Neuro-meningeal Tuberculosis in Adult Senegalese Patients: Profile and Outcome of Cases Diagnosed at a Referral Service, from 2015 to 2020 ()
1. Introduction
Tuberculosis is a major public health problem, especially in developing countries [1] . It is one of the top 10 causes of death worldwide [2] . Neuro-meningeal involvement is rare but represents one of the most severe forms with high morbidity and mortality. This localization represents 5% - 15% of extra pulmonary tuberculosis [3] . Among tuberculosis cases, 2% to 5% have a CNS lesion and its frequency rises to 10% in HIV-infected patients. The case fatality rate for untreated Neuro-meningeal tuberculosis (NMT) is almost 100% and delay in treatment often leads to permanent neurological damage in more than 50%, despite anti-tuberculosis treatment [4] . Prompt diagnosis is crucial for successful disease management. Unfortunately, the great clinical polymorphism and the lack of specificity of radiological and biological signs, apart from the identification of BK by the nucleic acid amplification test (GeneXpert MTB/RIF) in the cerebral spinal fluid (CSF), make diagnosis difficult and are frequently responsible for a delay in treatment [4] . Senegal is one of the African countries where tuberculosis has remained a concern until now. And there are no studies carried out on this subject. We undertook this study to describe the profile and the outcome of NMT cases diagnosed in the infectious diseases department (SMIT) of the Fann university hospital in Dakar from 2015 to 2020.
2. Methods
We carried out a retrospective, descriptive and analytical study, reviewing medical records of adults diagnosed with neuro-meningeal tuberculosis at the SMIT of Fann Hospital from January 2015 to December 2020. The SMIT is in a region with a high incidence of tuberculosis in Senegal and serves a population of low socio-economic status in Dakar. Cases were identified by reviewing the database of the hospital’s Department of Epidemiology and the discharge registry of the infectious diseases department. We included all cases that met the case definition.
Our cases were defined on the presence of at least one of the following criteria:
• Positive GeneXpert MTB/rif in cerebrospinal fluid (CSF) or,
• Lymphocytic meningitis is associated with another confirmed TB site that has progressed well under antituberculosis treatment.
We excluded patients who had incomplete records, patients under 18 years of age and those who had a final diagnosis other than NMT.
We recorded demographic and clinical information, laboratory results, drug treatment and outcome at discharge from the clinical files and microbiology department records. CSF macroscopic examination, protein and glucose quantification and cell count were performed at the hospital laboratory. HIV status was determined using HIV 1 + 2 ELISA test with a confirmatory immunofluorescent assay.
Descriptive statistics were performed for clinical, epidemiological and laboratory features. The statistical analyses were performed with Epi info 7.2 and SPSS V 21. Bivariate analysis was used to look for associations between variables, using Chi2 and Fisher tests. The alpha risk of error was set at 5%.
3. Results
Between 2015 and 2020, 5289 patients were admitted to the SMIT. We collected 55 cases of neuro-meningeal tuberculosis, representing a hospital frequency of 1.03% (Table 1) and an average of 9 cases per year. The median age was 38 years (range 16 - 77 years). The most affected age group was patients aged between 20 and 40 years (45.45%, n = 25) (Figure 1). Patients were predominantly male (76.36) with a sex ratio (M/F) of 3.23. Most of patients (45.45%) live in suburban localities of Dakar. A history of contact was found in 14.5% of cases. HIV-infected patients represented 41.82%. A history of tuberculosis was found in 25.5% of cases (Table 2).
![]()
Table 1. Annual incidence of neuro-meningeal tuberculosis at SMIT of Fann Hospital from 2015 to 2020.
![]()
Figure 1. Distribution of NMT cases by age group.
![]()
Table 2. Epidemiological characteristics of NMT at SMIT, Dakar (n = 55).
The delay in consultation was greater than one month in 60% of patients. Headache was the most constant reason for consultation (94.55%), Neck stiffness was present for 94.55% of patients, fever was found in 69.09% of cases and vomiting in 60%. Physical examination revealed meningeal signs in 94.55% of cases, consciousness disorders and intracranial hypertension were noted in 63.64% and 56.36% respectively. Nerve palsy was found in 38.18% and pyramidal syndrome in 36.36% of cases. Behavioral disorders were presented in 29.09% (Table 3).
Lumbar puncture was performed in 89.09%. The CSF was clear in 81.64%. The mean cytology value was 312.69 ± 614.83/mm3. Thirty-five patients (71.4%) had 100% lymphocytic CSF. High protein level (>1 g/L) was noted in 77.50%, and 71.40% had hypoglycorachia (<0.5 g/L). GeneXpert MTB/RIF in CSF was performed in 33 patients and was positive in 4 patients (12.12%); no cases of rifampicin resistance were detected (Table 4). CRP was elevated at 85.45%, the mean value was 59.19 ± 58.55 mg/L.
Brain CT performed in 78.18%, was abnormal in 72.09% of cases. Tuberculoma (27.90%), hydrocephalus (16.27%) and meningeal contrast enhancement (11.62%) were the main lesions. The neuro-meningeal localization was isolated in 61.18%, and in 32.7% it was associated with a pulmonary TB. All patients had been put on anti-tuberculosis treatment (2RHZE/10RH), which had been initiated in 83.64% of cases within a week of hospitalization.
Hyponatremia was the most frequent complication (50.9%). The median hospital stay was 19 days [range 2 - 96 days], and the average length of hospital stay was 22.75 days ± 18.14 days. Twelve patients (21.8%) died during hospitalization. There was no association between HIV status and death (p = 0.8). The presence of seizures was not statistically significantly associated with death (p = 0.46). The Lethality rate was higher in women (46.2% vs. 14.3%; p = 0.01), in patients whose delay in consultation was >1 month (p = 0.03), and in patients who presented consciousness disorders (p = 0.007). We did not find age, HIV status, or the presence of seizures as associated factors of death in our study (Table 5).
![]()
Table 3. Clinical characteristics of NMT at SMIT, Dakar (n = 55).
![]()
Table 4. CSF cytochemistry study in neuro-meningeal tuberculosis cases at SMIT, Fann Hospital (n = 49).
![]()
Table 5. Associated factors with death in neuro-meningeal tuberculosis cases at SMIT, Fann hospital (n = 55).
4. Discussion
In our study, we found a hospital frequency of 1.03% and an annual incidence of 9.16 cases per year. Incidences of neuro-meningeal tuberculosis vary from one series to another [5] [6] . This variability can be explained by the difference between the type of population studied, the conditions under which it was carried out, the study sites and the diagnostic tools used.
Patients were mainly young in our study. These data are similar to those observed in Africa [7] [8] [9] . Indeed, this age group corresponds to the active population, which is the most representative in our countries. The distribution of neuro-meningeal tuberculosis by sex varied according to the different series in the literature. The strong male predominance observed in our study confirms the data in the literature in which male predominance remains classic [5] [6] . However, Ossibibi in Congo in 2020 [6] reported a female predominance (67.9%), while Dollo and al., found in Casablanca in 2017, an equal distribution between men and women [7] .
A low rate of contact history was found by several authors [8] [9] , like our result, while the prevalence of the disease is high in our regions, this situation could be explained by ignorance of the disease by most of the population, the importance of undiagnosed cases making it difficult to assess the notion of contact. A history of TB was found in 25.4% of cases, this result was like the data found in the literature, according to which a history of tuberculosis is frequently found in subjects with meningitis tuberculosis [6] [10] .
The frequency of the association of tuberculosis and HIV infection was found in our study (41.82%), as reported by all authors, particularly the neuro-meningeal localization [6] [7] .
The lengthy consultation periods in this study are similar to those described by several African authors [6] [7] [11] . The long delay in consultation may be due to several factors, particularly the clinical polymorphism of the disease, the delay in suspicion of tuberculosis by clinicians, the lack of medical information or the difficulties of access to local medical services in our context.
The main symptoms found at admission headache were fever and consciousness disorder. These data have also been found by several authors [6] [12] . Cranial nerve palsy remains a frequent manifestation of neuro-meningeal tuberculosis and was found in 38.18% of cases in our study. Similar results were found by Amara et al. [13] . The extra-neurological locations were common in our cases. They were pulmonary in most of the cases. These results are consistent with those obtained by several authors [6] [7] [9] .
NMT is one of meningitis with clear or lymphocyte fluid as reported in our study to significant proportions. The local inflammatory reaction at the meningeal level reflects the hyperproteinorachie found. These results corroborate the data mentioned in the literature by several African authors [6] [7] .
GeneXpert MTB/rif in CSF was positive in 12.12%. The detection of Mycobacterium tuberculosis by microscopic examination of CSF is an important means of confirming the diagnosis, but its sensitivity remains low. Thus, the diagnosis of TNM remains difficult and in most cases relies on a range of arguments. This situation tends to delay the diagnosis and exposes patients to the occurrence of complications, which worsens the prognosis of the disease. Brain CT is an important diagnostic tool; however, abnormalities are highly variable. The incidence of hydrocephalus increases with the duration of the disease and decreases with age. It is particularly frequent in children, found in 87% of cases, whereas it is observed in only 12% of adults [14] .
Hyponatremia, the most frequent systemic complication, was found in our series. This biological abnormality is responsible for consciousness disorders [15] . The long delay in diagnosis is an associated factor in many series [5] [6] , as found in our study. Moreover, we found lethality higher in women (46.2% vs. 14.3%; p = 0.01) without explanation.
The SMIT of Fann University Hospital has been equipped with an archival unit for efficient data management. But this asset did not prevent from finding out certain files with missing information. However, this study has shed light on the epidemiological situation of neuro-meningeal tuberculosis at the Fann University Hospital while notifying the share of HIV infection in a country with limited resources.
5. Conclusion
Tuberculosis remains a public health problem and neuro-meningeal involvement represents one of the most severe forms. Its highly variable clinical expression and the low sensitivity of the GeneXpert MTB/RIF in the CSF often make diagnosis difficult, thus exposing patients to serious complications. Among the factors associated with death, we find consciousness disorders and long delays in diagnosis.
Consent
The patients had signed an informed consent form, which is available.
Authors’ Contributions
Daouda Thioub, and Papa Latyr Junior Diouf wrote the manuscript with input of Agbogbenkou Tevi Déla-dem Lawson and Viviane Marie Pierre Cisse Diallo.
Ndeye Aissatou Lakhe and Ndeye Fatou Ngom Gueye drafted the manuscript for important intellectual content. Moussa SEYDI and Sylvie Audrey Diop drafted and approved the final version to be published.
Funding Statement
This manuscript did not receive a grant.