In compensated cirrhosis, a spleen stiffness of 40 kPa or below has been accepted as sufficient to skip screening endoscopy. In chronic portal vein thrombosis without cirrhosis, everyone still gets scoped — because liver stiffness is normal in these patients, so the cirrhosis cut-offs cannot be transferred. This study tested whether spleen stiffness can do the job instead.
Investigators across 16 VALDIG centres retrospectively identified patients with chronic portal vein thrombosis and no cirrhosis who had upper endoscopy within two years of a spleen stiffness measurement by vibration-controlled transient elastography, splitting them into derivation and validation cohorts. High-risk varices were present in 43% of the 159 derivation patients and 32% of the 187 validation patients.
On multivariable analysis, only spleen stiffness remained associated with high-risk varices in both cohorts, despite many other candidates — myeloproliferative neoplasm, ascites, haemoglobin, bilirubin, albumin, splenomegaly, collaterals — being associated on univariable testing.
A threshold of 40 kPa or below gave 97% sensitivity in the derivation cohort, sparing 41% of endoscopies and missing 3% of high-risk varices, with a 97% negative predictive value. In validation it spared 43% of endoscopies, missed 5%, and gave a 96% negative predictive value.
These are young patients with a lifetime of surveillance ahead, and the cumulative burden of repeated endoscopy — sedation risk, time off work, cost — is substantial. In Indian practice, where extrahepatic portal venous obstruction is a common cause of variceal bleeding in young people, a non-invasive test that halves the endoscopy burden is a meaningful change if elastography is available.
- Measure spleen stiffness by transient elastography in chronic portal vein thrombosis without cirrhosis before booking screening endoscopy
- A value of 40 kPa or below identifies a group with high-risk varices probability of 5% or less, in whom endoscopy can be deferred
- Do not apply liver stiffness cut-offs in this population — liver stiffness is normal and carries no information
- Accept that about 3 to 5% of high-risk varices will be missed, and discuss that explicitly with the patient
- Repeat the measurement annually rather than treating a single normal value as permanent
The statistics, in plain English
Negative predictive value is the number that matters for a rule-out test: 96 to 97% means that among patients below the threshold, 3 to 4 in 100 still had high-risk varices. That is the miss rate you are accepting, and whether it is tolerable depends on how bad a missed variceal bleed is against how many endoscopies are avoided. Note that negative predictive value depends on how common the condition is — high-risk varices were present in 32 to 43% of these cohorts, and in a lower-prevalence population the negative predictive value would be higher, in a higher-prevalence one lower. The study is retrospective, with endoscopy performed up to two years from the elastography, which introduces some misclassification.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for gastroenterology & hepatology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free