Year-to-year variability of cholesterol levels in a pediatric practice
I. Benuck, S. S. Gidding and M. Donovan
Department of Pediatrics, Children's Memorial Hospital, Chicago, Ill.
OBJECTIVES: To study the impact of the variability of blood cholesterol
levels, which are known to vary spontaneously. The impact of this
variability on screening for hypercholesterolemia according to National
Cholesterol Education Program (NCEP) guidelines was reviewed in a private
pediatric practice. DESIGN: Retrospective chart review. SETTING: Private
pediatric practice. PATIENTS: Children (N = 646) aged 3 to 19 years.
INTERVENTION: Cholesterol measurements at a mean interval of 19 months
between visits. MAIN OUTCOME MEASURES: Year-to-year change in cholesterol
levels according to NCEP guidelines, regression to the mean, age, pubertal
status, body mass index, hematocrit, interval between measurements, and
season of the year were assessed for their contribution to cholesterol
level variability. RESULTS: Cholesterol level varied significantly with
both age (P < .001) and pubertal status (P < .01) with children aged
9 to 12 years; prepubertal children had the highest levels. Visit-to-visit
consistency of NCEP classification was poorer for younger children than for
older children (kappa = 0.21 for 3- to 6-year-olds, 0.39 for 6- to
10-year-olds, and 0.44 for those older than 10 years). Of children with
initial total cholesterol levels of 5.17 mmol/L (200 mg/dL) or greater,
only 40% continued to have high levels at follow-up. A child with an
initial cholesterol value of 5.17 mmol/L (200 mg/dL) showed an average
decline of 0.34 mmol/L (13 mg/dL) at follow-up by regression analysis.
CONCLUSION: Year-to-year variability in total cholesterol level has a
significant impact on the stability of NCEP classification.
