Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
From the Centers for Disease Control and Prevention
February 10, 2010

Agranulocytosis Associated With Cocaine Use—Four States, March 2008–November 2009

JAMA. 2010;303(6):504-512. doi:

MMWR. 2009;58:1381-1385

1 table omitted

In April 2008, a clinical reference laboratory in New Mexico notified the New Mexico Department of Health (NMDOH) of a cluster of unexplained agranulocytosis cases confirmed by bone marrow histopathology during the preceding 2 months. NMDOH began an investigation, which identified cocaine use as a common exposure in 11 cases of otherwise unexplained agranulocytosis during April 2008–November 2009. In the midst of the NMDOH investigation, in November 2008, public health officials in British Columbia and Alberta, Canada, reported detecting levamisole (an antihelminthic drug used mainly in veterinary medicine and a known cause of agranulocytosis1) from clinical specimens and drug paraphernalia of cocaine users with agranulocytosis. In January 2009, NMDOH posted a notification of its findings on CDC's Epidemic Information Exchange (Epi-X) and notified poison control centers. In a separate investigation during April-November 2009, public health officials in Seattle, Washington, identified 10 cases of agranulocytosis among persons with a history of cocaine use. Of the 21 cases, levamisole was detected from clinical specimens in four of the five patients tested. According to the Drug Enforcement Administration (DEA), as of July 2009, 69% of seized cocaine lots coming into the United States contained levamisole as an added agent. This report summarizes the investigations in New Mexico and Washington, which suggested that levamisole in cocaine was the likely cause of the agranulocytosis. Health-care providers should consider these findings in the differential diagnosis of agranulocytosis, and public health officials should be aware of cases of agranulocytosis associated with cocaine use.

New Mexico Investigation

After learning of the unexplained agranulocytosis in April 2008, NMDOH investigated the cases through medical record reviews and interviews with health-care providers. Four of the six patients had been undergoing treatments that were thought to have caused agranulocytosis (i.e., cancer treatment, gabapentin, sulfasalazine, and an unidentified herbal remedy obtained outside of the country). The remaining two patients (patients 1 and 2) had no known cause, although both patients were linked to illicit drug use (marijuana and cocaine for patient 1; heroin, and later, cocaine for patient 2). During the next 8 months, passive surveillance for additional cases resulted in seven additional cases of agranulocytosis reported to NMDOH, six from the same laboratory that sent the original alert to NMDOH, and one decedent (patient 3) from the New Mexico Office of the Medical Investigator. The seven additional cases included one Arizona resident examined in a New Mexico hospital (patient 9) and another (patient 10), whose bone marrow specimen was referred from Colorado.

To further investigate possible common exposures for patients with unexplained agranulocytosis, in June 2008 NMDOH developed a standardized questionnaire to include questions about illicit drug use and known causes of agranulocytosis. NMDOH conducted medical record reviews, physician interviews, and patient interviews for all patients with unexplained agranulocytosis reported to NMDOH. Of the 13 cases reported by January 2009, nine were deemed unexplained, and seven of these patients reported a history of cocaine use.

A review of the scientific literature revealed no reports of agranulocytosis associated with cocaine use. However, in November 2008, NMDOH investigators learned that levamisole* had been isolated from clinical specimens and drug paraphernalia of five cocaine-using patients with agranulocytosis in British Columbia and Alberta, Canada. Although levamisole had been isolated previously from cocaine, cocaine paraphernalia, and persons who used cocaine,24 agranulocytosis had not been associated previously with cocaine use. At the same time, the NMDOH Scientific Laboratory Division (SLD) reported that several unrelated specimens submitted for routine toxicology screening were positive for both cocaine and levamisole.

In January 2009, NMDOH SLD detected levamisole using gas chromatography/mass spectrophotometry (GC/MS) in a postmortem blood specimen from patient 3, who had a diagnosis of Serratia marcescans sepsis and agranulocytosis. The specimen had been collected in March 2008 and preserved as part of an investigation by the New Mexico Office of the Medical Investigator. The patient had been admitted to the hospital 5 months before death with a diagnosis of agranulocytosis and an absolute neutrophil count (ANC) of zero. No testing of the other cocaine-exposed patients for levamisole was conducted because levamisole has a half life of approximately 5 hours and was unlikely to be detected in blood or urine beyond 48 hours after the last exposure.5 The rest of the specimens from the seven patients with a history of cocaine use had been collected more than 48 hours after the last cocaine exposure.

On January 16, 2009, NMDOH issued a press release and notified health-care providers through the New Mexico Health Alert Network about the potential for agranulocytosis resulting from inadvertent levamisole exposure during cocaine use. Health-care providers were asked to report cases of unexplained agranulocytosis. One week later, NMDOH released the same information nationally through CDC's Epi-X and poison control centers. This action generated a report of one additional case (patient 10) in a cocaine user from Colorado, reported to NMDOH on February 28, 2009. A urine specimen from this patient was sent to NMDOH SLD, where levamisole was identified using GC/MS. Colorado law enforcement also detected levamisole using GS/MS in residue from the crack cocaine pipe that the patient submitted voluntarily. Since February 2009, three additional cases (patients 6, 7, and 8) have been detected in New Mexico. Levamisole testing was not conducted in any of these three patients because they were examined in the hospital >48 hours after last cocaine exposure. In total, 11 cases of agranulocytosis had been associated with cocaine use through the NMDOH investigation as of November 2009.

Washington Investigation

In April 2009, epidemiologists at Public Health – Seattle & King County (PHSKC) noted a published report from Canada describing agranulocytosis and infections in five users of cocaine contaminated with levamisole,6 and issued an alert to clinicians. Simultaneously, PHSKC received a report of three persons previously hospitalized with agranulocytosis (patients 12, 13, and 14) among persons with a history of cocaine use and initiated an investigation. A second PHSKC alert to local health-care providers and press release at the beginning of June 2009 generated five additional reports. As of November 2009, a total of 10 cases had been investigated in conjunction with the Washington State Department of Health.

As of November 2009, a total of 21 cases of cocaine-associated agranulocytosis had been investigated by NMDOH and PHSKC. Thirteen patients were women. The mean age was 42 years (range: 24-58 years). Five patients were whites, three were blacks, five were American Indian/Alaska Natives, three were Hispanics, and five were of unknown race/ethnicity. Both powder and crack cocaine use has been reported by these patients. Seven patients had at least one documented recurrence of agranulocytosis after repeated cocaine use, and eight patients had at least one documented incidence of agranulocytosis before they were reported to the health department. Of the 21 patients, five were tested by GC/MS for the presence of levamisole, and levamisole was isolated from four of the five patients.

Reported by:

M Brackney, MS, J Baumbach, MD, C Ewers, MSN, AL Martinez, J Hagan, MPH, New Mexico Dept of Health; D Czuchlewski, MD, K Foucar, MD, Univ of New Mexico Health Sciences Center; MH Fekrazad, MD, Univ of New Mexico Cancer Research and Treatment Center; SA Seifert, MD, New Mexico Poison and Drug Information Center; D Rimple, MD, Univ of New Mexico Hospital Dept of Emergency Medicine; KB Nolte, MD, Univ of New Mexico, Office of the Medical Investigator. JA Buchanan, MD, EJ, Lavonas, MD, Rocky Mountain Poison and Drug Center, Denver Health; C Nelson, MD, Colorado Dept of Public Health and Environment. RW Wood, MD, JS Duchin, MD, Public Health–Seattle & King County; J VanEenwyk, PhD, Washington State Dept of Health. N Reuter, Substance Abuse and Mental Health Svcs Admin; ML Ta, PhD, S Vagi, PhD, EIS officers, CDC.

CDC Editorial Note:

Agranulocytosis is an uncommon condition (7.2 cases per 1 million population per year, excluding patients with cancer and patients receiving cytotoxic drugs)7 that carries a risk for opportunistic infections and can be fatal in approximately 7%-10% of cases.8 Known causes include pharmaceutical drugs, toxins, ionizing radiation, autoimmune and genetic disorders, certain infections, and neoplasms.7 This report presents 21 cases of agranulocytosis for which, aside from cocaine exposure, no other common exposure was identified. Cocaine exposure has not been associated previously with agranulocytosis and, therefore, by itself, is not a likely cause of the agranulocytosis. However, agranulocytosis as a result of exposure to cocaine containing levamisole, a known cause of agranulocytosis, was reported recently in Canada.6 DEA has reported that, as of July 2009, 69% of the cocaine seized at U.S. borders contained levamisole, although the reason why levamisole is added to cocaine remains unclear. Levamisole also has been detected in cocaine obtained by law enforcement officers in New Mexico and Washington. These pieces of evidence suggest that exposure to levamisole through cocaine use was the likely cause of agranulocytosis in all 21 cases; however, surveillance and toxicologic data regarding additional cases are needed to better define a causal relationship.

Heroin use was reported in two of the 21 cases. DEA reported detecting levamisole in a handful of heroin seizures in 2008 but more frequently (<3%) in 2009 (DEA, unpublished data, 2009). Only trace amounts of levamisole have been detected in heroin, compared with an average concentration of approximately 10% detected in cocaine (DEA, unpublished data, 2009).

For multiple reasons, the 21 cases described in this report might represent a small portion of all agranulocytosis cases associated with cocaine (and potentially levamisole) in the United States. For example, agranulocytosis is not a reportable condition to health departments, patients might not disclose cocaine use to health-care providers, and patients who use cocaine might be less likely to seek health care.9 Agranulocytosis has been recognized as an idiosyncratic reaction to levamisole in 2.5%-13% of persons using levamisole for treatment of rheumatoid arthritis and in combined therapy for breast cancer.1 However, the proportion of cocaine users exposed to levamisole who might develop levamisole-induced agranulocytosis, is unknown.

Clinicians should be aware of the possible relationship between levamisole-associated agranulocytosis and use of cocaine, and possibly heroin, and should obtain a drug history in all potential cases routinely. Suspected cases should be reported to state or local health departments. Clinicians wishing to test patients for levamisole should have blood or urine collected promptly, because the likelihood of finding the drug decreases markedly after 48 hours.

CDC has begun national surveillance for agranulocytosis in association with suspected cocaine or heroin use, collecting information via medical abstraction form and patient interview. As of December 15, eight states had agreed to participate. The goals of surveillance are to characterize the extent of the problem, identify risk factors for exposure, and describe clinical presentation of patients with agranulocytosis associated with cocaine or heroin use. The Substance Abuse and Mental Health Services Administration is serving as a centralized source for disseminating relevant information regarding agranulocytosis associated with levamisole-contaminated cocaine. Additional information is available from Nicholas Reuter ( State and local health departments are encouraged to participate in the national surveillance effort and can report suspected cases to CDC at


This report is based, in part, on the contributions by J Buxton, P Kendall, L Knowles, D LeGatt, J Talbot, Canada; M Wilson, Denver Health Medical Center; N Shah, New Mexico Dept of Health; R Harruff, MD, PhD, King County Medical Examiner's Office; J Harlan, MD, Harborview Medical Center, and the DEA special testing and research laboratory.

What is already known on this topic?

In a recent report from Canada, agranulocytosis was associated with cocaine contaminated with levamisole.

What is added by this report?

Investigators from New Mexico and Washington identified an additional 21 cocaine users with unexplained agranulocytosis likely caused by exposure to levamisole.

What are the implications for public health practice?

Health-care providers should consider these findings in the differential diagnosis of agranulocytosis, and public health officials should be aware of cases of agranulocytosis associated with cocaine use.

*Levamisole is approved by the Food and Drug Administration as an adjuvant treatment for colon cancer and previously was used as an immunomodulator for various conditions. However, levamisole no longer is commonly used for these purposes. Today, levamisole primarily is used in veterinary practice as an antihelminthic agent.

Thompson JS, Herbick JM, Klassen LW,  et al.  Studies on levamisole-induced agranulocytosis.  Blood. 1980;56(3):388-396PubMed
Lintemoot J. ToxTalk. Levamisole: an unusual finding in a cocaine related fatality. Mesa, AZ: Society of Forensic Toxicologists; 2005. Available at Accessed December 15, 2009
Fucci N. Unusual adulterants in cocaine seized on Italian clandestine market.  Forensic Sci Int. 2007;172:2-3Article
Morley SR, Forest AR, Galloway JH. Levamisole as a contaminant in illicit cocaine. Proceedings of the International Association of Forensic Toxicologists (TIAFT) 44th International Meeting; Ljubljana, Slovenia; 2006. Available at Accessed December 15, 2009
Kouassi E, Caillé G, Léry L, Larivière L, Vézina M. Novel assay and pharmacokinetics of levamisole and p-hydroxylevamisole in human plasma and urine.  Biopharm Drug Dispos. 1986;7(1):71-89PubMedArticle
Zhu NY, LeGatt DF, Turner AR. Agranulocytosis after consumption of cocaine adulterated with levamisole [Clinical Observation].  Ann Intern Med. 2009;150(4):287-289PubMedArticle
Strom BL, Carson JL, Schinnar R, Snyder ES, Shaw M. Descriptive epidemiology of agranulocytosis.  Arch Intern Med. 1992;152(7):1475-1480PubMedArticle
Ibáñez L, Vidal X, Ballarín E, Laporte JR. Population-based drug-induced agranulocytosis.  Arch Intern Med. 2005;165(8):869-874PubMedArticle
Sterk CE, Theall KP, Elifson KW. Health care utilization among drug-using and non-drug-using women.  J Urban Health. 2002;79(4):586-599PubMedArticle