Probability of C282Y homozygosity decreases as liver transaminase activities increase in participants with hyperferritinemia in the hemochromatosis and iron overload screening study

Authors


  • Potential conflict of interest: Nothing to report.

  • The Hemochromatosis and Iron Overload Screening Study was initiated and funded by the National Heart, Lung, and Blood Institute, in conjunction with the National Human Genome Research Institute. The study was supported by contracts N01-HC-05185 (University of Minnesota), N01-HC-05186 (Howard University), N01-HC-05188 (University of Alabama at Birmingham), N01-HC-05189 (Kaiser Permanente Center for Health Research), N01-HC-05190 (University of California, Irvine), N01-HC-05191 (London Health Sciences Center), and N01-HC-05192 (Wake Forest University).

Abstract

Hemochromatosis is considered by many to be an uncommon disorder, although the prevalence of HFE (High Iron) 282 Cys → Tyr (C282Y) homozygosity is relatively high in Caucasians. Liver disease is one of the most consistent findings in advanced iron overload resulting from hemochromatosis. Liver clinics are often thought to be ideal venues for diagnosis of hemochromatosis, but diagnosis rates are often low. The Hemochromatosis and Iron Overload Screening (HEIRS) Study screened 99,711 primary care participants in North America for iron overload using serum ferritin and transferrin saturation measurements and HFE genotyping. In this HEIRS substudy, serum hepatic transaminases activities (e.g., alanine aminotransferase [ALT] and aspartate aminotransferase [AST]) were compared between 162 C282Y homozygotes and 1,367 nonhomozygotes with serum ferritin levels >300 μg/L in men and >200 μg/L in women and transferrin saturation >45% in women and 50% in men. The probability of being a C282Y homozygote was determined for AST and ALT ranges. Mean ALT and AST activities were significantly lower in C282Y homozygotes than nonhomozygotes. The probability of being a C282Y homozygote increased as the ALT and AST activities decreased. Conclusion: Patients with hyperferritinemia are more likely to be C282Y homozygotes if they have normal liver transaminase activities. This paradox could explain the low yields of hemochromatosis screening reported by some liver clinics. (HEPATOLOGY 2012;55:1722–1726)

One of the most common genetic disorders in Caucasians is hemochromatosis. Liver disease is the most prevalent, serious complication of iron overload resulting from hemochromatosis, and consequential cirrhosis and hepatocellular carcinoma are common causes of death.1 Hemochromatosis is not an inflammatory liver disease. Liver biopsies from patients with hemochromatosis typically show iron overload, with or without liver fibrosis, and an absence of lymphocytes, leucocytes, and eosinophils. Serum alanine aminotransaminase (ALT) and aspartate aminotransaminase (AST) leak into the circulation as a result of necrosis of hepatocytes and are routinely measured as markers of hepatocellular disease.

Many patients are referred to liver clinics for evaluation of elevations in serum ferritin. In such patients, it is common to measure serum transaminases. Other pertinent tests include serum transferrin saturation and HFE genotyping. Hepatitis B surface antigen (HBsAg) and anti-HCV (hepatitis C virus) are tested in many patients with an elevated serum ALT. Most physicians assume that elevations of serum transaminase activities increase the probability that a patient has hemochromatosis because this is the case with many liver diseases. We found that the probability of HFE 282 Cys → Tyr (C282Y) homozygosity decreases as the serum transaminase activities increase.

Abbreviations

ALT, alanine aminotransaminase; AST, aspartate aminotransaminase; C282Y, 282 Cys → Tyr; H63D, 63 His → Asp; HBsAg, hepatitis B surface antigen; HCV, hepatitis C virus; HEIRS, Hemochromatosis and Iron Overload Screening Study; HFE, High Iron Fe.

Patients and Methods

The study design and overall results of the Hemochromatosis and Iron Overload Screening (HEIRS) Study have been previously reported.2-4 The HEIRS Study was approved by all local investigational review boards. Participants ≥25 years of age who gave informed consent were recruited from five field centers that serve ethnically and socioeconomically diverse populations. All participants had random testing for serum transferrin saturation and serum ferritin levels (without intentional fasting) and genotyping to detect the common C282Y and H63D mutations of the HFE gene. Participants who reported a previous diagnosis of hemochromatosis or iron overload (treated or untreated) were excluded.

Postscreening clinical examinations were performed on participants with elevated transferrin saturation (>45% in women and >50% in men) and ferritin (>300 μg/L for men and >200 μg/L for women), all HFE C282Y homozygotes, and control participants (matched for age, gender, and race) with normal transferrin saturation and serum ferritin values, but without HFE C282Y or H63D mutations. Of 2,265 participants invited for clinical examinations, there was a 75% participation rate. Among C282Y homozygotes (n = 333), the participation rate was 91%. In this study, only participants with an elevated serum ferritin and transferrin saturation were analyzed, because participants with a normal serum ferritin level were considered to have a low probability of having liver disease. In the HEIRS Study, an elevated serum ferritin level was found in 88% of male and 57% of female C282Y homozygotes.2 These clinical examinations included measurements of serum ALT, AST, and ferritin. Self-reported daily ethanol consumption was collected and reported as g/day.

For analysis, intervals of serum ALT and AST activities were analyzed: (0,19), (20, 39), (40, 59), (60, 79), (80, 99), and >100 IU/L, respectively. There were no homozygotes with AST or ALT levels above 119 IU/L. The probability of being a C282Y homozygote was calculated for each ALT and AST interval and for gender-specific groups with and without an elevated AST and ALT level (>40 IU/L). The trend in probabilities was tested with a chi-square test for linear trend with 1 degree of freedom. All analyses were performed using OpenEpi software (version 2.3.1; Emory University, Atlanta, GA). A subgroup analysis was performed on only Caucasian participants. An adjusted Mantel-Haenszel chi-square test was used to determine whether the overall trend remained after adjustment for gender. Pearson's correlation coefficients were calculated for the relationship of ALT to ferritin.

Results

The participants included 80 female C282Y homozygotes, 82 male C282Y homozygotes, 575 female non-C282Y homozygotes, and 792 male non-C282Y homozygotes. All participants in this study had an elevated ferritin and transferrin saturation. Of C282Y homozygotes, 97% were Caucasian. In the nonhomozygotes, 41% were Caucasian.

Other genotypes in non-C282Y homozygous participants included wild type (i.e., no C282Y or H63D mutations) in 886, C282Y heterozygosity in 109, compound heterozygosity (C282Y/H63D) in 87, H63D homozygosity in 55, and H63D heterozygosity in 230. The profile of the participants is shown in Table 1. The investigation of the etiology of elevated ALT or AST activities in the non-C282Y homozygotes was beyond the primary scope of the HEIRS Study, although we previously reported the prevalence of viral hepatitis and the results of liver biopsies in selected HEIRS Study participants.5

Table 1. Participant Profile
Gender/genotypenAgeALT (IU/L)AST (IU/L)Ferritin (μg/L)Ethanol (g/day)HBsAg+Anti-HCV+
  1. Data are expressed as arithmetic mean (95% confidence interval of the mean). ALT was significantly greater in female non-C282Y homozygotes, compared to C282Y homozygotes (P = 0.003). ALT was significantly greater in male non-C282Y homozygotes, compared to C282Y homozygotes (P = 0.05). AST was significantly greater in female non-C282Y homozygotes, compared to C282Y homozygotes (P = 0.001). AST was significantly greater in male non-C282Y homozygotes, compared to C282Y homozygotes (P = 0.007).

Female C282Y homozygotes8055 (52-57)24 (19-30)25 (19-31)641 (539-742)7.8 (0-16)00
Female non-C282Y homozygotes57556 (55-57)37.2 (34-40)41.2 (38-45)526 (477-574)9.1 (5.8-12.3)1283
Male C282Y homozygotes8252 (49-55)37 (31-42)29 (26-33)1,118 (933-1,303)11 (5-16)00
Male non-C282Y homozygotes79253 (51.6-53.4)48.2 (44-52)43.2 (40-46)689 (645-733)12.8 (10-15.5)28127

Mean serum ALT and AST activities were significantly lower in C282Y homozygotes than in nonhomozygotes (Table 1). ALT and AST activities were significantly lower in female C282Y homozygotes than in male homozygotes. Among the female homozygotes, an ALT level <30 was observed in 65 of 80, with and AST level <30 in 69 of 80. The distribution of ALT values in relationship to serum ferritin in male and female C282Y homozygotes and nonhomozygotes is shown in Figs. 1A and 1B. In these figures, it is demonstrated that many C282Y homozygotes have normal ALT levels, but also that patients with an elevated ALT level are unlikely to be C282Y homozygotes. The correlation between ALT and ferritin was stronger in C282Y homozygotes than in nonhomozygotes, which is consistent with an inflammatory cause of the hyperferritinemia in nonhomozygotes.

Figure 1A.

Serum ALT in male C282Y homozygotes (•) (r = 0.44, P < 0.0001) and nonhomozygotes (○) (r = 0.22, P < 0.0001). Data displayed are excerpted from observations in participants with a serum ferritin level <1,000 μg/L and ALT level <300 IU/L. Solid line represents the upper limit of the reference range (40 IU/L).

Figure 1B.

Serum ALT in female C282Y homozygotes (•) (r = 0.63, P < 0.0001) and nonhomozygotes (○) (r = 0.31, P < 0.0001). Data displayed are excerpted from observations in participants with a serum ferritin level <1,000 μg/L and ALT level <300 IU/L. Solid line represents the upper limit of the reference range (40 IU/L).

The proportion of male C282Y homozygotes with ALT and AST levels <40 IU/L was 71% and 87%, respectively. The proportion of female C282Y homozygotes with ALT and AST levels <40 IU/L was 87% and 95%, respectively.

The decreasing probability of being a C282Y homozygote across groups in men and women with increasing ALT levels is shown in Fig. 2. Similar results were determined for AST. P values for chi-square tests for trends in proportions for ALT were 0.036 for men and 0.00017 for women. Mantel-Haenszel chi-square adjusted for gender was <0.0001. An unanticipated observation was that the probability of being a C282Y homozygote decreased as serum ALT and AST levels increased. The results of the subgroup analysis limited to Caucasians were similar.

Figure 2.

Probability of being a C282Y homozygote in six groups according to serum ALT levels in men (▪) and women (□). The chi-square test for the linear test for trend in proportions was 0.036 in men and 0.00017 in women. Numbers above the bar represent the number of participants in each group.

Discussion

It is widely believed that the probability of diagnosing many liver diseases increases as serum transaminases increase. In the present study of subjects with hyperferritinemia, the probability of being a C282Y homozygote decreased with increasing ALT and AST levels. This probably occurs because the deposition of excessive iron alone in hepatocytes of persons with hemochromatosis is not inflammatory. “Silent” hepatic fibrosis occurs in some subjects with hemochromatosis and normal serum transaminases.6, 7 On the other hand, some patients with hemochromatosis and HFE C282Y homozygosity have both hepatic iron overload and an inflammatory liver condition. For example, approximately 15% of C282Y homozygotes diagnosed in medical care have severe hepatic steatosis proven by liver biopsy. These patients had higher median serum ALT and ferritin levels than C282Y homozygotes without hepatic steatosis or other inflammatory liver disorder.8

In contrast, patients referred for evaluation of elevated serum ferritin levels usually have hyperferritinemia resulting from inflammatory liver disease, rather than iron overload resulting from HFE hemochromatosis.9 In prospective analyses of subjects with chronic elevation of serum transaminases, hepatic steatosis associated with or without excessive ethanol consumption was the predominant cause of elevated serum transaminases.10-13 Hemochromatosis was rare in these case series.9

In the present study, there was a potential bias wherein HEIRS Study non-C282Y homozygous participants were deliberately selected for postscreening clinical examinations because they had elevated serum transferrin saturation and ferritin measures. The present results demonstrate that these participants had higher mean serum transaminase activities than did HFE C282Y homozygotes. Another potential source of uncertainty was that elevations of ALT are intermittent or unreproducible in a majority of outwardly healthy subjects,11 whereas the present results are based on single measurements of serum transaminase activities in subjects selected for iron phenotypes and HFE genotypes. The C282Y homozygotes identified by screening in this study had relatively modest serum ferritin elevations, for the most part, and are not representative of patients diagnosed in practice. Homozygotes with heavier iron burdens and consequent hepatocellular damage may have elevated transaminases.

The present results demonstrate that participants who had C282Y homozygosity uncomplicated by a liver disorder associated with inflammation (e.g., steatosis or HCV) are more likely to have normal serum transaminases and elevated serum ferritin levels. Persons with both elevated serum transaminase and elevated serum ferritin levels are less likely to be C282Y homozygotes. Thus, it is also predicted that the proportion of patients who present with both elevated serum transaminases and hyperferritinemia who are C282Y homozygotes with iron overload without concomitant inflammatory liver disease is relatively small.8, 9, 11 Our observations and prediction are consistent with the low rates of detection of HFE C282Y homozygotes observed in liver clinics,14 because many of these homozygotes also have normal serum transaminases. In a retrospective analysis of physicians' evaluations of 100 consecutive patients in whom mild elevations of ALT and AST were observed, evaluation to exclude hemochromatosis was not performed in 90% of subjects.15 Taken together, these observations suggest that some physicians are reluctant to evaluate patients for HFE hemochromatosis because they erroneously believe that this condition is typically associated with elevated serum transaminases. We conclude that all Caucasian patients with hyperferritinemia should be evaluated for HFE hemochromatosis, regardless of serum transaminases. Other tools that can aid in the detection of HFE hemochromatosis include elevated serum transferrin saturation16 and family history.17, 18

Ancillary