Abstract

Evaluation of point-of-care CD4 and toxicity monitoring for resource-limited ART clinic settings in Mozambique

Presented by Ilesh Jani (Mozambique).

I. Jani¹, N. Sitoe¹, P. Chongo¹, J. Quevedo², O. Tobaiwa², J. Lehe², T. Peter<sup>2

1</sup>Instituto Nacional da Saude, Hospital Central de Maputo, Departamento de Imunologia, Maputo, Mozambique, ²Clinton Health Access Initiative, Maputo, Mozambique

Background: Point-of-care (POC) diagnostic technologies enable ART-related testing to be extended into rural areas with limited test access. Selection of appropriate POC platforms is difficult due to the wide range of available POC devices and because new technologies often have limited track record. We undertook an objective evaluation process to identify the most suitable POC devices for use in ART clinics in Mozambique.
Methods: A survey of available POC CD4, chemistry and hematology platforms was conducted. Information on their operational and technical features were collated and assessed using an objective scorecard, which grades devices against a set of ideal POC specifications, such as ease of use, portability, heat stability, throughput, etc. The highest scoring devices underwent a technical evaluation at two ART clinics. Finger prick blood was collected for POC tests and tested immediately at the clinics, while venous blood collected simultaneously rom the same patients underwent laboratory testing with conventional CD4, chemistry and hematology platforms: Becton Dickinson FACSCalibur^TM, Vital Selectra^TM and Sysmex KX21^TM, respectively.
Results: Of 50 devices screened, 18 platforms were selected and graded using the scorecard. The highest scoring CD4 (Inverness PIMA^TM), chemistry (Roche Reflotron^TM) and hematology (Hemocue^TM) devices proceeded to technical evaluation. All devices demonstrated acceptable performance against reference testing. Correlation, Bland-Altman bias and levels of clinical misclassification for each platform were as follows: PIMA (0.81; -26.5 cells/µl; 6.1%), Reflotron ALT (0.99; -1.1 IU; 1.0%), Reflotron AST (0.81; -3.6 IU; 3.0%), and Hemocue (0.93; 0.69 g/dl; 3.3%). These differences were not clinically significant.
Conclusions: This study has identified suitable platforms that can be used in ART clinic settings for staging and monitoring HIV patients. We recommend this methodology as an objective approach to selecting appropriate technologies, especially as new diagnostic platforms for major diseases become available.