Levamisole, a veterinary antihelminthic, was recently recognized
as an adulterant in cocaine and is known to cause severe adverse
reactions in some cocaine users. Because of the health concerns
involving levamisole-adulterated cocaine, we developed a liquid
chromatography–tandem mass spectrometry (LC–MS–MS) method
for the detection of levamisole in urine.This method was used to
determine the prevalence of levamisole in cocaine-positive patient
samples. All cocaine-positive urine samples that were sent to the
San Francisco General Hospital Clinical Laboratory were tested for
levamisole for one month. For LC, an Agilent 1200 series was used
with a C18column and a gradient of mobile phase A (0.05% formic
acid) and B (acetonitrile/methanol). Detection was carried out
with an Applied Biosystems QTRAP®LC–MS–MS.The levamisole
LC–MS–MS method was linear over the range of 5–2500 ng/mL
(r > 0.996). Interassay and intraassay CVs were < 6%.The lower
limit of detection for levamisole was 0.5 ng/mL. Out of 949 total
urine drug screens, 20% were positive for benzoylecgonine, and of
those, 88% were positive for levamisole.The high prevalence of
levamisole-adulterated cocaine and potential toxicity in cocaine
users is a serious public health concern.These findings validate the
utility of an LC–MS–MS method for the detection of levamisole.
Cocaine is frequently diluted throughout the chain of dis-
tribution with myriad less expensive cutting agents and adul-
terants. These substances are mainly used to increase the
weight of the product, resulting in a higher profit. Some adul-
terants are added in attempts to potentiate or mimic the effects
of cocaine or, in some cases, attenuate the deleterious side ef-
fects associated with cocaine use. Since 2005, there have been
multiple reports of the presence of levamisole, a veterinary
antihelminthic, in seized cocaine (1–3). Levamisole was pre-
viously used in humans as adjuvant therapy in the treatment
of colorectal cancer and nephrotic syndrome. It was also used
in the treatment of rheumatoid arthritis because of its im-
munomodulary effects. It is no longer available for human use
in North America because of its potential serious adverse side
effects. There has been a series of case reports describing ad-
verse side effects associated with levamisole-adulterated co-
caine over the past year (4–9). These side effects include, but
are not limited to, agranulocytosis, ANCA-associated vasculitis,
and retiform purpura. Cases often present with symptoms and
laboratory results that resemble an autoimmune disorder.
According to the United States Drug Enforcement Adminis-
tration (DEA) and state testing laboratories, the percentage of
cocaine specimens containing levamisole has increased steadily
over the past eight years. Levamisole was indentified in 30% of
the cocaine seized by the DEA from July to September 2008. In
July 2009, that number increased to 70% (6). The Wayne
County Medical Examiner’s Office has reported that, in 2008,
levamisole was present in 36.6% of all samples that contained
cocaine and/or its metabolites; however, no information about
the analytical method was provided (10). To date, there are no
published liquid chromatography–mass spectrometry (LC–
MS) methods specific for the detection of levamisole in bio-
logical samples, and the prevalence of levamisole in cocaine-
positive samples has not been reported since 2008. Because of
the potential public health concern related to levamisole-adul-
terated cocaine, we developed an LC–MS–MS assay for
screening cocaine-positive urine samples. This method was
then used to determine the prevalence of levamisole in co-
caine-positive patient samples in San Francisco.
Materials and Methods
All cocaine-positive urine samples that were sent to the San
Francisco General Hospital Clinical Laboratory for drug
screening in October 2009 were tested for levamisole. Assay re-
sults were not revealed to patients or health care providers.
This work was approved by the University of California San
Francisco Institutional Review Board, which determined that
Detection of Levamisole Exposure in Cocaine Users by
Liquid Chromatography–Tandem Mass Spectrometry
Kara L. Lynch1,*, Stephen S. Dominy2, Jonathan Graf3, and Alexander H. Kral4
1Department of Laboratory Medicine, University of California San Francisco, San Francisco, California;2Department of Psychiatry,
University of California San Francisco, San Francisco, California;3Department of Medicine, University of California San Francisco,
San Francisco, California; and4Urban Health Program, RTI International, San Francisco, California
Reproduction (photocopying) of editorial content of this journal is prohibited without publisher’s permission.
Journal of AnalyticalToxicology,Vol. 35, April 2011
* Author to whom correspondence should be addressed. University of California San Francisco,
San Francisco General Hospital, 1001 Potrero Ave. NH2M16, San Francisco, CA 94110.
Journal of AnalyticalToxicology,Vol. 35, April 2011
patient consent was not necessary.
Cocaine screening was performed using the CEDIA®Co-
caine Assay (Microgenics, Fremont, CA). The assay was per-
formed on a Siemens Advia®1800 Chemistry System (Deer-
field, IL) following manufacturer’s instructions with a cutoff
value of 300 ng/mL. Confirmatory testing was conducted using
an LC–MS–MS method. For levamisole testing, urine was di-
luted 1:10 with 0.05% formic acid and spiked with aminorex,
the internal standard. All organic solvents and reagents were of
analytical grade and were purchased from Fisher Scientific
(Fair Lawn, NJ). The levamisole standard was purchased from
Sigma-Aldrich (St. Louis, MO), and aminorex was obtained
from Cerilliant (Round Rock, TX). For LC, an Agilent 1200
series was used with a Phenomenex (Torrance, CA) Kinetex™
C182.6-µm (50 × 2.1 mm) column, maintained at 25°C and a
gradient of mobile phase A (0.05% formic acid) and mobile
phase B (acetonitrile/methanol, 50:50, v/v). The flow rate was
400 µL/min, and the LC program was 0–0.5 min, 10% B; 0.5–
1.5 min, 10% to 40% B; 1.5–1.75 min, 40% B; 1.75–2.25 min,
40% to 10% B; 2.25–5 min, re-equilibration with 10% B. Lev-
amisole and the internal standard eluted at 1.1 and 1.0 min, re-
spectively (Figure 1A).
Detection was carried out with an Applied Biosystems
QTRAP LC–MS–MS system equipped with a TurboIon Spray™
ionization source, controlled by Analyst 1.5 software (Life
Technologies/Applied Biosystems, Foster City, CA). The ion
transitions used were 205.2/178.1 (levamisole) and 163.1/120.1
(internal standard). Positive ionization was performed, and
the following parameters were used: ion spray voltage, 5000 V;
curtain gas, 15 psi; ion source gas 1, 50 psi; ion source gas 2,
50 psi; CAD gas, low; and temperature, 600°C. The compound
dependent parameters for levamisole were as follows: declus-
tering potential 55 V, entrance potential 7 V, and collision en-
ergy 33 V. The data acquisition was performed using an infor-
mation-dependent acquisition (IDA) method. The precursor/
product ion transitions were first monitored by a selected re-
action monitoring survey scan. This was followed by the gen-
eration of mass spectra by way of a product ion scan in Q3,
functioning as a linear ion trap, when IDA criteria were met
(detection of a peak > 1000 cps). With every batch of samples,
a blank (drug-free urine), two quality control samples (20 and
200 ng/mL), and a calibration curve from 5 to 1000 ng/mL
were run. In addition to identification of a peak at the correct
retention time, a library search of the acquired product ion
spectra was performed and a match factor (purity) > 80% be-
tween the unknown and the levamisole library product ion
spectra was required to report levamisole as positive. Figure 1B
shows a representative product ion spectrum for levamisole.
Results and Discussion
The levamisole assay was linear from 5 to 2500 ng/mL (r >
0.996). The interassay imprecision was < 3% (n = 10) and the
intraassay imprecision was < 6% (n = 20) (controls at 20 and
200 ng/mL). The accuracy and recovery of controls at 20 and
200 ng/mL were > 95.8% based on the weigh-in value. Carry-
over was not detected at 30,000 ng/mL. Matrix effect experi-
ments were conducted as described previously (11). The ion
suppression was 19.7% (n = 20) for levamisole and was com-
pensated for by use of an internal standard. The limit of de-
tection, defined as the concentration that provides a signal-to-
noise ratio of 3, was 0.5 ng/mL.
There were a total of 949 urine drug screens ordered
through the San Francisco General Hospital Clinical Labora-
tory during the levamisole screen. The cocaine immunoassay,
which targets benzoylecgonine, was used to establish if these
patients were positive for cocaine. It was determined that 191
(20%) of all the urine drug screens were cocaine positive.
These results were confirmed by LC–MS–MS. Levamisole was
detected in 169 (88%) of the cocaine-positive samples. The
levamisole concentrations in these samples ranged from 5 to
32,720 ng/mL, with a mean and standard deviation of 1887 ±
4238 ng/mL. The corresponding benzoylecgonine concentra-
tions ranged from 4 to 588,000 ng/mL, with a mean and stan-
dard deviation of 44,451 ± 76,100 ng/mL. There were no sta-
tistically significant differences (p < 0.05) in the prevalence of
levamisole by any of the demographic or drug use variables
(Table I), including sex, race/ethnicity, age, or drug use.
This study identified that the prevalence of levamisole in
cocaine-positive patient samples at San Francisco General
Hospital was 88%. This suggests that the majority of cocaine
used in San Francisco is adulterated with levamisole. The half-
life of levamisole is on average 5.6 h and approximately 2–5%
Figure 1. Chromatogram and mass spectrum of levamisole in a cocaine-
positive urine sample.The extracted ion chromatogram of the survey scan
(inset: chemical structure of levamisole) (A) and the product ion spectra for
levamisole (B) are shown.
Journal of AnalyticalToxicology,Vol. 35, April 2011
is excreted unchanged in the urine (12,13). It is possible that
the prevalence of levamisole-adulterated cocaine is greater
than 88%. A percentage of the urine samples could have been
being within the window of detection for benzoylecgonine.
Further studies are required to determine the rate of lev-
amisole elimination in the urine. Some of the metabolites of
levamisole are not available for purchase commercially, making
it difficult to develop targeted analytical methods for the de-
tection of these compounds in addition to levamisole. One
limitation of this study is that levamisole is the l-isomer of
tetramisole, and the method described here does not distin-
guish between the l- and d-isomers of tetramisole. Therefore,
this study cannot rule out the possibility that the d-isomer of
tetramisole is also present in the patient samples. Another
limitation is that the sampling in this study was not conducted
at random; therefore, it is not possible to generalize these re-
This study suggests that clinicians should consider lev-
amisole-adulterated cocaine in users with unexplained atypical
symptoms, such as idiopathic agranulocytosis. Further studies
are required to determine why only a subset of cocaine users
exposed to levamisole develops adverse reactions that mimic an
autoimmune disorder. Some patients may be genetically pre-
disposed to developing these symptoms. Previous studies sug-
gest that levamisole-induced agranulocytosis may be associated
with the HLA-B27 genotype (7,14,15).
This is the first published study to date that determines the
prevalence and concentration of levamisole in cocaine-positive
patient samples in a hospital setting. These findings validate
the utility of an LC–MS–MS method for the detection of lev-
amisole. Given the high prevalence and underappreciated risks
associated with exposure to levamisole, this is a serious public
health concern in geographical regions with a high incidence
of cocaine use, such as San Francisco.
The authors would like to thank Ryk Sheppard for technical
1. A.M.Valentino and K. Fuentecilla. Levamisole: an analytical profile.
Microgram J. 3: 134–137 (2005).
2. N. Fucci. Unusual adulterants in cocaine seizure on Italian clandes-
tine market. Forensic Sci. Int. 172(2-3): e1 (2007).
3. J.F. Casale, E.M. Corbeil, and P.A. Hays. Identification of levamisole
impurities found in cocaine exhibits. http://www.justice.gov/dea/pro-
4. N.Y. Zhu, D.F. Legatt, and A.R. Turner. Agranulocytosis after con-
sumption of cocaine adulterated with levamisole. Ann. Intern. Med.
150: 287–289 (2009).
5. L. Knowles, J.A. Buxton, N. Skuridina, I. Achebe, D. Legatt, S. Fan,
N.Yan Zhu, and J. Talbot. Levamisole tinted cocaine causing severe
neutropenia in Alberta and British Columbia. Harm Reduct. J. 6: 30
6. Centers for Disease Control and Prevention (CDC). Agranulocytosis
associated with cocaine use—four states, March 2008–November
2009. Morb. Mortal. Wkly. Rep. 58: 1381–1385 (2009).
7. D.R. Czuchlewsk, M. Brackney, C. Ewers, J. Manna, M.H. Fekrazad,
A. Martinez, K.B. Nolte, B. Hjelle, I. Rabinowitz, B.R. Curtis, J.G. Mc-
Farland, J. Baumbach, and K. Foucar. Clinicopathologic features of
agranulocytosis in the setting of levamisole-tainted cocaine. Am. J.
Clin. Pathol. 133: 466–472 (2010).
8. J.M.Waller, J.D. Feramisco, L.Alberta-Wszolek,T.H. McCalmont, and
L.P. Fox. Cocaine-associated retiform purpura and neutropenia: is lev-
amisole the culprit? J.Am.Acad. Dermatol. 63(3): 530–535 (2010).
9. M. Bradford, B. Rosenberg, J. Moreno, and G. Dumyati. Bilateral
necrosis of earlobes and cheeks: another complication of cocaine
contaminated with levamisole. Ann. Intern. Med. 152: 758–759
10. L.P Raymond and D.S. Isenschmid.The possible role of levamisole in
illicit cocaine preparations. J.Anal. Toxicol. 33(9): 620–622 (2009).
11. B.K. Matuszewski, M.L. Constanzer, and C.M. Chavez-Eng. Strategies
for the assessment of matrix effect in quantitative bioanalytical
methods based on HPLC–MS/MS. Anal. Chem. 75: 3019–3030
12. E. Kouassi, G. Caille, L. Lery, L. Lariviere, and M. Vezina. Novel
assay and pharmacokinetics of levamisole and p-hydroxylevamisole
in human plasma and urine. Biopharm.Drug Dispos. 7: 71–89 (1986).
13. J.M. Reid, J.S. Kovach, M.J. O’Connell, P.G. Bagniewski, and
C.G. Moertel. Clinical and pharmacokinetic studies of high-dose
levamisole in combination with 5-fluorouracil in patients with ad-
vanced cancer. Cancer Chemother.Pharmacol. 41: 477–484 (1998).
14. J. Symoens, E.Veys, M. Mielants, and R. Pinals. Adverse reactions to
levamisole. Cancer Treat. Rep. 62: 1721–1730 (1978).
15. S.I. Drew and B.M. Carter. Granulocytotoxic autoantibodies in HLA-
B27 neutropenic patients. Transplantation 27: 436–437 (1979).
Manuscript received August 4, 2010;
revision received October 21, 2010.
Table I. Levamisole Screen Results by Demographic and
Drug Use Characteristics of Cocaine-Positive Patients at
San Francisco General Hospital (n = 191)
< 30 years
30 to 39 years
40 to 49 years
50 to 59 years
60 years or older
Drug Use Detected by Screening