This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Objectives: Delay in diagnosis of tuberculosis (TB) is an important but under-appreciated problem. Our study aimed to analyse the patient pathway and possible risk factors of long diagnostic delay (LDD). Methods: We enrolled 400 new bacteriologically diagnosed patients with pulmonary TB from 20 hospitals across China. LDD was defined as an interval between the initial care visit and the confirmation of diagnosis exceeding 14 days. Its potential risk factors were investigated by multivariate logistic regression and multilevel logistic regression. Hospitals in China were classified by increasing size, from level 0 to level 3. TB laboratory equipment in hospitals was also evaluated. Results: The median diagnostic delay was 20 days (IQR: 7e72 days), and 229 of 400 patients (57.3%, 95% CI 52.4e62.1) had LDD; 15% of participants were diagnosed at the initial care visit. Compared to level 0 facilities, choosing level 2 (OR 0.27, 95%CI 0.12e0.62, p 0.002) and level 3 facilities (OR 0.34, 95%CI 0.14

Background Multidrug-resistant tuberculosis (MDR-TB) are unsatisfied to treat, pressing more effective and innovative treatment regimens. New efficient regimens for MDR-TB have obtained high treatment success rates. However, those regimens without drug susceptibility testing (DST) are also likely to contribute to the emergence of resistance. Precision treatments guided by DST might optimize the patients’ treatment outcome individually and minimize resistance amplification. Methods TB-TRUST is a phase III, multicenter, open-label, randomized controlled clinical trial of non-inferiority comparing the treatment success rate between the World Health Organization (WHO) shorter regimen and the refined ultra-short regimen for fluoroquinolones and second-line injectable drugs susceptible rifampicin-resistant TB. The control arm uses the WHO injectable-containing shorter regimen for 36–44 weeks depending on time of sputum smear conversion. The investigational arm uses a refined ultra-short regimen guided by molecular DST to pyrazinamide via whole-genome sequencing (WGS) to optimize the treatment of pyrazinamide-susceptible patients with levofloxacin, linezolid, cycloserine and pyrazinamide for 24–32 weeks and pyrazinamide-resistant with levofloxacin, linezolid, cycloserine and clofazimine for 36–44 weeks. The primary outcome is the treatment success rate without relapse at 84 weeks after treatment initiation. Secondary outcomes include the time of sputum culture conversion and occurrence of adverse events. Assuming α = 0.025 level of significance (one-sided test), a power of 80%, a < 10% difference in treatment success rate between control arm and investigational (80% vs. 82%), and a 5% lost follow-up rate, the number of participants per arm to show non-inferiority was calculated as 177(354 in total). Discussion Rapid molecular testing distinguishes patients who are eligible for shorter regimen with fluoroquinolone and the WGS-guided results shorten the treatment to 6 months for pyrazinamide susceptible patients. It’s foreseeable that not only novel developed medicines, but also traditional powerful medicines with the susceptibility confirmed by DST are the key factors to ensure the effect of anti-MDR-TB drugs. As a DST-guided precision treatment, TB-TRUST are expected to optimize therapy outcome in more patients who cannot afford the expensive new medicines and minimize and even avoid resistance amplification with the rational use of anti-TB drugs. Trail registration ClinicalTrial.gov, NCT03867136. Registered on March 7, 2019.