Surfactant proteins (SP) are multi-systemic proteins playing crucial roles in the regulation of rheological properties of physiological fluids, host defense, and the clearance of potentially harmful metabolites. Hydrocephalus patients suffer from disturbed central nervous system (CNS) fluid homeostasis and exhibit remarkably altered SP concentrations within the cerebrospinal fluid (CSF). A connection between CSF-SPs, CSF flow, and ventricular dilatation, a morphological hallmark of hydrocephalus, has been reported previously. However, currently there are no studies investigating the link between rheologically active SPs and periventricular white matter changes caused by impaired CSF microcirculation in hydrocephalic conditions. Thus, the aim of this study was to assess their possible relationships. The present study included 47 individuals (27 healthy subjects and 20 hydrocephalus patients). CSF specimens were analyzed for concentrations of SP-A, SP-C, and SP-D by using enzyme-linked immunosorbent assays (ELISAs). Axial T2w turbo inversion recovery magnitude (TIRM) magnetic resonance imaging was employed in all cases. Using a custom-made MATLAB-based tool for quantification of magnetic resonance signal intensities in the brain, parameters related to disturbed deep white matter CSF microcirculation were estimated (TIRM signal intensity (SI)-mean, minimum, maximum, median, mode, standard deviation, and percentiles, p10th, p25th, p75th, p90th, as well as kurtosis, skewness, and entropy of the SI distribution). Subsequently, statistical analysis was performed (IBM SPSS 24™) to identify differences between hydrocephalic patients and healthy individuals and to further investigate the connections between CSF-SP changes and deep white matter signal intensities. SP-A (0.38 ± 0.23 vs. 0.76 ± 0.49 ng/ml) and SP-C (0.54 ± 0.28 vs. 1.27 ± 1.09 ng/ml) differed between healthy controls and hydrocephalus patients in a statistically significant manner. Also, corresponding quantification of white matter signal intensities revealed statistically significant differences between hydrocephalus patients and healthy individuals: SImean (370.41 ± 188.15 vs. 222.27 ± 99.86, p = 0.001), SImax (1115.30 ± 700.12 vs. 617.00 ± 459.34, p = 0.005), SImedian (321.40 ± 153.17 vs. 209.52 ± 84.86, p = 0.001), SImode (276.55 ± 125.63 vs. 197.26 ± 78.51, p = 0.011), SIstd (157.09 ± 110.07 vs. 81.71 ± 64.94, p = 0.005), SIp10 (229.10 ± 104.22 vs. 140.00 ± 63.12, p = 0.001), SIp25 (266.95 ± 122.62 vs. 175.63 ± 71.42, p = 0.002), SIp75 (428.80 ± 226.88 vs. 252.19 ± 110.91, p = 0.001), SIp90 (596.47 ± 345.61 vs. 322.06 ± 176.00, p = 0.001), skewness (1.19 ± 0.68 vs. 0.43 ± 1.19, p = 0.014), and entropy (5.36 ± 0.37 vs. 4.92 ± 0.51, p = 0.002). There were no differences regarding SP-D levels in hydrocephalus patients vs. healthy controls. In the acute hydrocephalic subgroup, correlations were as follows: SP-A showed a statistically significant correlation with SImax (r = 0.670, p = 0.024), SIstd (r = 0.697, p = 0.017), SIp90 (r = 0.621, p = 0.041), and inverse correlation with entropy (r = - 0.700, p = 0.016). SP-C correlated inversely with entropy (r = - 0.686, p = 0.020). For the chronic hydrocephalus subgroup, the following correlations were identified: SP-A correlated with kurtosis of the TIRM histogram (r = - 0.746, p = 0.021). SP-C correlated with SImean (r = - 0.688, p = 0.041), SImax (r = - 0.741, p = 0.022), SImedian (r = - 0.716, p = 0.030), SImode (r = - 0.765, p = 0.016), SIstd (r = - 0.671, p = 0.048), SIp25 (r = - 0.740, p = 0.023), SIp75 (r = - 0.672, p = 0.048), and SIp90 (r = - 0.667, p = 0.050). SP-D apparently does not play a major role in CSF fluid physiology. SP-A and SP-C are involved in different aspects of CNS fluid physiology. SP-A appears to play an essential compensatory role in acute hydrocephalus and seems less involved in chronic hydrocephalus. In contrary, SP-C profile and white matter changes are remarkably connected in CSF of chronic hydrocephalus patients. Considering the association between CSF flow phenomena, white matter changes, and SP-C profiles, the latter may especially contribute to the regulation of paravascular glymphatic physiology.
Keywords: CSF; Glymphatic; Histogram analysis; MRI; Surfactant proteins; TIRM.