Biuret values are extremely high (Table 1) and in the same
range as those measured previously when the peptide
fragments were taken into account (1).
The absorbance spectra of the Biuret chromogen in a
urine sample and an albumin solution are shown in Fig.
1A. The profiles are directly comparable, and there were
no anomalous peaks at 550 nm, the wavelength used in
The Biuret assay showed quantitative recovery (99.3%
? 0.7%; n ? 6) of albumin added to human diabetic urine
(for controls, the recovery was 98.7% ? 0.5%; n ? 5),
demonstrating that the urine matrix does not affect the
chromogen and thus validating the use of the Biuret assay
to measure total protein concentration of urine samples.
To test whether the matrix of the urine (or urinary
pigment) contributes to the Biuret chromogen, diabetic
urine samples were filtered through an Amicon/Milli-
pore membrane with a molecular mass cutoff of 500 (cat.
no. 13022). This filtration indicated that most of the
proteinaceous material (85.8% ? 9.4%; n ? 5) was re-
tained. Filtration of urine through an Amicon/Millipore
membrane with a molecular mass cutoff of 10 000 (cat. no.
13622) gave a recovery of 96.5% ? 0.9% (n ? 5) of the
Biuret-reactive material in the filtrate. These experiments
demonstrate that the protein fragments in diabetic human
urine for the samples studied have molecular masses
mainly between 500 and 10 000 Da. Similar results were
obtained for control urines, where the membrane with a
molecular mass cutoff of 500 retained 81.8% ? 6.4% (n ?
6) of the material, whereas filtration through the mem-
brane with a cutoff of 10 000 recovered 97.7% ? 0.8% (n ?
6) in the filtrate.
The HPLC profile of a control urine sample is shown in
Fig. 1B (peaks measured by absorbance at 214 nm were
also apparent when measured at 278 nm; not shown).
Urine retained by the filter after filtration through a
membrane with a molecular mass cutoff of 500 gave a
similar profile (not shown). The whole profile was altered
substantially by proteolytic digestion of the urine sample
by trypsin (Fig. 1C) or Glu C, another endoproteinase (not
shown). These studies demonstrate that the HPLC peaks
in human urine are proteinaceous materials.
Overall, these studies demonstrate that large quantities
of low-molecular mass protein-derived material exists in
urine, which had not been recognized previously.
We would like to acknowledge the kind assistance of
Shane Reeve and Dr. Ian Smith of the Baker Institute,
Melbourne, Australia, in performing the proteolytic di-
1. Osicka TM, Houlihan CA, Chan JG, Jerums G, Comper WD. Albuminuria in
patients with type 1 diabetes is directly linked to changes in the lysosome-
mediated degradation of albumin during renal passage. Diabetes 2000;49:
2. Osicka TM, Pratt LM, Comper WD. Glomerular capillary wall permeability to
albumin and horseradish peroxidase. Nephrology 1996;2:199–212.
3. Eppel GA, Nagy S, Jenkins MA, Tudball RN, Daskalakis M, Balazs NH, et al.
Variability of standard clinical protein assays in the analysis of a model urine
solution of fragmented albumin. Clin Biochem 2000;33:487–94.
4. Gornall AG, Bardawill CJ, David MM. Determination of serum proteins by
means of the Biuret reaction. J Biol Chem 1949;177:751–66.
5. Iwata J, Nishikaze O. New micro-turbidimetric method for determination of
protein in cerebrospinal fluid and urine. Clin Chem 1979;25:1317–9.
6. Lowry OH, Rosebrough NJ, Farr AI, Randall RT. Protein measurement with the
folin-phenol reagent. J Biol Chem 1951;193:265–75.
Integration of Olanzapine Determinations in a HPLC-
Diode Array Detection System for Routine Psychotropic
Drug Monitoring, Pierre M. Llorca,1Franc ¸ois Coudore,2*
Christophe Corpelet,2Aurelie Buyens,2Monique Hoareau,2and
Alain Eschalier2(1Centre Medico Psychologique B and
2Clinical Pharmacology and Toxicology Laboratory,
CHU G. Montpied, BP39 63003, Clermont-Ferrand Cedex
3, France; * author for correspondence: fax 33-4-73-751-
823, e-mail firstname.lastname@example.org)
Olanzapine is an atypical antipsychotic drug now consid-
ered as a first-line agent to treat schizophrenia and
psychotic mood disorders (1). Plasma concentrations in-
dicative of a clinical response are known to be ?9 ?g/L
(2). Knowledge of plasma concentrations, to check com-
pliance or drug-drug interactions, is also necessary in
treatment of schizophrenia (3). Moreover, olanzapine
toxicity may appear at blood concentrations that are
considerably lower than those observed in antidepres-
sant-related deaths (4).
Numerous publications have described the usefulness
of a HPLC silica column and aqueous methanol eluents
for the analysis of many basic drugs in plasma (5, 6). The
use of this type of column provides more reproducible
results, making it possible to obtain a sensitive HPLC-
ultraviolet (UV) procedure that is easier to perform and
less expensive than electrochemical (7, 8) or mass spec-
trometric (9) detection. We have adapted this HPLC
system for routine therapeutic drug monitoring of psy-
chotropic drugs (10–12). To update this system, it is
important to include new antipsychotic drugs, such as
The Dionex HPLC system used consisted of a Dionex
injector (Model ASI-100), an isocratic pump (Model
P-580A), and a photodiode array detector (Model UVD-
170S). Chromatograms were processed using the Chrome-
leonTMchromatographic data collection and analysis sys-
Blood samples were collected in tubes containing lith-
ium heparin as an anticoagulant. After centrifugation for
10 min at 3000g, the plasma was immediately separated,
supplemented with 250 g/L ascorbic acid (10 ?L/mL of
plasma), and stored at ?20 °C until analysis.
Olanzapine was extracted from 1 mL of serum after the
addition of 0.2 mL of bicarbonate buffer (pH 10.5), 20 ?L
of internal standard working solution (demethylated me-
tabolite of trimipramine; 2 mg/L), and 5 mL of a mixture
of hexane–isoamyl alcohol (98:2 by volume). The mixture
was shaken for 15 min and centrifuged at 3000g for 5 min.
The aqueous layer was discarded, and the organic layer
Clinical Chemistry 47, No. 9, 2001
was transferred to another glass tube and back-extracted
after acidification with 1 mL of 0.2 mol/L sulfuric acid.
After shaking and centrifugation, the aqueous layer was
alkalinized and reextracted with the same mixture. The
organic phase was evaporated to dryness under a stream
of nitrogen, the residue was dissolved in 100 ?L of mobile
phase, and 50 ?L was injected into the chromatographic
All analyses were performed on an Ultremex silica
column [250 ? 4.6 mm (i.d.); Phenomenex]. The mobile
phase consisted of a methanol–deionized water mixture
(70:30 by volume) containing 0.110 mL/L butylamine. The
mobile phase was filtered through a 0.22 ?m filter and
degassed before use. The chromatography was carried
out at ambient temperature at a flow rate of 1 mL/min.
Peaks were monitored at 273 nm.
For calibration, a six-point calibration curve was con-
structed before each series of assays with calibrators
prepared by adding different volumes of olanzapine
working solution into drug-free serum to obtain a concen-
tration range of 1.25–80 ?g/L. Olanzapine plasma con-
centrations were quantified using linear regression of
response (drug/internal standard peak height ratios) vs
concentration. Serum samples, prepared in advance by
adding 5 and 40 ?g/L olanzapine to a pooled serum and
then aliquoting in Eppendorf tubes and storing at ?20 °C,
were used as quality-control samples.
The retention times of olanzapine and the internal
standard were 3.9 and 11.6 min, respectively (Fig. 1). The
limit of quantification, defined as the lowest concentration
that could be calculated with a CV ?10%, was 1 ?g/L
(n ? 10). This detection limit was similar to those reported
for an electrochemical detector, i.e., 1 ?g/L (8) and 1.2
?g/L (3), but was lower than the limit of quantification
for UV detection (6), i.e., 1.56 ?g/L. However, some
authors have obtained a lower quantification limit (0.25
?g/L) with electrochemical (7) and mass spectrometric
The linearity of the extraction procedure was verified
over the calibration range by measuring drug-free plasma
supplemented with known concentrations of olanzapine.
The slope, y-intercept, and correlation coefficient for dif-
ferent calibration curves were 0.012 ? 0.004, 0.069 ? 0.269
?g/L, and 0.9962 ? 0.0037, respectively. Thus, calibration
curves were linear over the range 1.25–80 ?g/L.
The absolute recovery of olanzapine was obtained by
comparing the peak height of extracted and nonextracted
supplemented solutions. The mean extraction recovery
was 67.7% ? 10.5% (n ? 5). These values were compara-
ble to those obtained by the same extraction method with
tricyclic antidepressants (10, 11) or clozapine (12). The
three extraction steps, although time-consuming, pro-
vided high-purity extracts and increased column longev-
Precision was estimated from intra- and interday assay
variations. The intraday assay variation was determined
by analyzing six aliquots of supplemented samples con-
taining 5 and 40 ?g/L olanzapine with a calibration curve
on the same day. The interday variation was determined
by analyzing supplemented serum (5 and 40 ?g/L) on 10
different days with an independent calibration curve on
each day (Table 1). No chromatographic interference was
observed between olanzapine and commonly used psy-
chotropic drugs (amitriptyline, clomipramine, fluoxetine,
clozapine, flunitrazepam, levopromazine). The analysis
method was applied to monitor plasma samples collected
from a small number of patients with chronic schizophre-
nia. A wide concentration range was observed (3.6–89.4
?g/L) for administered doses between 10 and 20 mg. For
a complete analysis, it is important to add more individ-
ual data and elements of clinical response in a larger
number of patients (13).
As illustrated in our study, the HPLC-diode array
detection method with an unmodified silica column and
hydrophilic eluents is a powerful and sensitive tool for the
efficient separation and identification of psychotropic
drugs in plasma. The HPLC analysis requires only 15 min
for each sample. With the concomitant use of UV spectral
analysis, this system is well suited for routine drug
monitoring of multiple, coadministered medications, such
as sedatives and antidepressants.
Fig. 1. Chromatograms of extracted blank serum (A), a serum supplemented with 20 ?g/L olanzapine (B), and a sample from a patient (C).
IS, internal standard (demethylated metabolite of trimipramine).
Table 1. Precision study.
Intraday (na? 6)
Interday (na? 10)
an, number of determinations.
This work was supported by a grant from Eli-Lilly France.
1. Green B. Focus on olanzapine. Curr Med Res Opin 1999;15:79–85.
2. Perry JP, Sanger T, Beasley C. Olanzapine plasma concentrations and
clinical response in acutely ill schizophrenic patients. J Clin Psychopharma-
3. Aravagiri M, Ames D, Wirshing WC, Marder SR. Plasma level monitoring of
olanzapine in patients with schizophrenia: determination by high-perfor-
mance liquid chromatography with electrochemical detection. Ther Drug
4. Robertson MD, McMullin MM. Olanzapine concentrations in clinical serum
and postmortem blood specimens-when does therapeutic become toxic? J
Forensic Sci 2000;45:418–21.
5. Smith RM, Westlake JP, Gill R, Osselton MD. Retention reproducibility of
basic drugs in high-performance liquid chromatography on a silica column
with a methanol-high pH buffer eluent. J Chromatogr 1992;592:85–92.
6. Olesen OV, Linnet K. Determination of olanzapine in serum by high-
performance liquid chromatography using ultraviolet detection considering
the easy oxidability of the compound and the presence of other psychotropic
drugs. J Chromatogr B Biomed Sci Appl 1998;714:309–15.
7. Catlow JT, Barton RD, Clemens M, Gillespie TA, Goodwin M, Swanson SP.
Analysis of olanzapine in human plasma utilizing reversed-phase high-
performance liquid chromatography with electrochemical detection. J Chro-
matogr B Biomed Sci Appl 1995;668:85–90.
8. Chiu JA, Franklin RB. Analysis and pharmacokinetics of olanzapine
(LY170053) and two metabolites in rat plasma using reversed-phase HPLC
with electrochemical detection. J Pharm Biomed Anal 1996;14:609–15.
9. Berna M, Shugert R, Mullen J. Determination of olanzapine in human plasma
and serum by liquid chromatography/tandem mass spectrometry. J Mass
10. Coudore F, Ardid D, Eschalier A, Fialip J, Lavarenne J. High-performance
liquid chromatographic determination of amitriptyline and its main metabo-
lites using a silica column with reversed-phase eluent. Application in mice.
J Chromatogr 1992;584:249–55.
11. Coudore F, Hourcade F, Molinier-Manoukian C, Eschalier A, Lavarenne
J. Application of HPLC with silica-phase and reversed-phase eluents for the
determination of clomipramine and demethylated and 8-hydroxylated me-
tabolites. J Anal Toxicol 1996;20:101–5.
12. Coudore F, Nicolay A, Hoareau M, Eschalier A. Another use of silica gel and
aqueous eluent for HPLC analysis of clozapine and desmethylclozapine. J
Anal Toxicol 1999;23:195–9.
13. Kurz M, Hummer M, Kemmler G, Kurzthaler I, Saria A, Fleischhacker WW.
Long-term pharmacokinetics of clozapine. Br J Psychiatry 1998;173:341–4.
Diurnal Variations in Serum and Urine Markers of Type
I and Type III Collagen Turnover in Children, Ole D.
Wolthers,1*Carsten Heuck,2and Lene Heickendorff,3(1Chil-
dren’s Clinic Randers, DK-8900 Randers, Denmark;2De-
partment of Paediatrics and Institute of Experimental
Clinical Research, Medical Research Laboratories, and
3Department of Clinical Biochemistry, Aarhus University
Hospital, DK-8000, Aarhus, Denmark; * address corre-
spondence to this author at: Children’s Clinic Randers,
Dytmaersken 9, DK-8900 Randers, Denmark; fax 45-86-43-
33-95, e-mail email@example.com)
New serum and urine markers of type I and type III
collagen turnover have recently been introduced in chil-
dren. These markers include the formation markers of
type I collagen turnover, serum N-terminal (PINP) and
C-terminal (PICP) propeptides of type I procollagen, and
serum N-terminal propeptide of type III procollagen
(PIIINP), as well as the resorption markers serum cross-
linked C-terminal telopeptide of type I collagen (ICTP)
and urine cross-linked N-telopeptides of type I collagen
(Ntx) and deoxypyridinoline (DPD) (1–5). The aim of the
present study was to assess diurnal variations in serum
PINP, PICP, ICTP, PIIINP, and urine DPD and Ntx in
Two boys and five girls 10.4–14.4 years (mean, 12.2
years) were studied. One boy and three girls were in
puberty. Height SD scores varied from ?2.5 to 2.2 (mean,
0.6) and weight SD scores from ?1.4 to 1.0 (mean, ?0.4).
The study was approved by the local ethics committee,
and informed consent was obtained from all children and
The first sample was urine collected from 2400 to 0800
on the morning of the day of investigation. Thereafter,
urine was collected in 4-h intervals until 2400 and in
another 8-h interval from 2400 to 0800. Blood samples
were taken at 0900 and every 2 h thereafter until 0700 the
following morning. The samples were centrifuged at
3000g for 10 min within 1 h after they were collected. After
centrifugation, the samples were stored at ?80 °C and
batch-assayed at the completion of the study.
Each child received breakfast at 0815, lunch at 1300, an
ice cream at 1630, and dinner at 1900. Sleep was permitted
from 2400 to 0730.
Serum concentrations of PICP, PINP, ICTP, and PIIINP
were determined by specific RIAs based on human anti-
gen (Orion Diagnostica) (1–3). Intra- and interassay vari-
ations were 3.5–3.9% and 4.1–7.2%, respectively. Urine
DPD was measured by a solid-phase chemiluminescent
enzyme immunoassay on an automated instrument (Im-
mulite Pyrilinks-D; Diagnostica Products Corporation)
(4). Urine Ntx was measured by the Osteomark immuno-
assay (Ostex) (5). Intra- and interassay variations were 8%
and 9% for the DPD assay, respectively, and 8% and 12%
for the Ntx assay, respectively.
Data are described as percentages of the overall day
mean ? SE of the mean in the 24-h profile. To evaluate the
24-h profiles one-way ANOVA for repeated measure-
ments was performed followed by the Student-Newman-
Keuls method for all pairwise multiple comparisons. The
5% level of significance was used.
PICP and ICTP were relatively low during the day (Fig.
1) with increased PICP concentrations from 0100 to 0500
(P ? 0.006; F ? 5.0) and ICTP concentrations from 0100 to
0700 (P ? 0.002; F ? 6.2). Peak concentrations of PICP
(mean ? SE) occurred at 0500 [342.0 ?g/L (91.6 ?g/L)]
and trough concentrations at 1100 [286.0 ?g/L (46.4
?g/L); P ? 0.01]. Peak concentrations of ICTP were
detected at 0700 [12.1 ?g/L (1.4 ?g/L)] and trough
concentrations at 2100 [10.3 ?g/L (1.3 ?g/L); P ? 0.02].
No significant variations in PINP (F ? 2.1; P ? 0.17) or
PIIINP (F ? 2.1; P ? 0.15) were detected.
A significant diurnal variation in urine DPD (F ? 15.1;
P ?0.001) and Ntx (F ? 8.2; P ?0.001) was found. Peak
concentrations of DPD occurred in urine collected at
0800–1200 [19.7 nmol/mmol (1.6 nmol/mmol)] and
trough concentrations in urine collected at 2000–2400
[12.4 nmol/mmol (1.0 nmol/mmol); P ?0.01]. DPD in the
urine collected at 0800–1200 and two samples collected at
2400–0800 did not vary, whereas DPD concentrations in
each of these periods were higher than in the samples
collected at 1200–1600, 1600–2000, and 2000–2400 (P
Clinical Chemistry 47, No. 9, 2001