Chongqing Public Health Medical Center Research Articles

Explainable machine learning for early predicting treatment failure risk among patients with TB-diabetes comorbidity

The present study aims to assess the treatment outcome of patients with diabetes and tuberculosis (TB-DM) at an early stage using machine learning (ML) based on electronic medical records (EMRs). A total of 429 patients were included at Chongqing Public Health Medical Center. The random-forest-based Boruta algorithm was employed to select the essential variables, and four models with a fivefold cross-validation scheme were used for modeling and model evaluation. Furthermore, we adopted SHapley additive explanations to interpret results from the tree-based model. 9 features out of 69 candidate features were chosen as predictors. Among these predictors, the type of resistance was the most important feature, followed by activated partial throm-boplastic time (APTT), thrombin time (TT), platelet distribution width (PDW), and prothrombin time (PT). All the models we established performed above an AUC 0.7 with good predictive performance. XGBoost, the optimal performing model, predicts the risk of treatment failure in the test set with an AUC 0.9281. This study suggests that machine learning approach (XGBoost) presented in this study identifies patients with TB-DM at higher risk of treatment failure at an early stage based on EMRs. The application of a convenient and economy EMRs based on machine learning provides new insight into TB-DM treatment strategies in low and middle-income countries.

Prevalence of, risk factors for, and target organ damage from metabolic syndrome among people living with HIV on ART: A cross-sectional analysis in Chongqing, China.

The incidence of metabolic syndrome (MetS) in people living with HIV is significantly higher than in people without HIV. MetS is not only a major driver of cardiovascular disease (CVD) but is also closely related to the development of chronic kidney disease (CKD). The aim of this study was to investigate the prevalence of and risk factors for MetS and to further understand the degree of damage to target organs. This was a cross-sectional descriptive study conducted at Chongqing Public Health Medical Center, China. Information was collected via questionnaire survey, physical examination, and laboratory tests. We used the China Diabetes Society guidelines to define MetS. Pooled cohort equations were calculated to compare CVD risk in the next 10 years in people living with HIV aged ≥40 years with or without MetS. We used Student's t-test, the chi-squared test, Fisher's exact test, binary logistic regression, and multiple linear regression in the statistical analysis. The study included 979 people living with HIV, including 13 who have experienced CVD, receiving antiretroviral therapy (ART). The median age was 43.0 years, 20.9% were female, and the median ART time was 45.0 months. The prevalence of MetS was 33.9%. The components of MetS criteria were hyperglycaemia (50.4%), hypertriglyceridaemia (48.4%), hypertension (46.8%), low concentrations of high-density lipoprotein cholesterol (28.2%), and abdominal obesity (25.0%). Higher body mass index (odds ratio [OR] 1.266; 95% confidence interval [CI] 1.203-1.333), higher total cholesterol (OR 1.267; 95% CI 1.011-1.588), high alcohol consumption (OR 1.973; 95% CI 1.009-3.859), and family history of diabetes (OR 1.726; 95% CI 1.075-2.770) were independent risk factors for MetS. Compared with the non-MetS group, the MetS group had a higher rate of urine albumin (23.8% vs 14.8%, p = 0.001), and the estimated glomerular filtration rate <90 mL/min/1.73 m2 (18.37% vs. 12.8%, p = 0.020) and β2-microglobin (p = 0.004) increased more markedly in the MetS group. Regarding the risk of developing CVD events in the next 10 years, 38.5% of those in the MetS group were at high or very high risk, which was significantly higher than in the non-MetS group (p < 0.001). In addition, age (p < 0.001) and sex (p = 0.002) are independent risk factors for developing CVD events in the next 10 years. The prevalence of MetS in people living with HIV on ART is high in Chongqing, China. Risk factors for the development of MetS are high alcohol consumption, family history of diabetes, higher body mass index, and higher total cholesterol levels. In addition, MetS is associated with a risk of CKD and the incidence of 10-year CVD.

Prevalence and analysis of acquired and transmitted integrase strand transfer inhibitor-associated HIV-1 drug resistance in Chongqing, China

ABSTRACT In this study, we examined the occurrence of acquired and transmitted drug resistance to integrase strand transfer inhibitor (INSTI) in HIV-1 strains in Chongqing (China) for guiding for the routine testing of INSTI-associated HIV-1 genotype resistance. Plasma samples were obtained from HIV-1 patients at Chongqing Public Health Medical Center from July 2019 to August 2022. Besides, amplification, sequence, and analysis of the portion of the HIV-1 pol gene that encodes the integrase protein were implemented to identify INSTI resistance. Integrase sequence data was harvested for a comprehensive cohort of 1032 patients infected with HIV-1. This cohort consisted of 564 ART-naive patients, 465 ART-treated patients, and 3 patients with an unknown treatment history. Within the study group, we identified INSTI resistance in 21 patients (2.03%, 21/1032), including 17 ART-treated patients (3.66%, 17/465). Among the ART-treated patients, 12 were INSTI-treated (11.76%, 12/102), 5 were INSTI-naive (1.38%, 5/363), and 4 were ART-ineffective patients (0.71%, 4/564). The prevalent major resistance mutation was Q148R (0.48%, 5/1032), while the most prevalent accessory resistance mutation was E157Q (1.65%, 17/1032). In light of the above, it is recommended that the incidence of accessory genotype analysis should be considered before starting any future INSTI-based therapy, especially in patients with drug resistance to NRTIs and NNRTIs and the reduction of INSTI sensitivity should be carefully monitored and investigated. Regular monitoring for resistance should be implemented after the use of INSTIs, and, importantly, ongoing monitoring of the decreasing susceptibility to INSTIs is crucial following the initiation of treatment with INSTIs.

Establishment and evaluation of an overlap extension polymerase chain reaction technique for rapid and efficient detection of drug-resistance in Mycobacterium tuberculosis

BackgroundRapid and accurate detection of drug resistance in Mycobacterium tuberculosis is critical for effective control of tuberculosis (TB). Herein, we established a novel, low cost strategy having high accuracy and speed for the detection of M. tuberculosis drug resistance, using gene splicing by overlap extension PCR (SOE PCR).MethodsThe SOE PCR assay and Sanger sequencing are designed and constructed to detect mutations of rpoB, embB, katG, and inhA promoter, which have been considered as the major contributors to rifampicin (RFP), isoniazid (INH), and ethambutol (EMB) resistance in M. tuberculosis. One hundred and eight M. tuberculosis isolates came from mycobacterial cultures of TB cases at Chongqing Public Health Medical Center in China from December 2018 to April 2019, of which 56 isolates were tested with the GeneXpert MTB/RIF assay. Performance evaluation of the SOE PCR technique was compared with traditional mycobacterial culture and drug susceptibility testing (DST) or GeneXpert MTB/RIF among these isolates. Kappa identity test was used to analyze the consistency of the different diagnostic methods.ResultsWe found that the mutations of S531L, S315T and M306V were most prevalent for RFP, INH and EMB resistance, respectively, in the 108 M. tuberculosis isolates. Compared with phenotypic DST, the sensitivity and specificity of the SOE PCR assay for resistance detection were 100.00% and 88.00% for RFP, 94.64% and 94.23% for INH, and 68.97% and 79.75% for EMB, respectively. Compared with the GeneXpert MTB/RIF, the SOE PCR method was completely consistent with results of the GeneXpert MTB/RIF, with a concordance of 100% for resistance to RFP.ConclusionsIn present study, a novel SOE PCR diagnostic method was successfully developed for the accurate detection of M. tuberculosis drug resistance. Our results using this method have a high consistency with that of traditional phenotypic DST or GeneXpert MTB/RIF, and SOE PCR testing in clinical isolates can also be conducted rapidly and simultaneously for detection of drug resistance to RFP, EMB, and INH.Graphical

Distribution of nontuberculous mycobacteria in patients with and without HIV/AIDS in Chongqing.

In this study, the distribution of nontuberculous mycobacteria (NTM) strains in patients with and without HIV/AIDS in Chongqing, China was evaluated. A retrospective study was performed in January-December 2020 at Chongqing Public Health Medical Center. NTM strains were assessed by a multi locus phylogenetic analysis. The distribution of NTM strains in HIV/AIDS and non-HIV/AIDS groups was compared. CD4+ cell counts, imaging changes, and characteristics of mycobacterial species were determined. In total, 324 patients with NTM infection (50 patients with HIV/AIDS and 274 patients without HIV/AIDS) were included. The most common etiological agent was M.abscessus (29%), followed by M.paraintracellulare (12%) andM.colombiense (11%). Predominant NTM species were M.avium (26%), M.colombiense (24%), and M.kansasii (18%) in patients with HIV/AIDS and wereM.abscessus (32%), M.paraintracellulare (13%), M.fortuitum (10%), and M. intracellulare (10%) in patients without HIV/AIDS. For a CD4+ cell count of <200/μl, the predominant species were M.aviumin the HIV/AIDS group and M.abscessus in the non-HIV/AIDS group. With respect to radiologic characteristics, different NTM strains were associated with distinct imaging manifestations; for example, M.marseillense, M.kansasii, and M.parasenchytosis were more likely to induce cavities. Imaging cavities, bronchiectasis, and acinar-like changes were more common in the non-HIV/AIDS groups. The infection rates of HIV and NTM in Chongqing are high, while M.abscessus, M.paraintracellulare, and M.colombiense are the main pathogens causing NTM diseases in Chongqing, and NTM strains differed significantly between patients with and without HIV/AIDS. Monitoring these indicators can help develop prevention strategies.

Characteristics of tuberculosis patients in the integrated tuberculosis control model in Chongqing, China: a retrospective study

BackgroundChina ranks second in the world in terms of numbers of tuberculosis (TB) cases and is one of the top three countries with the largest number of multidrug-resistant and rifampicin-resistant TB (MDR/RR-TB). It also has high mortality and low cure rates of human immunodeficiency virus (HIV)-positive TB patients. This study aimed to analyse, under the integrated TB control model, the characteristics of TB patients seeking healthcare in the largest designated TB hospital in Chongqing.MethodsThis was a retrospective study of TB registers in a health facility. Record data of 1827 TB patients who had attended the Chongqing Public Health Medical Center (CPHMC) from 1 January to 31 December 2018 were included. The Statistical Package for Social Science (SPSS 18.0; IBM Corporation, Armonk, NY, USA) was used to analyse the data. Counting data were compared using the chi-square test or Fisher’ s exact test. Among the results of the univariate analysis, the variables with statistical significance were included in the binomial stepwise logistic regression, with odds ratio and 95% confidence interval calculated. A two-tailed probability level of P < 0.05 was considered statistically significant.ResultsThe majority of registered patients were men (1197), of Han ethnicity (1670), aged 21–60 years (1331), farmer/unemployed (1075), and living in county/district (1207). Approximately 24.9% of patients (455/1827) contracted DR-TB, 6% (110/1827) were co-infected with HIV, and 41.0% (749/1827) had drug-related hepatotoxicity. Among those patients, DR-TB was more likely to develop among farmers who received retreatment and had drug-related hepatotoxicity (P < 0.05). Women who received retreatment and lived in county/district were less likely to be HIV positive (P < 0.05). Compared with farmers, patients who were unemployed were more likely to be HIV positive, and those aged 21–60 years had a higher risk of being tested as HIV positive (P < 0.05).ConclusionFarmers who received retreatment and had drug-related hepatotoxicity are more susceptible to DR-TB; young unemployed men have a higher risk of contracting HIV-positive TB. The demographic and clinical characteristics of TB patients should be taken into consideration in DR-TB and HIV-positive TB screening in the future.

High Incidence of Asymptomatic SARS-CoV-2 Infection, Chongqing, China

Background: SARS-CoV-2 has been a global pandemic, but the emergence of asymptomatic patients has caused difficulties in the prevention of the epidemic. Therefore, it is significant to understand the epidemiological characteristics of asymptomatic patients with SARS-CoV-2 infection. Methods: In this single-center, retrospective and observational study, we collected data from 167 patients with SARS-CoV-2 infection treated in Chongqing Public Health Medical Center (Chongqing, China) from January to March 2020. The epidemiological characteristics and variable of these patients were collected and analyzed. Findings: 82.04% of the SARS-CoV-2 infected patients had a travel history in Wuhan or a history of contact with returnees from Wuhan, showing typical characteristics of imported cases, and the proportion of severe COVID-19 patients was 13.2%, of which 59% were imported from Wuhan. For the patients who was returnees from Wuhan, 18.1% was asymptomatic patients. In different infection periods, compared with the proportion after 1/31/2020, the proportion of asymptomatic patient among SARS-CoV-2 infected patient was higher(19% VS 1.5%). In different age groups, the proportion of asymptomatic patient was the highest(28.6%) in children group under 14, next in elder group over 70 (27.3%). Compared with mild and common COVID-19 patients, the mean latency of asymptomatic was longer (11.25 days VS 8.86 days), but the hospital length of stay was shorter (14.3 days VS 16.96 days). Conclusion: The SARS-CoV-2 prevention needs to focus on the screening of asymptomatic patients in the community with a history of contact with the imported population, especially for children and the elderly population. Funding Statement: This work was supported by the National Key Research and Development Program (No.2016YFC1101100) and National Science Fund for Distinguished Young Scholars (No. 31625011). Declaration of Interests: The authors declare no competing interests. Ethics Approval Statement: This study was approved by the Ethics Committee of Chongqing Public Health Medical Treatment Center (NO.2020-018-01-KY), informed consent was allowed to be waived.

Chest computed tomography findings and dynamic changes of severe coronavirus disease 2019

Objective To investigate the features of chest CT imaging and dynamic changes of severe coronavirus disease 2019 (COVID-19). Methods The clinical and computed tomography (CT) data of 17 patients diagnosed with severe COVID-19 admitted to Chongqing Public Health Medical Center from January 24 to February 6, 2020 were collected. The first chest CT manifestations and the dynamic changes of imaging during treatment were retrospectively analyzed. Results The first chest CT manifestations of the 17 patients showed that 16 cases presented with peripheral and subpleural distributions, and 2 cases presented with 3 lobes involved, one case with 4 lobes involved and 14 cases with 5 lobes involved, and 17 cases presented with ground-glass opacities, ten cases with consolidation, seven cases with subpleural line, nine cases with air bronchogram, 3 cases with thickened lobular septum, two cases with bronchiectasis, two cases with pleural effusion, two cases with lymphadenopathy with the short diameter of 1.0-1.2cm. Among 16 patients who underwent repeated CT examination, the lesions of 8 patients showed continuous improvement, and those of the other 8 patients showed fluctuating changes. Conclusions The CT findings of severe COVID-19 patients are mainly ground-glass opacities and consolidation, with the peripheral distribution. The range of lesions is wide, with 5-lobe involvement mostly. Lymphadenopathy or pleural effusion is rare. Chest CT is useful for the evaluation for the therapeutic effects. Key words: Coronavirus Infections; Pneumonia; 2019 novel coronavirus; Corona virus disease 2019; CT features; Tomography, X-ray computed

Analysis of clinical features of 153 patients with novel coronavirus pneumonia in Chongqing/ 中华临床感染病杂志

Objective To analyze the clinical data of 153 patients with novel coronavirus pneumonia (COVID-19) in chongqing ,and provide reference and thinking for the diagnosis and treatment. Methods Analyze the clinical data, laboratory examination and chest imaging characteristics of 153 COVID-19 patients in Chongqing Public Health Medical Center from January 26 to February 5, 2020. According to the relevant diagnostic criteria ,patients were divided into non-severe group(n=132) and severe group(n=21),and analyze the correlation between serum index changes and disease severity. Results Combined with diabetes and chronic respiratory diseases, the severity of the disease was statistically significant (χ2=11.04和6.94, P<0.05). No symptoms were found in patients with mild illness (χ2=4.09, P<0.05) .The proportion of fever and muscle soreness in the severe group was higher than that in the non-severe group (χ2=4.40 and 22.67,P<0.05).Among the concomitant symptoms, the proportion of cough and shortness of breath in the severe group was higher than that in the non-severe group (χ2=8.46 and 4.80,P<0.05).C-reactive protein and d-dimer were higher in the severe group than in the non-severe group (t=43.44 and 37.13, P<0.05), and the number of CD3+T lymphocyte cells, CD4+T lymphocyte cells and CD8+T lymphocyte cells in the severe group was lower than that in the non-severe group (Z=27.25, 20.60 and 17.36, P<0.05).Compared with the non-severe group, both lungs and the right lung lower lobe were more susceptible to involved (χ2=6.95和20.39, P<0.05) . Conclusion Severity of COVID-19 was associated with underlying disease, symptoms, site of involvement, C-reactive protein, d-dimer, CD3+T lymphocyte count, CD4+T lymphocyte count, and CD8+T lymphocyte count. Key words: 2019 novel Coronavirus; Clinical features; Severe disease

Molecular characterization of para-aminosalicylic acid resistant Mycobacterium tuberculosis clinical isolates in southwestern China.

PurposeThe aim of this study is to investigate para-aminosalicylic acid (PAS) resistance-related gene mutations in clinical Mycobacterium tuberculosis (MTB) isolates and analyze the associated risk factors in southwestern China.Patients and methodsTotal 122 PAS-resistant and 55 PAS-susceptible clinical isolates were obtained from Chongqing Public Health Medical Center between April 2014 and January 2018. Drug susceptibility test was performed, and the PAS resistance-related genes were sequenced.ResultsPAS-resistant strains were more likely to resist streptomycin (OR: 9.5, 95% CI: 3.87-23.3; P<0.01), isoniazid (OR: 5.98, 95% CI: 2.14–16.76; P<0.01), rifampin (OR: 5.01, 95% CI: 2.11–11.88; P<0.01), ethambutol (OR: 2.79, 95% CI: 1.44–5.4; P<0.01), levofloxacin (OR: 2.56, 95% CI: 1.33–4.93; P<0.01), and amikacin (OR: 4.29, 95% CI: 1.70–10.83; P<0.01). The sequencing results showed that 112 (91.8%) PAS-resistant strains harbored 30 different mutations in folC, thyA, and ribD. Mutations in folC were the most commonly observed in PAS-resistant isolates (54.5%, 61/112), followed by mutations in thyA and ribD. Residues I43 in folC, R235 in thyA, and −11G in upstream of ribD were hotspots for mutation sites.ConclusionPAS drug resistance in MTB in southwestern China is mainly caused by mutations in folC, thyA, and ribD, among which folC was the most frequent mutation. Some mutation hotspots exist in the three genes, which accounts for about 80% of total mutations. These results highlight the possibility of developing molecular diagnostic methods for PAS-resistant tuberculosis in the future.