Accuracy of waste blood measurement in critically ill patients

Abstract

Dear Editor, Timely identification and enrollment of critically ill patients into clinical trials is a challenge. Residual blood samples (‘‘waste’’ blood) obtained for routine clinical tests is a precious and potentially underutilized source of biospecimens. The accurate measurement of specific biomarker using waste blood may be compromised due to sample handling, storage, and the quality control of the platform used [1, 2]. We hypothesized that waste blood samples, if appropriately processed, provide accurate results and are comparable to fresh samples. Adults with increased risk for acute lung injury (ALI), using a lung injury prediction score (LIPS) [3], were identified daily within 12 h of hospital admission. EDTA plasma samples were collected and processed within 2 h, and divided in two aliquots, one immediately frozen at -80 C (fresh blood sample), and the other kept at 4 C for 24 h before being frozen at -80 C (waste blood sample). The plasma was used for multiplex cytokine analysis (Procarta Cytokine kit, Affymetrix Panomics, Fremont, CA, USA) and evaluation of soluble receptor for advanced glycation end-products (sRAGE) (R&D Systems, Minneapolis, MN, USA). Bland–Altman analysis [4] was performed by using JMP 8 (SAS Institute Inc., Cary, NC, USA) to determine the agreement between paired comparisons of fresh and waste blood for the cytokines and sRAGE. The study was approved by the Institutional Review Board (09-000881) and was performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. Between May and December 2009, plasma samples of 29 patients were analyzed: median LIPS score of 4 (IQR 3.5–4.5). Sepsis was the most common risk factor (76%). The Bland–Altman plot (mean bias ? SD, limits of agreement) showed good agreement for IL-1ra (-9 ± 79 pg/ml, -166 to 148 pg/ml), IL-6 (0.9 ± 16 pg/ml, -15 to 33 pg/ml), IL-8 (-0.3 ± 1.6 pg/ml, -3.4 to 2.8 pg/ml), IL-12 (p40) (-0.2 ± 1.4 pg/ml, -2.8 to 2.6 pg/ml), MCP-1(-0.2 ± 11 pg/ml, -22.2 to 21.8 pg/ml), and sRAGE (-24 ± 325 pg/ml, -626 to 673 pg/ml) between fresh blood and waste blood samples (Fig. 1). In this prospective study of critically ill patients at risk of ALI, cytokines and sRAGE measurements from fresh and waste blood samples provided comparable results. This observation has important implications for future mechanistic studies allowing for more liberal use of rapidly separated and properly stored clinical ‘‘waste’’ samples. This strategy can increase the time window to obtain informed consent and limit the amount of blood collected, a common cause of anemia in critically ill patients [5]. This study has several limitations including the small sample size of this cohort, the timing of blood collection, and a biomarker panel limited to epithelial injury and inflammation. Nevertheless, the ability to use properly collected and stored blood from clinical samples for investigational use broadens the possibility of obtaining samples that would otherwise be scarce or impossible to collect for research purposes only. However, it remains prudent to evaluate and validate the stability of each biomarker of interest before using stored blood samples for research.