Elevated Serum Transferrin Saturation and Mortality

The two companion papers of Dr. Mainous and colleagues in the March/April 2004 issue of Annals of Family Medicine [1,2] provide additional evidence that an elevation of serum transferrin saturation is associated with a significant increase in all-cause mortality. They also show that persons self reporting a high dietary intake of iron, particularly with high red meat consumption, had a significant increased risk of death from all causes.

Of course, it would be useful to know more details about the subjects, particularly those with elevated serum transferrin saturations. In particular, in the NHANES surveys, serum transferrin saturations were measured only once, and they were not necessarily measured in a fasting state or at the same time of day. Both of these are important, because there is a diurnal variation in serum iron concentrations, and recent ingestion of iron-containing foods can increase serum iron concentrations substantially. Indeed, we sometimes measure serum iron levels before and after administration of iron supplements in order to assess whether patients are able to respond normally to such supplements with increased iron absorption from the gastrointestinal tract.

Then, too, we have no data about total body iron stores or about HFE gene status in the subjects studied. It is highly likely, however, that patients with one or more mutations in HFE, associated with classical hereditary hemochromatosis, were over represented in the subjects with serum transferrin saturations greater than 55%. As Mainous, et al., point out, heavy iron overload, such as occurs in classical hereditary hemochromatosis (usually due to homozygosity for the C282Y HFE mutation) clearly leads to increased mortality, especially due to hepatic cirrhosis, diabetes mellitus, cardiac disease, and development of hepatocellular carcinoma.

An important question is whether increased mortality is restricted to those with only the very highest body iron levels as reflected in transferrin saturations of 80 to 90%. It appears that this is not the case. After excluding from analysis those subjects who died or were diagnosed with cancer within 4 years of blood draw, we reported that cancer risk in the NHANES I cohort was elevated in subjects with baseline transferrin saturation of between 50 and 60% compared to <30% (RR=1.4), as well as in those with transferrin saturation >60% compared to <30% (RR=1.8) [3]. Similarly, Mainous et al. [1] reported that, for all cause mortality in the same cohort, the relative risk for those with baseline transferrin saturation between 55 and 60% was 1.5, and for those with transferrin saturation >60% it was 1.67.

Another finding in our original analysis of NHANES I [4], was an inverse association of baseline serum albumin level and cancer risk, which we speculated might be related to the relationship of body iron stores and risk. Others reported an inverse association of serum albumin with all- cause mortality in Britain that rivaled that of smoking and overall mortality (after excluding deaths within 5 years of blood draw). It would be interesting to know whether serum albumin is related to overall mortality in the NHANES cohorts as well.

In the second paper by Mainous et al. [2], an interesting interaction of high transferrin saturation with high iron intake is reported; mortality is increased only in those with both. If elevated transferrin saturation is a marker for HFE mutations, these results suggest that the mutations by themselves are not necessarily deleterious, and only become so when high amounts of iron are consumed. An extension of this idea is that HFE mutation carriers may be at increased sensitivity to prooxidant xenobiotic exposures, such as ionizing radiation [5].

The main take home message for family medicine physicians is that measurement of serum iron and iron binding capacity, with calculation of the serum transferrin saturation, might reasonably be added to screening lab tests for all adults. Such testing is best done in the morning after at least an 8-hour fast, to avoid possibly spurious increases. Those found to have increases, if confirmed, should undergo additional evaluation, including measurement of serum ferritin, assessment of liver status, more extensive dietary histories, and HFE mutational analyses. Long term interventional trials of dietary restriction and/or iron reduction by therapeutic phlebotomy would be logical and important extensions of the elegant work of Mainous and colleagues [1,2].

References

1. Mainous, III, AG, Gill, JM, Carek, PJ: Elevated serum transferrin saturation and mortality. Ann Fam Med 2004; 2: 133-138.

2. Mainous, III, AG, Wells, B, Carek, PJ, Gill, JM, Geesey, ME: The mortality risk of elevated serum transferrin saturation and consumption of dietary iron. Ann Fam Med 2004; 2: 139-144

3. Stevens, RG, Graubard, BI, Micozzi, MS, Neriishi, K, Blumberg, BS: Moderate elevation of body iron level and increased risk of cancer occurrence and death. Int J Cancer. 1994; 56:364-9.

4. Stevens, RG, Jones, DY, Micozzi, MS, Taylor, PR: Body iron stores and the risk of cancer. New Engl J Med. 1998; 319:1047-52.

5. Stevens, RG, Morris, JE, Anderson, LE: Hemochromatosis heterozygotes may constitute a radiation-sensitive subpopulation. Radiat Res. 2000; 153:844-7.

Competing interests:   None declared