A care path for the detection of advanced NAFLD-fibrosis: the Nijmegen-Leiden-Amsterdam 2-tiered care path study

Non-alcoholic fatty liver disease (NAFLD) is a disease of alarmingly increasing
prevalence. Progression along the NAFLD spectrum often goes unnoticed since it
is often asymptomatic. Awareness among health care workers and implementation
of care paths to detect progressing NAFLD stages are limited. Without clear
guidance papers or robust care pathways for risk stratification, the current
diagnostic approach for NAFLD is highly variable, leading to both
underdiagnosis of advanded stages of disease, and unnecessary referrals for
mild stages of disease. This calls for a comprehensive care path consisting of
non-invasive alternatives to detect those patients with severe cases of NAFLD.
Particularly the use of a sequential, two-tiered care path algorithm is
promising as it has the ability to detect underlying advanced cases of
fibrosis, and has previously been shown to be cost-effective. This was shown by
dr. Ankur Srivastava, who designed a pathway consisting of FIB4-score and
ELF-test that led to a reduction of unnecessary referrals to the hepatologist
by 80%, whilst improving the detection of advanced fibrosis and cirrhosis 5-
and 3-fold, respectively (8). In this study we propose the investigation of
several two-tiered sequential care path algorithms, comprised of the
FIB4-score, VCTE and the ELF-test, for the detection of advanced stages of
NAFLD-fibrosis: the Nijmegen-Leiden-Amsterdam NAFLD-NASH 2-tiered care path
study: NLA2-study.

The aim of the study is to improve case finding of advanced cases of NAFLD (>=F3
fibrosis), whilst simultaneously reducing unnecessary referrals for mild cases
(NAFLD-related complications, and to assess the cost-effectiveness of the
different proposed care paths compared to current regular care.

This is a care innovation study, with an estimated duration of three years. We
intend to commence the study at three academic medical centres namely in
Nijmegen, Leiden and Amsterdam, with the intention to include other
non-academic hospitals after the initial roll-out. The study has both a
prospective and a retrospective part. The prospective part consists of
participants who are deemed by their treating physician to be at risk of severe
NASH fibrosis. Participants will be invited to attend a study visit at the
local hospital.

This study visit will consist of, among others: anthropometric measurements,
blood pressure measurement, blood sampling and VCTE. The diagnostic testing for
potentially underlying advanced (>=F3) liver fibrosis consists of the
FIB4-score, VCTE and the ELF-test. A blood sample will be stored for additional
biomarker testing. Based on predefined cut-offs for the FIB4-score and liver
stiffness measurement (LSM) (measured using VCTE), participants will be
classified as being at low or high risk of advanced (>=F3) fibrosis (see figure
1). The ELF-test will be analysed in bulk and will thus not be used for risk
assessment. Participants classified at low risk will remain under the care of
their treating physician. Participants classified at high risk of advanced
(>=F3) fibrosis will be referred to a hepatologist.

Read-outs of the electronic health records (EHR) of all participants will be
performed at 24 months after inclusion in the study, and at six months for
those classified at high risk. Read-outs will be performed to assess the
correctness of the risk assessment and subsequent referral to the hepatologist.

The three different sequential, two-tiered care path algorithms will be
evaluated upon completion of the study. The diagnostic accuracy, defined as
sensitivity, specificity, predictive values and AUROCs, of the three different
care path algorithms will be calculated. The diagnostic performance will be
expressed as the percentage of correct referrals and the percentage of
unnecessary referrals of the different care path algorithms and the individual
non-invasive tests, compared to regular care.

All tests are non-invasive tests and pose minimal burden on and minimal to no
risk to participants. There is the possibility that severe cases of NAFLD,
including cirrhosis, will be diagnosed as a result of the study procedures.
This would result in participants remaining under care of a hepatologist for
screening of HCC and esophageal varices and could imply repetitive invasive and
uncomfortable procedures such as esophagogastroduodenoscopy. Benefits from
participation include comprehensive diagnostic workup for potentially
underlying severe NAFLD, which is currently highly variable. Moreover, on the
population level, the implementation of a clear care path for the detection of
advanced (>=F3) NAFLD-fibrosis would allow for the identification of those
patients who would benefit from additional treatment options (e.g. bariatric
surgery) or intensification of treatment regiments. This would decrease the
number of patients progressing to end-stage liver disease, such as cirrhosis
and HCC, and would decrease both liver related and overall mortality.