Prematurity as Predictor of Children's Cardiovascular-renal Health (PREMATCH)

STATE OF THE ART The cardiovascular system (both macro- and microcirculation) undergoes extensive maturation throughout fetal, perinatal and pediatric life. Extreme preterm birth interferes with the normal development of the cardiovascular and microcirculatory systems. Disruption in vascular ontogenesis leads to abnormalities in the microvascular structure and circulation in various organs, such as retina (retinopathy of prematurity [O'Connor et al.]), kidney (abnormal glomerulogenesis [Sutherland et al.]) and glycocalyx in sublingual capillaries [Nieuwdorp et al.], among microvascular-driven disruptions observed in other organs (e.g. periventricular leukomalacia [Takashima et al.], bronchopulmonary dysplasia [Gien et al.]). Besides the well known retinopathy of prematurity, microvascular growth drives glomerulogenesis in the kidney and terminates after 34-36 weeks of gestational age under normal fetal conditions.

Perinatal (fetal or neonatal) growth restriction or preterm birth therefore impairs glomerulogenesis [Sutherland et al., Abitbol et al., Barker et al., Faa et al., Gubhaju et al., Zaffanello et al.]. Other cardiovascular abnormalities following from premature birth in later life include decreased heart rate variability [Rakow et al.], endothelial dysfunction [Norman et al.] and hypertension [Abitbol et al., Keijzer-Veen et al.].

Epidemiological observations further confirm that former preterm born infants are indeed at increased risk to develop cardiovascular disease and chronic kidney disease during adulthood [Brenner et al., Carmody et al., Vieux et al., Zandi-Nejad et al.]. The concept that fetal and perinatal conditions affect normal cardiovascular and renal ontogeny is in itself not new. Epidemiological studies showed that there is an association between low birth weight and vascular dysfunction in later life, suggesting that vascular impairment in early life is a harbinger of a poorer long-term prognosis [Sutherland et al., Bacchetta et al., Keijzer-Veen et al., Puddu et al.]. Intra-uterine growth retardation and children small for gestational age can be regarded as a failure of a fetus to reach the genetic potential of growth due to nutritional deprivation, the so-called Barker hypothesis [Barker et al.]. In adult life, this early growth retardation is associated with impaired renal and (micro)vascular function, hypertension, glucose intolerance and cardiovascular disease [Sutherland et al., Abitbol et al., Faa et al., Zaffanello et al., Carmody et al., Keijzer-Veen et al.].

The same sequence of vascular impairment serving as an indicator for long-term prognosis applies to the (early) postnatal life of preterm infants. According to the Brenner hypothesis, the decreased number of nephrons causes hyperfiltration, sodium loss with activation of the renin-angiotensinaldosterone system and hypertension [Brenner et al.], a process that entails further nephron loss, predisposition to develop proteinuria and possibly chronic kidney disease [Vieux et al., Puddu et al.].

MOVING BEYOND THE CURRENT STATE OF THE ART This project aims to move beyond the state-of-the-art by studying association between macro- and microvascular structure and function in children (8-13 years) born prematurely (extremely low birth weights, ELBW, i.e. birth weight below 1000 grams), and sex- and age-matched controls. The specific strengths hereby are that this ELBW cohort has been well characterized on its perinatal aspects [George et al.]. The characterization in the postnatal period includes biometry, perinatal characteristics (e.g. Apgar score, drugs, respiratory support), creatinine trends in the first 6 weeks of postnatal life and psychomotor development (Bayley Scales of Infant Development) at the age of nine months and two years.

The phenotypes of interest for the current project include the micro- and macrocirculation, cardiac structure and function, endothelial function and renal anatomy and function during childhood. We hereby aim to apply state of the art methods (cf methodology section) to assess micro- and macrocirculatory function, combined with state of the art statistics and advanced tools (metabolomics, epigenetics) to explore epidemiological findings and its pathogenesis.

The expectation is that the results of this project will identify and quantify risk factors in former preterm infants for cardiovascular-renal disease when compared to control children, while we can also map early risk factors for cardiovascular-renal disease in adult life and pave the way for a better informed prevention of these complications. Finally, data collected in this specific cohort can be compared to data collected in other cohorts.

HYPOTHESIS We hypothesize that former ELBW infants, compared to controls following term birth, will be associated with changes in the macro- and microcirculation of the cardiovascular-renal system already in young children over and beyond what is currently known. These changes - even if subtle - are probably forerunners of cardiovascular-renal complications in adulthood. In addition, confounders (e.g. neonatal nutrition, neonatal treatment) documented in the early life may identify new approaches for early prevention.

OBJECTIVES Using a case-control design in a 1/2 proportion, this study aims to detect functional and structural changes in children after preterm birth (ELBW) compared with children born after normal gestation with normal weight. The phenotypic aspects considered cover micro- and macrovascular structures and function.

1. Endothelial function
2. Sublingual capillary glycocalyx and density
3. Retinal imaging and visual acuity
4. Left ventricular function
5. Renal anatomy and function
6. Structure and function of the carotid artery (intima-media thickness, distensibility, Young's elastic modulus), aortic pulse wave velocity and the systolic augmentation index.

MODULATORS Next, this project will search for host characteristics, life style and environmental factors (see below) that may further modulate the differences in youngsters born either prematurely (case, ELBW) and at term, as well as for circulating and urinary biomarkers that are associated with the observed differences, and may provide insight into the pathogenesis involved.

IDENTIFICATION OF PREDICTORS Based on the existing database of the early perinatal follow-up of the prematurely born infants, this project will attempt to construct models predicting increased risk of potentially clinically significant changes associated with preterm birth and vice versa (mediators in neonatal life may predict cardiovascular-renal outcome in adult life).

INTENTION TO LINK THESE DATASETS WITH OTHER COHORTS WITH SIMILAR OBSERVATIONS Finally, pooling of these cardiovascular phenotypic data in children with other cohorts, either or not former ELBW cases may provide opportunities for additional prediction model development, validation and subsequent secondary preventive strategies.

METHODOLOGY Methodology-related issues include phenotyping, database construction/quality control, modulators, predictors and statistics.

PHENOTYPING

1. Endothelial function will be assessed by 24h urinary microalbumin excretion and digital pulse wave amplitude hyperemic response (photoplethysmography,PPG). To determine amplitude changes of the digital pulse, the response of the PPG pulse wave amplitude to hyperemia will be calculated from the hyperemic fingertip as ratio of post-deflation PPG pulse to baseline amplitude (PAht/PAh0, PA=pulse amplitude, h=hyperemic finger, t=time interval, 0=baseline). To obtain this ratio, we will divide the PAht/PAh0 ratio by the corresponding ratio at the control hand (PAct/PAc0, c=control finger)[Kuznetsova et al.].
2. To measure sublingual capillary glycocalyx and density, videos (10 sec) of sublingual capillaries in 2 areas lateral of the frenulum and 3-4 cm anterior to the tongue base will be recorded, using orthogonal polarization spectral (OPS) and side stream dark field (SDF) imaging [Hubble et al.]. In our hands, the intra- and inter-observer reproducibility of capillary density is 10.2 and 13.4% respectively. The erythrocyte-endothelium gap is the gold standard for glycocalyx measurement in vivo [Nieuwdorp et al.], as endothelial glycocalyx allows limited access to erythrocytes. The perfused boundary region (PBR) hereby reflects glycocalyx thickness and integrity, increased PBR glycocalyx loss.
3. Retinal imaging will be performed using a Canon Cr-DGi (Canon Co Ltd, Kyoto, Japan) nonmydriatic retinal visualization system. After accommodation to darkness, 1 image/eye will be obtained [Liu et al.]. Trained observers will identify individual arterioles and venules, using a validated computer-assisted program IVAN (Vasculo-matic Nicola, Ophthalmology and Visual Science, University of Wisconsin-Madison[Sherry et al.]). In addition, we will investigate visual acuity (clearness of vision, spatial resolution of the visual processing system) by the non-invasive adapted Snellen charts without visual aids.
4. All children will undergo detailed assessment of left ventricular (LV) function through echocardiography. PREMATCH will hereby focus on early changes in diastolic and systolic LV function. In combination with tissue Doppler imaging (TDI), transmitral and pulmonary vein blood flows will be used to detect LV filling changes.
5. Renal anatomy and function will be assessed by 2-dimensional measurement, renal arterial Doppler blood flow measurement and 3-dimensional calculations. Creatinine clearance, 24h microalbuminuria and Cystatin C on peripheral blood (serum) will be quantified.
6. The structure and function of the carotid artery, aortic pulse wave velocity (PWV) and systolic augmentation index will be assessed as measures of macrovascular function and structure. We will use ultrasound to measure the local properties of the common carotid artery. diameter, distension, and intima-media thickness will be measured and averaged over 3 cardiac cycles in recordings consisting of >4 consecutive beats. Distensibility (103/kPa), compliance (mm²/kPa) and Young's elastic modulus will be calculated. Local application tonometry will be performed (SpyghmoCor system). Measurements will be performed at carotid, radial and femoral arteries. The local arterial pulse wave will be recorded as well as carotid-to-femoral and carotid-to-radial PWV (cf- and cr-PWV). The carotid and radial augmentation indexes will be measured directly at the carotid and radial arteries and the aortic augmentation index will be calculated from the radial signal by the validated generalized transfer function [Richart et al., Seidlerova et al., Mischak et al.].Furthermore, we will implement ambulatory assessment of central hemodynamics, using the Mobil-O-Graph 24h PWA Monitor (IEM GmbH, Stolberg, Germany), a validated monitor for 24h blood pressure monitoring, including the ARCSolver application, which allows pulse wave analysis of the central blood pressure and measuring of aortic PWV [Luzardo et al.].

DATABASE CONSTRUCTION/QUALITY CONTROL Trained nurses will code questionnaires, technicians will enter the data. For quality assurance, 10% of questionnaires will be randomly selected and recoded by another nurse. All data will be inputted twice by different technicians. Duplicate datasets will be compared with the PROC COMPARE application (SAS software) to trace input errors. Data coders and SAS programs will check for internal consistency of questionnaire replies. Non-Gaussian distributions will be normalized by proper transformation. As part of the quality control, descriptive statistics will be generated at 6 month intervals.

MODULATORS

• Anthropometric characteristics, by sex, age, height, weight, body mass index, span width, waist-tohip ratio, skinfolds (Harpenden Skinfold Caliper, Bedfordshire, UK).
• Matrix reasoning and spatial orientation tests (Wechsler), indicators of mental capability.
• Muscle strength, by grip strength.
• Sexual maturation, by Tanner scale.
• Office and self-measured home blood pressure, by Mobil-O-Graph 24h PWA monitor (IEM, Stolberg, Germany)
• Body composition, by Bodystat4000 (Bodystat Ltd, Douglas, UK)
• Questionnaire (education, medication use, habits, menarche (girls) and familial and personal history).
• Measurements on blood and 24h urine samples. Plasma, serum and urine samples will be divided into aliquots and stored (-20/-80°C) at the bio-bank of the Studies Coordinating Centre (SCC). Routine measurements include hemoglobin, hematocrit, red and white blood cell counts, mean corpuscular volume, mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, differential white blood cells count, serum creatinine, uric acid, serum lipids (total, HDL-cholesterol), glycaemia and insulin. Measurements of metabolic, inflammatory and oxidative stress include: 8-hydroxy -2'-deoxyguanosine (8-OHdG), interleukin-6 and high sensitivity C-reactive protein[Pearson et al.]. Other measurements will be considered if enough samples are available (e.g. homeostasis model assessment index (HOMA)[Marcelis et al.], leptin, adiponectin [Marcelis et al., Yeon et al.], E-selectin, P-selectin, vascular adhesion molecule-1, plasma fibrinogen, tumor necrosis factor, superoxide dismutase or serum malondialdehyde[Rao et al., Sharma et al.]).

Measurements of 24h urine samples include volume, electrolytes, creatinine, micro-albumin, and aldosterone. Urinary proteomics will be done in collaboration with Prof Harald Mischak, SME Mosaiques (mosaiques-diagnostics.de) by capillary electrophoresis coupled to mass spectrometry according to standard operating procedures in an environment with proper quality control [Mischak et al.].

- GPS coordinates of residence. Meteorological data and data on airborne pollutants and fine particulate collected from the appropriate sources.

PREDICTORS In our published cohort [George et al.] on creatinemia in ELBW infants in the first 6 weeks of life, raised creatinemia reflected immaturity (e.g. gestational age, weight) and morbidity (Apgar, ventilation, retinopathy of prematurity, intraventricular hemmorrhage), but also treatment modalities (e.g. ibuprofen, steroids, parenteral nutrition). We will link the perinatal covariates and creatinine trends to the dataset of this study to explore to what extent perinatal data predict cardiovascular and renal outcome. Since also treatment modalities are included, this study will provide first data on long term cardiovascular and renal outcome following drug exposure.