108.8) | 0.249 |
| Behavioral | 93.2 ± 10.7 | 97.3 ± 15.7 | 0.342 | |
| Conceptual | 98.7 ± 13.6 | 99.2 ± 13.1 | 0.906 | |
| Social | 97.2 ± 11.4 | 100.7 ± 15.6 | 0.423 | |
| Practical | 89.5 ± 9.1 | 92.8 ± 12.6 | 0.352 |
Abbreviations: HLHS, hypoplastic left heart syndrome; D-TGA, dextrotransposition of the great arteries; AA+VSD, hypoplastic aortic arch/ventricular septal defect. Data are mean ± standard deviation if normally distributed;
, data not normally distributed: median (25th–75th percentiles).
Table 4a.
Bayley Scales of Infant and Toddler Development III Scores by Anatomic/Surgical Group
| Parameter | EPO Group | Placebo Group | P Value |
|---|---|---|---|
| 12 Month Bayley IIICognitive HLHS† (n= 11 EPO, 6 placebo) | 100 (87.5–113.8) | 100 (100–100) | 0.990 |
| 12 month Bayley IIILanguage HLHS | 90.3 ± 15.1 | 90.0 ± 9.7 | 0.969 |
| 12 month Bayley IIIMotor HLHS† | 91.0 (91.0–97.0) | 92.5 (79.0–110.0) | 0.919 |
| 12 Month Bayley IIICognitive D-TGA (n= 7 EPO, 9 placebo) | 103.6 ± 14.1 | 110.0 ± 11.7 | 0.335 |
| 12 month Bayley IIILanguage D-TGA† | 91 (84.5–97.8) | 94 (92.8–98.5) | 0.284 |
| 12 month Bayley IIIMotor D-TGA | 92.4 ± 12.0 | 97.3 ± 12.3 | 0.394 |
| 12 Month Bayley IIICognitive AA+VSD/other 2V (n= 4 EPO, 5 placebo) | 97.5 ± 6.4 | 107.0 ± 13.0 | 0.228 |
| 12 month Bayley IIILanguage AA+VSD/other 2V† | 78.5 (69.5–87.5) | 91.0 (76.3–91.8) | 0.190 |
| 12 month Bayley III Motor AA+VSD/other 2V | 85.0 ± 12.7 | 86.8 ± 11.7 | 0.832 |
Abbreviations: HLHS, hypoplastic left heart syndrome; D-TGA, dextrotransposition of the great arteries; AA+VSD, hypoplastic aortic arch/ventricular septal defect. Data are mean ± standard deviation if normally distributed;
, data not normally distributed: median (25th–75th percentiles).
Table 4c.
Bayley III Scores by Aprotinin Administration Status
| Parameter | EPO Group | Placebo Group | P Value |
|---|---|---|---|
| 12 Month Bayley III Cognitive with aprotinin (n= 8 EPO, 8 placebo) | 106.9 ± 15.3 | 103.1 ± 12.8 | 0.604 |
| 12 month Bayley III Language with aprotinin | 88.1 ± 14.8 | 87.4 ± 8.3 | 0.902 |
| 12 month Bayley III Motor with aprotinin | 94.3 ± 12.3 | 87.3 ± 15.3 | 0.335 |
| 12 Month Bayley III Cognitive no aprotinin (n= 14 EPO, 12 placebo) | 97.9 ± 11.9 | 108.3 ± 9.1 | 0.020* |
| 12 month Bayley III Language no aprotinin | 88.7 ± 12.1 | 95.7 ± 13.8 | 0.184 |
| 12 month Bayley III Motor no aprotinin | 87.4 ± 12.1 | 96.2 ± 12.4 | 0.080 |
Abbreviations: EPO, erythropoietin. Data are mean ± standard deviation.
p<0.05.
Table 4b.
Bayley III Scores by EPO Dose: 1000 units/kg Every Day vs. 500 units/kg Every Other Day vs. Placebo
| Parameter | EPO 1000 QD, n = 11 | EPO 500 QOD, n = 11 | Placebo, n = 20 | P Value |
|---|---|---|---|---|
| 12 Month Bayley III Cognitive | 101.4 ±16.9 | 100.9 ±10.2 | 106.3 ±10.7 | 0.422 |
| 12 month Bayley III Language | 85.0 ± 16.3 | 92.0 ± 7.3 | 92.4 ± 12.4 | 0.268 |
| 12 month Bayley III Motor | 89.3 ± 15.7 | 90.5 ± 8.6 | 92.6 ± 14.1 | 0.787 |
EPO, erythropoietin; EPO 1000 QD; EPO 1000 units/kg every day for 3 doses; EPO 500 QOD, EPO 500 units/kg every other day for 3 doses. Comparison by one-way analysis of variance.
Three patients had pharmacokinetic data obtained; 1 placebo and 2 patients who received EPO 1000 units/kg. Pharmacokinetic modeling was not performed because of these small patient numbers; however, maximum EPO plasma concentrations (Cmax) were 8415 and 5447 mIU/ml in the 2 patients receiving EPO.
Discussion
EPO treatment was not associated with an overall difference in complications, including major intracranial thromboses, hemorrhage and other MRI findings, clinical adverse outcomes, and death. 12 month Bayley III scores were not different with EPO treatment; because of the Phase I/II pilot study design, this trial was not powered to demonstrate a neurodevelopmental outcome difference. Maximum EPO plasma levels in the two patients where they were obtained demonstrated concentrations consistent with neuroprotective levels in animal models for one patient, and concentration below these levels in the other.[11]
This EPO neuroprotection study is the first published study in the neonatal and infant cardiac surgery population to apply a pharmacologic neuroprotection strategy, measure cerebral physiology with near infrared spectroscopy, evaluate the structural changes of the brain with MRI, and assess the functional outcomes with longer term neurodevelopmental outcomes at age 12 months. The only previously published randomized trial of pharmacologic neuroprotection in this population was Clancy et al’s study of allopurinol as an oxygen free radical scavenger vs. placebo for neonates undergoing deep hypothermic circulatory arrest for Norwood stage I palliation or other complex repairs. [12] There was no effect of allopurinol treatment on the primary neurological outcome: seizures, coma, or death. When cardiac outcomes were included, the hypoplastic left heart syndrome group did have an improvement with allopurinol treatment. No imaging or longer term outcomes were assessed.
EPO treatment shows promise to improve neurodevelopmental outcomes in other neonatal clinical settings of brain injury. A prospective randomized controlled trial in 153 full term neonates diagnosed with hypoxic ischemic encephalopathy demonstrated an improvement in neurodevelopmental outcomes at age 18 months after EPO treatment every other day for 2 weeks.[10] Death or moderate/severe disability occurred in 44% infants in the control group and 25% infants in the EPO group (p=0.017). EPO improved outcomes only for infants with moderate hypoxic ischemic encephalopathy (p= 0.001) and not those with severe injury (p=0.227) A second small case control trial (n=45) in birth asphyxiated full term neonates found that EPO, 2500 units/kg subcutaneously for 5 days, resulted in fewer neurological and developmental abnormalities at age 6 months, than control patients.(27% vs. 71%, p=0.03)[13] In an EPO neuroprotection study in extremely low birthweight neonates (<1000g), 17 survivors received either 500, 1000, or 2500 units/kg IV EPO daily for the first 3 days of life. Compared to the 18 controls receiving no EPO, both Cognitive (p=0.044) and Motor (p=0.026) scores were higher with EPO, from 8 to 36 months corrected gestational age. [14]
There is recent data demonstrating apoptosis in response to common anesthetic agents that bind γ-aminobutyric acid and N-methyl-D-aspartate receptors to produce their effects, including halogenated anesthetic gases, benzodiazepines, and ketamine. These drugs are frequently used in large doses in neonatal cardiac surgery, and increased exposure to these agents is associated with worse neurodevelopmental outcomes in neonatal cardiac surgery, even after adjustment for important covariates. [15] There is now animal model data demonstrating that EPO protects against anesthetic induced apoptosis and that anesthetics themselves suppress endogenous EPO response to hypoxia.[16] These potential benefits of EPO treatment, as well as the demonstration of adequate safety profiles of EPO in the current trial, suggest that EPO in adequate doses should be studied in a larger cohort of neonatal cardiac surgery patients, with uniform dosing and treatment protocols, to determine its value as a neuroprotectant in this setting.
Limitations of this study that affect the conclusions reached with regard to the original intent of this phase I/II trial can be summarized as follows. First is the change in EPO dosing to levels that may not be neuroprotective. Second is the reduction in the planned number of patients due to the limited study period available. Third, timing of EPO dosing with regard to the potential insult: a broader dosing window could be more desirable, given the evidence for significant incidence of preoperative MRI brain injury, and the possibility that injury occurred later than postoperative day 3. Fourth, the lack of adequate numbers of patients for pharmacokinetic data, and lack of data with regard to EPO pharmacokinetics with cardiopulmonary bypass. Finally, six of 42 patients had chromosome 22q11.2 partial deletion, with known but variable association with neurodevelopmental problems. [17]
The limitations noted above, and the small size of this EPO study preclude making any conclusions about EPO neuroprotection in the neonatal cardiac surgery population. However, as a phase I/II trial it is useful in that it demonstrates no safety concerns of higher dose EPO administration, and gives important insight into optimized study design features for a future EPO neuroprotection trial. Besides studying a population large enough to determine a meaningful difference in the primary neurodevelopmental outcome measure, these include the following:
More optimal EPO dosing, with the higher doses more likely to provide benefit.
More complete pharmacokinetic data, necessary because of the unique features in this population, i.e. bypass.
More effective exclusion criteria and/or screening to account for enrollment of subjects with chromosome 22q11.2 microdeletions and other chromosomal syndromes, at least assuring equivalent numbers in each treatment group. These patients clearly affect the outcomes. [17]
Plan for stratification based on rate of preoperative MRI injury.
Refinement of EPO dosing regimen given the issue of preoperative injury, with consideration to starting EPO dosing shortly after birth/diagnosis/randomization, and a longer period of treatment, i.e. up to 2 weeks.
Optimal primary outcome variable: timing of neurodevelopmental testing at 12 months vs. later for greater predictive ability,[18] issues with Bayley III calibration (newer version of the test yields higher Cognitive Score than the Mental Development Index of the Bayley II) [19], and utility of MRI (recent data demonstrates an association with perioperative MRI brain injury and lower neurodevelopmental scores).[20]
Stratification and randomization according to anatomic diagnoses with known worse neurodevelopmental outcomes, i.e. hypoplastic left heart syndrome vs. other diagnoses.[21]
Sample size considerations for an optimized EPO study include choosing an appropriate primary outcome variable, i.e. Cognitive Composite Score of the Bayley III at 18–24 months, powered to detect a difference of 5 points higher with EPO treatment, with a standard deviation of 15. Assuming a power of 85% and alpha level of 0.05, 326 evaluable subjects would be required, using 2-sided T-test for comparison. This in turn would likely mean that 435 patients would be required to enroll in the study, accounting for a 25% death or study dropout rate. This number would increase with stratification by anatomic subgroup, or for interim analyses for safety purposes. Such a study very likely requires multicenter participation to enroll patients over a brief enough window of time to ensure that outcomes do not change substantially because of other unrelated improvements in therapy.
Conclusions
In this phase I/II trial, EPO treatment as a neuroprotectant for complex neonatal cardiac surgery was not associated with a different safety profile than placebo, including major intracranial thromboses, hemorrhage, other MRI injuries, and death. EPO treatment was not associated with an overall difference in 12 month Bayley Scales of Infant and Toddler Development III scores in this small group after analysis by the intent to treat basis; this study was not powered to demonstrate an outcome difference. Because of the limitations of changing EPO dose and aprotinin strategy during the study, a larger controlled trial would be required to definitively address the status of EPO neuroprotection for neonatal cardiac surgery. In addition, the significant incidence of well established preoperative brain injury could potentially be addressed with earlier EPO treatment, i.e. immediately after birth if the patient had a prenatal diagnosis. Because we observed no differences in adverse effects with EPO 1000 units/kg vs. 500 units/kg, and doses of at least 1000 units/kg are likely to be needed for neuroprotection, we advocate a higher dose EPO strategy for future trials in neonatal cardiac surgery. An optimized study design, likely in a multicenter setting, will be required to define the utility of EPO neuroprotection in the neonatal cardiac surgery population. We advocate the pursuit of such a study because of the many desirable properties of EPO for neuroprotection, and its demonstrated efficacy in other neonatal settings of cerebral injury.
Acknowledgments
Data and Safety Monitoring Board: Donna Ferreiro (Chair), Stephen Roth (Secretary), Robin Ohls, Emad Mossad, James DiNardo, Patti Massicotte, Tom Karl, J. William Gaynor
Funding Sources: NIH Eunice Kennedy Shriver National Institute of Child Health and Development Grant 1R21- HD55501-01, Baylor College of Medicine General Clinical Research Center Grant #0942, funded by NIH M01 RR00188, Charles A. Dana Foundation Brain and Immuno-Imaging Grant, and Texas Children’s Hospital Anesthesiology Research Fund (PI D. Andropoulos).
Footnotes
Financial Disclosures: None.
Clinical Trial Registration: www.clinicaltrials.gov; NCT00513240
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errorsmaybe discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Menache CC, du Plessis AJ, Wessel DL, Jonas RA, Newburger JW. Current incidence of acute neurologic complications after open-heart operations in children. Ann Thorac Surg. 2002;73:1752–8. doi: 10.1016/s0003-4975(02)03534-8. [DOI] [PubMed] [Google Scholar]
- 2.Austin EH, 3rd, Edmonds HL, Jr, Auden SM, Seremet V, Niznik G, Sehic A, et al. Benefit of neurophysiologic monitoring for pediatric cardiac surgery. J Thorac Cardiovasc Surg. 1997;114:707–15. doi: 10.1016/S0022-5223(97)70074-6. [DOI] [PubMed] [Google Scholar]
- 3.Snookes SH, Gunn JK, Eldridge BJ, Donath SM, Hunt RW, Galea MP, et al. A systematic review of motor and cognitive outcomes after early surgery for congenital heart disease. Pediatrics. 2010;125:e818–27. doi: 10.1542/peds.2009-1959. [DOI] [PubMed] [Google Scholar]
- 4.Jonas RA, Wypij D, Roth SJ, Bellinger DC, Visconti KJ, du Plessis AJ, et al. The influence of hemodilution on outcome after hypothermic cardiopulmonary bypass: results of a randomized trial in infants. J Thorac Cardiovasc Surg. 2003;126:1765–74. doi: 10.1016/j.jtcvs.2003.04.003. [DOI] [PubMed] [Google Scholar]
- 5.Wypij D, Newburger JW, Rappaport LA, duPlessis AJ, Jonas RA, Wernovsky G, et al. The effect of duration of deep hypothermic circulatory arrest in infant heart surgery on late neurodevelopment: the Boston Circulatory Arrest Trial. J Thorac Cardiovasc Surg. 2003;126:1397–403. doi: 10.1016/s0022-5223(03)00940-1. [DOI] [PubMed] [Google Scholar]
- 6.McQuillen PS, Barkovich AJ, Hamrick SE, Perez M, Ward P, Glidden DV, et al. Temporal and anatomic risk profile of brain injury with neonatal repair of congenital heart defects. Stroke. 2007;38:736–41. doi: 10.1161/01.STR.0000247941.41234.90. [DOI] [PubMed] [Google Scholar]
- 7.Andropoulos DB, Hunter JV, Nelson DP, Stayer SA, Stark AR, McKenzie ED, et al. Brain immaturity is associated with brain injury before and after neonatal cardiac surgery with high-flow bypass and cerebral oxygenation monitoring. J Thorac Cardiovasc Surg. 2010;139:543–56. doi: 10.1016/j.jtcvs.2009.08.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.McPherson RJ, Juul SE. Erythropoietin for infants with hypoxic-ischemic encephalopathy. Curr Opin Pediatr. 2010;22:139–45. doi: 10.1097/MOP.0b013e328336eb57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Xiong T, Qu Y, Mu D, Ferriero D. Erythropoietin for neonatal brain injury: opportunity and challenge. Int J Devl Neuroscience. 2011;29:583–591. doi: 10.1016/j.ijdevneu.2010.12.007. [DOI] [PubMed] [Google Scholar]
- 10.Zhu C, Kang W, Xu F, Cheng X, Zhang Z, Jia L, et al. Erythropoietin improved neurologic outcomes in newborns with hypoxic-ischemic encephalopathy. Pediatrics. 2009;124:e218–226. doi: 10.1542/peds.2008-3553. [DOI] [PubMed] [Google Scholar]
- 11.Statler PA, McPherson RJ, Bauer LA, Kellert BA, Juul SE. Pharmacokinetics of high-dose recombinant erythropoietin in plasma and brain of neonatal rats. Pediatr Res. 2007;61:671–675. doi: 10.1203/pdr.0b013e31805341dc. [DOI] [PubMed] [Google Scholar]
- 12.Clancy RR, McGaurn SA, Goin JE, Hirtz DG, Norwood WI, Gaynor JW, et al. Allopurinol neurocardiac protection trial in infants undergoing heart surgery using deep hypothermic circulatory arrest. Pediatrics. 2001;108:61–70. doi: 10.1542/peds.108.1.61. [DOI] [PubMed] [Google Scholar]
- 13.Elmahdy H, El-Mashad AR, El-Bahrawy H, El-Gohary T, El-Barbary A, Aly H. Human recombinant erythropoietin in asphyxia neonatorum: pilot trial. Pediatrics. 2010;125:e1135–1142. doi: 10.1542/peds.2009-2268. [DOI] [PubMed] [Google Scholar]
- 14.McAdams RM, McPherson RJ, Mayock DE, Juul SE. Outcomes of extremely low birth weight infants given early high-dose erythropoietin. J Perinatol. 2012 Jun 21; doi: 10.1038/jp.2012.78. [Epub ahead of print] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Andropoulos DB, Easley RB, Brady K, Stayer S, Meador M, Eisenman C, et al. Cumulative volatile anesthetic exposure and neurodevelopmental outcomes at age 12 months after neonatal cardiac surgery. Anesthesiology. 2012;117:A798. (Abstract) [Google Scholar]
- 16.Tanaka T, Kai S, Koyama T, Daijo H, Adachi T, Fukuda K, et al. General anesthetics inhibit erythropoietin induction under hypoxic conditions in the mouse brain. PLoS One. 2011;6(12):e29378. doi: 10.1371/journal.pone.0029378. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Atallah J, Joffe AR, Robertson CM, Leonard N, Blakley PM, Nettel-Aguirre A, et al. Two-year general and neurodevelopmental outcome after neonatal complex cardiac surgery in patients with deletion 22q11.2: a comparative study. J Thorac Cardiovasc Surg. 2007;134:772–9. doi: 10.1016/j.jtcvs.2007.03.007. [DOI] [PubMed] [Google Scholar]
- 18.Harris SR, Megens AM, Backman CL, Hayes VE. Stability of the Bayley II Scales of Infant Development in a sample of low-risk and high-risk infants. Dev Med Child Neurol. 2005;47:820–3. doi: 10.1017/S0012162205001738. [DOI] [PubMed] [Google Scholar]
- 19.Lowe JR, Erickson SJ, Schrader R, Duncan AF. Comparison of the Bayley II Mental Developmental Index and the Bayley III Cognitive Scale: are we measuring the same thing? Acta Paediatr. 2012;101:e55–8. doi: 10.1111/j.1651-2227.2011.02517.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Andropoulos DB, Easley RB, Brady K, McKenzie ED, Heinle JS, Dickerson HA, et al. Changing expectations for neurological outcomes after the neonatal arterial switch operation. Ann Thorac Surg. 2012 Jun 28; doi: 10.1016/j.athoracsur.2012.04.050. [Epub ahead of print] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Gaynor JW, Gerdes M, Nord AS, Bernbaum J, Zackai E, Wernovsky G, et al. Is cardiac diagnosis a predictor of neurodevelopmental outcome after cardiac surgery in infancy? J Thorac Cardiovasc Surg. 2010;140:1230–7. doi: 10.1016/j.jtcvs.2010.07.069. [DOI] [PMC free article] [PubMed] [Google Scholar]