Improving the diagnostic work-up of neurological disorders in rural Central Africa

Neurological disorders are a complex of symptoms and signs related to the nervous system that may be due to a wide variety of etiologies. Most causative conditions can be severe if left untreated and contribute to the major global mortality and morbidity of this syndrome. In sub-Saharan Africa, the burden is even more considerable because of a wider spectrum of infections, substantial delays in management and lack of diagnostic facilities and specialized care. The main purpose of this PhD dissertation is to improve the diagnostic workup of neurological disorders in rural areas of the Democratic Republic of the Congo (DRC). A large prospective etiological and diagnostic study was conducted from 2012 to 2016 in 351 children > 5 years and adults presenting with a pre-defined set of neurological complaints at the “Hôpital Général de Réference” of Mosango, Kwilu province, DRC. This study on NDs in DRC was part of a larger multicentric research project, conducted by the NIDIAG (Neglected Infections DIAGnosis; htpps://nidiag.eu) consortium, that aimed at improving the diagnosis of Neglected Tropical/Infectious Diseases in challenging syndromes. Considering its main study purpose, this PhD thesis had three specific objectives: •Firstly, to investigate the clinical spectrum, etiologies, and outcome of neurological disorders observed in the rural hospital of Mosango, DRC •Then to assess the diagnostic performance and contribution of clinical symptoms, first-line laboratory results and novel point-of-care RDTs for the diagnosis of HAT and other priority NTDs/IDs in patients with neurological disorders in the rural hospital of Mosango •Finally, to support the elaboration of an innovative integrated syndromic diagnostic tool (multi-disease approach) for neurological disorders, designed to first-line clinicians in rural DRC Reference diagnostic methods were systematically used for the diagnosis of a set of severe and treatable neuro-infections considered as priority conditions and for the field evaluation of new rapid diagnostic tests (RDTs). Pre-established clinical case definitions and post-hoc expert consensus were used for diagnostic ascertainment of the remaining etiologies. No neuroimaging was available in the study hospital. The core prospective study revealed that (i) the patterns of clinical presentation were very diverse; (ii) both infectious and noncommunicable etiologies had varied presentations; and (iii) the frequency of seven priority neuro-infections (Second-stage HAT, cerebral malaria, bacterial meningitis, CNS tuberculosis, neurosyphilis, HIV-related neurological disorders, cryptococcal meningitis in HIV-positive individuals) were respectively low (between 1 and 5% for each of these diagnoses). Death occurred in 8% of the patients and was more frequent in cases diagnosed with neuro-infection. An additional post-hoc analysis showed that about 13% of the patients enrolled with NDs had serological evidence of circulating Taenia solium antigen. This suggests, for the first time, that active neurocysticercosis might be endemic in this part of DRC and supports the need for additional neuroimaging studies. We demonstrated that lumbar puncture (LP), a key procedure to diagnose neuro-infections, but with which most first-line clinicians feel uncomfortable, was safe in the absence of neuroimaging, provided that the locally elaborated protocol was strictly adhered to. In particular, no major procedure-attributable adverse event was observed. Several predictors of cerebrospinal (CSF) pleocytosis (> 5 white blood cell/µL), a surrogate marker for infectious meningo-encephalitis, were identified, such as fever, altered consciousness, HIV infection and positive screening serology for human African trypanosomiasis (HAT). The study findings may assist in the selection of a subgroup of patients who would benefit most a diagnostic LP, and to withhold this procedure when it might be too risky. An additional diagnostic study demonstrated that a novel Human African trypanosomiasis (HAT)- Sero K-SeT rapid diagnostic test (RDT) was highly sensitive 100% (95% confidence interval: 67.6 to 100.0%) and specific 97% (95% confidence interval: 94.2% to 98.5%), with similar performance to that of the card agglutination test for trypanosomiasis (CATT), the currently routinely used screening assay. The high sensitivity and ease of use of this new RDT allows its current deployment in the most remote heath centers as screening tool for both clinical care and control activities. A post-hoc analysis evaluated the potential value of two bacterial biomarkers (C-reactive protein, CRP and procalcitonin, PCT) for the diagnosis of invasive bacterial infection (defined as bacterial meningitis or bacteremia or severe pneumonia; n=19 in the study population). Areas under the curve were 94.3% and 91.7% for CRP and PCT, respectively. More importantly, no single case of invasive bacterial infection was observed when the CRP value was normal (<10 mg/ml), suggesting that this biomarker, if available at the point of care at this cutoff, could reduce immediate antibiotic treatment in a substantial number of neurological patients who do not need it. This research has generated important knowledge on the pre-test probabilities of priority neuro-infections (in other words the frequencies of these diseases) among patients with NDs in the rural hospital of DRC. This has been translated in a preliminary diagnostic panorama with not-to-miss diagnoses, which has been presented to, and welcomed by, first-line care workers in the field. Through the calculation of likelihood ratios, the confirming and excluding powers of several key clinical features, first-line laboratory results and RDTs could be estimated for each priority disease. All these data have been finally integrated in a new electronic panorama on NDs, with the key support of the IT service of ITM. The elaboration of this e-panorama is described step by step in the last manuscript which is now under submission. In the final Chapter, we discuss the perspectives and challenges related to several outputs of this thesis such as the programmatic adoption of the new HAT screening tool in clinical settings, the wider implementation of electronic formats of the elaborated panoramic aid and the improvement of neurological teaching and training in my Department of Neurology, University of Kinshasa, DRC.