1447Childhood risk factors and cardiovascular disease outcomes in adulthood. Preliminary findings from the International Childhood Cardiovascular Cohort (i3C) Consortium

Background

Atherosclerosis develops decades before clinical cardiovascular disease (cCVD) occurs. Longitudinally, childhood risk factors predict adult pre-clinical atherosclerosis. There is currently no evidence directly linking childhood risk factors to cCVD.

Purpose

To provide the first direct evidence of any association between known risk factors for CVD when measured in childhood and adult CVD incidence and death.

Methods

Using i3C Consortium data, we linked childhood risk factors to adult cCVD. cCVD events were ascertained by participant re-contact in the US and Australia, medically adjudicated hospital records; and using the Finnish national health registry. Of 16,964 adult participants (mean age 49 years) examined during ages 3–19, 201 people with any cCVD event (70% coronary artery, 25% cerebrovascular, and 5% peripheral artery disease) have been determined. The analysis included Cox proportional hazard models. Each model was adjusted for childhood age, age at followup, sex and cohort/race. Continuous childhood variables were z-scored for each participant's last repeated measure during childhood.

Results

Childhood body mass index (BMI), serum total cholesterol (TC) and triglycerides, and systolic blood pressure were positively associated with adult cCVD events (P<0.0001). Smoking in childhood was associated with nearly 50% increased risk of adult cCVD (P=0.08). BMI; TC remained significant in the simultaneous risk factor model. The adjudication pipeline suggests that over 500 hospitalized cCVD events will be found on completion. Regression using the full set of imputed events yielded similar findings. Analysis of deaths is in process.

Childhood risk factor link to adult CVD Childhood risk variable Single risk factors models Simultaneous risk factor model n cCVD events/N at risk Hazard ratio (95% Confidence limits) p n cCVD events/N at risk Hazard ratio (95% Confidence limits) p Body Mass Index 201/16964 1.52 (1.33–1.73) <0.0001 142/11124 1.37 (1.14–1.64) 0.0008 Total cholesterol 191/13778 1.32 (1.14–1.52) 0.0001 " 1.21 (1.02–1.43) 0.03 Triglycerides 191/13654 1.17 (1.04–1.33) 0.01 " 1.04 (0.88–1.24) 0.6 Systolic blood pressure 190/14883 1.28 (1.11–1.48) 0.0007 " 1.18 (0.99–1.42) 0.07 Regular smoking ≥1/day 151/13436 1.44 (0.96–2.16) 0.08 " 1.43 (0.94–2.17) 0.10 Hazard ratios = increased risk per one standard deviation increase in continuous risk variables. E.g. every ∼0.9 mmol/L or ∼33 mg/dL increase in childhood total cholesterol is associated with a ∼32% and 21% increase in adult CVD risk in single and simultaneous risk factor models respectively. “Simultaneous risk factor model” recognizes that the risk factors are causally connected.

Conclusion

Childhood CVD risk factors predicts adult cCVD with implications for primordial CVD prevention.

Acknowledgement/Funding

National Heart, Lung, and Blood Institute (NHLBI)

Growth hormone deficiency and cardiovascular risk factors in childhood cancer survivors

6614

Background: Childhood cancer survivors (CCS) have a high frequency of growth hormone deficiency (GHD) and risk of early cardiovascular disease (CVD). This study examined the relations between GHD and risk factors for CVD in CCS. Methods: Anthropometrics, blood pressure, lipids, growth hormone (GH) stimulation test, dual-energy x-ray absorptiometry, abdominal CT, and insulin resistance (IR) (euglycemic, hyperinsulinemic clamp - low M/lbm signifies IR) were obtained in 174 CCS, mean age 15±2 years and 89 healthy sibling controls, mean age 13.5±3 years. Linear regression evaluated the relations between GHD and CVD risk factors, adjusted for sex, age, pubertal stage, and body mass index (BMI) or visceral fat. Results: 62 CCS (36%) had GHD. There were no significant measurement differences between non-GHD CCS and controls. Compared to controls, GHD CCS who never received GH (N = 34) had greater BMI (24.8 vs 20.8 kg/m2, p < 0.0001), percent body fat (36.1% vs 25.8%, p < 0.0001), visceral fat (34.8 vs 19.6 cm2, p < 0.0001), and triglycerides (TG) (120.2 vs 83.8 mg/dL, p = 0.001) and were more IR (M/lbm 11.1 vs 14.2 mg/kg/min, p = 0.0006). Adjustment for BMI and visceral fat did not change the IR or TG results. GHD CCS currently on GH had lower BMI (21.9 kg/m2, p = 0.02), percent body fat (31.2%, p = 0.08), and visceral fat (26.5 cm2, p = 0.03) compared to those not treated. IR and TG were not different between treated and not treated GHD CCS. Conclusions: GHD is a common finding in CCS and is significantly associated with adiposity, IR, and elevated TG.There is a suggestion that GH treatment had a positive impact on adiposity, but not IR and TG levels. These study findings imply that CVD risk factors are present in CCS with GHD independent of body fatness, suggesting that the cancer diagnosis or treatments received may lead to early cardiovascular disease in childhood cancer survivors.

No significant financial relationships to disclose.

Blood pressure and arterial compliance in young adults: the Minnesota Children's Blood Pressure Study

The aim of this study was to assess the relation between blood pressure (BP) and arterial compliance in a healthy sample of young adults. School children (aged 10 to 14 years at entry) were surveyed in 1977 to 1978, and 1,207 were followed once to twice yearly until age 23 years. Arterial compliance was measured in 179 adults at the last follow-up visit. The sample included individuals in the upper tertile of systolic BP during the last three follow-up visits and race- and sex-matched individuals in the lower two tertiles. We obtained radial artery waveforms using a calibrated tonometer device and characterized waveform morphology to determine large artery (C1) and oscillatory (C2) compliance. Blood pressure was measured using random zero sphygmomanometers. The mean and standard deviation of C1 was 2.13 +/- 0.59 mL/mm Hg and of C2 was 0.083 +/- 0.02 mL/mm Hg. Systolic BP was inversely related to C1 (P < .001) and C2 (P < .01) after adjustment for gender, height, weight, insulin, and HDL and LDL cholesterol. After adjustment, a 1 SD change in systolic BP was associated with a -0.30 mL/mm Hg change in C1 and a -.008 mL/mm Hg change in C2. Data from the Minnesota Children's Blood Pressure Study indicate that systolic BP is inversely related to arterial compliance, particularly C1 (the large artery, or capacitive compliance).

Unbalanced 4;6 translocation and progressive renal disease

Two sibs are described with an unbalanced 4;6 translocation resulting in partial trisomy 6p and monosomy for distal 4p. Growth retardation, psychomotor retardation, and characteristic facial appearance are present. The facial anomalies include high prominent forehead, blepharoptosis, blepharophimosis, high nasal bridge, bulbous nose, long philtrum, small mouth with thin lips, and low-set ears. Both children have small kidneys and have had proteinuria since early childhood. The older boy developed progressive renal disease including hypertension and renal failure necessitating renal transplantation at age 18 years. Renal biopsy of the younger girl also indicates significant renal involvement. Progressive renal disease is likely an important part of the trisomy 6p phenotype.

Cystic renal lymphangiectasia: a distinctive clinicopathologic entity in the pediatric age group

This report describes the features of unilateral cystic renal lymphangiectasia in a 2-year-old child who presented with hypertension, massive ascites, a left flank mass, and no evidence of familial renal cystic disease. The child became normotensive and is now asymptomatic more than 3 years after surgery. The clinical presentation and diffuse pathologic involvement are similar to findings for the few pediatric patients with cystic lymphangiectasia described in the literature and appear distinct from the more localized form of the disease seen in adults.

Antineutrophil cytoplasmic autoantibody-positive crescentic glomerulonephritis as a complication of treatment with propylthiouracil in children

Propylthiouracil, which is commonly used in the treatment of hyperthyroidism, has been associated in adults with antineutrophil cytoplasmic autoantibody, a serologic marker of vasculitis. Severe renal disease has not been reported as a complication of therapy with this drug. We report severe antineutrophil cytoplasmic autoantibody-positive vasculitis in children receiving propylthiouracil, as well as rapidly progressive crescentic glomerulonephritis after administration of this drug.

Cholesterol screening by primary care pediatricians: a study of attitudes and practices in the Minneapolis-St Paul metropolitan area

A telephone survey of the 197 board-certified pediatricians actively engaged in primary care in the Minneapolis-St Paul metropolitan area was conducted to assess their cholesterol screening practices and hypercholesterolemia management. The response rate was 95%. Nearly all the pediatricians (90%) do some cholesterol screening, with the majority (58%) screening only children with a strong family history of coronary heart disease. Though only 33% screen all their patients, 66% advocate universal pediatric screening. Most of the pediatricians indicated they would manage hypercholesterolemia patients themselves, nearly always with dietary means. Despite their strong support for screening, the pediatricians expressed skepticism about the significance of childhood cholesterol level as a predictor of adult cardiovascular disease and doubted their effectiveness in getting patients to adopt a cholesterol-reducing diet. Their definition of elevated total cholesterol level in childhood was consistent with published recommendations, but only 29% could define elevated low-density lipoprotein cholesterol level. The pediatricians expressed strong opposition to pediatric cholesterol screening in schools or in any setting other than clinics and hospitals.

Urinary prostaglandin E and kallikrein activity in glomerular hyperfiltration induced by a meat meal in man

Because renal vasodilator systems may be involved in the physiologic response to a protein meal, we studied the relationship of prostaglandin E (PGE) and kallikrein (KE) activity to the renal hemodynamic changes induced by a meat meal. Ten normal subjects on a maintenance diet providing 1 g protein per kg and 100 mmol Na daily were studied before and after a meal of 86 g of meat protein, once without medication, and again after treatment with indomethacin (150 mg daily for 3 days before and 50 mg the morning of the test). A carbohydrate meal of similar caloric, Na and K content was used in control studies. Glomerular filtration rate (GFR, inulin clearance), renal blood flow (RBF, para-aminohippurate [PAH] clearance), PGE and kallikrein urinary excretion were determined. All studies were performed during water diuresis to avoid bladder catheterization. The protein load induced an increase in GFR (ml/min, mean +/- s.e.m.: baseline 107.2 +/- 6.05, peak postmeal 146.4 +/- 6.79, p less than 0.01) and RBF (baseline 529.7 +/- 42.9, postmeal 678.9 +/- 61.9, p less than 0.05). Renal hemodynamic changes were unrelated to changes in urinary PGE and KE excretion. Indomethacin treatment inhibited PGE excretion by 73% during the test meal but did not modify the protein-induced hyperfiltration. Our results suggest that these renal vasodilator systems are not primarily responsible for the hyperfiltration response. In addition, the data show that inhibition of prostaglandin synthesis is not a practical approach to prevent glomerular hyperfiltration in clinical practice.

Percutaneous transluminal angioplasty for treatment of renal artery stenosis in children

Angioplasty was carried out in four children with hypertension and renal artery stenosis, including one patient with neurofibromatosis and one patient with multiple arterial stenoses. Two patients did not respond. In the patient with neurofibromatosis the stenosis persisted unchanged despite multiple balloon inflations. In one patient hypertension persisted despite successful dilatation, and was probably related to longstanding contralateral kidney disease. Two patients had excellent blood pressure response to dilatation. One of these patients required redilatation of bilateral recurrent stenoses.

Idiopathic hypercalciuria associated with hyperreninemia and high urinary prostaglandin E

A patient with idiopathic hypercalciuria and some features suggestive of Bartter syndrome is reported. Excessive urinary prostaglandin E (PGE) excretion and renal calcium leak were documented in this child. Treatment with aspirin and indomethacin reduced urinary PGE excretion and was associated with a decrease in daily calcium excretion. At the lowest levels of urinary PGE, the renal calcium leak was no longer evident although mild hypercalciuria persisted. These results suggest that PGE may play a role in some cases of idiopathic hypercalciuria.

Ontogenesis of the renin-angiotensin system in spontaneously hypertensive and normal Wistar rats

Kidney renin activity was investigated in spontaneously hypertensive and normotensive control Wistar rats. Kidneys were obtained from fetal rats after 18 and 21 days of gestation and from neonatal rats 1, 7, 14, and 21 days old. Kidney renin activity was determined by a modification of the radioimmunoassay for angiotensin I developed in this laboratory, and tissue from each age group was stained for juxtaglomerular granules. Kidney renin activity progressively increased in both strains from day 18 of fetal life until day 1 of neonatal life when a plateau level developed. In 1-day-old neonatal rats, significant differences in the kidney renin activity of the spontaneously hypertensive and the normotensive strains were initially noted. Kidney renin activity was maintained at significantly higher levels in the spontaneously hypertensive strain under 21 days of age when, because of a rise in the kidney renin activity of the normotensive control strain and a fall in the kidney renin activity of the spontaneously hypertensive strain, the difference between the two strains became statistically insignificant. In both strains, juxtaglomerular granules were first noted in 1-day-old neonatal rats and in each age group thereafter.