Open Collections

UBC Faculty Research and Publications

Chasing the dragon - characterizing cases of leukoencephalopathy associated with heroin inhalation in… Buxton, Jane A; Sebastian, Renee; Clearsky, Lorne; Angus, Natalie; Shah, Lena; Lem, Marcus; Spacey, Sian D Jan 21, 2011

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12954_2009_Article_201.pdf [ 280.65kB ]
JSON: 52383-1.0223908.json
JSON-LD: 52383-1.0223908-ld.json
RDF/XML (Pretty): 52383-1.0223908-rdf.xml
RDF/JSON: 52383-1.0223908-rdf.json
Turtle: 52383-1.0223908-turtle.txt
N-Triples: 52383-1.0223908-rdf-ntriples.txt
Original Record: 52383-1.0223908-source.json
Full Text

Full Text

RESEARCH Open AccessChasing the dragon - characterizing cases ofleukoencephalopathy associated with heroininhalation in British ColumbiaJane A Buxton1,2*, Renee Sebastian1,3, Lorne Clearsky2, Natalie Angus1,4, Lena Shah1,3, Marcus Lem2,Sian D Spacey5AbstractAn association between leukoencephalopathy, a disease of the white matter of the brain, and smoking heroin iswell recognized. This paper describes 27 cases of leukoencephalopathy identified in two cities in British Columbia,Canada 2001-2006; the largest number of geographically and temporally defined reported cases in North America.Twenty cases of leukoencephalopathy were identified in and around Vancouver with onset dates December 2001to July 2003; seven further cases were identified in Victoria September 2005-August 2006. Twenty (74%) of all caseswere male, two couples were reported and eleven cases (55%) had Asian ethnicity. One case reported smokingheroin on a single occasion and developed mild symptoms; all other cases were hospitalized. Thirteen (48%) casesdied; all had smoked heroin for a minimum of 3 years. Testing of one available heroin sample identified nosubstance other than common cutting agents.Although a specific etiology was not identified our study supports the theory of an intermittent exposure to atoxic agent added to the heroin or a combustion by-product. It also suggests a dose response effect rather thangenetic predisposition. Collaboration with public health, health professionals, law enforcement and persons whouse illegal drugs, will facilitate the early identification of cases to enable timely and complete follow-up includingobtaining samples. Testing of implicated heroin samples may allow identification of the contaminant and thereforeprevent further cases. It is therefore important to ensure key stakeholders are aware of our findings.IntroductionLeukoencephalopathy refers to disease of the white mat-ter of the brain and therefore can involve motor, sen-sory, and visual systems. Leukoencephalopathy can alsodisrupt cognitive and emotional function. There aremany etiologies of leukoencephalopathy, includinggenetic disorders, cerebrovascular disease, eclampsia andtoxic exposures. The clinical manifestation of the diseaseis a reflection of the areas of the brain involved in thedisease process. Clinical features range from inattention,forgetfulness and personality changes, to dysarthria,ataxia, dementia, coma and death [1]. Toluene, ethanol,cocaine, methylenedioxymethamphetamine (MDMA or“ecstasy”) and heroin have all been associated with toxicleukoencephalopathies [1]. Toxic exposure from heroininduced leukoencephalopthy typically involves the occi-pital lobes and cerebellum bilaterally, characteristic sym-metric patterns can be seen on neuroimaging [2-5].An association between leukoencephalopathy andsmoking heroin has been recognized for over 25 years,although the exact pathogenesis is still not well under-stood. In ‘chasing the dragon’, heroin is placed on apiece of aluminium foil, heated with a flame frombelow, and the resulting vapour (pyrolysate) is inhaledwith a straw or other tube-like structure. The practicewas first recognized in Hong Kong in the 1950’s but hasnow spread to users worldwide [6].Although the practice of ‘chasing the dragon’ is notuncommon, the associated leukoencephalopathy hasbeen rarely reported. The first and largest outbreak ofleukoencephalopathy linked to chasing the dragon wasreported 1982 in the Netherlands, and included 47* Correspondence: jane.buxton@bccdc.ca1Epidemiology Services, British Columbia Centre for Disease Control,Vancouver, BC, CanadaFull list of author information is available at the end of the articleBuxton et al. Harm Reduction Journal 2011, 8:3© 2011 Buxton et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative CommonsAttribution License (, which permits unrestricted use, distribution, and reproduction inany medium, provided the original work is properly cited.cases, 11 (23%) of whom died [6]. Since 1982, sporadiccases and small case clusters (ranging from 1-4 cases)have been report ed in Taiwan [7,8], Hong Kong [9],other European countries [4,5,10-13], the United States[3,14,15] and Canada [16]. Three larger clusters havebeen identified in China [17-20]. The primary hypothesisis that leukoencephalopathy is caused by a contaminantin the heroin or a combustion by-product [6,21,22].Despite multiple attempts to identify a contaminant inheroin samples; no causative agent has yet beenidentified.Between December 2001 and July 2003, 20 cases ofleukoencephalopathy linked to heroin inhalation wereidentified in and around Vancouver, British Columbia(BC) Canada. An investigation was initiated at the time;but no causative agent was identified. Case reports werecreated to describe the clinical, pathologic, and imagingfindings and awareness campaigns were initiated.In the fall of 2005, further cases of leukoencephalopa-thy were reported in Victoria, BC on Vancouver Island.An investigation was initiated in order to describe theepidemiology of the new cases. The purpose of thispaper is to characterize all cases of heroin associatedleukoencephalopathy identified since 2001 to date in BCin order to guide future research and public healthactions.MethodsA case of heroin associated leukoencephalopathy wasdefined as:A person with clinical features of toxic leukoencepha-lopathy +/- neuroimaging with white matter changestypical of heroin-associated leukoencephalopathyAND a history of chasing the dragonAND a resident of BC or reported obtaining heroinin BCA case report form (which collected demographic,drug use, and clinical information) and a fact sheet forphysicians and the public were developed. The Provin-cial Health Officer notified BC neurologists about thecases of leukoencephalopathy associated with ‘chasingthe dragon’ through the provincial specialty society listserve. Medical Health Officers throughout the provincewere also informed; they in turn notified the hospitalEmergency Department heads and family physicians intheir health regions. All physicians were requested tonotify cases to local public health, which ensured thecompleted case report form was sent to the BC Centrefor Disease Control.Where possible, charts of cases were reviewed andinterviews were conducted with the case or next of kin.Abstracted information included: sex, date of birth,residence and ethnicity, medical, social and drug his-tories, date of symptom onset, date of hospital admis-sion, clinical course and outcome. Cases were asked if aheroin sample was available for testing. A process wasarranged for samples to be transported by the policeto the Health Canada Drug Analysis Laboratory forcontent analysis using liquid chromatography-massspectrometry.ResultsIn addition to 20 cases identified in and around Van-couver with onset dates between December 2001 - July2003, a further 7 cases were identified in Victoria withonset between September 2005- August 2006 (Figure 1).All cases, except one, were hospitalized. Thirteen (48%)died and 20 (74%) were male (Table 1). Two heterosex-ual case-couples were reported; one pair in Vancouverand one in Victoria. Eleven Vancouver cases (55%)reported Asian ethnicity; while all Victoria cases wereCaucasian.Patients typically presented with symptoms of cerebel-lar dysfunction such as ataxia, and all cases reported dif-ficulty with speech. The clinical and imaging findings ofthree of the Vancouver cases are presented elsewhere[2]. The date of death was available for seven of thedeceased, for these the median time between symptomonset and death was 54 days (range: 16-408 days).Drug histories were obtained for 18 (67%) cases. Onecase that reported smoking on a single occasion devel-oped mild symptoms and was not hospitalized. Exclud-ing this case the mean duration of chasing the dragonwas 9.5 years (range 0.5 - 30 years). Of the cases thatdied, the minimum time of smoking heroin was threeyears. Use of other illicit drugs was reported in bothVancouver and Victoria cases (cocaine, marijuana,ecstasy, and crystal methamphetamine). Three (43%) ofthe Victoria cases reported that smoking heroin wastheir only type of illicit drug use. At least six (86%) ofthe Victoria cases were taking methadone at the time ofsymptom onset; and five (25%) Vancouver casesreported using methadone, although these data areincomplete.Drug supply information was available for four (57%)of the Victoria cases; all reported obtaining herointhrough telephone order and home delivery (dial-a-dope) from a male Asian supplier. The first threeVancouver cases also reported being supplied byan Asian supplier through dial-a-dope but data areincomplete for the rest of Vancouver cases.Three (47%) of the Victoria cases reported commercialpainting as an occupation. No associations were foundwith type of aluminium foil (commonly purchased at thelocal supermarket); or with other underlying conditionsor medication other than methadone. No differenceBuxton et al. Harm Reduction Journal 2011, 8:3 2 of 5in colour, texture or smell of the heroin was reported bythe cases. Apart from the two couples no cases reportedknowing anyone else with similar symptoms. One heroinsample from a Vancouver case was tested with only com-mon cutting agents identified.DiscussionWe have characterized 27 cases of leukoencephalopathyassociated with ‘chasing the dragon’. This is the largestnumber of cases reported in North America which aretemporally and geographically defined. Although anetiologic agent has not been identified, we have a betterunderstanding of the population at risk. The age, sexand ethnicity of our cases are consistent with the demo-graphic profiles in other published reports. Forty-onepercent of BC cases were Asian; this preponderance,also found in other studies, is likely representative ofpersons who ‘chase the dragon’ [6].Identification of two heterosexual case-couples sug-gests that the risk factors for leukoencephalopathy aremore likely to be substance related rather than due togenetic predisposition. The distribution of the cases inplace and time suggests a common intermittent expo-sure. Substances added to the heroin may be an inert‘cutting agent’ such as caffeine, lactose or mannitol toincrease the volume and hence profit, or an ‘adulterant’,012345Dec MarJunSepDec MarJunSepDec MarJunSepDec MarJunSepDec MarJunYear and month of symptom onsetNumber of casesVictoria casesn=7Vancouver cases n=202001 2002 2006200520042003Figure 1 Leukoencephalopathy cases associated with chasing the dragon in British Columbia, 2001-2006 by symptom onset date(n = 27).Table 1 Demographic and outcome variables among leukoencephalopathy cases in BC (n = 27)Variable Vancouver (n = 20) Victoria (n = 7) All cases (n = 27) Vancouver vs. VictoriaMean Range Mean Range Mean Range P-value†Age (years) 36 32 - 42 32 21 - 51 33 21 - 51 0.220N % N % N % P-value‡Sex: Male 15 75 5 71 20 74 1.000Ethnicity: Asian 11 55 0 0 11 41 0.022*Hospitalized 19 95 7 100 26 96 1.000Deceased 10 50 3 43 13 48 1.000† P-value determined using t-test.‡P-value determined using Fisher’s exact test.*Statistically significant difference (p < 0.05).Buxton et al. Harm Reduction Journal 2011, 8:3 3 of 5which is added for its pharmacological effect [23]. Thedial-a-dope delivery system identified by some casesmay involve additional persons in the supply and deliv-ery chain (i.e., from the dealer to the deliverer) andincrease the risk of contaminants being added, either forprofit or to retain some drug for personal use. It is unli-kely that the contaminant is added to cause intentionalharm as it is in the best interest of the dealer to main-tain his/her client base[24].According to a recent report from the UN Office ofthe Drug Commission, 96% of heroin seizures (2002-2007) in the US originated from Mexico and Columbia;whereas 98% of heroin seized in Canada originated fromSouthwest Asia [25]. Although the source of heroin dif-fers between Canada and US, and the epidemiology andprevalence of ‘chasing the dragon’ in BC is poorlyunderstood, the incidence and risk of the resultant leu-koencephalopathy is clearly low. We believe this indi-cates the contaminant is likely added close to the finaldelivery stage, rather than at the original source. How-ever there is likely under-reporting as physicians arerequired to actively report the condition to publichealth, some cases may have been mild and sponta-neously recovered and others attributed to otheretiologies.The purity of street heroin in BC, determined byHealth Canada, Drug Analysis Service, has increasedfrom 5-10% in the 1970’s, to greater than 60% [26].However, heroin used for smoking is usually 30% to40% pure as higher grade cuts char too quickly for effec-tive smoking. Heroin is also reported to be increasinglyavailable in the base form which is not amenable forinjection [27]. Smoking heroin in North America wasbecoming established prior to the knowledge of therisks of HIV associated with injecting [28]. Gossop et alfound that chasing the dragon was a well-establishedmethod of using heroin in certain populations, notmerely a pre-injection phase of heroin addiction [29].Therefore, drug availability, attitudes to using needles,stigma and the potential of disease transmission relatedto injection drug use may have led to increased smokingrather than injection of heroin. This illustrates thepotential for further cases to occur.One case with a single brief exposure to inhaling her-oin pyrolysate required outpatient support only. Thisfinding was similar to a case report in the literature of apatient with an isolated exposure who had a completerecovery, and is consistent with a dose-response rela-tionship [3,12]. Clarity of quantity and purity of heroinused by cases would allow a better understanding of adose-response relationship.Reported clusters of leukoencephalopathy have beenassociated with smoking heroin. However isolated casesof leukoencephalopathy associated with using heroinand heroin and cocaine intravenously [30-32], and onecase of leukoencephalopathy associated with heroiningestion occurring in a 2 year old child have beenreported [33]. The development of disease associatedwith other routes of administration highlights the lackof knowledge about the etiological agent and the impor-tance of determining its identity.The etiology of heroin-related toxic leukoencephalo-pathy requires further research and public health invol-vement. The severity of the outcome and lack ofcurative treatment highlights the importance of futureinvestigations. Current therapy with coenzyme Q andvitamin supplements is anecdotal only [34]. Previous lit-erature has been mostly published in neurology andradiology journals as clinical case reports; isolated casesmake it difficult to determine risk factors for this condi-tion. Research into the prevalence of ‘chasing the dra-gon’ will help determine the potential risk for furtheroutbreaks and may indicate a need to modify both edu-cational, treatment and support services for this groupof heroin users. Although a specific etiology has notbeen identified, a toxic agent added to the heroin, or acombustion by-product, remain the leading theories[6,21,22].Limitations of our study include incomplete case anddrug information. This may be related to the inability ofcases to communicate at presentation and the rapiddecline in mental state of some. Also, the illicit natureof drug use may cause concerns about sharing informa-tion regarding other drug users or the source of heroin.Family members may have little prior knowledge of thecase’s drug use, limiting the information providedthrough collateral history. The delay between heroin useand symptom onset also reduces the likelihood thatimplicated heroin is available for testing.Recognizing the difficulties inherent in studying such asporadically occurring condition, a serious effort todetermine etiology may require both reactive and pro-spective approaches. Collaboration with public health,health professionals, law enforcement and persons whouse illegal drugs, would facilitate the early identificationof cases to enable timely and complete follow-up includ-ing obtaining heroin samples. A pre-arranged processfor transporting and testing implicated heroin samplesmay allow identification of the contaminant and there-fore prevent further cases. Purity and contaminantsampling programs for street drugs could also be con-sidered. It is therefore important to ensure key stake-holders are aware of these findings and the associationof leukoencephalopathy and heroin smoking.AcknowledgementsThe authors would like to thank Dr. Mark Gilbert for his assistance and inputin the follow-up of the Vancouver Island cases.Buxton et al. Harm Reduction Journal 2011, 8:3 4 of 5Author details1Epidemiology Services, British Columbia Centre for Disease Control,Vancouver, BC, Canada. 2School of Population and Public Health, Universityof British Columbia, Vancouver, BC, Canada. 3Canadian Field EpidemiologyProgram, Public Health Agency of Canada, Ottawa, ON. 4School of PublicHealth University of Saskatchewan, SK, Canada. 5Division of Neurology, Dept.of Medicine, University of British Columbia, Vancouver, BC, Canada.Authors’ contributionsJAB oversaw and gave input at all levels of the investigations. ML and LCfollowed up the cases in Vancouver. RS and LS followed up the cases inVictoria. LC wrote a report on Vancouver cases. RS wrote the first draft ofthe paper. NA performed an updated literature review and edited the paper.SDS provided clinical expertise. All authors have read, given input andapproved the final manuscript.Competing interestsThe authors declare that they have no competing interests.Received: 11 December 2009 Accepted: 21 January 2011Published: 21 January 2011References1. Filley CM, Kleinschmidt-DeMasters BK: Toxic leukoencephalopathy. N Engl JMed 2001, 345:425-432.2. Keogh CF, Andrews GT, Spacey SD, Forkheim KE, Graeb DA: Neuroimagingfeatures of heroin inhalation toxicity: “chasing the dragon”. AJR Am JRoentgenol 2003, 180:847-850.3. Kriegstein AR, Armitage BA, Kim PY: Heroin inhalation and progressivespongiform leukoencephalopathy. N Engl J Med 1997, 336:589-590.4. Tan TP, Algra PR, Valk J, Wolters EC: Toxic leukoencephalopathy afterinhalation of poisoned heroin: MR findings. AJNR Am J Neuroradiol 1994,15:175-178.5. Weber W, Henkes H, Moller P, Bade K, Kuhne D: Toxic spongiformleucoencephalopathy after inhaling heroin vapour. Eur Radiol 1998,8:749-755.6. Wolters EC: “Heroin” leukoencephalopathy. Ned Tijdschr Geneeskd 1982,126:111-112.7. Chen CY, Lee KW, Lee CC, Chin SC, Chung HW, Zimmerman RA: Heroin-induced spongiform leukoencephalopathy: value of diffusion MRimaging. J Comput Assist Tomogr 2000, 24:735-737.8. Chang YJ, Tsai CH, Chen CJ: Leukoencephalopathy after inhalation ofheroin vapor. J Formos Med Assoc 1997, 96:758-760.9. Au-Yeung K, Lai C: Toxic leucoencephalopathy after heroin inhalation.Australas Radiol 2002, 46:306-308.10. Ryan A, Molloy FM, Farrell MA, Hutchinson M: Fatal toxicleukoencephalopathy: clinical, radiological, and necropsy findings in twopatients. J Neurol Neurosurg Psychiatry 2005, 76:1014-1016.11. Gacouin A, Lavoue S, Signouret T, Person A, Dinard MD, Shpak N,Thomas R: Reversible spongiform leucoencephalopathy after inhalationof heated heroin. Intensive Care Med 2003, 29:1012-1015.12. Vella S, Kreis R, Lovblad KO, Steinlin M: Acute leukoencephalopathy afterinhalation of a single dose of heroin. Neuropediatrics 2003, 34:100-104.13. Celius EG, Andersson S: Leucoencephalopathy after inhalation of heroin:a case report. J Neurol Neurosurg Psychiatry 1996, 60:694-695.14. Halloran O, Ifthikharuddin S, Samkoff L: Leukoencephalopathy from“chasing the dragon”. Neurology 2005, 64:1755.15. Long H, Deore K, Hoffman RS, Nelson LS: A fatal case of spongiformleukoencephalopathy linked to “chasing the dragon”. J Toxicol ClinToxicol 2003, 41:887-891.16. Hill MD, Cooper PW, Perry JR: Chasing the dragon–neurological toxicityassociated with inhalation of heroin vapour: case report. CMAJ 2000,162:236-238.17. Zhou L, Lu BX, Yin J, Luo YF, Wang Q, Liu XJ: [Glucocortioid treatment forheroin-induced spongiform leucoencephalopathy: a clinical controlledstudy]. Di Yi Jun Yi Da Xue Xue Bao 2003, 23:172-174.18. Wang Q, Lu BX: Single photon emission computerized tomography ofspongiform leukoencephalopathy heroin addicts: analysis of 10 cases. DiYi Jun Yi Da Xue Xue Bao 2002, 22:659-660.19. Lu B, Zhou L, Pan S: Clinical and pathological characteristics of heroinspongiform leukoencephalopathy in China. Zhonghua Nei Ke Za Zhi 2001,40:753-756.20. Zheng W, Zhang X: Characteristics of spongiform leukoencephalopathyinduced by heroin: MRI detection. Chin Med J (Engl) 2001, 114:1193-1195.21. Neiman J, Haapaniemi HM, Hillbom M: Neurological complications of drugabuse: pathophysiological mechanisms. Eur J Neurol 2000, 7:595-606.22. Brenneisen R, Hasler F: GC/MS determination of pyrolysis products fromdiacetylmorphine and adulterants of street heroin samples. J Forensic Sci2002, 47:885-888.23. Brunt TM, Rigter S, Hoek J, Vogels N, van Dijk P, Niesink RJ: An analysis ofcocaine powder in the Netherlands: content and health hazards due toadulterants. Addiction 2009, 104:798-805.24. Coomber R: Dangerous drug adulteration - an international survey ofdrug dealers using the internet and the world wide web (WWW).International Journal of Drug Policy 1997, 8:18-28.25. United Nations Office on Drugs and Crime. World Drug Report 2009Vienna Austria; 2009 [].26. Buxton JA: Vancouver drug use epidemiology. Site Report for the CanadianCommunity Epidemiology Network on Drug Use. Vancouver 2005 [].27. Huizer H: Analytical studies on illicit heroin. V. Efficacy of volatilizationduring heroin smoking. Pharm Weekbl Sci 1987, 9:203-211.28. Griffiths P, Gossop M, Strang J: Chasing the dragon: the development ofheroin smoking in the United Kingdom. Heroin Addiction and Drug Policy:The British System Oxford: Oxford Universtiy Press; 1997.29. Gossop M, Griffiths P, Strang J: Chasing the dragon: characteristics ofheroin chasers. Br J Addict 1988, 83:1159-1162.30. Rizzuto N, Morbin M, Ferrari S, Cavallaro T, Sparaco M, Boso G, Gaetti L:Delayed spongiform leukoencephalopathy after heroin abuse. ActaNeuropathol 1997, 94:87-90.31. Barnett MH, Miller LA, Reddel SW, Davies L: Reversible delayedleukoencephalopathy following intravenous heroin overdose. J ClinNeurosci 2001, 8:165-167.32. Maschke M, Fehlings T, Kastrup O, Wilhelm HW, Leonhardt G: Toxicleukoencephalopathy after intravenous consumption of heroin andcocaine with unexpected clinical recovery. J Neurol 1999, 246:850-851.33. Perez ER, Maeder P, Riovier L, Deonna T: Toxic leukoencephalopathy afterheroin ingestion in a 2.5 year old child [letter]. Lancet 1992, 340:729.34. Kriegstein AR, Shungu DC, Millar WS, Armitage BA, Brust JC, Chillrud S,Goldman J, Lynch T: Leukoencephalopathy and raised brain lactate fromheroin vapor inhalation ("chasing the dragon”). Neurology 1999,53:1765-1773.doi:10.1186/1477-7517-8-3Cite this article as: Buxton et al.: Chasing the dragon - characterizingcases of leukoencephalopathy associated with heroin inhalation inBritish Columbia. Harm Reduction Journal 2011 8:3.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at et al. Harm Reduction Journal 2011, 8:3 5 of 5


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items