PWE-073 Spleen stiffness via acoustic radiation force impulse in HIV patients with non cirrhotic portal hypertension

    loading  Checking for direct PDF access through Ovid



Several HIV centres describe small cohorts of patients with non-cirrhotic portal hypertension (NCPH). Didanosine (ddI) exposure has been associated and the optimal monitoring and management of these patients remains unclear. Non-invasive assessment of the spleen is now being utilised to assess splenic stiffness in cirrhotic portal hypertension and predict outcomes. This technique has not previously been studied in patients with HIV and NCPH.


Acoustic Radiation Force Impulse (ARFI) elastography was performed using Philips EPIQ7TM to simultaneously assess liver and spleen stiffness in 3 patient groups. Group 1: HIV and NCPH, defined as the presence of portal hypertension manifestations in the absence of cirrhosis; Group 2: HIV and past ddI exposure (without known NCPH), Group 3: HIV and no history of liver disease. Groups were matched for age, HIV chronicity and antiretroviral treatment (including cumulative ddI exposure in Groups 1 and 2). Clinical and demographic information was collected. Differences in liver and spleen stiffness (in kPa) between groups were analysed using the Mann-Whiney U test


25 patients were recruited (Group 1: n=11, Group 2: n=5, Group 3: n=9). Patients were well matched for age, HIV chronicity and all had HIV RNA levels<20 copies/mL. Cumulative ddI exposure in Groups 1 and 2 was 56 and 53 months respectively (p=0.91). Median (IQR) ARFI liver and spleen stiffness in Group 1, 2 and 3 was 5.5 (4.8–9.8), 4.3 (4.0–5.3) and 4.8 (3.8–5.2) kPa (p=0.031) and 46.3 (29.5–143.2), 21.3 (14.6–26.8) and 18.3 (14.6–21.6) kPa (p=0.001) respectively. Liver and spleen stiffness were both significantly higher in NCPH vs ddI-exposed (p=0.019 and p=0.005) and ddI-unexposed controls (p=0.038 and p<0.001). Spleen stiffness was more effective than liver stiffness at predicting NCPH, AUROC 0.812 vs 0.948. Combining the two variables improved the diagnostic performance, AUROC 0.961. The optimal cut-off for predicting NCPH using splenic stiffness was 25.4 kPa, with sensitivity 91%, specificity 93%, PPV 91%, NPV 93%, positive likelihood ratio 12.73, negative likelihood ratio 0.10. Spleen and liver stiffness scores were strongly correlated (p=0.0004 95% CI 18, 59).


Elevated spleen stiffness is observed in HIV patients with NCPH and can be quantified easily using ARFI with high diagnostic accuracy. Novel strategies such as ARFI for longitudinal monitoring of patients with HIV and NCPH should be considered.

Related Topics

    loading  Loading Related Articles