High prevalence of hepatic fibrosis in the setting of coexisting diabetes and hepatic steatosis: A case for selective screening in the general population?

Potential conflict of interest: Dr. Harrison consults, advises, and is on the speakers’ bureau for Gilead. He consults for NGM, Nimbus, and Fibrogen. The opinions in this article do not constitute endorsement by the San Antonio Military Medical Center, the US Army Medical Department, the US Army Office of the Surgeon General, the Department of the Army, the Department of Defense, or the US government of the information contained therein. See Article on Page 138 Chronic liver disease, accounting for one in 40 deaths worldwide, is a major cause of morbidity and mortality globally.1 With rising rates of obesity, diabetes mellitus (DM), and metabolic syndrome, nonalcoholic fatty liver disease (NAFLD) has become the leading etiology of chronic liver disease worldwide.2 Prevalence estimates of NAFLD vary depending on the population studied and the accuracy of the diagnostic test.2 Despite these limitations, it is currently estimated that the global prevalence of NAFLD is as high as 1 billion.2 In the United States, there are an estimated 75million to 100 million individuals with NAFLD.3 Moreover, nonalcoholic steatohepatitis (NASH), the more advanced form of NAFLD, is the most rapidly growing indication for liver transplantation in the United States.4 To counter these alarming trends, improved understanding of predictive factors associated with poor outcomes is needed. Fibrosis stage is the strongest predictor for disease‐specific mortality in NAFLD, and no other histologic features are associated with long‐term outcomes of patients with NAFLD.5 In the largest cross‐sectional study to date, Koehler et al.7 investigated the prevalence of and factors associated with clinically relevant liver fibrosis diagnosed by employing transient elastography (TE) in a general population of older (≥45 years) Caucasians living in Ommoord, a district of Rotterdam, The Netherlands. It must be noted that limited literature exists on this important but difficult‐to‐study group. Their study underscores the utility of TE as a noninvasive technique to detect clinically relevant liver fibrosis while avoiding the risks associated with liver biopsy. Between January 2011 and September 2013, 3041 participants underwent TE with liver stiffness measurement (LSM). LSM ≥8.0 kPa was defined as the surrogate marker of clinically relevant liver fibrosis. Of 3041 participants (age 66.0 ± 7.6 years) with reliable LSM, 169 (5.6%) demonstrated LSM ≥8.0 kPa. In a multivariate analysis, increased age, higher alanine aminotransferase level, current or former smoking, larger spleen size, positive viral serologies (hepatitis B surface antigen or anti‐hepatitis C virus positivity), and combined presence of DM and steatosis (detected by ultrasound) were independently associated with LSM ≥8.0 kPa. However, only increased age, higher alanine aminotransferase, and combined DM and steatosis demonstrated statistically significant P values <0.001 (see Koehler et al.,7 table 3). In addition, this study demonstrated the independent relationship of aging and advanced hepatic fibrosis (higher LSM values). Notably, LSM increased with each higher age decade among participants without DM or steatosis but not among those with DM and steatosis.7 While the finding that LSM increased with age among those without DM or steatosis was not surprising, the latter result raises opportunities for further studies. Although there may admittedly be an element of survival effect bias influencing the reduced impact of age on LSM among participants with DM and steatosis in the oldest age groups, further investigation is warranted to confirm these observations. In comparison to other population‐based cross‐sectional studies with similar conclusions, a selection bias may have occurred in this study due to increased mean age, a disproportionately higher number of women, and a greater prevalence of metabolic syndrome.7 More importantly, this study demonstrates the significant associations between DM and hepatic steatosis with clinically relevant liver fibrosis as LSM ≥8.0 kPa was noted in 17.2% of participants with a combined presence of DM and steatosis.7 These findings highlight the increasing burden of liver disease that is emerging in parallel with high rates of metabolic syndrome and is consistent with a recent prospective study in diabetic patients from Hong Kong which found that 17.7% of patients had an increased LSM (≥9.6 kPa by M probe and ≥9.6 kPa by XL probe). The majority of these patients underwent liver biopsy, and it was found that 56% had NASH and 50% had advanced fibrosis.10 The authors performed an additional analysis on a subgroup of participants with ultrasonographically defined NAFLD. Among participants with NAFLD, the prevalence of LSM ≥8.0 kPa was 8.4%, a relative estimation of NASH.7 However, both steatosis and NASH may have been underestimated in this study. Limitations of employing ultrasound may result in underestimation of NAFLD due to poor detection capability of ultrasound in individuals with 5%‐30% hepatic steatosis and in 22.3% of participants in this study with an obesity‐range body mass index of ≥30 (see Koehler et al.,7 table 1). The other caveat is that steatosis tends to regress and disappear with progression to advanced fibrosis in the NASH population. Therefore, one of the major limitations of this study was the lack of histopathological data. As correctly pointed out by the authors, a liver biopsy in this cohort was not feasible. The TE‐based LSM cutoff value for advanced liver fibrosis may vary with etiology of liver disease, body mass index, impact of steatosis on LSM values, and level of probe sensitivity. Therefore, the role of TE needs to be further delineated in the setting of prospective longitudinal study design and compared to histopathological data in patients with NAFLD. While the authors demonstrate significant associations with LSM, further validation of TE‐based LSM as a surrogate for clinically relevant liver fibrosis should be corroborated by histopathological data, the gold standard. Previous studies identify TE as a sensitive modality for ruling out advanced hepatic fibrosis and cirrhosis, but data demonstrating the specificity of LSM for ruling in advanced fibrosis are less robust.9 Nevertheless, given the unethical nature of exposing a large population‐based cohort to liver biopsy, the authors provide valuable insights to the existing body of literature regarding the utility of TE.7 Another interesting finding of this study involves the link between previous or current smoking and elevated liver stiffness.7 Although this result differs from the recent study by Roulot et al.,9 the finding is plausible from a pathophysiological perspective. Population‐based cohort studies have demonstrated a link between smoking and risk of cirrhosis.11 Data on coffee consumption were not collected and may have been valuable as well. In summary, Koehler et al.7 have performed a large population‐based, cross‐sectional study in older adults demonstrating the presence of clinically relevant liver fibrosis in up to 5.6% of the general population, in 8.4% of participants with evidence of NAFLD on ultrasonography in the setting of DM and/or glucose intolerance, and in an alarming 17.2% of participants with combined presence of DM and steatosis. This is an important contribution, further confirming that presence of DM, a component of metabolic syndrome in patients with NAFLD, is associated with clinically relevant liver fibrosis. Moreover, the results forewarn about the rising tide of NAFLD, the hepatic manifestation of metabolic syndrome expected to emerge as a global public health issue in the near future. Policy makers and stakeholders must pay attention and invest in preventative measures. Despite decades of clinical trials, no single treatment can be recommended to all patients with NASH. A practical approach is warranted to identify patients with NAFLD with the highest likelihood for histologic evidence of NASH. The primary and secondary prevention of NAFLD may require selective screening strategies (i.e., TE) in the general population with a special focus on high‐risk individuals who present with obesity, DM, and/or metabolic syndrome. One approach may be to have radiology technicians, trained to perform shear wave measurements, obtain them when they would ordinarily perform a right upper quadrant ultrasound in those patients referred for imaging.