Transfusion Medicine, 1992, 2, 189- I 94
ORIGINAL ARTICLES
Analysis of fresh frozen plasma administration with suggestions for ways to reduce usage J. N. Shanberge and T. Quattrociocchi-Longe Deportment of Clinical Pathology, Wiilium Beaumont Hospiral, Royal Oak, Michigan, U.S.A. Received 30 July 1991; accepted for publication 2 January I992
A programme for the daily monitoring of Fresh Frozen Plasma (FFP) usage, combined with continuous eduction in the correct use of FFP, was started at William Beaumont Hospital in 1985. In 2 years, this had resulted in a 77% reduction in FFP usage. An analysis of the type of cases which received FFP, after the major reduction had occurred, from July, 1985through June, 1989is presented. During this time 2,612 units were administered to 873 patients, an average of 54 units per month. According to the accepted criteria established by the Hospital Transfusion Committee, 67% of the transfused units on the
medical service were deemed indicated, compared with 54% on the surgical service. Most of the FFP was used to treat patients with liver disease, or receiving coumadin, or undergoing coronary bypass surgery. Conditions which will decrease the need for FFP administration are also presented for consideration. The results indicate that a consistent monitoring and education programme can keep blood and blood component usage at a defensible minimum.
Shortly after the NIH convened a Consensus Development Panel in September, 1984,to devise guidelines for the use of Fresh Frozen Plasma (FFP) (NIH Consensus Conference 1985), a concurrent monitoring and education programme was started at William Beaumont Hospital; a 930 bed, tertiary care, general hospital (Shanberge, 1987). At no time did the Blood Bank refuse to provide any order for FFP. In less than 2 years, this resulted in a 77% decrease in FFP administration. Since then, the use of FFP has averaged only 55 units per month (Fig. 1). An analysis of FFP usage at Beaumont, from July 1985 through to June, 1989 is presented as are the conditions considered necessary for a decrease in the need for FFP administration.
the patient’s chart and determines whether usage conformed with the Transfusion Committee’s Guidelines. If not, the ordering physician is contacted, and the case discussed with him. These ‘plasma rounds’ are used as an educational process for both the laboratory residents as well as the clinical staff and are supplemented by seminars for the various Services on the correct use of blood components.
SUMMARY.
Keywords: blood component monitoring, fresh frozen plasma, FFP usage.
RESULTS This survey covers all of the FFP usage for 4 years, July 1985 to June 1989. It represents 2,612 units given to 873 in-patients: a monthly average of 18-2patients and 54.4 units. 1,548 units or 59% of the units given were deemed indicated according to the criteria listed in Table 1. Table 2 gives a breakdown of FFP usage according to clinical services. The patients who received FFP and the number of FFP units administered were almost equally divided between the Medical and Surgical Services. However, 67% of the units given on the Medical Service were indicated, compared with only 52% on the Surgical Service. During the 4 years of this study, the cases in which the use of FFP did not meet the accepted guidelines remained relatively constant (mean 47%) as did the number of units (mean 4 I I%).
METHODS The programme for the daily monitoring of FFP usage and education in the correct use of FFP was carried out as described previously (Shanberge, 1987). Briefly, each morning the Transfusion Service provides the Chief of the Hemostasis and Coagulation Laboratory (JNS) with a list of patients who have received FFP during the previous day or weekend. He then reviews Correspondence: Dr J. N. Shanberge, William Beaumont Hospital. 3601 West 13 Mile Road, Royal Oak, MI 48073, U S A .
189
I90
J . N . Shanberge and T. Quattrociocchi-Longe 25C
20c
-” .-a
15C
c C J
I&
I oc
5c
C 1984
1985
1986
1987
I988
I989
I990
Time ( y e a r )
Fig. 1. FFP usage at William Beaumont Hospital, 1984 until the end of 1990. (Does not include FFP used in apheresis.)
Table 3 shows FFP usage according to diagnosis. The majority of patients h a d liver disease or underwent cardiovascular surgery or were on coumadin. During the time of the survey, 2,609 patients underwent
coronary bypass procedures (CABG), of whom only 118 (4.5%) were given FFP; 593 patients had cardiac valve replacement, of whom only 27 (4.6%) received FFP.
Table 1. Criteria for the use of fresh frozen plasma
A.
History or clinical course suggestive of a coagulopathy, due to a deficiency of coagulation factors, with bleeding, or prior to an invasive procedure. Must be documented by at least one of the following: Prothrombin time (PT) > 18 s. Activated partial thromboplastin time (PTT) > 55 s. Coagulation factor assay of 18 s) Congenital coagulation factor deficiency when used for bleeding, or prophylactically for surgery or an invasive procedure. Deficiency of factor 11, V, VII, IX,X,XI or XI11 von Willebrand’s disease (if Cryo is not available). Deficiency of factor VIII (if Cry0 or factor VIII concentrate is not available)
E. F. G.
Deficiency of antithrombin 111, protein C or protein S Immunodeficiency syndromes Thrombotic thrombocytopenic purpura; Haemolytic uraemic syndrome.
Fresh frozen plasma usage Table 2. FFP usage July I985 to June 1989:indicated (Y) and not indicated (N)
Service
-
Medical Patients 426 (48.8%)
Y
N
255 171
60% 40%
FFP units 1,269 Y 845 67% N 424 33% Surgical Patients 404 (46.3%) Y 185 46% N 219 54% FFP units 1,242 Y 647 52% N 595 48% Obstetric Patients 16 (1.8%) Y 7 44% N 9 56% FFP units 59 Y 28 47% N 31 53% Paediatric Patients 27 (3.1%) Y 18 67% N 9 33% FFP units 42 Y 28 67%
N
14
33%
DISCUSSION Much of the inordinate increase in FFP usage which started, both nationally and at Beaumont, in 1972and reached its peak in the early 1980s could be attributed to the shift to component preparation and a decreasing availability of whole blood. Many physicians felt compelled to ‘reconstitute’ the red blood cell units with FFP in order to replace those components which had been removed. Another reason for the increase was the advent of coronary bypass surgery. Originally, every patient at Beaumont was given routinely 2 units of FFP at the end of the procedure whether there was any
191
excess bleeding or not. This was abandoned when it was pointed out that not only was it not indicated but that excess plasma increases antithrombin I11 availability which may produce a heparin rebound (Shanberge et al., 1987). Recently, Martinowitz et al. (1990) also questioned the need for FFP after cardiopulmonary bypass procedures. Other abuses of F F P at Beaumont were similar to those reviewed by Oberman (1985). The criteria for the use of FFP, established by the Transfusion Committee in 1985 (Table l), follow the guidelines published by the NIH Consensus Conference (1 985). The haemostatic levels of a prothrombin time (PT) < 18 s and an activated partial thromboplastin time (PTT) < 5 5 s are simiar to those published previously (Blumberg et al., 1986; Coffin, 1987; Mozes er al., 1989; Renner et al., 1987; Silberstein et al., 1989; Willis, 1989). Although our educational programme has resulted in an overall decrease in F F P usage of 75%, the use of FFP still does not meet the accepted guidelines in over 40% of cases. Part of this is due to the constant turnover of house staff together with the new attending staff members who have come from other institutions where the accepted practice may have been different. Previously published reviews of the use of FFP present far fewer cases than are reported here (Silbert et al., 1981; Shaikh et al., 1985; Blumberg et al., 1986; Snyder et al., 1986; Solomon el al., 1988; Brien et al., 1989; Mozes et al., 1989; Silberstein et al., 1989). However, reviews which have delineated FFP use by Clinical Service show that the vast majority was given by the Surgical Service (Silbert et al., 1981; Shaikh el al., 1985; Blumberg et at., 1986; Brien et al., 1989). At Beaumont, the use by Surgery and Medicine was almost the same with a slightly greater use by Medicine (Table 2). Moreover, the acceptable use of F F P was greater for medical than surgical cases (60 us 46%), and this was also true for the number of units used (67 us 52%). The reason for this is that F F P is given more often empirically during various surgical procedures, because the turn-around-time to obtain results of coagulation tests is too long for them to be used to guide therapy. However, in the presence of excessive surgical bleeding, FFP is usually not indicated unless blood loss exceeds 5,000 ml (NIH Consensus Conference, 1985). In addition, in contrast to the previously published reports, in which large amounts of FFP were given to maintain blood volume (Silbert et al., 1981; Shaikh et al., 1985; Blumberg, 1986; Snyder et al., 1986), at Beaumont FFP was given to increase plasma volume in only three cases over the 4 years, two of which were on the new-born intensive care service. In our survey, most of the FFP used was for patients
192
J . N . Shanberge and T. Quattrociocchi-Longe Table 3. FFP usage according to
Liver disease Pre-procedure Coumadin Pre-procedure CABG Aortic aneurysm Valve replacement Massive bleed ICU syndrome Miscellaneous*
FFP units
% of total
Patients
% of total
429 55 322 159 365 216 105 74 66 588
16.4 2.1 12.3 6.1 14-0 8.3 4.0 2.8 2.5 22.5
97 29 142 67 118 61 21 18 34 215
11.1 3.3 16.3 7.7 13.5 7.0 3.1 2.1 3.9 24.6
diagnosis
* Includes 63 general surgery, 28 G1bleed, 28 cancer-related cases. 14haematological disease,4 thrombotic thrornbocytopenic purpura, 4 coagulation defect (3 factor XI deficiency, I factor IX deficiency)and 3 volume expansion.
Table 4. Points to reduce FFP usage 1.
2. 3. 4.
5. 6. 7. 8. 9.
Avoid empiric use of FFP. Formulas for use of FEP/packed red cells should not be used Give only enough FFP needed for haemostasis. Do not treat laboratory values FFP is not needed for minor procedures Prolonged PT or PTT values may be due to a low fibrinogen level or the presence of heparin Obtain PT and/or PTT shortly after FFP administration to determine adequacy of dosage Five units of platelet concentrate have a plasma equivalence of 1 unit of FFP Cryoprecipitate is more efficient than FFP to raise fibrinogen levels Prevent iatrogenic vitamin K deficiency (ICU syndrome) Always consider the possibilities of disease transmission with each unit
administered
with liver disease, on coumadin therapy, or during and after cardiovascular surgery (Table 3). There were, among the miscellaneous group, several instances when the FFP was given for unusual reasons: in two cases with hepatorenal syndrome to promote glomerular filtration, in one case of chronic thrombocytopenia to simulate platelet production, and in one case of septicaemia to increase the opsonic index. The number of units of F F P given per month at Beaumont for the 4 years averaged 54, in contrast to an average of 214 units per month in 1984. This resulted in considerable financial savings for the Hospital as well as a decrease in the possible unwanted complications of transfusion (Walker, 1987). If the use of FFP were to follow the criteria exactly, this average would drop to around 32 units per month. How can the use of FFP be decreased further, other than by instituting a prospective regulatory pro-
gramme? The important points are summarized in Table 4. First, even when F F P is indicated, only that amount should be given which is necessary to bring about haemostasis. It must be emphasized that a 16- or 17-s PT, although not ‘normal’ still represents a coagulation activity well within the haemostatic range (White et al., 1987). A clotting activity of at least 30%, with rare exceptions, is adequate even for most surgical procedures (Salzman, 1987). Therefore, it is not necessary, in the majority of cases, to reduce the PT or P?T to ‘normal’ levels. This is particularly true with minor surgical or invasive procedures. In a retrospective study of 608 patients undergoing paracentesis or thoracentesis, it was found that no significant bleeding occurred in the presence of prolonged PT, even as long as 24 s (McVay & Toy, 1991). Secondly, the empirical use of formulas, whereby a certain number of units of F F P are given for so many
Fresh frozen plasma usage 193 units of red cells, should be discontinued. Even with a large loss of plasma volume, the clotting activity may not fall below the haemostatic level (Counts et al., 1979; Mannucci et al., 1982; Reed ef al., 1986; Ciavarella et al., 1987; Phillips et al., 1987; Murray et al., 1988). Thirdly, it is often not realized that the prolonged PT or PTT value, on which FFP administration is based, may be erroneous due to a decreased fibrinogen level, as with massive haemorrhage or disseminated intravascular coagulation. In such instances, cryoprecipitate (cryo), a better source of fibrinogen, volumewise, is indicated for the initial therapy rather than FFP. A prolonged PTT may also be due to contamination of the blood sample by heparin which must first be removed in order to obtain an accurate value. Fourthly, in order to monitor the effectiveness of components such as FFP, the PT and/or PTT should be ordered soon after its administration. Because of the short half-life of factor VII (Salzman, 1987), a factor which is less essential for haemostasis, a delay in ordering a PT may lead to a lengthening of the PT giving a false impression of the inadequacy of dosage. Moreover, it is equally important to know whether the calculated number of units of FFP given is sufficient to reach the desired level of activity. Fifthly, when a patient needs both platelets and coagulation factors, for every 5 units of platelets administered, the patient wilI have received the equivalent of one unit of FFP, so that additional FFP may not be needed. Sixthly, prevention of a vitamin K deficiency, as occurs in severely ill patients who are given broadspectrum antibiotics and only intravenous fluids, the so-called ICU syndrome (Pineo et al., 1973), will prevent a few cases of unexpected bleeding requiring FFP. Finally, when there is a significantly large hole in a blood vessel, no amount of FFP or, for that matter, any blood component, is going to substitute for ligation, suturing, cautery, or sclerosis. The success of our programme has been based on education and consultation rather than regulation. More recently, it has been applied to the use of other blood components with an appreciable reduction in the number of units used. It is now proposed to apply a similar programme to red cell transfusion. REFERENCES Blumberg, N., Laczin, J., McMican, A., Heal, J. & Arvan. D. (1986) A critical survey of fresh-frozen plasma use. Transfusion, 26, 51 1-513. Brien, W.F., Butler, R.J. & Inwood, M.J. (1989). An audit of
blood component therapy in a Canadian general teaching hospital. Canadian Medical Association Journal, 140,8 12815.
Ciavarella, D., Reed, R.L., Counts, R.B., Baron, L., Pavlin, E., Heimbach, D.M. & Carrico, C.J. (1987) Clotting factor levels and the risk of diffuse microvascular bleeding in the massively transfused patient. British Journal of Haematology, 67,365-368. Coffin, C.M. (1987) Current issues in transfusion therapy. 2. Indications for use of blood components. Postgraduate Medicine, 81, 343-350. Counts, R.B., Haisch, C., Simon, T.L., Maxwell, N.G., Heimbach, D.M.& Carrico, C.J. (1979) Hemostasis in massively transfused trauma patients. Annals of Surgery, 190,91-99.
McVay, P.A. & Toy, P.T.C.Y. (1991) Lack of increased bleeding after paracentesis and thoracentesis in patients with mild coagulation abnormalities. Transfusion, 21, 164-171.
Manucci, P.M., Federici, A.B. & Sirchia, G. (1982) Hemostasis testing during massive blood replacement. A study of 172 cases. Vox Sanguinis, 42, 1 13- 123. Martinowitz, U., Goor, D.A., Ramot, B. & Mohr, R. (1990) Is transfusion of fresh plasma after cardiac operations indicated? Journal of Thoracic and Cardiovascular Surgery, 100,92-98. Mozes, B., Epstein, M., Ben-Bassat I., Modan, €3. & Halkin, H. (1989) Evaluation of the appropriateness of blood and blood product transfusion using preset criteria. Transfusion, 29,473-76. Murray, D.J., Olson, J., Strauss, R. & Tinker, J.H. (1988) Coagulation changes during packed red cell replacement of major blood loss. Anesthesiology, 69, 839-45. NIH Consensus Conference (1 985) Fresh-frozen plasma. Indications and risks. Journal of the American Medical Association, 253, 551-553. Oberman, H.A. (1985) Uses and abuses of fresh frozen plasma. In: Current Concepts in Transfusion Therapy (ed. Garratty, A.), 109-124. American Association of Blood Banks, Arlington, VA. Phillips, T.F.,Soulier, G. &Wilson, R.F. (1987) Outcome of massive transfusion exceeding two blood volumes in trauma and emergency surgery. Journal of Trauma, 27, 903-910.
Pineo, G.F., Brain, M.C. & Hirsh, J. (1973) Unexpected vitamin K deficiency in hospitalized patients. Canadian Medical Association Journal, 109, 880-883. Reed, R.L. 11, Ciavarella, D., Heimbach, D.M., Baron, L., Pavlin, E., Counts, R.B. & Carrico, C.J. (1986) Prophylactic platelet administration during massive transfusion. A prospective, randomized, double-blind clinical study. Annals of Surgery, 203, 40-48. Renner, S.W., Howanitz, J.H. & Fishkin, B.G. (1987) Toward meaningful blood usage review: Comprehensive monitoring of physician practice. Quality Review Bulletin. 13,76-80.
194
J . N . Shanberge and T. Quattrociocchi-Longe
Salzman, E.W. (1987). Hemostatic problems in surgical patients. In: Hemostasis and Thrombosis. Basic Principles and Clinical Practice (eds Colrnan, R.W., Hirsh, J., Marder, V.J. and Salzman, E.W.), 920-925. J.B. Lippincott, Philadelphia, PA. Shaikh, B.S., Wagar, D., Lau, P.M. & Campbell, E.W. Jr. (1985) Transfusion pattern of fresh frozen plasma in a medical school hospital. Vox Sanguinis, 48, 366-369. Shanberge, J.N. (1987) Reduction of fresh-frozen plasma use through a daily survey and education program. Transfusion, 27, 226-227. Shanberge, J.N., Murato, M., Quattrociocchi-Longe, T. & Van Neste, L.(1987) Heparin-protarnine compkxes in the production of heparin rebound and other complications of extracorporeal bypass procedures. American Journal of Clinical Pathology, 87,210-217.
Silberstein, L.E., Kruskall, M.S.,Stehling, L.C., Johnston, M.F.M., Rutman, R.C., Samja, C.T., Ramsey, G. & Eisenstaedt, S. (1989) Strategies for the review of transfusion practices. Journal of the American Medical Association, 262, 1993- 1997.
Silbert, J.A., Bove, J.R., Dubin, S. & Bush, W.S. (1981) Patterns of frozen plasma use. Connecticut Medicine, 45, 507-5 1 1.
Solomon, R.R., Clifford, J.A. & Gutman, S.I. (1988) The use of laboratory intervention to stem the flow of fresh-frozen plasma. American Journal of Clinical Pathology, 89,51821.
Snyder, A.J., Gottschall, J.L. & Menitove, J.E. (1986) Why is fresh-frozen plasma transfused? Transfusion,26, 107- 1 12. Walker, R.H. (1987) Special report: Transfusion risks. American Journal of Clinical Pathology, 88,374-378. White, G.C. 11. Marder, V.J., Colman, R.W., Hirsh, J. & SaIzman, E.W.(1987) Approach to the bleeding patient. In: Hemostasis and Thrombosis. Basic Principles and Clinical Practice (eds Colman, R.W., Hirsh, J., Marder, V.J. & .Salzman, E.W.), 1048-1060. J.B. Lippincott, Philadelphia, PA. Willis, J.L. (4. (1989) ) Use of blood components. FDA Drug Bulletin, Public Health Service, Rock, MD.
ORIGINAL ARTICLES
Analysis of fresh frozen plasma administration with suggestions for ways to reduce usage J. N. Shanberge and T. Quattrociocchi-Longe Deportment of Clinical Pathology, Wiilium Beaumont Hospiral, Royal Oak, Michigan, U.S.A. Received 30 July 1991; accepted for publication 2 January I992
A programme for the daily monitoring of Fresh Frozen Plasma (FFP) usage, combined with continuous eduction in the correct use of FFP, was started at William Beaumont Hospital in 1985. In 2 years, this had resulted in a 77% reduction in FFP usage. An analysis of the type of cases which received FFP, after the major reduction had occurred, from July, 1985through June, 1989is presented. During this time 2,612 units were administered to 873 patients, an average of 54 units per month. According to the accepted criteria established by the Hospital Transfusion Committee, 67% of the transfused units on the
medical service were deemed indicated, compared with 54% on the surgical service. Most of the FFP was used to treat patients with liver disease, or receiving coumadin, or undergoing coronary bypass surgery. Conditions which will decrease the need for FFP administration are also presented for consideration. The results indicate that a consistent monitoring and education programme can keep blood and blood component usage at a defensible minimum.
Shortly after the NIH convened a Consensus Development Panel in September, 1984,to devise guidelines for the use of Fresh Frozen Plasma (FFP) (NIH Consensus Conference 1985), a concurrent monitoring and education programme was started at William Beaumont Hospital; a 930 bed, tertiary care, general hospital (Shanberge, 1987). At no time did the Blood Bank refuse to provide any order for FFP. In less than 2 years, this resulted in a 77% decrease in FFP administration. Since then, the use of FFP has averaged only 55 units per month (Fig. 1). An analysis of FFP usage at Beaumont, from July 1985 through to June, 1989 is presented as are the conditions considered necessary for a decrease in the need for FFP administration.
the patient’s chart and determines whether usage conformed with the Transfusion Committee’s Guidelines. If not, the ordering physician is contacted, and the case discussed with him. These ‘plasma rounds’ are used as an educational process for both the laboratory residents as well as the clinical staff and are supplemented by seminars for the various Services on the correct use of blood components.
SUMMARY.
Keywords: blood component monitoring, fresh frozen plasma, FFP usage.
RESULTS This survey covers all of the FFP usage for 4 years, July 1985 to June 1989. It represents 2,612 units given to 873 in-patients: a monthly average of 18-2patients and 54.4 units. 1,548 units or 59% of the units given were deemed indicated according to the criteria listed in Table 1. Table 2 gives a breakdown of FFP usage according to clinical services. The patients who received FFP and the number of FFP units administered were almost equally divided between the Medical and Surgical Services. However, 67% of the units given on the Medical Service were indicated, compared with only 52% on the Surgical Service. During the 4 years of this study, the cases in which the use of FFP did not meet the accepted guidelines remained relatively constant (mean 47%) as did the number of units (mean 4 I I%).
METHODS The programme for the daily monitoring of FFP usage and education in the correct use of FFP was carried out as described previously (Shanberge, 1987). Briefly, each morning the Transfusion Service provides the Chief of the Hemostasis and Coagulation Laboratory (JNS) with a list of patients who have received FFP during the previous day or weekend. He then reviews Correspondence: Dr J. N. Shanberge, William Beaumont Hospital. 3601 West 13 Mile Road, Royal Oak, MI 48073, U S A .
189
I90
J . N . Shanberge and T. Quattrociocchi-Longe 25C
20c
-” .-a
15C
c C J
I&
I oc
5c
C 1984
1985
1986
1987
I988
I989
I990
Time ( y e a r )
Fig. 1. FFP usage at William Beaumont Hospital, 1984 until the end of 1990. (Does not include FFP used in apheresis.)
Table 3 shows FFP usage according to diagnosis. The majority of patients h a d liver disease or underwent cardiovascular surgery or were on coumadin. During the time of the survey, 2,609 patients underwent
coronary bypass procedures (CABG), of whom only 118 (4.5%) were given FFP; 593 patients had cardiac valve replacement, of whom only 27 (4.6%) received FFP.
Table 1. Criteria for the use of fresh frozen plasma
A.
History or clinical course suggestive of a coagulopathy, due to a deficiency of coagulation factors, with bleeding, or prior to an invasive procedure. Must be documented by at least one of the following: Prothrombin time (PT) > 18 s. Activated partial thromboplastin time (PTT) > 55 s. Coagulation factor assay of 18 s) Congenital coagulation factor deficiency when used for bleeding, or prophylactically for surgery or an invasive procedure. Deficiency of factor 11, V, VII, IX,X,XI or XI11 von Willebrand’s disease (if Cryo is not available). Deficiency of factor VIII (if Cry0 or factor VIII concentrate is not available)
E. F. G.
Deficiency of antithrombin 111, protein C or protein S Immunodeficiency syndromes Thrombotic thrombocytopenic purpura; Haemolytic uraemic syndrome.
Fresh frozen plasma usage Table 2. FFP usage July I985 to June 1989:indicated (Y) and not indicated (N)
Service
-
Medical Patients 426 (48.8%)
Y
N
255 171
60% 40%
FFP units 1,269 Y 845 67% N 424 33% Surgical Patients 404 (46.3%) Y 185 46% N 219 54% FFP units 1,242 Y 647 52% N 595 48% Obstetric Patients 16 (1.8%) Y 7 44% N 9 56% FFP units 59 Y 28 47% N 31 53% Paediatric Patients 27 (3.1%) Y 18 67% N 9 33% FFP units 42 Y 28 67%
N
14
33%
DISCUSSION Much of the inordinate increase in FFP usage which started, both nationally and at Beaumont, in 1972and reached its peak in the early 1980s could be attributed to the shift to component preparation and a decreasing availability of whole blood. Many physicians felt compelled to ‘reconstitute’ the red blood cell units with FFP in order to replace those components which had been removed. Another reason for the increase was the advent of coronary bypass surgery. Originally, every patient at Beaumont was given routinely 2 units of FFP at the end of the procedure whether there was any
191
excess bleeding or not. This was abandoned when it was pointed out that not only was it not indicated but that excess plasma increases antithrombin I11 availability which may produce a heparin rebound (Shanberge et al., 1987). Recently, Martinowitz et al. (1990) also questioned the need for FFP after cardiopulmonary bypass procedures. Other abuses of F F P at Beaumont were similar to those reviewed by Oberman (1985). The criteria for the use of FFP, established by the Transfusion Committee in 1985 (Table l), follow the guidelines published by the NIH Consensus Conference (1 985). The haemostatic levels of a prothrombin time (PT) < 18 s and an activated partial thromboplastin time (PTT) < 5 5 s are simiar to those published previously (Blumberg et al., 1986; Coffin, 1987; Mozes er al., 1989; Renner et al., 1987; Silberstein et al., 1989; Willis, 1989). Although our educational programme has resulted in an overall decrease in F F P usage of 75%, the use of FFP still does not meet the accepted guidelines in over 40% of cases. Part of this is due to the constant turnover of house staff together with the new attending staff members who have come from other institutions where the accepted practice may have been different. Previously published reviews of the use of FFP present far fewer cases than are reported here (Silbert et al., 1981; Shaikh et al., 1985; Blumberg et al., 1986; Snyder et al., 1986; Solomon el al., 1988; Brien et al., 1989; Mozes et al., 1989; Silberstein et al., 1989). However, reviews which have delineated FFP use by Clinical Service show that the vast majority was given by the Surgical Service (Silbert et al., 1981; Shaikh el al., 1985; Blumberg et at., 1986; Brien et al., 1989). At Beaumont, the use by Surgery and Medicine was almost the same with a slightly greater use by Medicine (Table 2). Moreover, the acceptable use of F F P was greater for medical than surgical cases (60 us 46%), and this was also true for the number of units used (67 us 52%). The reason for this is that F F P is given more often empirically during various surgical procedures, because the turn-around-time to obtain results of coagulation tests is too long for them to be used to guide therapy. However, in the presence of excessive surgical bleeding, FFP is usually not indicated unless blood loss exceeds 5,000 ml (NIH Consensus Conference, 1985). In addition, in contrast to the previously published reports, in which large amounts of FFP were given to maintain blood volume (Silbert et al., 1981; Shaikh et al., 1985; Blumberg, 1986; Snyder et al., 1986), at Beaumont FFP was given to increase plasma volume in only three cases over the 4 years, two of which were on the new-born intensive care service. In our survey, most of the FFP used was for patients
192
J . N . Shanberge and T. Quattrociocchi-Longe Table 3. FFP usage according to
Liver disease Pre-procedure Coumadin Pre-procedure CABG Aortic aneurysm Valve replacement Massive bleed ICU syndrome Miscellaneous*
FFP units
% of total
Patients
% of total
429 55 322 159 365 216 105 74 66 588
16.4 2.1 12.3 6.1 14-0 8.3 4.0 2.8 2.5 22.5
97 29 142 67 118 61 21 18 34 215
11.1 3.3 16.3 7.7 13.5 7.0 3.1 2.1 3.9 24.6
diagnosis
* Includes 63 general surgery, 28 G1bleed, 28 cancer-related cases. 14haematological disease,4 thrombotic thrornbocytopenic purpura, 4 coagulation defect (3 factor XI deficiency, I factor IX deficiency)and 3 volume expansion.
Table 4. Points to reduce FFP usage 1.
2. 3. 4.
5. 6. 7. 8. 9.
Avoid empiric use of FFP. Formulas for use of FEP/packed red cells should not be used Give only enough FFP needed for haemostasis. Do not treat laboratory values FFP is not needed for minor procedures Prolonged PT or PTT values may be due to a low fibrinogen level or the presence of heparin Obtain PT and/or PTT shortly after FFP administration to determine adequacy of dosage Five units of platelet concentrate have a plasma equivalence of 1 unit of FFP Cryoprecipitate is more efficient than FFP to raise fibrinogen levels Prevent iatrogenic vitamin K deficiency (ICU syndrome) Always consider the possibilities of disease transmission with each unit
administered
with liver disease, on coumadin therapy, or during and after cardiovascular surgery (Table 3). There were, among the miscellaneous group, several instances when the FFP was given for unusual reasons: in two cases with hepatorenal syndrome to promote glomerular filtration, in one case of chronic thrombocytopenia to simulate platelet production, and in one case of septicaemia to increase the opsonic index. The number of units of F F P given per month at Beaumont for the 4 years averaged 54, in contrast to an average of 214 units per month in 1984. This resulted in considerable financial savings for the Hospital as well as a decrease in the possible unwanted complications of transfusion (Walker, 1987). If the use of FFP were to follow the criteria exactly, this average would drop to around 32 units per month. How can the use of FFP be decreased further, other than by instituting a prospective regulatory pro-
gramme? The important points are summarized in Table 4. First, even when F F P is indicated, only that amount should be given which is necessary to bring about haemostasis. It must be emphasized that a 16- or 17-s PT, although not ‘normal’ still represents a coagulation activity well within the haemostatic range (White et al., 1987). A clotting activity of at least 30%, with rare exceptions, is adequate even for most surgical procedures (Salzman, 1987). Therefore, it is not necessary, in the majority of cases, to reduce the PT or P?T to ‘normal’ levels. This is particularly true with minor surgical or invasive procedures. In a retrospective study of 608 patients undergoing paracentesis or thoracentesis, it was found that no significant bleeding occurred in the presence of prolonged PT, even as long as 24 s (McVay & Toy, 1991). Secondly, the empirical use of formulas, whereby a certain number of units of F F P are given for so many
Fresh frozen plasma usage 193 units of red cells, should be discontinued. Even with a large loss of plasma volume, the clotting activity may not fall below the haemostatic level (Counts et al., 1979; Mannucci et al., 1982; Reed ef al., 1986; Ciavarella et al., 1987; Phillips et al., 1987; Murray et al., 1988). Thirdly, it is often not realized that the prolonged PT or PTT value, on which FFP administration is based, may be erroneous due to a decreased fibrinogen level, as with massive haemorrhage or disseminated intravascular coagulation. In such instances, cryoprecipitate (cryo), a better source of fibrinogen, volumewise, is indicated for the initial therapy rather than FFP. A prolonged PTT may also be due to contamination of the blood sample by heparin which must first be removed in order to obtain an accurate value. Fourthly, in order to monitor the effectiveness of components such as FFP, the PT and/or PTT should be ordered soon after its administration. Because of the short half-life of factor VII (Salzman, 1987), a factor which is less essential for haemostasis, a delay in ordering a PT may lead to a lengthening of the PT giving a false impression of the inadequacy of dosage. Moreover, it is equally important to know whether the calculated number of units of FFP given is sufficient to reach the desired level of activity. Fifthly, when a patient needs both platelets and coagulation factors, for every 5 units of platelets administered, the patient wilI have received the equivalent of one unit of FFP, so that additional FFP may not be needed. Sixthly, prevention of a vitamin K deficiency, as occurs in severely ill patients who are given broadspectrum antibiotics and only intravenous fluids, the so-called ICU syndrome (Pineo et al., 1973), will prevent a few cases of unexpected bleeding requiring FFP. Finally, when there is a significantly large hole in a blood vessel, no amount of FFP or, for that matter, any blood component, is going to substitute for ligation, suturing, cautery, or sclerosis. The success of our programme has been based on education and consultation rather than regulation. More recently, it has been applied to the use of other blood components with an appreciable reduction in the number of units used. It is now proposed to apply a similar programme to red cell transfusion. REFERENCES Blumberg, N., Laczin, J., McMican, A., Heal, J. & Arvan. D. (1986) A critical survey of fresh-frozen plasma use. Transfusion, 26, 51 1-513. Brien, W.F., Butler, R.J. & Inwood, M.J. (1989). An audit of
blood component therapy in a Canadian general teaching hospital. Canadian Medical Association Journal, 140,8 12815.
Ciavarella, D., Reed, R.L., Counts, R.B., Baron, L., Pavlin, E., Heimbach, D.M. & Carrico, C.J. (1987) Clotting factor levels and the risk of diffuse microvascular bleeding in the massively transfused patient. British Journal of Haematology, 67,365-368. Coffin, C.M. (1987) Current issues in transfusion therapy. 2. Indications for use of blood components. Postgraduate Medicine, 81, 343-350. Counts, R.B., Haisch, C., Simon, T.L., Maxwell, N.G., Heimbach, D.M.& Carrico, C.J. (1979) Hemostasis in massively transfused trauma patients. Annals of Surgery, 190,91-99.
McVay, P.A. & Toy, P.T.C.Y. (1991) Lack of increased bleeding after paracentesis and thoracentesis in patients with mild coagulation abnormalities. Transfusion, 21, 164-171.
Manucci, P.M., Federici, A.B. & Sirchia, G. (1982) Hemostasis testing during massive blood replacement. A study of 172 cases. Vox Sanguinis, 42, 1 13- 123. Martinowitz, U., Goor, D.A., Ramot, B. & Mohr, R. (1990) Is transfusion of fresh plasma after cardiac operations indicated? Journal of Thoracic and Cardiovascular Surgery, 100,92-98. Mozes, B., Epstein, M., Ben-Bassat I., Modan, €3. & Halkin, H. (1989) Evaluation of the appropriateness of blood and blood product transfusion using preset criteria. Transfusion, 29,473-76. Murray, D.J., Olson, J., Strauss, R. & Tinker, J.H. (1988) Coagulation changes during packed red cell replacement of major blood loss. Anesthesiology, 69, 839-45. NIH Consensus Conference (1 985) Fresh-frozen plasma. Indications and risks. Journal of the American Medical Association, 253, 551-553. Oberman, H.A. (1985) Uses and abuses of fresh frozen plasma. In: Current Concepts in Transfusion Therapy (ed. Garratty, A.), 109-124. American Association of Blood Banks, Arlington, VA. Phillips, T.F.,Soulier, G. &Wilson, R.F. (1987) Outcome of massive transfusion exceeding two blood volumes in trauma and emergency surgery. Journal of Trauma, 27, 903-910.
Pineo, G.F., Brain, M.C. & Hirsh, J. (1973) Unexpected vitamin K deficiency in hospitalized patients. Canadian Medical Association Journal, 109, 880-883. Reed, R.L. 11, Ciavarella, D., Heimbach, D.M., Baron, L., Pavlin, E., Counts, R.B. & Carrico, C.J. (1986) Prophylactic platelet administration during massive transfusion. A prospective, randomized, double-blind clinical study. Annals of Surgery, 203, 40-48. Renner, S.W., Howanitz, J.H. & Fishkin, B.G. (1987) Toward meaningful blood usage review: Comprehensive monitoring of physician practice. Quality Review Bulletin. 13,76-80.
194
J . N . Shanberge and T. Quattrociocchi-Longe
Salzman, E.W. (1987). Hemostatic problems in surgical patients. In: Hemostasis and Thrombosis. Basic Principles and Clinical Practice (eds Colrnan, R.W., Hirsh, J., Marder, V.J. and Salzman, E.W.), 920-925. J.B. Lippincott, Philadelphia, PA. Shaikh, B.S., Wagar, D., Lau, P.M. & Campbell, E.W. Jr. (1985) Transfusion pattern of fresh frozen plasma in a medical school hospital. Vox Sanguinis, 48, 366-369. Shanberge, J.N. (1987) Reduction of fresh-frozen plasma use through a daily survey and education program. Transfusion, 27, 226-227. Shanberge, J.N., Murato, M., Quattrociocchi-Longe, T. & Van Neste, L.(1987) Heparin-protarnine compkxes in the production of heparin rebound and other complications of extracorporeal bypass procedures. American Journal of Clinical Pathology, 87,210-217.
Silberstein, L.E., Kruskall, M.S.,Stehling, L.C., Johnston, M.F.M., Rutman, R.C., Samja, C.T., Ramsey, G. & Eisenstaedt, S. (1989) Strategies for the review of transfusion practices. Journal of the American Medical Association, 262, 1993- 1997.
Silbert, J.A., Bove, J.R., Dubin, S. & Bush, W.S. (1981) Patterns of frozen plasma use. Connecticut Medicine, 45, 507-5 1 1.
Solomon, R.R., Clifford, J.A. & Gutman, S.I. (1988) The use of laboratory intervention to stem the flow of fresh-frozen plasma. American Journal of Clinical Pathology, 89,51821.
Snyder, A.J., Gottschall, J.L. & Menitove, J.E. (1986) Why is fresh-frozen plasma transfused? Transfusion,26, 107- 1 12. Walker, R.H. (1987) Special report: Transfusion risks. American Journal of Clinical Pathology, 88,374-378. White, G.C. 11. Marder, V.J., Colman, R.W., Hirsh, J. & SaIzman, E.W.(1987) Approach to the bleeding patient. In: Hemostasis and Thrombosis. Basic Principles and Clinical Practice (eds Colman, R.W., Hirsh, J., Marder, V.J. & .Salzman, E.W.), 1048-1060. J.B. Lippincott, Philadelphia, PA. Willis, J.L. (4. (1989) ) Use of blood components. FDA Drug Bulletin, Public Health Service, Rock, MD.
