Predictive model and risk analysis for coronary heart disease in people living with HIV using machine learning.

OBJECTIVE: This study aimed to construct a coronary heart disease (CHD) risk-prediction model in people living with human immunodeficiency virus (PLHIV) with the help of machine learning (ML) per electronic medical records (EMRs). METHODS: Sixty-one medical characteristics (including demography information, laboratory measurements, and complicating disease) readily available from EMRs were retained for clinical analysis. These characteristics further aided the development of prediction models by using seven ML algorithms [light gradient-boosting machine (LightGBM), support vector machine (SVM), eXtreme gradient boosting (XGBoost), adaptive boosting (AdaBoost), decision tree, multilayer perceptron (MLP), and logistic regression]. The performance of this model was assessed using the area under the receiver operating characteristic curve (AUC). Shapley additive explanation (SHAP) was further applied to interpret the findings of the best-performing model. RESULTS: The LightGBM model exhibited the highest AUC (0.849; 95% CI, 0.814-0.883). Additionally, the SHAP plot per the LightGBM depicted that age, heart failure, hypertension, glucose, serum creatinine, indirect bilirubin, serum uric acid, and amylase can help identify PLHIV who were at a high or low risk of developing CHD. CONCLUSION: This study developed a CHD risk prediction model for PLHIV utilizing ML techniques and EMR data. The LightGBM model exhibited improved comprehensive performance and thus had higher reliability in assessing the risk predictors of CHD. Hence, it can potentially facilitate the development of clinical management techniques for PLHIV care in the era of EMRs.

Efficacy of dolutegravir + lamivudine q.d. with food in people with TB/HIV using a rifampicin-based regimen: A retrospective observational case series.

OBJECTIVES: Dolutegravir + lamivudine (DTG + 3TC) is a first-line regimen for people with HIV. However, there are still concerns about its efficacy in people with tuberculosis (TB)/HIV due to the lack of available evidence and drug-drug interaction with rifampicin. METHODS: A single-centre retrospective observational case series was conducted in Guangxi Zhuang Autonomous Region, China. We included all people with TB/HIV on combined use of once-daily (q.d.) dosing DTG + 3TC and rifampicin (RIF)-containing anti-TB regimens between 2020 and 2022. HIV-RNA, CD4 cell counts were collected and analysed. RESULTS: In all, 21 people with HIV (PWH) were included in this study. All the PWH were treatment-naïve and told to take DTG + 3TC q.d. with food. The median age was 53 years, and 71.43% were male. A total of 71.43% PWH had baseline viral load (VL) > 100 000 copies/mL, and 33.33% had baseline VL greater than 500 000 copies/mL. Only one PWH had CD4 cell count greater than 200 cells/µL, and the median CD4 count was 20 cells/µL. A total of 16 PWH started DTG + 3TC after initiation of the RIF-based anti-TB regimen, and the other five PWH initiated DTG + 3TC before the treatment of TB. All the PWH had at least 24 weeks of follow-up visits and all of the TB treatments were successful. A total of 20 PWH (95.24%) achieved viral suppression (VL <50 copies/mL). All detected viral loads between weeks 24 and 48 were less than 200 copies/mL. Among the PWH who started DTG + 3TC after the initiation of RIF-based anti-TB regimen, all achieved viral suppression by week 24 except the non-suppressed PWH. CD4 counts were greatly improved after antiretroviral treatment: the median CD4 counts were raised from 20 to 171 cells/µL at week 48. No serious adverse events were reported. CONCLUSIONS: This case series preliminarily validates the efficacy of DTG + 3TC q.d. with food when combined with RIF-based anti-TB regimens in people with TB/HIV.

Prediction of the risk of cytopenia in hospitalized HIV/AIDS patients using machine learning methods based on electronic medical records.

Background: Cytopenia is a frequent complication among HIV-infected patients who require hospitalization. It can have a negative impact on the treatment outcomes for these patients. However, by leveraging machine learning techniques and electronic medical records, a predictive model can be developed to evaluate the risk of cytopenia during hospitalization in HIV patients. Such a model is crucial for designing a more individualized and evidence-based treatment strategy for HIV patients. Method: The present study was conducted on HIV patients who were admitted to Guangxi Chest Hospital between June 2016 and October 2021. We extracted a total of 66 clinical features from the electronic medical records and employed them to train five machine learning prediction models (artificial neural network [ANN], adaptive boosting [AdaBoost], k-nearest neighbour [KNN] and support vector machine [SVM], decision tree [DT]). The models were tested using 20% of the data. The performance of the models was evaluated using indicators such as the area under the receiver operating characteristic curve (AUC). The best predictive models were interpreted using the shapley additive explanation (SHAP). Result: The ANN models have better predictive power. According to the SHAP interpretation of the ANN model, hypoproteinemia and cancer were the most important predictive features of cytopenia in HIV hospitalized patients. Meanwhile, the lower hemoglobin-to-RDW ratio (HGB/RDW), low-density lipoprotein cholesterol (LDL-C) levels, CD4+ T cell counts, and creatinine clearance (Ccr) levels increase the risk of cytopenia in HIV hospitalized patients. Conclusion: The present study constructed a risk prediction model for cytopenia in HIV patients during hospitalization with machine learning and electronic medical record information. The prediction model is important for the rational management of HIV hospitalized patients and the personalized treatment plan setting.

Epidemiology of extrapulmonary tuberculosis in central Guangxi from 2016 to 2021.

The burden of extrapulmonary tuberculosis (EPTB) has gradually increased in recent years, but not enough epidemiological data is available from central Guangxi. To better understand the epidemiology of EPTB in central Guangxi and identify risk factors associated with them, we retrospectively investigated the epidemiology of tuberculosis (TB), especially EPTB, among patients admitted to the Chest Hospital of Guangxi Zhuang Autonomous Region between 2016 and 2021. We excluded those infected with both pulmonary tuberculosis (PTB) and EPTB, reported the proportion and incidence of PTB or EPTB, and compared the demographic characteristics and risk factors of EPTB and PTB cases using univariate and multivariate logistic regression models. Among 30,893 TB patients, 67.25% (20,774) had PTB and 32.75% (10,119) had EPTB. Among EPTB, pleural, skeletal, lymphatic, pericardial, meningeal, genitourinary, intestinal, and peritoneal TB accounted for 49.44%, 27.20%, 8.55%, 4.39%, 3.36%, 1.48%, 0.87%, and 0.79%, respectively. Patients who were younger (age < 25), from rural areas, Zhuang and other ethnic groups, and diagnosed with anemia and HIV infection were more likely to develop EPTB. However, patients with diabetes and COPD were less likely to have EPTB. From 2016 to 2021, the proportion of PTB cases decreased from 69.73 to 64.07%. The percentage of EPTB cases increased from 30.27 to 35.93%, with the largest increase in skeletal TB from 21.48 to 34.13%. The epidemiology and risk factors of EPTB in central Guangxi are different from those of PTB. The incidence of EPTB is increasing and further studies are needed to determine the reasons for it.

[Laboratory diagnosis of HIV/AIDS patients complicated with Pneumocystis jirovecii pneumonia].

Pneumocystis jirovecii was detected in sputum samples and bronchoalveolar lavage fluid (BALF) obtained from HIV/AIDS patients complicated with Pneumocystis jirovecii pneumonia by Giemsa staining. CD4+ T lymphocytes of 500 patients were counted by flow cytometer. P. jirovecii positive rate in sputum samples (46.8%, 845/1 806) significantly lower than that of BALF (55.8%, 10(6)/190) (P < 0.05). The proportion of patients developing clinical symptoms in P. jirovecii positive cases (96.6%, 816/845) was higher than that of P. jirovecii negative cases (64.0%, 615/961) (P < 0.05). P. jirovecii positive rate increased with the decrease of CD4+ T lymphocyte number. P. jirovecii positive rates in cases with CD4+ > 200 x 10(6)/L, CDC 200 x 10(6)/L-100 x 10(6)n/L, and CD4+ < 100x10(6)/12.0% (6/50), 39.0%( 39/100), 54.6% (191/350), respectively (P < 0.05). Giemsa staining is an efficient, simple and feasible method for P. jirovecii detection, relying on the experience and skill of the operator.