Stability of canine factor VIII: coagulant activity in vitro.

A pilot study was undertaken to assess the stability of canine factor VIII:coagulant (FVIII:C) activity over three days, under various storage conditions (plasma at 4, 20 and 37 degrees C, whole blood at 4 and 20 degrees C). Blood collected from normal and hemophiliac dogs was used. Both plasma and whole blood samples appeared to be stable for up to 48 h at 4 and 20 degrees C. A subsequent study evaluated FVIII:C stability at 4 and 20 degrees C when stored as whole blood only. Samples were tested at 0, 24 and 48 h after collection. At 4 degree C there was a significant decline at 24 h (p less than 0.05), from 110% to 97% (mean values). Although the mean value was further decreased at 48 h (89%) this was not significant (p greater than 0.05). No significant change in FVIII:C activity was observed in whole blood stored at 20 degrees C for 24 or 48 h (110% and 107% respectively). These results suggest that canine whole blood samples collected into sodium citrate stored at 20 degrees C are adequate for routine FVIII:C assay for up to 48 h after collection.     

Effect of acepromazine, xylazine and thiopentone on factor VIII activity and von Willebrand factor antigen concentration in dogs

The effect of acepromazine maleate, xylazine and thiopentone on the packed cell volume, plasma protein content, factor VIII activity and von Willebrand factor antigen concentration of blood was studied in normal dogs. The same variables were measured in dogs with haemophilia A given acepromazine maleate and thiopentone. Both the packed cell volume and plasma protein content decreased after the administration of either acepromazine maleate or xylazine. Values were not changed further after administration of thiopentone. Changes in the haemostatic variables measured were generally small. Consequently, blood samples collected from dogs under the influence of premedicant doses of acepromazine maleate or xylazine, and when subsequently anaesthetised with thiopentone, are adequate for the assay of factor VIII activity and von Willebrand factor antigen concentration for establishing an animal's haemophilia A and von Willebrand's disease status.     

Effect of desmopressin on plasma factor VIII and von Willebrand factor concentrations in Greyhounds

Objective To determine the effect of 1-Deamino-8-D-argi-nine vasopressin on plasma concentrations of von Willebrand factor and factor VIII in Greyhound blood donors, and to compare the response of 1-Deamino-8-D-arginine vaso-pressin injection on plasma concentrations of von Willebrand factor between groups with different resting plasma concentrations of von Willebrand factor.
Animals Fifteen Greyhound blood donors were used. Dogs were grouped into three categories depending on their von Willebrand factor concentrations.
Procedure Desmopressin was administered subcuta-neously at 1 mg/kg to all dogs. Plasma von Willebrand factor and factor VIII concentrations were measured before and 10, 20, 30, 45, 60, 90 and 120 min after desmopressin injection.
Results The von Willebrand factor and factor VIII concentrations in all dogs increased significantly and remained higher than base-line throughout the 2 h period.
Conclusion Desmopressin is useful in increasing von Willebrand factor concentrations in Greyhound blood donors, including those with low resting concentrations.     

Failure of sodium pentobarbital anesthesia to alter 1-desamino-8-D-arginine vasopressin-induced elevations of plasma factor VIII/ von Willebrand factor in normal dogs.

The vasopressin analog 1-desamino-8-D-arginine stimulates elevations in plasma Factor VIII/ von Willebrand factor in normal dogs. In order to study the effects of general anesthesia on this response, six dogs were anesthetized with sodium pentobarbital or given an equivalent amount of saline then challenged with an intravenous dose of 1-desamino-8-D-arginine (0.6 micrograms/kg body weight). Factor VIII coagulant activity, von Willebrand factor antigen, and ristocetin cofactor activity were quantitated before anesthesia (or saline infusion), 20 min after induction (pre-1-desamino-8-D-arginine), and at 30 and 60 min post-1-desamino-8-D-arginine. Anesthesia did not significantly affect the elevations in plasma Factor VIII/ von Willebrand factor induced by 1-desamino-8-D-arginine. Sodium pentobarbital appeared however to prevent the rise in Factor VIII coagulant activity seen following saline treatment. The results of this study suggest that when 1-desamino-8-D-arginine is to be used in normal dogs to boost basal plasma von Willebrand factor levels, it is not necessary to administer it prior to induction of general anesthesia with sodium pentobarbital.     

Stability of stored canine plasma for hemostasis testing

BACKGROUND: A review of the literature revealed limited information about the stability of samples for coagulation testing in dogs. OBJECTIVE: The aim of this study was to evaluate the stability of individual coagulation factors, clotting times, and other parameters of hemostasis in stored canine plasma. METHODS: Citrated plasma samples were obtained from 21 dogs. Prothrombin time (PT), activated partial thromboplastin time (aPTT), fibrinogen concentration, and factor I, II, V, VII, VIII, IX, X, XI, and XII activities were measured on an automated coagulation analyzer with commercially available reagents. Antithrombin (AT) activity and D-dimer concentration were measured on an automated chemistry analyzer using validated kits. Samples were analyzed within 1 hour after collection (initial analysis) and once daily for 2 or 4 consecutive days following storage at room temperature (RT) or 4 degrees C, respectively. RESULTS: Storage time at either temperature did not have any effect on PT, factor II, V, VII, X, or XII activities, D-dimer concentration, or AT activity. In contrast, aPTT was significantly prolonged after 72 and 96 hours at 4 degrees C; fibrinogen concentration was decreased after 48 hours at RT; the activities of factors VIII and IX were decreased after 48, 72, and 96 hours at 4 degrees C; and factor XI activity was decreased after 72 hours at 4 degrees C. CONCLUSIONS: Results suggest that storage of canine plasma for 2 days at RT does not have a significant effect on hemostasis test results with the exception of a slight decrease in fibrinogen concentration. In contrast, aPTT and factors VIII, IX, and XI were unstable in refrigerated plasma after 48 or 72 hours of storage.     

DDAVP-induced increases in coagulation factor VIII and von Willebrand factor in the plasma of conscious dogs

Infusion of the vasopressin analogue DDAVP into five normal dogs at doses of 0.1-2.0 micrograms DDAVP per kg body weight induced dose-dependent increases in the plasma content of coagulation factor VIII and von Willebrand factor. Plasma concentrations of von Willebrand factor (determined antigenically as factor VIII-related antigen and functionally as coagglutinin cofactor activity) and coagulation factor VIII were measured immediately before and at 10, 30, and 120 min after 10-min intravenous infusions of DDAVP. The greatest increases in coagulation factor VIII were produced with the 2.0 micrograms/kg dose. Ten minutes after infusion the mean increase in coagulation factor VIII was 32 units/dl (concentrations of all indices were reported relative to concentrations in a standard canine plasma pool, arbitrarily assigned a concentration of 100 units/dl) and this increase did not change significantly throughout the duration of the experiment. At 10 min post-infusion, the mean factor VIII-related antigen concentration increased 81 units/dl (dose = 2.0 micrograms/kg) and did not change significantly for the duration of the experiment. The maximum mean increase in coagglutinin cofactor activity, 141 units/dl, occurred 10 min after infusion (dose = 1.0 microgram/kg). Coagglutinin cofactor activity decreased significantly from peak activity by 120 min post-infusion.     

Hemostatic response to surgical neutering via ovariectomy and ovariohysterectomy in dogs

Objective-To investigate the hemostatic response to surgery and compare the response for ovariohysterectomy with that for ovariectomy and to evaluate the usefulness of thromboelastography on plasma samples. Animals-42 female dogs. Procedures-Dogs were assigned to undergo ovariohysterectomy or ovariectomy. Blood samples were collected immediately before and 1, 6, and 24 hours after surgery and stored at -80°C for subsequent analysis. Plasma samples were subjected to thromboelastography after thawing. In addition, coagulation variables were measured, including concentrations of von Willebrand factor antigen, fibrinogen, antithrombin, and protein C; activity of factor VIII; activated partial thromboplastin time; prothrombin time; and thrombin time. The fibrinolytic response was assessed via concentrations of D-dimer, plasminogen, and α-2-antiplasmin (plasmin inhibitor). Results-Substantial hemostatic and fibrinolytic activation was evident after surgery in both groups, as characterized by significantly increased global clot strength and an overall hypercoagulable state at 4 hours after surgery in addition to decreases in von Willebrand factor antigen and factor VIII concentrations and shortened prothrombin and thrombin times. The dogs also typically had activation of the fibrinolytic system, as evidenced by increased postoperative concentrations of D-dimer, plasminogen, and plasmin inhibitor. Differences between the 2 groups could not be detected for any variables. Conclusions and Clinical Relevance-Elective surgery with limited tissue trauma induced hemostatic activation in dogs, which led to hypercoagulability after surgery. A difference between the ovariohysterectomy and ovariectomy groups was not detected. Thromboelastography can be used on plasma samples and may be useful for evaluating patterns over time.     

Plasma von Willebrand factor concentration and thyroid function in dogs

Plasma von Willebrand factor antigen concentration was determined in 15 dogs with suspected hypothyroidism, in 1 dog with hyperthyroidism, and in 14 euthyroid dogs. The mean +/- SEM von Willebrand factor:antigen concentration in hypothyroid dogs (47.1% +/- 12.6%) was significantly decreased (P less than 0.0005), compared with that in euthyroid dogs (94.7 +/- 5.6%). Four hypothyroid dogs were given thyroxine for 1 month and all 4 had an increase in von Willebrand factor:antigen concentration. The plasma von Willebrand factor:antigen concentration was 200% in the hyperthyroid dog. Seemingly, reduced concentrations of plasma von Willebrand factor:antigen can be found in dogs in association with congenital von Willebrand disease or with von Willebrand disease acquired through hypothyroidism.     

Stability of hemostatic proteins in canine fresh frozen plasma units

Abstract: The stability of hemostatic proteins, including coagulation factors II, VII, VIII, IX, and X and von Willebrand factor (vWf), in canine fresh frozen plasma (FFP) units stored for up to 1 year was studied. Plasma units from 7 donor dogs were subjected to 4 treatments following collection, including storage at −30°C for 3 months, 6 months, and 1 year and storage at −20°C for 6 months. Coagulant factor activity and vWf concentrations were measured at these times. Significant differences between prestorage and poststorage values were noted for factors VIII, IX, and X, and vWf. Differences seemed to be least pronounced for plasma stored for 3 months; however, a significant interaction between prestorage and poststorage differences and the 4 treatment groups could not be demonstrated. On the basis of factor content in the present study, 15–20 mL/kg of FFP stored for up to 1 year was capable of providing approximately 10–15 U/kg of vWf and factors VIII, IX, X, and II, whereas 10 mL/kg FFP provided at least 10 U/kg of factor VII.     

Von Willebrand's disease in dogs in the United Kingdom

An assay for the measurement of von Willebrand factor antigen has been established. In a period of 18 months, 13 dogs have been identified as suffering from von Willebrand's disease. The affected animals had levels of von Willebrand factor antigen which ranged from undetectable to 43 per cent of normal. Factor VIII levels were also reduced. Haemorrhagic episodes were usually associated with trauma or surgery, and often required transfusion with fresh blood or plasma to arrest haemorrhage.     

Effect of temperature,storage time,and sample type on sorbitol dehydrogenase activity in llama serum and plasma

Abstract: Serum and heparinized plasma samples were collected from 11 adult, clinically healthy llamas. Aliquots were assayed for sorbitol dehydrogenase (SDH) activity after storage at room temperature (20°C), 4°C, or −20°C for defined time intervals up to 1 week postcollection. Sorbitol dehydrogenase activity in all samples was within reference intervals for our laboratory. No difference was found between serum and plasma SDH activity when measured immediately (within 1 hour) after collection. Sorbitol dehydrogenase activity decreased to 79% of initial activity by 24 hours in serum stored at room temperature; plasma had 94% of initial SDH activity under the same conditions. Sorbitol dehydrogenase activity was stable in both plasma and serum stored for up to 1 week at 4°C or −20°C. With the exception of serum stored at 20°C for > 8 hours, in vitro stability of llama SDH was adequate for its use in diagnostic testing.     

Effects of season and sample handling on measurement of plasma alpha-melanocyte-stimulating hormone concentrations in horses and ponies

OBJECTIVE: To investigate effects of sample handling, storage, and collection time and season on plasma alpha-melanocyte-stimulating hormone (alpha-MSH) concentration in healthy equids. ANIMALS: 11 healthy Standardbreds and 13 healthy semiferal ponies. PROCEDURE: Plasma alpha-MSH concentration was measured by use of radioimmunoassay. Effects of delayed processing were accessed by comparing alpha-MSH concentrations in plasma immediately separated with that of plasma obtained from blood samples that were stored at 4 degrees C for 8 or 48 hours before plasma was separated. Effects of suboptimal handling were accessed by comparing alpha-MSH concentrations in plasma immediately stored at -80 degrees C with plasma that was stored at 25 degrees C for 24 hours, 4 degrees C for 48 hours or 7 days, and -20 degrees C for 30 days prior to freezing at -80 degrees C. Plasma alpha-MSH concentrations were compared among blood samples collected at 8:00 AM, 12 noon, and 4:00 PM. Plasma alpha-MSH concentrations were compared among blood samples collected in January, March, April, June, September, and November from horses and in September and May from ponies. RESULTS: Storage of blood samples at 4 degrees C for 48 hours before plasma was separated and storage of plasma samples at 4 degrees C for 7 days prior to freezing at -80 degrees C resulted in significant decreases in plasma alpha-MSH concentrations. A significantly greater plasma alpha-MSH concentration was found in September in ponies (11-fold) and horses (2-fold), compared with plasma alpha-MSH concentrations in spring. CONCLUSIONS AND CLINICAL RELEVANCE: Handling and storage conditions minimally affected plasma alpha-MSH concentrations. Seasonal variation in plasma alpha-MSH concentrations must be considered when evaluating pituitary pars intermedia dysfunction in equids.     

Changes in factor VIII activity and von Willebrand factor antigen concentration with age in dogs.

The factor VIII activity of 38 German shepherd puppies, 6-12 weeks old, submitted for diagnosis of haemophilia A was measured. Eight of these puppies had values higher than would be expected for haemophiliacs, but less than the reference range for adult dogs. A further sequential study of 21 puppies (6-26 weeks of age) indicated that the factor VIII activity of puppies is generally less than that of adult dogs until about 14 weeks of age. Changes in the concentration of von Willebrand factor antigen in the puppies were irregular. These variations are probably not sufficient to interfere with accurate diagnosis of haemophilia A in most affected young dogs, but may interfere with the detection of heterozygotes in young bitches.     

von Willebrand disease phenotype and von Willebrand factor marker genotype in Doberman Pinschers

OBJECTIVE: To define the relationship between clinical expression of a type-1 von Willebrand disease phenotype and genotype at 2 von Willebrand factor marker loci in Doberman Pinschers. ANIMALS: 102 client-owned Doberman Pinschers. PROCEDURES: Dogs were recruited on the basis of plasma von Willebrand factor concentration, clinical history, and pedigree. Blood samples and response to a history questionnaire were obtained for each dog. Plasma von Willebrand factor concentration was measured by use of an ELISA, and genotyping was performed via polymerase chain reaction for 1 intragenic and 1 extragenic von Willebrand factor marker. Amplification product size was determined by use of polyacrylamide gel electrophoresis (intragenic marker) or automated sequence analysis (extragenic marker). Western blots were prepared from a subset of dogs with low plasma von Willebrand factor concentration to evaluate multimer distribution. RESULTS: Strong associations were detected between plasma von Willebrand factor concentration and von Willebrand factor marker genotype. Twenty-five dogs had substantial reduction in plasma von Willebrand factor concentration and multiple hemorrhagic events. All were homozygous for a 157-base-pair intragenic marker allele and homozygous or compound heterozygous for 1 of 4 extragenic marker alleles. These marker genotypes were exclusively detected in dogs with low plasma von Willebrand factor concentration, although some dogs with these genotypes did not have abnormal bleeding. CONCLUSIONS AND CLINICAL RELEVANCE: Type-1 von Willebrand disease in Doberman Pinschers is associated with the von Willebrand factor gene locus; however, the expression pattern in this breed appears more complex than that of a simple recessive trait.     

Artifactual changes in PCV, hemoglobin concentration, and cell counts in bovine, caprine, and porcine blood stored at room and refrigerator temperatures

BACKGROUND: Blood samples collected from farm animals for hematology testing may not reach the laboratory or be examined immediately upon collection, and in some cases may need to be transported for hours before reaching a laboratory. OBJECTIVE: The objective of this study was to investigate the artifactual changes that may occur in PCV, hemoglobin (Hgb) concentration, and cell counts in bovine, caprine, and porcine blood samples stored at room (30 degrees C) or refrigerator (5 degrees C) temperature. METHODS: Baseline values for PCV, Hgb concentration, and RBC and WBC counts were determined immediately after blood collection from 36 cattle, 32 goats, and 48 pigs using manual techniques. Blood samples were split into 2 aliquots and stored at 30 degrees C or 5 degrees C. Hematologic analyses were carried out at specified intervals during 120 hours of storage. Results were analyzed by repeated measure ANOVA; results at different temperatures were compared by paired t-tests. RESULTS: Compared to baseline values, there were no significant changes in Hgb concentration, RBC count, or WBC count in samples from cattle; in Hgb concentration and RBC count in samples from goats; and in Hgb concentration and WBC count in samples from pigs throughout the 120 hours of storage at both 30 degrees C and 5 degrees C. Significant changes (P <.05) from baseline occurred in PCV after 14 hours of storage at 30 degrees C and after 19 hours of storage at 5 degrees C in cattle and goats; and after 10 hours of storage at 30 degrees C and 14 hours of storage at 5 degrees C in pigs. Significant changes also were observed in Hgb concentration at 96 hours at 30 degrees C and 5 degrees C, and in RBC counts at 48 hours at 30 degrees C and 96 hours at 5 degrees C in porcine samples; and in total WBC counts at 120 hours at 30 degrees C and 5 degrees C in caprine samples. Artifactual changes were more pronounced in the samples stored at 30 degrees C. CONCLUSIONS: At both 30 degrees C and 5 degrees C, blood samples from cattle and goats can be stored for up to 12 hours, while blood samples from pigs can be stored for up to 8 hours without any significant changes in PCV. Blood samples from all 3 species can be stored for more than 24 hours without significant changes in Hgb concentration, RBC count, and total WBC count.     

Effect of storage on measurement of ionized calcium and acid-base variables in equine, bovine, ovine, and canine venous blood.

The stability of blood ionized calcium (Ca2+) and acid-base variables in equine, bovine, ovine, and canine venous blood samples (n = 15, in each group) stored at 4 C for 3, 6, 9, 24, or 48 hours was studied. Variables included blood Ca2+ and standard ionized calcium (Ca2+ corrected to pH 7.4) concentrations, pH, blood carbon dioxide and oxygen tensions, base excess, bicarbonate concentration, and total carbon dioxide content. Results indicate that storage of blood samples at 4 C for up to 48 hours, despite appreciable acid-base changes, is associated with less than 1.5% change in equine, bovine, and ovine blood Ca2+ concentrations. Similar changes were observed in canine blood during the first 9 hours' storage. After 24 and 48 hours' storage, clinically relevant decrease (10.5 and 15.5%) in canine blood Ca2+ concentration was measured. Therefore, Ca2+ concentration in equine, bovine, and ovine venous blood samples stored up to 48 hours, and in canine blood samples stored up to 9 hours at 4 C is of diagnostic use.     

Effect of serum storage, anti-inflammatory oral doses of prednisone, and spontaneous hyperadrenocorticism on serum glutamate dehydrogenase activity in dogs

BACKGROUND: Glutamate dehydrogenase (GLDH) is a mitochondrial enzyme with highest activity in periacinar hepatocytes. It is reported to be a sensitive indicator of hepatic injury; however, results of studies regarding tissue specificity are contradictory. OBJECTIVES: The purpose of the study reported here was to examine the effect of 3 factors on serum GLDH activity in dogs: serum storage, anti-inflammatory oral doses of prednisone, and spontaneous hyperadrenocorticism (HAC). METHODS: Stability of enzyme activity was determined by comparing serum samples stored at approximately 20 degrees C, 4 degrees C, and 20 degrees C for 4, 24, 48, and 72 hours, 1 week, and 6 months. To determine whether orally administered prednisone affected GLDH activity, the median difference in serum GLDH activity was compared between 5 untreated control dogs and 8 dogs that had received a tapering oral dose of prednisone. Lastly, GLDH enzyme activity was compared between 17 dogs with HAC and 16 age-matched controls. RESULTS: GLDH activity remained stable for 48 hours, 1 week, and 6 months, in serum stored at approximately 20 degrees C, 4 degrees C, and 20 degrees C, respectively. The median change in GLDH activity was not significantly different between dogs receiving prednisone and controls; however, dogs with HAC had significantly higher values than those of age-matched controls. CONCLUSIONS: Serum samples should be maintained at 4 degrees C if analysis of GLDH activity will be delayed by >48 hours; serum stored at 20 degrees C yields reliable results for up to 6 months. Serum GLDH activity was not increased in most dogs receiving short-term, anti-inflammatory oral doses of prednisone, in contrast to its increased activity in dogs with HAC.     

Stability of sorbitol dehydrogenase activity in bovine and equine sera

Serum sorbitol dehydrogenase (SDH) activities in 10 cows and nine horses were measured using an automated clinical analyzer. The serum samples were divided into aliquots that were stored at room temperature (21 degrees C), refrigerated (0-5 degrees C), or frozen (-30 degrees C). The stability of the SDH activity was monitored at various intervals. SDH activity in bovine sera remained stable for at least 5 hours at room temperature, 24 hours refrigerated, and 72 hours frozen without any significant (p < 0.05) differences from the initial serum values. In equine sera, SDH activity remained stable for at least 5 hours at room temperature and 48 hours frozen. The activity of the refrigerated equine sera was stable for at least 5 hours but less than 24 hours. An evaluation of fresh bovine serum and heparinized plasma samples indicated that there was no significant difference (p < 0.05) between the two sampling methods and that either may be employed for automated measurement of SDH activity following the established protocol. Sample type comparison indicated that there was a small but statistically significant (p < 0.05) difference between the results obtained comparing fresh serum and heparinized plasma samples for the horse. A reference range for Holstein cows was established using sera from 71 clinically healthy cattle (mean -/+ 2 SD = 32 -/+ 26 U/L).     

The effectiveness of gentamicin or polymyxin B for the control of bacterial growth in equine semen stored at 20 degrees C or 5 degrees C for up to forty-eight hours.

Semen from three stallions was used to evaluate the effectiveness of two antibiotics added to semen extender for samples stored at 20 degrees C or 5 degrees C for up to 48 hours. Each ejaculate was divided into six different treatments: semen+extender (SE); SE+gentamicin (100 micrograms/mL); SE+polymyxin B (1000 units/mL); and each of the above treatments inoculated with Pseudomonas aeruginosa ATCC 27853. Sampling of diluted semen for bacteriological analysis was performed after 2, 8, 24 and 48 hours of preservation at either temperatures. The presence of nonspecific bacteria was noted after two hours in all SE aliquots. The number of bacteria did not change in samples stored at 5 degrees C, while in samples preserved at 20 degrees C, it increased by three to four times after 48 hours. In semen aliquots treated with either of the antibiotics, the number of nonspecific bacteria was very low after two and eight hours at both temperatures. This number remained stable up to 48 hours at 5 degrees C, while an increase was noted at 24 and 48 hours at 20 degrees C. At 5 degrees C, the number of P. aeruginosa cells tended to decrease between 24 and 48 hours in SE aliquots. The presence of gentamicin or polymyxin B appeared to rapidly inhibit growth of P. aeruginosa. At 20 degrees C, growth of P. aeruginosa increased between 8 and 24 hours in SE, while the presence of antibiotics almost completely inhibited the growth of the bacterium.(ABSTRACT TRUNCATED AT 250 WORDS)     

Submitting canine blood for prothrombin time and partial thromboplastin time determinations

Practitioners commonly submit samples from dogs for partial thromboplastin time and prothrombin time determinations. Controversy exists as to the necessity for rapid separation of plasma and cells, and submission of the plasma on ice (or frozen). The purpose of this study was to address three questions. First, is it better to submit plasma or is whole blood satisfactory? Second, is it necessary to refrigerate the sample or is maintenance at room temperature (20° C) adequate? Third, does the sample have to arrive at the laboratory within a few hours of collection or can reliable partial thromboplastin time/prothrombin time determinations be made on samples up to 48 hours old?It has been shown by this study that reliable partial thromboplastin time and prothrombin time determinations can be carried out on canine plasma for up to 48 hours after collection regardless of whether or not the plasma is separated immediately; however the samples must be kept at 4°C. If the samples are maintained at room temperature, reliable prothrombin time determinations can be obtained for up to six hours after collection regardless of whether or not the plasma is separated immediately. Reliable partial thromboplastin time determinations can be made on plasma stored at 20°C for up to 24 hours after collection and possibly longer (up to 48 hours) if the plasma has been separated immediately.