Cardiovascular Disorders and Risk Factors in Childhood Cranial and Craniospinal Tumors Survivors

Cardiovascular disease, after recurrence of the original cancer and development of second primary cancers, has been reported to be the leading cause of premature mortality among long-term childhood cancer survivors. Childhood cancer survivors after chemotherapy and radiotherapy for cranial and craniospinal tumors are at increased risk of transient ischemic attack, ischemic stroke. The mechanisms of vascular damage are not well understood. Preventive measures are not developed.

The purpose of my research is to study to adverse cardiovascular disorder and risks factors in childhood cranial and craniospinal tumors survivors.

Cardiovascular risk factors monitoring and modification, development of nursing and follow-up guides of adult patients may improve outcomes in these patients after treatment for cranial and craniospinal tumors in childhood. Attempts to develop more individualized risk prediction for cardiovascular disease may help refine surveillance and counseling in the future.

This was an open prospective observational study. 68 people were enrolled in the study: 48 childhood cancer survivals after complex management of cranial and craniospinal tumors formed the main group, and 20 healthy people formed the control group.

Criteria of inclusion to the main group were: age 16-40 years, radiation and chemotherapy for cranial and craniospinal tumors in childhood or adolescence, completion of the therapy at least one year before enrollment, signed informed consent. There were three separate types of informed consent - for adults, for minors and for their parents. Minors were signing informed consent in the presence of their parents.

Criteria of inclusion to the control group for healthy volunteers were demographics comparable to the main group.

Criteria of exclusion were: standard contraindications to exercise testing, anemia, pregnancy, psychiatric disease, alcohol or drug abuse, active malignancy, acute infection.

The study has been approved by the local ethics committee. At baseline, all patients underwent comprehensive evaluation including collection of medical history, complete physical exam, 12-lead ECG, hormone and lipid blood tests, echocardiography (Echo-CG), cardiopulmonary exercise testing (CPET) and pulse wave analysis. Additional carotid duplex scanning was performed in patients after craniospinal radiation therapy.

Echocardiography Left ventricular (LV) systolic and diastolic function, cardiac chamber volumes, valve function and systolic pulmonary artery pressure were evaluated using conventional transthoracic 2D-echocardiography (Dimension/Vivid 7 PRO, General Electric Medical System, Norway) according to current guidelines. Left ventricle (LV) end diastolic volume (LVEDV), end systolic volume (LVESV) and ejection fraction (LVEF) were measured with the modified Simpson method.

Cardiopulmonary exercise testing CPET was conducted on a treadmill (Intertrack 8100, Schiller, Switzerland) with a breath-by-breath algorithm of gas exchange analysis (Cardiovit CS-200 Ergo-Spiro, Schiller, Switzerland), using Bruce, modified Bruce or Naughton protocols depending on the level of physical tolerance. The researchers evaluated VO2 peak (highest oxygen uptake attained during exercise) and METs as its derivative, percentage from predicted VO2 peak (based on age, sex, height, and weight), anaerobic threshold (AT, defined by the V-slope method as the level of oxygen uptake at the moment of dislinear rise in minute ventilation and CO2 output relative to VO2, reflecting the shift to anaerobic metabolism and bicarbonate buffering of increased lactic acid), O2 pulse (VO2 peak/peak heart rate), ventilatory equivalent for carbon dioxide (VE/VCO2 slope), end-tidal CO2 pressure at rest and at peak exercise (PetCO2 rest and PetCO2 peak). Patient's effort was considered substantial if the self-reported number using a 1-10 modified Borg scale was ≥ 7 points and respiratory exchange ratio (RER, defined as VCO2/VO2 ratio) was ≥1.0.

Endothelial function assessment Pulse wave characteristics were obtained using finger photopletysmography (PPG) (Angioscan-01, AngioScan-Electronics) from an infrared light sensor. The amount of light transmitted to the finger directly depends on the alterations in its blood flow. The signal obtained during PPG is displayed as a digital volume pulse (DVP) waveform. This wave usually demonstrates two deflections: an early systolic peak (a), reflecting a direct pressure wave propagation from the left ventricle to the finger, and the diastolic peak (b), that reflects backward pressure wave propagation to the aorta. The time between these two peaks (peak-to-peak time, PPT) is proportional to the subject's height. We measured large artery stiffness index (SI, a ratio between the length of aorta calculated automatically based on height to PPT), reflection index of small resistive arteries (RI, calculated as b/a x 100%), augmentation index (AI, defined as the ratio between late and early systolic pulse wave peaks, expressed as percent from the pulse pressure) and augmentation index normalized for pulse rate of 75 beats per minute (AIp75%) with regard to the fact that the heart rate has a major impact on the augmentation of pulse pressure. Endothelial function was also evaluated using occlusion index and shear rate during post-occlusive reactive hyperemia (PORH) representing the magnitude of limb reperfusion after a brief period of brachial artery occlusion. The probe was performed in the morning hours in supine position after 15 minutes of rest in a comfortable quiet warm environment according to a standardized protocol on subject preparation (fasting state, restriction of moderate-to-high physical activity, nicotine, alcohol, caffeine or other vasoactive drugs 24 hours before the probe).

Statistical analysis Microsoft Excel spreadsheet application was used for data storage and preparation for future analysis. Statistical analysis was performed using Prism 9.2.0 software. Python software was applied for multiple logistic regression models. The continuous data were expressed as a mean value (M) ± standard deviation (SD) with a 95% confidence interval (CI). Categorical variables were presented in absolute numbers and percentage and computed by using two-tailed Fischer's exact test. Normality of distribution for quantitative variables was tested using Anderson-Darling test as the first step and using Student t-test or Mann-Whitney U-test as the second step. Normality of distribution was initially evaluated as well before correlation analysis with a Pearson coefficient for normal data and with a Spearmen coefficient for non-normal data. Multiple regression models were performed considering multiple testing of hypotheses. The results of the study were presented as histograms with pairwise comparison of correlation matrices. P-value <0.05 was considered significant.