| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1
Clinical Pathology, PathCentre, Nedlands Western Australia 6009, Australia, and
2
University Department of Medicine, Sir Charles Gairdner Hospital, Nedlands, Western Australia 6009, Australia;
a author
for correspondence: fax 61 8 93463882, e-mail John.Beilby{at}health.wa.gov.au
Serum iron, transferrin, transferrin saturation, and ferritin are used to determine the iron status of an individual. When iron overload is suspected because of abnormal iron indices, the accurate determination of hepatic iron content (HIC) is important for the diagnosis of genetic hemochromatosis (GH). The measurement of HIC and the calculation of the hepatic iron index (HII), defined as HIC/age, enables the discrimination between patients with homozygous GH and those with either the heterozygous form of the disease or secondary causes of hepatic iron deposition. Despite the advent of DNA genotyping for the diagnosis of GH, liver biopsies are sometimes required for histological grading of liver disease and biochemical analysis to quantify the HIC. Genotyping will now identify many homozygotes before the development of clinical symptoms and an increased HII, and may reduce the need for obtaining a liver biopsy in some patients. However, in our experience the number of specimens submitted for HIC analysis has not changed over an 8-month period since the introduction of genotyping for GH.
Two techniques are used for the determination of liver iron content. The more common technique is histological grading of stainable iron, using Prussian blue dye; the other technique is the quantitative chemical determination of iron. The latter requires the liver sample to be homogenized or digested with acid before iron determination by either colorimetry or atomic adsorption spectroscopy. The diagnosis of GH can be established reliably using the HII calculated from the hepatic iron concentration determined by atomic absorption spectroscopy. Although an age-corrected histological HII may be derived from histological staining for iron, it does not reliably distinguish GH from other causes of increased HIC (1). Quantitative HIC is usually determined on fresh tissue submitted to the laboratory at the time of biopsy. Occasionally, only the paraffin-embedded (histological) liver biopsy specimens are available for quantitative iron determination because the analysis was not anticipated at the time of biopsy and only deemed necessary after histological examination. In addition, in some instances the fresh sample submitted might be too small for analysis or was contaminated. Olynyk et al. (2) used a colorimetric assay to determine HIC and reported that HIC in deparaffinized specimens agreed with that in fresh tissue in a series of 44 biopsies. The aim of our study was to extend the findings of Olynyk et al. in a larger series of samples, using graphite furnace atomic absorption spectroscopy.
Liver biopsies were obtained during routine clinical evaluation from 100 patients. The biopsies were taken percutaneously with a Tru-cut needle. A portion of the tissue (~5 mm length) was sent fresh to the laboratory in a clean, dry container and stored at 4 °C until analysis (within 1 week). The other piece was fixed in formaldehyde (40 g/L) and embedded in paraffin as part of routine histological processing. For this study, embedded tissue was dissected from the paraffin block, and the wax removed by washing (2 x 30 min) in xylene. Only ~50% of the embedded tissue was removed from the paraffin block, thus enabling the fixed specimen to be returned to the archives for future reference. The fresh and deparaffinized specimens were dried at 80 °C for 24 h and weighed. The dry weights were 0.31–3.30 mg (median, 0.80 mg) for the fresh tissue and 0.13–1.18 mg (median, 0.50 mg) for the deparaffinized tissue. The dried tissue (fresh and deparaffinized) was digested with a mixture of nitric acid (200 µL) and hydrogen peroxide (400 µL) at 100 °C in a glycol bath for 15–20 min. A stock digest was made by diluting each digested sample to 5 mL with deionized water. The iron concentration was determined by graphite furnace atomic absorption spectroscopy on a Varian SpectrAA300 spectrometer with Zeeman background correction using pyrolytically coated graphite tubes. The stock digest was diluted to bring the iron concentration within the calibration range. Dilutions were performed with deionized water, with nitric acid added to maintain an acid concentration of 40 mL/L. Calibrators were also prepared in 40 mL/L nitric acid; standard additions were not used. Triplicate analyses were performed on each solution. The furnace settings were as follows: for drying, ramp to 85 °C (5 s), ramp to 95 °C (40 s), and ramp to 120 °C (10 s); for ashing, ramp to 800 °C (5 s) and hold at 800 °C (3 s); for atomization, ramp to 2300 °C (1.1 s, read signal), hold at 2300 °C (2 s, read signal), and hold at 2300 °C (2 s, tube clean). The absorbance was determined at 372 nm.
The interassay imprecision (CV) was determined using the National
Institute of Standards and Technology standard reference material
1577b, bovine liver that has a certified iron concentration of
3.29 ± 0.27 mmol/kg dry weight. The mean liver iron concentration
obtained was 3.36 mmol/kg (n = 30), with a CV of 4.2%. The
precision of liver iron determination on embedded tissue was assessed
by cutting a number of different size tissue samples from a single,
relatively large histological sample. The smaller the piece of
deparaffinized liver tissue used for the analysis, the more imprecise
were the results for that piece (Table 1
). When we interpolated from the data in Table 1
, we found a CV
of ~10% for a dry weight of 0.4 mg. This dry weight of
paraffin-embedded tissue is the same as that defined by Olynyk et al.
(2) as the weight below which results agreed poorly with
results on samples of fresh liver specimens.
|
HIC was determined in both fresh and deparaffinized liver specimens
(n = 100). There was a significant linear relationship between HIC
determined in fresh and deparaffinized tissue (Fig. 1
). We observed a positive bias for HIC determined in
deparaffinized tissue when the HIC was >100 mmol/kg; however, only
eight samples fell into this category. When the deparaffinized HIC
result was <100 mmol/kg (n = 92), no significant difference
between HIC determined in deparaffinized and fresh tissue from the same
patient was observed using the Wilcoxon signed-rank test. For 85% of
the HIC measurements on deparaffinized specimens, the result was either
within 20% of the fresh result or both results fell within the
reference interval of 5–24 mmol/kg dry weight (92% for 65 samples
with dry weights 
0.4 mg and 71% for 35 specimens with dry weights
<0.4 mg). We calculated the HII for a group of 51 patients with HIC of
2–306 mmol/kg dry wt. HII values ranged from 0.1 to 5.6. Determination
of the HII from deparaffinized specimens correctly classified 96% of
patients (49 of 51) as GH or non-GH, using an HII value >1.9 as
diagnostic of GH. For specimens weighing 0.4 mg, 98% (44 of 45) of
patients were classified correctly, with one patient having borderline
results (deparaffinized HII = 1.9; fresh tissue HII = 1.1).
|
In summary, determination of the HII of the deparaffinized specimens
provided a high diagnostic accuracy when the sample weight was 0.4
mg. In contrast to other methods (2)(3), in most
cases it is not necessary to use the entire histological specimen in
our method. When iron overload is suspected after histological
examination of a fixed liver sample, it is possible to quantify the
iron concentration on the same material. Moreover, material from the
archives of anatomic pathology departments can be used for chemical
iron determination many years after biopsy or autopsy.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  N. J. Foot, H. E. Dalton, L. M. Shearwin-Whyatt, L. Dorstyn, S.-S. Tan, B. Yang, and S. Kumar Regulation of the divalent metal ion transporter DMT1 and iron homeostasis by a ubiquitin-dependent mechanism involving Ndfips and WWP2 Blood, November 15, 2008; 112(10): 4268 - 4275. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Fischer, P. Harmatz, P. Nielsen, and M. D. Cappellini Does liver biopsy overestimate liver iron concentration? Blood, September 1, 2006; 108(5): 1775 - 1776. [Full Text] [PDF]
|
|
|
|
|
|
R. Fischer, P. Harmatz, and P. Nielsen Does liver biopsy overestimate liver iron concentration? Blood, July 15, 2006; 108(2): 778 - 778. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
