Osteoporosis International

, Volume 16, Issue 12, pp 1809–1814

Bone mineral density in men with genetic hemochromatosis and HFE gene mutation

Authors

    • Rheumatology DepartmentUniversity Hospital
  • Y. Deugnier
    • Hepatology DepartmentUniversity Hospital
  • J. F. Boisdet
    • Rheumatology DepartmentUniversity Hospital
  • Y. Rolland
    • Radiology DepartmentUniversity Hospital
  • A. Perdriger
    • Rheumatology DepartmentUniversity Hospital
  • Y. Pawlotsky
    • Rheumatology DepartmentUniversity Hospital
  • G. Chalès
    • Rheumatology DepartmentUniversity Hospital
Original Article

DOI: 10.1007/s00198-005-1934-0

Cite this article as:
Guggenbuhl, P., Deugnier, Y., Boisdet, J.F. et al. Osteoporos Int (2005) 16: 1809. doi:10.1007/s00198-005-1934-0

Abstract

Genetic hemochromatosis (GH) is an iron overload disorder mainly due to the C282Y mutation of the HFE gene. The possibility of bone involvement was only recently recognized. The aims of this study were to assess bone mineral density (BMD) and bone remodeling in men with GH, and to examine the influence of iron overload. Thirty-eight men (mean age 47.2±9.4 years) with well-defined HFE-related GH were studied. They had an important iron overload with liver iron concentration to age ratio >2.5, no previous venesection therapy and were C282Y homozygotes ( n =37) or compound C282Y/H63D heterozygote ( n =1). BMD measured by DXA was 0.925±0.15 g/cm2 at the lumbar spine (LS) and 0.778±0.13 g/cm2 at the femoral neck (FN). Osteopenia (T-score <−1 SD) was observed in 78.9% of patients and osteoporosis (T-score <−2.5 SD) in 34.2%. Vitamin D levels were normal, and no 1–84 parathyroid hormone dysfunction was found. Hypogonadism was found in only 13.2% of patients. Patients with hypogonadism had lower LS BMD than eugonadal patients (0.788±0.16 and 0.954±0.14 g/cm2). Bone remodeling and parathyroid hormone levels were lower in patients with cirrhosis, but BMD values were similar to those in patients without cirrhosis. FN BMD appeared to fall with rising hepatic iron concentrations ( r =−0.399). We conclude that there is significant bone loss in HFE-related hemochromatosis that cannot solely be explained by hypogonadism or cirrhosis. Further investigations are needed to determine the role of iron overload itself.

Keywords

Bone mineral densityHemochromatosisIron overloadOsteopeniaOsteoporosis

Introduction

Genetic hemochromatosis (GH) is an autosomal recessive disease mainly due to the C282Y mutation of the HFE gene [1, 2]. This mutation leads to inappropriate iron absorption through mechanisms involving impaired hepcidin secretion [3]. In Caucasian populations, 10% of subjects are C282Y heterozygous and 0.5% homozygous. C282Y homozygosity is found in more than 90% of patients with phenotypic hemochromatosis [4]. However, the penetrance of C282Y homozygosity is much lower than that expected from previous HLA-based studies. It ranges from 1 to 75% according to gender and disease definition, with an accepted rate of 50%. This suggests that other genes could contribute to the hemochromatosis phenotype [5]. C282Y heterozygosity does not result in iron overload except when associated with a minor HFE mutation such as H63D mutation. The H63D/C282Y compound may develop mild iron overload [5].

The life-threatening complications and articular changes associated with GH are well documented, whereas bone involvement was only recently recognized [6, 7, 8]. Osteopenia and osteoporosis occurred in 28 to 50% of patients in recent series [9, 10, 11], and the risk of vertebral fracture can reach 20% [10]. Both articular and bone changes may affect the functional prognosis and quality of life, whereas life-threatening complications can now be prevented by early diagnosis and iron depletion.

The controversy surrounding the existence of bone changes in GH was due to several factors. First, methods used to measure bone mass varied including histomorphometry, single-photon absorptiometry and double-photon absorptiometry. Second, measurements were made at different sites (radius, hip or lumbar spine). Third, osteoporosis and osteopenia were not clearly defined. Fourth, the GH definition did not rely upon genetic testing, iron excess was not determined accurately, i.e., using direct methods, and most series involved a few patients with a large spectrum of clinical stages. Finally, other risk factors for osteoporosis, particularly in women, were not taken into account.

The aim of this study was to assess bone mineral density (BMD) in men with well-defined HFE-related hemochromatosis due to homozygous HFE mutations. Women were excluded in order to avoid the influence of menopause.

Patients and methods

Patients

From 1995 to 1998, all patients diagnosed at the Hepatology Unit of our hospital by one of us (YD) as suffering from GH were referred to our Rheumatology Unit when they fulfilled the following criteria: (1) male sex, (2) an important iron overload with liver iron concentration to age ratio >2.5 in the absence of any known cause of iron overload, and (3) no previous venesection therapy. Eight additional untreated patients were included during the period 1999–2001 on the same basis, but retrospectively. From the 38 patients referred, all were validated retrospectively as C282Y homozygotes ( n =37) or compound C282Y/H63D heterozygote ( n =1) [12].

Laboratory studies

Hepatic iron concentration (HIC) was determined either biochemically on liver biopsy [13] and/or by magnetic resonance imaging [14, 15]. Most of our patients (27/38) had liver biopsy since, at the time of study initiation, genetic testing was not available, and the study of Guyader et al. [16] to determine the non invasive predictive factors of the absence of severe fibrosis in GH patients was not published. Eleven patients had cirrhosis, of whom nine were diagnosed by liver biopsy and two on the basis of clinical evidence. The nine remaining cases fulfilled the Guyader’s criteria for no severe fibrosis (no hepatomegaly, normal serum AST and serum ferritin <1,000) [16].

Standard automated analyzers were used to determine serum iron levels and transferrin saturation. Radioimmunoassay was used for serum ferritin determination (Chiron Diagnostics, Cergy-Pontoise, France). The C282Y mutation was screened for by restriction-enzyme digestion of polymerase chain reaction-amplified genomic DNA [2, 4]. Immunoradiometric assays were used to measure 1–84 intact parathyroid hormone (Nichols Institute Diagnostics, Paris) and 25-hydroxycholecalciferol (vitamin D) (Sorin Diagnostics France, Anthony, France). Markers of bone remodeling were measured as follows: osteocalcin by radioimmunoassay (Elsa-Osteo Cisbio), bone-specific alkaline phosphatase (BSAP) by enzyme immunoassay (Alkphase B Metrer Bio systems) and urinary C-telopeptides by enzyme immunoassay (CTX) (Cisbio). Testosterone was measured with a radioimmunoassay (Biomerieux). Patients were considered hypogonadal when their total serum testosterone level was below the lower limit of the normal range (3,500 pg/ml).

BMD measurement

Bone mineral density was measured at the lumbar spine and femoral neck by dual-energy X-ray absorptiometry with a Hologic QDR 1000 Plus device (Waltham, Mass.) in most patients, and a QDR 4500 device (Waltham, Mass.) in six patients; cross-calibration between the two devices showed no significant differences in the measurements. T-scores were calculated from the manufacturers’ references. Osteopenia and osteoporosis were defined according to WHO criteria [17].

Statistical analysis

Statistical analysis was performed using SPSS 11.5.1 software. The chi-square test was used to compare qualitative data, and the Mann-Whitney test was used to compare group means. Correlations were sought by using bivariate analysis (simple and multiple linear regression). Differences and correlations were considered statistically significant when the P value was below 0.05.

Results

The characteristics of the 38 patients are summarized in Table 1. Most subjects were middle-aged males (47.2±9.4 years) with important iron excess as indicated by a mean HIC of 339±147.9 µmol/g and with normal body mass index (BMI) (25.6±4.4 kg/m2). Eleven patients (28.9%) had HBV and HCV negative cirrhosis. Six patients (15.8%) were excessive alcohol drinkers of whom 2/6 (33.3%) had a cirrhosis versus 9/32 (28.1%) in other GH patients. There was no significant BMD difference between these two groups. Twenty-five patients (65.8%) complained about joint symptoms, and 18 (47.4%) had joint radiograph signs of hemochromatic rheumatism.
Table 1

Characteristics of the 38 men with genetic HFE-related hemochromatosis

Datum (units)-(normal range)

Mean

±SD

Age (years)

47.2

9.4

Serum ferritin (ng/ml)-(25–375)

1,819

1,333

Transferrin saturation (%)-(20–40)

81.4

16.7

Hepatic iron concentration (µmol/g)-(<36)

339.7

147.9

Albumin (43–51 g/l)

44.2

3.8

Total alkaline phosphatase (37–111 UI/l)

85.2

35.9

Serum total testosterone (pg/ml)-(3,500–11,500)

6,021

2,723

Parathyroid hormone (pg/ml)-(10–55)

29.8

10.4

25-hydroxycholecalciferol (ng/l)-(15–45)

22.9

10.1

Lumbar spine BMD (g/cm2)

0.925

0.15

Lumbar spine BMD (T-score)

−1.51

1.35

Femoral neck BMD (g/cm2)

0.778

0.13

Femoral neck BMD (T-score)

−1.83

1.19

Urinary CTX (ng/mmol creatinine)-(140–380)

174.15

114.1

Osteocalcin (ng/ml)-(5–35)

21.5

6.4

Bone-specific alkaline phosphatase (U/l)-(12–23)

22.7

10.4

Mean BMD was 0.925±0.15 g/cm2 at the lumbar spine (LS: L1-L4) (mean T-score = 1.5±1.35) and 0.778±0.13 g/cm2 at the femoral neck (FN) (mean T-score = 1.83±1.19) (Table 1). LS osteopenia (T-score <−1 SD) was observed in 22/38 (57.9%) patients and FN osteopenia in 27/36 (75%). Osteopenia was found at LS or FN in a total of 30/38 patients (78.9%). Osteoporosis (T-score<−2.5 SD) was found at LS in 10/38 patients (26.3%) and at FN in 9/36 (25%). Osteoporosis was found at LS or FN in a total of 13/38 patients (34.2%). Eight patients (21.1%) had a T-score <−3 SD at one or both sites: 6/38 (15.8%) at LS and 6/36 (16.6%) at FN.

Patients with osteoporosis (T-score <−2.5 SD at one site) tended to be older (50.8±7.9 versus 45.3±9.6 years, P =0.85) and to have slightly higher osteocalcin levels (24.3±6.2 versus 20±6.1 ng/ml, P =0.07). No other differences with non-osteoporotic patients were found, particularly regarding testosterone levels, hypogonadism, cirrhosis and iron parameters (Table 2).
Table 2

Characteristics of the 38 men with genetic HFE-related hemochromatosis according to the presence of osteoporosis

Datum (units)-(normal range)

Mean±SD

P

Osteoporosis

No osteoporosis

( n =13)

( n =25)

Age (years)

50.8±7.9

45.3±9.6

NS

Serum ferritin (ng/ml)-(25–375)

1,768±1,410

1,846±1,320

NS

Transferrin saturation (%)-(20–40)

81±15.8

81.6±17.5

NS

Hepatic iron concentration (μmol/g)-(<36)

366.2±193.9

328.9±128.5

NS

Serum total testosterone (pg/ml)-(3,500–11,500)

5,360.7±2,517.8

6,364.9±2,811.6

NS

Parathyroid hormone (pg/ml)-(10–55)

28.3±7.4

30.6±11.8

NS

25-hydroxycholecalciferol (ng/l)-(15–45)

23.3±8.8

22.7±10.9

NS

Lumbar spine BMD (g/cm2)

0.766±0.11

1.007±0.87

<0.0001

Lumbar spine BMD (T-score)

−2.95±0.99

−0.764±0.79

<0.0001

Femoral neck BMD (g/cm2)

0.660±0.11

0.844±0.08

<0.0001

Femoral neck BMD (T-score)

−2.894±1.04

−1.224±0.77

<0.0001

Urinary CTX (ng/mmol creatinine) (140–380)

201.6±124.7

161.3±110.4

NS

Osteocalcin (ng/ml)-(5–35)

24.3±6.2

20±6.1

0.07

Bone-specific alkaline phosphatase (U/l)-(12–23)

23.6±9.9

22.3±10.9

NS

NS: non significant

Serum calcium and phosphorus, parathyroid hormone, 25OH vitamin D and bone remodeling markers (urinary CTX, serum osteocalcin and bone-specific alkaline phosphatase) were normal. There was no evidence of hyperparathyroidism or osteomalacia.

Five patients (13.16%) were hypogonadal. They had lower BMD values at LS (0.788±0.16 versus 0.954±0.14 g/cm2, P =0.04) and lower T-scores (−2.75±1.5 versus −1.32±1.25, P =0.04); the difference was not significant at FN (0.713±0.14 versus 0.788±0.13 g/cm2, P =0.37). They also had higher ferritin levels (2,702±1,285 versus 1,680±1,306 ng/ml, P =0.14) and higher BSAP levels (35.6±9.6 versus 20.9±9.1 U/l, P =0.007). HIC tended to be higher (417.6±234.9 versus 324.7±126.4 μmoles/g, P =0.6) (Table 3). Three (60%) of the five patients with hypogonadism were osteoporotic, compared to 30.3% of the non-hypogonadal patients ( P =0.32). In other words, 76.9% of osteoporotic patients were not hypogonadal.
Table 3

Characteristics of the 38 men with genetic HFE-related hemochromatosis according to the presence of hypogonadism

Datum (units)-(normal range)

Mean±SD

P

Hypogonadism

No hypogonadism

( n =5)

( n =33)

Age (years)

48±14.9

47.1±8.6

NS

Serum ferritin (ng/ml)-(25–375)

2,702±1,285

1,680±1,306

0.14

Transferrin saturation (%)-(20–40)

81.7±9.2

81.4±17.8

NS

Hepatic iron concentration (μmol/g)-(<36)

417.6±234.9

324.7±126.4

NS

Serum total testosterone (pg/ml)-(3,500–11,500)

2,322.4±1,283.2

6,581.8±2,430.6

<0.0001

Parathyroid hormone (pg/ml)-(10–55)

29.9±8.1

29.7±10.8

NS

25-hydroxycholecalciferol (ng/l)-(15–45)

24.2±12.5

22.7±9.9

NS

Lumbar spine BMD (g/cm2)

0.788±0.16

0.945±0.14

0.04

Lumbar spine BMD (T-score)

−2.75±1.48

−1.33±1.25

0.04

Femoral neck BMD (g/cm2)

0.713±0.14

0.788±0.13

NS

Femoral neck BMD (T-score)

−2.42±1.25

−1.73±1.17

NS

Urinary CTX (ng/mmol creatinine) (140–380)

237.5±75.7

162.1±117.5

NS

Osteocalcin (ng/ml)-(5–35)

25.7±8.6

20.9±6

NS

Bone-specific alkaline phosphatase (U/l)-(12–23)

35.6±9.6

20.9±9.1

0.007

NS: non significant

Eleven patients (28.9%) had cirrhosis. There were no significant differences in BMD between cirrhotic and non-cirrhotic patients; LS BMD was 0.941±0.16 versus 0.918±0.15 g/cm2 ( P =0.8), and FN BMD was 0.789±0.13 versus 0.773±0.13 ( P =0.8), respectively. Cirrhotic patients were older (51.2±5.6 versus 45.6±10.2 years, P =0.08) and had higher ferritin levels (2,784±1,472 versus 1,461±1,104 ng/ml, P =0.014), lower urinary CTX levels (88.2±60.5 ng/mmol versus 214.6±111.8 ng/mmol creatinine, P =0.002), lower osteocalcin levels (18.5±5.2 versus 22.8±6.5 ng/ml, P =0.13), and lower PTH levels (24.3±10.6 versus 32±9.6, P =0.05); 25 OH vitamin D levels were normal and similar to those in non-cirrhotic patients. The hepatic iron concentration was slightly, but not significantly higher (387.6±152.9 versus 320.1±144.8 μmol/g, P =0.17) (Table 4). Osteoporosis was present in 27.2% of patients with cirrhosis compared to 37% of non-cirrhotic patients ( P =0.71).
Table 4

Characteristics of the 38 men with genetic HFE-related hemochromatosis according to the presence of cirrhosis

Datum (units)-(nomal range)

Mean±SD

P

Cirrhosis

No cirrhosis

( n =11)

( n =27)

Age (years)

51.2±5.6

45.6±10.2

0.08

Serum ferritin (ng/ml)-(25–375)

2,784±1,472

1,461±1,104

0.014

Transferrin saturation (%)-(20–40)

80.3±19.6

81.8±16.1

NS

Hepatic iron concentration (μmol/g)-(<36)

387.6±152.9

320.1±144.8

0.17

Serum total testosterone (pg/ml)-(3,500–11,500)

6,762.8±2,781.5

5,719.3±2,692.9

NS

Parathyroid hormone (pg/ml)-(10–55)

24.3±10.6

32±9.6

0.05

25-hydroxycholecalciferol (ng/l)-(15–45)

24.2±11.7

22.4±9.5

NS

Lumbar spine BMD (g/cm2)

0.941±0.16

0.918±0.15

NS

Lumbar spine BMD (T-score)

−1.37±1.45

−1.57±1.33

NS

Femoral neck BMD (g/cm2)

0.790±0.13

0.773±0.13

NS

Femoral neck BMD (T-score)

−1.72±1.2

−1.87±1.2

NS

Urinary CTX (ng/mmol creatinine) (140–380)

88.2±60.5

214.6±111.8

0.002

Osteocalcin (ng/ml)-(5–35)

18.5±5.2

22.8±6.5

0.13

Bone-specific alkaline phosphatase (U/l)-(12–23)

19.9±10

23.9±10.6

NS

NS: non significant

We found a negative correlation between the hepatic iron concentration and BMD at the FN ( r =−0.399, P =0.029), but not at the LS ( r =−0.113, P =0.546). This relationship was not affected when BMI was introduced into the model (Table 5); however, when age was introduced, only BMI remained significant, while HIC was not far from significance (Table 6).
Table 5

Linear regression analysis of femoral neck bone mineral density (BMD), body mass index (BMI) and hepatic iron concentration (HIC) in 38 men with genetic HFE-related hemochromatosis

Femoral neck BMD

β

t

P

BMI

0.397

2.432

0.022

HIC

−0.399

−2.448

0.022

Table 6

Linear regression analysis of femoral neck bone mineral density (BMD), body mass index (BMI), hepatic iron concentration (HIC) and age in 38 men with genetic HFE-related hemochromatosis

Femoral neck BMD

β

t

P

BMI

0.439

2.596

0.0159

Age

−0.185

−0.967

0.343

HIC

−0.310

−1.649

0.112

Discussion

This is the first study of bone metabolic status in a homogeneous cohort of patients with well-defined and untreated HFE-related genetic hemochromatosis. Although GH is known to be associated with osteopenia [6, 7, 9, 10, 11, 18], little is known of the prevalence and pathogenesis of this association. Previous studies have given inconsistent results, largely owing to the use of different definitions of GH and osteopenia.

BMD was evaluated with the DXA method (at two sites in most cases), which is the gold standard for non-invasive assessment of osteopenia. Based on WHO criteria [17], osteopenia was found in 78.9% of the patients and osteoporosis in 34.2%. In the absence of local references for males, T-scores were calculated from the manufacturers’ references, which could have slightly overestimated the number of osteoporosis and osteopenia cases in our series. However, these results are consistent with those of a recent study, respectively 71 and 29% among male GH patients [9], based on the same DXA measurement criteria (T-scores <−1 and <–2.5 SD). In our patients, the decrease in BMD was more pronounced at the femoral neck than at the lumbar spine, suggesting the greater impact of GH on cortical bone, as previously described [18]. We found no significant differences between osteoporotic (T <−2.5 SD) and non-osteoporotic patients (T >−2.5 SD) as regards biological data, cirrhosis or hypogonadism. In contrast, osteoporotic patients were slightly older and had higher osteocalcin levels.

It has been suggested that 25-hydroxyvitamin D deficiency [10] might explain osteoporosis in GH, but we found no difference in mean values between osteoporotic and non-osteoporotic patients; levels were generally in the lower part of the normal range (median 24.3 ng/ml, range 5 to 40 ng/ml). Diamond [10] reported low 25-hydroxyvitamin D concentrations in hypogonadal patients with GH, but found no osteomalacia at bone biopsy. We found no difference between hypogonadal and eugonadal patients as regards 25-hydroxyvitamin D levels. Cirrhosis did not appear to influence 25-hydroxyvitamin D levels.

Hyperparathyroidism has been found in patients with GH [19], but none of our patients had hyperparathyroidism or hypercalcemia. A recent study showed no increase in intact 1–84 parathyroid hormone in GH, while 44–68 PTH fragment levels were associated with joint status and ferritin levels. This could result from increased parathyroid hormone cleavage in parathyroid glands and liver, with release of N- and C-terminal fragments, which are more frequent in bone [8]. Parathyroid hormone levels were lower (but normal) in our cirrhotic patients than in their non-cirrhotic counterparts; this might have contributed to slower bone turnover, or might suggest a lack of response to low bone turnover.

Hypogonadism is a major concern in GH osteoporosis. Previous studies showed that 50 to 100% of patients with GH osteoporosis had hypogonadism [6, 10]. Diamond [10] found that all male patients with hypogonadism had osteoporosis, with increased bone resorption parameters and decreased bone formation parameters on bone biopsy. It seems that the increase in bone resorption may be linked to hypogonadism itself and not necessarily to iron overload. Indeed, hypogonadism in males, as in females, is characterized by an increase in bone activation frequency and resorption, and leads to deficits in both cortical and trabecular bone, loss of trabecular connectivity and trabecular perforation; in contrast, glucocorticoid-induced osteoporosis leads to trabecular thinning due to reduced bone formation [20]. In pigs, experimental iron overload led to a marked decrease in bone formation without affecting bone resorption; iron deposits are found in osteoblasts and osteoclasts and in the bone matrix along trabecular surfaces, the cement line and the osteoid-mineralized bone interface. Mineralization is not impaired, and levels of 25-hydroxyvitamin D and 1–25 hydroxyvitamin D are normal [21]. In rats, iron overload reduces osteoblast activity by promoting lipid peroxidation [22]. Interestingly, most of our patients with osteoporosis or osteopenia were eugonadal, in keeping with a recent report showing no link between free testosterone levels and BMD in males with hemochromatosis; it is noteworthy that higher amounts of removed iron (reflecting higher initial iron overload) were associated with osteoporosis in this latter study [9]. Thus, while hypogonadism clearly worsens osteoporosis, it is unlikely to account entirely for the decreased BMD in GH.

Many confounding factors, such as the severity of liver disease, can interfere with studies of osteoporosis. We found no significant difference in BMD between patients with and without cirrhosis. However, urinary CTX levels were significantly lower in cirrhotic patients, and osteocalcin levels were also subnormal, pointing to a decrease in bone remodeling. Bone remodeling parameters are difficult to interpret in patients with cirrhosis, and collagen-related markers such as CTX are clearly linked to liver fibrosis rather than osteopenia; nevertheless, decreased osteocalcin levels are considered to reflect impaired osteoblastic function, even in patients with liver disease [23]. In one series [10] all the hypogonadal patients had cirrhosis, making it difficult to determine the respective influences of hypogonadism, cirrhosis and iron overload on bone status [24]. In another series [9], cirrhotic patients had lower BMD and, once again, most of those with osteoporosis also had hypogonadism. None of our cirrhotic patients had hypogonadism, possibly because they were studied at diagnosis. The decrease in bone remodeling observed in our patients could have a negative impact on bone status later in the course of hemochromatosis. The impact of cirrhosis on bone is difficult to appreciate because of the presence of numerous cofactors for osteoporosis, such as hypogonadism, in the end-stage disease. Furthermore, it is difficult to separate the effect of cirrhosis itself from the effect of the underlying cause of cirrhosis. Patients with hemochromatosis and cirrhosis have been found to have lower forearm BMD and cortical osteoporosis at bone biopsy than patients with alcoholic cirrhosis and controls [18]. In addition, iron bone deposits are more abundant in patients with hemochromatosis than in patients with post-hepatitic cirrhosis [11].

We found no direct link between BMD and either transferrin saturation or ferritin levels. Cirrhotic and hypogonadal patients had higher ferritin levels, but the difference was only significant in those who also had cirrhosis. The hepatic iron concentration, which is a good index of total body iron overload, was slightly increased in both these subgroups and also in the osteoporotic subgroup, but the difference with other subjects was not significant. However, we did find a negative correlation between the hepatic iron concentration and BMD at the femoral neck, where the mean T-score was lower than at the lumbar spine. This statistical link was not affected by the introduction of BMI into the model. In contrast, when age was introduced into the model, only BMI remained significant (although the hepatic iron concentration remained close to significance). DXA might thus be most useful in patients with high HIC. Iron overload has previously been observed at bone biopsy in patients with hemochromatosis and positive Perls staining of osteoblasts, osteoclasts, the osseous matrix and trabeculae [11, 25], a finding confirmed by animal studies [21, 22]. Finally, histological studies of iron-overloaded bone essentially showed a decrease in bone formation and, sometimes, an increase in bone resorption in patients with associated hypogonadism. Osteomalacia has never been found [7, 11, 18, 26].

To summarize, in a population of 38 men with well-defined genetic HFE-related hemochromatosis, osteopenia was found in 78.9% of cases and osteoporosis in 34.2%. The decrease in BMD was more pronounced at the femoral neck than at the lumbar spine. Few patients had hypogonadism, and none had low 25 OH vitamin D levels or 1–84 parathyroid hormone dysfunction. Bone remodeling was decreased in patients with cirrhosis. The hepatic iron concentration seemed to have a negative impact on BMD. Thus, HFE-related genetic hemochromatosis appears to be associated with significant bone loss. Hypogonadism and cirrhosis cannot alone account for this osteopenia, although both may be contributory factors. Further investigations are needed to understand the impact of iron overload itself on bone status in patients with genetic hemochromatosis.

Copyright information

© International Osteoporosis Foundation and National Osteoporosis Foundation 2005