The Problem of Macrocephaly in Infancy

This submission has open access
Submission ID :
ESPR243
Submission Type
Submission Topic
Abstract: :

Introduction

Macrocephaly in infants is defined as a head circumference (HC) greater than 2 standard deviations above the mean, or above the 97.7th percentile. Since the differential diagnosis of macrocephaly includes neurosurgical conditions, macrocephaly is commonly encountered by primary care physicians with frequent referral to pediatric neurosurgeons, as there is no guidance including guidelines related to head imaging in these 2.3% of all infants.


Objectives

Our study aims to identify clinical associations with actionable imaging findings among infants and children under 2 years of age who have macrocephaly. Based on these clinical associations, our secondary aim is to develop a practical scoring system with high sensitivity and acceptable specificity to improve the yield of imaging for macrocephaly in primary care.


Methods

A case-control study based on electronic records in the Nemours Children's Health System was undertaken by querying a radiology registry for head imaging studies with 'macrocephaly,' 'macrocephalia,' or 'head growth' in the requisition among children 2 years of age or younger completed during the period October 2011 to February 2020. Four neurosurgeons reviewed the images with abnormal findings in the reports, determining cases to be a 'concern' if neurosurgical referral, follow-up or intervention were indicated. Control cases were selected randomly from the remaining imaging studies. Chart review was completed to collect patient-level data including measurements of HC, weight, length as well as details related to pregnancy, delivery, medical history, and presenting signs and symptoms prior to imaging. Controls were matched 3:1 to cases of 'concern' with covariates significant in the univariate model included in the multivariate model using manual, step-wise backward, conditional logistic regression, with the goal of developing an additive score predicting neurosurgical concern. 


Results

In the study sample of children with macrocephaly who had head imaging before 24 months of age (N = 1381), 46 (3.3%) were concern cases with 138 matched controls. Step-wise backward conditional logistic regression yielded a model with the following covariates: bulging fontanel, prematurity, any delay, and accelerated head growth. To develop an additive predictive score, integer points were assigned to each covariate proportionate to the model parameters: bulging fontanel (3 points), prematurity (3 points), any delay (1 point), and accelerated head growth (1 point). A test threshold of 1 point yields a sensitivity of 0.96 (0.86 – 0.99) and a specificity of 0.21 (0.15 – 0.29). 


Conclusions

Of head imaging studies done, there were few patients with findings of concern (46/1381, 3.3%) and even fewer (15/1381, 1.1%) requiring surgery. Overall, the decision rules developed related to findings requiring follow-up can eliminate 10 to 20% of negative scans with sensitivity of ~95%.

Nemours/Alfred I DuPont Hospital for Children
Sidney Kimmel Medical College, Thomas Jefferson University
Nemours/AIDHC
Nemours, CHLA
Nemours/AIDHC
Nemours Children's Hospital
Nemours Children's Hospital
Nemours/AIDHC

Similar Abstracts by Type

Submission ID
Submission Title
Submission Topic
Submission Type
Corresponding Author
ESPR157
Clinical Research
Original science
Aditya Chhikara
ESPR302
Epidemiology
Original science
Natasha Jouk
ESPR74
Clinical Research
Original science
Alexandra Mazo
37 hits