Z. Arvanitakis et al., MYCOBACTERIUM-TUBERCULOSIS MOLECULAR VARIATION IN CNS INFECTION - EVIDENCE FOR STRAIN-DEPENDENT NEUROVIRULENCE, Neurology, 50(6), 1998, pp. 1827-1832
Objective: To determine the molecular diversity among Mycobacterium tu
berculosis isolates associated with central nervous system tuberculosi
s (CNS TB) in a defined cohort of HIV uninfected patients. Design/meth
ods: A retrospective analysis was performed of clinical and laboratory
data for all patients with CNS TB diagnosed in Manitoba, Canada, betw
een 1979 and 1996. Restriction fragment-length polymorphisms (RFLP) of
archival isolates of M. tuberculosis from CNS TB patients were determ
ined and interpreted against the frequency of different isolates from
all TB patients in the years 1992 to 1996, Results: Among 2,334 patien
ts with active TB, CNS TB was diagnosed in 42 (1.8%); meningitis with
or without tuberculoma in 76%; and tuberculoma alone in 24%. CNS TB pa
tients were significantly more likely to be young (<40 years old), fem
ale, and of Aboriginal origin. Morbidity (fixed/recurrent CNS deficit)
rate was 29% and mortality rate was 26%. An adverse outcome, either m
orbidity or mortality, was significantly more common in those with men
ingitis. RFLP analysis of isolates (n = 19) from CNS TB patients revea
led 13 distinct restriction patterns with a predominance of the type 1
pattern (n = 6). The frequency of type 1 restriction pattern was sign
ificantly greater in patients with CNS TB compared to all TB patients
in Manitoba. Conclusions: CNS TB continues to have a high morbidity an
d mortality despite modern methods of detection and treatment. Althoug
h several strains of hi. tuberculosis cause CNS TB, the current study
suggests that the occurrence of CNS TB may be strain-dependent.