A Rare Case of Mycobacterium Chelonae Infection in an Immunocompromised Adult with Cavernous Sinus Syndrome
Ye Li,Le Zhang,Chunhui Li,Xuelun Zou,Yi Zhang,Zhiping Hu,Bo Xiao,Lili Long
DOI: https://doi.org/10.1111/cns.13808
2022-01-01
CNS Neuroscience & Therapeutics
Abstract:Mycobacterium chelonae is a non-tuberculous mycobacterium (NTM) that primarily causes skin and soft tissue infections in immunocompromised adults.1 Herein, we report a rare case of cavernous sinus syndrome (CSS) caused by M. chelonae infection in an immunocompromised adult female, which led to focal stenosis of the adjacent internal carotid artery (ICA) and pseudoaneurysm formation. On July 4, 2020, a 43-year-old woman was admitted to a local hospital with fever and headache. Her symptoms began with peristomal herpes 3 days earlier. Lumbar puncture (LP) showed high levels of leukocytes (54 cells/µL, polymorphonuclear cells were slightly dominant) and a protein concentration of 663 mg/dl in cerebrospinal fluid (CSF) samples. Magnetic resonance imaging (MRI) and MR angiography of the brain revealed no abnormalities. Her symptoms eased within 2 weeks of antiviral and empirical antibiotic therapy. However, starting July 20, she had recurrent febrile episodes. Considering continued suspicion of immune response secondary to central nervous system (CNS) infection, dexamethasone (40 mg/day), a hormone immunoregulatory drug, was added to the treatment regime. On the sixth day after dexamethasone administration, the patient developed sixth nerve palsy (abduction deficit in the left eye) and loss of pinprick and thermal sensations in the territories of the ophthalmic (V1) and maxillary (V2) divisions of the left trigeminal nerve. MRI revealed enlargement of the left cavernous sinus (CS) with an enhancing lesion, secondary infectious arteritis, and infectious pseudoaneurysm formation (Figure 1). Thus, a clinical diagnosis of CSS was made. Multiple tests for potential autoimmune and neoplastic causes came back negative. After her fever subsided, the patient was discharged on August 5. However, she still had clinical manifestations of CSS. On September 6, she was transferred to our hospital. She underwent LP for CSF sample analysis, which demonstrated a leukocyte count of 70 cells/mm3 (58% lymphocytes), a protein level of 380 mg/dl, and a glucose level of 2.78 mg/dl. Microbial culture results of the CSF samples were negative for bacteria, including mycobacteria, and fungi. However, the CSF sample was positive for modified acid-fast staining. Moreover, negative results of the purified protein derivative and tuberculosis infection T-cell spot test (T-SPOT. TB) and lung CT may not necessarily suggest tuberculosis infection. She had no other laboratory abnormalities. CSF samples were collected for next-generation sequencing (NGS) analysis on Day 4 of hospitalization. Libraries were multiplexed and sequenced using an Illumina platform, along with water and CSF from an uninfected clinical sample as negative controls (Appendix S1). Written consent for NGS analysis of clinical samples was obtained from the patient. M. chelonae was detected in the CSF from the initial rapid sequencing run completed on Day 6 of hospitalization. Sequencing identified 269 (out of 15,214,873) sequence reads uniquely corresponding to the M. chelonae genome (Figure 2A), and these reads covered a high percentage of the genome. When the reads from the human host were excluded, M. chelonae reads accounted for 0.46% of total microbial and unknown or unclassified reads (Figure 2A). NGS analysis performed on the CSF sample collected on Day 21 of hospitalization identified 40 (out of 17,569,288) sequence reads uniquely corresponding to the M. chelonae genome. The detection of M. chelonae in this subject was also confirmed by PCR and Sanger sequencing (Figure 2B,C; Figure S1). The aforementioned results indicated that the patient was virtually infected with M. chelonae. She was then started on intravenous linezolid and cefoxitin, followed by oral doxycycline. Her symptoms gradually improved, and an MRI scan at 3 weeks showed that the enhancing lesion in the left CS had significantly reduced in size. LP suggested that the intracranial infection was largely improved, and a repeat NGS analysis of the CSF sample indicated that the copy number of M. chelonae had indeed decreased (40 reads). The patient was discharged 3 weeks later, with a dramatic response to therapy. At the 1-month follow-up, contrast-enhanced scans showed no abnormal enhancement in the CS areas. At the 3-month follow-up, all CSF parameters had returned to normal. Examination at 6 months showed no apparent positive signs. In this study, we used NGS analysis to identify a rare case of CSS caused by M. chelonae infection. NTM infections are always challenging and involve the lungs, skin, soft tissue, and CNS.2 Although NTM CNS infections are rare, the mortality rate is high.3 Infections due to this pathogen are on the rise, and skin and soft tissue infections have been reported, involving skeletal tissue, eye, ear, lung, and rare cases of endocarditis and meningitis.1 Mycobacterium chelonae is an opportunistic pathogen that infects immunocompromised individuals.4 Therefore, in the present case, the patient's previous history of glucocorticoid therapy cannot be ignored. Empirical treatment with corticosteroids often clouds the diagnosis of CNS infection, as it transiently improves systemic features and occasionally improves focal neurological signs, but later accelerates the progression of infection.5 Cavernous sinus is a complex venous space surrounded by the dural folds and contains important neurovascular structures (III, IV, and VI nerves; first and second trigeminal branches; ICA, sympathetic plexus). The valveless venous communication system comprises almost every important venous structure in the head and neck. Different pathologic conditions, such as infectious, inflammatory, vascular, and neoplastic diseases, can give rise to CSS with unique clinical presentations. Although infectious and inflammatory etiologies are difficult to distinguish using imaging data, clinical symptoms and laboratory tests can help narrow down the differential diagnosis, which includes Tolosa-Hunt syndrome, sarcoidosis, granulomatosis with polyangiitis, IgG-4-related disease, and invasive fungal infections, such as aspergillosis and mucormycosis.6 However, to the best of our knowledge, CSS caused by M. chelonae infection has not been reported previously. In suspected cases, close attention should be paid to assess secondary infectious vasculitis and infectious pseudoaneurysm formation. In the present case, NGS analysis played a critical role in the accurate diagnosis of M. chelonae infection. Hence, NGS can be a valuable tool in the diagnosis of CNS infection, as it can detect causative pathogens even when routine tests fail to do so.7 However, the core challenge with regard to using NGS is to develop faster, more accurate, and more complete identification of pathogens in a large human background and ubiquitous bacteria background. In addition, for different pathogens detected by NGS, it is worth further discussion to determine the diagnostic value of their sequence number. In the present case, we could only determine M. chelonae infection after comprehensive consideration of all analyses, including the results of the patient's two NGS analyses, Sanger sequencing, combined with the patient's clinical manifestations, imaging, CSF characteristics, responses to treatment, and outcome. In the future, the NGS sequence number of M. chelonae and other pathogens in the CSF may indicate that CNS infection needs to be studied in a larger sample size. The study was supported by grants from the National Natural Science Foundation of China(No. 82171454 to Lili Long; No. 81571151 to Le Zhang), the Natural Science Foundation of Hunan Province Project (2020JJ5914), and the National Science and Technology Basic Resources Survey(2018FY100900). We are grateful to staff at Hugobiotech Co., Ltd., (Beijing, China) for sequencing. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Y.L. and L.L. wrote the manuscript. Y.L., L.L., and C.L. were the physicians who treated the patient and provided professional comments to the manuscript. Y.L., L.Z., Z.H., Y.Z., X.Z., and B.X. analyzed the clinical data and the NGS data. All authors have read and approved the final manuscript. The datasets presented in this study can be found in online repositories. The names of the repository and accession number can be found below: National Genomics Data Center, accession No: PRJCA007940. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.