This study is a multicenter randomized controlled trial. The purpose of this study is to assess the efficacy of the combination of PCR and CRISPR/Cas12a in alveolar lavage fluid for early targeted anti-infective therapy for patients with severe pneumonia. Hosted by the Department of Critical Care Medicine, Affiliated Drum Tower Hospital of Nanjing University Medical College, 5 adult ICU units participate in 3 hospitals. All patients are randomly assigned to the experimental group and the control group. For experimental group, the combined detection of PCR and CRISPR/Cas12a in the alveolar lavage fluid was carried out in the early stage, and the antibiotic scheme is changed base on the results of PCR-CRISPR/Cas12a.The patients in the control group were adjusted according to the traditional microbial detection methods. The types of early antibiotics, the proportion of target antibiotics, the duration of anti-infective treatment, the length of hospital stay in ICU, the mortality rate, the secondary antibiotic-associated diarrhea, and the incidence of new multidrug-resistant infections were recorded.

ICU patients have a high incidence of bacterial infection in the lower respiratory tract, mainly with severe pneumonia, often causing severe sepsis and septic shock, which is one of the main causes of death in patients. At present, the biggest difficulty faced by clinicians is the continuous increase of bacterial resistance rate and the increase of patient mortality due to the early inadequacy empirical anti-infective treatment. Studies have shown that patients with VAP(Ventilator Associated Pneumonia) have irrational drug use in the early stage, with a mortality rate of more than 50%. When the rate of appropriate drug use has dropped to 33%, while mechanical ventilation time and ICU hospitalization time have been significantly shortened. Therefore, identifying pathogenes as early as possible and shortening the time of empirical anti-infective treatment are very important for improving the prognosis of patients with severe pneumonia and reducing the incidence of bacterial resistance.

There are three traditional methods for detecting pathogenic microorganisms: 1. microbial culture method is the most traditional means of identifying pathogen. It is necessary to inoculate the patient's body fluid, blood, etc. in a suitable medium, incubate in a suitable incubator, and then pass the drug. Sensitivity tests determine the resistance of microorganisms, usually takes 3-7 days. For some specific types of pathogenic microorganisms or microorganisms with harsh growth conditions, there may be negative culture results. Therefore, the traditional culture methods have disadvantages such as poor timeliness, relatively high requirements, and low positive culture rate (30-40%). 2. time-of-flight mass spectrometry: the mass spectrometry technique is used to analyze and detect the protein components of the strain, and the characteristic peak spectrum is obtained. Compared with the bacterial map in the database, the bacteria can be judged by matching. The method can be shortened by about 6-8 hours compared with the conventional culture method, but since the detection of the colony needs to reach a certain amount, the specimen can not be directly detected after obtaining the specimen, and the preliminary microbial culture is required. Therefore, the detection time still takes 1-2 days or more, and there is also the disadvantage of low timeliness. In addition, it is necessary to compare the expansion and standardization of the database, and the inability to analyze the resistance of microorganisms is also the inadequacy of the detection technology. 3. High-throughput sequencing technology: With the rapid development of molecular biology in recent years, high-throughput sequencing technology is widely used in the early diagnosis of clinical microbiology, the principle is mainly through the connection of the universal linker to the fragmentation to be sequenced. Genomic DNA, which produces tens of millions of single-molecule polyclonal polymerase chain reaction arrays, then performs large-scale primer hybridization and enzyme extension reactions, and obtains complete DNA sequence information by computer analysis. However, this technology is difficult to effectively distinguish between pathogenic bacteria and background bacteria, technology and database to be standardized, detection time still takes about 2 days, can not obtain microbial resistance, expensive and other shortcomings At the office. In summary, the current time limit for targeted anti-infective treatment is stopped 2 days after the specimen is taken. Therefore, the search for new, pathogenic microbial detection technology that is faster, more accurate and more sensitive is a hotspot and a difficult point in the field of microbial and anti-infective research in recent years.

The PCR-CRISPR/Cas12a combination technology of alveolar lavage fluid developed by the College of Life Sciences of Nanjing University is based on PCR amplification and fluorescence signal detection twice to achieve the detection of the presence and absence of specific DNA sequences in the test sample. technology. The determination of the detection result of the clinical sample pathogen is based on the comparison of the fluorescence results of the PCR product of the sample DNAD with the fluorescence detection results of the positive control (PC) and the negative control (NC) as a standard. The specific recognition function of the CRISPR/Cas12a system relies on the specific guidance and binding of the crRNA to specific DNA, and the specificity of the crRNA is determined by detection of a positive control of a common pathogen by a single crRNA. The detection technology is highly specific and takes only 2-3 hours, which is a qualitative leap in the detection time compared to the conventional technology.

In order to verify the feasibility and accuracy of the technology, the Center and the Nanjing University of Life Sciences for the common pathogens of ICU pneumonia (Acinetobacter baumannii, MRSA, Klebsiella pneumoniae, Pseudomonas aeruginosa, etc.) for alveolar A preliminary study of the PCR-CRISPR/Cas12a combined detection technique for lavage fluid. Twenty-nine specimens of lower respiratory tract alveolar lavage fluid were cultured by conventional bacterial culture method and combined with PCR-CRISPR/Cas12a. The results showed that the accuracy of detection and identification of pathogens based on PCR-Cas12a technology reached 93.10% (27/29). For the 27 specimens, the pathogens infected by the traditional isolation culture method can be detected by PCR-CRISPR/Cas12a technology. The two exceptions were the detection of Acinetobacter baumannii in the No. 6 sample by the traditional isolation culture method and the detection of Proteus mirabilis in the No. 13 sample (not within the range of pathogens detected). Moreover, the pathogens identified by the PCR-CRISPR/Cas12a combination technique were more than one or two different than the traditional culture methods, which was consistent with the PCR, suggesting that the sensitivity is much higher than that of conventional microbial culture, and the results are reliable. These preliminary results indicate that the PCR-CRISPR/Cas12a combined detection technique has good accuracy and high sensitivity.

Based on those, the research team speculated that the combination of PCR and CRISPR/Cas12a detection technology of alveolar lavage fluid to guide anti-infective treatment of pneumonia patients can achieve targeted anti-infective treatment and improve patient prognosis. To validate the above hypothesis, we designed a multicenter randomized prospective study comparing the effects of PCR-CRISPR/Cas12a combined detection with alveolar lavage fluid and traditional microbial detection techniques on antimicrobial adjustment and prognosis in patients with ICU pneumonia. It aims to find more rapid, accurate and sensitive microbial detection technology for patients with pneumonia, and to achieve earlier precision treatment.

Inclusion Criteria:

• age ≥ 18 years
• patients with artificial airway and expected artificial airway for more than 48 hours
• patients with suspected pneumonia or clear pneumonia
• signed informed consent
• expected ICU hospitalization more than 3 days.

Exclusion Criteria:

• pregnant women
• lactating women
• considered by the doctors for bronchoscopy moderate to severe asthma
• airway stenosis
• tracheal fistula, bronchopleural fistula
• expected to die or give up treatment within 72 hours
• participate in other clinical research