Derivation of a pediatric growth curve for inferior vena caval diameter in healthy pediatric patients: brief report of initial curve development

Abstract

Abstract Background A validated tool has long been sought to provide clinicians with a uniform and accurate method to assess hydration status in the pediatric emergency medicine population. Outpatient clinicians use CDC height- and weight-based curves for the assessment of physical development. In hospital, daily weights provide objective data; however, these are usually not available at presentation. One of the most promising techniques for the rapid assessment of volume is ultrasound (US) to obtain an indexed inferior vena cava diameter (IVCDi); as previously described. Prior studies have focused on IVCDi in dehydrated patients and have shown that it provides accurate estimates of right atrial pressure and volume status. The objective of this study is to derive an IVC growth curve in healthy pediatric patients. Methods Prospective cohort design enrolled healthy children between the ages of 4 weeks and 20 years. Patients presenting with fever, illnesses, or diagnoses known to affect the volume will be excluded. All eligible patients under 21, who have provided self or parental written consent, will undergo a brief ultrasound to obtain transverse and long images of both the IVC and the aorta; all scans will be digitally saved. Image quality will be subjectively rated as poor, fair, or good based on wall clarity. Poor quality images will be recorded but may be omitted from our analysis. Five clinicians completed a 1-h introduction to IVC-US and ten supervised scans prior to enrollment. Still images will be measured in order to determine IVCDi in both transverse and longitudinal planes. To assess inter-rater reliability, in 10% of cases, two clinicians will complete scans. All study scans will be over-read by a fellowship-trained sonologist. IVCDi will be plotted independently as functions of age, gender, BMI, and aortic diameter. Within each group, means with means or medians with 95% CIs will be calculated. Following uni- and bivariate analyses and assessment for colinearity, a variety of parametric and nonparametric regression procedures will be conducted. The smoothed curves will be approximated using a modified LMS estimation procedure. Results Data for the initial curve derivation includes 25 patients ranging from 13 months to 20 years (mean 102 months or 8.5 years). Sixty-five percent of patients were enrolled from the ED, while 35% were enrolled from well-child clinic visits. When evaluating the size of IVC as a function of time linear growth, increasing size was found to proportionately increase with age of patient in months. Conclusions Data suggest a linear correlation between IVC size and age. Such data, when plotted as a new growth curve, may allow clinicians to plot a patient's sonographic measurements in order to assess hydration health.