Quantitative assessment of intracranial pressure by the tympanic membrane displacement audiometric technique in children with shunted hydrocephalus

Abstract

The objective of this prospective study was to compare the clinical features at presentation, tympanic membrane displacement test results and direct intracranial pressure measurements in children with shunted hydrocephalus to procure a quantitative measure of the intracranial pressure by tympanic membrane displacement test. A prospective comparative evaluation of 61 clinical episodes of shunt malfunction was assessed by volume displacement of the tympanic membrane and direct intracranial pressure measurements in 40 patients with shunted hydrocephalus between January 1995 and June 1996. The volume displacement of the tympanic membrane (Vm) on stapedial contraction was inward for raised intracranial pressure in 27 episodes and ranged from -120 nl to -506 nl (mean = -250 nl). This was confirmed by direct intracranial pressure monitoring, which ranged from 23 to 40 mm Hg (mean = 29 mm Hg). The tympanic membrane displacement test measurement in 30 episodes of low intracranial pressure ranged from +263 nl to +810 nl (mean = +530 nl), and this was corroborated by direct intracranial pressure measurement ranging from 1 to 6 mm Hg (mean = 3.8 mm Hg). The normal baseline Vm values obtained when the subjects were asymptomatic ranged from +58 nl to +175 nl (mean = +115 nl). The tympanic membrane displacement test as a non-invasive diagnostic tool in predicting changes in intracranial pressure had a sensitivity of 93% and specificity of 100%. The predictive value of the test was 100%, and the negative predictive value was 73%. The kappa statistical analysis was used to measure the agreements between the groups. The strength of the agreement was very good, kappa = 0.88 and the P value was < 0.001. The objective measure of intracranial pressure by tympanic membrane displacement test with the Vm value of -200 nl and more negative was indicative of raised intracranial pressure and a Vm value of +200 nl and greater, for low intracranial pressure. The intracranial pressure measurements made on an individual subject basis were reliable and accurate. The test can therefore be used for regular assessment of shunted hydrocephalics to enable correlation of intracranial pressure with symptoms in individual patients.