Kara Rye

Towards noninvasive detection of oesophageal varices.

Current guidelines recommend that all cirrhotic patients should undergo screening endoscopy at diagnosis to identify patients with varices at high risk of bleeding who will benefit from primary prophylaxis. This approach places a heavy burden upon endoscopy units and the repeated testing over time may have a detrimental effect on patient compliance. Noninvasive identification of patients at highest risk for oesophageal varices would limit investigation to those most likely to benefit. Upper GI endoscopy is deemed to be the gold standard against which all other tests are compared, but is not without its limitations. Multiple studies have been performed assessing clinical signs and variables relating to liver function, variables relating to liver fibrosis, and also to portal hypertension and hypersplenism. Whilst some tests are clearly preferable to patients, none appear to be as accurate as upper GI endoscopy in the diagnosis of oesophageal varices. The search for noninvasive tests continues.

Screening for Liver Disease - are LFTs Old HAT?

The economic burden of end stage liver disease is set to increase due to the rising prevalence of cirrhosis secondary to alcohol, viral hepatitis and fatty liver disease. Screening for liver disease has been advocated, as most cases of cirrhosis are preventable with early interventions. Liver function tests (LFTs) are routinely used as a first line investigation to screen for liver diseases but can be normal despite significant underlying liver fibrosis, hepatitis, steatohepatitis or even cirrhosis. Their relationships are far from linear and with little predictive value in some cases. Newer non-invasive modalities are emerging but currently their roles are largely experimental. This review will discuss the role of serum biomarkers and imaging techniques as new modalities to screen for liver disease.

P37 IDENTIFYING CIRRHOTICS AT RISK OF PARACENTESIS-INDUCED CIRCULATORY DYSFUNCTION (PICD): THE SIGNIFICANCE OF AN EARLY FALL IN STROKE VOLUME

Background and Aims: In studies using invasive techniques with intermittent measurements, a proportion of patients undergoing large volume paracentesis (LVP) with albumin infusion appear to still develop PICD. The mechanisms for circulatory disturbance in this situation are not fully understood. We conducted a pilot study to characterise in detail haemodynamics during drainage using continuous, non-invasive monitoring. Methods: 10 patients with tense ascites were recruited prospectively. Participants underwent non-invasive haemodynamic monitoring during LVP, with albumin given at 6 hours, using a FinometerTM to provide beat-by-beat measurement of blood pressure, cardiac output (CO), stroke volume (SV), and systemic vascular resistance (SVR). Results: Finometry enabled assessment of haemodynamic parameters at multiple time points, of which a sample is shown in the table.

P24 Presence of impaired baroreceptor sensitivity is a poor prognostic marker in cirrhosis

Introduction Autonomic function is essential for blood pressure control and baroreceptor sensitivity (BRS) acts as a composite marker of overall function. Both sympathetic and parasympathetic function is impaired in cirrhosis. Impaired BRS predicts death in cardiovascular diseases and chronic kidney disease. Aim The aim of the present study was to assess the relationship between BRS and liver disease severity, systemic and portal haemodynamics and mortality in cirrhosis. Method Prospective study of 29 cirrhotic patients. Systemic haemodynamics and BRS were assessed non-invasively using the Finometer.® Spontaneous BRS was calculated from the regression of pulse interval on systolic blood pressure. Portal pressure was assessed by measurement of the hepatic venous pressure gradient (HVPG). Gastroscopy assessed variceal size and 1-year probability of bleeding according to the NIEC index. Results 20 male (69%), median age 47 years (42–55), Child-Pugh score 6 (Class A 18, B 10, C 1) and MELD 10 (8–13). BRS was impaired in cirrhosis (median 4.2 ms/mm Hg, IQR 2.5–6.2 ms/mm Hg) but was not related to Child-Pugh score or MELD. Significant differences in BRS were seen with respect to gender (Female 2.0 ms/mm Hg vs male 5.8 ms/mm Hg, p=0.0125), presence of varices (present 3.8 ms/mm Hg vs absent 8.3 ms/mm Hg, p=0.0206), and ascites (ascites 2.0 ms/mm Hg vs no ascites 5.3 ms/mm Hg, p=0.0433). No significant differences in BRS were seen according to alcohol intake. A significant negative correlation was seen between BRS and age (r=−0.46, p=0.0130), heart rate (r=−0.56, p=0.0015), HVPG (r=−0.69, p<0.0001) and 1-year probability of variceal bleeding (r=−0.43, p=0.0199). Over a median follow-up of 710 days, 8/29 (27.6%) patients died with median time to death 854 days. Mortality was significantly higher in patients with HVPG greater than or equal to 10 mm Hg than without clinically significant portal hypertension (p=0.0024) and with a MELD score greater than or equal to 17 (p<0.0001). When stratified according to BRS (less than, or greater than or equal to 6 ms/mm Hg), mortality was significantly higher in patients with more severe impairment (<6 ms/mm Hg) than with preserved BRS (33% vs 12.5%, p=0.0034) despite no significant difference in Child-Pugh score, MELD, systemic haemodynamics, HVPG or length of follow-up. Patients with impaired BRS had significantly higher probability of variceal bleeding, and were more likely to have ascites (24% vs 0%). Conclusion BRS is not related to liver disease severity as assessed by Child-Pugh or MELD but a significant inverse correlation exists between BRS and HVPG. Compared to patients with preserved BRS, patients with significant impairment of BRS have a higher mortality, a higher probability of variceal bleeding and are more likely to have ascites. Impairment of BRS appears to be a poor prognostic factor independent of liver disease severity or portal pressure, possibly relating to a failure to respond to the haemodynamic challenges associated with complications in cirrhosis.

PTU-066 Non-invasive assessment of systemic haemodynamics to determine variceal bleeding risk in cirrhotic patients: Abstract PTU-066

Introduction Variceal bleeding is one of the most severe complications of portal hypertension. Universal endoscopic variceal screening is recommended and primary prophylaxis given based on oesophageal variceal size. Risk of bleeding also relates to other factors such as portal pressure, liver disease severity, and red wale markings at endoscopy. Currently there is no non-invasive way of measuring portal pressure or bleeding risk. The aim of this study was to evaluate whether non-invasive assessment of systemic haemodynamics in cirrhosis can identify bleeding risk in portal hypertension. Methods We studied 29 cirrhotic patients. Systemic haemodynamics and baroreceptor sensitivity (BRS) were assessed non-invasively using the Finometer® (TNO instruments, Amsterdam), and analysed with Beatscope® software. Spontaneous BRS was assessed by studying the relationship between inter-beat variability and beat-to-beat changes in systolic blood pressure. Portal pressure was assessed by measurement of the hepatic venous pressure gradient (HVPG). Gastroscopy assessed variceal size, Japanese score and 1-year probability of bleeding according to the NIEC index. Results 69% male, median age 47 (42–55) years, Child-Pugh (CP) score 6 (Class A 18, Class B 10, Class C 1) and MELD 10 (8–13). 90% alcoholic cirrhosis, 66% abstinent. HVPG correlated positively with variceal size (r=0.64, p<0.001), and all measures of variceal bleeding risk (Abstract 066). Significant positive correlations were seen with CI, HR and SVR and variceal bleeding risk, while a significant negative correlation was seen with BRS and bleeding risk. HVPG correlated positively with both CI (r=0.53, p=0.005) and HR (r=0.62, p<0.001) and negatively correlated with BRS (r=−0.69, p<0.001). BRS correlated negatively with HR (r=−0.56, p=0.002). Abstract PTU-066 Japanese score NIEC score 1 year probability of bleeding Child-Pugh score Cardiac output l/min r=0.38, p=0.042 r=0.46, p=0.012 r=0.49, p=0.006 r=0.31, p=0.103 Cardiac index l/min/m2 r=0.50, p=0.006 r=0.53, p=0.003 r=0.52, p=0.004 r=0.36, p=0.055 Peripheral vascular resistance MU r=−0.26, p=0.165 r=−0.32, p=0.093 r=−0.36, p=0.053 r=−0.18, p=0.355 Heart rate bpm r=0.27, p=0.160 r=0.39, p=0.038 r=0.35, p=0.065 r=0.37, p=0.048 BRS ms/mm Hg r=−0.41, p=0.026 r=−0.33, p=0.077 r=−0.43, p=0.020 r=−0.18, p=0.338 HVPG mm Hg r=0.61, p<0.001 r=0.65, p<0.001 r=0.72, p<0.001 r=0.58, p=0.005 Child-Pugh score r=0.49, p=0.008 r=0.75, p<0.001 r=0.72, p<0.001 Conclusion Non-invasive assessment of systemic haemodynamics in cirrhosis may be useful in predicting HVPG and bleeding risk from oesophageal varices. This study highlights that other factors such as baroreceptor sensitivity are important in determining portal pressure and bleeding risk and may open up potential new treatment options.

P16 Non-invasive detection of oesophageal varices: comparison of non-invasive assessment of systemic haemodynamics with laboratory parameters and predictive scores

Introduction Endoscopic screening for varices (OV) is advised in cirrhosis, repeated every 1–3 years, with primary prophylaxis given to large OV. This is costly to endoscopy units, unpleasant for patients and multiple procedures may affect compliance. Cirrhosis is characterised by a hyperdynamic circulation; novel tools make non-invasive assessment possible. Aim Toassess accuracy of non-invasive tests of systemic haemodynamics (Cardiac output and index (CO, CI), systemic vascular resistance (SVR), baroreceptor sensitivity (BRS) to detect OV and compare with other non-invasive methods (Child–Pugh score (CPS), albumin, platelet count, prothrombin time (PT), AST/ALT ratio, platelet count/spleen diameter ratio). Method Prospective study of 29 cirrhotic patients. Systemic haemodynamics were assessed non-invasively with the Finometer®, hepatic venous pressure gradient (HVPG) assessed portal pressure and gastroscopy for variceal size (none/small (absent), medium/large (large)). Results 69% male, median age 47 (42–55) years, CPS 6 (Class A 18, B 10, C 1) and MELD 10 (8–13). Prevalence OV 79%, large 38%. Significant differences in haemodynamics were seen between patients classified as absent or large OV (CO 5.6 vs 8.0lpm, CI 3.0 vs 4.5 l/min/m2, SVR 1.17 vs 0.77MU, HVPG 14 vs 19 mm Hg, BRS 5.8 vs 3.2 ms/mm Hg, CPS 5 vs 7, respectively). Comparisons summarised in the Abstract P16 Table 1. At a cutoff of 7.15lpm, CO predicted large OV with 73% sensitivity, 78% specificity and correctly classified 76% of patients. At a cutoff of 3.66 l//min/m2, CI predicted large OV with 82% sensitivity, 83% specificity and correctly classified 79% of patients. This compares to 78% correctly classified using HVPG, 76% CPS and 59% PT.Abstract P16 Table 1 Results Non-invasive test Presence OV AUROC (95% CI) Statistical significance Large OV AUROC (95% CI) Statistical significance CO 0.71 (0.49 to 0.92) p=0.029 0.84 (0.69 to 0.99) p<0.001 CI 0.76 (0.51 to 0.95) p=0.004 0.86 (0.71 to 1.0) p<0.001 SVR 0.63 (0.40 to 0.86) NS 0.77 (0.59 to 0.94) p=0.002 HVPG 0.91 (0.74 to 1.0) p<0.001 0.81 (0.64 to 0.98) p<0.001 BRS 0.81 (0.53 to 0.96) p=0.014 0.81 (0.64 to 0.98) NS Platelet count 0.75 (0.53 to 0.96) p=0.011 0.59 (0.36 to 0.82) NS Albumin 0.32 (0.02 to 0.61) NS 0.63 (0.42 to 0.84) NS PT 0.89 (0.79 to 0.99) p=0.015 0.70 (0.52 to 0.88) p=0.015 CPS 0.85 (0.75 to 0.95) p<0.001 0.74 (0.54 to 0.94) p=0.009 AST/ALT ratio 0.73 (0.55 to 0.91) p=0.006 0.68 (0.46 to 0.91) NS Platelet/spleen 0.78 (0.59 to 0.97) p=0.002 0.69 (0.48 to 0.90) p=0.040 Conclusion Non-invasive assessment of systemic haemodynamics appears a promising technique to identify cirrhotic patients at risk of having large oesophageal varices. Larger prospective validation studies need to be performed. Standard laboratory tests and predictive scores (except Child–Pugh score) are not reliable to accurately predict large oesophageal varices.