Clinical and Quality Evaluation of Red Blood Cell Units Collected Via Apheresis Versus Those Obtained Manually Eiman Hussein, MD,1* Azza Enein, MD2 Lab Med Summer 2014;45:238-243 DOI: 10.1309/LMKXJ0Y44GPRSXFG ABSTRACT To evaluate the impact of collection procedure on the in vitro quality of red blood cells (RBC), we studied 30 units of apheresis-prepared RBC (ARBC) and 30 units of manually collected RBC (MRBC). We performed assays on day 1 and day 21 of the study, evaluating red cell mass volume (RCM); rate of hemolysis; pH, and levels of sodium, potassium, adenosine triphosphate (ATP), 2,3-diphosphoglycerate (2,3-DPG) and glucose. Eight patients with aplastic anemia received RBC transfusions of both components and their post-transfusion hematocrit (HCT) levels were compared. On day 21, we observed a significant drop of sodium and glucose levels in the ARBC group, compared New transfusion technology allows selective collection of different components via apheresis. This process involves collecting combinations of blood components, including red blood cells (RBC), from a single donation, which can reduce donor exposure and transmission of infectious diseases. Most RBC units used for transfusion are prepared from anticoagulated whole blood. Automated collection of RBC via apheresis can Abbreviations RBC, red blood cells; Hb, hemoglobin; HCT, hematocrit; FDA, United States Food and Drug Administration; ARBC, apheresis-prepared RBC; MRBC, manually collected RBC; CPDA-1, citrate-phosphate-dextrose adenine; RCM, red cell mass volume; Na, sodium; K, potassium; ATP, adenosine triphosphate; 2,3-DPG, 2,3-diphosphoglycerate; ACD-A, acid citrate dextrose; SAGM, saline-adenine-glucose-mannitol; PAGGGM, phosphate-adenine-guanosine-glucose-gluconate-mannitol Cairo University Blood Bank, Department of Clinical Pathology, Cairo University, Egypt; 2Hematology laboratory, Department of Clinical Pathology, Cairo University, Egypt. 1 *To whom correspondence should be addressed. E-mail: [email protected] 238 Lab Medicine Summer 2014 | Volume 45, Number 3 with the MRBC group (P <.05). ARBC group demonstrated higher RCM that provided significantly higher HCT values to our group of anemic patients (P <.05). Hemolysis was significantly lower in the ARBC group, compared with the MRBC group (P <.05). At day 21, both groups had no detectable 2,3-DPG. Specimens from both groups retained ATP in sufficiently healthy amounts. The ARBC group demonstrated higher RCM and lower hemolysis levels compared with the MRBC group. Keywords: apheresis-prepared red cells, manually prepared red cells, hemolysis, in vivo assay, in vitro assay, storage days. potentially provide a standardized, high-quality product with consistent RBC content.1 Hemolysis is an important parameter for measuring the quality of RBC. Hemolysis is caused by disruption of the RBC membrane and the subsequent release of hemoglobin (Hb). Hemolysis of RBC can occur during preparation and during storage. The extent of hemolysis is often defined as the percentage of free hemoglobin in relation to the hematocrit (HCT). According to the United States Food and Drug Administration (FDA) guidelines, hemolysis in stored RBC units should not exceed 1%. Recently, the FDA has added that 95% of units should meet quality standards and that the data for these units must show statistical certainty. To evaluate the impact of the collection procedure on the quality of packed RBC, we studied apheresis-prepared RBC (ARBC) prepared via the Gambro Trima Accel device (Gambro BCT, Lakewood, Colorado, USA) and manually collected RBC (MRBC) stored in bags containing citratephosphate-dextrose adenine (CPDA-1) anticoagulant. Another objective of our study was to investigate the impact of collection procedure on the viability of RBC after transfusion by measuring post-transfusion HCT levels. www.labmedicine.com Science Materials and Methods We designed and performed the study according to the tenets of the Declaration of Helsinki. A total of 30 ARBC units versus 30 MRBC units were analyzed. RBC units were stored for 21 days, under standard blood-banking conditions at a mean (SD) temperature of 4 (2) Cº. None of the units were leukoreduced. before operating the machine. Users load the tubing set on the top of the device; then, they snap a cassette into the device to make the machine ready for use. We performed in vitro studies of RBC on day 1 and day 21, evaluating absolute red cell mass volume (RCM); pH; rate of hemolysis; and levels of HCT, sodium (Na), potassium (K), adenosine triphosphate (ATP), 2,3-diphosphoglycerate (2,3-DPG), glucose, and plasma Hb. Whole blood is pumped into the machine and mixed with acid citrate dextrose (ACD-A) anticoagulant at a controlled ratio near the access needle. The blood components separate into layers over the entire circumference of the centrifuge, according to their density. Each component leaves the centrifuge via its own channel. RBC collection is initiated after platelet collection is completed. The components that are not collected are returned to the donor via return pump. In this study, we performed sessions using an initial anticoagulant ratio of 8:1 of whole blood to ACD-A. Collection and Preparation of RBC Units In Vitro RBC Analysis All RBC units were collected from donors who all met the American Association of Blood Banks (AABB) eligibility criteria for donation. We recruited donors from the apheresis center and the Blood Bank at Cairo University Hospital, Egypt. Men who weigh at least 155 pounds qualify for automated donation at our center. Trima Accel Apheresis Protocol for RBC Collection Samples of approximately 5 mL were collected anaerobically for testing from units immediately after collection, on day 1 and then subsequently on day 21. We performed in vitro assays of HCT, RCM, Na, K, ATP, 2,3-DPG, pH, glucose, plasma Hb, and rate of hemolysis using standard procedures. Specimens were immediately injected into a blood-gas analyzer to determine pH at 37 Cº. Na+ and K+ concentrations were determined via flame photometry. For K+ levels of greater than 20 mmol/L, we used a urine standard for calibration. ATP levels were measured enzymatically, as described by Adams. 2 2,3-DPG levels were measured as described by Rose and Liebowitz.3 Plasma Hb was measured using a modification of the Drabkin cyanmethemoglobin method4 which is read at 540 and 680 nm via spectrophotometer (Beckman Coulter, Inc, Brea, CA). We performed apheresis via a single-needle procedure using a Gambro Trima Accel machine. The target end points were set to 250 mL of RBC and collected the RBC along with 6- × 1011 to 9 × 1011 platelets, depending on donors’ data We calculated the absolute RBC mass volume using the following equation: Absolute RBC mass volume = total RBC volume × product HCT level Manual RBC Preparation We collected a mean (SD) of 450 (20 mL) of whole blood in triple blood bags, containing 63 mL of (CPD-A). We separated plasma from the RBC via centrifugation at 4200 g for 10 minutes. The Trima Accel system is designed to optimize the efficiency of component collection while maintaining donor safety; it is an automated continuous-flow centrifuge that separates whole blood into components for collection based on donor blood volume and complete blood count. The system has a single-stage channel with all blood components flowing in one direction. The Trima Accel collects platelets, RBC, and plasma. Users interface with a touch-screen graphical monitor to manually configure and adjust collections www.labmedicine.com Percent hemolysis was determined using the following equation: Percentage hemolysis = (1 - Hct) × plasma Hb (g/L) / Total Hb (g/L) In Vivo RBC Analysis We evaluated a total of 60 RBC transfusions in 8 patients, of whom 3 were women and 5 were men; their mean (SD) age was 29 (6) years. All patients were being Summer 2014 | Volume 45, Number 3 Lab Medicine 239 Science treated for aplastic anemia and had thrombocytopenia. Their mean (SD) pretransfusion HCT level was 18 (3)%. None of the patients had bleeding, hemolysis, or sepsis. Each patient received transfusions of both products; all 8 patients received 4 transfusions each of MRBC. Six patients received 4 ARBC transfusions; the remaining 2 patients received transfusions of 2 units each of ARBC. All transfused units were stored for less than 7 days. We drew peripheral blood samples from patients with anemia 24 hours after transfusion of 1 unit and assessed their HCT levels using a calibrated cell counter (Cobas Micros, F. Hoffman La Roche, Ltd., Basel, Switzerland). Statistical Analysis We performed standard statistical analysis of the test results, including calculation of descriptive statistics, mean, and SD. We used the Student’s t-test for comparative studies. A P value of less than .05 was considered to be statistically significant. Results In Vitro Evaluation of RBC Count MRBC had a volume of 240 to 300 mL per unit (mean [SD], 275 [35.5] mL/unit), RCM values of 158.5 to 240.0 mL/unit (192.5 [3.4] mL/unit), and HCT values of 66% to 80% (70% [9.5%]). ARBC had a volume of 240 to 250 mL/unit (mean [SD], 240 [8.5] mL/unit), RCM values of 204 to 237.5 mL/ unit (207.84 [7.7] mL/unit), and HCT values ranging from 85% to 95% (86.6% [9%]). ARBC showed less variability in HCT level and volume, compared with MRBC (P <.05). We observed a significant increase in hemolysis with storage in both products (P <.05). The platelet count was below the detection limit in both components. In vitro mean (SD) values for ARBCs and MRBCs on day 1 and day 21 are compiled in Table 1 and Table 2, respectively. In Vivo Evaluation of RBC Count The mean (SD) increase in HCT level after transfusion of ARBC was 4.5 (1.2%). The mean (SD) increase in HCT level after transfusion of MRBC was 2.4 (1.3%). The difference was statistically significant (P <.05). 240 Lab Medicine Summer 2014 | Volume 45, Number 3 Discussion We investigated the impact of collection procedure and storage on the in vitro characteristics of RBC, evaluating RCM; pH; rate of hemolysis; and levels of HCT, Na, K, ATP, 2,3-DPG, pH, glucose, and plasma Hb. We also prospectively studied the post-transfusion effectiveness of manually collected versus apheresis-collected RBC in patients with aplastic anemia. Our evaluation of the HCT level of ARBC units revealed that the HCT of these units was not dependent on donor HCT as it was in whole-blood donation; this was due to the fact that apheresis machines collect RBC according to absolute RBC volume. ARBC showed significantly higher RBC mass that provided higher HCT values to our group of patients who received transfusions. The RBC mass difference between the 2 collections can also be explained by the presence of women in the whole-blood collection group or possibly a greater number of lower-weight individuals in that group. We also observed low variability in the volume of ARBC in our study. Similar findings have been reported in previous studies.5,6 We observed a significant drop of Na and glucose on day 21 in the ARBC group compared with the MRBC group (P <.05). These data contrast with those reported by Picker et al,7 which revealed that the amounts of glucose consumed during 49 days of storage were higher for the ARBC group compared with the MRBC group, using saline-adenine-glucose-mannitol (SAGM) as an RBC-preservative solution.7 However, the lower levels of glucose in our ARBC group at day 21 may reflect the amount of glucose in the original anticoagulant rather than differences in the RBC metabolism between both products. Storage of both components showed no relevant drop in pH, and all units had a pH of 6.5 or greater. It has been reported8 that damage in ARBC is minimal compared with that in MRBC. During manual RBC collection, the whole blood is typically poured onto the anticoagulant, and some of the RBC are instantly damaged by the acid, due to change in osmolarity. This phenomenon has been described as collection injury.8 By contrast, in automated collection, the anticoagulant is continuously added at a constant small ratio,2 which minimizes collection injury. www.labmedicine.com Science Table 1. In Vitro Characteristics of ARBC and MRBC on Study Day 1 Mean (SD) In Vitro Variable (Units of Measure) ARBC MRBC P Value Hematocrit % Absolute RBC mass volume (ml/U) Plasma Hg (mg/dL) K (mmol/L) Hemolysis corrected for hematocrit % Na (mmol/L) Glucose (mg/dL) ATP (mmol/L) 2,3-DPG (mmol/L) pH 86.6 (9) 207.8 (7.7) 37 (14) 3.29 (0.13) 0.09 (0.03) 130 (5) 440 (13) 3.4 (0.03) 2.25 (0.081) 7.2 (0.08) 70 (9.5) 192.5 (3.4) 25 (17) 1.29 (0.18) 0.19 (0.05) 133 (5) 541 (20) 3.1 (0.1) 2.3 (0.09) 7 (0.06) <.05 <.05 <.05 <.05 <.05 .19 >.05 .27 .21 .18 ARBC, apheresis-collected red blood cells; MRBC, manually collected red blood cells; RBC, red blood cell; Hg, hemoglobin; K, potassium; Na, sodium; ATP, adenosine triphosphate; 2,3-DPG, 2,3-diphosphoglycerate Table 2. In Vitro Characteristics of ARBC and MRBC on Study Day 21 Mean (SD) In Vitro Variable (Units of Measure) ARBC MRBC P Value Hematocrit % 89 (7) 73 (9) <.05 Absolute RBC mass volume (ml/U) 213.6 (0.6) 200.7 (3.2) <.05 Plasma Hg (mg/dL) 170 (15.5) 80 (9) <.05 K (mmol/L) 55 (7) 40 (1.8) <.05 Hemolysis corrected for hematocrit % 0.41 (0.22) 0.7 (0.18) <.05 Na (mmol/L) 67 (4) 118 (10) <.05 Glucose (mg/dL) 180 (35) 270 (20) <.05 ATP (mmol/L) 3 (0.04) 3 (0.3) .2 2,3-DPG (mmol/L) 00NA pH 6.7 (0.04) 6.7 (0.04) .21 ARBC, apheresis-collected red blood cells; MRBC, manually collected red blood cells; RBC, red blood cell; Hg, hemoglobin; K, potassium; Na, sodium; ATP, adenosine triphosphate; 2,3-DPG, 2,3-diphosphoglycerate Holme et al9 reported a slightly lower rate of hemolysis (mean [SD], 0.44 [0.26] vs 0.61% [0.50%]), and lower supernatant K+ levels (50 [3] vs 53 [3] mEq/L) for ARBC prepared via the Haemonetics apheresis machine (Haemonetics Corporation, Braintree, MA) compared with manual units.9 These findings are in accordance to those of Moog and colleagues,5 who compared filtered SAGM ARBC units collected by Haemonetics machine with those collected manually. Our results demonstrated significantly higher levels of K+ and plasma Hb for the ARBC group on day 1 and day 21, when compared to the MRBC group. The plasma Hb values that we report for the ARBC group on day 1 (mean [SD], 37 [17] mg/dL) contrast with values reported in previous studies that used Trima10 and Amicus5 apheresis machines (Fenwal, Inc., Lake Zurich, IL). www.labmedicine.com Because of the different HCT values demonstrated in our study, it was essential to correct hemolysis for the HCT, to avoid overestimating the hemolysis percentage. We observed significantly lower rates of hemolysis in ARBC compared with MRBC. Picker et al7 reported different results, in which MRBC had an advantage for hemolysis at the end of specimen storage and were superior in energy maintenance. This was indicated by less ATP degradation and K+ leakage, which could be a possible consequence of less citrate. The authors also demonstrated higher methemoglobin formation with ARBC, compared with MRBC, due to differences in glucose metabolism.7 Summer 2014 | Volume 45, Number 3 Lab Medicine 241 Science Mean percentages of hemolysis for stored ARBC, as reported in the literature, ranged from .05% to .8%.1,7,10-12 This variation may likely be related to the different apheresis machines and to the different additive solutions that various study groups used. In the current study, we observed significant increases in hemolysis with storage in specimens from the ARBC and MRBC groups: from .09 (.03%) and .19 (.05%) to 0.41 (.22%) and .7 (.22%), respectively. All units met the United States standards of less than 1% hemolysis. None of our RBC units were leukoreduced. It has been demonstrated13-15 that leukocytes can significantly contribute to RBC hemolysis during storage, due to the release of enzymes, especially proteases. Other studies16,17 showed an increase in hemolysis even with leukoreduction. Many factors can also affect RBC quality, including the duration and force of centrifugation. The temperature of blood components during processing and storage can also contribute to hemolysis. The significant increase in hemolysis with MRBC may indicate that processing and/or storage of MRBC were not optimized. Hemolysis in our MRBC group can also be related to the low Hb content of the specimens from those RBC units. A previous study18 demonstrated higher rates of in vitro hemolysis in RBC with low Hb content. Several biochemical changes occur during storage that can affect recovery and survival of RBC in the recipient. The most dramatic changes are rapid decrease of ATP and 2,3-DPG. 2,3-DPG is important in the regulation of oxygen delivery by RBC.19 After transfusion, low levels of 2,3-DPG will increase in vivo. It takes at least 24 hours before normal levels are reached in healthy volunteers, although slower recoveries have been described in some patients. 20 ATP is a critical energy source for the overall functioning of RBC. Heaton21 demonstrated that ATP levels should be higher than 2.7 µmol per g of Hb, to yield more than 90% chance for a 24-hour recovery of 75% or higher. 21 Our ATP values of both components were well preserved during storage. Our data on ATP levels are consistent with those obtained in previous studies.5,22,23 The initial mean (SD) concentrations of 2,3-DPG for specimens from our MRBC and ARBC groups were 2.25 (0.08) and 2.3 (0.09) mmol/L, respectively. On day 242 Lab Medicine Summer 2014 | Volume 45, Number 3 21, both components had no detectable 2,3-DPG. We suggest that the high citric acid content and associated acidity in ACD-A explains the poor maintenance of 2,3DPG levels in both groups. Knutson and colleagues23 suggested that 2,3-DPG levels can be maintained by changing the anticoagulant used or by rapid cooling of the RBC unit immediately after preparation. Recently, it has been demonstrated19 that resuspension of RBC in phosphate-adenine-guanosine-glucose-gluconatemannitol (PAGGGM) can provide a high-quality unit throughout 35 days of storage, with 2,3-DPG levels higher than 10 µmol per g of Hb, ATP levels higher than 5 µmol per g of Hb, and hemolysis levels of less than 0.2%. Although the storage lesion is well documented, whether the age of RBC can adversely affect the outcome of transfusion remains in dispute.24-27 A 2006 study28 reported that transfusion of stored or recently donated blood to hypoxic patients has the same ability to restore evoked brain-stem potentials. In a previous study,29 however, it had been reported that transfusion of older stored RBCs produces extravascular hemolysis and circulating non– transferrin-bound iron, which may enhance complications such as infection. Four large multicenter trials are currently studying the impact of storage days on clinical outcomes. Strategies regarding storage age of RBC should await the results of such ongoing trials.30 One limitation of the current study is that although measurement of HCT levels 24 hours after transfusion can reflect short-term viability, tagging studies may be required to draw definitive conclusions regarding RBC viability. Conclusion Compared with the MRBC group, the ARBC group demonstrated less variability in volume, higher RCM, and lower rates of hemolysis. Transfusion of ARBCs that had been stored for less than 1 week provided significantly higher HCT levels than transfusion of MRBCs for our cohort of patients with anemia. LM Acknowledgments We thank the staff members of the Department of Internal Medicine, Hematology division, Cairo University, Egypt, for their cooperation. We also thank the staff members of the Cairo University Apheresis Center and the Blood Bank for their technical support. www.labmedicine.com Science References 17. Müller-Steinhardt M, Janetzko K, Kandler R, Flament J, Kirchner H, Kluter H. Impact of various red cell concentrate preparation methods on the efficiency of prestorage white cell filtration and on red cells during storage for 42 days. Transfusion. 1997;37(11-12):1137-1142. 1. Snyder EL, Elfath MD, Taylor H, et al. Collection of two units of leukoreduced RBCs from a single donation with a portable multiplecomponent collection system. Transfusion. 2003;43(12):1695-1705. 18.Zimmermann R, Heidenreich D, Weisbach V, Zingsem J, Neidhardt B, Eckstein R. In vitro quality control of red blood cell concentrates outdated in clinical practice. Transfus Clin Biol. 2003;10(4):275–283. 2. Adams H. ATP determination with phosphoglycerate kinase. In: Bergmeyer HU, ed. Methods of Enzymatic Analysis. New York:Academic Press,1963;539-543. 19.de Korte D, Kleine M, Korsten HGH, Verhoeven AJ. Prolonged maintenance of 2,3-diphosphoglycerate acid and adenosine triphosphate in red blood cells during storage. Transfusion. 2008;48(6):1081-1089. 3. Rose ZB, Liebowitz J. Direct determination of 2,3-diphosphoglycerate. Anal Biochem. 1970;35:177-180. 4. Moore GL, Ledford ME, Merydith A. A micromodification of the Drabkin hemoglobin assay for measuring plasma hemoglobin in the range of 5 to 2000 mg/dl. Biochem Med. 1981;26:167-173. 5. Moog R, Frank V, Pierce J, Muller N. Evaluation of a concurrent multicomponenet collection system for the collection and storage of WBC-reduced RBC apheresis concentrates. Transfusion. 2001;41(9):1159-1164. 20.Valeri CR, Hirsch NM. Restoration in vivo of erythrocyte adenosine triphosphate, 2,3-diphosphoglycerate, potassium ion, and sodium ion concentrations following the transfusion of acid-citrate-dextrosestored human red blood cells. J Lab Clin Med. 1969;7(5):722-733. 21.Heaton WA. Evaluation of posttransfusion recovery and survival of transfused red cells. Transfus Med Rev. 1992;6(3):153-169. 22. Zeiler TA, Kretschmer V. Automated blood component collection with the MCS 3P separator: Evaluation of three protocols for buffy coat-poor and white cell–reduced packed red cells and plasma. Transfusion. 1997;37(8):791-797. 6. Högman CF. Preparation and preservation of red cells. Vox Sang. 1998;74(S2):177-187. 23.Knutson F, Rider J, Franck V, Joie M, Högman CF, Pamphilon D. A new apheresis procedure for the preparation of high-quality red cells and plasma. Transfusion. 1999;39(6):565-571. 7. Picker SM, Radojska SM, Gathof BS. In vitro quality of red blood cells (RBCs) collected by multicomponent apheresis compared to manually collected RBCs during 49 days of storage. Transfusion. 2007;47:687-696. 24.Winslow RM, Intaglietta M. Red cell age and loss function: Advance or SNO-job? Transfusion. 2008;48(3):411-414. 8. Heaton WA. Enhancement of cellular elements. In: Wallas CI, McCarthy LJ, eds. New Frontiers in Blood Banking. Arlington, TX: AABB;1986:89-125. 25.Aubron C, Nichol A, Cooper DJ, Bellomo R. Age of red blood cells and transfusion in critically ill patients. Ann Intensive Care. 2013;3(1):2. 9. Holme S, Elfath MD, Whitley P. Evaluation of in vivo and in vitro quality of apheresis-collected RBC stored for 42 days. Vox Sang. 1998;75(3):212-217. 26.Koch CG, Li L, Sessler DI, et al. Duration of red-cell storage and complications after cardiac surgery. N Engl J Med. 2008;358(12):1229-1239. 10.Rugg N, Pitman C, Menitove JE, Greenwalt TJ, McAteer MJ. A feasibility evaluation of an automated blood component collection system platelets and red cells. Transfusion. 1999;39(5):460-464. 27. Wang D, Sun J, Solomon SB, Klein HG, Natanson C. Transfusion of older stored blood and risk of death: a meta-analysis. Transfusion. 2012;52(6):1184-1195. 11. Aubuchon JP, Dumont LJ, Herschel L, et al. Automated collection of double red blood cell units with a variable-volume separation chamber. Transfusion. 2008;48(1):147-152. 28.Weiskopf RB, Feiner J, Hopf H, et al. Fresh blood and aged stored blood are equally efficacious in immediately reversing anemiainduced brain oxygenation deficits in humans. Anesthesiology. 2006;104(5):911-920. 12.Picker SM, Radojska SM, Gathof BS. Evaluation of concurrent collection of in-line filtered platelets and packed red blood cells by multicomponent apheresis with three last-generation apparatuses. Vox Sang. 2006;91(1):47-55. 29.Hod EA, Brittenham GM, Billote GB, et al. Transfusion of human volunteers with older, stored red blood cells produces extravascular hemolysis and circulating non–transferrin-bound iron. Blood. 2011;118(25):6675-6682. 13.Cardigan R, Smith K. Evaluation of the HemoCue plasma haemoglobin analyser for assessing hemolysis in red cell concentrates during storage. Vox Sang. 2002;82(2):76-79. 14.Rogers SE, Edmondson D, Goodrick MJ, et al. Prestorage white cell reduction in saline-adenine-glucose-mannitol red cells by use of an integral filter: evaluation of storage values and in vivo recovery. Transfusion. 1995;35(9):727-733. 30.Dzik WH, Beckman N, Murphy MF, et al. Factors affecting red blood cell storage age at the time of transfusion. Transfusion. 2013;53(12):3110-3119. 15.Hess JR, Sparrow RL, van der Meer PF, Devine DV. Red blood cell hemolysis during blood bank storage: using national quality management data to answer basic scientific questions. Transfusion. 2009;49:2599-2603. 16.Gammon RR, Strayer SA, Avery NL, Mintz PD. Hemolysis during leukocyte-reduction filtration of stored red blood cells. Ann Clin Lab Sci. 2000;30(2):195-199. To read this article online, scan the QR code, http://labmed. ascpjournals.org/content/45/3/238. full.pdf+html QR code generated on http://qrcode.littleidiot.be www.labmedicine.com Summer 2014 | Volume 45, Number 3 Lab Medicine 243
© Copyright 2026 Paperzz