Evaluation of a POCT device for C-reactive protein, hematocrit and leukocyte differential.

The area of point-of-care testing (POCT) is rapidly growing and holds great promise in terms of speed, accessibility, patient empowerment and improved dialogue between health care practitioners and patients [1]. Although for many POC tests the added value in clinical decision making has not yet been firmly established [2], a number of studies have recently appeared on the value of POCT measurement of biomarkers of inflammation [mainly C-reactive protein (CRP)] in patients presenting to the general practitioner with signs of a respiratory tract infection [3, 4]. It was found that POC testing for CRP changed the decision whether or not to prescribe antibiotics in a significant proportion of cases and led to an overall reduction in antibiotic usage (depending on the baseline prescribing behavior of the general practitioners) [5]. Other commonly ordered tests are the blood count and leukocyte differential. Although hemoglobin POCT systems have already found widespread acceptance, leukocyte counts and differential are still almost exclusively carried out in centralized laboratories. One obstacle to their introduction as a POC test has been the requirement for ‘yet another meter’. Spinit (Biosurfit SA, Lisboa, Portugal) is a POCT device combining multiple detection modes (i.e. surface plasmon resonance, spectroscopy, microcentrifugation and digital microscopy with image recognition) in one device. The disposable measurement cartridges that the equipment uses are actually compact disks (CDs) containing microfluidic channels. CDs for CRP have been on the market since 2011, and CDs for blood count have recently been introduced, measuring hematocrit (Ht), white blood cell count (WBC) and 5-part differential. We present here the first report (to the best of our knowledge) of an independent evaluation of Spinit, including the precision of the CRP and blood count measurements and comparison with our laboratory methods. Spinit was evaluated in a primary care setting; therefore, a sufficient range of eosinophil and basophil counts could not be obtained. Because Spinit requires whole blood, EDTA-anticoagulated hematology samples were used for all measurements on Spinit (blood count and CRP), whereas for the measurement of CRP using the laboratory method, heparin plasma was routinely used. This approach ensures optimal comparability to standard practice. The precision of the CRP measurement was assessed by a modified Evaluation Protocol 5 of the Clinical and Laboratory Standards Institute (CLSI EP5 – Complex Precision) in which “between-lot” precision was substituted for “between-run” precision. Briefly, using one instrument, a high (155 mg/L) and a low (14 mg/L) sample were measured on three consecutive days using two different lots of CDs in duplicate. An EP5 complex precision experiment could not be performed for the blood count because the Spinit digital image recognition algorithm is at present not yet compatible with stabilized quality control materials because of differing morphological properties compared with native blood cells. Therefore, a “simple precision” experiment was carried out. Briefly, two samples (Ht 0.44/WBC 25 · 109 L − 1 and Ht 0.29/WBC 4 · 109 L − 1) were measured 13 times on 1 day. The results of the precision experiments are given in Table 1. Except for CRP at the low level, all CVs are *Corresponding author: Albert J. de Graaf, PhD, Medlon B.V., PO Box 50000, 7500 KA Enschede, The Netherlands; and Ziekenhuisgroep Twente, Almelo, The Netherlands, E-mail: [email protected]. http://orcid.org/0000-0001-5451-3010 Sietske W. Hiemstra: Medlon B.V., Enschede, The Netherlands Evelien W.M. Kemna: Medlon B.V., Enschede, The Netherlands; and Medisch Spectrum Twente, Enschede, The Netherlands Johannes G. Krabbe: Medlon B.V., Enschede, The Netherlands; Ziekenhuisgroep Twente, Almelo, The Netherlands; and Medisch Spectrum Twente, Enschede, The Netherlands