Determination of Aluminum in Bone by Atomic Absorption Spectroscopy by olr10626


									CLIN.CHEM.31/7, 1172-1174(1985)

            of                                 Spectroscopy
Determination Aluminumin Boneby AtomicAbsorption
Johanna Smeyers-Verbeke’              and Dierik Verbeelen2

In developing thismethod     fordetermination Al in bone we #{149}
                                               of                thoroughly rinse all laboratory ware in dilute (10 niJJL)
paid special attention to the homogenization bone sam-           nitric acid
ples,whichpresentsgreat difficulties    fortrace-elementanaly-   screen
                                                               #{149} all materials and reagents for possible sources of Al
sis. To minimizethe riskof contamination,we preferredlow-        contamination
temperature ashing over classicalwet-digestiontechniques          Instrumentation.     For all determinations we used a Model
for destruction of the organic material. Graphite-furnace      460 atomic absorption spectrometer         equipped with a HGA
atomicabsorption spectroscopy is used formeasurement        of 76B graphite furnace and a Model 56 recorder (all from
Al. ForAl concentrations               15
                           exceedIng g/g, directstandard-      Perkin-Elmer      Corp., Norwalk, CT 06856). Argon was used
ization against a calibrationline can be used (between-run     as the purging gas in the furnace. For background correc-
CV, 5.9%). For Al concentrationswithin the normal range        tion we used a deuterium background corrector. A Perkin-
(<15 g/g) the standard-addition technique should be ap-        Elmer AS-i autosampler delivered 20 4 of sample solu-
plied. Comparison of resultsby the method with those by a      tions into the furnace. The furnace conditions were as
                                 of Al
procedurebased on extraction witha saturated          solution follows: wavelength 309 nm; automatic sample injection 20
of EDTA revealed that, although the latter method gave         4; drying 100#{176}C 30 s; ashung 1450 or 1550#{176}C 20 s;
                                                                                     for                                for
considerably   lower results three of 13 samples, therewas
                             for                               atomization, with gas interrupt, 2650 #{176}C   for 20s.
no statistical difference between resultsy the two methods.
                                           b                      Reagents.    The water used to prepare all solutions was
                                                                       doubly distilled in a quartz device just before use. It con-
AddItIonal    Keyphrues:  trace elements        analytical error       tained no detectable Al. Al standards were prepared from a
extraction with EDTA compared     .hemodiafysis                        commercial 1 g/L stock solution of Al (E. Merck, Darmstadt,
                                                                       F.R.G.), with HC1 added to give a concentration          matching
    Aluminum toxicity is often observed in patients with               that in the samples.
chronic renal failure who are being treated by hemodialysis                Collection  of bone samples. Bone samples were obtained
(1, 2). In both dialysis encephalopathy          and osteomalacic      from rate that had been injected intraperitoneally        with an
bone disease, Al concentrations       are increased, especially in     AlCl3 solution containing 2mg of Al per liter, for one month.
bone and brain tissues. Accurate methods for Al determina-             The cumulative dose of Al administered         to each rat was 40
tions in body fluids and tissue are important in the preven-           rag. The bone samples were stored frozen until analysis.
tion and early diagnosis of Al intoxication, but such analy-           This separate study was designed to investigate the elimina-
ses are not easy because contaminating         Al is so ubiquitous.    tion of Al from bone after an exposure to Al and to evaluate
    In recent years several methods for determining            Al in   the effect of desferrioxamine, a chelating agent reported to
serum or plasma have been described that are based on                  be an effective treatment for severe dialysis osteomalacia (5,
graphite-furnace   atomic absorption spectrometry. However,            6).
little systematic investigation     has been directed to measur-           Normal bone samples were obtained from untreated rats.
ing Al in tissues. Frequently,     reports of results do not refer         Homogenization.     After lyophilization,   the bone samples
to the techniques used for homogenizing and decomposing                 were homogenized. Homogenization of hard tissue for trace-
the material or for evaluating the results obtained.                    element analysis presents great difficulties. To avoid con-
    We have studied the analysis of bone, giving special                tamination in trace-element analyses, one obviously needs
attention to the homogenization step. We used low-tempera-              to use non-metallic-component      homogenizers, but these are
ture aahing to decompose the bone samples, which then                   not always effective with tissues as hard as bone. We
were analyzed by graphite-furnace         atomic absorption spec-       pulverize bone samples with the brittle fracture technique
trometry. We compared our results by this method, which                 first proposed by Iyengar and Kasperek (7, 8), using the
includes a complete dissolution of the bone sample, with                Model U Microdismembrator         (Braun, Melsungen, F.R.G.)
those by a simpler procedure (3) in which Al is extracted               for this purpose. The method exploits the brittleness of bone
from dried, homogenized         bone with a saturated EIYFA             at the temperature of liquid nitrogen for pulverizing bone
solution. The minimum sample handling makes this extrac-                samples more easily. After a first coarse fracturing of the
tion method very attractive, but only if Al is completely               bone in a polyvinyl chloride vessel fabricated           for that
extracted from the bone.                                                purpose in our laboratory, we place the bone material in a
                                                                        Teflon vessel along with a tungsten carbide ball of 9 mm
Materials    and Methods                                                diameter. The vessel is cooled in liquid nitrogen for 10 miii,
   Precautions   to avoid contamination.   Elsewhere (4) we             then placed on the Microdismembrator          and vibrated for 1
reported the precautions to be taken to avoid contamination             mm at about 3000 cycles/rain. We found it necessary to
with extraneous Al when this element is measured            in          repeat the cooling and vibration steps three times to obtain
biological fluids:                                                      a powder fine enough to pass a 500-pm (pore size) sieve.
  use high-purity reagents and plastic or quartz materials in              Low-temperature    ashing. We ashed 100mg of the homoge-
  preference to glassware                                               nized bone sample in the Model 504 low-temperature asher
                                                                        (LFE Corp., Waltham, MA 02154). At a generator power of
                                                                        400 W, the sample was completely decomposed after 6 h.
   ‘Fsrmaceutisch Instituut, Laboratory for Pharmaceutical    and       The ash residue was dissolved in 3 mL of 100 mJJL HCL
Biomedical Analysis, and2 Academisch Ziekenhuis, Renal Unit of
the Department of Medicine, Vrije Universiteit Brussel, Laarbeek-       Analyses were then perfonned           after appropriate sample
lass 103, B-1090 Brussels, Belgium.                                     dilution, depending on the Al concentration.
  Received February 15, 1985; accepted April 16, 1985.                     EDTA extraction procedure.        We applied the procedure

1172 CUNICAL CHEMISTRY, Vol. 31, No. 7, 1985
described by LeGendre and Alfrey (3) to extract Al from
bone samples that we had homogenized       with the brittle-
fracture technique: 25 mg of the homogenized bone samples
were extracted     for 2 h with saturated   EDTA solution.
Measurements     were performed with standards prepared in
the EDTA solution. Sample solutions with Al concentra-
tions exceeding 100 g/L were appropriately     diluted before

Results and Discussion
   We could homogenize       the bone samples by the brittle
fracture technique without any difficulties if we first broke
up the bone by coarse fracturing.                                                                                    50              iSO       Al(r,g/nIJ
                                                                                                                     100           .200
   To decompose the organic material,          we preferred low-
                                                                            Fig. 2. Companson of a calibrabon line (#{149})
temperature     ashing over the classical wet-digestion      tech-
                                                                            fora bonesample diluted(0)10-fold, (x) 30-fold, and (0) 50-fold
niques because the ash residue can be dissolved in a smaller
volume of acid. Blanks, taken through the entire process to
evaluate possible Al contamination,       revealed that less than           these patients should allow for measurement           of the diluted
30 ng of Al per grain of bone was added during the                          bone samples by direct standardization.         But the normal Al
procedure.                                                                  content is <10 pg/g (9), so the standard-addition         technique
    First, we determined the optimal aahing temperature         for         should be applied to evaluate these concentrations.
use in the electrothermal      (graphite furnace) analysis. As                 Table 1 compares the direct estimation          of bone Al with
Figure 1 shows, standard solutions should not be ached at                   the standard-addition       technique for bone samples with
temperatures exceeding 1450 #{176}C,    whereas undiluted    bone           markedly      increased   Al concentrations.      The comparison
samples are not completely atomized until higher tempera-                   yielded the linear regression equation y = 0.99x             0.011 (r

                                                                            = 0.9901, S       = 4.03) wherex   is the standard addition andy
tures (1550 #{176}C)reached. This suggests that direct stan-
dardization    of undiluted bone samples will be impossible                 the direct estimation. Statistical       analysis by paired Stu-
because samples and standard solutions have no optimal                      dent’s t-test (10) showed no difference in results by the two
ashing temperature in common. Bone samples diluted 50-                      methods, indicating that bone Al can be directly estimated
fold after low-temperature       ashing give maiimum        abeor-          against a calibration     line at these concentrations       without
bance when ached at a temperature between 1450 and                          risk of interferences.
1600        Thus
        #{176}C. for direct standardization       the aching tem-              The method proposed here includes a complete dissolution
perature should be 1450 #{176}C.                                            of the bone sample. In the simple method based on an
   Direct standardization    will, however, be possible only if             extraction of Al from the dried, homogenized bone with a
the sample matrix does not influence the absorption signal.                 saturated solution of EJYFA (3), the supernate is directly
Figure 2 illustrates the effect of the matrix on additions of               injected into the graphite furnace. We compared results by
standard to increasing dilutions of the same bone sample.                   the method proposed here with those by this extraction
Standardization     with aqueous standards will be possible                 method (Fable 2). The comparison yielded the following
only if the Al concentration in the bone is great enough to be              linear regression equation: y = 0.851x + 5.09 (r = 0.8307,
stifi detected if the sample is diluted 50-fold. Standard                   S,    =   15.2), where x is the present method and y the
addition lines for lesser dilutions do not parallel the calibra-            extraction method. For three samples indicated in the table,
tion curve. We calculated         that Al bone concentrations               the extraction method yielded considerably lower results
exceeding 15 zg/g can be determined with direct standard-                   than our method. These are mainly responsible for the
ization, but for lower concentrations the standard-addition                 relatively low values of the slope and the correlation coeffi-
technique should be used.                                                   dent in the regression analysis. Nevertheless, the statistical
   Because uremic patients frequently have high concentra-                  comparison by use of the paired Students          t-test showed no
tions of bone aluminum->100           pg/g (5, 6)-samples    from           difference between results by the two methods (t = 1.74).
                                                                               Analytical recoveries of aluminum added to normal bone
                                                                            samples were nearly 100% for both methods (Table 3).
                                                                               In a recent article Stevens (11) compared the EDTA
                                                                            extraction method with a method in which soft tissue (bone

                                                                            Table 1. ComparIson of Direct Estimation (DE) of
                                                                              Bone Al and Standard Addition (SA) In 11 Al-
                                                                                       Loaded Rat-Bone Samples
                                                                                                     Al, pg/g dry wlght

                                                                                                  SA                        DE
                                                                                                 91.5                       93.4
                                                                                                111.1                      112.2
                                                                                                103.3                       96.5
                                                                                                 80.2                       81.4
                                                                                                 84.5                       78.3
                                   1500                     2000                                100.3                      105.5
                          Ash.n9,thr,   (‘CI                                            92.1                       99.9
                                                                                                 74.5                       71.7
FIg. 1. E1!ect of ashing temperature on the absorption signal of (#{149})
                                                                     a                          114.7                      112.3
100 og/L standard solution of Al, (x) an undiluted normal bone sample                             75.6                      76.9
containIng40 pg of Alper liter, and (0) a bone sample diluted50-fold
         1                                                                                        85.7                      84.6
containing 40 pg of Al per liter

                                                                                            CUNICAL CHEMISTRY, Vol. 31, No.           7, 1985         1173
                                                                     reference materials    for use in      checking the accuracy of
Table 2. Results by the Present Method (I) and the                   aluminum analyses. Only when           such materials, with con-
  Extraction Method (II) Compared (13 Samples)                       centrations  close to those found       in real samples, become
                          Al, pg/g dry wslght                        available will it be possible really    to be sure of the accuracy
                        I                   II                       of Al determinations.
                       93.4                89.9                        We thank Prof. D. L. Massart for valuable advice and Mrs.
                      112.2               104.4                      Schoonjans for technical assistance. The Fonda voor Geneeskundig
                       96.5               119.9                      Wetenschappelijk Onderzoek supplied financial support.
                      105.5                60.5*
                       84.0                92.9                      References
                       99.9                88.0                      1. Alfrey AC, Legendre GR, Kaehny WD. The dialysis encephalop-
                       71.7                53.6*                     athy syndrome. Possible aluminum intoxication. NEng1J Med 294,
                      112.3                83.0                       184-188 (1976).
                       76.9                70.8                      2. Parkinson IS, Ward MK, Feest TG, et al. Fracturing dialysis
                       90.4                90.7                      osteodystrophy and dialysis encephalopathy: An epidemiological
                      84.6                 79.3                      survey. Lancet i, 406-409 (1979).
                      75.4                 63.9                      3. LeGendre GR, Alfrey AC. Measuring picogram amounts of
                      87.3                 87.7                      aluminum     in biological tissue by flameless atomic absorption
  See discussion Intext                                              analysis of a chelate. Clin Chem 22, 53-56 (1976).
                                                                     4. Smeyers-Verbeke      J, Verbeelen D, Massart DL The determina-
                                                                     tion of aluminium in biological fluids by means of graphite furnace
                                                                     atomic absorption spectrometry. Clin Chim Acts 108,67-73(1980).
  Table 3. AnalytIcal Recovery of Added Al from
                                                                     5. Ackrill P, Day JP, Gasting FM, et aL Treatment of fracturing
                       Bone                                          renal osteodystrophy by desferrioxamine. Proc Eur Dial Transpi
     Al, pg sddsd                                  AscoYsly, S       Assoc 19, 203-207 (1982).
   p.r grim st ban.                                  (n 6)           6. Bile BU, Buchanan MRC, Stevens B, et al. The efficacy of
Present ethod
      m                                                              various treatment modalities in aluminium          associated bone dis.
           20                                       97.6 ± 7.8       ease. Ibid., 19, 195-201 (1982).
           100                                     100.6 ± 5.7       7. Iyengar GV. Homogenised sampling of bone and other biological
Extraction method                                                    materials. Radiochem Radioanal Lett 24,35-42(1976).
           20                                      100.6   ±   2.1   8. Iyengar GV, Kasperek K. Application of the brittle fracture
          100                                       98.8   -   22    technique (BV1’)to homogenise biological samples, and some obser-
                                                                     vations regarding the distribution behaviour of the trace elements
                                                                     at different concentration levels in a biological matrix. JRo4ioanal
was not included in the study) is solubilized by means of a          Chem 39, 301-316 (1977).
tissue solubilizer, tetramethylammonium         hydroxide, and       9. Boyce BF, Elder NY, Nicholson GD, at al. Quantitation and
                                                                     localisation of aluminium in human cancellous bone in renal
also compared results with those by classical wet digestion.
                                                                     osteodystrophy. Scanning Electron Microscopy 3,329-337 (1981).
The latter two methods yielded results that are completely
comparable, but the extraction method gave systematically            10. Maasart DL, Derde MP, Michotte Y, Kaufman L awcn.                 A
                                                                     Program to Compare Two Series of Measurements, Elsevier Science
lower results, attributed to incomplete extraction of Al. The        Publ., Amsterdan 1984.
differences we found are less evident and therefore can              11. Stevens BJ. Electrothermal      atomic absorption determination of
hardly invalidate the extraction procedure.                          aluminum in tissues dissolved in tetramethylammonium            hydrox-
   These method-comparison      studies again show the need for      ide. Clin Chem 30,745-747(1984).

1174 CUNICAL CHEMISTRY, ol.31, No. 7, 1985

To top