The (II)logic of iron reduction therapy for steatohepatitis

Potential conflict of interest: Nothing to report. See Article on Page 751 View this article online at wileyonlinelibrary.com. A tool in the technique of logic that dates back to Aristotle is the use of a syllogism to prove an arguable point. In the context of iron reduction for steatohepatitis, it could unfold as follows: Major premise: An elevated serum ferritin can be associated with liver iron overload. Minor premise: Steatohepatitis is often associated with an elevated serum ferritin. Conclusion: Steatohepatitis can be associated with liver iron overload. The strength of the conclusion depends on the power or accuracy of the major and minor premise. There is no doubt that steatohepatitis patients usually have an elevated serum ferritin. However, the major premise that a high serum ferritin is an indicator of liver iron overload is where the problem begins with steatohepatitis (Fig. 1A). The overinterpretation of serum ferritin is a major clinical problem that leads to the overdiagnosis of hemochromatosis and or iron overload.1Figure 1: (A) A typical patient with NASH, showing 2+ iron staining; serum ferritin was 839 µg/L with a normal liver iron concentration at 29 µmol/g dry weight (reference range: 0‐35). (B) An 89‐year‐old female who is a C282Y homozygote for the HFE gene. Liver iron is 4+, and liver iron concentration is greatly elevated at 329 µmol/g. Of interest is the fact that there is no liver fibrosis, and this liver was successfully transplanted into a 52‐year‐old female.Therapeutic bleeding has a long tradition in medieval medicine, and previous studies have shown reductions in liver transaminases postphlebotomy in a variety of liver diseases not commonly associated with iron overload. The assessment of liver iron overload is traditionally done by liver biopsy and, ideally, liver iron concentration. Semiquantitative iron staining (Prussian Blue 0‐4) has usually shown that only those with 3 or 4+ staining will have elevations in liver iron concentration. Magnetic resonance imaging (MRI) scanning is a noninvasive indirect estimate of liver iron stores. In the study of Adams et al.,3 the treated patients had a mean serum ferritin of 255 µg/L, which is within the reference range. Estimates of liver iron concentration by MRI were in the middle of the normal range at 23 µmol/g. In the study of Hoki et al.,4 mean serum ferritin was 182 µg/L, and of the 40 patients reported, there were only 2 of 40 (5%) with 3+ iron staining and none with 4+ staining. Therefore, it may not be surprising that iron reduction therapy had no obvious benefits given that the patients did not have iron overload. There are still investigators that emphasize the regional distribution of iron in fatty liver disease in macrophages, rather than hepatocytes, and speculate that this could be contributing to oxidative stress. It is important to remember that it is very difficult to induce liver fibrosis in experimental animals with massive iron loading, and there are clinical patients with extreme iron overload without fibrosis or cirrhosis (Fig. 1B). The findings in these studies are similar to a study in 43 patients by Beaton et al.5 that did paired liver biopsies before and after phlebotomy therapy. Only 2 of 43 (5%) patients had an elevated liver iron concentration initially and there were no beneficial effects on a range of histological parameters, including fibrosis. Serum ferritin decreased with phlebotomy therapy as liver iron concentration decreased from the normal range to the low normal range. There was not a strong correlation between liver inflammation, C‐reactive protein, body mass index, and serum ferritin.6 Hoki et al. also suggest that their patients with nonalcoholic steatohepatitis (NASH) had increased duodenal iron absorption mediated though the iron‐regulatory protein and divalent metal transporter 1 (DMT1). They studied nonheme iron absorption with an oral iron tolerance test and showed differences between NASH, simple steatosis, and control patients. Given that their patients did not have hepatic iron overload, it is not clear whether this is a significant finding. Iron absorption is just an initial phase of a complex orchestration of iron proteins, and genes and iron uptake by the liver, iron storage, metabolism, and excretion are also part of the cascade of events. An interesting observation in this study was that increased iron absorption occurred in the presence of an elevated serum hepcidin. Hepcidin had originally been considered to be the conductor of the iron orchestra, but it has been shown that you can have iron overload with a low serum hepcidin or iron overload with a high serum hepcidin. An analogy would be the observations that clinical diabetes can occur with a low or high serum insulin. Incubation with serum from NASH patients suggested that there may be other humoral mediators that can increase DMT1 and iron absorption. Erythroferrone is the newest mediator that could regulate hepcidin and is produced by erythroblasts.7 It is possible that new factors will be discovered that will add to the complexity of human iron metabolism. All of these observations shed light on a common clinical scenario. The obese patient arrives in the clinic with mild elevations in aspartate aminotransferase, alanine aminotransferase, and serum ferritin. They have two glasses of wine per day and have already been told by their other physicians that they clearly have hemochromatosis. Genetic testing for HFE mutations is normal and they are not interested in a liver biopsy. You tell them not to worry about the ferritin elevation because it may be related to inflammation of fatty liver. However, they want to die with normal numbers and you suggest they become a voluntary blood donor. There is some evidence that an elevated serum ferritin is a marker of increased mortality, but this is likely related to the underlying disease,8 rather than a toxic effect of circulating ferritin. After donating six times, the ferritin is low normal and the patient asks why the ferritin went down if it was related to inflammation. It seems now that the ferritin decreased because the liver iron decreased and the serum ferritin was higher than expected for a given liver iron concentration (higher set point). Many of these management dilemmas would be resolved by expanding the reference range for serum ferritin (widening the goal posts), as suggested by McKinnon et al.10 Thus, it seems illogical to treat most NASH patients by phlebotomy because they never had iron overload. The initial flaw in logic was the assumption that an elevated serum ferritin represented iron overload. The misinterpretation of serum ferritin by clinicians and patients remains and could be improved by education and newer methods to noninvasively assess iron overload.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.