Environmental Health Perspectives Accumulation of Di-2ethylhexyl Phthalate (DEHP) in Whole Blood, Platelet Concentrates, and Platelet-Poor

The widespread use of plastic bags in routine blood bank procedures has been attended by concern about the accumulation of substances from the plastic containers in the blood products and the biologic effects of these substances (1-11). Most poly(vinyl chloride) plastic bags contain di-2-ethylhexyl phthalate (DEHP) to make them pliable enough to prevent breakage (2, 5). This material has been shown to accumulate in the plasma of stored whole blood (12).


Introduction
The widespread use of plastic bags in routine blood bank procedures has been attended by concern about the accumulation of substances from the plastic containers in the blood products and the biologic effects of these substances (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11). Most poly(vinyl chloride) plastic bags contain di-2-ethylhexyl phthalate (DEHP) to make them pliable enough to prevent breakage (2,5). This material has been shown to accumulate in the plasma of stored whole blood (12). 02115. *This work was supported by the U. S. Navy. The opinions or assertions contained herein are those of the authors, and are not to be construed as official or reflecting the views of the Navy Department or Naval Service at large.
Our study was undertaken to measure the amount of DEHP in the plasma separated from whole blood that had been anticoagulated with either acid-citrate-dextrose (ACD) or citrate-phosphate-dextrose (CPD) and stored in poly(vinyl chloride) plastic containers. We also measured the amount of DEHP in bioriented polyolefin and in polyethylene plastic containers used in freezing human red cells and platelets, and the accumulation of DEHP in four different anticoagulants stored at room temperature and in 0.9% sodium chloride solution and solutions used for rejuvenation of human red cells stored in poly(vinyl chloride) plastic containers. The amount of DEHP present in thawed glycerolized red cells stored in plastic containers at either -80°C or -150°C for up to 2 years, and the amount of DEHP in washed liquid-stored red cells and previouslyfrozen deglycerolized red cells were also measured.
Because CPD is now suggested as the anticoagulant of choice for collection of blood (13), components prepared from CPD whole blood were studied, i.e., platelet concentrates (PC), platelet-rich plasma (PRP), platelet-poor plasma (PPP), washed platelets, and washed red cells. Storage of platelet concentrates at 40C and 220C was studied because it has been recommended that platelet concentrates be stored at 220C for up to 72 hr (14). The effects of DEHP on platelet viability and function in vivo were also studied.

Collection and Preparation of the Whole Blood Fractions
Approximately 450 ml of whole blood was collected in double-pack poly(vinyl chloride) (PVC) blood bags containing either ACD or CPD anticoagulants. The plastic used in the collection bag containing the anticoagulant was Travenol PL-130, and that in all the transfer packs was Travenol PL-146. The whole blood was stored at 4°C for up to 34 days. Each of 10 units of ACD-stored whole blood was washed by continuous-flow centrifugation by use of either Haemonetics Corporation's reusable stainless steel washing bowls or Baxter Laboratories' collapsible disposable poly(vinyl chloride) bags (15)(16)(17). Two liters of buffered saline (0.9 gm/100 ml NaCl containing 0.2 g-% glucose and buffered to pH 6.8 with 0.065 g-% disodium phosphate) were used to wash each unit within 15 min (Table 1) (17). Samples were taken during storage of the whole blood at 40C and before and after washing of each unit.

Freeze-Preservation
The red cells were usually stored in either ACD or CPD at 40C for about 7 days prior  (18)(19)(20). The methods for addition and removal of the glycerol were as described below. Outdated red cells were rejuvenated prior to freezing (21). After the ACD or CPD whole blood or the concentrated red cells had been stored at 40C for at least 21 days and as long as 36 days, a 50-ml volume of a rejuvenation solution was added to each unit. The rejuvenation solution contained per liter either (a) 50 mmole pyruvate (sodium salt), 50 mmole inosine, 100 mmole glucose, and 50 mmole phosphate, and the pH was adjusted to 7.2 (PIGP) or (b) 50 mmole pyruvate, 50 mmole inosine, 100 mmole glucose, 50 mmole phosphate, and 5 mmole adenine, and the pH was adjusted to 7.2 (PIGPA). Each unit of blood was then placed in a water-tight plastic envelope and incubated at 370C with agitation for 60 min.
Units of whole blood and of concentrated red cells incubated without the rejuvenation solution served as controls.
Meryman method-Within 6 days of storage as whole blood at 40C in either ACD or CPD and after incubation with or without the rejuvenation solution, the outdated red cells were concentrated and all the visible plasma removed. A 100-ml portion of a glycerol solution containing per 100 ml 57.1 g glycerol, 1.6 g sodium R-lactate, 0.03 g KC1, 0.04 g MgCl2, and 0.06 g disodium phosphate buffered to pH 7.2, was added to 250 ml of concentrated red cells (22). A total of 400 ml of the glycerol solution was added to the concentrated red cells whose weight ranged from 275 to 325 g. The red cells were then frozen to -80°C and stored up to 20 months in a poly(vinyl chloride) plastic bag (TA-20, P.L-146) ( Table 2). The red cells were thawed, and deglycerolization was accomplished by the continuous-flow centrifugation wash method with the use of hypertonic NaCl solutions. The thawed red cells were diluted with 150 ml of 12 g-% NaCl buffered to pH 7.2 with 0.15 g-% disodium phosphate. They were then washed with 2 liters of 1.6 g-% NaCl buffered to pH 7.2 with 0.018 g-% disodium phosphate and 1 liter of 0.9 g-% NaCl containing 0.2 g-% glucose buffered to about pH 6.8 with 0.065 g-% disodium phosphate.
Huggins method-Concentrated red cells were prepared from ACD anticoagulated whole blood after storage for up to 3 days at 4°C. The red cells were glycerolized by adding an equal volume of glycerolizing solution containing 8.6M glycerol, 8% glucose, 1% fructose, and 0.3% disodium ethylenediaminetetracetic acid (Na2 EDTA) (23). The glycerolized red cell mass was slowly frozen to -80°C and stored at that temperature for up to 27 months in a poly(vinyl chloride) plastic container ( Table 2). The red cells were deglycerolized by using the continuousflow centrifugation wash method with a glycerol-lactate-NaCl solution containing 1 liter of 10 g-% glycerol and 5.6 g-% sodium lactate, 2 liters of 3.5 g-% sodium R-lactate, and a liter of 0.9 g-% NaCl containing 0.2 g-% glucose buffered to pH 6.8 with 0.065 g-% disodium phosphate (19).
Rowe method-Concentrated red cells were prepared from ACD-anticoagulated blood after storage at 40C for up to 10 days and from ACD blood stored at 40C for 27-29 days after incubation with PIGPA solution. An equal volume of a solution containing per 100 ml 2.88 g mannitol, 0.65 g NaCl, and 35 g glycerol was added to the red cell volume (20,24). The glycerolized red cells were transferred to a bioriented polyolefin plastic container held in an aluminum container during immersion in liquid nitrogen for 5 min. The red cells were stored for about 19 months in gas phase of liquid nitrogen at -1500C in an LR-1000 Linde liquid nitrogen freezer. After thawing, deglycerolization was accomplished by a continuous-flow centrifugation wash method by use of the ADL-10 washing machine with either reusable stainless steel or rigid disposable polycarbonate bowls. The wash solution consisted of 500 ml of 5.6 g-% sodium Rlactate. The thawed diluted glycerolized red cells were washed with 2 liters of 0.9% NaCl containing 0.2 g-% glucose buffered to pH 6.8 with 0.065 g-% disodium phosphate Platelet-Rich Plasma, Platelet Concentrates, Washed Red Cells Within 4 hr of collection platelet concentrates were prepared from CPD whole blood at room temperature (22 ± 1°C) by centrifugation. The platelet-poor plasma, platelet-rich plasma, and platelet concentrates were stored at either 4°C or 22CC in transfer packs (Fenwal TA-2 and TA-3) made of the plastic PL-146. Some of the units of whole blood were centrifuged at 4500g for 3 min, and the platelet-rich plasma was removed. A 20 ml volume of concentrated red cells was washed with an equal volume of buffered 1.6 g-% sodium chloride on three separate occasions, and the supernatant medium and the washed red cells were analyzed for DEHP.
Platelet counts were made on the plateletrich plasma. The platelet-rich plasma was then certrifuged at 4500g for 5 minutes to concentrate the platelets. The platelet concentrate and the platelet-poor plasma were then stored at 22°C or 40 C for up to 3 days. The platelet concentrate was centrifuged at 6975g for 5 min, and the supernatant of the platelet concentrate was removed for DEHP assay. The platelet button was then suspended in 10 ml of 0.9% nonbuffered isotonic sodium chloride and centrifuged at 6975g for 5 min. The supernatant medium of the washed platelets and the washed platelets themselves were assayed for DEHP. The number of platelets in the platelet button was estimated from the platelet count of platelet-rich plasma and the volume of platelet-rich plasma used to prepare the platelet concentrate.

DEHP in Solutions Stored in Poly(vinyl chloride) (PVC) Plastic Containers, in Other
Plastics, and in Anticoagulants Stored in PVC at Room Temperature Samples were obtained for measurement of DEHP content in sodium chloride solu-Environmental Health Perspectives tion (0.9%) that was stored in poly(vinyl chloride) plastic (PL-146) for up to 6 months at room temperature, and in the rejuvenation solutions (PIGP and PIGPA) that were stored in poly(vinyl chloride) plastic transfer packs for up to 3 months at room temperature. The ACD, CPD, heparin, and sodium citrate anticoagulants that were stored in poly(vinyl chloride) plastic containers (PL-130) at room temperature for up to 427 days were also analyzed for DEHP. Bioriented polyolefin (Hemoflex, Union Carbide Corp.) and polyethylene (Cryocyte, Fenwal Laboratories) were cut into 1-mm squares and put into a 2:1 chloroformmethanol mixture for up to 30 hr at room temperature, and the extracts were analyzed for DEHP.

Preparation of the Samples for DEHP Analysis
Aliquots of platelet-rich plasma, plateletpoor plasma, the supernatant of the platelet concentrates, washed red cells, and the supematant solution of washed red cells and washed platelets were lyophilized. A volume of chloroform-methanol mixture (2:1) was added to the lyophilized sample and mixed by vigorous stirring. The suspension was filtered and the residual material washed with CHCl3:CH30H. The liquid phase was then mixed with an equal volume of distilled water and centrifuged. The upper layer containing the methanol and water was discarded. Silicic acid was added to the remaining chloroform phase and mixed for 1 min with a vortex mixer. After centrifugation, the chloroform was decanted and evaporated, leaving a white powder that was dissolved in reagent-grade methanol. The samples were then stored at 4°C for 1 hr. The clear supernatant prepared by centrifugation was analyzed. The washed platelets were extracted without lyophilization in the manner described above.
Separation of DEHP from methanol was achieved by using a 6-ft column of 3% SE 30 on acid-washed Chromosorb W at 2000C oven temperature, 3000 C injection port temperature, and 3000 C flame detection temperature in a Hewlett-Packard Model 7620A gas chromatograph. Identification of the material extracted as DEHP was made on the basis of retention time and by the addition of DEHP to nonbuffered 0.9 g-% sodium chloride which was extracted in a similar manner (R. J. Jaeger, to be published). Recovery of the additional DEHP averaged 87%. This was quantitated by measuring the area of the peak (calculated by multiplying its width at half the height of the peak by the peak height). The DEHP values were not corrected for the incomplete recovery of DEHP.

Effects of DEHP on Platelet Aggregation in Vitro
Fresh platelet-rich plasma was prepared from sodium citrate-anticoagulated blood obtained from normal volunteers. DEHP (13 g-%) purchased from Polysciences Inc. was mixed with 12-82 mmole methanol. The DEHP-methanol mixture was added to the platelet-rich plasma to achieve final DEHP levels of 5-40 mg-%. The maximum amount of methanol used with the DEHP was added to the platelet-rich plasma to measure the effect of methanol on the aggregation patterns. In vitro aggregation of the platelet-rich plasma with and without DEHP to epinephrine, collagen, and ADP was studied.
Aggregation studies were performed in a modified Kalmedic Instruments Fragiligraph at 370 C by means of the technic of Born (25). Collagen extracts were prepared as described previously (26), and each batch was diluted so that an aliquot of 5 ml produced maximal aggregation with 2 ml of citrated platelet-rich plasma. Adenosine diphosphate, collagen, and epinephrine were diluted in 0.05M imidazole-buffered saline, pH 7.4. Phase microscopy platelet counts of whole blood were performed, and the platelet-rich plasma was counted with the Coulter Counter Model B.

Platelet Viability and Function
Platelet concentrates prepared from CPD whole bood were stored at either 40 C or 220 C for up to 72 hr before transfusion.
The recovery in vivo and lifespan were studied by 51Cr labeling of the platelets as described by Murphy and Gardner (14). The index of therapeutic effectiveness (ITE) was estimated from the recovery in vitro multiplied by the 51Cr recovery in vivo. The hemostatic effectiveness of preserved platelets was evaluated by their ability to correct the prolonged bleeding time produced by administration of 650 mg of aspirin to healthy volunteers. The bleeding time of these volunteers was measured by the procedure described by Mielke et al. (27) before and after aspirin ingestion and after infusion of the platelet concentrates. Twenty-four hours after ingestion of the aspirin the bleeding time was measured, and the volunteer was given autologous stored platelets. The bleeding time was then measured 2, 24, 48, 72, and 96 hr after infusion of the platelet concentrate. The platelets were obtained prior to aspirin treatment and stored in either PL-146 or bioriented polyolefin for at least 24 hr at 40C or at 22°C. After storage of the platelet concentrates for 48 hr at either 40 C or 22°C the supernatant of the platelet concentrates was added to fresh autologous platelets. These platelets were then stored for 24 hr at either 40C or 22°C to determine the effect of the additional DEHP in the supematant on the platelet viability in vivo and the ability to correct a prolonged bleeding time induced by aspirin in healthy individuals (26)(27)(28)(29).

Results
DEHP was recovered in significant quantities from plasma prepared from whole blood stored in poly(vinyl chloride) bags (PL-130) at 40C. The DEHP accumulation was greater in the whole blood stored in the ACD anticoagulant than in that stored in the CPD anticoagulant; 0.360 mg-% DEHP accumulated per day in ACD, and 0.268 mg-% per day in CPD (Fig. 1). The amount of DEHP in CPD whole blood on the day of collection was about half that found in the ACD whole blood; 2.07 mg-% in CPD, and 3.14 mg-% in ACD (Fig. 1). The plasma separated from ACD whole blood had significantly more DEHP on the day of collection than did the plasma obtained from CPD whole blood, and the increase in DEHP in the plasma of ACD whole blood after storage at 40C for up to 30 days was significantly greater than in the units of CPD whole blood stored at 40C for the same period (F = 18.68, p <0.01, n = 94). Table 1 shows that after washing liquidstored ACD whole blood by continuous-flow centrifugation procedures about 0.2 mg DEHP remained in the unit. Removal of about 98% of the DEHP was achieved by the continuous-flow centrifugation procedures. Table 2 shows that after storage in the frozen state at either -80°C or -150°C for at least 2 years in either bioriented polyolefin or poly(vinyl chloride) plastic containers, the amount of DEHP found in the thawed glycerolized red cells containing either 40% (w/v) or 20% (w/v) glycerol correlated with the amount of DEHP present at the time of freezing. The data suggest that the DEHP level did not increase during storage in the frozen state. Tables 3 and 4 show the amount of DEHP in the washed previously frozen red cells preserved by using either 40% (w/v) glycerol and storage at -80°C or 20% (w/v) glycerol and storage at ---150°C contained about 1 mg per unit. The amount of DEHP in the unit prior' to glycerolization and freezing was calculated from data reported in Figure 1. Table 5 shows that no DEHP was found in either the polyethylene or the bioriented polyolefin plastic containers. No DEHP was found after storage of the sodium chloride solution (0.9%) or the PIGP or PIGPA rejuvenation solutions in the poly(vinyl chloride) plastic bags at room temperature for up to 6 months. Table 6 shows that no   DEHP was found in any of the four anticoagulants stored at room temperature for up to 427 days in the poly(vinyl chloride) plastic containers (PL-130).

10, using reusable stainless steel bowls
The supernatant of the platelet concentrates prepared from CPD blood contained the highest concentration of DEHP of all the components reported in Table 7; after storage at 40 C for up to 3 days it contained about five times the amount found in platelet-poor plasma, and about four times the amount in platelet-rich plasma. After storage at 22°C for up to 3 days it contained about nine times as much DEHP as did the platelet-poor plasma, and three times as much as the platelet-rich plasma ( Table  7). There was a threefold increase in the amount of DEHP that accumulated in the supernatant of the platelet concentrate after storage at 220 C over that after storage at 40 C. The washed red cells and washed platelets contained a very small amount of DEHP (Table 7). Figure 2 shows the 51Cr survival of autologous platelet concentrates prepared from CPD whole blood. When platelet concentrates were stored at 22°C for up to 72 hr, the recovery in vivo of 5 1Cr radioactivity Environmental Health Perspectives     5A). Platelet concentrates stored in bioriented polyolefin containing less than 1 mg-% DEHP had 51Cr recovery in vivo and lifespan values similar to those of platelet concentrates stored at 220C for 24 hr with 35 mg-% DEHP (Fig. 5B). Figure 6 shows the effect of storage of platelet concentrates at 40 C for 24 hr in either poly(vinyl chloride) plastic (PL-146) (Fig. 6A, C) or bioriented polyolefin containers (Fig. 6B) on the platelet survival and the ability of the platelet concentrates to correct the prolonged bleeding time induced by aspirin treatment of a healthy volunteer. Three separate studies over a 1-yr period were performed on R.P. (Fig. 6). Platelet concentrates stored at 40 C for 24 (Fig. 6A).

Discussion
Significant quantities of DEHP were present in the plasma of whole blood collected in ACD and in CPD anticoagulants and stored at 4°C. Whole blood collected and stored in the ACD anticoagulant had significantly more DEHP on the day of collection and after storage at 4°C than did the CPD whole blood (Fig. 1). The reason for this difference is not known. Our findings on DEHP accumulation in ACD whole blood were similar to those of Jaeger and Rubin (12). We found very little DEHP in the washed red cells, a finding similar to that of Jaeger and Rubin (12). We found that when whole blood is stored in ACD for about 33 days and washed by continuous-flow centrifugation washing systems, about 98% of the DEHP is removed (Table 1). When red cells were frozen with either 40% (w/v) glycerol or 20% (w/v) glycerol in poly(vinyl chloride) plastic bags or in bioriented polyolefin plastic containers, the amount of DEHP in the supernatant of the thawed glycerolized red cells was the same as that present in the ACD and CPD blood at the time of freezing (Table 2). About 1 mg DEHP remained in the washed previously frozen red cells after storage at -80°C in 40% (w/v) glycerol or at -150°C in 20% (w/v) glycerol (Tables 3 and   4).
No DEHP was found in bioriented polyolefin or polyethylene plastic containers, in the 0.9% sodium chloride solution, the rejuvenation solutions, or in any of the four anticoagulants stored in poly(vinyl chloride) Vplastic bags (Tables 5 and 6). D-r_ Table 7 show that DEHP accumu-I-ted n the platelet-rich plasma and inite uperiiat of the platelet concentrate an the DIKP accumulation was increased after d -t-t the blood components at 220 C. The uper-taiiL ofJ,he platelet concentrates accumulated tIgtest amount of DEHP. Platelet-poor plma,m washed red cells, and washed platelets contained little DEHP; the reason for this is not known. Jaeger and Rubin have suggested that the DEHP is extracted from the poly-(vinyl chloride) plastic container by lipoprotein substances (12). Our data suggest that the platelets themselves are involved in the extraction of DEHP from the poly(vinyl chloride) containers. Our data also suggest that the platelets do not phagocytize the DEHP, since this material accumulated primarily in the supernatant of the platelet concentrates and the washed platelets contained only a small amount (Table 7).
Various investigators have studied the effect of plastic containers on platelet survival and function (8)(9)(10)(11). The accumulation of DEHP in the supernatant of platelet concentrates stored at 4°C and at 22°C for up to 72 hr prompted us to study the effects in vitro of the addition of DEHP to plateletrich plasma. Our data show that DEHP had no significant effects in vitro on the response of fresh platelet-rich plasma to ADP, epinephrine, or collagen (Figs. 3 and 4).
Our data also show that DEHP had no effect on 51Cr platelet survival or on the hemostatic effectiveness of platelets in correcting the prolonged bleeding time induced by aspirin administration to healthy volunteers. When platelet concentrates were stored at 220C for up to 72 hr there was no relation between the lifespan of the platelets and the preservation injury (Fig. 2). The addition of DEHP to the platelet concentrates did not adversely affect the recovery in vivo of 51Cr-labeled platelets (Figs. 5 and 6). These data suggest that the reduced Environmental Health Perspectives recovery in vivo noted during the 2-hr period after the transfusion of platelet concentrates stored at 22°C for up to 72 hr ( Fig. 2) was not related to the DEHP concentration (Fig. 5). In our studies, the presence of DEHP had no apparent effect on platelet function or survival in vivo (Figs. 5 and 6). The accumulation of significant amounts of DEHP observed in whole blood after storage in poly(vinyl chloride) plastic containers may be related to the microaggregates that form in the blood during storage at 40 C (30)(31)(32). The microaggregates may be a factor in the pulmonary insufficiency that occurs after massive transfusion of stored blood, and they may also affect the microcirculation of other organs (30).
Our data show that the accumulation of DEHP during storage of blood products in poly(vinyl chloride) plastic bags does not adversely affect platelet function or survival. However, the potential toxic effects of DEHP in vivo and its metabolism in man have not yet been defined.