Background Image
Table of Contents Table of Contents
Previous Page  320 / 658 Next Page
Information
Show Menu
Previous Page 320 / 658 Next Page
Page Background

Results and Discussion

Figures 1 and 2 demonstrate the separation of 32 illicit

drugs in less than 10 minutes. Using an SRM dwell time

of 10 ms per transition yielded a minimum of 15 data

points across an LC peak. The limits of quantitation

(LOQs) were determined as either 0.5 ng/mL (lowest cali-

brator concentration used) or as the concentration where

the percent relative errors and %CVs were less than 20%

for five replicate injections.

As shown in Figure 3, most calibration curves were

fit using linear regression. Some standards (for example,

cocaine) yielded better statistical calibration curves using

quadratic regression. In these select cases, the target

compound used a structurally different isotopically labeled

internal standard (for example, cocaine used D5-nor-

diazepam as internal standard).

The assay of biological sample extracts identified

multiple drugs of abuse and related metabolites. Figures

4A and B demonstrate examples of urine and whole blood

extracts assayed for the presence of illicit drugs with the

developed LC-MS/MS method. Note that cocaine and

benzoylecgonine were detected and qualified below the

assay LOQs in a whole blood extract (Figure 4B), indi-

cating that lower LOQs are possible for these compounds.

Conclusion

An LC-MS/MS method for assaying illicit drugs and their

metabolites at an LOQ of 0.5–2.5 ng/mL in biological

fluids for forensic use has been demonstrated.

Confirmation of the drugs of abuse was achieved by moni-

toring two SRM transitions per compound and measuring

their area ratios to within ±20%. Utilizing a low SRM

dwell time of 10 ms per transition to achieve sufficient

data points across a chromatographic peak had no adverse

effects, such as SRM cross-talk, on the quantitation and

confirmation of these illicit drugs. To authenticate this

assay, extracts from biological fluids were analyzed,

showing the presence of several drugs of abuse and their

metabolites.

:TR

00.01 - 08.5

:MS

G5

6

8

01

)nim( emiT

05

001

05

001

05

001

0

05

001

05

001

05

001

05

001

05

001

277529 :AA

419593 :AA

9980771 :AA

462729 :AA

565898 :AA

0865799 :AA

0848161 :AA

98302051 :AA

:TR

00.5 - 00.1

:MS

G5

1

2

3 4

5

)nim( emiT

05

001

05

001

05

001

05

001

05

001

RelativeAbundance

05

001

05

001

05

001

893932 :AA

3130902 :AA

054287 :AA

751529 :AA

3782942 :AA

970633 :AA

868769 :AA

821183 :AA

:TR

08.7 - 08.3

:MS

G5

4 5

6

7

)nim( emiT

05

001

05

001

05

001

05

001

05

001

05

001

05

001

05

001

6352445 :AA

3488331 :AA

7373258 :AA

2039544 :AA

5804561 :AA

648246 :AA

4259048 :AA

874866 :AA

:TR

03.6 - 03.2

:MS

G5

3 4

5

6

)nim( emiT

05

001

05

001

05

001

05

001

05

001

05

001

05

001

05

001

0152373 :AA

917698 :AA

303059 :AA

829559 :AA

2512231 :AA

8947312 :AA

8547522 :AA

3451031 :AA

5.2 = QOL

5.0 =QOL

A

0.1 = QOL

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

R

S

T

U

V

W

X

FF

Y

Z

BB

EE

CC

DD

AA

0.1 = QOL

0.1 = QOL

0.1 = QOL

0.1 = QOL

0.1 = QOL

0.1 = QOL

0.1 = QOL

5.0 = QOL

5.2 = QOL

5.2 = QOL

0.1 = QOL

0.1 = QOL

0.1 = QOL

5.0 = QOL

5.0 = QOL

0.1 = QOL

0.1 = QOL

5.0 = QOL

5.0 = QOL

5.0 = QOL

5.0 = QOL

5.0 = QOL

0.1 = QOL

0.1 = QOL

0.1 = QOL

0.1 = QOL

0.1 = QOL

5.0 =QOL

5.0 =QOL

Figure 1: Quantifier SRM transitions for the 2.5 ng/mL standard. For the compound designators, refer to the legend in Table 1.