• View in gallery

    Photograph of a modified collection set installed onto an automated plasmapheresis instrument. The saline (0.9% NaCl) solution line (A) passes behind the saline solution clamp (1); the line is then sealed. The transmembrane pressure line (B) is lengthened and then connected to the appropriate pressure sensor (2). The cell line (C) is sealed near its entrance into the reservoir (3); however, instead of entering the reservoir, the cell line passes through the cell pump (4) and then joins the return tubing line (D). The plasma line is welded to the collection tubing line (E), and the donor line is welded to the extraction tubing line (F).

  • View in gallery

    Photograph of a dual-instrument, continuous-flow, automated plasmapheresis procedure for harvest of plasma from horses. Whole blood (red arrows) is removed from a donor horse via a catheter placed in the right jugular vein and enters the extraction tubing line (START). The line divides and becomes the donor line for each instrument. Sodium citrate (4%) is infused as an anticoagulant into each donor line via the anticoagulant pump (a) before whole blood reaches each instrument. Anticoagulated whole blood passes through the air detector (b) and the whole blood pump (c). The anticoagulated whole blood is pumped to the separation device (d). Plasma (yellow arrows) exits the bottom port of the separation device and passes down to the 20-L collection bag (not shown). Simultaneously, concentrated blood cells (black arrows) exit the side port of the separation device and pass through the concentrated cell pump (e). The concentrated blood cells are pumped through the return tubing lines, which join and carry concentrated blood cells back to the donor horse via a catheter placed in the left jugular vein.

  • 1 Abbas A, Lichtman A. Section V, Chapter 18. Diseases caused by immune responses: hypersensitivity and autoimmunity. Cellular and molecular immunology. 5th ed. Philadelphia: Elsevier Saunders, 2005;414415.

    • Search Google Scholar
    • Export Citation
  • 2 Brook D. Aspects of plasma production. J Equine Vet Sci 1989; 9:303306.

  • 3 Magdesian K, Brook D, Wickler S. Temporal effects of plasmapheresis on serum proteins in horses. Am J Vet Res 1992; 53:11491152.

  • 4 Porter M, Green E. Blood and blood component therapy. In: Robinson N, ed. Current therapy in equine medicine. 5th ed. Philadelphia: Elsevier Saunders, 2003;355357.

    • Search Google Scholar
    • Export Citation
  • 5 Feige K, Ehrat F, Kastner S, et al. Automated plasmapheresis compared with other plasma collection methods in the horse. J Vet Med A Physiol Pathol Clin Med 2003; 50:185189.

    • Search Google Scholar
    • Export Citation
  • 6 Levine L, Broderick E. The plasmapheresis of hyperimmunized horses. Bull World Health Organ 1970; 42:9981000.

  • 7 Morris D. Therapy in hemolymphatic diseases. In: Colahan P, Mayhew I, Merritt A, et al, eds. Equine medicine and surgery. 5th ed. St Louis: Mosby, 1999;20032007.

    • Search Google Scholar
    • Export Citation
  • 8 Green E, Ward G. A simple method for repeated plasmapheresis of the horse. Lab Anim Sci 1974; 24:948951.

  • 9 Bartges J. Therapeutic plasmapheresis. Semin Vet Med Surg (Small Anim) 1997; 12:170177.

  • 10 Klages C. Repeated automated plasmapheresis in goats (Capra hircus): a clinically safe and long-term refinement to current antibody recovery techniques. J Am Assoc Lab Anim Sci 2006; 45:4953.

    • Search Google Scholar
    • Export Citation
  • 11 Feige K, Ehrat F, Kastner S, et al. The effects of automated plasmapheresis on clinical, haematological, biochemical and coagulation variables in horses. Vet J 2005; 169:102107.

    • Search Google Scholar
    • Export Citation
  • 12 Gutierrez J, Rojas E, Quesada L, et al. Pan-African polyspecific antivenom produced by caprylic acid purification of horse IgG: an alternative to the antivenom crisis in Africa. Trans R Soc Trop Med Hyg 2005; 99:468475.

    • Search Google Scholar
    • Export Citation
  • 13 Isbister G. Snake antivenom research: the importance of case definition. Emerg Med J 2005; 22:399400.

  • 14 BBC News. Worries over snake anti-venom shortage. Available at: news.bbc.co.uk/2/hi/health/890305.stm. Accessed Jul 14, 2009.

  • 15 ScrippsNews. Coral snake antivenin supply is dangerously low. Available at: www.scrippsnews.com/node/32096. Accessed Jul 14, 2009.

  • 16 Plumb D. Detomidine. In: Plumb's veterinary drug handbook. 6th ed. Ames, Iowa: Blackwell Publishing, 2008;263264.

  • 17 McVey D, Loan R. Total and antigen-specific serum immunoglobulin isotype concentrations in hyperimmunized cattle that have undergone plasmapheresis. Am J Vet Res 1989; 50:758761.

    • Search Google Scholar
    • Export Citation

Advertisement

Development of an automated plasmapheresis procedure for the harvest of equine plasma in accordance with current good manufacturing practice

Sara M. Ziska DVM, PhD1, John Schumacher DVM, MS2, Sue H. Duran PhD3, and Kenny V. Brock DVM, PhD4
View More View Less
  • 1 Department of Pathobiology, College of Veterinary Medicine, Auburn University, Auburn, AL 36849.
  • | 2 Department of Clinical Sciences, College of Veterinary Medicine, Auburn University, Auburn, AL 36849.
  • | 3 Department of Clinical Sciences, College of Veterinary Medicine, Auburn University, Auburn, AL 36849.
  • | 4 Department of Pathobiology, College of Veterinary Medicine, Auburn University, Auburn, AL 36849.

Abstract

Objective—To develop a high-speed, continuous-flow, automated plasmapheresis procedure for the high-volume harvest of equine plasma in accordance with current good manufacturing practice.

Animals—143 horses (predominantly draft breeds) between 3 and 10 years of age at the time of purchase.

Procedures—Adaptations were made to automated plasmapheresis instruments and sterile disposable collection sets, which allowed for dual-instrument, continuous-flow operation. Donor horses were connected to the apparatus via 2 catheters (1 inserted in each jugular vein). The instruments removed whole blood from donors, fractionated the blood, diverted plasma to collection bags, and simultaneously returned concentrated cells to the donors. Plasmapheresis was performed on donor horses at 14-day intervals with a maximum of 22 mL of plasma/kg of donor body weight harvested during each plasmapheresis procedure.

Results—During a 5-year period, 3,240 plasmapheresis procedures were performed and > 50,000 L of sterile equine plasma was harvested in accordance with current good manufacturing practice. Donors typically remained calm during the plasmapheresis procedures and tolerated the procedures well. The high-volume and frequent plasma harvest did not result in sustained hypoproteinemia in donor horses. Adverse events associated with the automated plasmapheresis technique were infrequent, and the recurrence of adverse events was minimized by making minor adjustments to the procedure.

Conclusions and Clinical Relevance—The automated plasmapheresis procedure described in this report can be used to safely harvest equine plasma or to perform therapeutic plasmapheresis in horses.

Abstract

Objective—To develop a high-speed, continuous-flow, automated plasmapheresis procedure for the high-volume harvest of equine plasma in accordance with current good manufacturing practice.

Animals—143 horses (predominantly draft breeds) between 3 and 10 years of age at the time of purchase.

Procedures—Adaptations were made to automated plasmapheresis instruments and sterile disposable collection sets, which allowed for dual-instrument, continuous-flow operation. Donor horses were connected to the apparatus via 2 catheters (1 inserted in each jugular vein). The instruments removed whole blood from donors, fractionated the blood, diverted plasma to collection bags, and simultaneously returned concentrated cells to the donors. Plasmapheresis was performed on donor horses at 14-day intervals with a maximum of 22 mL of plasma/kg of donor body weight harvested during each plasmapheresis procedure.

Results—During a 5-year period, 3,240 plasmapheresis procedures were performed and > 50,000 L of sterile equine plasma was harvested in accordance with current good manufacturing practice. Donors typically remained calm during the plasmapheresis procedures and tolerated the procedures well. The high-volume and frequent plasma harvest did not result in sustained hypoproteinemia in donor horses. Adverse events associated with the automated plasmapheresis technique were infrequent, and the recurrence of adverse events was minimized by making minor adjustments to the procedure.

Conclusions and Clinical Relevance—The automated plasmapheresis procedure described in this report can be used to safely harvest equine plasma or to perform therapeutic plasmapheresis in horses.

Equine plasma has been recognized as a valuable resource for > 100 years. In the early 20th century, people with diphtheria infections were treated with plasma obtained from horses that had been vaccinated with diphtheria toxin, which provided passive immunity in the recipients.1 Currently, equine plasma is harvested and incorporated into culture media used by diagnostic laboratories. Equine plasma serves as the matrix in pharmacokinetic studies conducted to determine the half-life of therapeutic compounds. Highly specific antibodies used in scientific assays are isolated from equine plasma. Plasma is harvested from equine donors for transfusion into equine patients for the management of protein-losing enteropathies, nephropathies, coagulopathies, failure of passive transfer, and other medical conditions. Hyperimmune equine plasma is commercially manufactured and sold for prophylactic administration or treatment of several clinically important infectious diseases of horses.2–4 Equine plasma is the starting material in the production of hyperimmune equine fragment antigen-binding products (eg, Fab and F[ab]2), which serve as critical pharmaceuticals in the human health-care industry. These products neutralize viruses, toxins, and venoms and are often the only treatment available for life-threatening medical conditions.

Various techniques have been used over the past 100 years to harvest plasma from equine donors. Initially, investigators simply relied on the rapid erythrocyte sedimentation rate of equine whole blood to harvest the plasma.5–7 They aseptically placed large-bore stainless steel needles into jugular veins of equine donors. Tubing sets were attached to the needles, and whole blood was harvested into sterile glass collection jars containing an anticoagulant.4 Approximately 10% to 20% of the donors' blood volume was removed, which resulted in an anticoagulant-to-whole blood ratio of 1:10.4,8 After the whole blood was collected, the anticoagulated blood was refrigerated for 12 to 24 hours to allow for gravity sedimentation. After the cellular mass gravitated to the bottom of the collection jars, plasma was removed by use of a siphon or other technique. Once plasma was isolated, the remaining cellular elements were discarded. In these early investigations, plasma was harvested at intervals of approximately 30 days, which yielded 2.5 to 5 L of plasma for a typical 500-kg adult horse, without complications to the donor horses.4,7,8

In the 1970s, 2 teams of investigators each described the technique of manual plasmapheresis for harvest of horse plasma.6,8 The process involved the sequential steps of phlebotomy, fractionation, plasma extraction, resuspension of the cellular mass, and return of the resuspended mass to the respective donors.9 Whole blood was collected by use of established techniques. After gravity sedimentation for 12 to 24 hours, plasma was harvested, and the concentrated cellular elements then were resuspended in a volume of saline (0.9% NaCl) solution similar to the volume of plasma harvested. The cells and saline solution were warmed in a water bath to 37°C, then transfused back into the donors. The process was lengthy, with a considerable amount of time required for sedimentation, cellular resuspension, and transfusion. Over time, investigators began collecting whole blood in bags, which were centrifuged to substantially decrease the amount of time required for fractionation. Both teams of investigators performed manual plasmapheresis approximately every 30 days, which yielded 10 to 20 L of plasma for a typical 500-kg horse, without complication to donors.6,8

The introduction of automated in-line blood cell separators in the early 1980s revolutionized plasmapheresis procedures.9,10 These instruments attach via sterile disposable collection sets to catheters inserted in donors. Instrument pumps remove whole blood from donors and infuse anticoagulant into extracted blood at a controlled rate. The anticoagulated whole blood is then fractionated via centrifugation or filtration techniques.5,7,9 Isolated plasma is diverted to attached bags or bottles. Simultaneously, concentrated cells are returned to donors. Some of the automated instruments provide fluid replacement during collection, and others provide fluid replacement after plasma harvest is complete. Although designed for use in humans, these machines have been modified for use in automated plasmapheresis procedures of equine donors in limited settings.3,5,10,11

Automated plasmapheresis procedures offer advantages over previously used plasma harvesting methods in horses. Automated plasmapheresis procedures result in harvest of plasma in a closed system, whereas manual procedures result in harvest of plasma in an open system, which allows for possible bacterial contamination of plasma.5,7 Automated plasmapheresis procedures are a more efficient method in horses, which results in harvest of a greater volume of plasma and return of cells to donors in a shorter period.5,10 With automated plasmapheresis procedures, blood is processed through sterile disposable tubing sets and separation devices. Because the equipment is for single use only, the risk of cross-contamination to donors is eliminated, along with the need to resterilize materials.10 The automated procedure also minimizes extracorporeal deficits by simultaneous harvest of plasma and return of blood cells.9 A study5 conducted to evaluate methods for harvest of equine plasma revealed nearly complete removal of erythrocytes and leukocytes from plasma harvested via automated centrifugation. In comparison, a considerable number of erythrocytes and leukocytes remained in plasma collected via gravity sedimentation or blood bag centrifugation. Also, plasma harvested via the automated technique had greater factor VIII activity, compared with that in plasma harvested via gravity sedimentation. Investigators have evaluated automated plasmapheresis procedures for harvest of 20 mL of plasma/kg from donor horses at 30-day intervals.3,5,11 No adverse events related to the automated procedures were described in any of those reports.3,5,11

Although automated plasmapheresis represents a superior technique for the harvest of equine plasma, there is little information available on the details of the collection procedures. Companies that manufacture plasmapheresis instruments market machines specifically for use in humans, rather than in horses. Consequently, numerous logistic obstacles must be overcome before a substantial volume of equine plasma can be harvested with these instruments. Operating manuals that accompany the machines do not address automated plasmapheresis procedures in horses. Moreover, the sterile disposable collection sets and other supporting materials are designed for harvest of 500 to 800 mL of human plasma, not 20 L of equine plasma. The 3 studies3,5,11 conducted to investigate automated plasmapheresis procedures in horses include brief descriptions of the harvest techniques and indicate the instruments used for the procedures. Those reports provide valuable information regarding automated plasmapheresis procedures in horses but do not include sufficient detail to allow others to replicate the procedures.

The purpose of the study reported here was to develop and describe a closed-loop, high-speed, continuous-flow, automated plasmapheresis procedure for high-volume harvest of plasma from horses in accordance with CGMP. Current good manufacturing practices are guidelines on the manufacture of products in a high-quality system; in the United States, these guidelines are enforced by the FDA. There is continual demand for equine plasma and its constituents. Critical shortages in equine-derived antivenom products have been reported in national and international news.12–15 Availability of a reliable method for automated plasmapheresis procedures in horses may assist in meeting global demands for equine plasma and for products derived from equine plasma.

Materials and Methods

Animals—Horses included for the study were 3 to 10 years old at the time of purchase. Draft breeds were preferred over other breeds because of their docile temperament and large total blood volume. All research and development procedures were approved by the Auburn University Institutional Animal Care and Use Committee.

All horses underwent a physical examination immediately after arrival at the collection facility. Horses received a microchip implant for identification purposes. Blood samples were collected and submitted for a CBC and serum biochemical analysis to establish baseline values. Separate blood samples were collected and submitted to test for equine infectious anemia via agar gel immunodiffusion and equine herpes virus-1 via PCR assay. Horses then were immediately placed in pasture isolation for a minimum quarantine period of 30 days. During the quarantine period, horses received moxidectina and were vaccinated against Eastern equine encephalomyelitis, Western equine encephalomyelitis, West Nile virus, tetanus,b rabies,c equine influenza, equine herpes virus-1, equine herpes virus-4,d and Streptococcus equi.e During the study, horses were maintained on mixed grass pastures with ad libitum access to water and alfalfa hay. In addition, approximately 2 kg of pelleted feedf was provided daily to each horse.

Instruments and scales—Before automated plasmapheresis procedures were performed, a number of modifications were made to the plasmapheresis instruments.g The internal scale associated with each instrument was removed because the weight of the plasma harvested during equine plasmapheresis exceeded the capabilities of the instrument's internal scale. Instead, an external scaleh was used to weigh plasma during plasmapheresis. Switch 8 of the DIP switch S1 located on each instrument's central processing unit board was placed in the on position. This prevented a message alerting the investigators to check donor blood flow from being repeatedly displayed throughout the procedure. Programmable read-only memory chips U14 and U15 were removed from the central processing unit board; only chips U16 and U17 were necessary for plasmapheresis procedures in horses. Softwarei for veterinary applications was installed on each instrument.

Two modified instruments were used simultaneously for each donor horse during plasmapheresis. The 2 instruments were positioned side by side at each collection station. The pressure cuff of each instrument was not used during plasma harvest but instead was wrapped around a portion of the adjacent stocks. The external scale used to weigh product during plasmapheresis was positioned next to the instruments.

Modification of collection sets—Two sterile disposable collection setsj designed specifically for the study reported here were adapted to create 1 modified set for dual-instrument, continuous-flow operation. Adaptations were made by use of a medical tubing sealer,k a medical tubing welder,l 3 pieces of custom-designed tubing,m–o and a custom-designed 20-L collection bag.p All adaptations were performed in a manner to maintain sterility of components. The cell line of each collection set was heat sealed by use of the medical tubing sealer at the point where it entered the reservoir. The saline solution line of each set was sealed approximately 30 to 35 cm from the spiked end. The spiked end of each sealed saline solution line was then separated from the set and discarded. Next, the donor line of 1 set was welded to a 90-cm-long line of the custom-designed extraction tubing.o The procedure was repeated with the remaining donor line and the remaining 90-cm-long line of that same extraction tubing. Next, each sealed cell line was welded to each 90-cm-long line of the custom-designed return tubing.m Then, the transmembrane pressure line of both collection sets was lengthened by welding approximately 45 to 50 cm of sterile tubing into that line. The modified set was installed onto each of the side-by-side plasmapheresis instruments (Figure 1). Once the modified set was installed onto the instruments, each plasma line was welded to each 70-cm-long line of the custom-designed collection tubing.n The remaining 90-cm-long line of collection tubing was welded to the centrally located port of the 20-L collection bag. The 20-L bag was placed into a protective liner and set on top of a cool pack resting at the bottom of a polyethylene plastic tub. The tub and its contents were set on the external scale to weigh the amount of plasma harvested into the 20-L bag.

Figure 1—
Figure 1—

Photograph of a modified collection set installed onto an automated plasmapheresis instrument. The saline (0.9% NaCl) solution line (A) passes behind the saline solution clamp (1); the line is then sealed. The transmembrane pressure line (B) is lengthened and then connected to the appropriate pressure sensor (2). The cell line (C) is sealed near its entrance into the reservoir (3); however, instead of entering the reservoir, the cell line passes through the cell pump (4) and then joins the return tubing line (D). The plasma line is welded to the collection tubing line (E), and the donor line is welded to the extraction tubing line (F).

Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.762

Schedule for donor horses—Horses entered the plasmapheresis schedule on the basis that they had achieved benchmark antibody titers against specific immunogens. Potential donors were identified from the schedule, and a record review was conducted for each horse. Horses that had medical criteria for deferral from plasma donation were deemed unsuitable for collection. All suitable donor horses were brought to the treatment barn. Identification of each donor was verified; each horse was weighed and underwent a physical examination. Rectal temperature was measured by use of a National Institute of Standards and Technology–certified traceable thermometer, and an ECG was recorded. Hair was clipped from a 20 × 12-cm area over each jugular vein. Blood was collected from each potential donor into 1 evacuated blood-collection tube containing EDTA and 1 evacuated blood-collection tube containing no anticoagulant. Total plasma protein concentration for each horse was determined via refractometry of plasma from EDTA-anticoagulated blood tubes. Healthy horses with a rectal temperature < 39.5°C and total plasma protein concentration > 5.5 g/dL were approved for plasmapheresis.

Approved donor horses were mildly sedated with detomidineq (6 μg/kg, IV) and led into the collection building. Each donor was guided into a set of stocks and secured with ropes placed over their dorsum and behind their hind limbs. The head of each horse was placed in a sling made from padded saddle girths, and the cranial aspect of the thorax of each horse rested against the front door of the stocks. Previously positioned plasmapheresis instruments and scales were located near the right shoulder of each donor. The clipped area over each jugular vein was prepared for catheterization by washing with 2% chlorhexidine acetater and a 1-step antiseptic sponge.s Approximately 1.5 mL of 2% lidocaine hydrochloridet was injected SC over each catheterization site, and a small skin incision was made with a No. 15 scalpel blade in the anesthetized areas. A 10-gauge, 76-mm catheteru was inserted through each skin incision and directed ventrally into each jugular vein of donor horses. The catheters were capped and secured in place with 2-0 sutures.v

Plasmapheresis procedures—For each donor horse, the Luer end of the extraction tubing line was connected to the catheter in the right jugular vein, and the Luer end of the return tubing linem was connected to the catheter in the left jugular vein. Both tubing lines were secured to the neck of the donor with sutures. A semifirm 10-cm-long cylinder was placed in the right jugular furrow at a point caudal to the catheterization site, and 2 foam pads were positioned outside the left jugular furrow. A tourniquet consisting of elastic wrap was placed around the donor neck to hold the cylinder and pads in position and put pressure on the right jugular vein to prevent venous collapse during plasmapheresis. After placement of the wrap to induce moderate compression of only the right jugular vein, both plasmapheresis instruments were advanced simultaneously to initiate the priming sequence, which was performed at the start of every procedure.

Once priming was complete, both instruments performed collection and reinfusion cycles concurrently (Figure 2). Whole blood was removed via the catheter in the right jugular vein of each donor and entered the extraction tubing line. The extraction tubing line divided and delivered blood to both donor lines of the modified disposable collection set. An anticoagulant solution of 4% sodium citratew was infused into the whole blood, which resulted in an anticoagulant-to-whole blood ratio of 1:16. This ratio was achieved by programming the instruments to operate at an anticoagulant setting of 6%. Next, the anticoagulated whole blood entered each plasmapheresis instrument via the blood lines. The blood pump sent anticoagulated blood to the separation device, which acted as a rotating membrane filter to separate plasma from cellular components of blood. Plasma exited the bottom port of the separation device, passed through the instrument's refractometer, and entered the 20-L collection bag. Simultaneously, concentrated blood cells exited the side port of the separation device and traveled through the cell pump and immediately back to the donor via the catheter in the left jugular vein. The reservoir was completely bypassed, which eliminated the need for intermittent collection and reinfusion cycles. The plasmapheresis procedure was terminated by the project director or on the basis of the following conditions: a target of 22 mL of plasma/kg of donor body weight was harvested, the 20-L capacity of the bag was reached, the donor's health appeared in jeopardy, the instruments were unable to maintain plasma separation, or there was power loss in the facility. The procedure was ended by pressing the stop button on both instruments. The modified collection set was sealed along both plasma lines, both anticoagulant lines, the extraction line, and the return line. This allowed the donor and collection bag to be separated from the disposable collection set; the disposable collection set then was discarded as medical waste.

Figure 2—
Figure 2—

Photograph of a dual-instrument, continuous-flow, automated plasmapheresis procedure for harvest of plasma from horses. Whole blood (red arrows) is removed from a donor horse via a catheter placed in the right jugular vein and enters the extraction tubing line (START). The line divides and becomes the donor line for each instrument. Sodium citrate (4%) is infused as an anticoagulant into each donor line via the anticoagulant pump (a) before whole blood reaches each instrument. Anticoagulated whole blood passes through the air detector (b) and the whole blood pump (c). The anticoagulated whole blood is pumped to the separation device (d). Plasma (yellow arrows) exits the bottom port of the separation device and passes down to the 20-L collection bag (not shown). Simultaneously, concentrated blood cells (black arrows) exit the side port of the separation device and pass through the concentrated cell pump (e). The concentrated blood cells are pumped through the return tubing lines, which join and carry concentrated blood cells back to the donor horse via a catheter placed in the left jugular vein.

Citation: American Journal of Veterinary Research 73, 6; 10.2460/ajvr.73.6.762

The collection bag was transported to a separate room for further processing. There, the plasma was homogenized and aseptically placed into 1-L high-density polyethylene bottles.x Appropriate labels were applied to the bottles; bottles were stored at −35°C in a continuously monitored freezer.

Donor horses were closely observed throughout the plasmapheresis procedures. Pulse rate, respiratory rate, characteristics of the mucous membranes, and an ECG were recorded at approximately the midpoint and end of each plasmapheresis procedure. Donors were administered detomidine (3 to 6 μg/kg, IV) or detomidine and butorphanoly (3 to 6 μg of each/kg, IV) as needed to prevent restlessness and excessive movement during plasmapheresis. Sedation was administered approximately once every hour. Sedatives were administered IV through an infusion port located in the return tubing line.

At the end of the plasmapheresis procedure, the tourniquet was removed from the neck of each horse. Each horse then received 15 L of fluidsz IV via gravity flow. Catheters were removed from the jugular veins after fluid administration; hemostasis at the catheterization sites was achieved by use of 4 × 4-inch gauze sponges and slight manual pressure. The donors then were washed, visually inspected, and returned to their designated pastures.

Numerous minor adjustments were implemented to prevent recurrence of postprocedure reperfusion injury. The tourniquet used during plasmapheresis was redesigned to prevent occlusion of the carotid artery, and employees were trained on proper application of the tourniquet. Two 15-cm-long pieces of foam padding were placed on each side of the jugular furrow of the jugular vein that received the returned concentrated cells to remove unnecessary pressure from this side of the horse's neck. Tourniquet removal was modified so that it became a gradual process and no longer was an abrupt event. Instead of administering replacement fluids immediately after the conclusion of plasmapheresis, IV administration of fluids was delayed for a minimum of 10 minutes.

Statistical methods—Any incidents were recorded, and the frequency of adverse events were determined. By use of statistical software,aa a Fisher exact 1-tail test was performed to determine whether there was a significant (P < 0.05) difference between the reported frequency of neurologic abnormalities before and after minor adjustments were made to the plasmapheresis and postplasmapheresis procedures.

Results

Donor horses consisted of 88 mares and 55 geldings (98 Belgians, 30 Percherons, 4 Standardbreds, 3 warmbloods, 2 Percheron-crossbred horses, 2 Belgian-crossbred horses, 2 Standardbred-crossbred horses, 1 Paint-crossbred horse, and 1 Shire-crossbred horse). Horses were 4 to 14 years old (mean ± SD age, 8 ± 2 years) and had a mean body weight of 804 ± 97 kg.

For the 5-year study period beginning in 2005, the plasmapheresis technique was used to perform > 3,200 plasmapheresis procedures on the 143 donor horses. Procedures were performed at 14-day intervals. A maximum of 22 mL of plasma/kg was harvested from each donor horse during each procedure. A maximum of 20 L of plasma (the capacity of the collection bag) was harvested during each procedure from horses that weighed ≥ 910 kg. Duration of each donation session was typically 3 to 6 hours. Use of this plasmapheresis procedure during the 5-year period yielded > 50,000 L of sterile equine plasma harvested in accordance with CGMP.

Serial automated plasmapheresis procedures performed in accordance with the described technique and schedule were tolerated well by almost all donor horses. Throughout the study period, no horse was deemed unsuitable for plasmapheresis because of a total plasma protein concentration < 5.6 g/dL. Most donors remained calm during the procedure, with only mild excitement or head shaking observed. Although nearly all donors tolerated plasmapheresis procedures and became more comfortable with subsequent plasmapheresis sessions, a small number of donors repeatedly resented the procedures. Conditioning and multiple plasmapheresis sessions did not improve the tolerance of these few horses. Signs of resentment in these donors included moderate to severe head shaking, foot stomping, striking, biting, and jumping out of the stocks.

Several adverse events related to plasmapheresis were observed in the horses. Inadvertent return of concentrated blood cells into the subcutaneous tissues was recognized as firm perijugular swelling > 5 × 5 cm, which was associated with the jugular vein that received the concentrated blood cells. Return of concentrated cells into the subcutaneous tissues was detected during 22 of 3,240 (0.7%) plasmapheresis procedures. Development of excessive scar tissue, defined as the formation of scar tissue > 4 cm in diameter at the site of a previous catheter insertion, was considered an adverse event. Of the 143 donor horses, 1 (0.7%) developed excessive scar tissue. Increased respiratory rates (≥ 30 breaths/min that continued to increase for sustained periods) resulted in premature termination of the plasmapheresis procedure. Twenty-two of 3,240 (0.7%) plasmapheresis procedures were prematurely terminated because of increased respiratory rate.

Immediately following plasmapheresis and removal of the compression wrap from the neck, 11 donors had ≥ 1 neurologic abnormality once (11/2,850 [0.4%]). These abnormalities included ataxia, inability to stand, nystagmus, and unilateral or bilateral blindness. Each of the described abnormalities was temporary and resolved within hours to days after the procedure, except for 1 horse with bilateral blindness, which was permanent. All of the donors, including the bilaterally blind horse, continued to be used in plasmapheresis procedures following the neurologic episode. After minor adjustments were made to the plasmapheresis and postplasmapheresis procedures, no additional neurologic abnormalities were detected (0/390 [0%] procedures). No significant (P = 0.24) difference was found between the frequency of neurologic abnormalities before and after the minor changes were implemented. During the investigation, 3 horses died after the plasmapheresis procedure (3/3,240 [0.09%]). These 3 donor horses developed signs of weakness and subsequently collapsed within 20 minutes after plasmapheresis and died within 12 hours after the plasmapheresis procedure. No horses died during the plasmapheresis procedures.

Discussion

A high-speed, continuous-flow, automated plasmapheresis procedure was developed to harvest a high volume of equine plasma in a closed and sterile system. Performance of plasmapheresis on horses required modification of instruments and sterile collection sets that were designed for use on humans. High-speed and high-volume plasmapheresis was accomplished by simultaneous operation of 2 modified instruments/donor horse. Each of the instruments was programmed to process 135 to 150 mL of whole blood/min, with each instrument harvesting approximately 30 to 50 mL of plasma/min. In comparison, investigators in previous studies5,11 used a single separator instrument per donor, which processed 70 to 100 mL of whole blood/min. In 1 study,3 the investigators did not describe the volume of whole blood processed per minute by the instrument; however, only 4 to 11 L of plasma was harvested over a 3-hour period in that study. In contrast, 20 L of plasma was routinely harvested over a 3- to 6-hour period in the present study.

In the present study, a closed system was maintained through use of tubing sealers and welders to modify the collection sets. Continuous-flow operation was achieved by bypassing the collection set reservoir and immediately returning concentrated blood cells to the donor horse. This modification allowed the instruments to simultaneously perform collection and reinfusion cycles. Before this modification was made, the reservoir received concentrated cells, with the instrument alternating between collection and reinfusion cycles. A report10 on automated plasmapheresis procedures in goats contains a description of use of the instrument and alternating cycles to harvest goat plasma. Because the reservoir was used and cycles were alternated, those investigators described an intermittent-flow plasmapheresis procedure.

To the authors' knowledge, the present report is the first that contains a description of automated plasmapheresis procedures in horses repeated at 14-day intervals. A similar volume of plasma had been harvested from equine donors via automated plasmapheresis procedures in other studies3,5,11 but only at 30-day intervals. Those investigators recommended the 30-day interval between plasmapheresis procedures to allow donor horses time to replenish depleted total protein, albumin, and IgG concentrations. Despite the use of automated plasmapheresis procedures at approximately twice the previously reported frequency, all horses in the present study were deemed to be suitable for plasmapheresis on the basis of criteria, one of which was the requirement for a total plasma protein concentration > 5.5 g/dL. In fact, a total plasma protein concentration < 6.0 g/dL was never recorded in > 3,200 examinations for donor approval, which illustrated that total plasma protein concentration was always within an acceptable range (6.0 to 8.5 g/dL) 14 days after plasmapheresis. It is reasonable to conclude that harvesting up to 22 mL of plasma/kg every 14 days does not result in a sustained hypoproteinemic state.

In general, the automated plasmapheresis procedure was tolerated well by the donor horses. Most horses stood still during the 3 to 6 hours required for plasmapheresis and needed infrequent administration of sedation. In contrast, there were a few donors that repeatedly behaved poorly during plasmapheresis. These horses bit, struck at personnel, stomped, or jumped out of the stocks, which occasionally resulted in premature termination of the plasmapheresis session. As a result, sedation was administered more frequently to these few specific donor horses. Attempts to condition these horses to the plasmapheresis procedures were unsuccessful. It was determined that these horses displayed this behavior during plasmapheresis procedures, but also when they were placed in stocks or were handled for other reasons. Overall, these horses had poor temperaments and the recommendation was made to remove them from the program.

Investigators in 3 previous reports3,5,11 on automated plasmapheresis procedures in horses did not describe adverse events resulting from the procedures. In contrast, infrequent but clinically important adverse events related to the automated plasmapheresis procedure were recorded during the study reported here. These adverse events may have been attributable to the large number of horses used in the study, total number of plasmapheresis procedures performed, or use of a tourniquet. Although rehydration by IV administration of fluids was performed following plasmapheresis in the present study and during plasmapheresis in the previous studies,3,5,11 it appears unlikely that adverse events resulted from the delay in rehydration. Fluid replacement via IV administration in humans undergoing plasmapheresis routinely is performed after plasma harvest is complete. In the horses of the present study, adverse reactions ranged from minor issues to grave complications. Regardless of the severity of the reaction, each incident was recorded and investigated, and attempts were made to prevent incident recurrence.

In a few rare instances, concentrated blood cells returned to donor horses were reinfused SC rather than IV. It was determined that this adverse event was related to inaccurate catheter placement or to catheter dislodgement as a result of the donor horse shaking its head during the plasmapheresis procedures. Catheter sites were closely monitored throughout the procedure for evidence of unexpected swelling. On recognition of swelling over the jugular vein, which indicated the perivascular return of concentrated cells into the subcutaneous tissues, plasma harvest was temporarily stopped, and the donor horse was evaluated. If the problem was identified quickly and there was minimal swelling in the area, a new catheter was inserted in the original site. If swelling prevented use of the original site, a new catheterization site was selected cranial or caudal to the original site. Occasionally, the new catheter was inserted in the contralateral jugular vein at a point caudal to the catheter used for removal of whole blood for plasmapheresis. If the catheter could not be replaced, the procedure was terminated at that point. To prevent such incidents, catheters were always sutured in place at the insertion site of the jugular vein. If there were any problems when placing a catheter, proper positioning of the catheter was confirmed by use of a syringe to manually remove whole blood from the catheter. The head of each horse was placed in a padded sling, and cross ties were applied to reduce head shaking during plasmapheresis. Although return of concentrated cells into the subcutaneous tissues was recognized as an adverse event, it occurred with an extremely low frequency (22/3,240 [0.7%] procedures), did not cause subsequent complications in the donor horse, and did not interfere with participation of the horse in future plasmapheresis procedures.

Because the jugular veins were catheterized every 14 days, formation of scar tissue was expected. However, most sites healed quickly and evidence of frequent catheterization was not readily apparent. The rapid healing and lack of gross scar formation were attributed to the use of aseptic techniques, skilled placement of catheters, and gentle removal of catheters. Despite the success observed for most horses, 1 donor horse developed diffuse scars along both jugular furrows with mild signs of pain after repeated catheter placement. Subsequently, catheters were placed cranial or caudal to the area of scar tissue that developed in this donor horse. However, because serial plasmapheresis procedures were performed, the location at which to place catheters in the jugular veins became limited and, over time, it became necessary to insert catheters through the scar tissue. This horse was monitored particularly closely during plasmapheresis because of the difficulty in accurate catheterization of the jugular veins. Although the scar tissue that developed in this horse was problematic, it never resulted in the inability to perform plasmapheresis.

The cause of the increased respiratory rates in association with plasmapheresis was never determined, but several possibilities were considered, including excitement of the donor horse, RBC lysis, undetected cardiac disease, and temporary hypovolemia. The plasmapheresis session was allowed to proceed despite the increased respiratory rate. However, the procedure was halted and the horse evaluated when the respiratory rate increased to ≥ 30 breaths/min. If the respiratory rate remained increased, plasmapheresis was terminated. If the respiratory rate decreased to within acceptable physiologic limits and no other clinical abnormalities were detected, plasmapheresis was resumed without complication. To reduce excitement of the donor horses, visual and auditory stimuli were kept at a minimum while the horses were in the collection room. To prevent RBC damage during plasmapheresis, care was taken to properly install tubing onto pump rollers and to protect tubing lines from crushing forces. Cardiac auscultation and ECG were performed on donor horses before and during plasmapheresis to identify cardiac abnormalities, but evidence of cardiac disease was not detected in any of these horses. Unfortunately, echocardiography was not available for use in cardiac evaluation. To reduce the potential for development of temporary hypovolemia, the speed of plasmapheresis was decreased in donors that developed increased respiratory rates. Other clinical signs associated with hypovolemia, such as colic and piloerection, were not observed throughout the study. Sedation with detomidine was used conservatively in these horses because detomidine causes an initial hypertension, which is followed by bradycardia and second-degree atrioventricular block.16 Therefore, the potential existed to exacerbate signs associated with underlying cardiac disease and hypovolemia.

Neurologic abnormalities after plasmapheresis and tourniquet removal were not frequently detected; however, the gravity of the events was substantial. A diagnosis of cerebral and cerebellar reperfusion injury was made on the basis of clinical signs and elimination of other potential causes. The abnormal neurologic events were attributed to inadvertent occlusion of the carotid artery, abrupt tourniquet removal, and prompt IV administration of fluids after plasmapheresis. After implementation of all changes, neurologic abnormalities were no longer observed following plasmapheresis and tourniquet removal (0/390 [0%] procedures). Investigators for the other studies3,5,11 on plasmapheresis in horses did not describe the use of a tourniquet, which may explain the reason that those investigators did not observe signs associated with reperfusion injury to the brain. A tourniquet was applied in the study reported here to provide adequate blood flow and to prevent collapse of the jugular vein during the plasmapheresis procedures. A study17 on manual plasmapheresis procedures in cattle included a description of the use of a tourniquet on the neck of donor cattle to provide adequate blood flow during plasma harvest. No adverse events associated with plasmapheresis were reported, but only 3 cattle were used in that short-term study.17

In the present study, 3 donor horses died within 12 hours after completion of the plasmapheresis procedure. The donor horses became weak and subsequently collapsed within 20 minutes after completion of the plasmapheresis procedure. After they collapsed, 2 donor horses had increased pulse rates with sustained ventricular tachycardia as determined by use of ECG. Results of CBCs and serum biochemical analysis performed before and after plasmapheresis did not yield information regarding potential causes of death. Necropsy results from one of the donor horses that developed cardiac abnormalities revealed acute, multifocal, moderate myocardial necrosis. Other necropsy findings for these 3 donor horses were unremarkable. Although gross and histologic necropsy findings did not reveal evidence of cerebral or cerebellar reperfusion injury, no horses died as a result of plasmapheresis after the changes to prevent reperfusion injury were implemented. Since completion of this investigation, > 1,000 additional plasmapheresis procedures have been performed at the same facility by use of the same technique without any deaths of horse donors following the procedures.

In the present report, we describe a safe and reliable method for repeated harvest of 22 mL of plasma/kg from equine donors as frequently as every 14 days. Any adverse events related to the plasmapheresis procedure can be kept to a minimum by selection of donors with acceptable temperaments, careful catheterization, close monitoring of donors during collection, and prevention of reperfusion injury. This method of automated plasmapheresis can be used to collect high volumes of equine plasma in accordance with CGMP.

This method could also be used to perform therapeutic plasmapheresis on equine patients, which is a form of treatment never reported in equine medicine. The theoretical basis for therapeutic plasmapheresis is the removal of pathogenic or deleterious materials from the circulation.9 Presumably, these materials are proteins or protein-bound or high–molecular-weight solutes and include autoantibodies, circulating immune complexes, excess lipids or hormones, exogenous toxins, and other abnormal molecules. Therapeutic plasmapheresis has been used in human and canine medicine for the management of autoimmune diseases, neoplasia, hyperviscosity, sepsis, intoxication, and other conditions. The technique potentially could be used to help in the management of hyperlipemia in ponies and miniature horses. Plasmapheresis theoretically would hasten removal of excess triglycerides from the blood. Plasmapheresis could be used in horses for the treatment of diseases (including glomerulonephritis, vasculitis, purpura hemorrhagica, or anterior uveitis) that result from deposition of circulating immune complexes.

ABBREVIATION

CGMP

Current good manufacturing practice

a.

Quest gel, Fort Dodge Animal Health, Fort Dodge, Iowa.

b.

West Nile Innovator + EWT, Fort Dodge Animal Health, Fort Dodge, Iowa.

c.

RabVac 3, Fort Dodge Animal Health, Fort Dodge, Iowa.

d.

Fluvac Innovator EHV-4/1, Fort Dodge Animal Health, Fort Dodge, Iowa.

e.

Pinnacle IN, Fort Dodge Animal Health, Fort Dodge, Iowa.

f.

Life Design Prime 14, Nutrena, Minneapolis, Minn.

g.

Autopheresis-C A-200, Baxter-Fenwal, Lake Zurich, Ill.

h.

Champ Square CQ50L scale with CD33 indicator, Ohaus Corp, Pine Brook, NJ.

i.

Autopheresis-C APA software for veterinary applications, version 2.0, Baxter-Fenwal, Lake Zurich, Ill.

j.

4R-2252 Plasmacell-C set, Baxter-Fenwal, Lake Zurich, Ill.

k.

Baxter AutoSeal tubing sealer, Baxter-Fenwal, Lake Zurich, Ill.

l.

TSCD sterile tubing welder, Terumo Medical Corp, Somerset, NJ.

m.

03-220-EPS1 tubing set, Charter Medical, Winston-Salem, NC.

n.

03-220-EPS2 tubing set, Charter Medical, Winston-Salem, NC.

o.

03-220-EPS3 tubing set, Charter Medical, Winston-Salem, NC.

p.

EPS-20 L collection bag, Charter Medical, Winston-Salem, NC.

q.

Dormosedan, Pfizer Animal Health, Exton, Pa.

r.

Nolvasan surgical scrub, Fort Dodge Animal Health, Fort Dodge, Iowa.

s.

ChloraPrep One-Step antiseptic sponge, Medi-Flex Hospital Products, Overland Park, Kan.

t.

Hospira Inc, Lake Forest, Ill.

u.

Becton-Dickinson, Franklin Lakes, NJ.

v.

Supramid, S. Jackson Inc, Alexandria, Va.

w.

Baxter-Fenwal, Lake Zurich, Ill.

x.

Plasmalink pooling bottle, Baxter-Fenwal, Lake Zurich, Ill.

y.

Torbugesic, Fort Dodge Animal Health, Fort Dodge, Iowa.

z.

Plasma-Lyte A, Baxter-Fenwal, Lake Zurich, Ill.

aa.

SAS, version 9.1, SAS Institute Inc, Cary, NC.

References

  • 1 Abbas A, Lichtman A. Section V, Chapter 18. Diseases caused by immune responses: hypersensitivity and autoimmunity. Cellular and molecular immunology. 5th ed. Philadelphia: Elsevier Saunders, 2005;414415.

    • Search Google Scholar
    • Export Citation
  • 2 Brook D. Aspects of plasma production. J Equine Vet Sci 1989; 9:303306.

  • 3 Magdesian K, Brook D, Wickler S. Temporal effects of plasmapheresis on serum proteins in horses. Am J Vet Res 1992; 53:11491152.

  • 4 Porter M, Green E. Blood and blood component therapy. In: Robinson N, ed. Current therapy in equine medicine. 5th ed. Philadelphia: Elsevier Saunders, 2003;355357.

    • Search Google Scholar
    • Export Citation
  • 5 Feige K, Ehrat F, Kastner S, et al. Automated plasmapheresis compared with other plasma collection methods in the horse. J Vet Med A Physiol Pathol Clin Med 2003; 50:185189.

    • Search Google Scholar
    • Export Citation
  • 6 Levine L, Broderick E. The plasmapheresis of hyperimmunized horses. Bull World Health Organ 1970; 42:9981000.

  • 7 Morris D. Therapy in hemolymphatic diseases. In: Colahan P, Mayhew I, Merritt A, et al, eds. Equine medicine and surgery. 5th ed. St Louis: Mosby, 1999;20032007.

    • Search Google Scholar
    • Export Citation
  • 8 Green E, Ward G. A simple method for repeated plasmapheresis of the horse. Lab Anim Sci 1974; 24:948951.

  • 9 Bartges J. Therapeutic plasmapheresis. Semin Vet Med Surg (Small Anim) 1997; 12:170177.

  • 10 Klages C. Repeated automated plasmapheresis in goats (Capra hircus): a clinically safe and long-term refinement to current antibody recovery techniques. J Am Assoc Lab Anim Sci 2006; 45:4953.

    • Search Google Scholar
    • Export Citation
  • 11 Feige K, Ehrat F, Kastner S, et al. The effects of automated plasmapheresis on clinical, haematological, biochemical and coagulation variables in horses. Vet J 2005; 169:102107.

    • Search Google Scholar
    • Export Citation
  • 12 Gutierrez J, Rojas E, Quesada L, et al. Pan-African polyspecific antivenom produced by caprylic acid purification of horse IgG: an alternative to the antivenom crisis in Africa. Trans R Soc Trop Med Hyg 2005; 99:468475.

    • Search Google Scholar
    • Export Citation
  • 13 Isbister G. Snake antivenom research: the importance of case definition. Emerg Med J 2005; 22:399400.

  • 14 BBC News. Worries over snake anti-venom shortage. Available at: news.bbc.co.uk/2/hi/health/890305.stm. Accessed Jul 14, 2009.

  • 15 ScrippsNews. Coral snake antivenin supply is dangerously low. Available at: www.scrippsnews.com/node/32096. Accessed Jul 14, 2009.

  • 16 Plumb D. Detomidine. In: Plumb's veterinary drug handbook. 6th ed. Ames, Iowa: Blackwell Publishing, 2008;263264.

  • 17 McVey D, Loan R. Total and antigen-specific serum immunoglobulin isotype concentrations in hyperimmunized cattle that have undergone plasmapheresis. Am J Vet Res 1989; 50:758761.

    • Search Google Scholar
    • Export Citation

Contributor Notes

The authors thank Drs. James C. Wright and Roberto Palomares-Naveda for assistance with the statistical analysis and interpretation.

Address correspondence to Dr. Ziska (ziskasm@auburn.edu).