UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)
UvA-DARE (Digital Academic Repository)
Prevalence morbidity and mortality among heroin users and methadone patients
Buster, M.C.A.
Publication date
2003
Link to publication
Citation for published version (APA):
Buster, M. C. A. (2003). Prevalence morbidity and mortality among heroin users and
methadone patients.
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.
0 0
G i e ll H . A . v a n B r u s s e l
DepartmentDepartment of Social and Mental Health, Municipal Health Service, Amsterdam.
M a r c e ll C A . B u s t e r
DepartmentDepartment of Social and Psychiatric Epidemiology, Municipal Health Service, Amsterdam
K a m b i zz N a s s e r l
DepartmentDepartment of Social and Psychiatrie Epidemiology, Municipal Health Service, Amsterdam
J a c q u e ss v a n L i m b e e k
DepartmentDepartment of Social and Psychiatric Epidemiology, Municipal Health Service, Amsterdam
H e n kk v a n O e u t e k o m
DepartmentDepartment of Tuberculosis, Municipal Health Service, Amsterdam
W i n tt v a n d e n B r i n k
P M A M A H U A M P P
N N CC I
D D
N N
O O
T U B E R C U L O S I SS A M O N G
D
R
U
GG
A
D
D
I
C
T
S I N
A
M
S
T
E
R
D
A
M M
M E T H A D O N EE P R O G R A M M E S
EuropeanEuropean Journal Of Public Health 199S; 4: 253-57
A b s t r a c t t
Thee objective was to determine the tuberculosis incidence and related factorss among drug users in 3 methadone maintenance programmes in Amsterdamm to see whether the present screening procedures are ad-equate.. The study population was selected from the Central Methadonee Register CMR. Cases were identified during hospitalization andd during routine screening of ambulatory methadone clients. The dataa were generated by drug and tuberculosis departments of the Amsterdamm Municipal Health Service MHS. From 5044 clients, 43 new casess were identified. The incidence of tuberculosis was 386 per 100100 000 person years py and increased with age and period of regis-trationn with the CMR. After correction for age, sex and period of reg-istration,, the incidence among clients of the general practitioners CP wass the lowest. The incidence rate among the general MHS clients is 9 timess higher and in the MHS programme for prostitutes and illegal for-eignerss it is 27 times higher than in the CP population. The increased riskk for drug users treated at the MHS indicates lifestyle as a risk fac-tor;; MHS clients are more involved in the drugscène than CP clients. On thee basis of this study, periodic screening limited to clients of the MHS wouldd seem to be adequate for the present.
I n t r o d u c t i o n n
Thee incidence of tuberculosis varies not only between countries but alsoo within countries. On the whole, the incidence is low in Western countries,, and high in many of the developing countries. However, this iss not the case in all Third World countries. In Morocco the incidence iss high: in 1990 there were 103 cases per 100000 inhabitants, com-paredd with Surinam with only 7.1 cases per 100 000 inhabitants dur-ingg the same year." In 1990 there were 9.2 tuberculosis cases in The
P M A M A H U A M P P
2 T u b e r c u l o s i s a m o n g o p l a t * u s e r s
Netherlandss and 10 cases in the USA, per 100 000 inhabitants. Until thee end of the 1980s, the number of new cases in both countries declinedd steadily, after which the decline came to a standstill in The Netherlandss and changed into an increase in the USA.
Thee observed decline in the USA was stronger among whites than amongg non-whites. The relative risk for non-white persons compared too white persons rose from 2.9 in 1953 to 5.3 in 1987.12 The incidence off tuberculosis among females is approximately half the incidence amongg males in both blacks and whites. In The Netherlands too, the incidencee among individuals with Dutch ethnicity is still declining
44 per 100000 in 1990-, whereas it is increasing slightly among foreign citizens,, in 1990, the Incidence among the non-Dutch part of the populationn was 87 per 100 000; this is almost 18 times the incidence amongg the Dutch.'3
Ann increase has been observed in subgroups of the Dutch as well.. In Amsterdam, the incidence among Dutch males aged 25-49 yearss rose from 16 per 100000 in 1984 to 35 per 100000 in 1990. Thee increase in this group is at least partly attributable to the HIV epi-demic.'** The elderly, prisoners, drug users, homeless and alcoholics aree particularly susceptible to tuberculosis. The vulnerability ties eitherr in a decline in physical condition causing tuberculosis reactiva-tion,, or in a lifestyle exposing them to new infection. Age, length of stayy in shelters and intravenous drug use appear to be related to activee tuberculosis among homeless people with physical ailments.IS In 19799 before the outbreak of the HfV epidemic Reichman et ai. stated thatt drug users were a new high-risk group for tuberculosis.'6 That timee the incidence in New York was 27 per 100 000 persons. In the municipall methadone maintenance programme, the incidence was almostt 20-fold higher, namely 510 cases per 100 000 persons.
Withh the HIV epidemic among drug users, the probability of activee tuberculosis increased as a result of an ineffective cellular immunee system. Among HIV positive drug users with a positive tuber-culinn test, the incidence of active tuberculosis is 896 per year.17 The overalll probability that a latent tuberculosis infection will express itselff in the form of active tuberculosis in an immunocompetent per-sonn is approximately 10%. Moreover, the development of Multidrug-resistantt tuberculosis MDRTB within this high risk population gives reasonn for concern. Inadequate therapy in the sense of insufficient drugs,, or too short therapy and lack of compliance of the patient is consideredd to be the cause of the MDR TB.|8)91,0
Inn Amsterdam, the total number of drug-addicts is estimated to bee 7000 'according to recent estimations this number should be 5000; seesee chapter 5 figure 2A- Approximately one third are from the Netherlands,, 25% from 'ethnic groups' -born in Surinam, the Dutch Antilles,Antilles, Turkey or Morocco- and 40% are from other foreign countries.
P M A M A H U A M P P
T a b a r c M i o s i ss a m o n g o p l a t * u s e r s 1 3
Thee majority of the 1700 ethnical drug users were born in Surinam. Onee third of the estimated 3000 foreign drug users has a residence permitt and is permanent resident of Amsterdam, The majority of the foreignn drug users has no residence permit, and most of them are tem-porarilyy residing in the city.1"
Amsterdamm has an extensive network of low-threshold drug relieff agencies. The MHS and 200 general practitioners are the main providers.. Harm reduction is the main objective and methadone pre-scriptionn is the core activity within these services. The services are limitedd to those citizens who are registered as citizens of Amsterdam. Basicc medical care, for the destabilised, active drug users, Is provided forr by the MHS. Methadone prescription by the CPs is limited to the moree stabilized drug users, with a diminishing addiction and a stable sociall life. Only within the programme of the MHS is screening for tuberculosiss mandatory. Every year approximately 3500 people use thesee agencies.
Thee objective of this study is to determine the incidence of tuberculosiss among drug users receiving methadone from the MHS and CPss in Amsterdam. Incidence rates in the period of 1989-1993 and dif-ferencess in incidence rates between subgroups are presented. As an additionall spin-off, the study tentatively answers the question whetherr the existing policy of methadone prescription to large num-berss of stabilized drug-users by the GPs without mandatory periodical screeningg for tuberculosis is acceptable.
M e t h o d s s
StudyStudy population
Informationn from the Central Methadone Registers CMR was used to definee the study population. Since 1980, the CMR has registered all methadonee prescriptions in Amsterdam. For each client, a number of individuall details are collected: gender, date of birth, country of birth andd methadone, e.g. dosage, date of first prescription and prescribing physician.. Methadone is dispensed in various programmes in different locations:: police stations, jails, therapeutic programmes by the Jellinekk Centre, and in methadone maintenance programmes by the GPs andd Municipal Health Service. Approximately 9096 of the methadone is dispensedd through the low-threshold programmes of the MHS and GPs."2 2
Methadonee can be supplied from different locations in the coursee of 1 year. The various locations are divided into 3 categories, eachh one excluding the other two. The first two are MHS locations: I .. the Prostitution and Foreigners Outpatient Clinic PFO for prostitutes andd illegal foreigners and 2. the general outpatient clinics or methadonee bus for the Dutch resident active drug users that are not eligiblee for the CP methadone programme. The third category is the
P M A M A H U A M P P
4 T u b e r c u l o s i s « m a n s o p i a t e u s « r *
GP-clients.. The clients in the worst circumstances are in the PFO; in additionn to their addiction they work as prostitutes and/or are illegal foreignerss to whom medical care and methadone is only provided in casess of serious health problems. The out-patient clinics provide intensivee medical and social care to clients with a problematic addic-tionn history who are not able or motivated to detoxify. The methadone buss programmes provide methadone exclusively to clients who need lesss help in the medical and social area. General practitioners offer helpp to drug users who are more stabilized and to some extent- inte-grated. .
Clientss receiving methadone maintenance treatment from GPs andd MHS and who were living in Amsterdam, were selected. Drug users withh an address outside Amsterdam and drug users being supplied withh methadone less than 7 times in the year of observation were excludedd from the study. Moreover, to prevent overestimation of the incidence,, persons who turned to the Drug department for help becausee of tuberculosis were excluded from the study. These clients aree defined as clients having tuberculosis diagnosed during the first monthh of the study. Finally those persons with a known history of tuberculosiss having a relapse within the observation period were excluded. .
DeterminationDetermination of the observation time
Thee study period runs from January 1989 through December 1992. Thee individual observation time is determined per calendar year. The startt of the individual observation time is the first month of the year concernedd in which methadone is supplied at least 7 times. The 31 Decemberr of the year concerned marks the end of the individual observationn time. In the event of death, the date of death marks the endd of the observation period and if tuberculosis is diagnosed in a person,, the date of the diagnosis marks the end of the observation time. .
Iff a client is provided with methadone at different locations in 1 calendarr year, the total observation time of that year is attributed to thee location at the top of the hierarchy, irrespective of the number of suppliess in that location. The order of rank is, successively, PFO, gen-erall outpatient clinic or bus and GP. This hierarchic division takes into accountt the fact that the client populations have a decreasing degree off social and medical problems.
Case-finding Case-finding
Thee information on tuberculosis cases comes from 3 sources. First, throughh the Hospital Project of the MHS, which counsels all drug-users hospitalizedd in Amsterdam. The second source of information is the 6 monthlyy tuberculosis screening by X-ray of all methadone clients of
P M A M A H U A M P P
T u b e r c u l o s i ss a m o n g o p i a t e u s a r s
19891989 1990 1991 1992
Totall number of persons Malee %
Meann age SD
Meann time with CMR SD
OrlglmOrlglm % Westernn countries Sur/Dutchh Ant. Mediterranean n Otherr countries 3268 8 71.1 1 32.88 6.6 522 2.8 3079 9 70.3 3 33.66 6.9 6.00 3.1 3076 6 71.4 4 34.33 7.0 6.66 3.4 3 3 1 8 8 70,6 6 34.SS 6.9 6.99 3.9 64.3 3 19.6 6 7.1 1 9.0 0 65.3 3 17.8 8 7.7 7 9.2 2 65.2 2 18.4 4 7,7 7 8.8 8 64.. 5 17.6 6 9.3 3 8.7 7
TableTable I Description of variables overr the study period
thee MHS at the tuberculosis department, a civil servant agency of the MHS.. Thirdly, all cases of tuberculosis in Amsterdam are reported to thee tuberculosis department, including drug users not known at the drugg department. Cases of tuberculosis reported in a year during whichh a client was not in treatment are omitted from the study.
Inn the period 1989-1992 a total of 80 drug users were diag-nosedd as having tuberculosis in Amsterdam. Ten cases were diagnosed inn 1989, 20 in 1990, 30 in 1991 and another 20 during 1992. In all casess tuberculosis was diagnosed through a microbiological specimen orr a bacterial culture. Eventually, 43 cases were included in the study.
Eightt patients had a relapse and were not included. Six of these casess obtained methadone at the MHS and 2 did not receive methadone duringg the study period. Of the 71 cases in which tuberculosis was diagnosedd for the first time, 10 were not registered with the CMR, but weree registered with the hospital project or tuberculosis department off the MHS. Three cases were registered but did not receive methadone inn the whole of the study period. Nine cases did receive methadone duringg the study period but did not during the year tuberculosis was diagnosedd and 1 case had been supplied with methadone during less thann 7 days in the year concerned. These 10 cases all received methadonee at the MHS, 4 of them at the PFO. In 6 cases, tuberculosis wass diagnosed within 1 month after the first supply of methadone. Fivee of them were clients of the MHS and 1 of the CP.
Analysis Analysis
Thee trend in incidence was studied using a Poisson regression tech-nique,, in which the 4 annual cohorts are grouped into a single cohort
P M A M A H U A M P P
T M b * r c u l o s l ss a m o n g o p i a t e u s e r s
PFÖ:PFÖ: Prostitutes and Foreigners Outpatient Clinic, CÖC:CÖC: General Outpatient Clinic,
CP:CP: Central Practitioner
totall group male e
PV V MethadoneMethadone programme
PFO O
GOCC not PFO
Onlyy GP age e >> 3S yrs registration n 55 5 yrs 1099 9 6487 7 3530 0 9.9 9 58.4 4 31.8 8 40,4 4 77.0 0 66.9 9 21.2 2 34.2 2 43.8 8 50.1 1 72.2 2 79.9 9
TableTable 2 Distribution of person yearss py per subgroup and
methadonee programme
withh different values on the dummy-variable year- To carry out a Poissonn regression the condition constant rate over time- has to be met.1'33 Analyses to distinguish sub groups were also carried out using Poissonn regression. Subgroups are distinguished by age, sex, country off origin and time and location of methadone distribution.
R e s u l t s s
DescriptionDescription of tftc study population
Thee study population consisted of 5044 individuals studied for 1 or moree years. Those who were observed for 1 year -33% of all clients- con-tributedd only 10% of the total observation time. Long-term methadone maintenancee clients predominate in the study.
Inn Table 1, the socio-demographic profile of the study population duringg the study period is presented. For each year, approximately 71%% of the study population are males. The mean age rises during the studyy period: in 1989 it was 32.8 years and in 1992 34.8 years. The meann length of registration with the CMR increases as well, from 5.2 yearss in 1989 to 6.9 years in 1992. This implies that the mean age of thee clients entering the study population during the period of study is lowerr than that of those leaving the study population.
Everyy year, approximately 65% of the study population origi-natess from Western countries including The Netherlands. The percent-agee of persons from Surinam or the Dutch Antilles fell from 19.6% in
19899 to 17.6% in 1992, whereas the contribution of persons from the Mediterraneann rises from 7.1% to 9.3%. As a whole the population is dynamic,, but quite stable in terms of sub-populations during the study-period. .
Tablee 2 shows that the PFO differs from the 2 other categories by itss relatively small contribution to the observation time of males,
per-T u b e r c u l o s i s s a m o n g g P M A M A H U A M P P o p i a t ee u s e r s 1200 0 en n 0 0 0 en n
[YY e *a r
ov v 0 0 0 en n ' ' <T> <T> 0 0 0 OT OT o> > 0 0 0 en n "** *FigureFigure I Annual incidence of tuberculosiss of drug users in
methadonee maintenance programmes withh 95% confidence intervals
sonss o l d e r than 35 years and persons w h o have been r e g i s t e r e d w i t h thee CMR f o r five years or m o r e . This was t o be e x p e c t e d , c o n s i d e r i n g thee PFO's specific t a r g e t g r o u p s . Most o f the GP clients b e l o n g t o t h e c a t e g o r yy older than 35 years and are registered over more t h a n 5 yearss w i t h the CMR.
TuberculosisTuberculosis incidence in the total study period
Forty-threee new cases o f t u b e r c u l o s i s were d i a g n o s e d . The t o t a l observ a t i o nn t i m e was 1 1 1 1 7 p e r s o n y e a r s . The mean i n c i d e n c e o f t u b e r c u -losiss in the t o t a l s t u d y p e r i o d 1989- 1992 t h e r e f o r e a m o u n t e d t o 3 8 6
p e rr 1 0 0 0 0 0 p y • 95% C.I.: 287-522 per 100000
py-TuberculosisTuberculosis incidence patterns
Thee annual incidence o f t u b e r c u l o s i s w i t h 95% c o n f i d e n c e i n t e r v a l s is s h o w nn in Figure 1 . The a n n u a l incidence rates are based o n 7, 10, 18 a n dd 8 cases r e s p e c t i v e l y . The f i g u r e s do not show a s t a t i s t i c a l l y s i g -n i f i c a -n tt i-ncrease o v e r t h e 4 y e a r s . The i-ncide-nce is l o w e s t i-n 1 9 8 9 ;
P M A M A H U A M P P
T u b e r c u l o s i ss « m a R fl o p i a t e u s e r s
CMR:CMR: Central Methadone Register
PFO:PFO: Prostitutes and foreigners Outpatient Clinic COC:COC: General Outpatient Clinic
CP:CP: General Practitioner Incidencee 9 5 % C.L perper J 00 000 py Sex Sex Matee 500 365 €84 Femalee 120 45-322 AgeAge years <SS 35 281 181-43S >> 35 576 383-866 CMRR registration years < 55 229 101-480 SSS 447 322-619 LandLand of origin Westernn countries 262 167-410 Mediterraneann 598 249-1438 Surinam/Dutchh Antilles 774 475-1265 Otherr countries 313 101-970 MethadoneMethadone programme PFOO 728 364-1456 GOCC not PFO 508 362-715 onlyCPP 0.57 14-227
TableTable 3 The incidence of tuberculosis in subgroupss
2 1 3 / 1 0 0 0 0 00 py. In the following years it was higher. Only in 1991 doess it deviate significantly with 670 per 100 000 py. This is seen as aa temporary upswing of the high endemic incidence of tuberculosis amongg the study population.
TuberculosisTuberculosis Incidence of different sub groups
Thiss is given in Table 3. The univariate analyses show that the inci-dencee is significantly • p < 0.05• higher among males, older clients •>> 35years- and clients who have been known with the CMR for 5 years orr more. Moreover, the incidence in clients originating from Surinam orr the Dutch Antilles is higher than in Westerners. Incidence is highest
TT m b T c a I ss 1 s RR g
P M A M A H U A M P P
o p l a tt • y » « r s
•• •
CMR:CMR: Centra) Methadone Register
PFO:PFO: Prostitutes and Foreigners Outpatient Clinic CP:CP: Central Practitioner
DevianceDeviance 537.8; df- 12735 LikelihoodLikelihood Ratio Statistic:
43.9;43.9; df- 5 p < 0.001 alpha - 7.61 (SB - 0.67)
Variable e Agee >35 years Sexx male
registeredd with CMR >5 years CP P PFO O Beta a SE E P-value e RR R 95XX CI 0.68 8 1.55 5 0.84 4 2.21 1 1.08 8 0.31 1 0.42 2 0.43 3 0.73 3 0.42 2 0.028 8 0.005 5 0.049 9 0.002 2 0.010 0 1.98 8 4.69 9 2.32 2 0.11 1 2.94 4 1.08*3.65 1.08*3.65 1.61*13.67 1.61*13.67 1.00-S.38 1.00-S.38 0.030.03 0.46 1.30-6.66 1.30-6.66
TabUTabU 4 The multivariate Poisson regressionn model for tuberculosis
incidencee among drug users in methadone maintenancee programmes
amongg the clients of the PFO programme, namely 728 per 100 000 py. Off the 43 cases 41 -95%- were clients of the MHS, and the other 2 cases weree from the general practitioners group.
Inn the multivariate model, shown in Table 4, the variables age, sex,, length of methadone registration and the location of treatment appearr to be related to the incidence of tuberculosis. The incidence increasess with age -assuming that all other variables remain the same-same-tt the risk of tuberculosis being twice as high for drug users
olderr than 35 years than for those of 35 years or younger. The risk appearedd to be 4.7 times higher for males than for females. Drug userss who have been registered with the CMR for more than 5 years havee a 2.3 times higher risk than those registered for less than 5 years.. The risk also varies with the location of treatment, with the risk off contracting tuberculosis for drug users attending the PFO being 2.9 timess higher than that for those attending a general out-patient clinic. Forr those who are not provided with methadone by the MHS the risk is 99 times lower. In the multivariate analysis no significant relation betweenn country of birth and tuberculosis and no interactions betweenn the variables in the model could be found.
D i s c u s s i o n n
Thee mean incidence of tuberculosis among the Amsterdam drug users inn the study population is 386 per 100 000 py in the study period. This isis 10 times as high as the incidence in the general population of this
P M A M A H U A M P P
T u b e r c u l o s i ss a m o n g o p i a t e u s e r s
TheThe Netherlands Amsterdam Amsterdam Study-population Study-population
700 0 600 0 500 0 400 0 300 0 200 0 100 0 o o CT1 CT1 CTl l — — en n <T> > r s i i en n en n
FigureFigure 2 The annual i n c i d e n c e of t u b e r c u l o s i ss in The N e t h e r l a n d s , A m s t e r d a mm and d r u g users in m e t h a d o n ee m a i n t e n a n c e p r o g r a m m e s 1 9 8 9 -- 1992
city.. In A m s t e r d a m the incidence is 4 times as h i g h as in The Netherlandss g e n e r a l l y . In the period 1 9 8 9 - 1992 a t o t a l of 904 cases o ff active t u b e r c u l o s i s was d i a g n o s e d in the city. '" Of these cases 80 weree k n o w n d r u g users. This implies t h a t d r u g users • making up an estimatedestimated 1% of the Amsterdam's population a c c o u n t f o r a p p r o x i m a t e l yy 9% of t u b e r c u l o s i s c a s e s . In Figure 2 the incidence of t u b e r c u -losiss in The N e t h e r l a n d s , A m s t e r d a m and the study p o p u l a t i o n is s h o w nn in a d i a g r a m . How the sudden increase in t u b e r c u l o s i s among d r u gg users in 1991 s h o u l d be i n t e r p r e t e d is not clear.
P M A M A H U A M P P
T u b e r c u l o s i ss * m o • £ o p i a t e u s e r s B I
Maless -Rate Ratio RR = 4.7 compared to females-, drug users older than 355 years -RR = 2.0 compared to those of 35 years or younger •, and drug users reg-isteredd with the CMR for 5 years or more • RR = 2.3 compared to those registered forr less than 5 years* run a higher risk of tuberculosis. The reasons for the associationn between the incidence and the variables of age and period off methadone consumption are t w o f o l d . First, the chance of contact withh mycobacterium tuberculosis increases with age and with the time spentt within the group of drug users. Secondly, a longer addiction periodd and higher age reduce the physical condition, increasing the probabilityy of disease. The difference between males and females is nott easy to explain. However, it is a known fact that in the general Amsterdamm population tuberculosis occurs more often in males than in females;; only 30% of all tuberculosis patients in Amsterdam in 1990 aree female.
Thee incidence of tuberculosis appears to be closely related to thee programme in which methadone is supplied. Clients of the General Practitionerss run a 9 times -95%Ci: 2.2-33- lower risk of tuberculosis thann those being provided for by the general out-patient clinic or the methadonee bus. Despite the fact that the PFO treats more females, moree drug users younger than 35 years and drug users registered with thee CMR for less than 5 years, the risk of contracting tuberculosis is 3 timess higher -95%Ct: 1.3-6,7- for this group than for the clients of gen-erall out-patient clinics or the methadone bus. The close relation betweenn incidence figures and methadone location suggests that the deviantt lifestyle related to drug use could account for the increased incidencee of tuberculosis among different drug users. This type of lifestylee is especially common amongst clients of the MHS. Clients of thee MHS generally have more police contacts than clients of the GPs."2 Thee clients of the PFO are persons who, in addition to their drug addic-t i o n ,, are also characaddic-terized by working as a prosaddic-tiaddic-tuaddic-te and/or being illegall foreigners. Their lifestyle is probably more hazardous than drugg addiction alone. Among these illegal foreigners there wilt be a negativee selection by health: on the one hand because of a higher reluctancee of this group to obtain care with a government agency; on thee other because the Amsterdam 'deterrent policy' for illegal foreign-erss increases this reluctance even more. Only in the case of serious healthh problems, often HIV-related, is a long term methadone pro-grammee offered.
Onee of the undesired effects with respect to tuberculosis, a communicablee disease, is the daily contact between the clients in the methadonee programmes of the MHS. General practitioners however, prescribee 'take home methadone' for a period of a few weeks at a time. Thus,, the CP-clients rarely get into contact with one another. This providess another reason to be alert for tuberculosis in the MHS pro-grammes. .
P M A M A H U A M P P
•• 2 T u b e r c u l o s i s a m o n g o p l a t * u s e r s
Althoughh HIV infection was not included as a variable in this study,, this infection is an important risk factor for tuberculosis. In 23 off the 43 cases -53%- infection with HIV is known, whereas among drugg users receiving methadone at the MHS an estimated 14% has been infectedd with HIV.|,S In the study of Selwyn et al., among intravenous drugg users, all tuberculosis cases were found to be HIV positive.17 The differencess in incidence between the various methadone locations can probablyy also, at least partly, be traced back to differences in HIV prevalence.. This, again, is connected with lifestyle.
Despitee the fact that the tuberculosis incidence among foreign-erss is 18 times higher than among the native population/3 the country off origin does not appear to be a significant factor. This is due to the highh tuberculosis incidence w i t h i n the population studied. Consequently,, mean age and period of registration with the CMR i.e. periodperiod of involvement in the drug scene- are more important risk fac-torss in this study.
C o n c l u s i o n n
Inn this study, the most important predictors for tuberculosis amongg drug users in Amsterdam were age, period of registration with thee CMR and type of methadone programme * MHS versus CP- The fac-torss age and period of registration are related to the period of drug use.. The factor 'type of programme' can be interpreted as a lifestyle indicator:: CP clients are less involved in the drugscène than MHS clients,, plus there is a probable impact of a difference in terms of HIV prevalence. .
Preventivee measures for tuberculosis through drug relief agen-ciess are necessary. On the basis of this study, it seems sufficient to limitt mandatory tuberculosis screening in Amsterdam to the clients of thee MHS. However, the fact that at least 30% of the drug users being diagnosedd as having tuberculosis for the first time were not in touch withh any of the drug relief agencies at that time is a cause for concern. Inn order to prevent the dissemination of tuberculosis among drug userss and the general public, the reach of the MHS methadone pro-grammess should be improved.
Acknowledgement Acknowledgement
TheThe authors would tike to thank Ms. I. Fellinger, case managermanager with the MHS Hospital Project and Mr.Mr. C. Brewster. Head of the MHS CMR for their support andand co-operation in collecting the data.
M a r c e ll C . A . B u s t e r
DepartmentDepartment of Epidemiology, Documentation and health Promotion EDC, MunicipalMunicipal Health Service, Amsterdam
L i e s b e t hh R o o k
DepartmentDepartment of pharmacology, Slotervaart Hospital, Amsterdam
G i e ll H . A . v a n B r u s s e l
DepartmentDepartment of Social and Mental Health, Municipal Health Service, Amsterdam.
J a nn v a n R e e
DepartmentDepartment of Pharmacology, University of Utrecht, Utrecht
W i mm v a n d e n B r i n k
P M A M A H U A M P P
C H A S I N GG T H E D R A G O N ,
R
E
L
A
T
E
DD T O
T
H
E
I
M
P
A
I
R
E
DD
L
U
N
G
F
U
N
C
T
I
O
NN
A
M
O
N
G
H
E
R
O
I
NN
U
S
E
R
S
?
DrugDrug and Alcohol Dependence 2002: 68; 219-26
A b s t r a c t t
Aim:Aim: To d e s c r i b e t h e p u l m o n a r y f u n c t i o n a n d prevalence of dyspnoea a m o n gg m e t h a d o n e p a t i e n t s and t o s t u d y the r e l a t i o n w i t h e x p o s u r e t o h e r o i nn b y i n h a l i n g .
StudyStudy population: A sample of 100 p a t i e n t s f r o m m e t h a d o n e m a i n t e n a n c ee t r e a t m e n t -84% male, average age 42
years-Measurements:Measurements: Questionnaires were used t o measure l i f e - t i m e e x p o s u r ee t o h e r o i n , cocaine, c a n n a b i s , t o b a c c o , and s y m p t o m s o f
d y s p n o e a .. S p i r o m e t r y was p e r f o r m e d a n d r e s i d u a l d i f f e r e n c e of mea-suredd FEVi Forced Expiratory Volume in 1 second- f r o m the age, sex, h e i g h tt and e t h n i c i t y p r e d i c t e d value AFEVi was used as a main o u t -comee parameter.
Findings:Findings: The median AFEVi was - 0 . 2 6 litres • interquartile range - 0.70; ++ 0.12- T w e n t y per cent experienced d y s p n o e a w h i l e 'walking at a normalmal pace with someone of their own age'. H i s t o r y o f c i g a r e t t e s m o k -i n gg was r e p o r t e d by 98%; hero-in s m o k -i n g by 88%. M u l t -i p l e l-inear r e g r e s s i o nn analysis s h o w e d a s t a t i s t i c a l l y s i g n i f i c a n t a s s o c i a t i o n b e t w e e nn h e r o i n - s m o k i n g and AFEVi, l o g i s t i c r e g r e s s i o n a n a l y s i s s h o w e dd an a s s o c i a t i o n between h e r o i n - s m o k i n g and prevalence o f d y s p n o e a . .
Conclusions:Conclusions: Chronic heroin s m o k i n g seems t o be related t o an i m p a i r e dd l u n g f u n c t i o n and h i g h e r prevalence o f d y s p n o e a . However,
partt o f t h e o b s e r v e d l u n g f u n c t i o n i m p a i r m e n t w i l l be caused by t o b a c c oo s m o k i n g . Further research is needed t o q u a n t i f y the effect o f heroinn s m o k i n g and d i s e n t a n g l e t h e effect o f s m o k i n g heroin and t o b a c c o . .
P M A M A H U A M P P
•• B C h a s i n g t h e d r a g o n i « d l u n g f u n c t i o n
I n t r o d u c t i o n n
'Chasing'Chasing the dragon' implies that heroin is heated on foil and heroin vapourr is inhaled through a straw. To enhance evaporation, a mixture
off base-heroin and caffeine is generally used." Since the start of the heroinn epidemic in the early seventies in Amsterdam, chasing the dragonn has become the dominant route of administration. Nowadays, 85%% of the heroin users in Amsterdam smokes its heroin. Trends towardss smoking heroin are observed in Spain, the UK and Ireland.1'13 Chasingg the dragon is considered to be a safer mode of heroin use thann injecting. Risks of fatal overdose, HIV, HBV and HCV infection are moree prevalent among injectors. Chasing the dragon however, is not withoutt risks. Overdose mortality may occur14 and in individual cases, lethall leucoencephalopathy has been reported.IS|6
Itt is known that heroin affects the pulmonary function. It affects thee respiratory control centres which may lead to fatal pulmonary depression.177 Like other opiates, heroin is known to release his-tamine18199 and case series of asthma triggered by inhalation of heroin havee been r e p o r t e d . "0" ' Injecting heroin may cause septic emboli
'originating'originating from an infection at the injection site or tricuspid valve endocarditis-endocarditis- or foreign body emboli 'originating from contaminants oror fillers in the injectate- Furthermore, infectious diseases such as TB
andd pneumonia are more prevalent among heroin users, especially amongg HIV infected drug users.1121*3 Lung-function studies among smalll samples of ex heroin injectors indicated a decrease of diffusing capacityy rather than obstructive complications.| U | , S Lung function studiess among samples of heroin inhalers are not known.
Att the Amsterdam methadone treatment centres there is an increasingg concern about the pulmonary function due to chronic use off heroin by chasing the dragon. However, little is known about the effectt of heroin smoking on pulmonary function. The Amsterdam populationn of heroin users may provide valuable information about thee chronic effects of smoking heroin on pulmonary function. The aim off this study is to describe the lung function and complaints of dys-pnoeaa among heroin addicts treated with methadone and to study the relationn with the exposure to heroin by chasing the dragon, control-lingg for other causes of lung function impairment.
M e t h o d s s
StudyStudy population and data collection
Thee study was conducted at a methadone maintenance outpatient clinicc of the Municipal Health Service MHS of Amsterdam. Patients of thiss treatment centre are generally not able or willing to stop using illicitt drugs. The major goal of methadone maintenance treatment is to reducee the harm that is caused by the use of these drugs.
P M A M A H U A M P P
CC h a • I n g t R • d r a g o n a n d l a n g f u n c t i o n 1 7
subjectedd to a periodical medical check up. For the purpose of this studyy the lung function of the patients was measured during this checkk up. The protocol was approved by the medical ethical commit-teee of the Municipal Health Service and the University of Amsterdam. AA trained research assistant asked 123 patients to perform a spiro-m e t r yy test before filling out a questionnaire. In this way the answers off the questions could not influence the technician when spirometry wass conducted. Three patients refused to co-operate. Twenty patients appearedd to be unable to conduct the spirometric test according t o cri-teriaa of quality and reproducibility of the American Thoracic Society butt did fill out the questionnaire. In this article we focus on the 100 patientss who performed a valid spirometric test.
MeasuresMeasures outcome
Twoo measures of outcome were used in this study: reported com-plaintss of dyspnoea based on the Modified Medical Research Council questionnaire1166 and the Forced Expiratory Volume in 1 second FEVi. Sixx hierarchical questions * shown at Table / • considering shortness of breathh were asked. Generally the two most severe categories of short-nesss of breath - while walking at their own pace and at rest- are sideredd as having dyspnoea. In order to prevent empty ceils when con-ductingg the logistic regression analysis, patients who answered 'Yes' too the question 'Are you suffering from shortness of breath when you walkwalk at a normal pace with someone of your own age?' were consid-eredd to have complaints of dyspnoea too. The spirometer • Vitaiograph 2170,2170, Spirotrac iv- was used to measure the FEVi. Only one spirome-terr was used and ail measures were performed by one technician. The residuall difference of the measured FEVi from the age, sex, height and ethnicityy predicted value AFEVi was used as main outcome parameter. Moreover,, the percentage of the client's predicted FEVi value was cal-culatedd %FEVi. Predicted values are based on the guidelines of the Europeann Respiratory Society.117 A race-adjusted correction factor was appliedd to the data obtained in black participants by multiplying the predictedd value of FEVi by a factor 0.9 as recommended by the Americann Medical Association.1"
Fr*éict*éFr*éict*é vmtmes FEVi
Maless Caucasian: FEVi predicted - 0-0430 length cm - 0,029 age years - 2.49 Femaless Caucasian: FEVi ptedicted " 0.0395 length cm - 0.025 age years - 2.60
Negroid:: FEVi pttdMtü - FEVi predlcted a^ta, 0-9
Exposure Exposure
Heroinn use by chasing the dragon is the exposure variable of interest inn this study. The variable indicating the life time exposure to heroin
P M A M A H U A M P P
•• • C h a s i n g t h • d r a g o n a n d l u n g f y n c t l o n
byy chasing the dragon was constructed by multiplying the frequency off heroin use -number of days during the tast thirty days of heroin useuse • and the total number of years that heroin was used. This variable rangedd from null never chasing the dragon- to a maximum value of 9000 30 years and 30 days a month- As 12 percent had a zero expo-sure,, an ordinal variable of 8 groups of approximately equal size were constructed.. The value 0.0 was directed to the lowest exposure group
equivalentequivalent to zero exposure', the value 1.0 to the highest exposure groupp -equivalent to a duration of exposure of more than 20 years and
dailydaily use
Confounders Confounders
Thee vast majority of patients uses other drugs than opiates that may havee additional harmful effects. This may bias the observed relation betweenn heroin and lung function. In this study exposure to heroin by smokingg is the variable of interest and exposure to other drugs • smok-inging of cocaine, cannabis and cigarettes- are potential confounders. Similarr to the exposure to heroin by chasing the dragon, the exposure too marihuana and cocaine by inhalation is constructed by multiplying thee period of use • in years - with the frequency of use • last 30 days of use-use- Subsequently, based on the percentage of non-exposure -21% amongg base-cocaine and 29% among marihuana- ordinal variables of 5 and 4 groupss of approximately equal size were constructed -highest catego-ryry was equivalent to a duration of more than 12.4 and 13.0 years and
dailydaily use of cocaine or marihuana
respectively-Exposuree to tobacco was calculated by multiplying duration and frequencyy of use • mean daily number of cigarettes during the last 30 days-days- and expressed in pack-years. One pack-year is the equivalent to thee exposure of one packet of cigarettes a day during one year. Tobaccoo exposure is expressed as a continuous variable in which one incrementt equals 10 packyears of exposure. Among unexposed sub-jectss the average AFEVi is expected to be null in all age categories.
However,, among heroin addicts, tobacco smoking is common and the averagee AFEVi is expected to show higher negative values with increasingg age. In the analysis, age is a continuous variable in which thee age of 20 is set to zero and one increment equals 10 years age dif-ference. .
Inn addition to exposure variables, other variables are related to thee exposure to heroin by inhaling and/or pulmonary function. The variabless that were evaluated are 'body-mass index BMI: Weight
(kg)/length(m)(kg)/length(m)22-- lower than 18', 'a lifetime history of tuberculosis TB oror pneumonia treated with antibiotics during the last two years' and
'reported'reported symptoms of bronchial allergenic or non-specific
hyper-responsiveness'.responsiveness'. These variables could be confounders but also inter-mediatee steps between exposure and disease or even symptoms of
P M A M A H U A M P P
C h a s i n gg t h * d r a g o n a n d l u n g f u n c t i o n 8 8
pulmonaryy impairment. Only in the first case statistical adjustment wouldd be necessary. Therefore two analyses were conducted, one with additionall adjustment for BMI, TB/pneumonia and bronchial hyper-responsivenesss and one without this adjustment."'
Analysts Analysts
Multiplee linear regression was applied to describe the relation betweenn the exposure to heroin by smoking and AFEVi, logistic regres-sionn was applied to describe thé relation between the exposure to heroinn by smoking and the occurrence of dyspnoea. The results of the linearr regression and logistic regression analysis are presented in threee steps. The first model shows univariate relations of exposure to smokingg of heroin, cocaine, cannabis and cigarettes. Controlled for eachh other's influence these relations are presented again in the sec-ondd model. In the third model additional variables were added: Body Masss Index < 18, complaints of bronchial hypersensitivity, history of TBB lifetime- or pneumonia two years- Moreover, crude data are presentedd in scatter plots showing the AFEVi and prevalence of dys-pnoeaa in each exposure category.
Nextt to the FEVi the FVC- Forced Vital Capacity: the total volume thatthat can be expired after maximum inhalation- is measured. The FVC Wass 7.10 sd:0.98- and the FVC % predicted was 10096 -sd:i5.9< The mean FEVi/FVCC ratio was 0.73 sd:i2.6- The maximum value of the FEVi/FVC ratioo was 0.93; indicating that the decreased FEVi was not the conse-quencee of a decreased FVC.
R e s u l t s s
DescriptionDescription of thm study population
Hundredd patients successfully completed the questionnaire and spirometry.. Their average age was 42.4 years - standard deviation sd: 6.7 • Thee youngest was 22, the oldest 57 years old. The majority was male •84%.. and white -77%.
Exposure Exposure
Alll patients reported that they had used heroin, 88% used heroin by chasingg the dragon. Moreover, 79% reported they ever inhaled cocaine andd 71% had a history of smoking cannabis. During the last 30 days, 522 patients inhaled heroin, 49 cannabis and 49 cocaine. Less than half -41.. had ever injected their drugs. Only 13 out of 100 patients had recentlyy injected heroin, 11 of them both heroin and cocaine.
Alll patients but 2 had ever smoked cigarettes. Only one of the everever smokers did not smoke during the month preceding the inter-view.. Among cigarette smokers, the median age of starting cigarette usee was 1 5 years. The average period of cigarette use was 26 years.
7 1 1
P M A M A H U A M P P
CC h • i i R g t h e d r a g o n a n d l u n g f u n c t i o n
Sold:Sold: considered as complaints of dyspnoea in this study
AreAre the following remarks applicable to you? II suffer from shortness of breath when
Mediann AFEVi titers
interquartile-range interquartile-range
Mediann %FEVi
interquartile-range interquartile-range
Degreee 0: 'No' to all following question 45 Degreee 1: I am in a hurry on flat ground 15 Degreee 2: I climb stairs or walk up a small hill at a normal pace 18
Degreee 3: I walk at a normal pace with someone of my own age 9
Degreee 4: I walk at my own pace on f l a t ground 6
Degreee 5: 1 am at rest 7 Totall 100 0.000 -0.31; 0,58 -0.266 -0.71; -0.03 ^0.366 -1.20; -0.07 -1.022 0.S1; 0.21 0.899 2.14; -0.45 SS -1.24,-0.26 -0.266 -0.70 ; 0.12 00 0 93 3 89 9 8 5 5 7 5 5 7 6 6 93 3 93-113 93-113 78-99 78-99 68-98 68-98 69-93 69-93 36-36- 89 67-93 67-93 78-104 78-104
TableTable 1 Definitions of different categoriess of dyspnoea and AFEVi. SSFEVi i
ComplaintsComplaints of dyspnoea and AFEVi
Thee outcome variables complaints of dyspnoea, AFEVi and %FEVi- are shownn in Table 1. Forty-five patients reported no complaints of dysp-noea.. Twenty-two patients reported dyspnoea of the third degree or higher.. The AFEVi ranged from -3.0 until +1,1 litre. The average value wass -0.32 sd:7.7- and thé median value was -0.26 litres 'interquartile rangee /Off:-0.70; +0.12- The fcFEVi ranged from 19 until 129 with an aver-agee value of 91 • sd: 20.1 • and a median of 93 IQR: 78-104- Table 1 shows aa decreasing AFEVi and «FEVi with increasing degree of dyspnoea. Normall AFEVi and KFEVi values are observed among those without any complaintss of dyspnoea. Spirometric results among methadone patientss with dyspnoea to the third degree and higher -mean AFEVi:-0.89 litre,, %FEVi: 74%- are lower than among patients with none or minor c o m p l a i n t ss • mean AFEVi: -0.t6 litre, %FEVi: 96%, both t-tests p < 0.001 •
FEVtFEVt and the use of drugs
Figuree 1 shows the relation of exposure to smoking heroin, cocaine or cannabiss and AFEVi. AFEVi values vary widely among the high heroin exposuree and low cocaine exposure categories. Two subjects with symptomss of dyspnoea and FEVi values that are more than 2.5 litres lowerr than expected, show extreme exposure values. They are includ-edd in the zero exposure group of cocaine and cannabis, and in the higherr exposure groups of heroin
Tablee 2 shows the results of the multivariate regression analysis. Inn Table 2A univariate relations are shown. Age was not significantly relatedd to a decreased AFEVi. Differences in exposure to cigarette
CC h a S nn g tt h e dd r a g o a n d d P M A M A H U A M P P l u n gg f u n c t i o n 7 1 1 1,5 5 I I ^0.5 5 0 0 5 5 -1 1 - 1 , 5 5 -- -2 -2.5 5 ** -3
BlackBlack dots; dyspnoea in this study
1.5 5 O O o o • •
9 9
_ Q _ _ • •e e
o o o o Q Q 0 0 8 8s s
o o 0 0i i
» » o o o o • • o o b b8 8
8 8 • • o o 0 0 —Q— — : : m m • • o o o o o o6 6
o o • • 8 8 o o o o8 8
<J J 0 0 100 10.2 10.4 10.6 10.8 |HH e r o i ji c t p c j s u r e |c a I e g }> r • \ \ "3 3 0.5 5 ~~ 0 •0.5 5 ** - I -C C -1.5 5 >> -2 u.. . -2,5 5 JJ -3I I
100 10.25 10,5 10,75 |CC o c a i n 4 e x p o s |u i' e ' a i p p r 1.5 5 \ \ .0.5 5 it it ~~ 0 >o.s s -1.5 5 >> -2 o. . -2,5 5 JJ -3 roo w, I C a n n a b t ss p * 333 I0.67 I I « p o s u r ee |c a I e y o r y | FigureFigure 1 O r d i n a l e x p o s u r e to heroin , cocaine -inhaling and cannabiss in r e l a t i o n t o t h e A F E V i and d y s p n o e a . .s m o k i n gg d i d not show a s i g n i f i c a n t r e l a t i o n s h i p w i t h the AFEVi. The degreee o f h e r o i n e x p o s u r e by s m o k i n g was s i g n i f i c a n t l y related t o a decreasingg AFEVi. The h i g h e s t exposure c a t e g o r y s h o w e d a FEVi value 0.577 l i t r e l o w e r t h a n t h e none-exposure category. We d i d not observe anyy s i g n i f i c a n t r e l a t i o n between exposure t o cannabis or cocaine and AFEVi. .
Tablee 2B shows t h e results after c o n t r o l l i n g f o r age and i n f l u -encee o f m u l t i p l e d r u g use. A l t h o u g h e x p o s u r e t o base-cocaine is posi-t i v e l yy c o r r e l a posi-t e d w i posi-t h exposure posi-t o heroine b y chasing posi-the d r a g o n
•• r2:0.4 p <0.001 associations between cocaine, h e r o i n and AFEVi appear
t oo be s t r o n g e r a f t e r a d j u s t m e n t . The FEVi values showed a decrease w i t hh i n c r e a s i n g h e r o i n i n h a l i n g e x p o s u r e and an increase w i t h increas-ingg e x p o s u r e t o cocaine by i n h a l i n g . This effect is p r o b a b l y caused by patientss w i t h an e x t r e m e l y low AFEVi w h o b e l o n g e d to the h i g h e r h e r o i nn i n h a l i n g e x p o s u r e c a t e g o r i e s b u t n e v e r i n h a l e d c o c a i n e •• Figure / •
Iff the t w o m o s t e x t r e m e values are o m i t t e d f r o m t h e a n a l y s i s , t h ee AFEVi s t i l l s h o w s a s i g n i f i c a n t decrease w i t h increasing exposure t oo heroin i n h a l a t i o n - p = 0.05- The s i g n i f i c a n t a s s o c i a t i o n w i t h cocaine, however,, d i s a p p e a r s -p = 0.32- Similarly, if t h e square %FEVi is used t o l i m i tt t h e i n f l u e n c e o f t h e l o w e r v a l u e s , t h e m u l t i v a r i a t e analysis s t i l l showss a s i g n i f i c a n t association p = 0.03- w i t h h e r o i n e x p o s u r e and increasee o f sguare %FEVi w i t h increasing cocaine i n h a l a t i o n disappears
7 2 2
P M A M A H U A M F F
C h a s i n gg t h * d r a g o n a n d I u a g f u n c t i o n
f:f: continuous variable: ast: 0~ 20 years, increment - 10 years; packyears: incrementincrement — JO years
*:: ordinal variables lowest through highest category ranging 0 through J #:: cfkhatomous variable; 'no'/"yes'values V/'l'
A»» Univariate relation with AFEVt litre Aget t
Cigarettes* * Cannabis* *
Cocainee smoking* H e r o i nn smoking*
B,B, Multivariate regression analyses adjustmentadjustment for exposure effects
Aget t Cigarettes* * Cannabis* *
Cocainee smoking* H e r o i nn smoking*
C.C. Additional adjustment for BMIBMI < 18 / history of T8 or pneumonia /
complaintscomplaints of bronchial hyper-responsiveness
B M I << 1 8 * Historyy TB / Pneumonia* Bronchiall Hyper-responsiveness* Aget t Clgarettest t Cannabis* * Cocainee smoking* H e r o i nn smoking**
EstimatedEstimated AFEVt among non-exposure group 2B:2B: 0.32 (95%Q Bèta a - 0 . 1 7 7 -Ö.0S S 0 J 1 1 0.15 5 - 0 . 5 7 7 - 0 . 1 5 5 -- 0.05 0.06 6 0.52 2 - 0 . 7 8 8 - 0 . 3 3 3 - 0 . 2 S S - 0 . 4 9 9 - 0 . 1 3 3 - 0 . 0 5 5 0.05 5 0.48 8 - 0 . 6 1 1 -0.34,0.98),-0.34,0.98), 2C: 0.57(95%Ci: 95%% CI - 0 . 4 22 ; 0.07 - 0 . 1 55 ; 0.05 -0.29-0.29 ; 0.5) ; 0.58 -1.03-1.03 ; -0.11 -0.40-0.40 ; 0.10 ; 0.05 -0.32-0.32 ; 0.44 0.050.05 ; 0.97 1.28;1.28; -0.28 -0.64-0.64 ; 0.02 -0.61-0.61 ; 0.11 0.760.76 ; -0.21 -- 0.36 : 0.11 ; 0.04 -0.30-0.30 ; 0.40 0.060.06 ; 0.90 1.081.08 ; -0.15 constantconstant -0.09:1.23) -0.09:1.23) p-value e 0.16 6 0.33 3 0.58 8 0.49 9 0.02 2 0.23 3 0.35 5 0.75 5 0.03 3 0 . 0 0 3 3 0.03 3 0 . 1 7 7 0.001 1 0 . 0 9 9 0 . 4 9 9 0.88 8 0.03 3 0.01 1
TableTable 2 Linear regression analysis: exposuree versus AFEV1
AA BMI lower than 18 was observed in 26% of the patients, where-ass 49% reported complaints of bronchial hyper-responsiveness and 2\%2\% a history of TB 7 patients lifetime- or Pneumonia - 17 patients duringduring past two years- Table 2C shows the influence of these poten-tiall confounders on the relation between heroin and FEVi. In the uni-variatee analysis the variables history of TB or pneumonia and com-plaintsplaints of bronchial hyper-responsiveness are significantly related to heroinn exposure - X* linear by linear association: p < 0.05, p < 0.01 - The variable
C h a s i n g g tt h • d r a g o n n a n d d
P M A M A H U A M P P
l u n gg f n n c t l o 7 3 3
f:f: continuous variable
ageage :0~2Q years, increment - i Q years; pack/ears: increment - 10 years *:*: ordinal variables lowest through highest category ranging 0 through t #;; dichotomous variable;'no'/' yes'values 'ff/'V
A.A. Univariate relation with dyspnoea
Aget t Cigarettess t Cannabis* * Cocainee smoking* Heroinn smoking* Oddss ratio 1.S S 1.3 3 0.9 9 t.5 5 5.2 2 95%% CI 0.70.7 ; 3.2 1.01.0 ; 1.7 0.30.3 ; 3.0 0.40.4 ; S.5 1.1;1.1; 26.7 p-value e 0.33 3 0.07 7 0.81 1 0.59 9 0.03 3
ff.ff. Multivariate regression analyses
adjustmentadjustment for exposure effects
Aget t Cigarettest t Cannabis* * Cocainee smoking* Heroinn smoking* 1.2 2 1.3 3 0.9 9 0.6 6 7.6 6 0.50.5 ; 2.9 0.90.9 i 1.8 0.20.2 ; 3.2 0.10.1 ; 2.6 1.31.3 ; ASA 0.67 7
o.n n
0.84 4 0.46 6 0.03 3C.C. Additional adjustment for
BMlBMl < 18/ history of TB or pneumonia /
complaintscomplaints of bronchial hypèr-rèsponsiveness
BMl<< 18* Historyy TB / Pneumonia* Bronchiall Hyper-responsiveness* Aget t Cigarettest t Cannabis* * Cocainee smoking* Heroinn smoking* 0.9 9 3.8 8 3.3 3 1.0 0 1.3 3 0.8 8 0.5 5 5.6 6 0.30.3 ; 3.3 1.11.1 ; 12.8 1.01.0 ; 10.3 0.40.4 ; 2.6 0.90.9 ; 1.9 0.20.2 ; 3.5 0.10.1 ; 2.6 0.80.8 ; 4 0 . 9 0.88 8 0.03 3 0.04 4 0.93 3 0 J 0 0 0.82 2 0.43 3 0.09 9
TableTable 3 Logistic regression analysis:: exposure versus dyspnoea
BMlBMl < 18 did not show a significant relation with the level of heroin exposure.. In the univariate analysis all 3 variables showed a signifi-cantt associated with AFEVi -data not shown- In the multivariate ana-lysiss only the variables complaints of bronchial hyper-responsiveness andd BMl<18 were significantly related to the AFEVi. The higher preva-lencee of methadone patients with symptoms of bronchial hyper-responsivenesss and exposure to heroin by inhaling partly explains the relationn between heroin and AFEVi.
P M A M A H U A M P P
7 44 C h a s i n g t h e d r a g o n • n d l u n g f u n c t i o n
DyspnoeaDyspnoea and exposure to drugs
Tablee 3 presents the results of the logistic regression analysis of the relationn between dyspnoea and the exposure to drugs. These results showw large similarities with those of the linear regression analyses describedd above. Among the methadone patients, heroin exposure was associatedd with an increased prevalence of dyspnoea. Reported differ-encess of exposure to cocaine, cannabis or cigarettes were not statisti-callyy significant related to the prevalence of dyspnoea. Dyspnoea was reportedd more frequently among patients with complaints of bronchial hyper-sensitivity.. In contrast to results of the linear regression analy-sis,, no significant association between dyspnoea and BMl is observed. Thee association between 'history of TB and/or pneumonia' and heroin exposuree by 'chasing the dragon' partly explains the relation between heroinn and prevalence of dyspnoea -data not shown- After additional adjustmentt the odds ratio of heroin exposure and dyspnoea is not sig-nificantlyy higher than one.
Analysiss performed with a more narrow definition of dyspnoea •• 13% that reported to experience a short of breath while walking at theirtheir own pace or in rest- showed similar results. However, due to the loww number of values per cell odds ratios and confidence intervals are inflatedd 'data not
shown-D i s c u s s i o n n
Thiss study demonstrates that methadone patients show an impaired lungg function both if we consider complaints of dyspnoea and AFEVi andd 96FEVi. In a sample of the general population with a comparable agee and gender distribution only 1% with serious complaints of dys-pnoeaa -when walking at his own pace or at rest- would have been expected.1200 In this study 13% of the patients reported such complaints off dyspnoea. The median %FEVi was 93 *sd = 20- The average 9éFEVi observedd at a population survey conducted within various regions of thee Netherlands - N = 2589 • was 108 • sd = 17 - "7
Furthermore,, data suggest a relation between heroin use by chasingg the dragon and an impaired lung function. We were able to constructt subgroups reflecting differences in life-time exposure to heroinn by chasing the dragon by taking into account differences in routess of administration -injecting versus chasing the dragon-, differ-encess in duration of inhaling heroin and frequency of inhaling heroin duringg the last month of use. AFEVi values decreased and prevalence off dyspnoea increased with increasing heroin exposure. Bias towards zeroo is possible if heroin inhalers lowered their consumption due to an impairedd lung function. Due to limitations of a transversal study and limitedd accuracy of life-time exposure to heroin quantification of thé effectt is not possible. The effect of patients who failed spirometric
P M A M A H U A M P P
C h a s i n gg t h • d r a g o n a n d l u n g f u n c t i o n 7 fl
testt is expected to be limited. They were older average 49 years- but exposuree to heroin or cigarettes nor the severity of the reported symp-tomss of dyspnoea significantly differed from the others.
Ann impaired pulmonary function was not shown among all sub-jectss of the highest category of exposure to heroin by inhaling. So
probably,, if inhaling of heroin is causally related to an impaired lung function,, not all heroin users are equally vulnerable. In this respect a similarityy with cigarette smokers may arise. Non-specific airway responsivenesss is considered to predispose some cigarette smokers to developp Chronic Obstructive Pulmonary Disease. Only \S% of the cigarettee smokers will eventually develop COPD.121 Among heroin inhalerss there may be a susceptible subpopulation too. For example, it couldd be hypothesised that those who inhale heroin and also show a histaminee release after exposure to opiates are specially vulnerable.
Bronchiall hyper-responsiveness and history of TB/pneumonia seemm to be partly responsible for the association between heroin and AFEVii and dyspnoea respectively. It is not exactly clear whether these variabless should be considered as confounders, intermediate factors, orr symptoms of pulmonary impairment. If we consider these variables ass intermediate factors or symptoms, additional adjustment would be incorrect.. The variables could also be confounders. A history of TB and pneumoniaa for example, could be indicative for poor living conditions andd therefore occur more prevalent among those with the highest exposuree to heroin administered by inhaling. If this is the case, adjust-mentt is justified. However, those who are not exposed to heroin by chasfngg the dragon, injected their heroin and their living conditions aree not expected to be better.
Consideringg a causal mechanism it may be important to realise thatt the purity of heroin that is inhaled in Amsterdam is approximate-lyy 309É. Caffeine is the main adulterant.1" However, caffeine, adminis-teredd orally, is a bronchodilator and appears to improve lung function inn people with asthma.123
Alll but two study participants smoked tobacco, the average exposure off cigarettes among the total group is estimated to be 20.2 packyears. Althoughh the validity of a retrospectively calculated exposure variable isis limited,'24 We may assume that part possibly most- of the pul-monaryy impairment in this population can be attributed to the smok-ingg of cigarettes. Moderate to heavy cigarette smoking men have, on average,, a 15 ml/year larger decline of FEVt than non-smokers.125 Unfortunately,, we cannot transpose these findings to this transversal studyy because people with a low %FEVi are at higher risk for mor-tality.1266 No statistical significant relation between exposure to tobac-coo and AFEVi or dyspnoea was determined. This should be interpreted ass the result of a lack of contrast among the methadone clients 'almost'almost everybody continuously smoked cigarettes- rather than a lack
P M A M A H U A M P P
7 11 C h a s i n g t h • d r a g o n a n d l u n g f u n c t i o n
off effect. Moreover, if heavy cigarette smokers lowered their con-sumptionn due to an impaired lung function, the reported contrast of tobaccoo exposure is likely to be lower than the reai contrast -again, biasbias towards zero- This could decrease the capability of the cigarette exposureexposure variable to control for potential confounding. In this study, thee relation between heroin smoking and AFEVi or dyspnoea did not
decreasee after adding the number of pack years to the analyses. Moreover,, the number of packyears of tobacco exposure as reported wass not significantly related to the level of heroin exposure rï:0.03; pp = 0J6- This suggests that the real confounding bias will be limited.
Nevertheless,, in order to improve the lung function, limitation off the exposure to cigarettes calls for special attention.t 2 7 m Cannabis exposuree was not significantly related tö a decreased AFEVi or increasedd prevalence of dyspnoea. Other studies concerning the effect off cannabis on pulmonary function offer conflicting results. Tashkin concludedd that the FEVt did not decline due to cannabis smoking'"but Taylorr observed altered spirometric results -FEVi/FVC- among
non-tobaccotobacco smoking- cannabis dependent individuals at age 21 years.130 Moreover,, it is suggested that airway inflammations are more
preva-lentt among combined marijuana and tobacco smokers.1" The appar-entlyy positive effect of cocaine on AFEVi is considered to be an arte-factt and caused by the effect of a few subjects with a severely impairedd lung function who never smoked base-cocaine. Possibly they didd not start using base-cocaine because of their impaired lung func-tion.. In Amsterdam, cocaine smoking started during the 1980s, and increasedd during the 1990s. Although a negative effect of base cocainee on FEVi has not been observed, other physical problems may hinderr oxygen uptake. Inhaling base-cocaine is associated with pul-monaryy inflammation and infiltration on X-ray1" and may cause a decreasedd diffusion of oxygen from the lungs to the b l o o d .1"I ï 4
C o n c l u s i o n n
Thee results of this transversal study suggests that chronic heroin use byy chasing the dragon is related to an impaired lung function and higherr prevalence of dyspnoea. More research is needed to quantify thee possible effect, to disentangle the effect of smoking heroin and tobaccoo and to identify whether and what kind of particularly vulner-ablee subpopulations exists.
Acknowledgement Acknowledgement
ThisThis study benefited from financial support of of the Central CommitteeCommittee Treatment of Heroin Addicts CCBH, Utrecht andand the Municipal Health Service, Amsterdam.
