|Home | About | Journals | Submit | Contact Us | Français|
Relationships of thyroid and iron measures in large cohorts are unreported. We evaluated thyroid-stimulating hormone (TSH) and free thyroxine (T4) in white participants of the primary care–based Hemochromatosis and Iron Overload Screening (HEIRS) Study.
We measured serum TSH and free T4 in 176 HFE C282Y homozygotes without previous hemochromatosis diagnoses and in 312 controls without HFE C282Y or H63D who had normal serum iron measures and were matched to C282Y homozygotes for Field Center, age group, and initial screening date. We defined hypothyroidism as having TSH >5.00mIU/L and free T4 <0.70ng/dL, and hyperthyroidism as having TSH <0.400mIU/L and free T4 >1.85ng/dL. Multivariate analyses were performed using age, sex, Field Center, log10 serum ferritin (SF), HFE genotype, log10 TSH, and log10 free T4.
Prevalences of hypothyroidism in C282Y homozygotes and controls were 1.7% and 1.3%, respectively, and of hyperthyroidism 0% and 1.0%, respectively. Corresponding prevalences did not differ significantly. Correlations of log10 SF with log10 free T4 were positive (p=0.2368, C282Y homozygotes; p=0.0492, controls). Independent predictors of log10 free T4 were log10 TSH (negative association) and age (positive association); positive predictors of log10 SF were age, male sex, and C282Y homozygosity. Proportions of C282Y homozygotes and controls who took medications to supplement or suppress thyroid function did not differ significantly.
Prevalences of hypothyroidism and hyperthyroidism are similar in C282Y homozygotes without previous hemochromatosis diagnoses and controls. In controls, there is a significant positive association of SF with free T4. We conclude that there is no rationale for routine measurement of TSH or free T4 levels in hemochromatosis or iron overload screening programs.
Hemochromatosis in Western European whites is typically associated with homozygosity for C282Y, a mutation in the HFE gene on chromosome 6p (1). Approximately 0.44–0.48% of non-Hispanic whites in North America are C282Y homozygotes (2,3). Some C282Y homozygotes develop iron overload and complications such as liver disease, diabetes mellitus, arthropathy, and hypogonadotrophic hypogonadism (1–4). Among 391 patients with hemochromatosis diagnosed in medical care combined from five studies, 4.1% had primary hypothyroidism and 0.3% had hyperthyroidism (5–9). To date, there is no report of thyroid-related laboratory measures in participants of large population or workplace hemochromatosis screening programs (2,10–12).
The Hemochromatosis and Iron Overload Screening (HEIRS) Study is a cross-sectional, multi-center, multi-ethnic study of primary care clinic attendees (13). The HEIRS Study performed initial screening for hemochromatosis and iron overload using transferrin saturation (TS) and serum ferritin (SF) measurements and HFE genotyping (C282Y and H63D alleles) in 101,168 participants (13). In this HEIRS substudy, we measured and compared serum concentrations of thyroid-stimulating hormone (TSH), free thyroxine (T4), and SF at a post-initial screening examination in 176 white participants with C282Y homozygosity and in 312 white participants without HFE C282Y or H63D designated as control subjects. The implications of the present results for understanding the prevalence and pathogenesis of thyroid abnormalities in whites with hemochromatosis and C282Y homozygosity are discussed.
Participants were recruited during the interval February 2001–February 2003 from HEIRS Study Field Centers (13). These clinics serve ethnically and socioeconomically diverse primary care patients. At initial screening, blood samples were obtained from participants for measurement of TS and SF without regard for state of fasting, and for HFE mutation analysis (13). The dataset excluded observations on participants for whom TS, SF, or HFE genotype data on initial screening were not available. All adult volunteers were eligible to participate in the HEIRS Study, but those who participated only because a family member participated or those who reported on their screening questionnaire that they had been previously diagnosed to have hemochromatosis or iron overload were excluded from the present analysis (3).
All C282Y homozygotes diagnosed in the HEIRS Study were invited to participate in a clinical examination that included listing of participants' medications and collection of blood samples for additional testing (13). Control subjects invited to attend a clinical examination were selected from initial screening participants who had (i) HFE genotype wt/wt (defined as absence of alleles C282Y and H63D) and (ii) both TS and SF in the eligible range. Eligible ranges for men were TS 20–34% and SF 87–247ng/dL. Eligible ranges for women were TS 16–28% and SF 19–121ng/dL. Among participants eligible to be control subjects, a subset was selected that was frequency matched in a 1:1 ratio to the group of C282Y homozygotes and provisional iron overload cases (defined as participants with confirmed elevations of both SF and TS without evidence of elevated C-reactive protein or elevated serum levels of hepatic transaminases). Variables used for the frequency matching were Field Center (UAB, UCI, Howard, KP-Portland, LHSC, and Toronto), age group (24–44, 45–64, and 65+ years), and date of initial screening visit. Potential control participants who could not be contacted or who declined to participate were replaced. New potential control subjects were selected using frequency matching in a 1:1 ratio to the group of the potential control subjects who declined. Variables used for this frequency matching were Field Center (as above), age group (as above), date of initial screening, gender, and race/ethnicity. Most C282Y homozygotes in the HEIRS Study were white (3). Therefore, we included only those participants and control subjects who reported that they were only “white or Caucasian” (13).
Serum concentrations of TSH and free T4 were measured at the HEIRS Study Central Laboratory at the University of Minnesota Medical Center-Fairview (13) using a Siemens/Bayer ADVIA Centaur® immunoassay analyzer (Siemens, New York, NY). The coefficients of variation for the TSH (TSH-3) and free T4 (FrT4) assays in the normal range were 5.0% and 4.0%, respectively. The TSH assay has a functional sensitivity of approximately 0.019mIU/L. Reference ranges were TSH 0.400–5.00mIU/L and free T4 0.70–1.85ng/dL. These ranges represent the central 95% confidence intervals (CIs) of serum TSH and free T4 measurements in ostensibly healthy young adults. We defined hypothyroidism as having TSH >5.00mIU/L and free T4 <0.70ng/dL, and hyperthyroidism as having TSH <0.400mIU/L and free T4 >1.85ng/dL, regardless of cause. SF levels were measured using a turbidometric immunoassay (Roche Diagnostics/Hitachi 911, Indianapolis, IN) (3,14). The SF reference ranges defined by the HEIRS Study were 15–200ng/mL (women) and 15–300ng/mL (men) (3,14). We defined severe hyperferritinemia as SF >1000ng/mL.
We evaluated observations on 176 C282Y homozygotes and 312wt/wt control subjects. Statistical analyses were performed using SAS (15), Excel 2000® (Microsoft, Redmond, WA), and GB-Stat® (v. 10.0, 2003; Dynamic Microsystems, Silver Spring, MD). Initial evaluations revealed that TSH and free T4 data were not normally distributed. This is characteristic of thyroid screening programs, largely due to the relatively high prevalence of hypothyroidism (16–20). Similarly, SF values are not normally distributed (3,14). Therefore, TSH, free T4, and SF data were normalized using log10 transformation for analysis; some results were reported as antilog values, as appropriate. For display of TSH values, we used three significant figures. Descriptive data are displayed as enumerations, percentages, or mean±1 SD (or 95% CIs, as appropriate). Frequency values were compared using chi-square analysis or Fisher exact test, as indicated. Mean values were compared using Student's t-test. We determined the correlations of log10 SF as a dependent variable with log10 free T4 and with log10 TSH as independent variables. Multivariate analyses were performed using age, sex (male or female), Field Center, HFE genotype (either C282Y/C282Y or wt/wt), log10 SF, and log10 TSH as independent variables, and log10 free T4 as the dependent variable. In other multivariate analyses, we evaluated the determinants of log10 SF as the dependent variable. For correlation and multivariate analyses, we included only those observations from the 437 study participants who reported taking neither thyroid supplements nor medications to suppress thyroid function. Values of p<0.05 were defined as significant.
There was a predominance of women among the C282Y homozygotes and wt/wt control subjects, consistent with the overall greater participation of women than men in the HEIRS Study (Table 1) (3). C282Y homozygotes were significantly younger and had a significantly higher mean free T4 than control subjects in univariate analyses, although the magnitudes of these respective differences were small (Table 1). The prevalence of abnormal values of TSH, free T4, or both did not differ significantly between HFE C282Y homozygotes and control subjects (Table 1). Mean SF was significantly greater in C282Y homozygotes than in control subjects (Table 1).
Three participants had TSH and free T4 levels defined as hypothyroidism; each was a control subject (Table 2). Seven participants had TSH and free T4 levels defined as hyperthyroidism: three were C282Y homozygotes and four were control subjects (Table 2). The mean SF in the three participants with hypothyroidism was significantly lower than the mean SF in the seven participants with hyperthyroidism (Table 2). The respective prevalence estimates for hypothyroidism and hyperthyroidism did not differ significantly between C282Y homozygotes and control subjects (Table 1), or between male and female C282Y homozygotes and control subjects of the corresponding sex (data not shown).
Twenty-two HFE C282Y homozygotes (12.5%; 18 men and 4 women) had SF >1000ng/mL. None had hypothyroidism as defined herein, although one man reported taking thyroid supplements. Another man had hyperthyroidism. By study design, no HFE wt/wt control subjects had hyperferritinemia.
Forty-nine participants (15 HFE C282Y homozygotes, 34 HFE wt/wt control subjects) reported that they took thyroid supplements (Table 3). These respective proportions of these participants were 8.5% and 10.9% (p=0.4019). Mean values of age, TSH, free T4, proportions of participants with elevated or subnormal TSH or free T4, or those who had hypothyroidism or hyperthyroidism as defined herein did not differ significantly between C282Y homozygotes and control subjects (Table 3). Mean SF was significantly higher in C282Y homozygotes than in control subjects (Table 3). Two control participants who reported taking thyroid supplements met the present definition for having hyperthyroidism, suggesting that their elevated T4 and subnormal TSH values were due to thyroid supplements (Table 2).
A man in his 70s with HFE C282Y homozygosity and SF 1462ng/mL reported taking propylthiouracil; he met the present definition for having hyperthyroidism. A woman in her 30s with HFE wt/wt and SF 22ng/mL reported that she took methimazole; her values of TSH and free T4 were within the respective reference limits. The proportions of C282Y homozygotes and control subjects who reported taking medications to suppress thyroid function did not differ significantly (p=0.5883).
The correlation of log10 SF with log10 free T4 in all 437 study participants who reported taking neither thyroid supplements nor medications to suppress thyroid function was significant (Pearson correlation coefficient 0.1148; p=0.0166) (data not shown). To determine if the relationship of log10 SF with log10 free T4 were significant in both C282Y homozygotes and in control subjects, we computed correlations in each group of subjects. In 160 C282Y homozygotes, the correlation was positive but not significant (Pearson correlation coefficient 0.0947; p=0.2368) (Fig. 1). In 277 control subjects, the correlation was positive and significant (Pearson correlation coefficient 0.1183; p=0.0492) (Fig. 1).
We used age, sex, Field Center, HFE genotype, log10 SF, and log10 TSH as independent variables, and log10 free T4 as the dependent variable. Values of p for these analyses were as follows: Field Center, 0.4687; age, 0.0142; gender, 0.3238; HFE genotype, 0.9689; log10 SF, 0.2102; and log10 TSH, <0.0001. These analyses reveal significant direct (positive) relationships of age and log10 TSH with log10 free T4.
We used age, sex, Field Center, HFE genotype, log10 TSH, and log10 free T4 as independent variables, and log10 SF as the dependent variable. Values of p for these analyses were as follows: Field Center, 0.8192; age, 0.0006; gender, <0.0001; HFE genotype, <0.0001; log10 TSH, 0.1571; and log10 free T4, 0.2102. These analyses reveal significant direct (positive) relationships of age, male gender, and HFE C282Y homozygosity with SF.
The prevalence of hypothyroidism or hyperthyroidism as defined in the present HEIRS substudy did not differ significantly between white HFE C282Y homozygotes and white HFE wt/wt control subjects, and is similar to that reported in other large population studies in which screening of whites for thyroid disorders was performed (16–20). Multivariate analyses demonstrated that there is a significant inverse relationship of age with log10 free T4 levels, consistent with other population studies of thyroid-related measures (16–20).
In a large population hemochromatosis screening program in Norway, 12.5% of women aged 20–49 years with C282Y homozygosity reported having hypothyroidism, whereas only 3.0% of the control participants reported having hypothyroidism (11). Thyroid-related laboratory measures were not quantified in participants in this study (11). By design, only study participants with elevated TS at initial screening underwent HFE genotyping, and therefore C282Y homozygotes without elevated TS would have been excluded from the Norway screening study (11). This difference of hypothyroidism reports between C282Y homozygotes and control subjects in the Norway screening study could be due to misunderstanding of the screening questionnaire by study participants, the higher proportion of thyroid peroxidase antibody positivity in younger than older women observed in other studies, or a consequence of multiple comparisons in any large study (11,13,16,20,21). The HEIRS Study identified all C282Y homozygotes at initial screening, and afterward obtained reports of thyroid supplement use and measurements of TSH and free T4 in selected participants. Therefore, it is not possible to compare the results of the present HEIRS substudy directly with the hypothyroidism reports in the Norway screening study.
There is evidence that iron deposits in the anterior pituitary gland or thyroid gland do not contribute to the pathogenesis of hypothyroidism or hyperthyroidism in most persons with hemochromatosis. In persons with severe iron overload due to hemochromatosis, iron deposits were visualized in a minority of thyrotrophs; the deposits were much less prominent than those in gonadotrophs (22,23). There are few well-documented hemochromatosis patients who had hypothyroidism due to impaired thyrotroph function; most of them also had hypogonadotrophic hypogonadism (24–29). In rare cases, secondary hypothyroidism resolves after therapeutic phlebotomy to achieve iron depletion (29). Iron deposits were detected in the thyroid gland in a majority of persons diagnosed to have hemochromatosis in medical care or at autopsy (23,30,31). Nonetheless, only 4.1% of 391 patients with hemochromatosis diagnosed in medical care had primary hypothyroidism and only 0.3% had hyperthyroidism (5–9). It has been suggested but remains unproven that iron deposits in the thyroid glands of persons with hemochromatosis cause Hashimoto thyroiditis or Graves disease to develop (6,23).
In persons with hemochromatosis, there is a significant positive correlation of SF with hepatic iron content and iron removed by phlebotomy to achieve iron depletion (32,33). Mean SF in the present HFE C282Y homozygotes was significantly greater than that of HFE wt/wt control subjects, but our prevalence estimates for hypothyroidism and hyperthyroidism did not differ significantly between these groups. The proportion of C282Y homozygotes who reported taking thyroid supplements did not differ significantly from that of control subjects, although two control subjects who reported that they took thyroid supplements met present criteria for having hyperthyroidism. Mean SF in participants with hyperthyroidism was significantly greater than mean SF in participants with hypothyroidism, but both means were within the reference range for SF. None of 22 C282Y homozygotes with severe hyperferritinemia had hypothyroidism; the only C282Y homozygote with hyperthyroidism did not report taking thyroid supplements. Although direct measures of hepatic iron content or quantitative phlebotomy in these substudy participants were not available, our SF observations agree with previous evidence that hypothyroidism or hyperthyroidism in persons with hemochromatosis is usually unrelated to the severity of iron overload (23).
We observed a significant positive correlation of log10 SF with log10 free T4 in control subjects, and a nonsignificant positive correlation of these variables in C282Y homozygotes. The mean SF in the three participants with hypothyroidism was significantly lower than the mean SF in the seven participants with hyperthyroidism. Nonetheless, multivariate analysis of observations from all subjects who reported taking neither thyroid supplements nor medications to suppress thyroid function did not reveal that log10 free T4 was a significant, independent determinant of log10 SF. Hashimoto reported that SF was significantly lower in patients with primary hypothyroidism than in those who were euthyroid, and that SF increased significantly in patients with primary hypothyroidism after normalization of thyroid function (34). In patients with hyperthyroidism due to Graves disease, SF was significantly higher than in patients with primary hypothyroidism (34). In rats made hypothyroid with propylthiouracil or hyperthyroid with L-T4, the liver ferritin synthesis rate was reduced by 36% in hypothyroid rats, and elevated by 38% in hyperthyroid rats; a similar trend was observed in liver ferritin concentration (35). Thus, the present results confirm and extend previous reports that SF is positively correlated with free T4 in control subjects without HFE C282Y homozygosity. The present observations also suggest that increased absorption of iron characteristic of many persons with C282Y homozygosity has a greater influence on SF than free T4, and that increasing quantities of storage iron in C282Y homozygotes are not typically associated with decreasing levels of free T4.
It is unlikely that a non-HFE determinant of either TSH or free T4 on chromosome 6 could explain the occurrence of hypothyroidism or hyperthyroidism in most patients with hemochromatosis. HFE C282Y occurs in linkage disequilibrium with class I alleles at human leukocyte antigen (HLA)-A and -B loci and with the hemochromatosis ancestral haplotype HLA-A*03,B*07 on chromosome 6p (1,36). In Western European whites, Hashimoto thyroiditis and Graves' disease have been associated with HLA class II loci (37–39), but there are few reports of these disorders in whites with hemochromatosis phenotypes or C282Y homozygosity (6,40,41). This suggests that linkage disequilibrium of HFE C282Y with HLA class II loci is not great, that the ancestral hemochromatosis haplotype does not include susceptibility alleles for common autoimmune thyroid disorders, or that there are other susceptibility factors for Hashimoto thyroiditis and Graves' disease. Thyroid-related abnormalities are not typical of chromosome 6p deletion syndromes that involve the region 6p24-pter (42). The CGA gene that encodes the alpha chain subunit common to TSH, gonadotrophin, luteinizing hormone, and follicle-stimulating hormone occurs on the long arm of chromosome 6 (6q12–q21) (43–45).
The HEIRS Study was not designed to determine the clinical severity of or need for treatment of thyroid dysfunction. Our review of medications reported by participants suggests that some were previously diagnosed and treated for thyroid dysfunction, causing their TSH and free T4 values to return to normal before they enrolled in this post-initial screening evaluation. Obtaining self-reports or family history of thyroid gland disease, examining the thyroid gland and related attributes, measuring thyroid peroxidase and thyroglobulin antibodies, and performing thyroid ultrasonography and biopsy were beyond the scope of the HEIRS Study. Long-term follow-up of participants that would have permitted estimation of incidence rates of thyroid conditions was not possible.
In routine medical care settings, some patients who report having fatigue, weakness, or malaise need evaluation for thyroid-related disease, iron overload, or both (8,23,46). In hemochromatosis and iron overload screening programs, however, the prevalence of these symptoms in participants with hemochromatosis phenotypes and C282Y homozygosity was similar to that in age- and sex-matched control subjects (11,47). In this HEIRS substudy, the prevalence of abnormal values of free T4, TSH, or both is similar in white persons with or without C282Y homozygosity. Based on these observations, we conclude that there is no rationale for routine measurement of TSH or free T4 levels in hemochromatosis or iron overload screening programs.
The HEIRS Study was initiated and funded by the National Heart, Lung, and Blood Institute, in conjunction with the National Human Genome Research Institute. The study is supported by contracts N01-HC05185 (University of Minnesota); N01-HC05186, N01-CM-07003–74, and Minority CCOP (Howard University); N01-HC05188 (University of Alabama at Birmingham); N01-C05189 (Kaiser Permanente Center for Health Research); N01-HC05190 (University of California, Irvine); N01-HC05191 (London Health Sciences Centre); and N01-C05192 (Wake Forest University). Additional support was provided by the University of Alabama at Birmingham General Clinical Research Center (GCRC) grant M01-RR00032, Howard University GCRC grant M01-RR10284, and the University of California, Irvine UCSD/UCI Satellite GCRC grant M01-RR00827, sponsored by the National Center for Research Resources, National Institutes of Health; Howard University Research Scientist Award UH1-HL03679-05 from the National Heart, Lung and Blood Institute and the Office of Research on Minority Health (VRG); and Southern Iron Disorders Center (JCB, RTA).
Participating HEIRS Study investigators and institutions
Birmingham, AL—University of Alabama at Birmingham:
Dr. Ronald T. Acton (Principal Investigator), Dr. James C. Barton (Co-Principal Investigator), Ms. Deborah Dixon, Dr. Susan Ferguson, Dr. Richard Jones, Dr. Jerry McKnight, Dr. Charles A. Rivers, Dr. Diane Tucker, and Ms. Janice C. Ware.
Irvine, CA—University of California, Irvine:
Dr. Christine E. McLaren (Principal Investigator), Dr. Gordon D. McLaren (Co-Principal Investigator), Dr. Hoda Anton-Culver, Ms. Jo Ann A. Baca, Dr. Thomas C. Bent, Dr. Lance C. Brunner, Dr. Michael M. Dao, Dr. Korey S. Jorgensen, Dr. Julie Kuniyoshi, Dr. Huan D. Le, Dr. Miles K. Masatsugu, Dr. Frank L. Meyskens, Dr. David Morohashi, Dr. Huan P. Nguyen, Dr. Sophocles N. Panagon, Dr. Chi Phung, Dr. Virgil Raymundo, Dr. Thomas Ton, Prof. Ann P. Walker, Dr. Lari B. Wenzel, and Dr. Argyrios Ziogas.
London, Ontario, Canada—London Health Sciences Center:
Dr. Paul C. Adams (Principal Investigator), Ms. Erin Bloch, Dr. Subrata Chakrabarti, Ms. Arlene Fleischhauer, Ms. Helen Harrison, Ms. Kelly Jia, Ms. Sheila Larson, Dr. Edward Lin, Ms. Melissa Lopez, Ms. Lien Nguyen, Ms. Corry Pepper, Dr. Tara Power, Dr. Mark Speechley, Dr. Donald Sun, and Ms. Diane Woelfle.
Portland, OR, and Honolulu, HI—Kaiser Permanente Center for Health Research, Northwest and Hawaii, and Oregon Health and Science University:
Dr. Emily L. Harris (Principal Investigator), Dr. Mikel Aickin, Dr. Elaine Baker, Ms. Marjorie Erwin, Ms. Joan Holup, Ms. Carol Lloyd, Dr. Nancy Press, Dr. Richard D. Press, Dr. Jacob Reiss, Dr. Cheryl Ritenbaugh, Ms. Aileen Uchida, Dr. Thomas Vogt, and Dr. Dwight Yim.
Washington, D.C.—Howard University:
Dr. Victor R. Gordeuk (Principal Investigator), Dr. Fitzroy W. Dawkins (Co-Principal Investigator), Ms. Margaret Fadojutimi-Akinsiku, Dr. Oswaldo Castro, Dr. Debra White-Coleman, Dr. Melvin Gerald, Ms. Barbara W Harrison, Dr. Ometha Lewis-Jack, Dr. Robert F. Murray, Dr. Shelley McDonald-Pinkett, Ms. Angela Rock, Dr. Juan Romagoza, and Dr. Robert Williams.
Minneapolis, MN—University of Minnesota and University of Minnesota Medical Center Fairview:
Dr. John H. Eckfeldt (Principal Investigator and Steering Committee Chair), Ms. Susie DelRio-LaFreniere, Ms. Catherine Leiendecker-Foster, Dr. Ronald C. McGlennen, Mr. Greg Rynders, Dr. Michael Y. Tsai, and Dr. Xinjing Wang.
Winston-Salem, NC—Wake Forest University:
Dr. David M. Reboussin (Principal Investigator), Dr. Beverly M. Snively (Co-Principal Investigator), Dr. Roger Anderson, Ms. Elease Bostic, Ms. Brenda L. Craven, Ms. Shellie Ellis, Dr. Curt Furberg, Mr. Jason Griffin, Dr. Mark Hall, Mr. Darrin Harris, Ms. Leora Henkin, Dr. Sharon Jackson, Dr. Tamison Jewett, Mr. Mark D. King, Mr. Kurt Lohman, Ms. Laura Lovato, Dr. Joe Michaleckyj, Ms. Shana Palla, Ms. Tina Parks, Ms. Leah Passmore, Dr. Pradyumna D. Phatak, Dr. Stephen Rich, Ms. Andrea Ruggiero, Dr. Mara Vitolins, Mr. Gary Wolgast, and Mr. Daniel Zaccaro.
NHLBI PROJECT OFFICE
Bethesda, MD—Ms. Phyliss Sholinsky (Project Officer), Dr. Ebony Bookman, Dr. Henry Chang, Dr. Richard Fabsitz, Dr. Cashell Jaquish, Dr. Teri Manolio, and Ms. Lisa O'Neill.
NHGRI PROJECT OFFICE
Bethesda, MD—Dr. Elizabeth Thomson.
Dr. Jean MacCluer, Southwest Foundation for Biomedical Research, also contributed to the design of this study.