Liver stiffness measurement selects patients with cirrhosis at risk of bearing large oesophageal varices

Background/Aims

Periodic endoscopic screening for oesophageal varices is recommended in patients with cirrhosis, but might be limited to a subgroup of patients if a simple non-invasive test was available to select those at risk of bleeding.

Methods

We studied in 165 patients with cirrhosis the relation between the presence of oesophageal varices assessed by endoscopy, and liver stiffness measurement by Fibroscan^®, a non-invasive parameter related to liver fibrosis. The results were compared to those of other parameters reflecting portal hypertension, splenic size, platelet count, and platelet count/spleen size ratio.

Results

Liver stiffness measurement was correlated to the grade of oesophageal varices (r = 0.6, p < 0.0001). AUROC values of liver stiffness measurement were 0.84 (95% CI: 0.78–0.90) for the presence of oesophageal varices and 0.83 (0.76–0.89) for varices grade ⩾ II. Liver stiffness measurement value <19 kPa was highly predictive of the absence of oesophageal varices grade ⩾ II (Se: 84%, PPV: 47%, NPV: 93%).

Conclusions

Liver stiffness measurement allows to predict the presence of large oesophageal varices in patients with cirrhosis, and may help to select patients for endoscopic screening.

Introduction

Mortality in patients with cirrhosis has continuously decreased during the past decades thanks to active prevention of complications mainly gastro-intestinal haemorrhage and bacterial infections [1]. These measures, particularly prevention of digestive haemorrhage from oesophageal varices (OV) which rely on either pharmacological treatments [2], [3], [4] or band ligations [5], are widely considered as effective and may reduce mortality.

The risk of variceal haemorrhage is clearly related to the size of OV [6], [7]. Therefore primary prevention of variceal bleeding applies to patients with previously diagnosed large OV (grade II or III) detected by periodical upper tract endoscopy. Several consensus conferences have stated that this periodical surveillance was justified and that the optimal periodicity for variceal screening was between one and three years [8], [9], [10], [11], [12], [13].

The number of patients eligible for variceal screening depends on the prevalence of cirrhosis in the general population and on putative criteria of selection. If all patients with cirrhosis were concerned, as cirrhosis is more and more screened for, the numbers would be considerable. A generalized program of periodical upper tract endoscopy in these patients might result in heavy economical burden even for developed countries [14]. Furthermore repeated examinations, when not performed under general anaesthesia or profound sedation, are often poorly accepted by patients who may refuse further followup [15]. For these reasons selection of patients with a high probability of bearing OV and especially large OV at risk of rupture has been proposed using various non-invasive criteria [11], [16], [17], [18], [19], [20]. These putative criteria of selection are either simple but poorly discriminant, or multiple but complex and hardly usable in clinical practice.

There is still a need for a simple reliable parameter that would allow limiting the indications of upper tract endoscopy to a subgroup of patients especially at risk. We propose here a new method for selecting patients for variceal screening. Liver stiffness measurement (LSM) relies on the calculation of liver elasticity (or stiffness) from the velocity of a low frequency elastic wave inside the liver [21]. The method is entirely non-invasive and well accepted by patients. It is correlated to the extent of fibrosis and is particularly reliable for the diagnosis of cirrhosis [22].

Our purpose in this study was to evaluate the value of LSM for selecting patients with cirrhosis at risk of bearing large OV and therefore eligible for upper tract endoscopy screening and to compare this new parameter to other previously proposed criteria of selection.