All sea turtle species in the waters of the US are listed as threatened or endangered under the Endangered Species Act due to a diverse array of threats, including habitat degradation, fisheries interactions, boat strikes, oil spills, and extreme weather events (eg, cold stunning).1 Consequently, rescue and rehabilitation efforts are undertaken for these long-lived and imperiled species. Stranded sea turtles admitted to rehabilitation may present with severe anemia for which a whole blood transfusion is indicated. Possible causes of anemia may include fisheries interactions, boat strikes, shark predation, and chronic debilitation.2,3 Additionally, perioperative hemorrhage from fibropapilloma tumor removal or other surgical procedures is possible during rehabilitation.2,3 Despite the clinical benefit of whole blood or packed RBC (pRBC) transfusions to help treat severe hemorrhage or anemia, transfusion research in sea turtles is sparsely represented in the literature.3–5
Previous sea turtle transfusion research has explored pretransfusion blood product storage and compatibility testing. Sodium heparin is the preferred anticoagulant for the collection and short-term storage of whole blood in loggerhead sea turtles (Caretta caretta) given that it resulted in fewer clinicopathological changes over 24 hours when compared to sodium citrate and citrate phosphate dextrose adenine.4 Regarding pretransfusion testing, major and minor crossmatches incubated at room temperature for 30 minutes were found optimal for detecting crossmatch incompatibilities in green (Chelonia mydas) and loggerhead sea turtles.3 In vitro testing of allogeneic whole blood transfusions among green sea turtles found only 54% were compatible by major and minor crossmatch, whereas in vitro testing of whole blood xenotransfusions between loggerhead and green sea turtles found all were incompatible by major and minor crossmatch.3 To the authors’ knowledge, there are no case series or prospective studies detailing in vivo blood product administration in sea turtles nor any subsequent associated transfusion reactions despite transfusions being reported anecdotally.2
Blood transfusion filters remove aggregates of cellular debris, platelets (or thrombocytes), WBCs, or fibrin that form during blood collection and storage. Consequently, the use of a blood filter during transfusion administration is considered a standard of care in human and veterinary medicine. In human medicine, commercially available standard blood filters contain pore sizes ranging from 170 to 260 µm. In domestic canine and feline patients, the Hemo-Nate blood filtration system (Utah Medical Products Inc), a disposable, stainless-steel filter with 18-µm pores designed for human neonatal blood transfusions, is frequently used as a postsyringe, inline microaggregate blood filter. The Hemo-Nate blood filtration system is advantageous in nondomestic veterinary patients, given its low priming volume (0.7 mL) and versatility to function as a postsyringe, inline filter without an extension set. Hemo-Nate blood filters could improve the safety of whole blood or pRBC transfusions in sea turtles; however, the filter’s 18-µm pores may cause unnecessary postfiltration hemolysis given the average diameter of RBCs in sea turtles ranges from 11 to 20 µm (Table 1). Hemolysis and the resulting free hemoglobin risk inducing oxidative stress, with potentially detrimental effects on renal, myocardial, and vascular systems; moreover, transfusion of hemolyzed blood carries an increased risk of adverse transfusion reactions.6
Compiled average RBC morphology for 6 of the 7 extant sea turtles from the present study and data available in prior peer-reviewed references.
| Sea turtle species | Maximum diameter (µm) | Minimum diameter (µm) | Cell area (µm2) | Reference |
|---|---|---|---|---|
| Green (Chelonia mydas) | 20 | 17 | 267 | Work et al11 |
| Hawksbill (Eretmochelys imbricata) | 20.45 | 13.74 | 221.19 | Zhang et al40 |
| Leatherback (Dermochelys coriacea) | 19.7 | 13.4 | 207 | Present study |
| Loggerhead (Caretta caretta) | 19.05 | 12.85 | 194.28 | Casal and Orós12 |
| Kemp’s ridley (Lepidochelys kempii) | 16.1 | 11.4 | 143.7 | Present study |
| Olive ridley (Lepidochelys olivacea) | 20.35 | 12.55 | 201.65 | Zhang et al13 |
To the authors’ knowledge, there are no reports of RBC morphology for flatback sea turtles (Natator depressus).
This study’s objectives were 1) to determine the hemolytic effect of Hemo-Nate blood filters for in vitro sea turtle whole blood transfusions, and 2) describe the average diameter of leatherback (Dermochelys coriacea) and Kemp’s ridley sea turtle (Lepidochelys kempii) RBCs adding to previously reported RBC morphometrics in other sea turtle species. We hypothesized there would be no significant hemolysis in postfiltration sea turtle plasma samples given Hemo-Nate blood filters do not cause significant postfiltration hemolysis in domestic chickens (Gallus domesticus) and American alligators (Alligator mississippiensis).7,8
Methods
Sea turtles and blood sampling
Five green, 5 loggerhead, and 5 Kemp’s ridley sea turtles were sourced from patients undergoing rehabilitation at the Karen Beasley Sea Turtle Rescue and Rehabilitation Center in Surf City, NC, between March and July 2023 or captured in July 2023 in actively fished pound nets in Core Sound near Cape Lookout National Seashore in conjunction with a concurrent sea turtle sex-determination study (L. Avens, PhD, Southeast Fisheries Science Center, NOAA Fisheries, unpublished data, 2023). Sea turtle age class ranged from juvenile to adult across the 3 species; median (minimum-maximum) weight for sampled green, loggerhead, and Kemp’s ridley sea turtles was 4.8 kg (3.4 to 6.8 kg), 30.0 kg (23.0 to 138.1 kg), and 3.7 kg (2.4 to 31.5 kg), respectively. All sampled sea turtles were deemed clinically healthy based on physical exam. An a priori power analysis in the statistical software R (v4.3.1)9 using the pwr package (v1.3.0)10 and its dependents determined a sample size of approximately 15 sea turtles would be sufficient to achieve a power of 80% given an alpha level of 0.05 (2-sided), mean difference of 3.2 mg/dL free plasma hemoglobin between pre- and postfiltration samples, and a SD of the differences of 4 mg/dL. The mean difference and SD of the differences were estimated using previously published data on American alligators.8
Six milliliters of venous blood was collected aseptically from the dorsal cervical sinus of each sampled sea turtle using either 3.81-cm (1.5-inch) 18-gauge or 2.54-cm (1-inch) 21-gauge needles and a plastic syringe containing a precalculated ratio of 25 U/mL sodium heparin based upon previous research.4 All samples were gently rolled 8 to 10 times to mix the anticoagulant solution and blood following phlebotomy. Sea turtle blood sampling was approved by the North Carolina State University IACUC (No. 22-457). Sea turtle rehabilitation at the Karen Beasley Sea Turtle Rescue and Rehabilitation Center is performed under a permit from the North Carolina Wildlife Resources Commission (No. 23ST05). Pound netting in Core Sound was performed under a permit from the National Marine Fisheries Service (No. 21233).
Blood filtration and data collection
Immediately after phlebotomy, the needle was removed from the syringe containing the heparinized blood sample and replaced with a microbore T-port extension set (B. Braun) cut proximal to the male connector (T-port) at a 45° angle to its lumen to allow for nonturbulent blood flow. The syringe was then placed in a calibrated Triumph SP12 Syringe Pump (Triumph Medical Services) set to 60 mL/h. A standardized syringe pump infusion rate (60 mL/h) was elected to limit any confounding influence of infusion rate variability if based on patient weight. The infusion rate of 60 mL/h was chosen based on a recommended maximum transfusion rate in sea turtles of 5 mL/kg/h and a theoretical sea turtle patient weighing 12 kg. Blood was infused for 2 minutes, and a total of 2 mL was collected into a plastic microcentrifuge tube. This sample was labeled as “prefiltration” and gently agitated by hand at room temperature until processing. The microbore T-port extension set was then replaced with a Hemo-Nate blood filter, attached to a new microbore T-port extension set cut as described earlier. Blood was then infused for another 2 minutes at the same rate, and another 2 mL of blood was collected into a plastic microcentrifuge tube. This sample was labeled as “postfiltration” and gently agitated at room temperature until processing.
For both pre- and postfiltration samples, PCV and total solids (TS) were measured by micro-Hct centrifugation and a calibrated refractometer. An i-STAT CHEM8+ (Abbott Laboratories) biochemistry panel was run on the pre- and postfiltration samples within 30 minutes of infusion using an i-STAT Alinity v point-of-care analyzer (Zoetis Inc) following manufacturer instructions. The i-STAT CHEM8+ biochemistry panel measures sodium, potassium, chloride, TCO2, anion gap, ionized calcium, glucose, BUN, creatinine, Hct, and hemoglobin. Pre- and postfiltration samples were centrifuged at 1,000 X g for 10 minutes. Free plasma hemoglobin in pre- and postfiltration samples was measured using the HemoCue Plasma/Low Hb System (HemoCue America). After centrifugation, a drop of plasma was transferred to waxed paper using a plastic disposable transfer pipette, and HemoCue Plasma/Low Hb System microcuvettes were filled and analyzed following manufacturer instructions. The HemoCue Plasma/Low Hb System is a portable photometer designed to assess human hemoglobin concentrations quickly and accurately from 0 to 3,000 mg/dL (0 to 3 g/dL) within a sample of 20 µL of plasma, serum, aqueous solutions, or blood products. Before data collection, the HemoCue Plasma/Low Hb System was calibrated following manufacturer instructions.
RBC morphology
Archived blood smears previously stained with a commercial Romanowsky stain variant (Hemacolor Stain Set; Sigma-Aldrich) following manufacturer’s instructions from leatherback and Kemp’s ridley sea turtles were utilized to determine average maximum and minimum diameter as well as the cell area of their RBCs. Blood smears were made using blood samples collected from free-ranging leatherback sea turtles captured during health assessments in North Carolina waters and Kemp’s ridley sea turtles who had undergone rehabilitation at the Karen Beasley Sea Turtle Rescue and Rehabilitation Center. With the use of light microscopy, blood smears from 5 individuals of both species were examined with a calibrated ocular micrometer using 400X magnification, and the maximum and minimum diameters of 20 RBCs were measured from each individual for a total of 100 RBCs per species similar to previous sea turtle hematology studies.11,12 RBC area was calculated according to the following formula as previously described as (maximum diameter X minimum diameter X π)/4.13 Historical sea turtle blood sampling of leatherback and Kemp’s ridley sea turtles was performed under appropriate North Carolina Wildlife Resources Commission (Nos. 05ST05, 07ST05, and 09ST05) and National Marine Fisheries Service (No. 21233) permits.
Statistical analyses
Pre- and postfiltration PCV, TS, i-STAT CHEM8+, free plasma hemoglobin, and RBC morphology results were summarized using descriptive statistics. The dependent variables PCV, TS, sodium, chloride, potassium, glucose, i-STAT Hct, and free plasma hemoglobin from pre- and postfiltration samples as well as their respective differences were assessed for normality. Given their non-Gaussian distribution, the difference in dependent variables between pre- and postfiltration samples was compared between sea turtle species included in this study using a Kruskal-Wallis test, which found no significant difference among the 3 species. Subsequently, results were aggregated regardless of species and dependent variables were analyzed by comparing paired pre- and postfiltration samples using a Wilcoxon signed-rank test. Diagnostic agreement between PCV as measured by micro-Hct centrifugation and Hct as measured by the i-STAT Alinity v point-of-care analyzer was assessed using Bland-Altman methods14 and Passing-Bablok regression.15,16 For Bland-Altman plots and statistics, bias was calculated by subtracting the PCV result from the i-STAT Hct result. The mean bias and the limits of agreement (LOA) were then calculated; upper and lower LOA were calculated as follows:
LOA = Meandifference ± (1.96 X SDdifference)
Bias was considered statistically significant if the mean bias’s 95% CI did not include 0. For Passing-Bablok regression, constant bias was present if the 95% CI for the y intercept did not include 0, whereas proportional bias was present if the 95% CI for the slope did not include 1. All analyses were performed in the statistical software R (v4.3.1).9 Data organization and summary statistics were performed using the dplyr (v1.1.2)17 and psych (v2.3.6)18 packages and their dependents. Bland-Altman methods and Passing-Bablok regression were performed using the blandr (v0.5.3)19 and mcr (v1.3.2)20 packages and their dependents. To account for multiple comparisons, Wilcoxon signed-rank test P values were adjusted using the Bonferroni correction. Null hypotheses were rejected if adjusted Wilcoxon signed-rank test P < .05.
Results
Morphometrics of leatherback and Kemp’s ridley sea turtle RBCs are presented along with previously reported values for other sea turtle species (Table 1). Prefiltration PCV, TS, and plasma biochemistry results were within reported ranges for the 3 sea turtle species included in this study (Table 2).21 There were no significant differences between pre- and postfiltration samples for TS, sodium, chloride, potassium, glucose, i-STAT Hct, and free plasma hemoglobin; however, there was a significant difference between pre- and postfiltration samples for PCV (adjusted P value = .029). Bland-Altman methods found the i-STAT Alinity v had significant negative bias (ie, underestimates) when Hct was measured compared to PCV with a mean bias of −5% (95% CI, −7% to −4%) and lower and upper LOA of −12% (95% CI, −15% to −10%) and 2% (95% CI, −1% to 4%), respectively (Figure 1). Passing-Bablok regression analysis found an intercept of 0% (95% CI, −7.9% to 3.7%) and a slope of 1 (95% CI, 0.7% to 1.1%; Figure 2) indicating the i-STAT Alinity v had no significant constant or proportional bias when measuring Hct.
Pre- and postfiltration PCV, total solids (TS), select plasma biochemical analytes, and free plasma hemoglobin (Hgb) from loggerhead (n = 5), Kemp’s ridley (5), and green sea turtles (5).
| Prefiltration | Postfiltration | ||||||
|---|---|---|---|---|---|---|---|
| Analyte | Mean | Median | (Minimum, maximum) | Mean | Median | (Minimum, maximum) | Adjusted P value |
| PCV (%) | 30 | 30 | (23, 43) | 27 | 26 | (18, 45) | .029* |
| iSTAT Hct (%) | 23 | 22 | (15, 34) | 23 | 22 | (15, 34) | .933 |
| TS (g/dL) | 5.0 | 5.0 | (2.6, 6.8) | 5.0 | 5.0 | (2.6, 6.8) | 1 |
| Sodium (mmol/L) | 148 | 148 | (142, 153) | 149 | 148 | (141, 154) | .693 |
| Potassium (mmol/L) | 3.5 | 3.4 | (2.6, 4.8) | 3.5 | 3.5 | (2.7, 4.8) | 1 |
| Chloride (mmol/L) | 105 | 107 | (89, 119) | 105 | 105 | (89, 119) | 1 |
| Anion gap (mmol/L) | 9 | 9 | (3, 17) | 9 | 9 | (−4, 16) | N/A |
| Ionized calcium (mmol/L) | 1.01 | 1.01 | (0.82, 1.22) | 1.00 | 1.00 | (0.79, 1.22) | N/A |
| Total CO2 (mmol/L) | 38 | 38 | (27, 50) | 38 | 38 | (29, 48) | N/A |
| Glucose (mg/dL) | 102 | 102 | (57, 165) | 103 | 103 | (58, 165) | 1 |
| BUN (mg/dL) | 68 | 66 | (3, 140) | 69 | 67 | (3, 140) | N/A |
| Creatinine (mg/dL) | 0.2 | 0.2 | (0.2, 0.6) | 0.2 | 0.2 | (0.2, 0.6) | N/A |
| Hgb (g/dL) | 7.9 | 7.7 | (5.8, 11.6) | 8.2 | 8.0 | (5.4, 11.6) | N/A |
| Plasma Hgb (mg/dL) | 19 | 10 | (0, 80) | 25 | 20 | (0, 90) | 1 |
Pre- and postfiltration samples were measured after infusing heparinized whole blood without and with a Hemo-Nate blood filter, a disposable stainless-steel filter with 18-µm pores, at 50 mL/h using a veterinary syringe pump. PCV and TS were measured by micro-Hct centrifugation and a refractometer. Plasma biochemical analytes, Hgb, and Hct were measured by an i-STAT Alinity v point-of-care analyzer. Free plasma hemoglobin was measured using the HemoCue Plasma/Low Hb System. To account for multiple comparisons, P values were adjusted using the Bonferroni correction.
Analytes that significantly differ pre- and postfiltration.
Bland-Altman plot of Hct measured by the i-STAT Alinity v point-of-care analyzer compared to PCV measured by micro-Hct centrifugation demonstrating a significant mean bias of −5% (95% CI, −7% to −4%). Black dots represent individual paired measurements. The mean difference (Bias, solid black line), 95% CI (red dashed lines), and the limits of agreement (black dotted-dashed lines), defined as the mean difference ± 1.96 times the SD of the differences, were calculated.
Citation: American Journal of Veterinary Research 85, 5; 10.2460/ajvr.23.12.0269
Passing-Bablok plot of Hct measured by the i-STAT Alinity v point-of-care analyzer compared to PCV measured by micro-Hct centrifugation. Passing-Bablok regression analysis found an intercept of 0% (95% CI, −7.9% to 3.7%) and a slope of 1 (95% CI, 0.7% to 1.1%) suggesting the i-STAT Alinity v had no significant constant or proportional bias when measuring Hct. Black dots represent individual paired measurements. The solid blue line and associated shading represent the estimated regression line and its 95% CI.
Citation: American Journal of Veterinary Research 85, 5; 10.2460/ajvr.23.12.0269
Discussion
Similar to morphology observed in other extant sea turtle species, this study found the average minimum-maximum diameter of leatherback sea turtle RBCs is 13.4 to 19.7 µm and the average minimum-maximum diameter of Kemp’s ridley sea turtle RBCs is 11.4 to 16.1 µm. Given sea turtle RBC maximum diameter may be ≥ 18 µm, there has been historical concern that microaggregate blood filters (eg, Hemo-Nate blood filter) may cause unnecessary and potentially harmful hemolysis during the administration of whole blood or pRBCs to a sea turtle recipient. While this study found a significant median postfiltration decrease in PCV of approximately 4% (ie, from 30% to 26% or a 13% decrease in the total RBCs transfused), there was no significant postfiltration change in free plasma hemoglobin to suggest hemolysis or any of the biochemical analytes evaluated. These results suggest that although the Hemo-Nate blood filter may not hemolyze transfused sea turtle RBCs, the 18-µm filter’s pores likely retain a small proportion of infused RBCs given their diameter.
While permutations exist, veterinary blood transfusions are delivered via 3 general administration setups: gravity flow using a blood administration set with a built-in blood filter, peristaltic pump infusion using a blood administration set with a built-in filter, and syringe pump infusion with a postsyringe, inline filter (eg, a Hemo-Nate blood filter). Due to necessary transfusion volume, gravity flow and peristaltic pump infusion are the most practical and prevalent setups in equine and large animal blood transfusions. However, blood transfusion administration setup varies in canine and feline patients due to hospital resources, standard operating procedures, as well as clinician preference. In terms of the current evidence base, one notable, in vivo experimental study22 found delivery using a syringe pump and Hemo-Nate blood filter was associated with the greatest loss of transfused RBCs after 24 hours posttransfusion, whereas subsequent in vitro canine and feline studies6,23 and 1 in vivo experimental feline study24 have found delivery using a syringe pump and Hemo-Nate blood filter have no significant impact on RBC survival posttransfusion. Although the aforementioned in vivo experimental canine study22 found delivery via gravity flow resulted in no loss of transfused RBCs after 24 hours posttransfusion, this method is potentially challenging due to its unpredictable delivery rate. Without vigilant staffing or drip infusion monitoring devices, rapid transfusion may result in transfusion-associated circulatory overload. Conversely, slow transfusion might be associated with increased blood contamination or failure to achieve the desired clinical endpoint.25 Syringe pump infusion with a Hemo-Nate blood filter provides the most customizable and accurate delivery, especially for small domestic and nondomestic veterinary patients.
It is unclear why no significant postfiltration hemolysis was observed in previous in vitro infusion studies in domestic chickens or American alligators and the present study in sea turtles despite these species having potentially larger diameter RBCs than the Hemo-Nate blood filtration system’s 18-µm pores. The lack of hemolysis observed in avian and reptile RBCs when using the Hemo-Nate filter could be due to their ability to deform, withstand shear stress, or multiple factors. Reptile RBCs have generally longer lifespans than mammalian RBCs and may need to be more resistant to hemolysis to account for longer circulation and survival.5 Ectothermic RBCs are more osmotically resistant to hemolysis than mammalian RBCs likely because they evolved to tolerate a wider preferred optimal temperature zone.26 Additionally, larger diameter RBCs are more osmotically resistant to hemolysis than smaller diameter RBCs.3,26 Although RBC osmotic resistance is not equivalent to mechanical resistance, it may correlate to the decreased mechanical fragility seen in ectothermic and larger diameter RBCs when infused through the Hemo-Nate filter. Given the apparent resistance of sea turtle RBCs to mechanical hemolysis and limited transfusion medicine research in nondomestic veterinary patients, extrapolation of the present study’s results may benefit other species, including elasmobranchs, terrestrial turtles, and lizards.27–29
The significant negative bias between PCV and Hct measured by the i-STAT Alinity v is not surprising given their underlying methodologies. Veterinary point-of-care analyzers measure Hct via whole blood conductometry in which a higher proportion of RBCs decreases electrical conductivity; subsequently, these analyzers likely underestimate sea turtle Hct due to differences in conductivity between nucleated reptilian RBCs and nonnucleated mammalian RBCs. Measurement of Hct performed by veterinary point-of-care analyzers have previously and repeatedly underestimated PCV in avian,30–32 terrestrial chelonian,33,34 and green, hawksbill (Eretmochelys imbricata), loggerhead, and Kemp’s ridley sea turtle patients.35,36 However, 1 previous study37 in Galápagos green sea turtles found no difference between PCV and veterinary point-of-care analyzer Hct. The discrepancy between the previous study and the present study may be due to differences in sample size or statistical analyses. In the present study, discordance between Bland-Altman methods and Passing-Bablok regression analysis for the assessment of constant bias is likely due to our small sample size (n = 30 paired PCV and i-STAT Hct measurements). Passing-Bablok regression analysis was performed as a robust, nonparametric method with less sensitivity to outliers to assess proportional bias which is not determined with Bland-Altman methods. While Passing and Bablok15,16 originally advised a minimum sample size of 30, Ludbrook38 suggests a minimum sample size of 50 to better narrow calculated 95% CIs. Future studies evaluating diagnostic agreement in sea turtles and other nondomestic veterinary patients should avoid inferential statistics (ie, demonstrating equivalency) and continue to judiciously employ agreement analyses that quantify constant and proportional bias.
Possible limitations in the present study include the i-STAT Alinity v and HemoCue Plasma/Low Hb System not being validated in sea turtles and the use of a standardized syringe pump infusion rate while using the Hemo-Nate blood filter. While the i-STAT Alinity v has not been formally validated in any sea turtle species, the analyzer and its predecessor (the i-STAT 1) are widely used in sea turtle rescue, rehabilitation, and research, and plasma biochemistry results were within reported ranges for the 3 sea turtle species. Given that the HemoCue Plasma/Low Hb System quantifies free hemoglobin using a modified azide-methemoglobin reaction with absorbance measured at 2 specific wavelengths (570 and 880 nm) optimized for human hemoglobin, species-specific absorbance characteristics of oxyhemoglobin and deoxyhemoglobin in sea turtles may affect its accuracy. While the HemoCue Plasma/Low Hb System is designed for very low hemoglobin concentrations, the analyzer demonstrates nonlinearity for hemoglobin concentrations between 0 and 30 mg/dL, which may have affected the accuracy of this study’s results. Given the variable weight of individual sea turtles admitted for rehabilitation depending on species and age class, a standardized syringe pump infusion rate (60 mL/h) was chosen a priori. Sea turtle whole blood transfusions are recommended to be initially delivered at a slow rate (0.25 mL/kg/h) for the first 30 minutes before increasing to 5 mL/kg/h for the remainder of the transfusion assuming no adverse transfusion reactions are observed (eg, apnea, tachypnea, arrhythmias, seizures, etc).2 Assuming a maximum transfusion rate of 5 mL/kg/h, the in vivo transfusion rates of the sea turtles sampled for this study could range from 12 to 690 mL/h (weight range, 2.4 to 138.1 kg), Based on the authors’ experience, 60 mL/h would likely be too fast of a transfusion for juvenile green and Kemp’s ridley sea turtles and a very slow transfusion rate for subadult to adult loggerhead sea turtles. Future investigations could explore the hemolytic effect, if any, of using the Hemo-Nate blood filter at the higher transfusion rates necessary for larger sea turtles.
Hemo-Nate blood filters are designed for small-volume transfusions; consequently, the manufacturer recommends a maximum filtration volume of 20 mL for pRBC transfusions and 50 mL for whole blood transfusions. If resistance is noted during transfusion (eg, an occlusion alarm), clinicians are advised to change the filter before continuing the transfusion. Interestingly, despite the Hemo-Nate blood filter being used most commonly (and arguably universally) postsyringe during blood transfusion administration in veterinary clinical practice and research, it is important to note this setup deviates from the recommendations provided by the manufacturer. Although the Hemo-Nate blood filter is bidirectionally supported, the manufacturer recommends the preferred approach for whole blood and pRBC transfusions is for the blood product to be aspirated through the filter (vs pushed through the filter) then transfused minus the filter postsyringe. Some have recommended prefiltering (aspirating through the filter) whole blood and pRBC transfusions using in-line filters with > 170-µm pores to avoid potential stress or damage caused to the RBCs by the Hemo-Nate blood filter39; however, what effect this has on sea turtle RBCs remains unknown. The authors elected to use the Hemo-Nate blood filter postsyringe and inline in the present study to reflect likely real-world clinical use so that results could be better compared to previous blood filter research. Using a 170- to 260-µm standard blood filter or a similar albeit larger 40-µm microaggregate blood filter (eg, Haemonetics SQ40 High Flow Blood Filter) could potentially avoid the risk of hemolysis or retention of transfused RBCs altogether. Nonetheless, the authors sought to determine the hemolytic effect of Hemo-Nate blood filter given its widespread commercial availability and versatility to be used postsyringe without the need for an extension or administration set.
To synthesize the present study’s results with the current sea turtle transfusion medicine literature, sodium heparin has been recommended as the preferred anticoagulant for the collection and short-term storage of sea turtle whole blood to be used for transfusion.4 Allogenic whole blood transfusions are safer than xenotransfusions between sea turtle species; however, major and minor crossmatches incubated at room temperature for 30 minutes should be performed before every whole blood transfusion regardless of donor species.3 The frequency and occurrence of aggregates in sea turtle blood donations and the clinical consequences of transfusion to a conspecific are currently unknown, but it is likely that blood filtration is beneficial for sea turtles receiving blood transfusions and provides a standard of care similar to domestic veterinary patients. Although not evaluated in the present study, for large sea turtles undergoing rehabilitation (eg, adult green or loggerhead sea turtles), gravity flow or a peristaltic pump infusion with a blood administration set with a built-in filter is likely the most effective delivery method for whole blood transfusions. For smaller sea turtles undergoing rehabilitation (eg, Kemp’s ridleys or juvenile green turtles), a syringe pump infusion with a postsyringe, inline Hemo-Nate blood filter changed every 50 mL is expected to cause no significant hemolysis based on the results of this study.
Acknowledgments
The authors thank Laura Ruterbories, Larisa Avens, Kathy Zagzebski, Maria Serrano, and the dedicated staff of the Karen Beasley Sea Turtle Rescue and Rehabilitation Center.
Disclosures
The authors have nothing to disclose. No AI-assisted technologies were used in the generation of this manuscript.
Funding
This study was made possible through the financial support of the North Carolina State University Department of Clinical Sciences’ VPP Resident Grant and Dissemination Fund.
References
- 1.↑
Wallace BP, DiMatteo AD, Bolten AB, et al. Global conservation priorities for marine turtles. PLoS One. 2011;6(9):e24510. doi:10.1371/journal.pone.0024510
- 2.↑
Norton TM, Innis CJ, Manire CA. Critical care and emergency medicine In: Manire CA, Norton TM, Stacy BA, Harms C, Innis C, eds. Sea Turtle Health and Rehabilitation. J. Ross Publishing; 2017:483–495.
- 3.↑
Donnelly KA, Norton TM, Zirkelbach B, Stacy NI. Advancing transfusion medicine in sea turtles: optimization of a cross-matching protocol. J Zoo Wildl Med. 2019;50(2):315–321. doi:10.1638/2018-0062
- 4.↑
Phillips BE, Stoskopf MK, Beasley JF, Harns CA. Evaluation of three anticoagulants used for short-term storage of loggerhead sea turtle (Caretta caretta) whole blood. J Herpetol Med Surg. 2017;27(3):97–103. doi:10.5818/17-06-117.1
- 5.↑
Radisic R, Owens SD, Manire CA, et al. Red blood cell osmotic fragility in healthy loggerhead and green sea turtles. J Vet Diagn Invest. 2020;32(6):908–911. doi:10.1177/1040638720957117
- 6.↑
Morse SA, Mooney ET. Effect of microaggregate filter passage on feline whole blood stored for 35 days. J Feline Med Surg. 2022;24(2):116–122. doi:10.1177/1098612X211009145
- 7.↑
Jankowski G, Nevarez J. Evaluation of a pediatric blood filter for whole blood transfusions in domestic chickens (Gallus gallus). J Avian Med Surg. 2010;24(4):272–278. doi:10.1647/2009-026.1
- 8.↑
Nevarez JG, Cockburn J, Kearney MT, Mayer J. Evaluation of an 18-micron filter for use in reptile blood transfusions using blood from American alligators (Alligator mississippiensis). J Zoo Wildl Med. 2011;42(2):236–240. doi:10.1638/2010-0062.1
- 9.↑
R Core Team. R: a Language and Environment for Statistical Computing. The R Foundation; 2023. Accessed November 1, 2022. https://www.r-project.org
- 10.↑
Champely S. pwr: Basic Functions for Power Analysis. Version 1.3.0. The R Foundation; 2020. Accessed November 1, 2022. https://cran.r-project.org/web/packages/pwr/index.html
- 11.↑
Work TM, Raskin RE, Balazs GH, Whittaker SD. Morphologic and cytochemical characteristics of blood cells from Hawaiian green turtles. Am J Vet Res. 1998;59(10):1252–1257. doi:10.2460/ajvr.1998.59.10.1252
- 12.↑
Casal AB, Orós J. Morphologic and cytochemical characteristics of blood cells of juvenile loggerhead sea turtles (Caretta caretta). Res Vet Sci 2007;82(2):158–165. doi:10.1016/j.rvsc.2006.07.017
- 13.↑
Zhang FY, Li PP, Gu HX, Ye MB. Hematology, morphology, and ultrastructure of blood cells of juvenile olive ridley sea turtles (Lepidochelys olivacea). Chelonian Conserv Biol. 2011;10(2):250–256. doi:10.2744/CCB-0890.1
- 14.↑
Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;327(8476):307–310. doi:10.1016/S0140-6736(86)90837-8
- 15.↑
Passing H, Bablok W. A new biometrical procedure for testing the equality of measurements from two different analytical methods. Application of linear regression procedures for method comparison studies in Clinical Chemistry, Part I. J Clin Chem Clin Biochem. 1983;709(11):709–720.
- 16.↑
Passing H, Bablok W. Comparison of several regression procedures for method comparison studies and determination of sample sizes. Application of linear regression procedures for method comparison studies in Clinical Chemistry, Part II. J Clin Chem Clin Biochem. 1984;22(6):431–445.
- 17.↑
Wickham H, François R, Henry L, Muller K, Vaughan D. dplyr: a Grammar of Data Manipulation. Version 1.1.2. The R Foundation; 2023. Accessed August 1, 2023. https://cran.r-project.org/web/packages/dplyr/index.html
- 18.↑
Revelle W. psych: Procedures for Psychological, Psychometric, and Personality Research. Version 2.3.6. The R Foundation; 2023. Accessed August 1, 2023. https://cran.r-project.org/web/packages/psych/index.html
- 19.↑
Datta D. blandr: a Bland-Altman Method Comparison Package for R. Version 0.5.3. The R Foundation; 2017. Accessed August 1, 2023. https://cran.r-project.org/web/packages/blandr/index.html
- 20.↑
Potapov S, Mode IF, Schuetzenmeister A, et al. mcr: Method Comparison Regression. Version 1.3.2. The R Foundation; 2023. Accessed August 1, 2023. https://cran.r-project.org/web/packages/mcr/index.html
- 21.↑
Stacy NI, Innis CJ. Clinical pathology In: Manire CA, Norton TM, Stacy BA, Harms C, Innis C, eds. Sea Turtle Health and Rehabilitation. J. Ross Publishing; 2017;147–208.
- 22.↑
McDevitt RI, Ruaux CG, Baltzer WI. Influence of transfusion technique on survival of autologous red blood cells in the dog. J Vet Emerg Crit Care (San Antonio). 2011;21(3):209–216. doi:10.1111/j.1476-4431.2011.00634.x
- 23.↑
Cooley-Lock KM, Williams JP, Williams ML. Assessment of erythrocyte damage and in-line pressure changes associated with simulated transfusion of canine blood through microaggregate filters. Am J Vet Res. 2019;80(9):852–861. doi:10.2460/ajvr.80.9.852
- 24.↑
Heikes BW, Ruaux CG. Effect of syringe and aggregate filter administration on survival of transfused autologous fresh feline red blood cells. J Vet Emerg Crit Care (San Antonio). 2014;24(2):162–167. doi:10.1111/vec.12115
- 25.↑
Rozanski EA. Blood transfusion methods: time to reevaluate? J Vet Emerg Crit Care (San Antonio). 2011;21(3):184–185. doi:10.1111/j.1476-4431.2011.00648.x
- 26.↑
Aldrich KJ, Saunders DK, Sievert LM, Sievert G. Comparison of erythrocyte osmotic fragility among amphibians, reptiles, birds and mammals. Trans Kans Acad Sci. 2006;109(3–4):149–158. doi:10.1660/0022-8443(2006)109[149:COEOFA]2.0.CO;2]
- 27.↑
Delaune AJM, Field CL, Clauss TM. Whole blood transfusions to treat severe anemia in seven cownose rays (Rhinoptera bonasus) and one short-tail stingray (Dasyatis brevicaudata). J Zoo Wildl Med. 2017;48(4):1172–1180.
- 28.
Louis MM, Cerreta AJ, Griffioen JA, et al. Homologous whole blood transfusion for treatment of severe anemia in five Eastern box turtles (Terrapene carolina carolina). J Exot Pet Med. 2020;35:27–33. doi:10.1053/j.jepm.2020.05.013
- 29.↑
Aguilar LAB, Divers SJ, Comolli JR, Kehoe S, Good J. Successful management of hemocoelom and marked anemia in a central bearded dragon (Pogona vitticeps) following multiple blood transfusions and surgery. J Herpetol Med Surg. 2021;31(2):111–118.
- 30.↑
Harter TS, Reichert M, Brauner CJ, Milsom WK. Validation of the i-STAT and HemoCue systems for the analysis of blood parameters in the bar-headed goose, Anser indicus. Conserv Physiol. 2015;3(1):cov021. doi:10.1093/conphys/cov021
- 31.
Dannemiller NG, Horak KE, Ellis JW, Barrett NL, Wolfe LL, Shriner SA. Effects of external oiling and rehabilitation on hematological, biochemical, and blood gas analytes in ring-billed gulls (Larus delawarensis). Front Vet Sci. 2019;6:405. doi:10.3389/fvets.2019.00405
- 32.↑
Yaw TJ, Gentry J, Ratliff C, et al. Venous blood analytes and osmolality of rehabilitated juvenile black-bellied whistling ducks (Dendrocygna autumnalis). J Avian Med Surg. 2019;33(2):123–132. doi:10.1647/2016-194
- 33.↑
Eshar D, Gancz AY, Avni-Magen N, King R, Beaufrére H. Hematologic, plasma biochemistry, and acid-base analysis of adult Negev Desert tortoises (Testudo werneri) in Israel. J Zoo Wildl Med. 2014;45(4):979–983. doi:10.1638/2014-0053.1
- 34.↑
Wong S, Di Girolamo N. Agreement between hematocrits measured with electrical conductivity on the i-STAT analyzer and manual packed cell volume measurements in 148 chelonian patients. Vet Clin Pathol. 2021;50(3):342–347. doi:10.1111/vcp.13014
- 35.↑
Muñoz-Pérez JP, Lewbart GA, Hirschfeld M, et al. Blood gases, biochemistry and haematology of Galápagos hawksbill turtles (Eretmochelys imbricata). Conserv Physiol. 2017;5(1):cox028.
- 36.↑
Wolf KN, Harms CA, Beasley JF. Evaluation of five clinical chemistry analyzers for use in health assessment in sea turtles. J Am Vet Med Assoc 2008;233(3):470–475. doi:10.2460/javma.233.3.470
- 37.↑
Lewbart GA, Hirschfeld M, Denkinger J, et al. Blood gases, biochemistry, and hematology of Galápagos green turtles (Chelonia mydas). PLoS One 2014;9(1):e96487. doi:10.1371/journal.pone.0096487
- 38.↑
Ludbrook J. Linear regression analysis for comparing two measurers or methods of measurement: but which regression? Clin Exp Pharmacol Physiol 2010;37(7):692–699. doi:10.1111/j.1440-1681.2010.05376.x
- 39.↑
Yagi K, Holowaychuk MK. Recipient monitoring. In: Yagi K, Holowaychuk MK, eds. Manual of Veterinary Transfusion Medicine and Blood Banking. John Wiley & Sons, 2016;172–188.
- 40.↑
Zhang FY, Gu HX, Chen HL, et al. Blood cells morphology and hematology of Eretmochelys imbricata and Chelonia mydas. Chin J Zool 2009;44:113–121.
