Application of high-throughput sequencing technology in pathogens detection for spinal infections
王冰,郑禹豪,吕国华,胡嘉瑞,王孝宾,张谦实,李晶,刘富兵
DOI: https://doi.org/10.3760/cma.j.cn121113-20200715-00449
2021-01-01
Abstract:Objective:To explore the application of high-throughput sequencing (HTS) technology in pathogens detection for spinal infection.Methods:From January 2019 to May 2020, a total of 41 patients including 31 males and 10 females with an average age of 59.7±11.9 years (29-75 years) were suspected of spinal infections. There were 37 patients with local pain, 15 with fever (≥38 ℃) and 18 with neurological dysfunction. The infected sites were as follows, 4 cases of cervical spine, 8 cases of thoracic spine and 29 cases of lumbar spine. There were 36 patients met the surgical indications and underwent open debridement, bone grafting, fusion and internal fixation, while the other 5 patients underwent conservative treatment (three received drug therapy and two were transferred to the internal department for chemotherapy). Lesions obtained from open surgery patients were underwent pathology and HTS examination. In 5 cases with conservative treatment, two of them underwent CT guided percutaneous puncture for samples, while one case underwent ultrasound guided percutaneous puncture for pus, one case for venous blood, and one case received lumbar puncture for cerebrospinal fluid. The samples were sent for pathological and HTS examination, while liquid specimens were sent for bacterial culture and HTS. The sensitivity and specificity of HTS results were determined according to pathological examination which was regarded as the "gold standard". Based on HTS results combined with the clinical manifestations, imaging examination and pathological results of the patients, targeted antibiotics or anti-tuberculosis drugs were selected for postoperative drug therapy. Patients with bacterial infection received anti-infection treatment for 3 months after operation. For tuberculosis patients, "tetrad" (isoniazid+rifampicin+pyrazinamide+ethambutanol) anti-tuberculosis treatments were underwent for one year. Inflammation indicators from the blood samples were observed before and after treatment, including white blood cell count (WBC), C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR). These indicators were used to monitor disease progression and the curative effects. All patients were followed up for at least 3 months after surgery.Results:A total of 41 patients with suspected spinal infection were included in this study. The HTS pathogen detection results were obtained within 48 h. For the initial 5 patients, first-generation sequencing verification was conducted with coincidence rate 100%. Further, no further verification was conducted in the rest patients. Among the 41 cases, a total of 26 cases had positive results with a positive rate of 63.4%(26/41). Among them, thirteen cases were with mycobacterium tuberculosis (31.7%) and 6 cases with staphylococcus (14.6%). Fungi and Brucellosis were diagnosed in 2 cases respectively, accounting for 4.9% respectively. The test were negative in 15 patients (36.6%), including 2 patients with tumor or tumor-like lesions (1 hematologic tumor and 1 eosinophilic granuloma). A total of 38 patients underwent pathological examination, which confirmed 7 cases of suppurative infection, 12 cases of tuberculosis, 2 cases of tumor or tumor-like lesions and the remaining 17 cases of inflammatory lesions. The sensitivity and specificity of HTS were 80%(16/20) and 55.6% (10/18) with positive predictive value (PPV) 66.7% (16/24) and negative predictive value (NPV) 71.4% (10/14). All patients were followed up for 3 months. The inflammation indicators of blood at 3 months were all lower than that at admission. WBC decreased from (7.50±3.26)×10 9/L at admission to (6.22±2.53)×10 9/L at 3 months after treatment without statistically significant difference ( t=1.082, P=0.290). The CRP decreased from (32.2±34.1) mg/L to (4.5±10.5) mg/L, and ESR from (44.2±26.5) mm/1 h to (18.6±12.1) mm/1 h with statistically significant difference ( t=8.963, P<0.001; t=5.421, P<0.001). Conclusion:High-throughput sequencing technology can be used in detection of spinal infection pathogens, due to its relatively high positive rate, satisfied sensitivity and good diagnostic value.