Clinical Analysis Of Infectious Diseases Of Central Nervous System Based On Metagenomic Next Generation Sequencing

Background and purpose:Central nervous system infection(infection of central nervous system,ICNS)is an acute or chronic inflammatory disease caused by various pathogens invading CNS parenchyma,capsule,blood vessels,etc.,including encephalitis,meningitis,meningoencephalitis,encephalomyelitis and brain abscesses.CNS infectious disease is one of the most important disabling and fatal diseases worldwide,and it is also a common disease in neurology.Its pathogens include viruses,bacteria,fungi,spirochetes,rickettsiae,parasites and prions.At present,the traditional methods of pathogen detection are still mainly based on smears for morphological identification,bacterial culture and antigen detection,as well as PCR-based pathogen-specific nucleic acid detection,but various methods have limitations,and the overall positive rate is still low,exceeding in 50% of ICNS cases,the pathogen diagnosis cannot be confirmed in the end.Molecular diagnostic methods can improve the detection rate of cerebrospinal fluid pathogens,but the current clinical application is not widespread.The purpose of the study: make use of mNGS technology to analyze the difference of positive rate with traditional microbial detection methods,and its clinical characteristics.Methods:(1)Research objects: In this retrospective case study,data of 166 patients with suspected central nervous system infection who were hospitalized in the Department of Neurology of the First Affiliated Hospital of Nanchang University from October2018 to November 2020 were collected.The history,signs,laboratory and imaging results of each patient were recorded in detail.According to the different detection methods of pathogenic microorganisms,the included patients were divided into CSF mNGS group and CSF conventional group.(2)Criterion of suspected ICNS:Signs and/or symptoms of ICNS include at least one: fever(>38 ℃),headache,nausea/vomiting,convulsions,meningeal irritation,focal neurological impairment,and altered consciousness;At least meet the following: a.Increased white blood cell count,increased protein and decreased glucose and chloride in CSF;b.Brain imaging examination shows pathological infection changes.(3)Research method: Collect 2 ～3ml CSF and store it at low temperature(-80 ℃)within 30 minutes in vitro.Take300μLCSF for nucleic acid extraction.The extracted nucleic acid is sonicated to200-300 bp gene fragments,and the sequencing adapter sequence and primer sequence are removed.After filtering the low-quality data,perform circular amplification and replication to generate DNA Nano Ball(DNA nanoball,DNB),load the prepared DNB on the chip,and use BGISEQ.500 for sequencing to remove the human reference genome sequence.After the host information interferes,the remaining data is compared with a dedicated microbial database.Results:1.Among 166 patients with suspected central nervous system infection,76 patients were in the CSF mNGS group,including 52 males(68.4%)and 24 females(31.6%).The median age of onset was 46.50(29.25,61.00)years old,and the interquaternary interval was 46.50(29.25,61.00).In the CSF conventional group,there were 56 males(62.2%)and 34 females(37.8%).The median age of onset was49.00(31.50,62.00)years old.There was no significant difference in gender(χ2 =0.697,P = 0.404)and age of onset(Z =-0.509,P = 0.611)between the two groups(P > 0.05).2.The CSF mNGS group and the CSF conventional group had statistical significance in the number of patients diagnosed with ICNS(χ2=19.433,P=0.000)(P<0.05).It was suggested that the number of confirmed infections in mNGS group were higher.Patients in both groups had no statistical significance in body temperature(Z=-1.342,P = 0.180),fever(χ2 = 0.107,P = 0.744),headache(χ2 =0.814,P = 0.367),changes in consciousness(χ2 = 0.271,P = 0.603),mental disorder(χ2 = 0.820,P = 0.365),cognitive impairment(χ2 = 0.795,P = 0.373),limb tic(χ2 =2.402,P = 0.121),disease outcome(χ2 = 1.769,P = 0.184)(P > 0.05).3.The CSF mNGS group and the CSF conventional group showed no significant differences in CSF white blood cell(Z =-0.020,P = 0.984),CSF protein(Z =-1.133,P = 0.257),CSF sugar to control serum sugar ratio(t=-0.167,P = 0.867),serum white blood cell(Z =-1.202,P = 0.229),neutrophils ratio(t=-0.362,P = 0.718),ferritin(Z =-0.919,P = 0.358)(P >0.05).4.In the CSF mNGS group,there were 33 cases of infective lesions and 43 cases of no sensorial lesions on the brain MRI.In the CSF conventional group,there were33 cases of infective lesions and 56 cases of no perceptual lesions on the brain MRI(χ2 = 0.687,P = 0.407).5.In the CSF mNGS group,30 patients were diagnosed with CNS infection,36 patients with suspected CNS infection,and 10 patients with non-CNS infection.In the CSF conventional group,11 patients were diagnosed with CNS infection,71 patients with suspected CNS infection,and 8 patients with non-CNS infection.Among the 41 patients diagnosed with CNS infection,43.90% were viral,21.95% fungal,19.51%mycobacterium,9.76% bacterial,and 2.44% were parasitic and spirochete.6.The diagnostic consistency of the two methods for central nervous system infection was poor(Kappa=0.084<0.2).The positive rate of mNGS was 43.9%,the positive rate of conventional method was 6.1%,and the total positive rate of the combination of mNGS and conventional method was 45.5%.There was a significant difference between the two etiological methods(P=0.001<0.05).The sensitivity,specificity,positive predictive value and negative predictive value of mNGS in the diagnosis of central nervous system infection were 96.67%,100%,100% and90.90%,the conventional method were 13.33%,100%,100% and 27.78%respectively.Among the 29 patients with positive metagenomic second-generation sequencing,there was no correlation between the sequence number of pathogenic microorganisms detected and the outcome of the disease(P=0.793>0.05).Conclusions:1.The number of patients finally diagnosed with central nervous system infection in the CSF mNGS group were higher than those in the CSF conventional group.There were no significant differences in other general clinical data,history characteristics,laboratory results and head imaging characteristics.2.Only less than 1/4 of all patients had clear pathogenic microbial infection,and more than half were not clear.Viral infection accounted for the largest proportion,and all of them were diagnosed by metagenomic second-generation sequencing,followed by fungi,mycobacteria,bacteria,spirochetes and parasites.3.The diagnostic consistency of metagenomic next generation sequencing method and conventional method for central nervous system infection is poor.The metagenomic next generation sequencing method has higher positive rate,sensitivity and higher negative predictive value.4.Metagenomic second-generation sequencing technology has good diagnostic value for identification of pathogenic microorganism species and rare pathogens,but it still needs to be further improved for diagnosis of intracellular bacteria.