Clin Chem Lab Med 2012;50(1):185–187 © 2011 by Walter de Gruyter • Berlin • Boston. DOI 10.1515/CCLM.2011.749 Letter to the Editor Validity of the measurement of capillary cholesterol in the diagnosis of defined hypercholesterolemia Jesús López-Torres Hidalgo*, Ignacio Párraga Martínez, Rafael Muñoz Sánchez-Villacañas, Jose María del Campo del Campo, Alejandro Villena Ferrer and Susana Morena Rayo Health Care Service of Castilla-La Mancha (SESCAM), Albacete, Spain Keywords: cholesterol; hypercholesterolemia; lipids; point- of-care testing; risk factors. Hypercholesterolemia is an important cardiovascular risk fac- tor that must be detected and requires regular tests to deter- mine its evolution. The US Preventive Services Task Force recommends that screening includes determination of total cholesterol (TC) and high density lipoprotein-cholesterol (HDL-C) concentrations in men over 35 years old and in women of 45 years and over (1). A patient is considered to have “limiting hypercholes- terolemia” if he/she presents TC levels between 200 and 249 mg/dL (1 mmol/L =39 mg/dL) and “defined hypercholes- terolemia” at levels >250 mg/dL (2). TC is usually quanti- fied in the laboratory by enzymatic methods, but equipment that is based on dry chemistry, such as Accutrend GC, uses a reflectometric technique. In a few minutes and with minimum discomfort for the patient, this apparatus can measure TC in a sample of capillary blood. These features make this easy to use in an ambulatory setting as a method to detect hypercho- lesterolemia (3, 4). Some studies defend the use of capillary tests to determine TC concentrations (5). However, there is still little informa- tion available to be able to establish whether or not it is a valid technique for the detection of TC levels. The purpose of this study is to analyze the validity of TC measurement in capillary blood as a method to identify defined hypercholesterolemia. This is an observational study carried out in a primary care setting (Medical Centers of Hellín and Zone V-B/VI of Albacete, Spain). The study was carried out in 112 patients with a history of hypercholesterolemia ( ≥200 mg/dL), which were selected consecutively. The patients were requested to make a blood extraction to determine lipid profile in the reference laboratory (Albacete University Hospital). The concentration of TC in serum was determined by the CHOD- PAP method (cholesterol oxidase-phenol+aminophenazone). The laboratory is accredited by the Spanish Society of Clinical Biochemistry (SEQC). Next, the TC in capillary blood measured by Accutrend GC (range of possible values: 150–300 mg/dL) was determined. Both measurements were recorded correctly in 91 patients (81.25%), and in the remain- ing patients could not be carried out owing to incorrect tech- nique, missing data for plasma levels or the patient’ s refusal to have the capillary prick test. Other parameters were also determined in participants by family physicians or were obtained from the clinical history, such as: cigarettes consumption (WHO definition), blood pressure, glycemia waist perimeter, level of cardiovascular risk (SCORE table) and the presence of metabolic syndrome (NCEP ATP-III) (6). To avoid possible sources of error, we proceeded to the proper calibration of the equipment used and the training of the nursing staff (a training session with volunteer patients). The study was approved by the Albacete Institutional Review Board. The statistical analysis consisted in a description of par- ticipants and a comparison of the mean levels of TC by Student’s t-test for paired values. The correlation was mea- sured between both measurements to verify their linear asso- ciation (Pearson’s r) and their level of consistency, estimating the mean correlation between all the possible arrangements of paired observations (intra-class correlation coefficient). Agreement was also determined by the Bland and Altman method, based on individual differences, representing graphi- cally the differences observed. The percentage of bias or sys- tematic error was determined by the formula: [(total capillary C – total plasma C)/total plasma C]×100. To assess the validity of the capillary determination, plasma determination was used as the gold standard, cal- culating the sensitivity, the specificity and the predictive values to identify a possible defined hypercholesterolemia. Other parameters calculated were Youden’ s J statistics (diagnostic safety of the test) and variation coefficient in both determinations. Among the participants, composed of 60.7% women and a mean age of 54.1 years (SD =14.0), 45.4% presented a moderate/high cardiovascular risk (SCORE) and 11.1% ful- filled metabolic syndrome criteria. *Corresponding author: Jesús López-Torres Hidalgo, Gerencia de Atención Primaria, c/Marqués de Villores n ° 6, 02001 Albacete, Spain Phone: +34 967195131, E-mail: jesusl@sescam.org Received September 13, 2011; accepted September 14, 2011; previously published online October 8, 2011 Brought to you by | Karolinska Institute Authenticated Download Date | 5/24/15 5:00 AM