The multiparametric magnetic resonance imaging evaluation of sinusoidal obstruction syndrome findings after oxaliplatin-based chemotherapy

Purpose: The aim of this study is to evaluate magnetic resonance elastography (MRE) derived liver stiffness measure (LSM) and changes in magnetic resonance imaging (MRI) T1-T2 relaxation times as well as quantitatively demonstrate the degree of parenchymal dysfunction using hepatobiliary phase and pre-contrast liver MRI in patients with sinusoidal obstruction syndrome (SOS) after oxaliplatin-based chemotherapy. Material and Method: Patients who underwent liver MRI and MRE between 1 January 2018 – 1 February 2020 and received oxaliplatin due to GIS cancer were examined in MRI Unit of Hacettepe University Faculty of Medicine, Department of Radiology. Patients with SOS were evaluated retrospectively, with the control group who received oxaliplatin without SOS findings. The presence of SOS in 18 patients was diagnosed by MRI, while8 patients had no findings. There were also 16 patients without any known disease with normal MRI and MRE findings. Forty-two patients (M/F:19/23) included in the study in total. The demographic characteristics, imaging findings and laboratory values were recorded for each patient. MRE-LSM, T1-T2 mapping images and liver MRI with hepatospecific contrast agents were evaluated. Parenchymal signal intensity changes were calculated using hepatobiliary phase and precontrast T1-weighted images. Patients were grouped by the lobe dominance and amount of reticular hypointensities in hepatobiliary phase within the SOS group. If parenchymal reticular hypointensities were less than 20%, they were categorized as "mild", between 20%-50% were "moderate", while above 50% were categorized as "advanced". Laboratory and imaging findings were compared according to the severity of the disease. Results: The average liver stiffness value in the patients with SOS was 3,14±0,44kPa, while it was 2,6±0,67kPa in the control group(P=0,036). The mean stiffness levels of the right lobe were also measured as 3,32±0,54kPa in the SOS group, and 2,7±0,7kPa (p=0,023) in the control group. No significant difference was noted between the two groups considering left lobe stiffness. The mean corrected T1 relaxation time was 1181.8±151,4 ms in the SOS group, as1024,4±156,5 ms in the control group (p=0,032). Mean T2 relaxation time was 50,29±3,6 ms in the SOS group and 44±3,9 ms in the control group (p=0,001). The right and left lobe signal intensity ratios were 2,66±0,64 and 2,84±0,51 in the SOS group, while it was 3,43±0,7and 3,4±0,7 in the control group (p=0,012, p=0,037). As it comes to the comparison of patients who received oxaliplatin treatment without any SOS findings and normal cases, the mean parenchymal stiffness values of the whole liver and each right-left lobes in the patient group who received oxaliplatin were found higher than the rest (p<0.05).However, there was no significant difference between these two groups in terms of T1-T2 relaxation time and parenchymal signal intensity ratio. Three patients classified as "mild", ten patients classified as "moderate" while five patients were classified as "advanced" regarding the degree of parenchymal heterogeneity of SOS in the hepatobiliary phase. According to the distribution of parenchymal heterogeneity, 9 patients demonstrated right lobe dominance, 2 patients demonstrated left lobe dominance as 7 patients showed no right or left lobe dominance. While the parenchymal stiffness was 2,8±0,22 kPa in patients with mild SOS findings, it was measured as 3±0,33 kPa and 3,65±0,28 kPa, in those with moderate and advanced SOS findings (p=0,002). Although T1 and T2 relaxation times increased as SOS severity increased, no statistical significance was noted. Conclusions: In patients with hepatic SOS, there is an increase in MRE-LSM and a good correlation between MRE-LSM and SOS severity. T1 and T2 relaxation times obtained by tissue mapping method were useful in the diagnosis of SOS but were found to be insufficient in determining SOS severity. Even if SOS findings do not develop in patients receiving oxaliplatin, liver stiffness increases. However, this increase is not as much as in patients with SOS. Parenchymal signal intensity ratios give quantitative measures of liver dysfunction in this patient group. MRE is especially effective in the diagnosis of SOS and determining SOS severity in patients who cannot be given contrast material and it may be useful in the follow-up of these patients. Keywords: Sinusoidal obstruction syndrome, MR elastography, T1-T2 mapping.