Giachino Marina, Vetter Beatrice, Perone Sigiriya Aebischer, Correia Jorge César, Erkosar Berra, Heller Olivia, Khanal Vijay Kumar, Lab Bruno, Pataky Zoltan, Poudel Sagar, Rai Mamit, Sharma Sanjib Kumar
Division of Tropical and Humanitarian Medicine, Geneva University Hospitals, Geneva, Switzerland.
FIND, Geneva, Switzerland.
PLOS Glob Public Health. 2024 Oct 3;4(10):e0003760. doi: 10.1371/journal.pgph.0003760. eCollection 2024.
Non-communicable diseases (NCDs), such as cardiovascular disease and diabetes, represent a serious global health concern. There is an urgent need for prompt diagnosis and effective monitoring at point of care, especially in low- and middle-income countries. Here we present the results of a study assessing the quantitative accuracy of two devices that may fit the target product profile for a cardiometabolic point-of-care device. This prospective, quantitative, accuracy study (NCT05257564) was conducted between March to May 2022, investigating the performance of the JanaCare Aina Blood Monitoring System (JCAina) and the Tascom SimplexTAS 101 device (TAS101) compared with local standard laboratory methods in rural Nepal. Using fingerstick capillary blood, cardiometabolic parameters were analysed using both devices. The quantitative accuracy was compared against a local laboratory reference assay. System usability was also assessed. For JCAina, the mean absolute biases (Bland-Altman analysis) for glucose, HbA1c and total cholesterol tests were -3.87 mg/dL (95% CI: -7.52--0.22), 1.34% (95% CI: 1.21-1.47), and -9.52 mg/dL (95% CI: -11.9--7.2), respectively, corresponding to mean percentage biases of 2.0%, 18.5%, and -6.4%. These indicate clinically small (<10% biases) differences from laboratory results for glucose and cholesterol, and a moderate (10-20%) positive bias for HbA1c. For TAS101, the mean absolute biases for glucose, HbA1c, total cholesterol and creatinine tests were 18.7 mg/dL (95% CI: 15.8-21.5), -0.2% (95% CI: -0.26--0.14), 29.8 mg/dL (95% CI: 27.0-32.6), and -0.02 mg/dL (95% CI: -0.05-0.01), respectively, corresponding to mean percentage biases of 12.1%, -2.6%, 15.8%, and -4.5%. These indicate clinically small differences for HbA1c and creatinine, and moderate positive biases for glucose and cholesterol. Both systems exhibited usability challenges. The JCAina and TAS101 point-of-care cardiometabolic devices were shown to have promising accuracy in environmental conditions such as in Nepal, though improvements are still needed for some parameters and for ease of use. Trial registration: NCT05257564 (ClinicalTrials.gov).
心血管疾病和糖尿病等非传染性疾病是全球严重的健康问题。迫切需要在医疗现场进行快速诊断和有效监测，尤其是在低收入和中等收入国家。在此，我们展示了一项研究的结果，该研究评估了两种可能符合心脏代谢即时检测设备目标产品轮廓的设备的定量准确性。这项前瞻性、定量准确性研究（NCT05257564）于2022年3月至5月进行，在尼泊尔农村地区将JanaCare Aina血液监测系统（JCAina）和Tascom SimplexTAS 101设备（TAS101）的性能与当地标准实验室方法进行了比较。使用手指刺血采集的毛细血管血，通过这两种设备分析心脏代谢参数。将定量准确性与当地实验室参考检测进行比较。还评估了系统的可用性。对于JCAina，葡萄糖、糖化血红蛋白（HbA1c）和总胆固醇检测的平均绝对偏差（Bland-Altman分析）分别为-3.87mg/dL（95%CI：-7.52--0.22）、1.34%（95%CI：1.21-1.47）和-9.52mg/dL（95%CI：-11.9--7.2），对应的平均百分比偏差分别为2.0%、18.5%和-6.4%。这表明葡萄糖和胆固醇与实验室结果在临床上差异较小（偏差<10%），而HbA1c存在中度（10-20%）正偏差。对于TAS101，葡萄糖、HbA1c、总胆固醇和肌酐检测的平均绝对偏差分别为18.7mg/dL（95%CI：15.8-21.5）、-0.2%（95%CI：-0.26--0.14）、29.8mg/dL（95%CI：27.0-32.6）和-0.02mg/dL（95%CI：-0.05-0.01），对应的平均百分比偏差分别为12.1%、-2.6%、15.8%和-4.5%。这表明HbA1c和肌酐在临床上差异较小，而葡萄糖和胆固醇存在中度正偏差。两个系统都存在可用性挑战。JCAina和TAS101即时检测心脏代谢设备在尼泊尔等环境条件下显示出有前景的准确性，不过在一些参数和易用性方面仍需改进。试验注册：NCT05257564（ClinicalTrials.gov）。
