THROMBOSIS RESEARCH 66; !%3-590,1992 0049-3848/92 $5.00 + .OOPrinted in the USA. Copyright (c) 1992 Pergamon Press Ltd. All rights reserved.
BEDSIDE MONITORING OF WARFARIN THERAPY BY A WHOLE BLOOD CAPILLARY
COAGULATION
MONITOR.
YOSHIKO YANO, JUN-ICHI KAMBAYASHI, KOUHEI MURATA, EIICHI SHIBA, MASATO SAKON, TOM10 KAWASAKI and TAKESADA MORI Department of Surgery II, Osaka University Medical School, Osaka 553, Japan. (Received 24.12.1991; accepted in revised form 4.3.1992 by Editor A. Takada) ABSTRACT A recently developed portable capillary coagulation monitor device allows rapid determination of prothrombin time (PT) in sec. #and in International Normalized Ratio (INR) from a drop of whole blood. Considering a possible application of this device for monitoring of warfarin therapy at bedside or at outpatient clinic, a comparative study was performed in patients receiving warfarin between the whole blood capillary PT(WBC-PT) and conventional laboratory tests such as plasma PT (LAB-PT) and Thrombotest (TTO). There was an excellent positive correlation (rz0.95, n=49) between LAB-PT(sec) and WBC-PT(sec). As the correlation coefficient between WBC-PT(sec) and the reciprocal of LAB-PT(%)was excellent (r=0.95+49), it is feasible to convert WBCPT(sec) to LAB-PT(%) by the following formula; y=313.48x+8.29 (x:reciprocal value of LAB-PT(%), y:WBC-PT). As the correlation between WBC-PT(sec) and TTO(sec) was also excellent (r=0.85, n=63),TTO(%) may be likewise estimated by the following formula; y=66.49x+11.36 (x:reciprocal value of TTO(%), y:WBC-PT). From these observations it is concluded that the determination of WBC-PT is simple and accurate and that the monitoring of warfarin therapy is easily performed by the present method.
INTRODUCTION Oral anticoagulation therapy with warfarin has been proven beneficial for the treatment and prevention of various thromboembolic disorders( 1 ). In arder to prevent bleeding complications, which occur in 4-40% of patients receiving long term ______________________ KEY WORDS : Warfarin monitoring, Prothrombin time, Thrombotest
583
584
BEDSIDE WARFARIN MONITORING
Vol. 66, No. 5
warfarin therapy( 2, 3 ), and to adjust the dosage properly, monitoring of the warfarin therapy is essential and has been practised by way of detecting ex vivo hypocoagulability. As the essential pharmacological effect of warfarin is the inhibition of vitamin K dependent carboxylation of coagulation factors resulting in the formation of PIVKA (Protein Induced Vitamin K Absence), frequent determination of prothrombin time (PT) and Thrombotest (TTO) has been the practical tool of the monitoring( 4, 5 ). We have been administering warfarin to our outpatients for the purpose of prevention and treatment of deep vein thrombosis and arterial graft occlusion and the monitoring has been performed by determination of PT or TTO in percent every 2 to 4 weeks. Because it has been difficult to obtain the result of the test while seeing patients , such a device to obtain the result at bedside has been sought. As a portable whole blood capillary coagulation monitor apparatus has been developed which allows the estimation of plasma equivalent PT in set and International Normalized Ratio (INR), attempts were made to investigate the correlation between the whole blood capillary PT (WBC-PT) and laboratory PT (LABPT) or TTO for possible application of this apparatus in warfarin monitoring at outpatient clinic. MATERIALS
AND METHODS
1. Comoarison between WRC-PT and I AR-PT First, 49 patients with warfarin therapy who visited our vascular surgery outpatient clinic were subjected to the study. Daily administration of warfarin ( mean 2.6mg/day, range l -6mg ) was assigned because of the following conditions: 17 patients .. ..deep vein thrombosis, 32 patients ....status post vascular reconstruction. After measuring WBC-PT in the outpatient clinic, each patient was sent to the central laboratory for the determination of LAB-PT(sec&%). Also,1 7 healthy volunteers and 28 inpatients admitted for elective surgery,who did not receive warfarin,were subjected to the simultaneous determination of WBC-PT and LAB-PT(sec&%).
2. Cornpa ‘son between WEE-PT and TTO Anotherngroup of 63 outpatients were subjected to the study, who were receiving warfarin ( mean 2.6mg/day, range l-6mg ) because of the following conditions: 23 patients....deep vein thrombosis, 37 patients . ...status post vascular reconstruction, 3patients .. ..cavenous hemangioma. Two sets of venous blood collection were performed at our outpatient clinic and one whole blood sample was immediately subjected to WBC-PT determination and another sample was processed to titrated plasma, which was stored in deep freezer within a week until determination of TTO (se&%) as described below. We also examined and confirmed that there is no difference between activity of TTO in the fresh samples and those which were dissolved after storage in deep freezer (below -20°C) within a week. 3. Determination of WBC-PT WBC-PT was determined by a small portable apparatus (Ciba Corning 512 Coagulation System, manufactured by the Ciba Corning Diagnostic, Inc. U.S.A) , which was kindly supplied by Ciba Corning, Tokyo. The apparatus is a laser photometer, powered by chargeable batteries and it allows us to measure not only PT but also APTT by use of respective cartridge. The cartridge for PT contains a dry rabbit brain thromboplastin. Although only a drop of fresh whole blood (minimum of 25p.l) which is enough to fill the small chamber in the cartridge is needed for the determination, about lml of whole blood was drawn into a plastic syringe by venipuncture at forearm. After discarding the first few drops, the chamber was filled. The coagulation reaction begins as the blood rehydrates the thromboplastin and the mixture continues to flow in the capillary until a clot is formed. The time between application of blood and cessation of blood flow is automatically converted to the
Vol. 66, No. 5
BEDSIDE WARFARIN MONITORING
585
prothrombin time and INR. To evaluate the precision of this monitor, we also examined two sets of the same samples measured by these two different monitors. 4. Determination of I AR-PT & TTQ In the central laboratory, a titrated plasma was immediately prepared after venipuncture, which was subjected to automated coagulation monitor system (Amelung Coagulometer KC-1 0,U.S.A). The method to determine LAB-PT is essentially based on Quick’s one step procedure and the following tissue thromboplastin reagent (Rabbit thromboplastin C ,DadeB Baxter Healthcare, Inc.in U.S.A.) was used. Because TTO is not regularly performed in the Central Laboratory,the determination was performed in our laboratory. Using a TTO reagent (Thrombotest OwrenB , NYCOMED AS , NORWAY), the end point was determined by a Clotek coagulation monitor system ( HYLAND DIVISION TRAVENOL LABORATORIES Inc.,U.S.A.). Standard curves were obtained at each set of the determination. 5. Statistical analvsis. Statistical analysis was carried out using programs in the Multi Variant software package( Micro Systems Inc. Japan). Correlation between two variables are expressed using the Pearson’s correlation coefficient. RESULTS . .
1 The evaluation of the p ec sron of this monw ’The PT values of two sits Lf the same samples measured by two different monitors gave similar results and two groups of PT values show an excellent correlation coefficient (r=0.95). 2. Inter elationshio bet een WBC-PT and I AR-PT First,lhe interrelatiorship between WBC-PT and LAB-PT(sec) was investigated. The correlation coefficient was 0.95 with a slope of 1.08 and an intercept of -0.38, indicating an excellent correlation(FlG.1). ThenLAB-PT expressed in percent was plotted against the values of WBC-PT. FIG.2 shows there was an inverse correlation with coefficient value of -0.88. .
0
45
I
I
1
I
I
8
10
12
14
16
I
18 LAB-PT
I
20 (set)
FIG 1. Interrelationship between WBC-PT and LAB-PT(sec). The correlation coefficient was 0.95 with a slope of 1.08 and an intercept of -0.38.
Vol. 66, No. 5
BEDSIDE WARFARIN MONITORING
. . .
.
*a
r = -0.88 n =49 I
0
1
I
I
I
1
20
40
60
80
100
LAB-PT
(%I
FIG. 3 Correlation between LAB-PT(%) and WBC-PT. There was an inverse correlation with coefficient value of -0.88.
30
1
10
r =0.95
i
Ts,
I
0
0.01
I
I
0.02
,
,
0.03
1
, n =49 , , 0.04 0.05
1/LAB-PT
,
,
0.06
(%I
FIG. 3 Interrelationship between the reciprocal values of LAB-PT(%) and WBC-PT. There was an excellent correlation with coefficient value of 0.95, and following conversion formula was obtained: y=313.48x+8.29 (x:reciprocal value of LAB-PT(%), y:WBC-PT)
BEDSIDE WARFARIN MONITORING
Vol. 66, No. 5
587
2.5
tY Z -
2 1.5
0.006
0.012
0.019
0.026
0.032 l/LAB-PT
0.039 (%)
~~~.l;Rlnterrelationship between the reciprocal.values of LAB-.PT(%) There was also an excellent correlation wtth coeffment valu’e of 0.93. .
.
y =0.09x +9.22 r so.85 n =63
0
I
I
30
60 TTO
I
90
(set)
FIG. 5 Correlation between WBC-PT and TTO(sec). These two values were significantly correlated (k0.85).
BEDSIDE WARFARIN MONITORING
Vol. 66. No. 5
IE-
0
41
I
0.01
I
0.03
I
0.05
I
1
0.07
0.09
I/TTO
(96)
FIG 6. Relationship between WBC-PT and reciprocal values of TTO(%). There was an excellent correlation coefficient of 0.86 and following conversion formula was obtained: y=66.49x+11.44 (x:reciprocal value of TTO(%), y:WBC-PT)
By replotting the reciprocal values of LAB-PT(%) against WBC-PT, an excellent correlation was obtained ( r=0.95 ) (FIG.3.). Thereby, it was found possible to estimate LAB-PT(%) from the value of WBC-PT by the following formula; y=313.48x+8.29 (x:reciprocal value of LAB-PT(%) ,y:WBC-PT) . There was also an excellent correlation between INR values and reciprocal values of LAB-PT(%) (r=0.93) (FIG.4), indicating that it is also possible to estimate LAB-PT(%) from the value of WBC-PT. We also examined 94 samples including volunteers and 28 inpatients who did not receive warfarin and similar excellent correlation coefficient was obtained (r=0.95, -0.84, 0.95, 0.93, rensp;tiv;y). 3. FIG.5 reflects a significant correlation between the values of WBC-PT and TTO (set) (r=0.83). Likewise, there was a excellent correlation between WBC-PT and the reciprocal values of TTO (%) ( FIG.6) and it was possible to estimate the value of TTO (%) from the value of WBC-PT by the following equation; y=66.49x+11.44 (x: reciprocal value of TTO(%), y:WBC-PT )
Although there have been considerable efforts to develop a rapid and precise method to measure clotting time from a drop of capillary blood for not only PT determination but also other coagulation tests such as APTT, the previous devices were not generally approved because of the lack of accuracy due to various problems such as lack of temperature control, use of a visual end point for clot formation and delivery of reagents( 6, 7 ). The present device has been reported to overcome the
Vol. 66, No. 5
BEDSIDE WARFARIN MONITORING
589
problems by providing temperature control, laser photometer which can detect the end point of clot formation and a disposable cartridge with a dried tissue thromboplastin. Actually, the determination of PT by this device was simple and reproducible as has been reported ( 8, 9 ). Normal value of plasma PT yields the shortest coagulation time in set among the available routine laboratory coagulation tests because the reagent of PT, tissue thromboplastin,strongly activates the extrinsic pathway for coagulation. Thereby, it is well known that the quality of the reagent is strictly standardized internationally and that INR is established to allow reasonable comparison of the results among different laboratories ( 4, 10 ). Because only a few seconds of difference in PT reflects a considerable difference in coagulability of the extrinsic pathway, it would be more practical to express the coagulability in percent PT and we have been using percent PT as a reliable indicator in the monito#ring of warfarin. In this viewpoint, TTO is also an ideal laboratory test for the monitoring which was developed to be more sensitive to the presence of PIVKA than PT and is expressed as percentage. It is not possible to draw standard lines of WBC-PT because the diluted whole blood samples may not be used to obtain the lines in the present monitor system. Thereby, we simply determined LAB-PT(%) or TTO simultaneously with WBC-PT and the interrelationship was analyzed. As expected, there were excellent correlations between WBC-PT(sec)and LAB-PT(sec) or TTO (set). When the reciprocal values of LAB-PT(%) or TTO (%) were plotted against the values of WBC-PT(sec), a linear correlation was obtained. From this observation, the formulae were established to estimate LAB-PT(%) or TTO(%)from the value of WBCPT. Therefore, this unique device may be of significant value in the bedside monitoring of warfarin therapy in any institutions regardless of the mode of monitoring, namely PT in set, %, INR and even lTO(%). REFERENCES 1 DALEN,J.E.,HIRSH,J. American College of Chest Physicians and the National Heart, Lung, and Blood Institute National Conference on Antithrombotic Therapy. Arch Intern Med,l46,462-472,1986. 2 COON,W.W.,WILLIS,PARK,W.lll. Hemorrhagic therapy. Arch Intern Med,l33,386-392,1974.
complications
of anti coagulant
3 PETITTI,D.B., STROM,B.L., MELMON, K.L. Duration of warfarin anticoagulant therapy and the probabilities of recurrent thromboembolism and hemorrhage. Am J Med,81,255-259,1986. 4 LOELIGER,E.A., VAN DEN BESSELAAR,A.M.,LEWIS,S.M. Reliability and clinical impact of the normalizatin of the prothrombin times in oral anticoagulant control. Thromb Haemost 53,148-l 54,1985. 5 LOELIGER,E.A. The optimal therapeutic range in oral anticoagulation. proposal. Thromb Haemost 42, 1141-l 152,1979.
History and
6 ULIN,A.W.,GOLLUB,S. New rapid and practical micromethods for prothmrombin determination. The watch glass method. J Lab Clin Med,50,323-325,1957. 7 KATO,K. Micro-prothrombin test with capillary whole blood. A modification of Quick’s quantitative method. Am J Clin Pathol,l 0,147-l53,194O.
590
BEDSIDE WARFARIN MONITORING
Vol. 66, No. 5
8 LUCAS,F.V., DUNCAN,A., JAYR., COLEMAN, R.,CRAFT,P., CHAN, B., WINFREY, L., MUNGALL, D.R. A Novel Whole Blood Capillary Technic for Measuring the Prothrombin Time. Am J Clin Pathol,88,442-446,1987. 9 WHITE,R.H., McCURDY,S.A.,HANS VON MARENSDORGG,WOODRUFF,D.E. Home Prothrombin Time Monitoring after the Initiation of Warfarin Therapy. Ann Intern Med,l 11,730-737 1989. 10 WHO Expert Committee on Biological Technical Report Series, 687, 81-l 05, 1983.
Standardization.
Thirty-third
Report.
BEDSIDE MONITORING OF WARFARIN THERAPY BY A WHOLE BLOOD CAPILLARY
COAGULATION
MONITOR.
YOSHIKO YANO, JUN-ICHI KAMBAYASHI, KOUHEI MURATA, EIICHI SHIBA, MASATO SAKON, TOM10 KAWASAKI and TAKESADA MORI Department of Surgery II, Osaka University Medical School, Osaka 553, Japan. (Received 24.12.1991; accepted in revised form 4.3.1992 by Editor A. Takada) ABSTRACT A recently developed portable capillary coagulation monitor device allows rapid determination of prothrombin time (PT) in sec. #and in International Normalized Ratio (INR) from a drop of whole blood. Considering a possible application of this device for monitoring of warfarin therapy at bedside or at outpatient clinic, a comparative study was performed in patients receiving warfarin between the whole blood capillary PT(WBC-PT) and conventional laboratory tests such as plasma PT (LAB-PT) and Thrombotest (TTO). There was an excellent positive correlation (rz0.95, n=49) between LAB-PT(sec) and WBC-PT(sec). As the correlation coefficient between WBC-PT(sec) and the reciprocal of LAB-PT(%)was excellent (r=0.95+49), it is feasible to convert WBCPT(sec) to LAB-PT(%) by the following formula; y=313.48x+8.29 (x:reciprocal value of LAB-PT(%), y:WBC-PT). As the correlation between WBC-PT(sec) and TTO(sec) was also excellent (r=0.85, n=63),TTO(%) may be likewise estimated by the following formula; y=66.49x+11.36 (x:reciprocal value of TTO(%), y:WBC-PT). From these observations it is concluded that the determination of WBC-PT is simple and accurate and that the monitoring of warfarin therapy is easily performed by the present method.
INTRODUCTION Oral anticoagulation therapy with warfarin has been proven beneficial for the treatment and prevention of various thromboembolic disorders( 1 ). In arder to prevent bleeding complications, which occur in 4-40% of patients receiving long term ______________________ KEY WORDS : Warfarin monitoring, Prothrombin time, Thrombotest
583
584
BEDSIDE WARFARIN MONITORING
Vol. 66, No. 5
warfarin therapy( 2, 3 ), and to adjust the dosage properly, monitoring of the warfarin therapy is essential and has been practised by way of detecting ex vivo hypocoagulability. As the essential pharmacological effect of warfarin is the inhibition of vitamin K dependent carboxylation of coagulation factors resulting in the formation of PIVKA (Protein Induced Vitamin K Absence), frequent determination of prothrombin time (PT) and Thrombotest (TTO) has been the practical tool of the monitoring( 4, 5 ). We have been administering warfarin to our outpatients for the purpose of prevention and treatment of deep vein thrombosis and arterial graft occlusion and the monitoring has been performed by determination of PT or TTO in percent every 2 to 4 weeks. Because it has been difficult to obtain the result of the test while seeing patients , such a device to obtain the result at bedside has been sought. As a portable whole blood capillary coagulation monitor apparatus has been developed which allows the estimation of plasma equivalent PT in set and International Normalized Ratio (INR), attempts were made to investigate the correlation between the whole blood capillary PT (WBC-PT) and laboratory PT (LABPT) or TTO for possible application of this apparatus in warfarin monitoring at outpatient clinic. MATERIALS
AND METHODS
1. Comoarison between WRC-PT and I AR-PT First, 49 patients with warfarin therapy who visited our vascular surgery outpatient clinic were subjected to the study. Daily administration of warfarin ( mean 2.6mg/day, range l -6mg ) was assigned because of the following conditions: 17 patients .. ..deep vein thrombosis, 32 patients ....status post vascular reconstruction. After measuring WBC-PT in the outpatient clinic, each patient was sent to the central laboratory for the determination of LAB-PT(sec&%). Also,1 7 healthy volunteers and 28 inpatients admitted for elective surgery,who did not receive warfarin,were subjected to the simultaneous determination of WBC-PT and LAB-PT(sec&%).
2. Cornpa ‘son between WEE-PT and TTO Anotherngroup of 63 outpatients were subjected to the study, who were receiving warfarin ( mean 2.6mg/day, range l-6mg ) because of the following conditions: 23 patients....deep vein thrombosis, 37 patients . ...status post vascular reconstruction, 3patients .. ..cavenous hemangioma. Two sets of venous blood collection were performed at our outpatient clinic and one whole blood sample was immediately subjected to WBC-PT determination and another sample was processed to titrated plasma, which was stored in deep freezer within a week until determination of TTO (se&%) as described below. We also examined and confirmed that there is no difference between activity of TTO in the fresh samples and those which were dissolved after storage in deep freezer (below -20°C) within a week. 3. Determination of WBC-PT WBC-PT was determined by a small portable apparatus (Ciba Corning 512 Coagulation System, manufactured by the Ciba Corning Diagnostic, Inc. U.S.A) , which was kindly supplied by Ciba Corning, Tokyo. The apparatus is a laser photometer, powered by chargeable batteries and it allows us to measure not only PT but also APTT by use of respective cartridge. The cartridge for PT contains a dry rabbit brain thromboplastin. Although only a drop of fresh whole blood (minimum of 25p.l) which is enough to fill the small chamber in the cartridge is needed for the determination, about lml of whole blood was drawn into a plastic syringe by venipuncture at forearm. After discarding the first few drops, the chamber was filled. The coagulation reaction begins as the blood rehydrates the thromboplastin and the mixture continues to flow in the capillary until a clot is formed. The time between application of blood and cessation of blood flow is automatically converted to the
Vol. 66, No. 5
BEDSIDE WARFARIN MONITORING
585
prothrombin time and INR. To evaluate the precision of this monitor, we also examined two sets of the same samples measured by these two different monitors. 4. Determination of I AR-PT & TTQ In the central laboratory, a titrated plasma was immediately prepared after venipuncture, which was subjected to automated coagulation monitor system (Amelung Coagulometer KC-1 0,U.S.A). The method to determine LAB-PT is essentially based on Quick’s one step procedure and the following tissue thromboplastin reagent (Rabbit thromboplastin C ,DadeB Baxter Healthcare, Inc.in U.S.A.) was used. Because TTO is not regularly performed in the Central Laboratory,the determination was performed in our laboratory. Using a TTO reagent (Thrombotest OwrenB , NYCOMED AS , NORWAY), the end point was determined by a Clotek coagulation monitor system ( HYLAND DIVISION TRAVENOL LABORATORIES Inc.,U.S.A.). Standard curves were obtained at each set of the determination. 5. Statistical analvsis. Statistical analysis was carried out using programs in the Multi Variant software package( Micro Systems Inc. Japan). Correlation between two variables are expressed using the Pearson’s correlation coefficient. RESULTS . .
1 The evaluation of the p ec sron of this monw ’The PT values of two sits Lf the same samples measured by two different monitors gave similar results and two groups of PT values show an excellent correlation coefficient (r=0.95). 2. Inter elationshio bet een WBC-PT and I AR-PT First,lhe interrelatiorship between WBC-PT and LAB-PT(sec) was investigated. The correlation coefficient was 0.95 with a slope of 1.08 and an intercept of -0.38, indicating an excellent correlation(FlG.1). ThenLAB-PT expressed in percent was plotted against the values of WBC-PT. FIG.2 shows there was an inverse correlation with coefficient value of -0.88. .
0
45
I
I
1
I
I
8
10
12
14
16
I
18 LAB-PT
I
20 (set)
FIG 1. Interrelationship between WBC-PT and LAB-PT(sec). The correlation coefficient was 0.95 with a slope of 1.08 and an intercept of -0.38.
Vol. 66, No. 5
BEDSIDE WARFARIN MONITORING
. . .
.
*a
r = -0.88 n =49 I
0
1
I
I
I
1
20
40
60
80
100
LAB-PT
(%I
FIG. 3 Correlation between LAB-PT(%) and WBC-PT. There was an inverse correlation with coefficient value of -0.88.
30
1
10
r =0.95
i
Ts,
I
0
0.01
I
I
0.02
,
,
0.03
1
, n =49 , , 0.04 0.05
1/LAB-PT
,
,
0.06
(%I
FIG. 3 Interrelationship between the reciprocal values of LAB-PT(%) and WBC-PT. There was an excellent correlation with coefficient value of 0.95, and following conversion formula was obtained: y=313.48x+8.29 (x:reciprocal value of LAB-PT(%), y:WBC-PT)
BEDSIDE WARFARIN MONITORING
Vol. 66, No. 5
587
2.5
tY Z -
2 1.5
0.006
0.012
0.019
0.026
0.032 l/LAB-PT
0.039 (%)
~~~.l;Rlnterrelationship between the reciprocal.values of LAB-.PT(%) There was also an excellent correlation wtth coeffment valu’e of 0.93. .
.
y =0.09x +9.22 r so.85 n =63
0
I
I
30
60 TTO
I
90
(set)
FIG. 5 Correlation between WBC-PT and TTO(sec). These two values were significantly correlated (k0.85).
BEDSIDE WARFARIN MONITORING
Vol. 66. No. 5
IE-
0
41
I
0.01
I
0.03
I
0.05
I
1
0.07
0.09
I/TTO
(96)
FIG 6. Relationship between WBC-PT and reciprocal values of TTO(%). There was an excellent correlation coefficient of 0.86 and following conversion formula was obtained: y=66.49x+11.44 (x:reciprocal value of TTO(%), y:WBC-PT)
By replotting the reciprocal values of LAB-PT(%) against WBC-PT, an excellent correlation was obtained ( r=0.95 ) (FIG.3.). Thereby, it was found possible to estimate LAB-PT(%) from the value of WBC-PT by the following formula; y=313.48x+8.29 (x:reciprocal value of LAB-PT(%) ,y:WBC-PT) . There was also an excellent correlation between INR values and reciprocal values of LAB-PT(%) (r=0.93) (FIG.4), indicating that it is also possible to estimate LAB-PT(%) from the value of WBC-PT. We also examined 94 samples including volunteers and 28 inpatients who did not receive warfarin and similar excellent correlation coefficient was obtained (r=0.95, -0.84, 0.95, 0.93, rensp;tiv;y). 3. FIG.5 reflects a significant correlation between the values of WBC-PT and TTO (set) (r=0.83). Likewise, there was a excellent correlation between WBC-PT and the reciprocal values of TTO (%) ( FIG.6) and it was possible to estimate the value of TTO (%) from the value of WBC-PT by the following equation; y=66.49x+11.44 (x: reciprocal value of TTO(%), y:WBC-PT )
Although there have been considerable efforts to develop a rapid and precise method to measure clotting time from a drop of capillary blood for not only PT determination but also other coagulation tests such as APTT, the previous devices were not generally approved because of the lack of accuracy due to various problems such as lack of temperature control, use of a visual end point for clot formation and delivery of reagents( 6, 7 ). The present device has been reported to overcome the
Vol. 66, No. 5
BEDSIDE WARFARIN MONITORING
589
problems by providing temperature control, laser photometer which can detect the end point of clot formation and a disposable cartridge with a dried tissue thromboplastin. Actually, the determination of PT by this device was simple and reproducible as has been reported ( 8, 9 ). Normal value of plasma PT yields the shortest coagulation time in set among the available routine laboratory coagulation tests because the reagent of PT, tissue thromboplastin,strongly activates the extrinsic pathway for coagulation. Thereby, it is well known that the quality of the reagent is strictly standardized internationally and that INR is established to allow reasonable comparison of the results among different laboratories ( 4, 10 ). Because only a few seconds of difference in PT reflects a considerable difference in coagulability of the extrinsic pathway, it would be more practical to express the coagulability in percent PT and we have been using percent PT as a reliable indicator in the monito#ring of warfarin. In this viewpoint, TTO is also an ideal laboratory test for the monitoring which was developed to be more sensitive to the presence of PIVKA than PT and is expressed as percentage. It is not possible to draw standard lines of WBC-PT because the diluted whole blood samples may not be used to obtain the lines in the present monitor system. Thereby, we simply determined LAB-PT(%) or TTO simultaneously with WBC-PT and the interrelationship was analyzed. As expected, there were excellent correlations between WBC-PT(sec)and LAB-PT(sec) or TTO (set). When the reciprocal values of LAB-PT(%) or TTO (%) were plotted against the values of WBC-PT(sec), a linear correlation was obtained. From this observation, the formulae were established to estimate LAB-PT(%) or TTO(%)from the value of WBCPT. Therefore, this unique device may be of significant value in the bedside monitoring of warfarin therapy in any institutions regardless of the mode of monitoring, namely PT in set, %, INR and even lTO(%). REFERENCES 1 DALEN,J.E.,HIRSH,J. American College of Chest Physicians and the National Heart, Lung, and Blood Institute National Conference on Antithrombotic Therapy. Arch Intern Med,l46,462-472,1986. 2 COON,W.W.,WILLIS,PARK,W.lll. Hemorrhagic therapy. Arch Intern Med,l33,386-392,1974.
complications
of anti coagulant
3 PETITTI,D.B., STROM,B.L., MELMON, K.L. Duration of warfarin anticoagulant therapy and the probabilities of recurrent thromboembolism and hemorrhage. Am J Med,81,255-259,1986. 4 LOELIGER,E.A., VAN DEN BESSELAAR,A.M.,LEWIS,S.M. Reliability and clinical impact of the normalizatin of the prothrombin times in oral anticoagulant control. Thromb Haemost 53,148-l 54,1985. 5 LOELIGER,E.A. The optimal therapeutic range in oral anticoagulation. proposal. Thromb Haemost 42, 1141-l 152,1979.
History and
6 ULIN,A.W.,GOLLUB,S. New rapid and practical micromethods for prothmrombin determination. The watch glass method. J Lab Clin Med,50,323-325,1957. 7 KATO,K. Micro-prothrombin test with capillary whole blood. A modification of Quick’s quantitative method. Am J Clin Pathol,l 0,147-l53,194O.
590
BEDSIDE WARFARIN MONITORING
Vol. 66, No. 5
8 LUCAS,F.V., DUNCAN,A., JAYR., COLEMAN, R.,CRAFT,P., CHAN, B., WINFREY, L., MUNGALL, D.R. A Novel Whole Blood Capillary Technic for Measuring the Prothrombin Time. Am J Clin Pathol,88,442-446,1987. 9 WHITE,R.H., McCURDY,S.A.,HANS VON MARENSDORGG,WOODRUFF,D.E. Home Prothrombin Time Monitoring after the Initiation of Warfarin Therapy. Ann Intern Med,l 11,730-737 1989. 10 WHO Expert Committee on Biological Technical Report Series, 687, 81-l 05, 1983.
Standardization.
Thirty-third
Report.
