Validation of a Non-Invasive Method Using Mechanical Extensometer for the Estimation of Intracranial Compliance Using Repeated Measures Agreement Analysis
Hayley Williams1, Sanem Uysal1, Nicholas Thompson2, Catherine Hassett3, Joao Gomes3
1Department of Neurology, Neurological Institute, Cleveland Clinic, 2Department of Quantitative Health Sciences, Lerner Research Institute, Cleveland Clinic, 3Neurointensive Care Unit, Cerebrovascular Center, Neurological Institute, Cleveland Clinic
Objective:
To validate a non-invasive intracranial (ICP) waveform monitoring device against the gold standard invasive ICP monitoring device for comparison of three methodologies calculating estimated intracranial compliance.
Background:
Cerebral compliance is the brain’s ability to accommodate volumes increases without an increase in pressure. Several methods utilize surrogate cerebral compliance estimations, but these require invasive ICP monitoring. For this study, intracranial compliance was calculated by transcranial doppler (TCD), as well as ICP waveform analysis (P1:P2 ratio, Time to Peak [TTP]).
Design/Methods:
Adult brain-injured patients were monitored with both invasive (extraventricular drain) and noninvasive (mechanical extensometer). TCD and ICP waveform analysis was performed by ICM+ and Brain4Care analytics. To examine agreement, a linear mixed-effects model accounted for repeated measurements to compute the concordance correlation coefficient (CCC), total deviation index (TDI), and coefficient of individual agreement (CIA). TCD values were log-transformed to mitigate non-normality. A coverage probability plot (CPP) was computed for the percent of observations within different cut points for each method. Computations were performed in R. All tests were two-sided and statistical significance was established at 0.05.
Results:
21 patients were monitored for median 36 minutes (IQR 27–42). Agreement statistics for TCD, P2/P1 ratio, and TTP were as follows: TTP: [CCC 0.33; CIA 0.25; TDI, 95% agreement 0.24 units], P2:P1: [CCC 0.42; CIA 0.05; TDI, 95% agreement 1.05], and log(TCD): [CCC: 0.50; CIA: 0.04; TDI, 95% agreement 2.94]. Although the CCC was highest for log(TCD) values, TTP and P2:1 ratio had better agreement with TDI and CIA analysis. This was further supported by the CPP, which showed TTP had the highest percent of cases within various cutoff difference values.
Conclusions:
Non-invasive and invasive evaluations of estimated intracranial compliance reveals discordance between the three methodologies. ICP waveform derived measurements may be a better estimate of compliance compared to TCD-derived methods; however, prospective validation studies are required.