Ethiopian Public Health Institute, Addis Ababa, Ethiopia.
St Paul's Hospital Millennium Medical CollegeAddis Ababa, Addis Ababa, Ethiopia.
PLoS One. 2021 Jun 28;16(6):e0253848. doi: 10.1371/journal.pone.0253848. eCollection 2021.
Even though the lives of millions have been saved in the past decades, the mortality rate in patients with drug-resistant tuberculosis is still high. Different factors are associated with this mortality. However, there is no comprehensive global report addressing these risk factors. This study aimed to determine the predictors of mortality using data generated at the global level.
We systematically searched five electronic major databases (PubMed/Medline, CINAHL, EMBASE, Scopus, Web of Science), and other sources (Google Scholar, Google). We used the Joanna Briggs Institute Critical Appraisal tools to assess the quality of included articles. Heterogeneity assessment was conducted using the forest plot and I2 heterogeneity test. Data were analyzed using STATA Version 15. The pooled hazard ratio, risk ratio, and odd's ratio were estimated along with their 95% CIs.
After reviewing 640 articles, 49 studies met the inclusion criteria and were included in the final analysis. The predictors of mortality were; being male (HR = 1.25,95%CI;1.08,1.41,I2;30.5%), older age (HR = 2.13, 95%CI;1.64,2.62,I2;59.0%,RR = 1.40,95%CI; 1.26, 1.53, I2; 48.4%) including a 1 year increase in age (HR = 1.01, 95%CI;1.00,1.03,I2;73.0%), undernutrition (HR = 1.62,95%CI;1.28,1.97,I2;87.2%, RR = 3.13, 95% CI; 2.17,4.09, I2;0.0%), presence of any type of co-morbidity (HR = 1.92,95%CI;1.50-2.33,I2;61.4%, RR = 1.61, 95%CI;1.29, 1.93,I2;0.0%), having diabetes (HR = 1.74, 95%CI; 1.24,2.24, I2;37.3%, RR = 1.60, 95%CI;1.13,2.07, I2;0.0%), HIV co-infection (HR = 2.15, 95%CI;1.69,2.61, I2; 48.2%, RR = 1.49, 95%CI;1.27,1.72, I2;19.5%), TB history (HR = 1.30,95%CI;1.06,1.54, I2;64.6%), previous second-line anti-TB treatment (HR = 2.52, 95% CI;2.15,2.88, I2;0.0%), being smear positive at the baseline (HR = 1.45, 95%CI;1.14,1.76, I2;49.2%, RR = 1.58,95%CI;1.46,1.69, I2;48.7%), having XDR-TB (HR = 2.01, 95%CI;1.50,2.52, I2;60.8%, RR = 2.44, 95%CI;2.16,2.73,I2;46.1%), and any type of clinical complication (HR = 2.98, 95%CI; 2.32, 3.64, I2; 69.9%). There are differences and overlaps of predictors of mortality across different drug-resistance categories. The common predictors of mortality among different drug-resistance categories include; older age, presence of any type of co-morbidity, and undernutrition.
Different patient-related demographic (male sex, older age), and clinical factors (undernutrition, HIV co-infection, co-morbidity, diabetes, clinical complications, TB history, previous second-line anti-TB treatment, smear-positive TB, and XDR-TB) were the predictors of mortality in patients with drug-resistant tuberculosis. The findings would be an important input to the global community to take important measures.
尽管在过去几十年中挽救了数百万人的生命,但耐多药结核病患者的死亡率仍然很高。不同的因素与这种死亡率有关。然而,目前还没有全面的全球报告来解决这些风险因素。本研究旨在确定使用全球水平产生的数据预测死亡率的因素。
我们系统地检索了五个主要电子数据库(PubMed/Medline、CINAHL、EMBASE、Scopus 和 Web of Science)和其他来源(Google Scholar、Google)。我们使用 Joanna Briggs 研究所的批判性评估工具来评估纳入文章的质量。使用森林图和 I2 异质性检验来评估异质性。使用 STATA 版本 15 进行数据分析。估计了合并的风险比、风险比和优势比及其 95%置信区间。
在审查了 640 篇文章后,49 项研究符合纳入标准并纳入最终分析。死亡的预测因素包括:男性(HR = 1.25,95%CI;1.08-1.41,I2;30.5%)、年龄较大(HR = 2.13,95%CI;1.64-2.62,I2;59.0%)、RR = 1.40,95%CI;1.26-1.53,I2;48.4%),包括年龄增加 1 岁(HR = 1.01,95%CI;1.00-1.03,I2;73.0%)、营养不良(HR = 1.62,95%CI;1.28-1.97,I2;87.2%)、RR = 3.13,95%CI;2.17-4.09,I2;0.0%)、任何类型的合并症(HR = 1.92,95%CI;1.50-2.33,I2;61.4%)、RR = 1.61,95%CI;1.29-1.93,I2;0.0%)、患有糖尿病(HR = 1.74,95%CI;1.24-2.24,I2;37.3%)、RR = 1.60,95%CI;1.13-2.07,I2;0.0%)、HIV 合并感染(HR = 2.15,95%CI;1.69-2.61,I2;48.2%)、RR = 1.49,95%CI;1.27-1.72,I2;19.5%)、TB 病史(HR = 1.30,95%CI;1.06-1.54,I2;64.6%)、先前的二线抗结核治疗(HR = 2.52,95%CI;2.15-2.88,I2;0.0%)、基线时痰液阳性(HR = 1.45,95%CI;1.14-1.76,I2;49.2%)、RR = 1.58,95%CI;1.46-1.69,I2;48.7%)、XDR-TB(HR = 2.01,95%CI;1.50-2.52,I2;60.8%)、RR = 2.44,95%CI;2.16-2.73,I2;46.1%)和任何类型的临床并发症(HR = 2.98,95%CI;2.32-3.64,I2;69.9%)。不同耐药类别之间存在死亡预测因素的差异和重叠。不同耐药类别中死亡的共同预测因素包括:年龄较大、任何类型的合并症和营养不良。
耐多药结核病患者死亡的预测因素包括患者相关的人口统计学因素(男性、年龄较大)和临床因素(营养不良、HIV 合并感染、合并症、糖尿病、临床并发症、TB 病史、先前的二线抗结核治疗、痰液阳性和 XDR-TB)。这些发现将为全球社会提供重要的信息,以采取重要措施。