Janakan Selvarajah, Maddison Terlato, Weilun Gao, Omar Salehi, Christine Lu, Faris Gondal, Karen Lu, Angela Zhu, Spiros Fourlanos, Ashok S Raj
In this quaternary diabetes clinic, ICD-10-AM coding substantially under-recognised MASLD compared with FibroScan®, which identified a high prevalence of MASLD and significant fibrosis. These findings highlight important gaps in diagnostic coding and clinical awareness of MASLD in T2D.
BACKGROUND: Metabolic dysfunction-associated steatotic liver disease (MASLD), previously termed non-alcoholic fatty liver disease, affects an estimated 30%-70% of people with type 2 diabetes (T2D), with prevalence likely to be higher in quaternary diabetes clinics. Documentation using the International Classification of Diseases, Tenth Revision, Australian Modification (ICD-10-AM) may substantially underestimate disease burden.
AIMS: To compare the prevalence of MASLD identified by ICD-10-AM coding with imaging-based diagnosis using transient elastography (FibroScan®) and determine disease severity and clinical characteristics associated with significant fibrosis.
METHODS: Adults attending a quaternary diabetes clinic underwent retrospective review of electronic health records (August 2020-December 2023) for ICD-10-AM-coded MASLD, excluding other chronic liver diseases. A prospective cohort of consecutive clinic attendees underwent FibroScan® assessment for imaging-based diagnosis of MASLD and fibrosis staging. Clinical characteristics associated with significant fibrosis were analysed.
RESULTS: ICD-10-AM coding identified MASLD in 3% (85/2589) of patients, compared with 83% (81/103) diagnosed by FibroScan®. Significant fibrosis was present in 45% (46/103). Patients with significant fibrosis had higher body mass index (34 vs 28 kg/m2, P < 0.01), alanine aminotransferase (40 vs 26 U/L, P < 0.001), gamma-glutamyl transferase (54 vs 30 U/L, P < 0.001), lower high-density lipoprotein cholesterol (1.03 vs 1.15 mmol/L, P = 0.04), younger age (62 vs 67 years, P = 0.05) and shorter diabetes duration (11 vs 18 years, P = 0.02).
CONCLUSIONS: In this quaternary diabetes clinic, ICD-10-AM coding substantially under-recognised MASLD compared with FibroScan®, which identified a high prevalence of MASLD and significant fibrosis. These findings highlight important gaps in diagnostic coding and clinical awareness of MASLD in T2D.