Using Ultrasound Elastography to Predict Development of Hepatic Sinusoidal Obstruction Syndrome

Hepatic sinusoidal obstructive syndrome (SOS), also known as veno-occlusive disease, is a complication of hematopoietic stem cell transplant (HCT) that is associated with high morbidity and mortality. 57,000 patients in the United States and Europe undergo HCT annually and SOS affects up to 15% of these patients. SOS pathogenesis is thought to be caused by damage to the hepatic venous endothelium due to the preparative regimen used before HCT. This damage results in obstruction of blood flow through the liver. Pathology shows collagen deposition in the sinusoids and fibrosis of venous lumens. The severity of the disease is not correlated to the number and severity of the histological changes. Mild and moderate SOS can resolve with supportive treatment. Severe SOS (30% of SOS) is commonly associated with multi-organ failure and has a mortality rate of 80% despite available prophylaxis and treatment.

SOS is most commonly defined by two clinical criteria: the modified Seattle criteria and the Baltimore criteria. The modified Seattle criteria state that at least two of the following criteria must be present within 20 days of HCT: bilirubin > 2mg/dL; hepatomegaly and/or ascites; and/or weight gain > 5% above baseline weight (6). Pediatric SOS incidence in HCT is 20% and is higher compared to adults. Death or multi-organ dysfunction affects 30-60% children who develop SOS. The most common definition of severe SOS is retrospective, namely death from SOS-related causes or persistent multi-organ dysfunction at 100 days post HCT. However, the European Society for Blood and Marrow Transplantation has proposed new prospective SOS diagnosis and grading schemes that could become standard of care since it can be performed prospectively and thus can guide treatment.

Defibrotide is a DNA derivative from porcine intestine that protects and repairs endothelial cells. Prior trials showed that defibrotide decreased the incidence of multi-organ failure and death from SOS. The main caveat is that treatment must be initiated very close to the time of clinical diagnosis using the Baltimore criteria to be effective (14). A study showed that 31/33 (94%) patients had complete remission of their SOS when treated with defibrotide <3 days after diagnosis, whereas only 3/12 (25%) patients had complete remission when treated >3 days of diagnosis. However, universal prophylaxis is difficult due to high drug costs ($155,000 for 21-day course). There is a critical need for an early and effective SOS diagnostic test that can identify patients who would benefit from defibrotide treatment.

Several adult and pediatric prospective studies have evaluated the efficacy of grayscale and Doppler ultrasound (US) in diagnosing SOS and have concluded that the clinical criteria are superior to US criteria for SOS diagnosis. The main reason for this conclusion is that conventional US is able to diagnose SOS only after the clinical diagnosis. This research has resulted in multiple recent guidelines recommending US only for confirming clinical diagnoses or following disease progression and not for primary diagnosis. Ultrasound shear wave elastography (SWE) has been shown to effectively diagnose passive hepatic congestion. Fontan physiology is the best studied example. SWE values markedly increased after the Fontan operation. This surgery connects the hepatic venous circulation to the pulmonary arteries exposing the liver to increased resistance from the pulmonary circulation thereby increasing hepatic venous congestion. Additionally, the effect sizes in the Fontan studies are large compared with the effect sizes in hepatic fibrosis studies. The common thread of hepatic venous congestion between Fontan physiology and SOS physiology led us to hypothesize that SWE could be useful in SOS diagnosis. Additionally, preliminary SWE studies in adults showed that it might be useful in the setting of SOS.

The investigators of this study recently conducted a single site prospective cohort study involving 25 patients undergoing myeloablative HCT patients from December 2015 through June 2017. The investigators found increased velocities in all patients who developed SOS. US SWE velocity values showed no difference between pre-conditioning median US SWE velocity in the SOS group (1.24 + 0.09 m/s) and non-SOS group (1.41 + 0.18 m/s) (p=0.06). By day +5, patients with SOS had US SWE velocities that significantly increased by 0.25 + 0.21 m/s from baseline compared to 0.02 + 0.18 m/s in patients without SOS from baseline (p=0.02). By day +14, patients with SOS had US SWE velocities that significantly increased by 0.91 + 1.14 m/s from baseline compared to 0.03 + 0.23 m/s in patients without SOS from baseline (0.01). These values are both clinically and statistically significant, demonstrating that patients with SOS have significantly increased liver stiffness as measured by US SWE compared to patients without SOS. Additionally, SWE changes happened on average 9 to 11 days before clinical diagnostic criteria became positive. The sensitivity and specificity of this test were 60-80% and 67-93% in our small cohort of 25 patients depending on the threshold used and the test timing.