Background

Depolymerized holothurian glycosaminoglycan (DHG) is a new agent with anticoagulant properties quite different from those of unfractionated heparin (UFH) and low-molecular-weight heparin (LMWH) in terms of antithrombin III–dependency, and exerts an antithrombotic effect with less bleeding than UFH and LMWH in vivo. In this study, the anticoagulant and hemorrhagic effects of DHG were investigated on hemodialysis in a dog model of renal failure and compared with those of UFH, LMWH, and nafamostat mesilate (FUT).

Methods

The dog renal failure model was prepared by 7/8 renal artery ligation. Effectiveness was based on completion of 3-hour hemodialysis, no marked clot deposition in the extracorporeal circuit, and permeability of blood urea nitrogen (BUN) and creatinine. Template bleeding was measured by determining the hemoglobin content of the blood from the wound.

Results

DHG induced no major bleeding or clot formation during 3-hour hemodialysis, in contrast to UFH and LMWH, each of which induced marked bleeding. These glycosaminoglycans (GAGs) were equally effective in decreasing plasma levels of BUN and creatinine. On the other hand, dogs treated with FUT failed to complete 3-hour hemodialysis. These anticoagulants prolonged activated partial thromboplastin time (APTT) to different extents and GAGs prolonged thrombin clotting time markedly but FUT did not.

Conclusion

Our findings suggest that thrombin clotting time prolongation can contribute to prevention of clot formation in extracorporeal circuits, and the non-antithrombin III–dependent activities of DHG may be related to its low risk of hemorrhage for hemodialysis. DHG appears to be promising as an alternative anticoagulant with low risk of hemorrhage for hemodialysis.

METHODS

Animal model

Male beagle dogs with a body weight between 9 and 12 kg were obtained from Yakken Farm Co., Ltd. These experiments were reviewed and approved by the Animal Ethics Committee of Taiho Pharmaceutical Co., Ltd. (Tokyo, Japan). In total, 7/8 renal mass reduction was performed in two surgeries performed 3 days apart; the first involved mass reduction by ligation (3/4) of the right arterial branch (stage 1), and the second involved total left-side nephrectomy (stage 2). Three days after the second-stage surgery, the model animals were confirmed to have plasma levels of blood urea nitrogen (BUN) elevated to approximately 50 to 100 mg/dL with development of moderate-to-marked azotemia, and were then randomly allocated to groups before hemodialysis.

Materials

The following materials were purchased as indicated: UFH (porcine mucosa; Nacalai Tesque, Inc., Kyoto, Japan), LMWH (porcine intestinal mucosa; Sigma Chemical Co., Ltd., St. Louis, MO, USA), and FUT (Torii Pharmaceutical Co., Ltd., Tokyo, Japan). DHG was prepared in-house from GAG extracted from sea cucumbers by depolymerization using hydrogen peroxide. The specific activities and molecular weight of the GAGs used were as follows: UFH [antithrombin III activity, 205.7 heparin units/mg; antithrombin activity (heparin cofactor II), 184.8 heparin units/mg; antifactor Xa activity (antithrombin III), 191.5 heparin units/mg; activated partial thromboplastin time (APTT) prolongation activity, 3702 DHG units/mg; average molecular weight of 15,000]; LMWH [antithrombin III activity, 58.7 heparin units/mg; antithrombin activity (heparin cofactor II), 21.6 heparin units/mg; antifactor Xa activity (antithrombin III), 74.4 heparin units/mg; APTT prolongation activity, 1033 DHG units/mg; average molecular weight of 8000]; and DHG [antithrombin III activity, 0.0 heparin units/mg; antithrombin activity (heparin cofactor II), 101.9 heparin units/mg; antifactor Xa activity antithrombin III), 0.0 heparin units/mg; APTT prolongation activity, 1013 DHG units/mg; average molecular weight of 12,500]^17,18.

GAGs were administered on a gravimetric basis, except for DHG, which was given on a unit basis with 1 mg/kg DHG equivalent to 1000 DHG units/kg based on APTT prolongation activity determined using a DHG standard.

Hemodialysis design

Anesthesia was induced with sodium pentobarbital (30 mg/kg, intravenously; Abbott Laboratories, North Chicago, IL, USA) and maintained (5 mg/kg/hour) during the experiment. Animals were placed on an operating table with a weighing facility, intubated, and their body temperature kept at 38°C with an electric blanket. Catheters inserted into the right femoral artery and left femoral vein were connected to an extracorporeal hemodialysis circuit system (DBB-22B; Nikkiso Co., Ltd., Tokyo, Japan), with its hollow-fiber dialyzer [A-04H of 0.4 m² (cuprophan); Kawasumi Lab., Tokyo, Japan] perfused with biocarbonate dialysate (Kindaly Solution AF-2; Fuso, Osaka, Japan). The extracorporeal circuit was primed prior to dialysis with UFH (2000 IU/L) containing saline. The blood flow rate was kept at 50 mL/minute, and the dialysate flow rate at 330 mL/minute. During hemodialysis, each arterial and venous chamber was connected to a spectrumed DTX-PLUS transducer (Becton Drive, Franklin Lakes, NJ, USA) to monitor pressure, which was recorded using a RMP-6018M 8-channel dymograph (Nihon Kohden, Tokyo, Japan). Another catheter was inserted into the left femoral artery and blood pressure was monitored in identical fashion.

Previously, we performed a preliminary study to establish the minimum effective doses of DHG and other drugs on hemodialysis for 5 hours in normal dog. It was possible to complete it with DHG at a dose of 5 mg/kg bolus + 2.5 mg/kg/hour drip infusion but not at 3 mg/kg bolus + 1.5 mg/kg/hour drip infusion; it could be completed with UFH at a dose of 1 mg/kg bolus + 0.5 mg/kg/hour drip infusion but not at 0.5 mg/kg bolus + 0.25 mg/kg/hour drip infusion; it could be completed with LMWH at a dose of 3 mg/kg bolus + 1.5 mg/kg/hour drip infusion but not at 1 mg/kg bolus + 0.5 mg/kg/hour drip infusion; it could be completed with FUT at a dose of 3 mg/kg/hour drip infusion but not at 1 mg/kg/hour drip infusion. To confirm the reproducibility, these hemodialysis studies were performed twice at each dose. Based on these results, we established the minimum effective doses for normal dog¹⁹. In this study, hemodialysis was carried out at the minimum effective dose of each anticoagulant as follows: DHG, 5 mg/kg bolus + 2.5 mg/kg/hour drip infusion; UFH, 1 mg/kg bolus + 0.5 mg/kg/hour drip infusion; LMWH, 3 mg/kg bolus + 1.5 mg/kg/hour drip infusion; and FUT, 3 mg/kg/hour drip infusion. DHG, UFH, and LMWH were administered as a bolus injection into the femoral vein followed by continuous infusion into the inlet of the circuit line. FUT was directly infused continuously into the extracorporeal circuit⁵ to avoid degradation in the dog. In the control group, 10 mL saline was intravenously injected as a bolus followed by continuous saline infusion (10 mL/hour) instead of anticoagulants.

The possible hemodialysis period (until the difference in pressure values between the inlet and outlet of the dialyzer reached 200 mm Hg) was determined and limited the maximum duration to 3 hours. Following dialysis, the inlet of the line was clamped, removed from the catheter, and attached to a saline bag in order to push the blood in the lines back into the dog. The venous line was then disconnected, both wounds were sutured, and the dog was returned to its cage.

At the end of hemodialysis, residual blood clot in the extracorporeal circuit was determined. Residual blood clot in the dialyzer was classified into four grades based on visual assessment as follows: 0 = no clot formation observed; 1 = clot formation observed in several fibers; 2 = clot deposits in hollow fibers in a fasciculus; 3 = clot formation in about half of the fibers; and 4 = dialysis not maintainable due to clot formation. Clot deposits remaining in the drip chamber were expressed as the hemoglobin content if blood clot was present in it. The blood clot was hemolyzed by sonication in a defined volume of saline, and its hemoglobin content measured using a hemoglobin kit (Hemoglobin B Test Wako, Wako Pure Chemical Co., Osaka, Japan).

Laboratory tests

Before, during, and after hemodialysis, blood was taken from the femoral artery into 3.8% sodium citrate (1 volume to 9 volumes of blood) or a sample bottle containing ethylenediaminediaminetetraacetic acid (EDTA) (Sysmex; Toa Medical Electronics Company, Ltd., Kobe, Japan). Plasma was separated by centrifugation at 1500 g for 10 minutes at 4°C, and stored at -80°C until assay.

APTT and thrombin clotting time were determined using a commercially available APTT reagent (Platelin® LS, Organon Teknika Co., Durham, NC, USA) and dog thrombin (final 1 unit/mL) purified in our laboratory^21,22.

White blood cell count, hematocrit, and platelet count were measured with an automatic analyzer (NE-6000, Toa Medical Electronics Company, Ltd.). BUN was measured using a Wako BUN Test and creatinine using a Wako CRN Test (Wako Pure Chemical Co., Osaka, Japan). Triglyceride was measured using a Wako Triglyceride E-HA Test (Wako Pure Chemical Co.) and nonesterified fatty acid (NEFA) using a Wako NEFA-HA Test (Wako Pure Chemical Co.).

At the end of dialysis, template bleeding of the ear surface was measured as follows. A template device (Simplate®, Organon Teknika Co., Durham, MC, USA) was applied to the surface of the ear, with care taken to avoid large veins. Blood from the wound was then carefully taken up with filter paper every 1 minute until complete arrest of bleeding for 1 minute. If bleeding from the incisions had not stopped after 120 minutes, measurement was stopped at that time. The paper was placed in a defined volume of saline and hemoglobin content was determined using a hemoglobin kit (Wako Hemoglobin B Test; Wako Pure Chemical Co.) and bleeding expressed as blood loss.

GAG plasma concentrations

GAG plasma concentrations were determined using a DS kit (Stago Diagnostica, Asnieres, France) by measuring antithrombin activity in the presence of bovine heparin cofactor II. Each GAG was used as the calibrator.

Statistical analysis

All group data were expressed as mean values and SD unless stated otherwise. Differences between groups in residual blood clot in the dialyzer and clot deposition in drip chambers were tested statistically by the Wilcoxon test. Student paired t test was used for analysis of blood loss. All other statistical testing was performed by one-way analysis of variance (ANOVA) followed by Dunnett's test. A P value of less than 0.05 was considered to be statistically significant.

RESULTS

Efficacy and safety of anticoagulants in experimental hemodialysis in dogs with renal failure

When hemodialysis in dogs with renal failure was conducted without an anticoagulant (control), the difference in pressure between the inlet and outlet of the dialyzer reached 200 mm Hg within approximately 60 minutes. With DHG (5 mg/kg bolus + 2.5 mg/kg/hour drip infusion), with UFH (1 mg/kg bolus + 0.5 mg/kg/hour drip infusion), and with LMWH (3 mg/kg bolus + 1.5 mg/kg/hour drip infusion), 3-hour hemodialysis was completed in all five dogs of each group Table 1. With FUT (3 mg/kg/hour drip infusion), the difference in pressure values between the inlet and outlet of the dialyzer reached 200 mm Hg. Additionally, even when hemodialysis with FUT (5 and 10 mg/kg/hour drip infusion) was performed twice at each dose, no dogs could undergo an entire 3 hour hemodialysis (data not shown).

Table 1 - The mean period of hemodialysis and the percent efficacy of dialysis with DHG, UFH, LMWH, and FUT.

Full table

With hemodialysis using DHG, UFH, and LMWH, the difference in pressure values between the inlet and outlet of the dialyzer was stably maintained at about 150 mm Hg throughout the 3-hour dialysis period (data not shown). The average duration of stable hemodialysis and effects as an anticoagulant on hemodialysis to reduce creatinine and BUN are shown in Table 1. Dialysis with DHG was completed effectively, as well as those with UFH or LMWH, since the percent decreases in plasma creatinine and BUN were equivalent with these GAGs. With FUT and saline control, 3-hour hemodialysis could not be completed, resulting in failure to reduce creatinine and BUN. There was no difference between FUT and the control in either possible period of hemodialysis or percentage decreases of creatinine and BUN. Table 2 shows values for residual blood clots in the dialyzer. Large thrombotic patches in the dialyzer plate were observed in the control and FUT groups. DHG, UFH, and LMWH decreased blood clots in the dialyzer, and the scores for them were below the control value. Figure 1 shows the amounts of blood clot remaining in the inlet and outlet of drip chambers. Significant clots were formed within the bubble trap in the control group. DHG, similar to UFH and LMWH, inhibited the clot formation in the chamber of the extracorporeal circuit. In contrast to the degree of residual clot in the dialyzer, blood clot in the chamber on both sides was less in the FUT group than in the control. These results indicate that FUT was unable to maintain extracorporeal circulation because of the complete occlusion of the dialyzer.

Figure 1.

Anticoagulant effects of depolymerized holothurian glycosaminoglycan (DHG), unfractionated heparin (UFH), low-molecular-weight heparin (LMWH), and nafamostat mesilate (FUT) on clot deposition in drip chambers of the extracorporeal circuit. The amount of residual blood clot in each chamber is presented as hemoglobin content (see Methods section). Results indicate the mean SD (N = 5). *P < 0.05, **P < 0.01 vs. control by Wilcoxon's test. There is no difference among the glycosaminoglycans (GAGs).

Full figure and legend (46K)

Table 2 - Anticoagulant effects of DHG, UFH, LMWH, and FUT in the dialyzer extracorporeal circuit.

Full table

Laboratory tests

The effects of anticoagulants on APTT and thrombin clotting time in blood obtained from the inlet of a dialysis circuit during hemodialysis are shown in Figure 2. The baseline values of APTT and thrombin clotting time in dogs with renal failure (APTT, 13.1 seconds; thrombin clotting time, 15.6 seconds on average) tended to be slightly shortened than in normal dogs (APTT, 14.0 seconds; thrombin clotting time, 16.9 seconds on average). At 1 hour after injection, APTT was prolonged by DHG to about 5.4 times (70.9 seconds) the baseline value. UFH prolonged APTT to 33.6 seconds, LMWH to 41.4 seconds, and FUT to 56.5 seconds. FUT resulted in an APTT of more than 200 seconds in the dialysis circuit (sampling at the inlet of the dialyzer). DHG, UFH, and LMWH prolonged thrombin clotting time more than 200 seconds during hemodialysis. FUT only minimally prolonged thrombin clotting time in the inlet and outlet of the circuit. Even with high doses (5 and 10 mg/kg/hour drip infusion), FUT only slightly prolonged thrombin clotting time (data not shown).

Figure 2.

Effects on activated partial thromboplastin time (APTT) (A) and thrombin clotting time (B) of depolymerized holothurian glycosaminoglycan (DHG), unfractionated heparin (UFH), low-molecular-weight heparin (LMWH), and nafamostat mesilate (FUT). All samples were taken directly from the femoral artery except as indicated. Each point indicates the mean SD of five dogs. Normal APTT, 13.1 seconds; normal thrombin clotting time, 14.9 seconds. Symbols are: () saline control; () DHG, 5 mg/kg bolus + 2.5 mg/kg/hour drip infusion; () UFH, 1 mg/kg bolus + 0.5 mg/kg/hour drip infusion; () LMWH, 3 mg/kg bolus + 1.5 mg/kg/hour drip infusion; () FUT, 3 mg/kg/hour drip infusion; () FUT, 3 mg/kg/hour drip infusion (sampled from extracorporeal circuit, inlet).

Full figure and legend (33K)

Effects on changes in white blood cell count, hematocrit, and platelet count during hemodialysis are shown in Figure 3. Although a transient decrease in platelet count was observed in surgical stage 2, platelet count recovered on the day of hemodialysis. During hemodialysis, platelet count exhibited no significant differences among GAGs, but constantly decreased during hemodialysis using FUT until 1 hour after injection. No significant differences among anticoagulants were observed in changes in white blood cell count. Compared with that in normal dogs, white blood cell count in dogs with renal failure tended to increase. In our hemodialysis model, leukopenia in the early phase of the hemodialysis session and rebound was observed, as in clinical hemodialysis²³. Although hematocrit did not differ greatly among GAGs during hemodialysis, it tended to decrease in the UFH and LMWH groups at 24 hours after hemodialysis. However, there was no difference among the GAGs in changes in hematocrit.

Figure 3.

Effects on platelet count (A), white blood cell count (B), and hematocrit (C) of depolymerized holothurian glycosaminoglycan (DHG), unfractionated heparin (UFH), low-molecular-weight heparin (LMWH), and nafamostat mesilate (FUT). Each point indicates the mean SD of five dogs. Symbols are: () saline control; () DHG, 5 mg/kg bolus + 2.5 mg/kg/hour drip infusion; () UFH, 1 mg/kg bolus + 0.5 mg/kg/hour drip infusion; () LMWH, 3 mg/kg bolus + 1.5 mg/kg/hour drip infusion; () FUT, 3 mg/kg/hour drip infusion. *P < 0.05, **P < 0.01 vs. control by Dunnet's test, and there is no difference among the glycosaminoglycans (GAGs).

Full figure and legend (35K)

Regarding lipid levels, nonesterified fatty acids transiently increased during hemodialysis using GAGs and FUT but remained nearly within the normal range (0.3 to 0.7 meq/L) (data not shown). No significant change in triglyceride was observed during hemodialysis with any anticoagulant (data not shown). There was no difference among the four anticoagulants in either nonesterified fatty acids or triglyceride levels.

The effects of anticoagulants on bleeding during hemodialysis are shown in Figure 4. In our study, blood loss was almost the same in normal dogs as in those with renal failure. Although DHG had no marked effect on bleeding during hemodialysis, UFH and LMWH each significantly enhanced bleeding during hemodialysis. The bleeding time with DHG was 6.9 1.0 minutes; that with UFH was 60.0 39.4 minutes, and with LMWH it was 94.0 26.1 minutes, with prehemodialysis values of 5.7 0.9 minutes (mean SD).

Figure 4.

Hemorrhagic effects of depolymerized holothurian glycosaminoglycan (DHG), unfractionated heparin (UFH), and low-molecular-weight heparin (LMWH). The template device was applied to the ear surface with care taken to avoid large veins, and blood from the wound was collected. Blood loss is expressed as hemoglobin content (see Methods section). Template bleeding was measured before surgery performed to induce renal failure (stage 1), prehemodialysis (HD-pre), and at 3 hours after the start of hemodialysis (HD-3 hr). Abbreviations are: D, DHG, 5 mg/kg bolus + 2.5 mg/kg/hour drip infusion; U, UFH, 1 mg/kg bolus + 0.5 mg/kg/hour drip infusion; L, LMWH, 3 mg/kg bolus + 1.5 mg/kg/hour drip infusion. Results indicate the mean SD (N = 5). *P < 0.05 vs. prehemodialysis values by paired t test.

Full figure and legend (17K)

GAG plasma concentrations

Figure 5 shows GAG plasma concentrations during and 24 hours after hemodialysis. Plasma concentrations of DHG, UFH, and LMWH during hemodialysis when these agents were administered as a bolus injection into the femoral vein followed by continuous infusion directly into the extracorporeal circuit were 73.4 g equivalent/mL, 13.9 g equivalent/mL, and 48.3 g equivalent/mL, respectively, at 3 hours after the start of hemodialysis. GAGs did not pass through the membrane of the dialyzer nor did they interact with it, since there were no differences in GAG plasma levels between the inlet and outlet of the dialyzer Figure 5.

Figure 5.

Plasma concentrations of depolymerized holothurian glycosaminoglycan (DHG) (A), unfractionated heparin (UFH) (B), and low-molecular-weight heparin (LMWH) (C) during hemodialysis in dogs with renal failure. Each point indicates the mean SD (N = 5). Plasma concentrations of DHG, UFH, and LMWH were assayed by measuring antithrombin activity in the presence of bovine heparin cofactor II (see Methods section). Symbols are: () sampled directly from the femoral artery; () sampled from the extracorporeal circuit (ahead of the dialyzer); () sampled from the extracorporeal circuit (beyond the dialyzer).

Full figure and legend (25K)

DISCUSSION

In the present study, we investigated the efficacy and safety of DHG as an anticoagulant for hemodialysis in an animal model of renal failure in order to mimic clinical conditions, and compared it to UFH, LMWH, FUT, and the control. At the minimum effective dose of each anticoagulant determined previously in normal dogs¹⁹, 3-hour hemodialysis was completed with DHG, UFH, and LMWH. The percent decreases in plasma creatinine and BUN during hemodialysis were equivalent for these GAGs, indicating that dialysis with DHG was completed effectively as well as those with UFH or LMWH. Moreover, effective dialysis with DHG was achieved with a low risk of bleeding, in contrast to UFH and LMWH, which featured large amounts of bleeding Figure 4, consistent with the findings of our previous studies in rats¹⁸ and dogs¹⁹. As mentioned above, the antithrombin III–independent inhibitory effects of DHG on the blood coagulation system differ markedly from those of both heparins. Although it has been speculated that prolonged bleeding caused by GAGs in animals correlates with antithrombin activity¹⁸, the difference in risk of hemorrhage may depend on thrombin inhibition via antithrombin III. Dermatan sulfate is known to accelerate thrombin inhibition by heparin cofactor II²⁴, and this is probably the main anticoagulant effect of this GAG. In one study, dermatan sulfate functioned as an effective anticoagulant agent for hemodialysis in patients with chronic renal failure²⁵. In that study, dermatan sulfate prolonged thrombin clotting time, but did not increase bleeding, supporting our hypothesis. Although in the present study, both heparins induced severe bleeding, this is rarely seen clinically. This may be because doses of both of the heparins used in dogs were much higher than those used in humans. Moreover, the enhancement of fibrinolysis by heparins^26,27 may also be related to the long bleeding time observed in this study.

In contrast to GAGs, no dogs in which FUT was used could undergo 3 hours of hemodialysis. Patients with renal failure are in a hypercoagulable state and hemodialysis promotes this state^28,29. Therefore, hemodialysis in dogs with renal failure may require more powerful anticoagulation treatment than in normal dogs. Based on a previous study¹⁹ and the present study, we speculate that prolongation of thrombin clotting time is a key factor in achievement of hemodialysis in our beagle model. Although sufficient prolongation of APTT by FUT was observed, FUT only minimally prolonged thrombin clotting time in the extracorporeal circuit in our hemodialysis model. Therefore, it was assumed that FUT was unable to inhibit blood clot formation in the dialyzer because of incomplete inhibition of thrombin. In agreement with our results, Matsuo et al³⁰ have reported that despite sufficient prolongation of APTT in the circuit, FUT was less effective in suppressing thrombin generation than UFH. From these findings, FUT might fail to allow completion of hemodialysis. On the other hand, this issue does not occur in humans. There are three possible reasons. First, the comparative activity of thrombin of dogs is higher than that of humans. Furthermore, it is supposed that renal failure status works more hypercoagulability than normal status, especially in hemodialysis in dogs with renal failure. Therefore, strong antithrombin activity may be needed to prevent thrombosis in the dialysis circuit in renal failure dogs. Second, dialyzer blood flow in dogs is much lower than in humans, which might increase clotting tendency. Finally, this study was performed with a cuprophan membrane, which might on its own induce coagulation, and the results might have been different with the use of more biocompatible membranes.

In the clinical use of anticoagulants, some problems include clearance of anticoagulant in renal failure, availability of antidotes, and monitoring tests. In this study, we measured GAG plasma concentrations during and at 24 hours after hemodialysis. Despite use of the same doses as in normal dogs, plasma concentrations of DHG and LMWH at 3 hours after the start of hemodialysis were higher than those in normal dogs (DHG, 38.2 versus 73.4 g equivalent/mL; LMWH, 32.9 versus 48.3 g equivalent/mL)¹⁹. These findings suggest that DHG and LMWH were retained in the blood of dogs with renal failure. However, they had been completely cleared by 24 hours after the hemodialysis session. DHG is partially depolymerized by cells of the monocyte-macrophage system and its products of degradation are excreted by the kidneys (data not shown). As there are some reports of delayed clearance of LMWH in renal failure with manifestation of adverse effects^31,32, careful pharmacokinetic studies of DHG in animals and patients with renal failure are required. With respect to neutralization of DHG, our previous study demonstrated that prolongation of bleeding time by overdose of DHG was neutralized by protamine sulfate with concomitant normalization of thrombin clotting time, and suggested that protamine sulfate may be useful as an antidote for DHG to prevent bleeding in cases of an overdose³³. Regarding the monitoring of DHG, APTT prolongation by DHG in plasma correlates with that in whole blood in a dose-dependent manner (0.3 to 10 g/mL), suggesting that APTT prolongation in whole blood can be used to monitor the effect of DHG (data not shown). In addition, immunogenicity and antigenicity of DHG were evaluated by the active systemic anaphylaxis (ASA) test and passive cutaneous anaphylaxis (PCA) test using guinea pigs and mice/rats. Neither immunogenicity nor antigenicity was observed in these animals after repeated immunization with DHG (data not shown). However, since some cases of hypersensitivity to heparinoid have been reported³⁴, subcutaneous testing of DHG may be necessary to determine whether this agent is acceptable for patients.

In the present study, it was demonstrated that DHG functioned as effectively as UFH and LMWH as an anticoagulant for experimental hemodialysis in dogs with renal failure, and that it had no major risk of inducing bleeding, unlike UFH and LMWH. These anticoagulants markedly extended thrombin clotting time, but FUT, which yielded insufficient thrombin clotting time prolongation, failed to enable completion of 3-hour hemodialysis. These findings suggest that efficacy of DHG is associated with thrombin clotting time prolongation and that its non-antithrombin III–dependent effects are related to its low risk of inducing bleeding. DHG should thus be useful as an anticoagulant for hemodialysis.

References

1. Swartz, RD: Hemorrhage during high-risk hemodialysis using controlled heparinization. Nephron 1981 28: 65–69,  | PubMed | ISI | ChemPort |
2. Agresti, J, Conroy, JD, Olshan, A, et al: Heparin-free hemodialysis with cuprophan hollow fiber dialyzer by a frequent-saline-flush, high-blood-flow technique. Am Soc Artif Intern Organs 1985 31: 590–594,  | ChemPort |
3. Bergrem, H, Leivestad, T: Dialysis death and increased free fatty acids. Lancet 1978 2: 1160,  | PubMed | ISI | ChemPort |
4. Kinugasa, E, Kojima, H, Tamura, K, et al: Utility of FUT-175, an enzyme inhibitor, in hemodialysis—Its anticoagulative and anticomplement actions. J Jpn Soc Dial Ther 1982 15: 546,
5. Akizawa, T, Sato, M, Kitaoka, T, et al: The usefulness of the multienzyme inhibitor nafamostat mesilate in high-bleeding-risk hemodialysis. Proc Eur Dial Transplant Assoc 1985 22: 334–338,
6. Akizawa, T, Koshikawa, S, Ota, K, et al: Nafamostat mesilate: A regional anticoagulant for hemodialysis in patients at high risk for bleeding. Nephron 1993 64: 376–381,  | PubMed | ISI | ChemPort |
7. Leu, JG, Chiang, SS, Lin, SM, et al: Low molecular weight heparin in hemodialysis patients with a bleeding tendency. Nephron 2000 86: 499–501,  | Article | PubMed | ISI | ChemPort |
8. Koshikawa, S, Akizawa, T, Mimura, N, et al: Effects of low molecular weight heparin as an anticoagulant for hemodialysis patients with bleeding risk. Multiinstitutional control study with low dose or regional heparinization. Clin Eval 1991 19: 541–571,
9. Akizawa, T, Kitaoka, T, Sato, M, et al: Comparative clinical trial of regional anticoagulation for hemodialysis. Am Soc Artif Intern Organs Trans 1988 34: 176–178,  | ChemPort |
10. Kitagawa, H, Chang, H, Fujita, T: Hyperkalemia due to nafamostat mesilate. N Engl J Med 1995 332: 687,  | Article | PubMed | ISI | ChemPort |
11. Higuchi, N, Yamazaki, H, Kikuchi, H, et al: Anaphylactoid reaction induced by a protease inhibitor, nafamostat mesilate, following nine administrations in a hemodialysis patient. Nephron 2000 86: 400–401,  | Article | PubMed | ISI | ChemPort |
12. Ueda, A, Nagase, S, Morito, N, et al: Anaphylactoid reaction induced by low-molecular-weight heparin in a hemodialysis patient. Nephron 2001 87: 93–94,  | Article | PubMed | ISI | ChemPort |
13. Gerlach, AT, Pickworth, KK, Seth, SK, et al: Enoxaparin and bleeding complications: A review in patients with and without renal insufficiency. Pharmacotherapy 2000 20: 771–775,  | Article | PubMed | ISI | ChemPort |
14. Yoshida, K, Minami, Y, Nemoto, H, et al: Structure of DHG, a depolymerized glycosaminoglycan from sea cucumber, Stichopus japonicus. Tetrahedron Lett 1992 33: 4959–4962,  | Article | ISI | ChemPort |
15. Nagase, H, Enjoji, K, Minamiguchi, K, et al: Depolymerized holothurian glycosaminoglycan with novel anticoagulant actions: Antithrombin III- and heparin cofactor II-independent inhibition of factor X activation by factor IXa-factor VIIIa complex and heparin cofactor II-dependent inhibition of thrombin. Blood 1995 85: 1527–1534,  | PubMed | ISI | ChemPort |
16. Nagase, H, Enjoji, K, Shima, M, et al: Effect of depolymerized holothurian glycosaminoglycan (DHG) on the activation of factor VIII and factor V by thrombin. J Biochem 1996 119: 63–69,  | PubMed | ISI | ChemPort |
17. Nagase, H, Kitazato, KT, Sasaki, E, et al: Antithrombin III-independent effect of a new depolymerized holothurian glycosaminoglycan (DHG) on acute thromboembolism in mice. Thromb Haemost 1997 77: 399–402,  | PubMed | ISI | ChemPort |
18. Kitazato, K, Kitazato, KT, Nagase, H, et al: DHG, a new depolymerized holothurian glycosaminoglycan, exerts an antithrombotic effect with less bleeding than unfractionated or low-molecular-weight heparin, in rats. Thromb Res 1996 84: 111–120,  | Article | PubMed | ISI | ChemPort |
19. Minamiguchi, K, Kitazato, KT, Sasaki, E, et al: The anticoagulant and hemorrhagic effects of DHG, a new depolymerized holothurian glycosaminoglycan, on experimental hemodialysis in dogs. Thromb Haemost 1997 77: 1148–1153,  | PubMed | ISI | ChemPort |
20. Minamiguchi, K, Nagase, H, Kitazato, KT, et al: Interaction of a new depolymerized holothurian glycosaminoglycan with proteins in human plasma. Thromb Res 1996 83: 253–264,  | Article | PubMed | ISI | ChemPort |
21. Hashimoto, N, Morita, T, Iwanaga, S: A method for systematic purification from bovine plasma of six vitamin-K-dependent coagulation factors: Prothrombin, factor X, factor IX, protein S, protein C, and protein Z. J Biochem 1985 97: 1347–1355,  | PubMed | ISI | ChemPort |
22. Owen, WG, Esmon, CT, Jackson, CM: The conversion of prothrombin to thrombin. J Biol Chem 1974 249: 594–605,  | PubMed | ISI | ChemPort |
23. Craddock, PR, Fehr, J, Dalmasso, AP, et al: Hemodialysis leukopenia. Pulmonary vascular leukostasis resulting from complement activation by dialyzer cellophane membranes. J Clin Invest 1977 59: 879–888,  | PubMed | ISI | ChemPort |
24. Tollefsen, DM, Petska, CA, Monafo, WJ: Activation of heparin cofactor II by dermatan sulfate. J Biol Chem 1983 258: 6713–6716,  | PubMed | ISI | ChemPort |
25. Lane, DA, Ryan, K, Ireland, H, et al: Dermatan sulphate in hemodialysis. Lancet 1992 339: 334–335,  | Article | PubMed | ISI | ChemPort |
26. Porta, R, Pescador, R, Niada, R, et al: Profibrinolytic activities of chemically modified heparins with very low anticoagulant activities. Thromb Res 1989 53: 435–445,  | Article | PubMed | ISI | ChemPort |
27. Poplawski, A, Glowinski, S: The effect of heparin on the plasma fibrinolytic activity. Acta Physiol Pol 1974 25: 39–42,  | PubMed | ISI | ChemPort |
28. Pop Al-Saady, NM, Leatham, EW, Gupta, S, et al: Monocyte expression of tissue factor and adhesion molecules: The link with accelerated coronary artery disease in patients with chronic renal failure. Heart 1999 81: 134–140,  | PubMed | ChemPort |
29. Ambuhl, PM, Wuthrich, RP, Korte, W, et al: Plasma hypercoagulability in hemodialysis patients: Impact of dialysis and anticoagulation. Nephrol Dial Transplant 1997 12: 2355–2364,  | PubMed | ISI | ChemPort |
30. Matsuo, T, Kario, K, Nakao, K, et al: Anticoagulation with nafamostat mesilate, a synthetic protease inhibitor, in hemodialysis patients with a bleeding risk. Haemostasis 1993 23: 135–141,  | PubMed | ISI | ChemPort |
31. Busby, LT, Weyman, A, Rodgers, GM: Excessive anticoagulation in patients with mild renal insufficiency receiving long-term therapeutic enoxaparin. Am J Hematol 2001 67: 54–56,  | Article | PubMed | ISI | ChemPort |
32. Smith, BS, Gandhi, PJ: Pharmacokinetics and pharmacodynamics of low-molecular weight heparins and glycoprotein IIb/IIIa receptor antagonists in renal failure. J Thromb Thrombolysis 2001 11: 39–48,  | Article | PubMed | ISI | ChemPort |
33. Sasaki, E, Minamiguchi, K, Kitazato, KT, et al: Neutralization of DHG, a new depolymerized holothurian glycosaminoglycan, by protamine sulfate and platelet factor 4. Haemostasis 1997 27: 174–183,  | PubMed | ISI | ChemPort |
34. Grassegger, A, Fritsch, P, Reider, N, et al: Delayed-type hypersensitivity and cross-reactivity to heparins and danaparoid: A prospective study. Dermatol Surg 2001 27: 47–52,  | Article | PubMed | ISI | ChemPort |

Acknowledgments

We thank Tomio Hirota, Safety Research Laboratory, Taiho Pharmaceutical Co., Ltd., for skillful animal care, and Professor Satoshi Teraoka, Department of Surgery Kidney Center, Tokyo Woman's Medical College, and Professor Fumiaki Marumo, Second Department of Internal Medicine, Tokyo Medical and Dental University School of Medicine, for their invaluable advice. The authors are indebted to Professor J. Patrick Barron of the International Medical Communications Center of Tokyo Medical University for his review of this manuscript.