Materials and Methods

Animals—The protocol was approved by the Institutional Animal Care and Use Committee at the University of California, Davis. Ten adult healthy staffowned dogs were studied. Dogs were determined to be healthy on the basis of history, physical examination findings, and CBC determination. Dogs had no history or evidence of recent illness or any chronic medical condition. Dogs were excluded if nonsteroidal antiinflammatory medications, propofol, synthetic colloids, or antimicrobials had been administered within the previous 2 weeks. Dogs were excluded if they had an Hct of < 35%, a platelet count of < 150,000 platelets/μL, or an abnormal leukogram.

Procedure—Fifteen milliliters of venous blood was obtained from each dog and placed as 1.5-mL aliquots into tubes containing 3.8% trisodium citrate and maintained at room temperature (25ºC). Baseline (ie, from undiluted blood samples) closure time was measured in duplicate by use of the bench-top platelet function analyzer^c and collagen and ADP-coated cartridges. Whole blood samples were diluted with saline solution, HES 600/0.7, or HES 670/0.75 at a ratio of 1:9 (1 part saline solution or colloid to 9 parts whole blood) and 1:3 (1 part saline solution or colloid to 3 parts whole blood). The HES 600/0.7 had a mean molecular weight of 600 kd and a degree of substitution of 0.7. The HES 600/0.7 contained 6 g of hetastarch; contained 0.9 g of 0.9% NaCl in water, NaOH, sodium (154 mEq/L), and chloride (154 mEq/L); and had an osmolality of 308 mOsm/L and a pH value of 5.5 (range, 3.5 to 7.0). The HES 670/0.75 had a mean molecular weight of 670 kd; had a degree of substitution of 0.75; and contained 6 g of hetastarch, sodium (143 mEq/L), chloride (124 mEq/L), lactate (28 mEq/L), calcium (5 mEq/L), potassium (3 mEq/L), and magnesium (0.9 mEq/L). All analyses were completed within 4 hours of blood sample collection. Samples associated with error readings were discarded and repeated. Duplicate measurements varying more than 17% were discarded.

Statistical analysis—Duplicate measurements were averaged and mean ± SD values were determined for each measurement.^d Normality of residuals and equal-variance tests were used to assess whether data were normally distributed. Change in closure time from baseline value for each fluid was evaluated by a repeated-measures ANOVA,^d applying the GeisserGreenhouse-Epsilon and Box-Epsilon probability adjustment with the dilution as the within-factor variable. When a significant effect was identified, the Bonferroni all-pairwise multiple comparison test was used to identify the significant differences.^d Differences in closure times between each fluid at each dilution were evaluated by a repeated-measures ANOVA applying the Geisser-Greenhouse-Epsilon and Box-Epsilon probability adjustment^d with fluid types as the within-factor variable. When a significant effect was identified, a paired t test was used to determine the significant differences. A value of P < 0.05 was considered significant.

Results

Closure times—The Hct (mean, 48.1 ± 4.1%; range, 39% to 54%) and platelet count (mean, 286,000 ± 74,129 platelets/μL) were within reference ranges for all dogs. Mean baseline closure time was 68.0 ± 15.3 seconds (range, 53 to 93 seconds). The 1:9 dilution of blood with saline solution, HES 600/0.7, and HES 670/0.75 resulted in closure times of 71.5 ± 10.6 seconds, 76.4 ± 10.7 seconds, and 76.8 ± 15.5 seconds, respectively. The 1:3 dilution of blood with saline solution, HES 600/0.7, and HES 670/0.75 resulted in closure times of 85.8 ± 15.7 seconds, 100.6 ± 18.6 seconds, and 101.8 ± 16.2 seconds, respectively. All data were normally distributed.

Effect of dilution—Closure time resulting from a 1:9 dilution of blood with saline solution (71.5 ± 10.6 seconds) was not significantly different from baseline (68.0 ± 15.3 seconds). Closure time resulting from a 1:3 dilution of blood with saline solution (85.8 ± 15.7 seconds) was significantly (P = 0.001) different from baseline (68.0 ± 15.3 seconds) and from closure time resulting from a 1:9 dilution of blood with saline solution (71.5 ± 10.6 seconds). Closure times resulting from a 1:9 dilution of blood with HES 600/0.7 (76.4 ± 10.6 seconds) or HES 670/0.75 (76.8 ± 15.5 seconds) were not significantly different from baseline (68.0 ± 15.3 seconds). The 1:3 dilutions of blood (HES 600/0.7, 100.6 ± 18.6 seconds; HES 670/0.75, 101.8 ± 16.2 seconds) resulted in closure times that were significantly (P = 0.002) different from baseline (68.0 ± 15.3 seconds) and from 1:9 dilutions of blood (HES 600/0.7, 76.4 ± 10.7 seconds; HES 670/0.75, 76.8 ± 15.5 seconds; Figure 1).

Mean ± SD closure time as an indication of canine platelet function versus fluid type (saline solution [striped bar], HES 600/0.7 [white bar] and HES 670/0.75 [black bar]) used for various dilutions (baseline [no dilution; gray bar], 1:3 dilution and 1:9 dilution) of whole blood samples from 10 dogs. *Significant (P < 0.05) dilutional effect, compared with baseline value and 1:9 dilution with the same fluid type. †Significant (P < 0.05) colloid effect, compared with saline solution at the same dilution.

Citation: American Journal of Veterinary Research 68, 6; 10.2460/ajvr.68.6.605

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Mean ± SD closure time as an indication of canine platelet function versus fluid type (saline solution [striped bar], HES 600/0.7 [white bar] and HES 670/0.75 [black bar]) used for various dilutions (baseline [no dilution; gray bar], 1:3 dilution and 1:9 dilution) of whole blood samples from 10 dogs. *Significant (P < 0.05) dilutional effect, compared with baseline value and 1:9 dilution with the same fluid type. †Significant (P < 0.05) colloid effect, compared with saline solution at the same dilution.

Citation: American Journal of Veterinary Research 68, 6; 10.2460/ajvr.68.6.605

• Download Figure
• Download figure as PowerPoint slide

Mean ± SD closure time as an indication of canine platelet function versus fluid type (saline solution [striped bar], HES 600/0.7 [white bar] and HES 670/0.75 [black bar]) used for various dilutions (baseline [no dilution; gray bar], 1:3 dilution and 1:9 dilution) of whole blood samples from 10 dogs. *Significant (P < 0.05) dilutional effect, compared with baseline value and 1:9 dilution with the same fluid type. †Significant (P < 0.05) colloid effect, compared with saline solution at the same dilution.

Citation: American Journal of Veterinary Research 68, 6; 10.2460/ajvr.68.6.605

• Download Figure
• Download figure as PowerPoint slide

Effect of fluid type—Closure times resulting from 1:9 dilutions of blood were not significantly different between the 3 fluid types. A significant (P = 0.03) difference was found in closure time between 1:3 dilutions of blood with saline solution (85.8 ± 15.7 seconds) and with HES 670/0.75 (101.8 ± 16.2 seconds). Closure times resulting from a 1:3 dilution of blood with HES 600/0.7 (100.6 ± 18.6 seconds) and HES 670/0.75 (101.8 ± 16.2 seconds) were not significantly different from each other (Figure 1).

Discussion

In this study, we evaluated the effects of saline solution and 2 commonly used synthetic colloids, HES 600/0.7 and HES 670/0.75, on platelet function in vitro. The 1:9 and 1:3 dilutions simulated a 10 mL/kg and 30 mL/kg dose, respectively, of fluids assuming no in vivo redistribution or excretion. Saline solution and the 2 synthetic colloids significantly prolonged closure times at the 1:3 dilution. No significant difference was found between the 2 HES preparations at any dilution. The prolongation of closure time associated with saline solution is attributed to simple dilution of coagulation precursors.^3,5,6,8,13 In the present study, though, the prolonged closure time for saline solution at the 1:3 dilution was still within the reported reference range for clinically normal dogs (saline solution, 85.8 seconds; reference range, 52 to 86 seconds).^15,22,24

Although the manufacturer of the platelet function analyzer will only validate the closure time as a measure of platelet function for an Hct of > 35% and platelet count of > 150,000 platelets/μL, there have been numerous human studies^18,20,25,29 and a veterinary study¹⁵ that have shown the analyzer to be accurate for Hcts as low as 25% and platelet counts as low as 100,000 platelets/μL. Closure times measured on blood with a > 30% Hct were consistently reported to be a reliable evaluation of platelet function.^15,18,25,29 In the present study, it can be calculated that the greatest dilution (1:3) would be associated with Hcts in the range of 29% to 40% with a mean of 36%. The range of platelet counts associated with the 1:3 dilution would be 120,000 to 316,000 platelets/μL with a mean of 214,000 platelets/μL. Consequently, the closure time measured on these diluted samples can be considered as an accurate evaluation of platelet function.

The closure time at the highest dilution (1:3) for HES 670/0.75 was significantly higher than for dilution with saline solution. This is attributed to a platelet dysfunction effect over and above simple dilution.^{1–4,6,9–11,13,14} Hydroxyethyl starch preparations are thought to bind directly to the glycoprotein IIb/IIIa receptor on the platelet surface, which inhibits the binding of fibrinogen to the receptor. The inability of fibrinogen to bind to the receptor prevents outside-to-inside signaling on the platelet membrane, platelet upregulation and, therefore, the eventual formation of the fibrin meshwork required for the formation of the platelet plug and a clot.^{1–4,6,9–11,13,14}

Multiple studies have evaluated various HES preparations and found that high–molecular-weight products (> 400 kd) and some low– (< 200 kd) and medium– (200 to 400 kd) molecular-weight solutions inhibit platelet function.^2,13 The degree of substitution is theorized to play a more substantial role in platelet inhibition than the mean molecular weight of the HES preparation.¹ Degree of substitution refers to the number of hydroxyethyl groups substituted for a hydroxyl group per unit of glucose.³⁰ Hydroxyethyl starch preparations with a higher molecular weight or a higher degree of substitution may be correlated with more platelet dysfunction.^1–4,13 The high molecular weights and high degree of substitution of HES 600/0.7 and HES 670/0.75 would be expected to influence platelet function considerably.

We expected HES 600/0.7 to decrease platelet function in dogs based on the multiple studies in people.^2–4,10,13 We proposed, though, that platelet function would not be as affected by HES 670/0.75, compared with HES 600/0.7, on the basis of a recent study that found increased platelet reactivity with HES 670/0.75.⁴ The increased platelet reactivity with HES 670/0.75 was proposed to result from the calciumcontaining solvent in the solution; calcium may increase the platelet reactivity to activation by increasing the intracellular calcium concentrations, theoretically overcoming the HES receptor binding effects.⁴ Deusch et al⁴ reported that platelet reactivity increased with HES 670/0.75 and the solvent of HES 670/0.75, which contains calcium. Other studies,^1–5,13,14 though, have not confirmed this finding, and have found a similar degree of impaired platelet function with HES 670/0.75 to other HES preparations. In the present study, HES 670/0.75 affected platelet function in a manner similar to HES 600/0.7. The increases in closure times over the established reference range are not known to be predictive for bleeding tendencies. Further research is needed in clinical patients comparing the values of the closure times along with more detailed platelet receptor analysis for the assessment of bleeding tendencies in the surgical setting.

The method of hemostatic analyzation may affect the study outcome. There are many different platelet functions analyzers including flow cytometry, platelet aggregometry, the bench-top platelet function analyzer, and thromboelastography. The bench-top platelet function analyzer was used in this study because it generates highly reliable and repeatable data and is easy to use in the clinical setting. The bench-top platelet function analyzer activates platelets under conditions of high shear stress in a capillary tube and stimulates adhesion and aggregation of platelets by use of collagen and ADP.^18,26,28 The bench-top platelet function analyzer is the only method using high shear stress for platelet activation and, therefore, is thought to best mimic in vivo physiologic conditions.^18,26,28 Flow cytometry can evaluate platelet activation by the binding of fluorescently labeled monoclonal antibodies to platelet surface antigens with or without the addition of exogenous platelet agonists (thrombin, collagen, thromboxane A₂, and ADP).^26,31 Platelet aggregometry is used to determine interactions of platelet agonists with either whole blood or platelet-rich plasma. The strong (collagen, arachidonate, and thrombin) or weak (ADP, epinephrine, and ristocetin) platelet agonists may stimulate platelet activation that is detected by changes in light absorbance or transmittance.^26,32 Thromboelastography is used to evaluate the clot strength and rate of clot formation. Whole blood is placed in a cup with a pin suspended in the cup. Calcium is added to promote coagulation as the cup is oscillated, and fibrin strands form between the cup and pin. The pin then begins to move with the cup, and measurements are taken from the changes in movement of the pin.³³

The various methods to analyze platelet function indicate that each method has strengths and weaknesses. The platelet function analyzer does not evaluate the molecular platelet components related to activation. Platelet aggregometry may not replicate physiologic conditions as a result of the use of platelet-rich plasma. The requirement of the thromboelastography for additional calcium may affect the evaluation of HES 670/0.75, compared with other HES preparations. Because the mechanism of the platelet upregulation of HES 670/0.75 is theorized to be caused by the calcium contained in the solvent, the addition of calcium to the starch solutions for testing may not allow for differentiation of the variable effects of this solution.

In conclusion, our findings did show that HES 600/0.7 and HES 670/0.75 prolonged closure times in vitro beyond a dilutional effect. Our findings did not support our hypothesis that the magnitude of platelet dysfunction associated with HES 600/0.7 would exceed that of HES 670/0.75 in vitro. These results suggest that the choice between these 2 colloids makes no difference with regard to platelet dysfunction. In vivo evaluation of platelet function after the administration of synthetic colloids is warranted.