Comparison of Differential Pulse Voltammetry (DPV)—a new method of carbamazepine analysis—with Fluorescence Polarization Immunoassay (FPIA)
ABSTRACT Carbamazepine is a widely used anti-epileptic drug with narrow therapeutic range. Many methods have been developed for monitoring
the serum drug level. Differential pulse voltammetry (DPV), an electrochemical method advantaged by simple, inexpensive, and relatively short analysis time, has recently been developed
for carbamazepine detection. We used a newly developed DPV method with glassy carbon as a working electrode to determine the
carbamazepine level. The performance of DPV is compared with the widely used fluorescence polarization immunoassay (FPIA) technique in precision, accuracy, linearity and detection limit. The precision, linearity and accuracy of the DPV and FPIA
techniques were comparable at most clinical used levels. The detection limit was 1 μg/mL for the DPV technique and 0.5 μg/mL
for the FPIA technique. The performance of the DPV technique was within the FDA guidelines for bioanalytical methods, which
ensures the clinical applicability of the DPV technique. The DPV technique may have the potential to be a good alternative
for carbamazepine analysis.
Keywordselectro-chemical method–differential pulse voltammetry–drug monitoring–immunoassay–carbamazepine
- [show abstract] [hide abstract]
ABSTRACT: Recently, there has been considerable discussion regarding the classification of drugs as a function of their therapeutic index, defined as the ratio between the upper and lower limits of the therapeutic range. Pharmacologic agents with a therapeutic index < 2 are classified as "narrow therapeutic index" (NTI) drugs. One of the agents classified as an NTI drug is carbamazepine. These recent developments led us to evaluate critically the evidence supporting the classification of carbamazepine as an NTI drug to address an old question: "Does carbamazepine have a narrow therapeutic plasma concentration range?"Therapeutic Drug Monitoring 02/1998; 20(1):56-9. · 2.23 Impact Factor
- The Lancet 08/1985; 2(8469-70):1432. · 39.06 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: A retrospective study of consecutive cases of massive carbamazepine poisoning treated in an intensive care unit during the period 1981-1991 was performed, mainly to determine whether serum carbamazepine levels were predictive of toxicity. Out of a total of 51 admissions with a diagnosis of carbamazepine self-poisoning, 28 (25 patients) were included. The reasons for exclusion were coingestion of other drugs (11 cases), incorrect diagnosis or inadequate information (6 cases), a peak observed serum concentration of carbamazepine below 76 mumol/L [18 mg/L] (4 cases), and lack of any documented serum carbamazepine assay (2 cases). The peak serum concentrations ranged from 78 to 285 mumol/L [18.4 to 67.4 mg/L]. It was found that serum levels equal to or above 170 mumol/L [40 mg/L] were significantly associated with an increased risk of serious complications such as coma, seizures, respiratory failure and cardiac conduction defects. In 60% of the 10 patients with a serum level > or = 170 mumol/L at least two of these symptoms occurred, in 50% at least three, and in 40% all four. There were two fatalities. Among the 16 patients (18 admissions) with a serum carbamazepine concentration below 170 mumol/L, only one was comatose and none had any of the other severe symptoms. It is concluded that serum carbamazepine levels accurately predict the severity of toxicity in massive carbamazepine poisoning in adults.Journal of toxicology. Clinical toxicology 01/1993; 31(3):449-58.
Comparison of Differential Pulse Voltammetry
(DPV) – a New Method of Carbamazepine Analysis--
with Fluorescence Polarization Immunoassay (FPIA)
W.Y. Lin, M.L.Pan and Y.L. Su
Department of Nuclear Medicine
Taichung Veterans General Hospital
160 Sec. 3, Chung-Gang Rd.
Taichung 407, Taiwan
Background: Differential pulse voltammetry (DPV) has recently been developed for
carbamazepine detection. DPV technique is simple, inexpensive, and has relatively short analysis
time. In this study, we compared the performance of the DPV technique in carbamazepine
detection with that of the widely used Fluorescence Polarization Immunoassay (FPIA) technique.
Methods: DPV uses glassy carbon as a working electrode and a platinum wire as the auxiliary
electrode. Various concentrations of the standard solutions of carbamazepine were prepared by
dissolving pure carbamazepine powder in Tetrabutylammonium Perchlorate (TBAP)/acetonitrile.
FPIA technique was performed by the Abbott TDx analyzer. Precision, accuracy, linearity and
detection limit were measured for comparison.
Results: The precision and linearity of the DPV and the FPIA techniques were found to be
comparable. The accuracy of the DPV technique was also comparable with that of the FPIA
technique. The detection limit was 1ug/mL for the DPV technique and 0.5 ug/mL for the FPIA
technique. Both were good enough for clinical use. In addition, the performance of the DPV
technique was within the FDA guidelines for bioanalytical methods, which ensures the clinical
applicability of the DPV technique.
Conclusion:The DVP technique may have the potential to be a good alternative for
• Carbamazepine is an anti-epileptic drug widely used for treatment of simple and complex
partial seizures, trigeminal neuralgia, and bipolar affective disorder. It selectively inhibits the
high frequency epileptic foci without affecting the normal neural activity by blocking sodium
channels [1-3]. The usual adult therapeutic levels are between 4 and 12 ug/mL . Serious
side effects, such as coma, seizures, respiratory failure and cardiac conduction defects,
develop more frequently when the serum level of carbamazepine is higher than 15 ug/mL [5-
7]. In one report, approximately 13% of patients died after massive carbamazepine overdose
. Thus, it is critical to monitor the serum drug level.
•Many analytic techniques have been applied in the determination of the carbamazepine level,
such as high performance liquid chromatography (HPLC), gas chromatography, gas
chromatography combined with mass spectrometry and fluorescence polarization
immunoassay (FPIA) [9-14]. Though the HPLC technique has the advantages of rapid run
times and excellent resolution, the expense of the equipment and the high maintenance
required to keep it running optimally limit the application in clinical labs.
• Gas chromatography has the advantage of very high resolving power provided by the
capillary columns that are typically used. However, the compounds must be not only
sufficiently volatile to be introduced in the gas phase when the sample is injected into the GC
but also stable so that they do not degrade at the temperatures required to vaporize them. As a
consequence, gas chromatography is generally limited to non-polar and slightly polar
molecules, which make up about 20% of the known organic molecules . In the clinical
laboratory, immunoassays are more widely used to monitor the carbamazepine concentrations
in serum or plasma due to the simplicity of use . Of these immunoassays, FPIA is
accepted in most clinical labs because it provides accurate and sensitive measurement of
small toxicology analytes, such as therapeutic drugs, narcotics, and some hormones, than
Electrochemical methods have been proved to be sensitive and reliable for detection of
several electroactive drugs such as abacavir and sildenafil citrate. In comparison with other
analytical techniques, electrochemical techniques are simple, inexpensive, and have relatively
short analysis time [17-19]. Only a few studies have investigated carbamazepine with
electrochemical techniques such as cyclic voltammetry, differential pulse polarography and
coulometry [20-21] and none of them compared the electrochemical technique with the
widely used FPIA technique. In this study, we evaluated the performance of differential pulse
voltammetry (DPV), one of the electrochemical methods, in carbamazepine analysis and
compared it with the performance of FPIA.
Materials and Methods (I)
• Pure carbamazepine in powder form was purchased from MP Biomedicals, Inc. (Germany).
Tetrabutylammonium Perchlorate (TBAP) was obtained from Toyo Kasei Kogyo Co., Ltd.
(Japan). The tested samples were prepared in 7 different concentrations (0, 2, 4, 8, 12, 20 and
23.6 ug/mL) by dissolving carbamazepine in 0.1M TBAP/ acetonitrile. All the other reagents
used were of analytical grade. Doubly distilled water was obtained by purification through a
Millipore water system.
Materials and Methods (II)
•For analytical application, the following parameters were employed: DPV pulse amplitude,
50mV; pulse width: 0.05(s); sample width: 0.0167(s); pulse period: 0.2 (s); scan rate: 20 mV/s−1.
DPV was performed with a CHI421A (CHI Model 660 series electroanalytical workstation). It
was conducted with the use of a three-electrode cell in which a BAS glassy carbon electrode (area
= 0.07 cm2) was used as a working electrode. The glassy carbon electrode was polished with
0.05μm alumina on Buehler felt pads and was ultrasonicated for 2 min to remove the alumina
residue. A platinum wire was used as the auxiliary electrode. All cell potentials were measured by
using a homemade Ag/AgCl, KCl (sat.) reference electrode. The buffer solution contained 0.1M
(C4H9)4NClO4 to support electrolytes in CH3CN.
•The typical current peak of carbamazepine is shown in Figure 1. The +1.24V peak was adapted in
our study for analysis since the peak is higher than that of -2.2V. The concentration-related
currents are represented in Figure 2. According to the calibration curve shown in Figure 2, the
concentration of the tested samples can be calculated from the current.
Fig.2. DPV voltammograms of carbamazepine
under optimum conditions: ( a ) 2.0 × 10 -4 M
(mol/L); ( b ) 1.0 × 10 -4 M (mol/L); ( c ) 4.0 ×
10 -5 M (mol/L); ( d )1.0 × 10 -5 M (mol/L);
( e )1.0 × 10 -6 M (mol/L) at +1.24V (A) and -
Materials and Methods (III)
Fig.1. Typical DPV potential of
carbamazepine occurs at +1.24V (A) and -
Materials and Methods (IV)
The FPIA technique was performed by a qualified technician using an Abbott TDx analyzer
with standard operating procedure. All the required reagents were purchased from Abbott
Laboratories (Chicago, IL, USA).
Precision. Tested samples of 4, 8, 12 ug/mL of carbamazepine were analyzed 20 times by the
DPV technique and four tests per run for 5 runs by the FPIA technique to evaluate the between-
run variation. Comparison of the with-in run variation is not possible since the DPV technique can
only test one sample per run. The results are expressed as coefficient of variation (CV, %).
Accuracy. Tested samples of 2, 4, 8, 12, 20 and 23.6 ug/mL of carbamazepine were analyzed 5
times by the DPV technique and FPIA technique, respectively. The results are expressed in bias,
the difference between the mean of each concentration and the reference concentration, and in the
percent recovery, which is calculated by dividing the bias by the reference concentration. The
positive bias indicates that the mean concentration is higher than the reference concentration,
whereas the negative bias indicates that the mean concentration is lower than the reference
Materials and Methods (V)
Linearity. The 7 different concentrations of the tested samples were tested three times with each
technique. The mean of the three results was correlated with the theoretical concentration of each
sample. The results are expressed as the coefficient of determination (R2, RSQ).
Detection limit. The tested sample with 0 ug/mL was analyzed 10 times with each technique. The
tested sample of 2 ug/mL was diluted serially and analyzed 10 times with each technique as well.
Then the mean and the CV of the results were calculated. The minimal concentrations indicating
that the mean – 3SD was higher than the mean + 3SD of the blank results were interpreted as the
detection limit of each technique.
Materials and Methods (VI)
According to the FDA guidelines for bioanalytical methods , it is recommended that the
accuracy and precision of the assay should be within 15% of the actual value except at the lower
limit of quantitation (LLOQ), where it should not deviate by more than 20%.
Using Pearson's correlation, a linear relationship was calculated between the results obtained with
the FPIA technique and the reference concentration as well as between the results obtained with
the DPV technique and the reference concentration. A t-test was used for the calculation of the
statistical significance of r (Pearson's Correlation Coefficient). A P value of less than 0.05 was
considered to be statistically significant. The coefficient of determination (R2, RSQ) was also
Agreement between the two methods and agreement of each method with the reference value
concentration are presented using the Bland–Altman approach, by plotting the percent difference
between the two methods versus the mean concentration determined. We considered outliers
values to be outside two standard deviations from the mean.
The results of the precision, accuracy, linearity and the detection limit of the DPV and the
FPIA techniques are listed in Table 1. The coefficient of variation for the DPV technique at the
concentrations of 4, 8, and 12 ug/mL was 3.27%, 2.35% and 2.87%, respectively. The coefficient
of variation for the FPIA technique at the concentrations of 4, 8, and 12 ug/mL was 3.26%, 2.65%
and 2.44%, respectively. The precision of the DPV and FPIA techniques was comparable.
The bias for the DPV technique at the concentrations of 2, 4, 8, 12, 20, and 23.6 ug/mL was -
0.38 ug/mL (-18.89%), -0.30 ug/mL (-7.38%), 0.53 ug/mL (6.62%), 0.59 ug/mL (4.88%), -0.29
ug/mL (-1.44%) and -0.14 ug/mL (-0.61%), respectively. The bias for the FPIA technique at the
concentrations of 2, 4, 8, 12, 20, and 23.6 ug/mL was -0.03 ug/mL (-1.4%), -0.13 ug/mL (-3.28%),
0.19 ug/mL (2.32%), -0.52 ug/mL (-4.31%), 1.55 ug/mL (7.76%) and 0.03 ug/mL (0.13%),
Table 1. The results of the precision, accuracy, linearity and detection limit of
the DPV and FPIA techniques
The performance of both techniques was within the FDA guidelines for bioanalytical methods,
which recommends that allowable deviation should not be more than 20% at the lower limit of
quantitation (LLOQ) ?2ng/mL in the current setting ?and should be within 15% of the actual value
for the other levels (4~23.6 ug/mL in the current setting). Although the bias in the DPV technique
seems bigger than that in the FPIA technique, it ensures the clinical applicability of the DPV
In comparison with the reference samples, the RSQ was 0.993 for the DPV technique and
0.994 for the FPIA technique. The t-test for the statistical significance of r (Pearson's Correlation
Coefficient) in both the DPV and FPIA techniques was P<0.001. The detection limit was 1ug/mL
for the DPV technique and 0.5 ug/mL for the FPIA technique.
The Bland-Altman approach for both the DPV and reference value concentration (Fig. 3) and
the FPIA and reference value concentration (Fig. 4) showed good agreement.
Fig.3. Bland-Altman plots of the DPV results and reference value
concentrations. The solid line represents the mean difference, and the dashed
line represents 1.96 SD.
Fig.4. Bland-Altman plots of the FPIA results and reference value
concentrations. The solid line represents the mean difference, and the
dashed line represents 1.96 SD.