Clinical Chemistry 50: 1797-1803, 2004.
First published August 12, 2004; 10.1373/clinchem.2004.034223
(Clinical Chemistry. 2004;50:1797-1803.)
© 2004 American Association for Clinical Chemistry, Inc.
|
Automation and Analytical Techniques |
Rapid and Simultaneous Determination of Coproporphyrin and Protoporphyrin in Feces by Derivative Matrix Isopotential Synchronous Fluorescence Spectrometry
Dan-Li Lin1,
Li-Fang He1,2 and
Yao-Qun Li1,a
1 Department of Chemistry and Key Laboratory of Analytical Sciences of MOE, Xiamen University, Xiamen, China.
2 Department of Chemistry, Longyan College, Longyan, China.
aAddress correspondence to this author at: Department of Chemistry and Key Laboratory of Analytical Sciences of MOE, Xiamen University, Xiamen 361005, China. Fax 86-592-2185875; e-mail yqlig{at}xmu.edu.cn.
 |
Abstract
|
|---|
Background: Measurement of fecal porphyrins is important in the diagnosis of porphyria, but conventional methods to measure them have drawbacks. We explored the use of derivative matrix isopotential synchronous fluorescence (MISF) spectrometry for the measurement of coproporphyrin and protoporphyrin.
Methods: The MISF scanning route was selected based on information from the three-dimensional fluorescence spectrum, which was a combination of the contour line of protoporphyrin via a detection point of coproporphyrin and that of coproporphyrin via a detection point of protoporphyrin. Derivative technique eliminated the constant interfering signals. MISF was used to measure porphyrins in stools from 2 pregnant women and 20 healthy volunteers.
Results: The coproporphyrin and protoporphyrin spectra were resolved with almost no mutual interference. The amplitudes of the derivative peaks were linearly related to the concentrations of coproporphyrin up to 310 nmol/L and protoporphyrin up to 590 nmol/L. The detection limits for coproporphyrin and protoporphyrin were 1.2 and 1.7 nmol/L, respectively. The within-run imprecision (CV; n = 6) was 2.2% at 175 nmol/L for coproporphyrin and 2.3% at 500 nmol/L for protoporphyrin. Bland–Altman analysis indicated no significant differences between the proposed MISF method and conventional spectrophotometry or fluorimetry. Mean (SD) recoveries of porphyrins added to fecal samples were of 98 (7)% for coproporphyrin and 102 (4)% for protoporphyrin.
Conclusions: This technique provides spectral resolution of coproporphyrin and protoporphyrin, obviating the need for chromatographic separation, and measurements can be made in a single scanning. The method also appears suitable for routine testing of large numbers of samples.
|
Introduction
|
|---|
The porphyrias are characterized by excessive production and excretion of porphyrins. The disorders may be inherited or may result from exposure to certain toxicants. Many foreign chemicals (e.g., herbicides and polyhalogenated aromatic compounds) and some toxic metals (e.g., lead and mercury) can affect the activity of one or more enzymes of heme biosynthesis, inducing alterations in the porphyrin profile (1). Pregnancy can also lead to higher than normal porphyrin concentrations (2)(3).
Classification and fundamental understanding of the mechanisms of the various disorders have been advanced by measurements of the relative concentrations of the porphyrins found in excreta, blood, and tissue. The measurement of fecal porphyrins is a central feature of the differential diagnosis of porphyria. Coproporphyrin and protoporphyrin, the most commonly measured porphyrins in feces, are the major components in fecal porphyrins. The ratio of fecal coproporphyrin to fecal protoporphyrin varies among the porphyrias. For example, fecal protoporphyrin always exceeds coproporphyrin in variegate porphyria, whereas the reverse is true in hereditary coproprophyria (4).
Many methods for the determination of fecal porphyrins have been proposed, such as spectrophotometry (5), fluorimetry (6)(7)(8)(9), thin-layer chromatography (10)(11), HPLC (7)(8)(12)(13)(14), and mass spectrometry (15)(16). Conventional spectrophotometry and fluorimetry can determine only the total amounts (5)(6) or determine the ratios of individual porphyrins based on the mathematical resolution of simultaneous equations (9), which is unfavorable for the differential diagnosis of porphyria. HPLC and mass spectrometric methods can separate and determine various porphyrins, but they usually require tedious pretreatments and expensive apparatus.
Matrix isopotential synchronous fluorescence (MISF) spectrometry, which was first described by Murillo-Pulgarin and Molina (17)(18)(19), is a novel technique for resolving badly overlapping mixtures or detecting analytes in complex fluorescence backgrounds without the need for preseparation procedures; it has been used to determine various components in urine or serum (17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27). This method is particularly suitable for two-component analysis; for multicomponent systems with more than two analytes, other techniques should be combined.
Our objective was to develop a rapid and simple method for the simultaneous and direct measurement of coproporphyrin and protoporphyrin in feces by use of a derivative MISF technique.
|
Materials and Methods
|
|---|
reagents
Coproporphyrin III tetramethyl ester and protoporphyrin IX dimethyl ester were purchased from Sigma Chemicals and used as received. Hydrochloric acid and diethyl ether were of analytical reagent grade. After hydrolysis of porphyrin methyl esters in a small volume of 5.8 mol/L hydrochloric acid for 24 h, stock solutions of each porphyrin were stored in the dark in a refrigerator (4 °C).
instrumentation
All fluorescence spectra were obtained on a laboratory-constructed, computer (IBM)-controlled spectrofluorometer that was similar to one described previously (20). The spectrofluorometer was equipped with a 350W xenon lamp, and the slit bandpasses of the excitation and emission monochromators were 5 nm. In addition to conventional fluorescence spectra, the apparatus could provide all types of synchronous spectra. An electronic differentiator was connected directly to the spectrofluorometer to obtain first- or second-derivative spectra. A quartz glass cuvette with a pathlength of 1 x 1 cm was used throughout.
sample preparation
Fresh feces (
50–100 mg) from health humans was dissolved with 1.0 mL of concentrated hydrochloric acid in a centrifuge tube. Diethyl ether (3.0 mL) was added and thoroughly mixed to give an emulsion, followed by 3.0 mL of distilled water and further mixing. The mixture was centrifuged, separating into an ether layer, a pad of insoluble material at the interface, and a layer of aqueous acid. Chlorophyll derivatives and carotenoid pigments were partitioned into the ether phase, and the fecal porphyrins remained in the aqueous acid (lower) layer (5). Approximately 4 mL of the acid layer was transferred to a tube and kept in the dark in a refrigerator (4 °C). We weighed 250 mg of feces, dried it for 2 h, and reweighed it to determine the ratio of the wet weight to the dry weight.
selection of misf scanning route
A suitable MISF scanning route is essential for MISF measurements. After recording the excitation and emission fluorescence spectra of coproporphyrin and protoporphyrin, we calculated their theoretical three-dimensional fluorescence spectra by use of a homemade data-processing program. A MISF scanning route for the mixture of coproporphyrin and protoporphyrin was selected based on information from the three-dimensional spectrum.
Protoporphyrin was first supposed as the interfering component. The point of maximum fluorescence intensity for coproporphyrin (
excitation = 401 nm; emission = 593 nm) was selected as a detection point. We then searched for the fluorescence intensity of protoporphyrin at this point and obtained a set of excitation and emission with the same intensity. This trajectory was simply a MISF scanning route for the detection of coproporphyrin (indicated by the "i" in Fig. 1
). In a similar manner, we obtained the detection route of protoporphyrin (indicated by the "ii" in Fig. 1
), which is a constant-value trajectory of coproporphyrin via a detection point (excitation = 407 nm; emission = 604 nm) of protoporphyrin. The two contour lines were united as an integrated detection route with a crossing point. The first segment was the contour line of protoporphyrin, and the second one was that of coproporphyrin. When combined with a derivative technique, the protoporphyrin signal was eliminated in the first segment, which then showed the net derivative signal of coproporphyrin. Similarly, we could obtain the net derivative signal of protoporphyrin in the second segment. The simultaneous determination of two components can therefore be achieved in a single scan.
View larger version (41K):
[in this window]
[in a new window]
|
Figure 1. Theoretical contour map of 175 nmol/L coproporphyrin (thin solid lines) and 500 nmol/L protoporphyrin (dashed lines) and the determination route (thick solid line).
The trajectories i and ii are the detection routes of coproporphyrin and protoporphyrin, respectively. ex, excitation wavelength; em, emission wavelength.
|
|
misf measurements
Using the above selected trajectory, we recorded MISF spectra of porphyrins; the scanning time for a MISF spectrum is only 20 s. The amplitudes of the positive- and negative-derivative peaks were measured for the detection of these two porphyrins. The peak heights were retrieved from the spectra via Origin 6 software, and then the absolute values of positive and negative peaks were added as the fluorescence intensity of each component. For convenient observation and quantitative measurement, the spectra were plotted with the determination series set as the x coordinate and relative fluorescence intensity as the y coordinate. The determination series denote the spatial order of appearance of all dots forming spectra.
method comparison
To test the validity of the proposed MISF method, we performed quantitative spectrophotometric and fluorimetric measurements of fecal porphyrins, as proposed by Lockwood et al. (5) and Pudek et al. (6). The fecal samples were scanned from 350 to 450 nm with a Beckman DU7400 spectrophotometer. After determining the height of the Soret band above the sloping background absorbance, we used the formula developed by Lockwood et al. (5) to calculate the total porphyrin concentration, which was expressed as nmol/g dry weight.
By the method of Pudek et al. (6), we scanned the excitation spectra of the coproporphyrin calibrator and the real fecal samples from 340 to 450 nm, while monitoring the emission at 668 nm, the emission isosbestic point for coproporphyrin and protoporphyrin. The heights of the fluorescent signals at 410 nm, which is the excitation isosbestic point for coproporphyrin and protoporphyrin, and 340 nm, which is required for baseline correction, were used to calculate the total porphyrin concentration.
|
Results
|
|---|
spectral characteristics
Coproporphyrin and protoporphyrin have similar molecular structures. The conventional fluorescence spectra of these compounds were similar and overlapped extensively, as seen from the contour maps in Fig. 1
. It was not possible to resolve them by direct conventional constant-wavelength synchronous fluorescence approaches (Fig. 2
), whereas MISF spectroscopy provides a simple method to selectively detect a fluorescent compound in the presence of another strongly interfering one.
View larger version (17K):
[in this window]
[in a new window]
|
Figure 2. Constant-wavelength synchronous fluorescence spectra of 130 nmol/L coproporphyrin (dashed line), 250 nmol/L protoporphyrin (dotted line), and a mixture of the two (solid line).
 = 192 nm.
|
|
With the selected trajectory, the spectra of coproporphyrin and protoporphyrin could be resolved with almost no mutual interference (Fig. 3
). MISF spectra can be plotted in either the three-dimensional (Fig. 3A
) or two-dimensional (Fig. 3B
) form. The three-dimensional MISF spectrum can provide more information, whereas the two-dimensional spectrum is more convenient for obtaining quantitative information on the compounds. The coproporphyrin and protoporphyrin peaks appeared in turn with no overlap. In the spectral region of coproporphyrin, the derivative signal intensity of protoporphyrin tended to zero. Similarly, the contribution of coproporphyrin was nearly zero in the spectral region of protoporphyrin. Peak-to-peak measurement was used for the detection of these two components with a-a' and b-b' corresponding to coproporphyrin and protoporphyrin, respectively.
View larger version (20K):
[in this window]
[in a new window]
|
Figure 3. First-derivative MISF spectra of coproporphyrin and protoporphyrin.
(A), three-dimensional first-derivative MISF spectrum of a mixture containing 110 nmol/L coproporphyrin and 250 nmol/L protoporphyrin, The thin line, corresponding to the baseline along the selected scanning route, is indicated in the plot for easy observation. ex, excitation wavelength; em, emission wavelength. (B), two-dimensional first-derivative MISF spectra of 110 nmol/L coproporphyrin (dotted line), 250 nmol/L protoporphyrin (dashed line) and a mixture of the two (solid line). The points a/a' and b/b' indicate the points used for peak-to-peak measurements.
|
|
First-derivative MISF spectra for coproporphyrin and protoporphyrin added in various concentrations to 1.0 mol/L hydrochloric acid are shown in Fig. 4
. At coproporphyrin concentrations up to 310 nmol/L, the fluorescence intensity of protoporphyrin (250 nmol/L) was basically invariable, indicating that coproporphyrin in this concentration range will not interfere with protoporphyrin measurements, based on the premise of self-linearity. When the protoporphyrin concentration was <590 nmol/L, its interference on coproporphyrin was negligible.
View larger version (23K):
[in this window]
[in a new window]
|
Figure 4. First-derivative MISF spectra of coproporphyrin (A) and protoporphyrin (B) with different concentrations.
(A) coproporphyrin was added at concentrations of 0 (trace 1), 45 (trace 2), 90 (trace 3), 130 (trace 4), 175 (trace 5), 220 (trace 6), and 310 (trace 7) nmol/L in the presence of 250 nmol/L protoporphyrin. (B), protoporphyrin was added in concentrations of 0 (trace 1), 85 (trace 2), 170 (trace 3), 250 (trace 4), 335 (trace 5), 420 (trace 6), 500 (trace 7), and 590 (trace 8) nmol/L in the presence of 130 nmol/L coproporphyrin.
|
|
linearity, detection limit, and precision
Individual calibrators were prepared by diluting stocks solutions in 1.0 mol/L hydrochloric acid. We carried out a statistical analysis of the experimental data by fitting the least-squares line according to the equation: y = a + bx. The linear equations were as follows: for coproporphyrin, y = 37.8 + 2.71x; for protoporphyrin, y = –24.9 + 1.24x. The correlation coefficients (r) were 0.9990 and 0.9979, respectively.
The detection limits of 1.2 and 1.7 nmol/L for coproporphyrin and protoporphyrin, respectively, were obtained based on 3 SD of the 1.0 mol/L hydrochloric acid blank according to the IUPAC definition, corresponding to 0.50 nmol of coproporphyrin and 0.72 nmol of protoporphyrin per gram (dry weight) of feces.
To assess the within-run imprecision of the method, we measured six solutions that all contained 175 nmol/L coproporphyrin and 500 nmol/L protoporphyrin. The CVs were 2.2% and 2.3% for coproporphyrin and protoporphyrin, respectively.
method comparison
We compared the proposed MISF method with conventional spectrophotometry (5) and fluorimetry(6). The fecal porphyrin results measured by these three methods in specimens from 2 pregnant women and 20 healthy volunteers are summarized in Table 1
. We observed increased porphyrin excretion in the two pregnant women. The porphyrin data obtained by the three different methods were in accordance with each other. To show the agreement of the methods, we analyzed the porphyrin data measured by three methods for 20 specimens from healthy donors by the Bland–Altman statistical method (28) (Fig. 5
). The sums of the amounts of the two porphyrins obtained by the MISF method were used for comparison. Shown in Fig. 5A
is the Bland–Altman plot of MISF vs spectrophotometry. The mean difference (
; MISF method minus spectrophotometry) was 0.8 nmol/g dry weight, and the standard deviation of the difference (S) was 1.3 nmol/g dry weight. Hence the limits of agreement were
– 2S to
+ 2S, or –1.8 to 3.4 nmol/g dry weight. We also checked the precision of the estimated limits of agreement. The 95% confidence intervals for the mean, upper limit, and lower limit of the differences were 0.8 ± 0.62, 3.4 ± 1.07 and –1.8 ± 1.07 nmol/g dry weight, respectively. The narrow range for the limits of agreement (–1.8 to 3.4 nmol/g dry weight) confirmed that the two methods were clinically comparable.
View this table:
[in this window]
[in a new window]
|
Table 1. Results for fecal porphyrins obtained by three different methods in specimens from 20 healthy volunteers and 2 pregnant women.
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Figure 5. Bland–Altman plot of the difference vs mean results for fecal porphyrin values obtained by the MISF method and spectrophotometry (A) or by the MISF method and conventional fluorimetry (B).
Spectro, spectrophotometry; Fluori, fluorimetry, wt, weight.
|
|
Similarly, we compared the MISF method and conventional fluorimetry (Fig. 5B
). The
was 0.9 nmol/g dry weight and the S was 1.0 nmol/g dry weight. The range for the limits of agreement (–1.0 to 2.8 nmol/g dry weight) was sufficiently narrow. The 95% confidence intervals for the mean, upper limit, and lower limit of the differences between these two methods were 0.9 ± 0.45, 2.8 ± 0.77, and –1.0 ± 0.77 nmol/g dry weight, respectively.
The absorbance, fluorescence excitation, and MISF spectra of a fecal specimen from a pregnant woman analyzed by these three methods are shown in Fig. 6
; as seen in Fig. 6
, spectrophotometry produces the least distinct spectrum, which is not good for analytical proposes. MISF spectrometry produces a well-structured spectrum and makes it possible to identify and quantify coproporphyrin and protoporphyrin simultaneously.
View larger version (14K):
[in this window]
[in a new window]
|
Figure 6. Absorbance (A), fluorescence excitation (emission = 668 nm; B), and MISF (C) spectra of a fecal sample from a pregnant woman containing 10.3 nmol of coproporphyrin and 27.0 nmol of protoporphyrin per gram dry weight.
|
|
recovery
We added stock solutions of porphyrins and fecal extract to a tube and diluted them with 1.0 mol/L hydrochloric acid. The derivative MISF signals of the samples were then measured. The results are summarized in Table 2
, the mean (SD) recoveries were 98 (7)% for coproporphyrin and 102 (4)% for protoporphyrin.
|
Discussion
|
|---|
MISF spectrometry is a useful technique for resolving mixtures with badly overlapping spectra. Its scanning route is a contour line of the interfering component or fluorescence background (matrix), which joins the identical intensity points on a fluorescence matrix three-dimensional spectrum. Because the background MISF signal and that of the interfering component are constant, they can be further compressed by use of a first-order derivative technique. The selectivity of fluorescence spectrometry is thus improved and its high sensitivity is maintained.
Our experimental results show the successful application of MISF spectrometry to the rapid quantification of coproporphyrin and protoporphyrin in feces. The Bland–Altman analysis demonstrates the agreement between the proposed MISF method and conventional spectrophotometry as well as fluorimetry for the total amounts of the two porphyrins. The MISF method has an advantage over conventional spectrophotometry and fluorimetry, which only determine the total amounts, not the individual signals of two porphyrins. The MISF method can therefore simultaneously and directly determine coproporphyrin and protoporphyrin concentrations in feces.
The method proposed here provides new possibilities for further improvement in the selectivity of clinical analyses. This inexpensive technique is also suitable for routine screening of large numbers of samples because only one scan and no highly specialized instrumentation are required for the direct assay of both porphyrins. The analytical recoveries and precision are satisfactory.
In conclusion, this rapid, simple method is suitable for the simultaneous and direct determination of these two porphyrins and could become a new, useful tool for detecting fecal porphyrins.
|
Acknowledgments
|
|---|
We thank the Educational Ministry Foundation of China, the National Natural Science Foundation of China (Grant 29875023), and the Natural Science Foundation of Fujian Government for financial support.
|
References
|
|---|
- Bowers MA, Aicher LD, Davis HA, Woods JS. Quantitative determination of porphyrins in rat and human urine and evaluation of urinary porphyrin profiles during mercury and lead exposures. J Lab Clin Med 1992;120:272-281.[ISI][Medline]
[Order article via Infotrieve]
- Sakaguchi S, Sakaguchi T, Nakamura I, Aminaka M, Tanaka T, Kudo Y. Effect of beryllium chloride on porphyrin metabolism in pregnant mice administered by subcutaneous injection. J Toxicol Environ Health 1997;50:507-517.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Jones KG, Sweeney GD. Quantitation of urinary of porphyrins by use by second-derivative spectroscopy. Clin Chem 1979;25:71-74.[Abstract/Free Full Text]
- Lim CK, Peters TJ. Urine and faecal porphyrin profiles by reversed-phase high-performance liquid chromatography in the porphyrias. Clin Chim Acta 1984;139:55-63.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Lockwood WH, Poulos V, Rossi E, Curnow DH. Rapid procedure for fecal porphyrin assay. Clin Chem 1985;31:1163-1167.[Abstract/Free Full Text]
- Pudek MR, Schreilber WE, Jamani A. Quantitative fluorometric screening test for fecal porphyrins. Clin Chem 1991;37:826-831.[Abstract/Free Full Text]
- Casini S, Fossi MC, Cavallaro K, Marsili L, Lorenzani J. The use of porphyrins as a non-destructive biomarker of exposure to contaminants in two sea lion populations. Mar Environ Res 2002;54:769-773.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Taylor C, Duffy LK, Plumley FG, Bowyer RT. Comparison of spectrofluorometric and HPLC methods for the characterization of fecal porphyrins in river otters. Environ Res 2000;84:56-63.[Medline]
[Order article via Infotrieve]
- Grandchamp B, Deybach JC, Grelier M, Verneuil HD, Nordmann Y. Studies of porphyrin synthesis in fibroblasts of patients with congenital erythropoietic porphyria and one patient with homozygous coproporphyria. Biochim Biophys Acta 1980;629:577-586.[Medline]
[Order article via Infotrieve]
- Henderson MJ. Thin-layer chromatography of free porphyrins for diagnosis of porphyria. Clin Chem 1989;35:1043-1044.[Abstract/Free Full Text]
- Lam CW, Lai CK, Chan YW. Simultaneous fluorescence detection of fecal urobilins and porphyrins by reversed-phase high-performance thin-layer chromatography. Clin Chem 1998;44:345-346.[Free Full Text]
- Taylor C, Duffy L, Bowyer RT, Blundell G. Profiles of fecal porphyrins in river otters following the Exxon Valdez oil spill. Mar Pollut Bull 2000;40:1132-1138.[CrossRef]
- Zuijderhoudt FMJ, Kamphuis JS, Kluitenberg WE, Dorresteijn-de BJ. Precision and accuracy of a HPLC method for measurement of fecal porphyrin concentrations. Clin Chem Lab Med 2002;40:1036-1039.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Taylor C, Ben-David M, Bowyer RT, Duffy LK. Response of river otters to experimental exposure of weathered crude oil: fecal porphyrin profiles. Environ Sci Technol 2001;35:747-752.[Medline]
[Order article via Infotrieve]
- Luo J, Lamb HJ, Lim CK. Analysis of urinary and faecal porphyrin excretion patterns in human porphyries by fast atom bombardment mass spectrometry. J Pharm Biomed Anal 1997;15:1289-1294.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Luo J, Lim CK, Matteis DF. Peroxylated and hydroxylated uroporphyrins: a study of their production in vitro in enzymic and chemical model systems. Biomed Chromatogr 1996;10:213-220.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Murillo-Pulgarin JA, Molina AA. Matrix isopotential synchronous fluorescence direct determination of gentisic acid in urine. Anal Chim Acta 1994;296:87-97.[CrossRef]
- Murillo-Pulgarin JA, Molina AA, Lopez PF. Direct determination of amiloride in urine using isopotential fluorimetry. Analyst 1997;122:247-252.[CrossRef][Medline]
[Order article via Infotrieve]
- Murillo-Pulgarin JA, Molina AA. Direct determination of salicylic acid in human serum by matrix isopotential synchronous fluorescence. Microchem J 1995;52:341-349.[CrossRef]
- Sui W, Wu C, Li YQ. Rapid simultaneous determination of four anthracene derivatives using a single non-linear variable-angle synchronous fluorescence spectrum. Fresenius J Anal Chem 2000;368:669-675.[CrossRef][Medline]
[Order article via Infotrieve]
- Murillo-Pulgarin JA, Molina AA, Lopez PF. Simultaneous direct determination of amiloride and triamterene in urine using isopotential fluorometry. Anal Biochem 2001;292:59-68.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Nevado JJB, Murillo-Pulgarin JA, Laguna, Gomez MA. Determination of pyridoxal in human serum by matrix isopotential synchronous fluorescence spectrometry. J Pharm Biomed Anal 1994;14:1487-1494.[CrossRef]
- Murillo-Pulgarin JA, Molina AA. Direct determination of quinidine in urine. Anal Chim Acta 1995;317:359-364.[CrossRef]
- Murillo-Pulgarin JA, Molina AA. Direct determination of salicylamide in serum by matrix isopotential synchronous fluorimetry. Talanta 2002;56:557-564.[Medline]
[Order article via Infotrieve]
- Murillo-Pulgarin JA, Molina AA, Lopez PF. Direct determination of nalidixic acid in urine by matrix isopotential synchronous fluorescence spectrometry. Talanta 1996;43:431-438.[CrossRef][Medline]
[Order article via Infotrieve]
- Murillo-Pulgarin JA, Molina AA, Lopez PF. Direct determination of triamterene in urine by matrix isopotential synchronous fluorescence spectrometry. Anal Chim Acta 1996;326:117-126.[CrossRef]
- Murillo-Pulgarin JA, Molina AA, Lopez PF. Direct determination of dipyridamole in serum. Anal Biochem 1997;245:8-16.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
- Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307-310.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
Top
Abstract
Introduction
