Lyme disease in British Columbia: Are we really missing an epidemic?
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
B. Henry, MD, MPH, FRCPC, M. Morshed, PhD, SCCM
Lyme disease in British
Columbia: Are we really
missing an epidemic?
Results from surveillance and research on Lyme disease suggest
there is a real but low risk of contracting this tick-borne illness in BC.
yme disease is a tick-borne
L
ABSTRACT: The risk of Lyme disease endemic areas of the eastern United
depends on climate, geography, the States (29 per 100 000). There is no zoonosis caused in North
abundance of specific insect vec- evidence to support an epidemic of America by infection with
tors, and human interaction with Lyme disease in BC. Responses to a the spirochete Borrelia burg-
these. In BC, Ixodes pacificus, the recent survey indicate that physi- dorferi. Humans acquire Lyme dis-
primary tick vector for the cause- cians generally are aware of the low ease through the bite of an infected
ative spirochete, Borrelia burgdor- but real risk of Lyme disease, know tick.1 The principal tick vector in BC
feri, has consistently been found in to treat patients with clinical symp- is the Pacific black-legged tick, Ixodes
low numbers in populous areas, and toms, and understand that Lyme dis- pacificus,2 which is found throughout
rates of infection in this tick remain ease is preventable and treatable. the highly populated areas of southern
at less than 1%. Correspondingly, Public health authorities will contin- BC. This situation is in contrast to
rates of human cases of Lyme dis- ue to remind residents and visitors eastern Canada and the US, where the
ease in BC are less than 0.5 per to BC of the simple measures they tick Ixodes scapularis is the most
100 000 per year; this is similar to can take to prevent tick bites and ex- common vector. The low incidence of
rates reported in US states with envi- posure, as well as which early signs Lyme disease in BC may be explained
ronmental epidemiology like BC’s and symptoms should lead them to by the fact that I. pacificus is a less
and considerably less than in high seek appropriate medical treatment. competent vector than I. scapularis, is
less abundant, and is less likely to feed
on deer mice.3-5 Studies have shown
that infectivity rates are lower in areas
where I. pacificus predominates than
in areas where I. scapularis predomi-
nates. Lyme disease advocacy groups
in BC have expressed concern that an
Dr Henry is medical director, Vectorborne
Diseases Program, BC Centre for Disease
Control. She is also an assistant professor
in the School of Population and Public
Health at the University of British Colum-
bia. Dr Morshed is program head of Zoonot-
ic Diseases and Emerging Pathogens,
Provincial Health Services Authority, Public
This article has been peer reviewed. Health Reference Laboratories.
224 BC MEDICAL JOURNAL VOL. 53 NO. 5, JUNE 2011 www.bcmj.orgLyme disease in British Columbia: Are we really missing an epidemic?
epidemic is being ignored. Surveil-
Ecological Niche
lance and research on Lyme disease in
BC indicate this is not the case. Optimal
Tick and mouse Potential
surveillance
The BC Centre for Disease Control Not suitable
has actively screened ticks in over 125
areas of the province. From 1993 to
1996, 10 056 ticks were tested and 40
were found positive for B. burgdorferi
(0.40%). From 1997 to 2007, 8602
ticks were tested and 30 ticks were
found positive (0.35%), demonstrat-
ing a stable, low prevalence of infec-
tion. I. pacificus ticks were found most
commonly in the Lower Mainland and
Vancouver Island; Dermacentor ander-
soni, which is not a competent vector
for Lyme disease, was the tick identi-
fied most commonly throughout BC.
0 100 200 300 400
On Vancouver Island, active drag-
ging for ticks at 17 sites yielded only KM
41 ticks, mostly I. pacificus, all of
Figure 1. Forecasted ecological niche of Borrelia burgdorferi in British Columbia.
which were found negative for B.
burgdorferi. Active solicitation of
ticks from veterinarians on Vancouver population we tested 3500 deer mice north of N51° latitude ( Figure 1 ).
Island and the Gulf Islands led to 115 by culture and found 30 (0.83%) pos- There is concern that global warm-
tick submissions, all of which were itive. We also tested 164 mice for anti- ing could lead to expansion of the eco-
found negative for B. burgdorferi. In bodies to B. burgdorferi and found 6 logical niche for the vector, resulting
the Okanagan over 2 years, 5557 D. (3.66%) positive, demonstrating a low in the potential for increased exposure
andersoni ticks were collected (no I. prevalence in this reservoir. to infected ticks in BC.7 Our models
pacificus ticks were found). Of 110 indicate a modest geographic range
ticks randomly tested for B. burgdor- Ecological niche modeling expansion for Ixodes ticks and B. burg-
feri by culture and PCR, all were In order to identify areas with risk of dorferi based on 2050 climate warm-
found negative. A total of 219 deer Lyme disease transmission in BC, we ing projections; however, the areas
mice were trapped from the same undertook ecological niche modeling where expansion might occur are
areas and tested for antibodies to B. for both ticks and B. burgdorferi infec- sparsely populated and the densely
burgdorferi by the National Microbi- tion and assessed the potential impact populated centres of southern and in-
ology Laboratory in Winnipeg, and all of climate change. Modeling identi- terior BC are already within the exist-
were found to be negative. fied optimal environmental condi- ing ecological niche of B. burgdor-
We receive 800 to 1000 ticks from tions in south coastal BC (i.e., Van- feri. There are also variable local
physicians, veterinarians, and the pub- couver Island, Lower Mainland, and habitats within regions, so exact risks
lic every year. Approximately half are Sunshine Coast) and interior BC val- within a region may vary greatly.
I. pacificus, of which one to two per ley regions.6 The habitat in these areas Lyme disease may also, though
year are found to be positive for B. is characterized by low-lying vegeta- rarely, be acquired from “adventi-
burgdorferi. tion such as high grass and brush, with tious” ticks that drop off migrating
The major mammalian reservoir abundant leaf litter and a nearby water birds. These ticks pose a theoretical
for B. burgdorferi in BC is the deer source. Niche modeling demonstrates risk of infection during the summer
mouse. To assess prevalence in this that B. burgdorferi is generally absent months throughout the province.8
www.bcmj.org VOL. 53 NO. 5, JUNE 2011 BC MEDICAL JOURNAL 225Lyme disease in British Columbia: Are we really missing an epidemic?
Physician awareness
In 2008 we conducted a survey of
16 Travel physicians in BC. We modified a pre-
Endemic viously validated questionnaire11 to
14
collect data on respondent demogra-
12 phics and general knowledge of Lyme
disease. The survey included ques-
10
tions about geographic risk percep-
Number
8 tion, laboratory testing, and three clin-
ical scenarios. Physicians were also
6 asked whether they were aware that
Lyme disease is reportable.
4
We sent questionnaires to all
2 pediatricians, internists, and family
practitioners who gave a BC address
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 as their practice address. The response
rate was 32% (1673/5199). Of these
Year respondents, 148 (8%) recalled diag-
nosing 221 cases of Lyme disease in
Figure 2. Number of endemic and travel-related cases of Lyme disease in BC between 1997 2007, while 58% of family physicians
and 2008. Travel-related cases include patients with exposure histories in countries with
endemic Lyme disease.
and 66% of specialists indicated they
knew Lyme disease is reportable.
Physicians scored high on the
Human surveillance knowledge questions, with over 90%
We used capture-recapture methodolo- to that of Washington and California, correctly identifying the signs and
gy and a review of BC’s three sources with yearly incidence rates of less than symptoms of Lyme disease as well as
of passive surveillance (laboratory 0.5 per 100 000 population (including the causative agent and incubation
data, enhanced surveillance forms, travel-related cases). These rates have period. Fewer were aware that erythe-
and cases reported on the public health remained stable over the past 10 years. ma migrans ( Figure 3 ) on its own is
information system) to estimate the In contrast, the incidence of Lyme dis- diagnostic. The mean overall knowl-
annual number of Lyme disease cases ease in the 10 highly endemic states in edge score was 74% (8.9/12).
in BC between 1997 and 2008, to US is 29.2 per 100 000 people,9 indi- Three clinical scenarios were pre-
develop a more accurate estimate of cating important differences in both sented: Scenario 1 involved a patient
the burden of disease, and to compare disease risk and burden of illness. with erythema migrans and no labor-
BC with Washington, California, and Capture-recapture methods show atory testing; more than half (57%)
high-endemic areas in the eastern US. underreporting of Lyme disease does of all respondents answered correctly
Ninety-three cases of Lyme disease occur in BC. The best model places (“give antibiotics at this time”), while
were identified over the 12-year period the corrected number of cases in BC one-third opted to first test for Lyme
( Figure 2 ); 45 patients (48.4%) were between 1997 and 2008 at 142 (95% disease. Scenario 2 involved an asymp-
male and most were between 41 and CI:111-224), for a maximum incidence tomatic patient with history of a tick
70 years of age (mean age 43.7 years, of 0.5 per 100 000 population. Under- bite; 56% indicated correctly they
range 3.5–80.6). One-third of patients reporting is common for rare diseases would educate and reassure the pa-
acquired their illness outside of BC. when surveillance is passive.10 Wheth- tient. Scenario 3 involved a patient
The mean age of patients with travel- er underreporting results from clinical with arthritis, no history of erythema
related disease was 40.4 years, young- cases being treated without testing migrans, and multiple negative tests
er than those who acquired Lyme dis- and not reported or whether cases are for Lyme disease; 82% correctly re-
ease in BC (mean age 44.8 years). truly not diagnosed is not known. To ported they would investigate causes
The annual incidence rate of Lyme help gain insight into this question other than Lyme disease, or refer the
disease in BC ranged from 0.1 to 0.3 we looked at physician awareness of patient to a specialist.
per 100 000 population, a rate similar Lyme disease in BC. Several questions addressed phy-
226 BC MEDICAL JOURNAL VOL. 53 NO. 5, JUNE 2011 www.bcmj.orgLyme disease in British Columbia: Are we really missing an epidemic?
sicians’ perceptions of risk in their or symptoms and positive laboratory
community of practice. When asked tests, establishes the diagnosis of Lyme
about their patients’ risk of develop- disease.13 Erythema migrans is an an-
ing Lyme disease after a tick bite, 94% nular, slowly expanding erythematous
of respondents indicated they believed lesion, usually 5 cm or greater in diam-
their patients faced some risk. Logis- eter, that may exhibit partial central
tic regression modeling showed that clearing or central necrosis, giving a
physicians have a good understanding bull’s-eye appearance. Erythema mi-
of the spatial distribution of risk with- grans typically occurs 7 to 14 days
in the province, with greater risk per- after infection (range 3 to 30 days),
ceived in areas where ecological con- and in some cases secondary lesions
ditions are most suitable for disease may occur.14 In contrast, a localized
transmission. tick-bite reaction occurs within hours
The final questions of the survey of the bite, expands over hours, and
addressed physician perceptions of resolves within 48 hours. In studies,
patients requesting evaluation for erythema migrans rash occurs in at
Lyme disease because of nonspecific least 80% of all patients with Lyme
symptoms such as fatigue and muscu- disease and 90% of children.15 Pa-
loskeletal pains. While a majority tients can also experience symptoms
(79% of family physicians and 72% of of fatigue, chills, fever, headache, and
specialists) indicated they believed migratory arthralgias, and lymphaden- Figure 3. Erythema migrans.
that their patients’ symptoms were opathy, which may last several weeks
Source: CDC/James Gathany
caused by something other than Lyme if untreated.
disease, 31% of family physicians and Untreated infection can spread over
12% of specialists reported they had several weeks or months and lead to Most cases of Lyme disease are
treated such patients for Lyme disease three main syndromes: successfully treated with antibiotics.
because of patient concern. • Neurological. Neurological abnor- Treatment is most effective if begun
This survey shows physicians are malities can include aseptic menin- early in the course of illness. Howev-
knowledgeable about and aware of the gitis, cranial neuritis, Bell palsy, and er, a small percentage of patients have
risk of Lyme disease in BC, despite radiculoneuritis. Such abnormalities lingering symptoms that last months
the province being a low endemic affect about 5% of untreated patients. to years even after appropriate treat-
area. It is also apparent physicians in • Musculoskeletal. Musculoskeletal ment. Symptoms include muscle and
BC are comfortable with treating pa- manifestations can include migrato- joint pain, arthritis, cognitive defects,
tients empirically, in many cases based ry joint and muscle pains without sleep disturbance, and fatigue. The
on patient concern. More cases are objective signs of swelling. cause of these symptoms is not known,
clinically diagnosed and treated than • Cardiac. Rarely occurring cardiac although there is some evidence that
are reported to public health. manifestations can include atrioven- they result from an autoimmune res-
tricular block and acute myoperi- ponse. Long-term antibiotic treatment
Clinical picture carditis. has been found to be of no benefit in
Ticks are most likely to transmit infec- Weeks to years after onset of in- patients with long-term symptoms,
tion after being attached for more than fection (mean 6 months) episodes of and has been associated with some-
24 hours of feeding, making prompt swelling and pain in large joints (espe- times severe adverse effects, includ-
detection and removal of ticks a key cially the knees) can occur in up to ing death.17,18
way to prevent Lyme disease. A tick 60% of untreated patients, leading to In addition, a group of patients
attached for less than 24 hours is un- chronic arthritis. Some patients devel- with nonspecific symptoms such as
likely to transmit infection, even if it op chronic axonal polyneuropathy or fatigue, memory changes, and muscu-
is infected with B. burgdorferi.12 encephalopathy. Lyme disease is rare- loskeletal pain have been identified
Erythema migrans diagnosed on ly fatal, although patients with late by some physicians as suffering from
physical examination, even in the disseminated disease can have severe, “chronic Lyme disease” despite mul-
absence of other Lyme-specific signs chronic, and disabling symptoms.16 tiple negative laboratory tests and no
www.bcmj.org VOL. 53 NO. 5, JUNE 2011 BC MEDICAL JOURNAL 227Lyme disease in British Columbia: Are we really missing an epidemic?
history of acute disease. This syndrome by the US Centers for Disease Control standard and unproven ways. As a
is the subject of ongoing scientific and Association of State and Territori- result, these laboratories can return a
research, with a number of possible al Public Health Laboratory Directors. positive result that is not reproducible
infectious and noninfectious causes • Step 1: Enzyme immnoassay (EIA) by public health laboratories follow-
being investigated. One recently dis- (VIDAS, BioMériux, France). This is ing the internationally recognized pro-
covered virus, Xenotropic murine leu- a very sensitive test, meaning it will tocols. For example, some private US
kemia virus, has been associated with detect almost all true cases of Lyme labs use only one marker for IgM and
a similar syndrome in some studies19 disease but will also react if a patient only three markers for IgG Western
blot tests, which can result in a false-
positive result and a high rate of cross-
reactivity with other infections.21 This
can lead not only to unnecessary treat-
ment for Lyme disease, but can also
Early treatment prevents late prevent patients from receiving treat-
ment for the condition that is actually
complications and should be causing their symptoms.
initiated on clinical suspicion Antibody tests may be negative
early after infection by B. burgdorferi,
pending laboratory results. as it may take several weeks to devel-
op antibodies. If the diagnosis is
unclear and acute serology is nega-
tive, a convalescent test 2 to 4 weeks
later may aid in diagnosis. Because
erythema migrans is considered diag-
but not in others.20 There is a need for has certain other diseases, including nostic for Lyme disease, a patient pre-
further research into the cause of this mononucleosis, lupus, and various senting with this distinctive rash
syndrome and therapeutic options for microbial infections. requires no further testing. However,
people who are suffering from its • S t e p 2 : We s t e r n b l o t ( W B ) a patient with erythema migrans treat-
debilitating symptoms. (MARDX, Trinity Biotech Co., CA, ed early with antibiotics may not
US). This test is conducted on a develop antibodies. If the patient pres-
Diagnostic testing in BC specimen that yields positive or ents with other symptoms of Lyme
With the exception of direct detection equivocal/indeterminate EIA results. disease and erythema migrans is atyp-
of B. burgdorferi from biopsy speci- It is a very specific test that can dis- ical or absent, serologic testing should
mens of erythema migrans rash, there tinguish between true-positive and be done at initial presentation and
is no validated direct test for the B. false-positive results from the EIA. repeated after 2 weeks. Early treat-
burgdorferi bacterium in blood or Western blot IgM is considered reac- ment prevents late complications and
other samples and the organism can- tive when two markers (out of three should be initiated based on clinical
not be easily cultured. Laboratory test- required markers) are identified, and suspicion pending laboratory results.
ing for Lyme disease relies on detec- Western blot IgG is considered reac- Even after curative antibiotic treat-
tion of antibodies in blood. Because tive when five markers (out of ten ment, antibodies may persist in the
antibodies to B. burgdorferi proteins required markers) are identified. blood for years, meaning that a posi-
can be induced by infection with These tests, taken together, show tive antibody test after treatment does
microbes other than B. burgdorferi,13 whether a patient has ever been ex- not indicate treatment failure. Because
antibody tests can yield false-positive posed to B. burgdorferi. Positive results of long-term persistence of antibod-
results unless properly interpreted. do not demonstrate active infection ies, asymptomatic patients should not
A two-step process to test for evi- and must be interpreted in light of be retested, as a positive test result can
dence of Lyme disease is used by the patient history and symptoms. be misleading.
Public Health Laboratories of the BC Some commercial laboratories use In BC for the past 10 years, PHSA
Provincial Health Services Authority either discredited tests (such as urine laboratories have consistently receiv-
(PHSA), following recommendations antigen tests) or interpret tests in non- ed about 3000 patient samples for
228 BC MEDICAL JOURNAL VOL. 53 NO. 5, JUNE 2011 www.bcmj.orgLyme disease in British Columbia: Are we really missing an epidemic?
serological testing for Lyme disease Emerg Infect Dis 2006;12:604-611. 14. Wormser GP, Dattwyler RJ, Shapiro ED,
yearly. Of these specimens, about 90 3. Steere AC, Coburn J, Glickstein L. The et al. The clinical assessment, treatment,
have had positive or indeterminate emergence of Lyme disease. J Clin and prevention of Lyme disease, human
results on EIA and 7 to 12 cases of Invest 2004;113:1093-1101. granulocytic anaplasmosis, and babesio-
Lyme disease have subsequently been 4. Dolan MC, Maupin GO, Panella NA, et al. sis: Clinical practice guidelines by the
confirmed by WB. Vector competence of Ixodes scapularis, Infectious Diseases Society of America.
I. spinipalpis, and Dermacentor ander- Clin Infect Dis 2006;43:1089-1134.
Conclusions soni (Acari:Ixodidae) in transmitting Bor- 15. Gerber MA, Shapiro ED, Burke GS, et al.
There is no evidence to support an epi- relia burgdorferi, the etiologic agent of Lyme disease in children in southeastern
demic of Lyme disease in BC. The pri- Lyme disease. J Med Entomol 1997; Connecticut. Pediatric Lyme Disease
mary vector, I. pacificus, is found in 34:128-135. Study Group. N Engl J Med 1996;335:
populous areas in consistently low 5. Casher L, Lane R, Barrett R, et al. Rela- 1270-1274.
numbers, and rates of infection in the tive importance of lizards and mammals 16. Feder HM Jr, Johnson BJ, O’Connell S,
tick population remain less than 1%. as hosts for ixodid ticks in northern Cali- et al. A critical appraisal of “chronic Lyme
Human case rates in BC are less than fornia. Exp Appl Acarol 2002;26:127-143. disease.” N Engl J Med 2007;357:1422-
0.5 per 100 000. 6. Mak S, Morshed M, Henry B. Ecological 1430.
A recent survey of clinicians con- niche modeling of Lyme disease in 17. Patel R, Grogg KL, Edwards WD, et al.
firms doctors have good knowledge British Columbia, Canada. J Med Ento- Death from inappropriate therapy for
of Lyme disease, are comfortable mol 2010;47:99-105. Lyme disease. Clin Infect Dis 2000;31:
making a diagnosis given clinical 7. Brownstein JS, Holford TR, Fish D. Effect 1107-1109.
signs and symptoms, and appropriate- of climate change on Lyme disease risk 18. Holzbauer S, Kemperman M, Lynfield R.
ly use laboratory testing to assist in in North America. Ecohealth 2005;2:38- Death due to community-associated
diagnosis for those patients suffering 46. Clostridium difficile in a woman receiv-
nonspecific signs and symptoms. 8. Morshed M, Scott J, Fernando K, et al. ing prolonged antibiotic therapy for sus-
Further research is needed to devel- Migratory songbirds disperse ticks pected Lyme disease. Clin Infect Dis
op diagnostic tests and treatment pro- across Canada, and first isolation of the 2010;51:369-370.
tocols for patients suffering from non- Lyme disease spirochete, Borrelia burg- 19. Lombardi V, Ruscetti F, Das Gupta J, et
specific, debilitating symptoms that dorferi, from the avian tick, Ixodes aurit- al. Detection of an infectious retrovirus,
some physicians attribute to “chronic ulus. J Parasitol 2005;91:780-790. XMRV, in blood cells of patients with
Lyme disease.” Otherwise, Lyme dis- 9. Centers for Disease Control and Preven- chronic fatigue syndrome. Science 2009;
ease is a preventable and treatable tion. Lyme disease—United States, 326: 585-595.
illness, and public health authorities 2003-2005. MMWR Morb Mortal Wkly 20. Erlwein O, Kaye S, McClure MO, et al.
will continue to remind residents and Rep 2007;56:573-576. Failure to detect the novel retrovirus
visitors to the province of the simple 10. Doyle TJ, Glynn MK, Groseclose SL. XMRV in chronic fatigue syndrome. PLoS
measures they can take to prevent tick Completeness of notifiable infectious One 2010;5:e8519.
bites and exposure as well as the signs disease reporting in the United States: 21. US Centers for Disease Control and Pre-
and symptoms of acute illness so they An analytical literature review. Am J Epi- vention. Notice to readers: Caution re-
can seek appropriate medical treat- demiol 2002;155:866-874. garding testing for Lyme disease.
ment. 11. Magri JM, Johnson MT, Herring TA, et al. MMWR Morb. Mortal Wkly Rep 2005;
Lyme disease knowledge, beliefs, and 54:125.
Competing interests practices of New Hampshire primary
None declared. care physicians. J Am Board Fam Pract
2002;15:277-284.
References 12. Bratton RL, Whiteside JW, Hovan MJ, et
1. Hengge UL, Tannapfel A, Tyring SK, et al. al. Diagnosis and treatment of Lyme dis-
Lyme borreliosis. Lancet Infect Dis 2003; ease. Mayo Clin Proc 2008;83:566-571.
3:489-500. 13.Aguero-Rosenfeld ME, Wang G,
2. Hanincova K, Kurtenbach K, Diuk- Schwartz I, et al. Diagnosis of Lyme
Wasser M, et al. Epidemic spread of Lyme borreliosis. Clin Microbiol Rev 2005;
borreliosis, northeastern United States. 18:484-509.
www.bcmj.org VOL. 53 NO. 5, JUNE 2011 BC MEDICAL JOURNAL 229You can also read
