TRANSFUSION MEDICINE Original Article
Laboratory Hemostatic Abnormalities in Massively Transfused Patients Given Red Blood Cells and Crystalloid SUSAN D. LESLIE, M.D., AND PEARL T.C.Y. TOY, M.D.
develop significant thrombocytopenia after 20 units. Importantly, probably clinically significant prothrombin time and partial thromboplastin time prolongations occurred consistently after transfusion of 12 units of relatively plasma-free red blood cells in unselected patients at an urban trauma hospital. These data suggest that coagulation factor replacement is necessary in patients who receive 12 or more units of packed red blood cells or cell-saver blood, and platelet replacement is necessary in patients who receive 20 or more units of any red blood cell product. A prospective study is needed to determine whether the expected abnormal clinical bleeding indeed occurs in patients with such laboratory coagulation abnormalities and to determine when plasma transfusion is indicated in patients massively transfused with red blood cells. (Key words: Dilutional coagulopathy; Massive transfusion; Trauma; Cell saver) Am J Clin Pathol 1991;96: 770-773
Studies on patients massively transfused with predominantly whole blood (WB) 1-3 showed that pathologic hemorrhage is caused more frequently by thrombocytopenia than by depletion of coagulation factors. However, WB is not widely available, and most bleeding patients are given crystalloid and packed red blood cells (pRBC). In addition, intraoperative blood salvage is often used, during which washed, relatively plasma-free, cell-saver red blood cells (CS-RBC) are reinfused into the patient. One group reported platelet count and coagulation changes in seven elective surgery patients who were given 0.8 to 2.0 blood volumes of pRBCs and crystalloid.4 Because such studies are rare, and have not been done in patients who receive emergency transfusions, we retrospectively reviewed the records of massively transfused patients at our hospital,
an urban trauma center, to determine prothrombin time (PT) and activated partial thromboplastin time (PTT) results after transfusion of relatively plasma-free red blood cell products (pRBCs and CS-RBCs), and to determine platelet counts after transfusion of any RBC products (WB, pRBCs, and CS-RBCs). MATERIALS AND METHODS
Included were all 39 consecutive patients receiving massive transfusions at our hospital during the 6-month study period from June 1 to November 30, 1989. Massive transfusion was defined as 10 or more units of any RBC product transfused within 24 hours. Patients were identified by review of blood bank records. Red blood cell products included pRBCs, WB, and CS-RBCs. Packed red blood cells were stored in Adsol (Fenwal Laboratories, Deerfield, IL) preservative for as long as 42 days. Whole From the Blood Bank, San Francisco General Hospital and Medical blood was stored in citrate-phosphate-dextrose-adenine Center, and Department of Laboratory Medicine, University of California, anticoagulant-preservative (CPD A-1) (Fenwal) and was San Francisco, California. used 2 to 14 days after collection. The following inforReceived February 11, 1991; received revised manuscript and accepted mation was sought in each study patient. for publication June 27, 1991. Address reprint requests to Dr. Toy: 2M2, Blood Bank, San Francisco General Hospital, 1001 Potrero Avenue, San Francisco, California 94110. 1. Platelet count after transfusion of any RBCs (pRBCs, 770
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
Most of the literature on massive transfusion concerns wholeblood replacement, whereas clinically, packed red blood cells are commonly given. To determine when hemostatic abnormalities occur in patients resuscitated primarily with packed red blood cells and crystalloid, the cases of 39 consecutive patients who were transfused with 10 or more red blood cell units of any kind within 24 hours were reviewed. After transfusion with 20 or more units of red blood cell products of any kind (packed red blood cells, cell-saver units, or whole blood), 75% (3 of 4) of patients had platelet counts less than SO X 109/L, compared to 0 of 29 patients given less than 20 units (P < 0.001). After transfusion of 12 units of relatively plasma-free red blood cell products (packed red blood cells or cell-saver units), 100% (8 of 8) of patients had prothrombin time prolonged by more than 1.5 times mid-range of normal, compared to 36% (5 of 14) of patients given less than 12 units (P = 0.012). These data confirm that patients massively transfused with red blood cells of any kind
771
LESLIE AND TOY Massively Transfused Patients Given RBCs and Crystalloid WB, or CS-RBCs) before platelet transfusion. 2. Prothrombin time and PTT after transfusion of relatively plasma-free RBC units (pRBCs or CS-RBCs) before the transfusion of coagulation factors in the form of fresh frozen plasma or WB.
RESULTS Among the 39 patients, 31 (79%) were men and the mean age was 40 years (range, 19-83 years). The primary diagnoses were trauma (n = 25), gastrointestinal bleeding (n = 5), and other diagnoses (n = 9). In the 24-hour massive transfusion period, the average total number of RBC units transfused (pRBC + WB + CS-RBC) was 14 in the first 4 hours (range, 2-56), and 21 RBC units in 24 hours
Is
100
.
I\
NUMBER OF RED BLOOD CELL UNITS (packed cells, whole blood, cell saver units)
m-
•
605040-
•
•
•
30 -
•
•
20 • 10 -
•
t
i
•••
•
:•
10
20
30
NUMBER OF HED BLOOD CELL UNITS (packed red cells, cell saver units)
•>150
Do
100 -
50
1
0
1
1
1
10 20 NUMBER OF RED BLOOD CELL UNITS (packed red cells, cell saver units)
1
1
30
FIG. 1 {upper). Platelet count after transfusion with relatively plateletfree red blood cell units, stored WB, CS-RBC, and pRBC. Horizontal line indicates platelet count of 50 X 109/L. FIG. 2 (middle). Prothrombin time after transfusion of crystalloid and relatively plasma-free red blood cell units, pRBC, and/or CS-RBC. Horizontal line indicates 1.5 times mid-range of normal. Open circle indicates one patient with prothrombin time longer than 150 seconds. FlG. 3 (lower). Partial thromboplastin time after transfusion with crystalloid and relatively plasma-free red cell units, pRBC, and/or CS-RBC. Horizontal line indicates 1.5 time mid-range of normal.
(range, 10-83). The average number of components and WB used in 24 hours is shown in Table 1. Platelet Count Figure 1 shows the platelet count compared to the total number of WB, pRBC, and CS-RBC units transfused until
Vol. 96 • No. 6
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
Thirty-two patients were included in the analysis of platelet counts (Fig. 1). Excluded were seven patients: patients who received platelet transfusions before platelet counts were drawn (n = 3), patients with abnormal platelet counts before massive transfusion began (n = 2), a patient who received platelets with the initial RBC transfusion (n = 1), and a patient for whom time of issue of the platelets from the blood bank was unknown (n = 1). Twenty-two patients were included in the PT or PTT analyses (Figs. 2 and 3). Excluded were 17 patients: those who received WB, fresh frozen plasma, or cryoprecipitate with their initial RBC transfusions (n = 6), those with abnormal PT/PTT results before RBC transfusion (n = 1), those who received WB before PT/PTTs were drawn (n = 5), or those whose PT/PTTs were not drawn (n = 5). In the group of 22 patients who remained, 21 patients received no platelets before PT/PTTs were drawn. One patient received 6 units of platelets before PT/PTT was drawn. Platelet count was performed on the Technicon H-l (Tarrytown, NY) or the Coulter S+V instrument (Coulter, Hialeah, FL). Normal range was 150-400 X 109/L. Prothrombin time was performed in a one-stage test adding thromboplastin reagent containing tissue extract and CaCl2 to citrated plasma and timing until a clot was formed. Normal range during the study period was 10.5— 13.0 seconds. Activated PTT was performed by activating the intrinsic clotting factors in citrated plasma with micronized silica suspended with a standard concentration of platelet phospholipid, then adding CaCh and timing until a clot was formed. The normal range during the study period was 26.5-36.8 seconds. Because D-dimers are elevated in most surgical patients,5 the test is not ordered by surgeons at this hospital in such patients. The test for fibrin degradation products has been replaced by the D-dimer test at this hospital. Differences in proportions were compared using the chi-square test.
§ a- 200
81
772
TRANSFUSION MEDICINE Article
TABLE 1. BLOOD AND BLOOD COMPONENTS USED IN 24 HOURS IN STUDY PATIENTS Mean No. Median No. Range of Units of Units (No. of Units)
Blood Product pRBC + WB + CS-RBC
1st 4 hours 24 hours Packed red blood cells (pRBC) Whole blood (WB) Cell saver units (CS-RBC) Fresh frozen plasma Platelet concentrates Cryoprecipitate Emergency Group 0
16.7 21.0 13.0 3.0 4.5 5.1 4.7 3.2 1.7
12 15 11 3 0 4 0 0 0
2-64 10-83 5-39 0-22 0-28 0-22 0-30 0-40 0-16
DISCUSSION For massive transfusion, defined as transfusion of one blood volume or more within a 24-hour period, most of the literature describes experience with WB or modified WB1"3 but rarely experience with pRBC and fresh frozen plasma6 or pRBC and crystalloid.4 In current practice, a limited amount of WB is available and used, and resuscitation usually is attempted with pRBCs and crystalloid. In our study, mild thrombocytopenia (47 to 100 X 109/ L) occurred in all patients after transfusion of 15 RBC units of any kind and more severe thrombocytopenia (25 to 61 X 109/L) occurred after 20 RBC units. Significant prolongation of PT/PTT (more than 1.5 X MRN) occurred in all patients transfused with 12 or more relatively A.J.C.P. •
Our study shows that significant prolongations in PT and PTT consistently occur when CS-RBCs, pRBCs, and crystalloid are used to replace more than one blood volume loss in patients at an urban trauma hospital. Some patients transfused with less than one blood volume also had prolonged PT/PTT. Patients in our study may be more likely than elective surgery patients to develop coagulopathy because of preexisting coagulopathy due to liver disease, dilution due to large amounts of crystalloid resuscitation,9 or disseminated intravascular coagulation due to prolonged hypotension or brain injury. This is a retrospective study with the possible pitfalls of a retrospective study. Complete laboratory data were not available on all patients. Most importantly, clinical coagulopathy was not consistently documented in the medical records, which made it impossible for us to correlate the laboratory abnormalities with clinical findings. However, this retrospective study contributes to the sparse literature in this area and underscores the need for further prospective studies. Our data suggest that transfusion with 20 or more RBC units of any kind results in significant thrombocytopenia, similar to transfusions with WB. Transfusion with more than 12 units of pRBCs, CS-RBCs, and crystalloid was cember 1991
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
the time the sample was obtained for the platelet count, before administration of platelets. After transfusion of 20 or more units of RBC products, 75% (3 of 4) of patients had platelet counts less than 50 X 109/L, compared to 0 of 29 patients given less than 20 units (P < 0.001). Figures 2 and 3 show PT and PTT levels compared to the number of relatively plasma-free RBC units before administration of plasma (WB or fresh frozen plasma). Plasma-free RBC units included pRBCs and CS-RBCs. Figure 2 shows that even with transfusion of less than 12 units of plasma-free RBCs, about one third of patients (5 of 14) had PT ;> 1.5 X mid-range of normal (MRN). In two of the four cases,fibrinogenconcentrations were 0.37 g/L and 0.51 g/L. After transfusion of 12 units or more of plasma-free RBC products, 100% (8 of 8) of patients had PT prolonged by more than 1.5 times MRN compared to 36% (5 of 14) of patients given less than 12 units (P = 0.012). Four of these eight patients had PT ^ 2 X MRN. Figure 3 shows that with transfusion of less than 12 units of plasma-free RBCs, 6 of 14 patients had PTT ^1.5X MRN. The range was large, from 1 to 5 X MRN. These abnormalities could not be correlated with histories of hypotension, previous liver disease, or head trauma.
plasma-free RBC units. Even with transfusion of less than 12 plasma-free RBC units, approximately one third of patients had significant prolongations of PT/PTT. Previous studies attest to the safety of administering pRBCs and crystalloid in patients who bleed less than one blood volume during aortic7 and cardiac surgery.8 In addition, the study of Murray and co-workers4 included seven elective surgery patients transfused with 1.3-2.0 blood volumes of pRBCs and crystalloid. Three of the seven patients (transfused with 1.3, 1.5, and 1.5 blood volumes) did well. Four of the seven patients (transfused with 1.3, 1.75, 1.9, and 2.0 blood volume) had abnormal bleeding. Of the four, two whose platelets counts were 69 and 62 X 109/L responded to platelet transfusion. The other two patients had low platelet counts (38 and 83 X 109/L) and lowfibrinogenconcentrations (0.40 and 0.73 g/L) and responded to fresh frozen plasma in addition to platelets. Thus, the presumed causes of abnormal bleeding in their patients massively transfused with pRBCs and crystalloid were either thrombocytopenia and/or hypofibrinogenemia. Further dilution might be expected in patients given plasma-free RBC transfusions that were two or more times their blood volume. Because Murray and co-workers4 found that one blood volume pRBC transfusion without plasma reduced fibrinogen levels to 50%, two blood volumes may reduce levels to 25%. Because the normal range of fibrinogen is 200-400 mg/dL, 25% would be 50-100 mg/dL, which is less than or equal to the minimum level adequate for hemostasis.
LESLIE AND TOY Given RBCs and Crystalloid Massively Transfused Patiet consistently associated with PT/PTT greater than 1.5 X MRN, values that have been associated with abnormal microvascular bleeding in massively transfused patients.10 Prospective studies of trauma patients given crystalloid and pRBCs or CS-RBCs are desirable to confirm that these laboratory hemostatic abnormalities are indeed associated with abnormal clinical bleeding. Such studies will better define the indications for fresh frozen plasma in massive transfusion in the 1990s because pRBCs and CS-RBCs often are used instead of WB. Acknowledgment. The authors thank Ronald Miller, M.D., for his helpful suggestions in reviewing the manuscript.
3. Counts RB, Haisch C, Simon TL, et al. Hemostasis in massively transfused trauma patients. Ann Surg 1979;190:91-99. 4. Murray DJ, Olsomn J, Strauss R. Tinker JH. Coagulation changes during packed red cell replacement of major blood loss. Anesthesiology 1988;89:839-845. 5. Hunt PA, Rylatt DB, Hart R, Bundesen PG. Serum crosslinked fibrin (XDP) andfibrinogen/fibrindegradation products (FDP) in disorders associated with activation of the coagulation and fibrinolytic systems. Br J Haematol 1985;60:715-722. 6. Courcey PA, Brotman S, Dawson B. Massive blood transfusion in acute trauma (letter). Transfusion 1983,23:404. 7. Virgilio RW, Rice RL, Smith DE, et al. Crystalloid vs. colloid resuscitation: Is one better? A randomized clinical study. Surgery 1979;85:129-139. 8. Umlas J, Sakhuja R. The effect on blood coagulation of the exclusive use of transfusions of frozen red cells during and after cardiopulmonary bypass. J Thorac Cardiovascular Surg 1975;70:519— 523. 9. Hewson JR, Neame PB, Kumar N, et al. Coagulopathy related to dilution and hypotension during massive transfusion. Crit Care Med 1985;13:387-391. 10. Ciaverella D, Reed RL, Counts RB, et al. Clotting factor levels and the risk of diffuse microvascular bleeding in the massively transfused patient. Br J Haemat 1987;67:365-368.
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
REFERENCES 1. Krevans JR, DP Jackson. Hemorrhagic disorder following massive whole blood transfusion. JAMA 1955;159:171-177. 2. Miller RD, Robbins TO, Tong MI, Barton SL. Coagulation defects associated with massive blood transfusions. Ann Surg 1971; 1174: 794-801.
773
Laboratory Hemostatic Abnormalities in Massively Transfused Patients Given Red Blood Cells and Crystalloid SUSAN D. LESLIE, M.D., AND PEARL T.C.Y. TOY, M.D.
develop significant thrombocytopenia after 20 units. Importantly, probably clinically significant prothrombin time and partial thromboplastin time prolongations occurred consistently after transfusion of 12 units of relatively plasma-free red blood cells in unselected patients at an urban trauma hospital. These data suggest that coagulation factor replacement is necessary in patients who receive 12 or more units of packed red blood cells or cell-saver blood, and platelet replacement is necessary in patients who receive 20 or more units of any red blood cell product. A prospective study is needed to determine whether the expected abnormal clinical bleeding indeed occurs in patients with such laboratory coagulation abnormalities and to determine when plasma transfusion is indicated in patients massively transfused with red blood cells. (Key words: Dilutional coagulopathy; Massive transfusion; Trauma; Cell saver) Am J Clin Pathol 1991;96: 770-773
Studies on patients massively transfused with predominantly whole blood (WB) 1-3 showed that pathologic hemorrhage is caused more frequently by thrombocytopenia than by depletion of coagulation factors. However, WB is not widely available, and most bleeding patients are given crystalloid and packed red blood cells (pRBC). In addition, intraoperative blood salvage is often used, during which washed, relatively plasma-free, cell-saver red blood cells (CS-RBC) are reinfused into the patient. One group reported platelet count and coagulation changes in seven elective surgery patients who were given 0.8 to 2.0 blood volumes of pRBCs and crystalloid.4 Because such studies are rare, and have not been done in patients who receive emergency transfusions, we retrospectively reviewed the records of massively transfused patients at our hospital,
an urban trauma center, to determine prothrombin time (PT) and activated partial thromboplastin time (PTT) results after transfusion of relatively plasma-free red blood cell products (pRBCs and CS-RBCs), and to determine platelet counts after transfusion of any RBC products (WB, pRBCs, and CS-RBCs). MATERIALS AND METHODS
Included were all 39 consecutive patients receiving massive transfusions at our hospital during the 6-month study period from June 1 to November 30, 1989. Massive transfusion was defined as 10 or more units of any RBC product transfused within 24 hours. Patients were identified by review of blood bank records. Red blood cell products included pRBCs, WB, and CS-RBCs. Packed red blood cells were stored in Adsol (Fenwal Laboratories, Deerfield, IL) preservative for as long as 42 days. Whole From the Blood Bank, San Francisco General Hospital and Medical blood was stored in citrate-phosphate-dextrose-adenine Center, and Department of Laboratory Medicine, University of California, anticoagulant-preservative (CPD A-1) (Fenwal) and was San Francisco, California. used 2 to 14 days after collection. The following inforReceived February 11, 1991; received revised manuscript and accepted mation was sought in each study patient. for publication June 27, 1991. Address reprint requests to Dr. Toy: 2M2, Blood Bank, San Francisco General Hospital, 1001 Potrero Avenue, San Francisco, California 94110. 1. Platelet count after transfusion of any RBCs (pRBCs, 770
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
Most of the literature on massive transfusion concerns wholeblood replacement, whereas clinically, packed red blood cells are commonly given. To determine when hemostatic abnormalities occur in patients resuscitated primarily with packed red blood cells and crystalloid, the cases of 39 consecutive patients who were transfused with 10 or more red blood cell units of any kind within 24 hours were reviewed. After transfusion with 20 or more units of red blood cell products of any kind (packed red blood cells, cell-saver units, or whole blood), 75% (3 of 4) of patients had platelet counts less than SO X 109/L, compared to 0 of 29 patients given less than 20 units (P < 0.001). After transfusion of 12 units of relatively plasma-free red blood cell products (packed red blood cells or cell-saver units), 100% (8 of 8) of patients had prothrombin time prolonged by more than 1.5 times mid-range of normal, compared to 36% (5 of 14) of patients given less than 12 units (P = 0.012). These data confirm that patients massively transfused with red blood cells of any kind
771
LESLIE AND TOY Massively Transfused Patients Given RBCs and Crystalloid WB, or CS-RBCs) before platelet transfusion. 2. Prothrombin time and PTT after transfusion of relatively plasma-free RBC units (pRBCs or CS-RBCs) before the transfusion of coagulation factors in the form of fresh frozen plasma or WB.
RESULTS Among the 39 patients, 31 (79%) were men and the mean age was 40 years (range, 19-83 years). The primary diagnoses were trauma (n = 25), gastrointestinal bleeding (n = 5), and other diagnoses (n = 9). In the 24-hour massive transfusion period, the average total number of RBC units transfused (pRBC + WB + CS-RBC) was 14 in the first 4 hours (range, 2-56), and 21 RBC units in 24 hours
Is
100
.
I\
NUMBER OF RED BLOOD CELL UNITS (packed cells, whole blood, cell saver units)
m-
•
605040-
•
•
•
30 -
•
•
20 • 10 -
•
t
i
•••
•
:•
10
20
30
NUMBER OF HED BLOOD CELL UNITS (packed red cells, cell saver units)
•>150
Do
100 -
50
1
0
1
1
1
10 20 NUMBER OF RED BLOOD CELL UNITS (packed red cells, cell saver units)
1
1
30
FIG. 1 {upper). Platelet count after transfusion with relatively plateletfree red blood cell units, stored WB, CS-RBC, and pRBC. Horizontal line indicates platelet count of 50 X 109/L. FIG. 2 (middle). Prothrombin time after transfusion of crystalloid and relatively plasma-free red blood cell units, pRBC, and/or CS-RBC. Horizontal line indicates 1.5 times mid-range of normal. Open circle indicates one patient with prothrombin time longer than 150 seconds. FlG. 3 (lower). Partial thromboplastin time after transfusion with crystalloid and relatively plasma-free red cell units, pRBC, and/or CS-RBC. Horizontal line indicates 1.5 time mid-range of normal.
(range, 10-83). The average number of components and WB used in 24 hours is shown in Table 1. Platelet Count Figure 1 shows the platelet count compared to the total number of WB, pRBC, and CS-RBC units transfused until
Vol. 96 • No. 6
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
Thirty-two patients were included in the analysis of platelet counts (Fig. 1). Excluded were seven patients: patients who received platelet transfusions before platelet counts were drawn (n = 3), patients with abnormal platelet counts before massive transfusion began (n = 2), a patient who received platelets with the initial RBC transfusion (n = 1), and a patient for whom time of issue of the platelets from the blood bank was unknown (n = 1). Twenty-two patients were included in the PT or PTT analyses (Figs. 2 and 3). Excluded were 17 patients: those who received WB, fresh frozen plasma, or cryoprecipitate with their initial RBC transfusions (n = 6), those with abnormal PT/PTT results before RBC transfusion (n = 1), those who received WB before PT/PTTs were drawn (n = 5), or those whose PT/PTTs were not drawn (n = 5). In the group of 22 patients who remained, 21 patients received no platelets before PT/PTTs were drawn. One patient received 6 units of platelets before PT/PTT was drawn. Platelet count was performed on the Technicon H-l (Tarrytown, NY) or the Coulter S+V instrument (Coulter, Hialeah, FL). Normal range was 150-400 X 109/L. Prothrombin time was performed in a one-stage test adding thromboplastin reagent containing tissue extract and CaCl2 to citrated plasma and timing until a clot was formed. Normal range during the study period was 10.5— 13.0 seconds. Activated PTT was performed by activating the intrinsic clotting factors in citrated plasma with micronized silica suspended with a standard concentration of platelet phospholipid, then adding CaCh and timing until a clot was formed. The normal range during the study period was 26.5-36.8 seconds. Because D-dimers are elevated in most surgical patients,5 the test is not ordered by surgeons at this hospital in such patients. The test for fibrin degradation products has been replaced by the D-dimer test at this hospital. Differences in proportions were compared using the chi-square test.
§ a- 200
81
772
TRANSFUSION MEDICINE Article
TABLE 1. BLOOD AND BLOOD COMPONENTS USED IN 24 HOURS IN STUDY PATIENTS Mean No. Median No. Range of Units of Units (No. of Units)
Blood Product pRBC + WB + CS-RBC
1st 4 hours 24 hours Packed red blood cells (pRBC) Whole blood (WB) Cell saver units (CS-RBC) Fresh frozen plasma Platelet concentrates Cryoprecipitate Emergency Group 0
16.7 21.0 13.0 3.0 4.5 5.1 4.7 3.2 1.7
12 15 11 3 0 4 0 0 0
2-64 10-83 5-39 0-22 0-28 0-22 0-30 0-40 0-16
DISCUSSION For massive transfusion, defined as transfusion of one blood volume or more within a 24-hour period, most of the literature describes experience with WB or modified WB1"3 but rarely experience with pRBC and fresh frozen plasma6 or pRBC and crystalloid.4 In current practice, a limited amount of WB is available and used, and resuscitation usually is attempted with pRBCs and crystalloid. In our study, mild thrombocytopenia (47 to 100 X 109/ L) occurred in all patients after transfusion of 15 RBC units of any kind and more severe thrombocytopenia (25 to 61 X 109/L) occurred after 20 RBC units. Significant prolongation of PT/PTT (more than 1.5 X MRN) occurred in all patients transfused with 12 or more relatively A.J.C.P. •
Our study shows that significant prolongations in PT and PTT consistently occur when CS-RBCs, pRBCs, and crystalloid are used to replace more than one blood volume loss in patients at an urban trauma hospital. Some patients transfused with less than one blood volume also had prolonged PT/PTT. Patients in our study may be more likely than elective surgery patients to develop coagulopathy because of preexisting coagulopathy due to liver disease, dilution due to large amounts of crystalloid resuscitation,9 or disseminated intravascular coagulation due to prolonged hypotension or brain injury. This is a retrospective study with the possible pitfalls of a retrospective study. Complete laboratory data were not available on all patients. Most importantly, clinical coagulopathy was not consistently documented in the medical records, which made it impossible for us to correlate the laboratory abnormalities with clinical findings. However, this retrospective study contributes to the sparse literature in this area and underscores the need for further prospective studies. Our data suggest that transfusion with 20 or more RBC units of any kind results in significant thrombocytopenia, similar to transfusions with WB. Transfusion with more than 12 units of pRBCs, CS-RBCs, and crystalloid was cember 1991
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
the time the sample was obtained for the platelet count, before administration of platelets. After transfusion of 20 or more units of RBC products, 75% (3 of 4) of patients had platelet counts less than 50 X 109/L, compared to 0 of 29 patients given less than 20 units (P < 0.001). Figures 2 and 3 show PT and PTT levels compared to the number of relatively plasma-free RBC units before administration of plasma (WB or fresh frozen plasma). Plasma-free RBC units included pRBCs and CS-RBCs. Figure 2 shows that even with transfusion of less than 12 units of plasma-free RBCs, about one third of patients (5 of 14) had PT ;> 1.5 X mid-range of normal (MRN). In two of the four cases,fibrinogenconcentrations were 0.37 g/L and 0.51 g/L. After transfusion of 12 units or more of plasma-free RBC products, 100% (8 of 8) of patients had PT prolonged by more than 1.5 times MRN compared to 36% (5 of 14) of patients given less than 12 units (P = 0.012). Four of these eight patients had PT ^ 2 X MRN. Figure 3 shows that with transfusion of less than 12 units of plasma-free RBCs, 6 of 14 patients had PTT ^1.5X MRN. The range was large, from 1 to 5 X MRN. These abnormalities could not be correlated with histories of hypotension, previous liver disease, or head trauma.
plasma-free RBC units. Even with transfusion of less than 12 plasma-free RBC units, approximately one third of patients had significant prolongations of PT/PTT. Previous studies attest to the safety of administering pRBCs and crystalloid in patients who bleed less than one blood volume during aortic7 and cardiac surgery.8 In addition, the study of Murray and co-workers4 included seven elective surgery patients transfused with 1.3-2.0 blood volumes of pRBCs and crystalloid. Three of the seven patients (transfused with 1.3, 1.5, and 1.5 blood volumes) did well. Four of the seven patients (transfused with 1.3, 1.75, 1.9, and 2.0 blood volume) had abnormal bleeding. Of the four, two whose platelets counts were 69 and 62 X 109/L responded to platelet transfusion. The other two patients had low platelet counts (38 and 83 X 109/L) and lowfibrinogenconcentrations (0.40 and 0.73 g/L) and responded to fresh frozen plasma in addition to platelets. Thus, the presumed causes of abnormal bleeding in their patients massively transfused with pRBCs and crystalloid were either thrombocytopenia and/or hypofibrinogenemia. Further dilution might be expected in patients given plasma-free RBC transfusions that were two or more times their blood volume. Because Murray and co-workers4 found that one blood volume pRBC transfusion without plasma reduced fibrinogen levels to 50%, two blood volumes may reduce levels to 25%. Because the normal range of fibrinogen is 200-400 mg/dL, 25% would be 50-100 mg/dL, which is less than or equal to the minimum level adequate for hemostasis.
LESLIE AND TOY Given RBCs and Crystalloid Massively Transfused Patiet consistently associated with PT/PTT greater than 1.5 X MRN, values that have been associated with abnormal microvascular bleeding in massively transfused patients.10 Prospective studies of trauma patients given crystalloid and pRBCs or CS-RBCs are desirable to confirm that these laboratory hemostatic abnormalities are indeed associated with abnormal clinical bleeding. Such studies will better define the indications for fresh frozen plasma in massive transfusion in the 1990s because pRBCs and CS-RBCs often are used instead of WB. Acknowledgment. The authors thank Ronald Miller, M.D., for his helpful suggestions in reviewing the manuscript.
3. Counts RB, Haisch C, Simon TL, et al. Hemostasis in massively transfused trauma patients. Ann Surg 1979;190:91-99. 4. Murray DJ, Olsomn J, Strauss R. Tinker JH. Coagulation changes during packed red cell replacement of major blood loss. Anesthesiology 1988;89:839-845. 5. Hunt PA, Rylatt DB, Hart R, Bundesen PG. Serum crosslinked fibrin (XDP) andfibrinogen/fibrindegradation products (FDP) in disorders associated with activation of the coagulation and fibrinolytic systems. Br J Haematol 1985;60:715-722. 6. Courcey PA, Brotman S, Dawson B. Massive blood transfusion in acute trauma (letter). Transfusion 1983,23:404. 7. Virgilio RW, Rice RL, Smith DE, et al. Crystalloid vs. colloid resuscitation: Is one better? A randomized clinical study. Surgery 1979;85:129-139. 8. Umlas J, Sakhuja R. The effect on blood coagulation of the exclusive use of transfusions of frozen red cells during and after cardiopulmonary bypass. J Thorac Cardiovascular Surg 1975;70:519— 523. 9. Hewson JR, Neame PB, Kumar N, et al. Coagulopathy related to dilution and hypotension during massive transfusion. Crit Care Med 1985;13:387-391. 10. Ciaverella D, Reed RL, Counts RB, et al. Clotting factor levels and the risk of diffuse microvascular bleeding in the massively transfused patient. Br J Haemat 1987;67:365-368.
Downloaded from http://ajcp.oxfordjournals.org/ by guest on June 6, 2016
REFERENCES 1. Krevans JR, DP Jackson. Hemorrhagic disorder following massive whole blood transfusion. JAMA 1955;159:171-177. 2. Miller RD, Robbins TO, Tong MI, Barton SL. Coagulation defects associated with massive blood transfusions. Ann Surg 1971; 1174: 794-801.
773
