Tuberculosis (TB), caused by Mycobacterium tuberculosis, remains a major global public-health problem, particularly in low- and middle-income countries. Although pulmonary tuberculosis is its most common manifestation, involvement of the central nervous system represents one of the most devastating forms of extrapulmonary tuberculosis. Tuberculous meningitis (TBM) accounts for approximately 1–5% of all tuberculosis cases but produces disproportionately high mortality and long-term neurological disability.¹ Children, older adults, people living with human immunodeficiency virus, malnourished individuals and immunocompromised patients are particularly vulnerable to the disease. Despite improvements in molecular diagnosis and antitubercular treatment, TBM continues to be associated with death or severe neurological disability in a substantial proportion of affected patients.¹ TBM usually develops following haematogenous dissemination of M. tuberculosis from a primary focus, resulting in the formation of small subpial or subependymal granulomas known as Rich foci. Rupture of a Rich focus into the subarachnoid space produces a dense gelatinous inflammatory exudate, predominantly involving the basal cisterns.² This inflammatory process may cause obstruction of cerebrospinal fluid pathways, vasculitis of the perforating cerebral arteries and entrapment of cranial nerves. Consequently, patients may develop hydrocephalus, cerebral infarction, cranial-nerve palsies, tuberculomas and other potentially irreversible neurological complications.² Clinically, TBM commonly presents with a subacute history of fever, headache, vomiting, altered sensorium, neck stiffness, seizures, focal neurological deficits or cranial-nerve involvement. Early diagnosis is frequently difficult because the initial symptoms are non-specific and microbiological confirmation from cerebrospinal fluid has limited sensitivity. The uniform research case definition classifies patients as definite, probable or possible TBM by integrating clinical findings, cerebrospinal-fluid parameters, neuroimaging features and evidence of extracranial tuberculosis.³ Nevertheless, diagnostic delay remains common and is strongly associated with presentation in an advanced clinical stage and an unfavourable outcome. Neuroimaging therefore has a central role in supporting early diagnosis, detecting complications and determining the anatomical extent of the disease. Although computed tomography can demonstrate hydrocephalus, basal enhancement, infarction and larger tuberculomas, magnetic resonance imaging (MRI) provides superior soft-tissue contrast and greater sensitivity for subtle pathological changes. The characteristic MRI manifestations of TBM include basal leptomeningeal enhancement, hydrocephalus, cerebral infarctions, tuberculomas, cerebritis, ventriculitis and cranial-nerve enhancement.⁴ Post-contrast T1-weighted images can delineate basal meningeal and cranial-nerve enhancement, whereas fluid-attenuated inversion recovery images can demonstrate meningeal inflammation, oedema and parenchymal abnormalities. Diffusion-weighted imaging is particularly sensitive for the early detection of acute ischaemic infarctions, while susceptibility-weighted imaging can identify haemorrhage and vascular abnormalities. Magnetic resonance angiography may additionally demonstrate arterial narrowing, irregularity or occlusion secondary to tuberculous vasculitis. The occurrence of multiple concurrent abnormalities on MRI has been associated with greater clinical severity, lower Glasgow Coma Scale scores and more extensive neurological involvement.⁴ Cerebral infarction is one of the most important predictors of poor outcome and may result from vasculitis, thrombosis, vasospasm or compression of vessels by basal inflammatory exudates.⁵ Indian studies have shown that diffusion-weighted MRI can identify acute and clinically unsuspected infarctions, particularly within the basal ganglia, internal capsule, thalamus and other perforator territories.⁶ India continues to carry a disproportionately high share of the global tuberculosis burden; consequently, TBM remains an important cause of infectious neurological morbidity and mortality in Indian clinical practice. Recent prospective Indian evidence has demonstrated that cerebral infarctions may be multiple and distributed across cortical, subcortical and deep grey-matter regions, with advanced TBM and infarction being associated with higher mortality.⁷ Therefore, systematic evaluation of MRI patterns and their relationship with clinical presentation and disease severity may facilitate earlier recognition, prognostic stratification and timely management; the present study was undertaken to assess the MRI patterns of tuberculous meningitis and determine their clinicoradiological correlation. AIM The aim of the study was to evaluate the magnetic resonance imaging patterns of tuberculous meningitis and their correlation with clinical manifestations and disease severity. OBJECTIVES 1. The primary objective was to describe the spectrum, frequency and anatomical distribution of MRI abnormalities in patients with tuberculous meningitis. 2. The secondary objective was to assess the association between MRI findings and clinical manifestations, including altered sensorium, seizures, focal neurological deficits and cranial nerve palsies.

Study design The study was designed as a hospital-based prospective observational study to evaluate MRI patterns in tuberculous meningitis and their clinicoradiological correlation. Study population The study population comprised adult patients aged 18 years and above with definite or probable tuberculous meningitis who underwent MRI of the brain. Sample size The sample size was fixed at 50 patients. This was a pragmatic, feasibility-based sample size for a descriptive and exploratory clinicoradiological study, rather than a sample derived from a formal hypothesis-testing calculation. Sampling technique Consecutive non-probability sampling was adopted. All eligible patients presenting during the recruitment period were approached sequentially until the sample size of 50 was reached. Recruitment was not restricted to patients with an abnormal MRI. Inclusion criteria 1. They were aged 18 years or above. 2. They had definite or probable tuberculous meningitis according to the uniform research case definition. 3. They could undergo the required brain MRI examination. 4. Written informed consent was provided by the patient or a legally authorised representative. Exclusion criteria 1. An alternative cause of meningitis, such as bacterial, viral, fungal or malignant meningitis, was established. 2. A contraindication to MRI prevented safe examination. 3. A contraindication to gadolinium administration prevented completion of the required contrast-enhanced protocol. 4. Persistent motion artefacts rendered the examination non-diagnostic. 5. Consent was declined or withdrawn. Patients with HIV infection, diabetes mellitus or previous tuberculosis were not excluded solely because of these conditions; these factors were recorded as relevant clinical characteristics. Diagnostic criteria Tuberculous meningitis was classified using the uniform research case definition, which incorporated clinical features, cerebrospinal fluid findings, cerebral imaging and evidence of tuberculosis elsewhere in the body.³ Definite TBM was identified through microbiological or pathological confirmation according to the published criteria. Probable TBM was assigned using the prescribed diagnostic score after consideration of alternative diagnoses. Diagnostic category and microbiological confirmation were recorded separately. Since neuroimaging contributed to the classification of probable TBM, its potential influence on the observed frequency of MRI abnormalities was acknowledged. This study evaluated imaging patterns and associations; it did not estimate the diagnostic sensitivity or specificity of MRI. Uniform research case definition. Clinical evaluation Information was collected prospectively using a structured study proforma. Demographic details included age and sex. Clinical history included symptom duration, previous tuberculosis, contact history, comorbidities and antitubercular treatment received before imaging. The presence of fever, headache, vomiting, neck stiffness, altered sensorium, seizures, limb weakness and visual disturbances was documented. Neurological examination findings included meningeal signs, cranial nerve palsies and focal motor deficits. The Glasgow Coma Scale score and clinical severity grade were recorded as close as practicable to the baseline MRI examination. The interval between clinical assessment and imaging was documented. Assessment of clinical severity Disease severity was assessed using the modified British Medical Research Council grading system. Grade Criteria used Grade I Glasgow Coma Scale score of 15 without focal neurological signs Grade II Glasgow Coma Scale score of 11–14, or a score of 15 with focal neurological signs Grade III Glasgow Coma Scale score of 10 or below Patients were assigned to one of these three grades for comparison with MRI findings. Laboratory investigations Laboratory findings were obtained from the clinical records. These included complete blood count, blood glucose, serum electrolytes and renal function tests. HIV status was recorded when testing had been performed with appropriate consent. Cerebrospinal fluid findings were documented when lumbar puncture had been considered clinically safe and performed by the treating team. The recorded parameters included: • Total and differential cell counts. • Protein and glucose concentrations. • Acid-fast bacillus microscopy. • Mycobacterial culture. • Molecular testing for Mycobacterium tuberculosis, where available. Evidence of extracranial tuberculosis was recorded from relevant clinical, microbiological and imaging investigations. Tests that were not performed were recorded as unavailable and were not classified as negative. MRI protocol The imaging protocol included: 1. T1-weighted images. 2. T2-weighted images. 3. Fluid-attenuated inversion recovery images. 4. Diffusion-weighted images with apparent diffusion coefficient maps. 5. Susceptibility-weighted or T2*-weighted images. 6. Post-contrast T1-weighted images in axial, coronal and sagittal planes. Gadolinium-based contrast was administered after appropriate safety screening, according to the institutional protocol and product-specific recommended dose. Magnetic resonance angiography was performed when clinically indicated and was analysed separately because it was not necessarily available for all participants. The baseline MRI was obtained as early as clinically feasible. Treatment was not delayed to obtain research imaging, and the interval between initiation of antitubercular therapy and MRI was recorded. MRI interpretation MRI examinations were evaluated using a structured reporting format. The following findings were recorded: MRI abnormality Parameters assessed Meningeal enhancement Presence, extent and distribution, including basal cisterns, Sylvian fissures and cerebral convexities Hydrocephalus Ventricular enlargement, apparent obstructive pattern and transependymal CSF seepage Cerebral infarction Number, anatomical location, laterality and diffusion characteristics Tuberculomas Number, size, location, signal characteristics, enhancement pattern and surrounding oedema Cranial nerve involvement Visible nerve thickening or enhancement Other abnormalities Ventriculitis, cerebritis, abscess formation and haemorrhage Vascular involvement Arterial narrowing or occlusion when angiographic images were available Acute infarction was identified by restricted diffusion with corresponding reduction in apparent diffusion coefficient values. Lesions suggestive of tuberculomas were characterised in the clinical context; imaging appearances were not treated as independent microbiological confirmation. Clinicoradiological correlation The presence and distribution of MRI abnormalities were compared with the clinical findings recorded around the time of imaging. The principal associations examined were: 1. Cerebral infarction with focal neurological deficits. 2. Hydrocephalus with altered sensorium and Glasgow Coma Scale score. 3. Tuberculomas with seizures. 4. Cranial nerve enhancement with corresponding clinical cranial nerve palsies. 5. Meningeal enhancement, hydrocephalus, infarction and tuberculomas with modified BMRC grade. Clinical altered sensorium was recorded separately from the Glasgow Coma Scale score and was defined as documented confusion, disorientation or reduced consciousness. Study outcome measures The primary outcome measures were the proportions of patients with meningeal enhancement, hydrocephalus, cerebral infarction and tuberculomas on baseline MRI. Secondary outcome measures were the associations between these MRI abnormalities and clinical manifestations, Glasgow Coma Scale scores and modified BMRC grades. The analysis focused on baseline clinicoradiological relationships. Treatment response and long-term neurological outcomes were not included unless follow-up assessments had been specified in the approved protocol. Statistical analysis Data were analysed using SPSS. Categorical variables were expressed as frequencies and percentages. Continuous variables were presented as mean and standard deviation when approximately normally distributed, or median and interquartile range when skewed. The proportions of patients with individual MRI abnormalities were reported with 95% confidence intervals. The chi-square test was used to examine associations between categorical variables when expected cell counts were adequate. Fisher’s exact test, or its extension for larger contingency tables, was used when cell counts were small. Continuous variables were compared between two groups using the independent-samples t-test or Mann–Whitney U test, as appropriate. Comparisons across three clinical severity grades were performed using one-way analysis of variance or the Kruskal–Wallis test, depending on the distribution of the variable. Two-sided p values below 0.05 were considered statistically significant.

Table 1. Sociodemographic characteristics Characteristic Category Frequency (n) Percentage (%) Age group (years) 18–29 14 28.0 30–39 13 26.0 40–49 10 20.0 50–59 8 16.0 ≥60 5 10.0 Sex Male 29 58.0 Female 21 42.0 Total 50 100.0 Interpretation: The largest age group was 18–29 years (28.0%), followed by 30–39 years (26.0%). Participants younger than 40 years accounted for 54.0% of the sample. Males constituted 58.0%, while females constituted 42.0%. Table 2. Clinical manifestations and disease severity Clinical characteristic Frequency (n) Percentage (%) Fever 45 90.0 Headache 42 84.0 Vomiting 31 62.0 Neck stiffness 35 70.0 Altered sensorium 24 48.0 Seizures 14 28.0 Focal neurological deficits 16 32.0 Cranial nerve palsies 8 16.0 Modified BMRC grade Grade I 12 24.0 Grade II 25 50.0 Grade III 13 26.0 Interpretation: Fever (90.0%) and headache (84.0%) were the most frequent manifestations. Altered sensorium occurred in 48.0%, focal neurological deficits in 32.0%, and seizures in 28.0%. Half of the participants had BMRC Grade II disease, while 26.0% had Grade III disease. Table 3. Distribution of MRI abnormalities MRI abnormality Frequency (n) Percentage (%) Meningeal enhancement, any distribution 43 86.0 Basal meningeal enhancement 39 78.0 Hydrocephalus 22 44.0 Cerebral infarction 18 36.0 Tuberculomas 25 50.0 Cranial nerve enhancement 6 12.0 Ventriculitis 4 8.0 Cerebritis 3 6.0 Intracranial abscess 1 2.0 Intracranial haemorrhage 1 2.0 No detectable MRI abnormality 2 4.0 Interpretation: Meningeal enhancement was the predominant finding, occurring in 86.0%. Tuberculomas, hydrocephalus and cerebral infarction occurred in 50.0%, 44.0% and 36.0%, respectively. Two participants (4.0%) had no detectable MRI abnormality. Table 4. Association between MRI findings and clinical manifestations MRI finding Clinical manifestation assessed Manifestation among patients with MRI finding, n/N (%) Manifestation among patients without MRI finding, n/N (%) p-value Cerebral infarction Focal neurological deficits 12/18 (66.7) 4/32 (12.5) <0.001 Hydrocephalus Altered sensorium 16/22 (72.7) 8/28 (28.6) 0.004 Tuberculomas Seizures 10/25 (40.0) 4/25 (16.0) 0.114 Cranial nerve enhancement Cranial nerve palsies 5/6 (83.3) 3/44 (6.8) <0.001 Interpretation: Focal neurological deficits were more frequent among patients with infarction than among those without infarction (66.7% versus 12.5%; p < 0.001). Hydrocephalus was associated with altered sensorium (72.7% versus 28.6%; p = 0.004). Cranial nerve enhancement was associated with clinical palsies (p < 0.001), although only six participants had enhancement. Seizures were more frequent with tuberculomas, but the association was not statistically significant (p = 0.114). Table 5. Association between MRI abnormalities and BMRC severity MRI abnormality Grade I (N = 12), n (%) Grade II (N = 25), n (%) Grade III (N = 13), n (%) p-value Meningeal enhancement 8 (66.7) 22 (88.0) 13 (100.0) 0.044 Hydrocephalus 2 (16.7) 10 (40.0) 10 (76.9) 0.009 Cerebral infarction 1 (8.3) 7 (28.0) 10 (76.9) <0.001 Tuberculomas 4 (33.3) 12 (48.0) 9 (69.2) 0.211 Interpretation: Meningeal enhancement, hydrocephalus and cerebral infarction were more frequent in higher BMRC grades. Associations with severity were statistically significant for meningeal enhancement (p = 0.044), hydrocephalus (p = 0.009) and infarction (p < 0.001). Although tuberculomas were more frequent in Grade III disease, their association with severity was not statistically significant (p = 0.211).

In the dataset, 54.0% of participants were younger than 40 years and 58.0% were male, demonstrating a predominantly young-adult demographic profile. This male proportion was comparable to the global modelling study by Dodd et al., which estimated approximately 164,000 adult TBM cases in 2019, with almost 60% occurring in males.⁸ In the Indian context, Modi et al. reported a mean age of 30.4 ± 13.8 years among 209 patients, similarly highlighting the involvement of younger adults.⁹ Fever, headache and neck stiffness occurred in 90.0%, 84.0% and 70.0% of the simulated sample, respectively, while altered sensorium occurred in 48.0%. Meningeal enhancement was the predominant simulated MRI abnormality, occurring in 86.0%, closely resembling the 84.4% leptomeningeal enhancement reported by Soni et al. in their 90-patient imaging series.¹⁰ Basal meningeal enhancement occurred in 78.0%, compared with 82% among presenting MRI findings in the 43-patient serial-imaging study by Thwaites et al.¹¹ Hydrocephalus occurred in 44.0%, lying between the reported frequencies of 29% in Soni et al. and 77% in Thwaites et al., although these populations and imaging schedules were not directly interchangeable.¹⁰˒¹¹ Within the data, altered sensorium was more frequent with hydrocephalus than without it—72.7% versus 28.6% (p = 0.004)—and hydrocephalus was associated with BMRC severity (p = 0.009). If reproduced in actual observations, these relationships would support assessing ventricular abnormalities alongside consciousness level, without proving that hydrocephalus alone caused neurological deterioration. Cerebral infarction occurred in 36.0% of the simulated sample, compared with 46 of 120 patients, approximately 38.3%, in the Indian study by Kumar et al.¹² The infarction frequency was higher than the 25.8% reported by Wasay et al. among 559 patients, but lower than the 57.7% reported by Soni et al.¹³¹⁰ Such differences could reflect case selection, disease severity, imaging timing and imaging methods; percentage comparisons alone could not determine the explanation. Focal neurological deficits were present in 66.7% of simulated patients with infarction compared with 12.5% without infarction (p < 0.001), while infarction was also associated with BMRC grade (p < 0.001). The clinical importance of infarction was supported by Chan et al., who identified infarction in 12 of 40 patients, with two deaths and six dependent survivors at one year among those affected.¹⁴ Tuberculomas occurred in 50.0% of the sample, compared with 100% in the selected imaging series of Soni et al., emphasising the limited generalisability of individual hospital-based estimates.¹⁰ Seizures occurred in 40.0% of simulated patients with tuberculomas versus 16.0% without them, but this difference was not statistically significant (p = 0.114), and neither was the association between tuberculomas and BMRC grade (p = 0.211). Thwaites et al. reported that tuberculomas developed in 74% during treatment without an association with poor clinical outcome.¹¹ Clinical cranial nerve palsies occurred in 16.0% of the simulated sample, compared with 33.3% reported by Wen et al.; the association with nerve enhancement (p < 0.001) remained limited by only six enhancement-positive cases.¹⁵ Any MRI abnormality occurred in 96.0%, compared with 26 of 29 patients, approximately 89.7%, in Abdelmalek et al., but neither proportion alone represented diagnostic sensitivity in an unselected suspected-TBM population.¹⁶ BMRC Grades II and III together accounted for 76.0% of the sample, while the association between meningeal enhancement and severity was relatively weak statistically (p = 0.044) and warranted caution because multiple comparisons were performed. Wang et al., analysing 2,437 adults across 22 studies, reported pooled mortality of 24.7% and neurological sequelae in 50.9% of survivors, providing context for the importance of assessing neurological complications.¹⁷ Similarly, Stadelman et al. analysed 5,752 adults across 39 studies and reported pooled six-month mortality of 24%, although mortality could not be evaluated from the present baseline-only.¹⁸ . A sample of 50, small subgroups, unadjusted comparisons and the contribution of MRI to probable-TBM classification would limit precision, causal interpretation and generalisability. Overall, this comparison demonstrated how MRI patterns could be discussed alongside clinical severity, but actual patient data and longitudinal outcomes were essential before claiming prognostic value or improved patient management.

In the dataset, meningeal enhancement was the most frequent MRI abnormality, followed by tuberculomas, hydrocephalus and cerebral infarction. Cerebral infarction was associated with focal neurological deficits, hydrocephalus with altered sensorium, and cranial nerve enhancement with clinical cranial nerve palsies. Meningeal enhancement, hydrocephalus and cerebral infarction were also associated with higher modified British Medical Research Council grades. However, the associations of tuberculomas with seizures and clinical severity were not statistically significant. These findings demonstrated how MRI abnormalities could be evaluated alongside clinical manifestations and disease severity in tuberculous meningitis.

1. Wilkinson RJ, Rohlwink U, Misra UK, van Crevel R, Mai NTH, Dooley KE, et al. Tuberculous meningitis. Nat Rev Neurol. 2017;13(10):581-598. doi: 10.1038/nrneurol.2017.120. 2. Davis AG, Rohlwink UK, Proust A, Figaji AA, Wilkinson RJ. The pathogenesis of tuberculous meningitis. J Leukoc Biol. 2019;105(2):267-280. doi: 10.1002/JLB.MR0318-102R. 3. Marais S, Thwaites G, Schoeman JF, Török ME, Misra UK, Prasad K, et al. Tuberculous meningitis: a uniform case definition for use in clinical research. Lancet Infect Dis. 2010;10(11):803-812. doi: 10.1016/S1473-3099(10)70138-9. 4. Dian S, Hermawan R, van Laarhoven A, Immaculata S, Achmad TH, Ruslami R, et al. Brain MRI findings in relation to clinical characteristics and outcome of tuberculous meningitis. PLoS One. 2020;15(11):e0241974. doi: 10.1371/journal.pone.0241974. 5. Tai MLS, Viswanathan S, Rahmat K, Nor HM, Kadir KAA, Goh KJ, et al. Cerebral infarction pattern in tuberculous meningitis. Sci Rep. 2016;6:38802. doi: 10.1038/srep38802. 6. Shukla R, Abbas A, Kumar P, Gupta RK, Jha S, Prasad KN. Evaluation of cerebral infarction in tuberculous meningitis by diffusion-weighted imaging. J Infect. 2008;57(4):298-306. doi: 10.1016/j.jinf.2008.07.012. 7. Aggrohia A, Bhatia V, Saroch A, Pannu AK. Predicting cerebral infarction in tuberculous meningitis and its prognostic significance. Biomark Med. 2024;18(8):407-417. doi: 10.1080/17520363.2024.2347194. 8. Dodd PJ, Osman M, Cresswell FV, Stadelman AM, Lan NH, Thuong NTT, et al. The global burden of tuberculous meningitis in adults: a modelling study. PLOS Glob Public Health. 2021;1(12):e0000069. doi: 10.1371/journal.pgph.0000069. 9. Modi M, Sharma K, Prabhakar S, Goyal MK, Takkar A, Sharma N, et al. Clinical and radiological predictors of outcome in tubercular meningitis: a prospective study of 209 patients. Clin Neurol Neurosurg. 2017;161:29-34. doi: 10.1016/j.clineuro.2017.08.006. 10. Soni N, Kumar S, Shimle A, Ora M, Bathla G, Mishra P. Cerebrovascular complications in tuberculous meningitis-a magnetic resonance imaging study in 90 patients from a tertiary care hospital. Neuroradiol J. 2020;33(1):3-16. doi: 10.1177/1971400919881188. 11. Thwaites GE, Macmullen-Price J, Chau TTH, Mai PP, Dung NT, Simmons CP, et al. Serial MRI to determine the effect of dexamethasone on the cerebral pathology of tuberculous meningitis: an observational study. Lancet Neurol. 2007;6(3):230-236. doi: 10.1016/S1474-4422(07)70034-0. 12. Kumar A, Mudassir S, Sinha N, Babanrao WB, Ranjan A. Stroke in tuberculous meningitis and its correlation with magnetic resonance angiography manifestations. J Neurosci Rural Pract. 2022;13(3):417-423. doi: 10.1055/s-0042-1745713. 13. Wasay M, Khan M, Farooq S, Khowaja ZA, Bawa ZA, Ali SM, et al. Frequency and impact of cerebral infarctions in patients with tuberculous meningitis. Stroke. 2018;49(10):2288-2293. doi: 10.1161/STROKEAHA.118.021301. 14. Chan KH, Cheung RTF, Lee R, Mak W, Ho SL. Cerebral infarcts complicating tuberculous meningitis. Cerebrovasc Dis. 2005;19(6):391-395. doi: 10.1159/000085568. 15. Wen A, Cao WF, Liu SM, Zhou YL, Xiang ZB, Hu F, et al. Incidence and risk factors of cranial nerve palsy in patients with tuberculous meningitis: a retrospective evaluation. Infect Drug Resist. 2023;16:829-841. doi: 10.2147/IDR.S396022. 16. Abdelmalek R, Kanoun F, Kilani B, Tiouiri H, Zouiten F, Ghoubantini A, et al. Tuberculous meningitis in adults: MRI contribution to the diagnosis in 29 patients. Int J Infect Dis. 2006;10(5):372-377. doi: 10.1016/j.ijid.2005.07.009. 17. Wang MG, Luo L, Zhang Y, Liu X, Liu L, He JQ. Treatment outcomes of tuberculous meningitis in adults: a systematic review and meta-analysis. BMC Pulm Med. 2019;19(1):200. doi: 10.1186/s12890-019-0966-8. 18. Stadelman AM, Ellis J, Samuels THA, Mutengesa E, Dobbin J, Ssebambulidde K, et al. Treatment outcomes in adult tuberculous meningitis: a systematic review and meta-analysis. Open Forum Infect Dis. 2020;7(8):ofaa257. doi: 10.1093/ofid/ofaa257.