The Immediate Impact of Therapeutic Hypothermia on Cardiac Function of Neonates with Hypoxic Ischemic Encephalopathy

Introduction

Perinatal hypoxia in the form of hypoxic-ischemic encephalopathy (HIE) occurs in 3–5 of 1,000 neonates in developed countries  and is significantly higher in developing countries up to 25 of 1,000 neonates [1] {Gupta, 2019 #20}. Cardiovascular impairment which may, be present sub clinically is a frequent complication among neonates with HIE up to 80% of infants [2, 3]. Hypoxia adversely affects the myocardium either in the primary insult and or during the reperfusion phases of injury. There is no doubt that, sick myocardium result in circulatory insufficiency which delays the brain reoxygenation and might aggravate multi-organ damage [4, 5].

Although therapeutic hypothermia (TH) had been established as the standard of care treatment for neonatal HIE as neuroprotective, the effect of whole-body cooling on myocardial function is still challenging [6, 7]. 

Whole body cooling could be achieved by many means, and Blanketrol-III hyper-hypothermia system offers simple programmable body temperature with minimal fluctuations in water temperature [8, 9].  However, it is still expensive and not available for all asphyxiated neonates especially in low- and middle-income countries. A reduced fractional shortening after perinatal asphyxia has been found in some studies [10] but not in others [11]. Tissue Doppler derived indices as myocardial performance index (MPI) could be more sensitive and reliable tools for assessment of cardiac function in newborn [11].

Aim of the study

The current study aimed to compare the myocardial function before and immediately after therapeutic hypothermia for neonates with HIE using echocardiography and TDI derived MPI.  

Patients and Methods

Study design and setting

The current study is a prospective study, approved by ethical committee of our institute according to Declaration of Helsinki with IBR number; 0107176. All cooled neonates with HIE, born at Alexandria Main Maternity Hospital (AMMH), admitted to the neonatal intensive care unit (NICU) of Alexandria main university hospital (AUH) over one year period; from April 2022 to April 2023 were enrolled in the present study after informed parental consent.

Patients: All the enrolled neonates with HIE between gestational age 35– 42 weeks, were selected consecutively and planned for therapeutic hypothermia with respect to the criteria of The American Academy of Pediatrics (AAP) and the American College of Obstetricians and Gynecologists (ACOG) of birth asphyxia [12].

According to NICU protocol in our institute, neonates with mild HIE and low Apgar score (< 6) and or poor arterial blood gases were also eligible for the present study [13]. Neonates with congenital cyanotic heart disease, multiple congenital anomalies, GA34 weeks or birth weight <1800 gram were excluded.

Methods: Neonates with HIE received whole-body cooling within the first 6 hours after birth, with a target rectal temperature between 33.5 °C and 34.5 °C for 72 h under close monitoring with a probe and were then rewarmed to 37 °C at a rate of 0.5 °C/h.

Active cooling is done either through Blanketrol-III device or through passive cooling combined with hot and cold jell packs. The choice between the two methods is related to the availability of the device. If the Blanketrol III is already was occupied by a patient, we made for use passive cooling combined with cold jell packs. The Blanketrol III hyper-hypothermia system is a product of GENTHERM Company (serial No: 106130311862072(21)202311419) with minimal fluctuations in water temperature. While, effective passive cooling method was done according to Queensland guidelines through turning off the warmer or incubator, removing clothes, and not covering the baby with a blanket and adding cool packs over his body [14].

Patients' data: In the current work, one hundred and eleven neonates with HIE were eligible. However, only one hundred patients completed the study till the third scan echocardiography as illustrated in figure (1). Patient data were obtained prospectively from medical records, including gestational age, Apgar score at 1, 5 and 10 minutes, complete blood count (CBC), initially measured cardiac Troponin I (cTnI) (Roche Diagnostics, GmbH, Mannheim, Germany; cTnI levels were determined with a one- step enzyme-linked immunosorbent assay (ELISA) applied to the RxL Dimension analyser. The lower limit of detection was 0.01 ng/ml as declared by the manufacturers.), arterial blood gas values, and history of the inotropic medication. For each participant, three echocardiography scans were done for assessment of myocardial function. The first was on day 1, in the first 6 hours of life and before cooling (intervention), the second scan was at 48 hours, during cooling, and the third scan was on day 4, after rewarming and all these data were compared.

Echocardiography: Echocardiography was performed using GE Vivid IQ premium machine (serial No: 5720863 S1727016 Rev6), through 5–11 MHz GE, 12 S-RS probes, with M mode, 2D-dimentional, Pulsed wave Doppler (PW) and Pulsed Wave Tissue Doppler (TDI) modalities by single operator. Biventricular systolic and diastolic function were evaluated through measuring the following parameters; Ejection fraction (EF: Normal >55%), Fractional shortening (FS: Normal >28%), Tricuspid annular plane systolic excursion (TAPSE: >8mm), E/A ratio (normal>0.8) and MPI by TDI for left and right ventricles (RV-MPI <0.42 ± 0.14) (LV-MPI <0.37 ± 0.10) in first days of life [15-18]. In addition to, measurement of parameters of pulmonary hypertension defined as tricuspid regurgitant velocity ≥2.5 m/s and/or bidirectional PDA with assessment of pulmonary acceleration time (PAAT) [19]

The patients were further subdivided into two subgroups according to the modality of TH: subgroup I (Ice/Pack) and subgroup II (Blanketrol-III). The clinical profile, laboratory data and the cardiac function of both subgroups were compared. 

Ethics Approval

The study was approved by the Faculty of Medicine, Alexandria University Research Ethics Committee, and the confidentiality of patients’ data was respected.  The ethical approval number is 0107176 on 21-4 -2022.

Informed parental consents were obtained from caregivers of all enrolled newborns.

Statistical analysis:

Data were fed to the computer and analyzed using IBM SPSS software package version 26.0. The Kolmogorov-Smirnov and Shapiro-Wilk tests were used to verify the normality of distribution. Qualitative data and quantitative data were described. Significance of the obtained results was judged at the 5% level. Chi-square test, Fisher’s Exact or Monte Carlo correction, Student t-test, Mann Whitney test, Repeated measures ANOVA Friedman’s, ANOVA test, Cochran’s Q, Pearson coefficient and Spearman coefficient.

Results

Echocardiography: The RV systolic and diastolic performance, including (TAPSE, E/A) ratio as well as RV-MPI increased significantly along the clinical phases of TH (before cooling, after 48 hours of cooling and after rewarming), P value< 0.001 as shown in table 3. Also, LV-EF and LVFS were significantly improved after rewarming when compared to corresponding values before TH, P value< 0.001 as shown in table (3). While, the mean of LV E/A and LV-MPI showed no statistical difference before cooling and after rewarming as shown in table (3).

A negative correlation between core body temperature and LV-EF (r = -0.204, P= 0.041) and positive correlation between core body temperature and LV-MPI (rho = 0.408, P= <0.001^*)

 No other correlations were found between echo variables and core body temperature during cooling or with oxygen saturation before cooling.

Of noted, in the current study, forty-seven cases had TH by Blanketrol III device (sub-group II) and another fifty three cases had TH by passive cooling/ Ice Jell packs (sub-group I). No significant difference was found in the clinical or laboratory profile of both subgroups I & II except in LDH, CRP, platelet count and AST.

Figure (2) showed the pattern of heart rate variability in both subgroups along phases of TH and it was almost similar with no significant difference.  Comparing echocardiographic variables of subgroups, I & II revealed no significant difference in echo variables of RV function during cooling or after rewarming as shown in tables (4 & 5). While, for LV, the EF and LV-MPI were significantly changed, respectively in subgroup II compared to subgroup I during TH and after rewarming as shown in tables (4 & 5).

Discussion

HIE is a public health problem in developing countries and many studies had reported the impaired cardiac performance in neonates with HIE [3, 5]. Although, TH  has been established as the standard of care treatment for neuroprotection for neonates with HIE, still it is debatable for cardiac protection. Optimizing cardiovascular system (CVS) during TH may improve the long-term outcomes of these infants. It remains unclear whether TH promotes CVS recovery in these infants or contributes to progressive CVS dysfunction especially with associated sinus bradycardia. This challenged us to conduct this work to explore the immediate effect of TH on the cardiac function of neonates with HIE post cooling.

The cardiac performance of one hundred neonates with HIE (mean gestational age 38±1 weeks and birth weight 3.2±0.4 Kg) was initially evaluated by monitoring heart rate, BP, cardiac enzymes and echocardiography. Whole body cooling started within 6 hours after birth then echocardiography was re-evaluated and compared during TH and after cooling.

In the current study, high troponin level sampled within the first 4 hours was found among 54% of the study population and this is explained by the severity of perinatal asphyxia and failure of adaptation of cardiovascular system to hypoxia thus, brain and myocardium develop ischemic lesions [10, 20]. Similarly found by Matter et al., who found significantly higher troponin I in neonates with HIE versus healthy controls.  It is important to mention that sepsis involved only 13% of the cases while PPHN involved 19% [11].

The initial levels of heart rate and blood pressure before TH were subnormal values as being affected by perinatal hypoxia [21]. In consistent to Vergales et al., who conducted a study over 37 neonates and found depressed heart rate in first 24 hour after birth explained by autonomic dysfunction secondary to HIE [22].  In the present study, evaluation of RV function by conventional echo and TDI-derived MPI showed significant improvement along the clinical phases of TH (before cooling, after 48 hours of cooling and after rewarming). On the other side, LV had showed significant improvement in systolic function (EF, FS) at day 4 after rewarming but not yet normalized. While TDI-derived MPI of LV and diastolic function showed no significant changes along the three echo scans. That might need longer time for follow up.

In agreement with Giesinger et al., who studied fifty-three neonate with HIE with a mean gestation age of 38.8± 2.0 weeks and birthweight  of 3.33±0.6 kg and reported that the measures of RV systolic performance (TAPSE and fractional area change) were initially  lower than published normative data then increased significantly by time during cooling and after rewarming [23]. However, no change was identified over time in EF or FS among the cases.  In contrast to our results of the LV. This might be explained by different severity of HIE among study population as his study included 25% mild, 50% moderate and 25% severe HIE cases versus 43% mild, 51% moderate and 6% severe HIE cases in the current study.

Yajamanyam et al., conducted a study over 20 term infants receiving TH on days 1–3 and demonstrated that TDI-derived MPI of LV was higher in cases compared to controls before cooling [24]. After rewarming, MPI of LV has improved but not yet normalized. While FS and TDI-derived MPI of RV was similar between cases and controls all through. The study population included only moderate and severe HIE cases and he stated that LV dysfunction might persist for longer duration post cooling.

Also, Nestaas et al., showed impaired myocardial function in Forty-four infants with HIE by low strain-rate indices but it improved significantly at day 4 cooled for 72hours [25].

On the other hand, Rodriguez et al., 2022 found significant improvement of TD-LV-MPI after cooling and not for RV-MPI [26]. They also concluded that FS was not useful to determine systolic dysfunction as it was not changed and recommended the use of EF.

Furthermore, TR gradient, PAAT and IVC-CI were assessed to estimate PH. Initially the values were low, then worsened by cooling and finally improved after rewarming. Thus, TH regardless the method of cooling improves pulmonary hypertension parameters [27].

In light of the above mention studies, impact of TH on the cardiovascular system was evidenced through improving the hemodynamic of the neonate, (HR) to normal values, BP and the cardiac dysfunction regardless the used tool for identification and also PHT. This could be explained as hypothermia possibly decreases the ATP consumption, decreases oxygen free radicals and inflammatory mediators involved in tissue injury [28]. Comparing the subgroup, I and II, there was no statistically significant difference as regard mode of delivery, anesthesia, resuscitation data and also APGAR score at 1 minute or at 5 minutes, but there was significant difference as regard APGAR score at 10 minutes. However, no significant difference was found between the two groups regarding patients scored below 7 or those scored above 7. No significant difference was found between the two groups regarding hypoxia parameters or cardiac enzymes. However, LDH, CRP, platelet count and AST were significantly different between the two groups but these are not specific for myocardial injury [29].

Surprisingly, the heart rates showed the same pattern of changes in both study subgroups.  Also, insignificant difference was observed in echocardiographic variables of cardiac function between Blanketrol-III subgroup when compared to Ice/pack subgroup during the second and third scan phase respectively, except for EF, LV-MPI and TR gradient.  In which, LV-MPI and TR were greater after rewarming of Ice/pack cooling rather than Blanket-Rol- III subgroup. However, longer-term follow up for larger scale study might show better improvement.

Up to our knowledge, this first study in literature demonstrating the comparable cardioprotective effect of the two cooling modalities.

Study limitation: This study has some limitation, First: short follow up time. Further longer follow up and larger scales studies are recommended to evaluate role of different cooling modality on the cardiac protection as well as on the other organs. Detailed Tissue Doppler and speckle tracking echocardiographic studies are recommended for further cardiac evaluation.

Conclusions

Therapeutic hypothermia might be cardioprotective in neonates with HIE proved by echocardiographic parameters especially MPI.  It is possible for perinatal hypoxia to harm a newborn's myocardium, despite mild encephalopathy. Ice/pack cooling has a very close protective effect on heart of neonate with HIE, almost similar to Blanket-Rol III modality that’s highly recommends its use especially in absence of device cooling modalities.