On the limitations of the tuberculous meningitis diagnosis

Dear Editor,

In the original article “Mortality and sequelae of tuberculous meningitis in a high-resource setting: A cohort study, 1990–2017”,¹ the authors study the mortality and sequelae of tuberculous meningitis in a retrospective cohort of patients from 1990 to 2017. To make the diagnosis of this clinical condition, a uniform case definition was used based on expert opinion,² which highlights the difficulties of diagnosis posed by tuberculous meningitis. This definition was intended to try to facilitate the comparison of research data, rather than for use in clinical practice. Persons with microbiological or molecular confirmation of M. tuberculosis in the CSF are assigned confirmed tuberculous meningitis. The remaining cases are defined as probable or possible, using an expert-generated classification system where higher scores designate an increased likelihood of tuberculous meningitis. True cases are expected to be represented by all definite cases, most probable cases and some possible cases.³ It is striking to us that the authors of the article published in your journal have conducted a joint analysis of patients with a confirmed diagnosis along with other cases where said diagnosis is only probable or possible. In Table 1 of their article, these last two groups represent 79 cases of a total of 135 patients analysed (nearly 60% of the total). Given that we do not have absolute certainty that these patients indeed suffer from tuberculous meningitis, we believe that it would have been more appropriate to study only patients with a confirmed diagnosis. Having included a heterogeneous group of patients in their analysis (with confirmed, probable and possible diagnoses), it is possible that the conclusions obtained have certain limitations.