Safety and Tolerability of Nafamostat Mesilate And

bs_bs_banner

Therapeutic Apheresis and Dialysis 2016; 20(2):197–204 doi: 10.1111/1744-9987.12357 © 2016 The Authors. Therapeutic Apheresis and Dialysis published by John Wiley & Sons Australia, Ltd on behalf of International Society for Apheresis, Japanese Society for Apheresis, and Japanese Society for Dialysis Therapy

Safety and Tolerability of Nafamostat Mesilate and Heparin as Anticoagulants in Leukocytapheresis for Ulcerative Colitis: Post Hoc Analysis of a Large-Scale, Prospective, Observational Study

Koji Sawada,1 Maiko Ohdo,1 Tomoko Ino,2 Takashi Nakamura,2 Toyoko Numata,2 Hiroshi Shibata,2 Jun-ichi Sakou,2 Masahiro Kusada,2 and Toshifumi Hibi3

1Dojima General & Gastroenterology Clinic, Osaka, 2Scientiﬁc and Technical Affairs Department, Japan Operation Division, Blood Puriﬁcation Business Unit, Asahi Kasei Medical Co. Ltd., and 3Center for Advanced IBD Research and Treatment, Kitasato University, Kitasato Institute Hospital, Tokyo, Japan

Abstract: Nafamostat mesilate is the first anticoagulant of reactions (8.6% vs. 7.1%) and intrafilter pressure in- choice for leukocytapheresis (LCAP) with a Cellsorba E creases (12.7% vs. 16.8%) between the nafamostat column for treating ulcerative colitis (UC). However, mesilate and heparin groups. Adverse reactions of hemor- because of complications, mainly due to allergy to rhage or blood pressure decreases associated with heparin nafamostat mesilate, heparin may be used as a substitute. use were not observed. There were no significant differ- To evaluate the safety and tolerability of nafamostat ences in rates of clinical remission (69.1% vs. 68.1%) and mesilate and heparin as anticoagulants in LCAP for UC, mucosal healing (62.9% vs. 63.6%) between the we conducted post hoc analysis of data from a large- nafamostat mesilate and heparin groups. Thus, the safety scale, prospective, observational study of LCAP, which and tolerability were comparable in the nafamostat was conducted at 116 medical facilities in Japan between mesilate and heparin groups, indicating that both May 2010 and December 2012. Of 832 patients included nafamostat mesilate and heparin can be well tolerated as in this analysis, nafamostat mesilate and heparin were anticoagulants in LCAP for UC. Key Words: Anticoag- used in 676 (81.3%) and 113 (13.6%), respectively. There ulant, Heparin, Leukocytapheresis, Nafamostat mesilate, were no significant differences in the incidence of adverse Ulcerative colitis.

Ulcerative colitis (UC) is a disease of unknown or- We established previously that leukocytapheresis igin that causes chronic inflammation in the mucosa (LCAP), a type of apheresis, is an effective treat- of the colon. The clinical symptoms are recurrent or ment for active UC (6,7). LCAP is an extracorpo- continuous diarrhea with bloody mucus in the stool, real circulation therapy that uses the Cellsorba E abdominal pain, and the presence of diffuse lesions column (Asahi Kasei Medical, Tokyo, Japan), that originate in the rectum and typically advance which is filled with a nonwoven polyester fiber that continuously in a proximal direction (1). removes activated leukocytes and activated plate- The main medical treatment for active UC is the lets from the peripheral blood to elicit anti- oral and topical administration of 5-aminosalicylic acid inflammatory effects (8–10). LCAP for active UC (5-ASA) (2); however, corticosteroids are often used received approval for medical insurance coverage in patients refractory to 5-ASA (3). Patients who are in Japan in 2001. also refractory to corticosteroids receive concurrent Our previous study revealed that the efficacy rate immunosuppressants or biological agents (4,5). of LCAP was 74% among 39 patients with steroid- resistant UC (6). In addition, the efficacy was 80% in a double-blind sham column control study of 19 Received March 2015; revised June 2015. patients with steroid-resistant active UC, which indi- Address correspondence and reprint requests to Dr. Koji Sawada, fi fi Dojima General & Gastroenterology Clinic, 2-4-27 Dojima, Kita-ku, cated that LCAP had a signi cantly higher ef cacy Osaka 530–0003, Japan. Email: [email protected] than a sham column (7).

197 This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modiﬁcations or adaptations are made. 198 K Sawada et al.

These studies used nafamostat mesilate, which has remove leukocytes. LCAP was performed 5–10 times a relatively short half-life, as the anticoagulant in the during the treatment period with a blood flow rate of LCAP treatment because UC is a hemorrhagic ill- 30–50 mL/min and a blood processing volume ness. In addition, polyester fiber has a negative elec- ≥30 mL/kg bodyweight. tric charge, which may induce the production of bradykinin in blood and thereby decrease blood Assessment of treatment outcomes pressure (11). Because nafamostat mesilate can in- The surveyed patient demographic data included hibit the production of bradykinin (12), it may be age, weight, gender, UC disease duration, response more suitable as an anticoagulant during LCAP to corticosteroids, and concomitant medications. treatment. However, several studies have reported The information from each LCAP session included that nafamostat mesilate is associated with adverse the date that LCAP was performed, the amount of events such as allergic symptoms and anaphylactic processed blood, the anticoagulant used, the reason shock (13,14). Thus, in an actual clinical setting, hep- for the change, the dose if the anticoagulant was arin could be used as a substitute anticoagulant in pa- changed, and the reason for discontinuation of tients who develop adverse events associated with LCAP where applicable. The following reasons for nafamostat mesilate. However, the tolerability of the discontinuation of LCAP were defined as heparin as an anticoagulant in LCAP for UC has “intrafilter pressure increase”: increased intrafilter not been investigated using data from a large-scale pressure or blood clotting. In addition, all adverse study. events noted during the observation period were re- We previously conducted a large-scale, prospec- corded in the CRFs, and any events for which a tive, observational study to evaluate the safety and causal relationship with LCAP could not be ruled efficacy of LCAP for active UC (15). In this large- out were considered adverse reactions (ARs) to scale study, nafamostat mesilate was used as the anti- LCAP. The AR terminology followed the Medical coagulant for LCAP treatment in 676 patients Dictionary for Regulatory Activities (MedDRA/J (81.3%), while heparin was used in 113 patients version 15.1). The incidence of all ARs and intrafilter (13.6%). Therefore, we analyzed the results of this pressure increases was investigated according to the study to evaluate the safety and tolerability of number of patients. nafamostat mesilate and heparin as anticoagulants The Lichtiger Clinical Activity Index (CAI) (16) in LCAP for UC. was used to determine efficacy and was assessed before the start of LCAP and 2 weeks after the last PATIENTS AND METHODS LCAP session. Clinical remission was defined as a CAI score ≤4 at 2 weeks after the last LCAP session. In addition, clinical improvement was defined as clin- Study design ical remission or a final CAI score at least 50% lower The present study used data from a large-scale, than that obtained before the start of LCAP. Mucosal prospective, observational study of LCAP, which in- healing was assessed in all patients with an endo- cluded 847 patients with active UC (15). The study scopic index (EI) within the disease activity index was conducted in accordance with the Good (DAI) (17) and was defined as an EI of ≤1or0at Postmarketing Study Practice (GPSP) ordinance 2 weeks after the last LCAP session. The of the Japanese Ministry of Health, Labor, and hematochezia scores before the start of LCAP and Welfare. All patients in this study underwent LCAP 2 weeks after the last LCP session were compared at one of the 116 participating medical facilities be- for those patients with a DAI hematochezia score. tween May 2010 and December 2012. The treatment strategy for each patient, including the course of LCAP, was determined by the attending physician. Statistical analysis The observation period began 2 weeks prior to the Patient demographic data, the incidence of AR start of LCAP and ended 2 weeks following the and intrafilter pressure increases, and treatment effi- conclusion of treatment. The treating physicians cacy were compared by analyzing continuous data then filled out a case report form (CRF) after the using Wilcoxon rank-sum tests and categorical data observation period. using Fisher’s exact test. The hematochezia scores before and after treatment were compared using LCAP treatment Wilcoxon signed-rank test. Any missing data were Leukocytapheresis was performed using Cellsorba excluded. In all the analyses, P < 0.05 (two-sided) E, a column filled with nonwoven polyester fiber to was considered to be statistically significant.

© 2016 The Authors. Therapeutic Apheresis and Dialysis published by John Wiley & Sons Australia, Ltd on behalf Ther Apher Dial, Vol. 20, No 2, 2016 of International Society for Apheresis, Japanese Society for Apheresis, and Japanese Society for Dialysis Therapy Anticoagulants in LCAP for UC 199

RESULTS appropriate treatment. The main ARs in the heparin group were a decreased platelet count (2.7%, 3 of 113), nasal congestion (1.8%, 2 of 113), and pain in Anticoagulant use and demographic profiles the vascular access site (1.8%, 2 of 113). No serious Of the 847 patients enrolled in this study, the ARs, including exacerbated hemorrhagic stool or de- data from 832 were analyzed after 15 patients creased blood pressure, were observed. No ARs were excluded because the anticoagulant used fi were observed in the two patients who received could not be identi ed. The anticoagulants used LMWH in all sessions or in the five individuals who in the 832 patients are shown in Table 1. In total, received both nafamostat mesilate and heparin. 676 patients (81.3%) received nafamostat mesilate There was no significant difference in the inci- in all LCAP sessions, and 113 (13.6%) received dence of intrafilter pressure increases between the heparin in all LCAP sessions. In addition, 36 indi- nafamostat mesilate and heparin groups (12.7%, 86 viduals (4.3%) started with nafamostat mesilate of 676 vs. 16.8%, 19 of 113, P=0.233; Fig. 2). and were later switched to heparin or low-molecular- weight heparin (LMWH). Of these, 31 patients were Efficacy in the nafamostat mesilate and heparin switched to heparin, and five were switched to groups LMWH. No patients switched from heparin or Efficacy was investigated in a total of 610 patients LMWH to nafamostat mesilate. Both nafamostat after 222 patients with either a CAI ≤4 or who used mesilate and heparin were used simultaneously in fl fi in iximab, tacrolimus, or cyclosporine were ex- ve patients. cluded. The clinical improvement and remission rates The demographic data of the patients who re- were similar between the nafamostat mesilate and ceived nafamostat mesilate as the anticoagulant in heparin groups (P=0.518 and 0.902, respectively; all LCAP sessions (nafamostat mesilate group) and Fig. 3a). The clinical improvement and remission the 113 patients who received heparin in all LCAP rates in the patients who switched from nafamostat sessions (heparin group) are shown in Table 2. The fi mesilate to heparin or LMWH were 70.4% (19 of nafamostat mesilate group had a signi cantly higher 27) and 63.0% (17 of 27), respectively. rate of concomitant steroid use. The mucosal healing rates of the 202 patients in the nafamostat mesilate group and the 22 individuals Safety in the nafamostat mesilate and heparin groups in the heparin group with EI scores are shown in There was no statistically significant difference in Figure 3b. Patients in the nafamostat mesilate and the incidence of AR between the nafamostat heparin groups had similar results for both EI = 0 mesilate and heparin groups (8.6%, 58 of 676 vs. and EI ≤ 1(P=1.00 for both). 7.1%, 8 of 113, P=0.715; Fig. 1). The main ARs in The mean hematochezia scores before the start of the nafamostat mesilate group were headache LCAP and 2 weeks after the last LCAP session were (2.2%, 15 of 676), nausea (1.3%, 9 of 676), and fever compared in the 458 patients in the nafamostat (0.9%, 6 of 676) (Table 3). Three patients (0.4%) in mesilate group and in the 76 patients in the heparin the nafamostat mesilate group experienced serious group with hematochezia scores. The scores in the ARs: deep vein thrombosis (N=2) and infective nafamostat mesilate group decreased from 1.7 ± 0.8 endocarditis/candidemia (N=1). However, all of to 0.4 ± 0.7, whereas the scores in the heparin group these patients achieved recovery following the decreased from 1.9 ± 0.7 to 0.5 ± 0.8. The decrease after LCAP treatment was statistically significant in both groups (P < 0.001). TABLE 1. The anticoagulants used in the 832 patients Safety in patients whose anticoagulant was changed Anticoagulant used for LCAP treatment No. patients (%) Next, we investigated ARs in the 36 patients who Nafamostat mesilate for all LCAP 676 (81.3) switched from nafamostat mesilate to heparin or sessions Heparin for all LCAP sessions 113 (13.6) LMWH. Among these, 18 patients experienced LMWH for all LCAP sessions 2 (0.2) ARs during nafamostat mesilate use and 12 individ- Anticoagulant was changed 36 (4.3) uals experienced no ARs after switching to heparin from nafamostat mesilate to heparin or LMWH or LMWH (Table 4). One patient who experienced Both nafamostat mesilate 5 (0.6) a serious AR of anaphylactic shock during and heparin were used nafamostat mesilate use was switched to heparin simultaneously and experienced no AR subsequently; thus, LCAP LCAP,leukocytapheresis; LMWH, low-molecular-weight heparin. was continued.

© 2016 The Authors. Therapeutic Apheresis and Dialysis published by John Wiley & Sons Australia, Ltd on behalf of International Society for Apheresis, Japanese Society for Apheresis, and Japanese Society for Dialysis Therapy Ther Apher Dial, Vol. 20, No 2, 2016 200 K Sawada et al.

TABLE 2. Baseline demographic data and concomitant medications

Item Nafamostat mesilate group (N=676) Heparin group (N=113) P-value † Age (years) 40.4 ± 16.1 (6–88) 41.0 ± 15.3 (15–77) 0.625 † Weight (kg) 56.7 ± 11.3 (13.3–117) 57.6 ± 10.4 (39.8–80) 0.378 ‡ Sex, male/female 56.5%/43.5% 61.1%/38.9% 0.412 † UC disease duration (years) 6.5 ± 7.0 (0–41.1) 8.1 ± 9.7 (0–64.4) 0.197 † Lichtiger CAI 9.7 ± 3.8 9.8 ± 3.3 0.529 ‡ Severity (CAI score) 0.279 CAI ≤ 4 9.6% 8.9% CAI = 5–6 11.0% 6.2% CAI = 7–11 48.2% 56.6% CAI ≥ 12 31.2% 28.3% ‡ Response to corticosteroid 0.500 Resistant 31.7% 27.4% Dependent 36.4% 35.4% Nonrefractory 31.9% 37.2% Concomitant medication ‡ 5-ASA 94.2% 96.5% 0.500 ‡ Corticosteroid 67.3% 46.0% <0.001 ‡ Thiopurine 32.2% 40.7% 0.085 ‡ Inﬂiximab 6.5% 2.7% 0.133 ‡ Tacrolimus 13.2% 8.8% 0.223 ‡ Cyclosporine 1.2% 0% 0.610

5-ASA, 5-aminosalicylic acid; CAI, clinical activity index; UC, ulcerative colitis. The data are presented as percentages (%) or means † ‡ ± standard deviations. The minimum and maximum values are given in parentheses. Wilcoxon rank-sum test. Fisher’s exact probability test.

(50.0%, 14 of 28) and “experienced intraﬁlter pressure increases in a previous treatment in which nafamostat mesilate was used or was worried about this” (42.9%, 12 of 28).

Heparin dose The heparin dose was indicated for 111 patients (Table 5). Doses varied among facilities, with one- shot use doses of 0–3000 IU, continuous administration ranging from 0 to 5000 IU/500 mL of saline, and a fixed infusion rate of 10%–18% for the circulating blood volume totaling 2000–5000 IU. The priming was 0–3000 IU/500 mL of saline. The standard doses used in the present study were as follows: none for FIG. 1. Incidence of adverse reactions according to the anticoag- one-shot use, 3000 IU for continuous administration, ulants used (based on the number of subjects). and 2000 IU for priming. These standard doses were administered to 29 patients at four facilities. Reasons for heparin use Of 113 patients in the heparin group, the reason DISCUSSION for this choice of anticoagulant was indicated in 93. The main reasons were “experienced an AR due to Here, we analyzed the results of a large-scale, pro- nafamostat mesilate in the past or were worried spective, observational study of LCAP to evaluate about experiencing an AR” (64.5%, 60 of 93), “hep- the safety and tolerability of nafamostat mesilate arin is used in all patients at our facility” (22.6%, 21 and heparin as an anticoagulant during LCAP treat- of 93), and “worried about intrafilter pressure in- ment for UC. There were no significant differences creases due to nafamostat mesilate” (8.6%, 8 of 93). in the incidence of AR and intrafilter pressure in- Of the 31 patients who began treatment with creases, and the efficacy was comparable between nafamostat mesilate and were later switched to the nafamostat mesilate and the heparin groups. heparin, the reasons for switching were indicated Therefore, both nafamostat mesilate and heparin for 28. These included “used nafamostat mesilate can be well tolerated as an anticoagulant in LCAP in a previous treatment and experienced an AR” for UC.

TABLE 3. Adverse reactions according to the anticoagulant used

No. patients (%) Nafamostat mesilate Heparin group Adverse reaction group (N=676) (N=113) General disorders and administration site conditions Fever 6 (0.9) 0 (0) Pain in vascular 4 (0.6) 2 (1.8) access site Chills 3 (0.4) 0 (0) Malaise 3 (0.4) 0 (0) Chest discomfort 3 (0.4) 0 (0) Gastrointestinal disorders Nausea 9 (1.3) 0 (0) Vomiting 2 (0.3) 0 (0) FIG. 2. Incidence of intrafilter pressure increases according to Epigastric discomfort 2 (0.3) 0 (0) the anticoagulants used (based on the number of subjects). Abdominal pain 1 (0.1) 1 (0.9) Diarrhea 1 (0.1) 0 (0) Nervous system disorders for LCAP in clinical practice. However, heparin was Headache 15 (2.2) 0 (0) used in 13.6% of patients in all LCAP sessions. The Hypesthesia 2 (0.3) 0 (0) Blood and lymphatic system main reason for heparin use from the start of the disorders treatment was “experienced an AR due to Anemia 2 (0.3) 0 (0) nafamostat mesilate in the past or was worried about Vascular disorders ” Deep vein thrombosis 2 (0.3) 0 (0) experiencing an AR, which accounted for Flushing 1 (0.1) 0 (0) approximately 60% of the patients. A comparison of Jugular vein 1 (0.1) 0 (0) the patient demographic data in the nafamostat thrombosis Skin and subcutaneous mesilate and heparin groups revealed that although tissue disorders concomitant steroid use was high in the nafamostat Rash 4 (0.6) 0 (0) mesilate group, there were no differences in the Erythema 2 (0.3) 0 (0) Itching 2 (0.3) 0 (0) CAI score before starting LCAP treatment and the Investigations severity of UC. Therefore, we speculated that heparin Platelet count 3 (0.4) 3 (2.7) is mainly used when there are concerns about safety decreased Blood pressure 1 (0.1) 0 (0) in the clinical setting. decreased The incidence of ARs in the nafamostat mesilate Leukocyte count 1 (0.1) 0 (0) and the heparin groups was similar although there decreased Cardiac disorders were some notable differences. For example, the Palpitations 2 (0.3) 0 (0) AR of “headache,” which was observed in several Tachycardia 1 (0.1) 0 (0) patients in the nafamostat mesilate group, was not Respiratory, thoracic, and mediastinal disorders reported at all in the heparin group. In addition, Dyspnea 4 (0.6) 0 (0) dyspnea, rash, chest discomfort, anaphylactic shock, Nasal congestion 2 (0.3) 2 (1.8) vomiting, itching, blood pressure decreased, and Cough 1 (0.1) 0 (0) fl Infections and infestations ushing, which were likely due to an anaphylactic Endocarditis infective 1 (0.1) 0 (0) reaction, were observed in 2.5% (17 of 676) of the Candidemia 1 (0.1) 0 (0) patients in the nafamostat mesilate group; however, Common cold 1 (0.1) 0 (0) Cellulitis of the right 1 (0.1) 0 (0) these were not observed in any patients in the hepa- lower limb rin group. Immune system disorders Nagase et al. (13) reported the rapid onset of al- Anaphylactic shock 2 (0.3) 0 (0) Musculoskeletal and connective lergy to nafamostat mesilate during the early stage tissue disorders of LCAP. In addition, Ashizuka et al. (14) reported Back pain 1 (0.1) 0 (0) the development of allergic and cardiopulmonary Muscle spasms 1 (0.1) 0 (0) symptoms in patients using nafamostat mesilate. This was resolved by switching to LMWH, which enabled treatment to be continued safely. The present analy- In the present study, nafamostat mesilate was used in sis revealed that nafamostat mesilate caused a de- 81.3% of patients in all LCAP sessions; this well repre- crease in blood pressure, anaphylactic shock, and sents nafamostat mesilate as a first-choice anticoagulant dyspnea; however, switching to heparin allowed the

FIG. 3. Efﬁcacy according to the anticoagulants used. (a) Clinical improvement and remission rates in the nafamostat mesilate and heparin groups. (b) Mucosal healing rates in the nafamostat mesilate and heparin groups.

TABLE 4. Adverse reactions in patients for whom the anticoagulant was changed from nafamostat mesilate during treatment

Anticoagulant No. Adverse reaction Severity Treatment Outcome after change Adverse reaction Severity Treatment Outcome 1 Headache Slight (À) Recovery Heparin ———— 2 Headache Slight (À) Recovery Heparin ———— 3 Headache, dizziness Slight (À) Recovery Heparin Platelet count Slight (À) Recovery postural decreased 4 Headache, dyspnea Slight (À) Recovery Heparin ———— 5 Fever Moderate (À) Recovery Heparin ———— 6 Fever, chills Slight ( + ) Recovery Heparin Fever Slight ( + ) Recovery 7 Fever, ﬂushing, Slight ( + ) Recovery LMWH Fever, ﬂushing, Slight (À) Recovery oropharyngeal pain oropharyngeal pain 8 Fever, chills, nasal Slight (À) Recovery Heparin ———— congestion 9 Anaphylactic shock Serious ( + ) Recovery Heparin ———— 10 Anaphylactic shock Slight ( + ) Recovery Heparin ———— 11 Nausea Slight (À) Recovery Heparin ———— 12 Nausea, cough Slight (À) Recovery LMWH Nausea Slight (À) Recovery 13 Blood pressure Moderate ( + ) Recovery Heparin ———— decrease 14 Blood pressure Slight ( + ) Recovery Heparin ———— decrease 15 Abdominal pain Slight (À) Recovery Heparin Fever Slight (À) Recovery 16 Abdominal pain Slight (À) Recovery Heparin Nasal congestion Slight (À) Recovery 17 Dyspnea Slight (À) Recovery Heparin ———— 18 Rash Slight (À) Recovery LMWH ————

LMWH, low-molecular-weight heparin.

LCAP treatments to continue without incidence in Ashizuka et al. (14) reported that gastrointestinal most patients, which suggests that heparin can be bleeding was not exacerbated in patients who re- used as a substitute anticoagulant in patients who ceived LMWH. In addition, in the present study, no develop ARs associated with nafamostat mesilate. hemorrhagic ARs were observed among patients in Compared with nafamostat mesilate, heparin has a the heparin group, and the mean hematochezia signiﬁcantly longer half-life (18,19). Therefore, scores for this group decreased signiﬁcantly after nafamostat mesilate is used currently as the anticoag- LCAP treatment. Therefore, the exacerbation of ulant of choice for the extracorporeal circulation hemorrhaging in patients with UC because of hepa- treatment of hemorrhagic diseases (20). However, rin use may not be problematic in clinical practice.

TABLE 5. Heparin doses

Total heparin One-shot usage Continuous administration No. patients Priming dosage (IU) (IU) (IU/500 mL of saline) (No. facilities) (IU/500 mL of saline) 2000 0 2000 6 (2) 2000 2000 0 2 (1) Unknown 2200 1000 1200 1 (1) 2000 2500 0 2500 20 (2) 0 or nafamostat mesilate 20 mg 3000 0 3000 29 (4) 2000 1500 1500 9 (1) Unknown 2000 1000 4 (1) Unknown 3000 0 1 (1) Unknown 4000 0 4000 1 (1) 3,000 1000 3000 3 (2) 1000–2000 2000 2000 16 (3) 1000–2000 5000 0 5000 16 (3) 1700–2000 or nafamostat mesilate 20 mg 2000 3000 3 (2) Unknown

In the present study, we found no subject in the et al. (21) reported that factor XII, which is activated heparin group who experienced a decreased blood by negative charge and related to blood coagulation, pressure. Kanke et al. (12) reported that when hep- may be less generated in the Cellsorba column with arin was used as anticoagulant for LCAP, bradyki- the new filling solution. This result supports the find- nin, which is involved in lowering the blood ings of our study. However, there tended to be a high pressure, significantly increased. However, after incidence of intrafilter pressure increases in the hep- that, in 2008, the filling solution for Cellsorba col- arin group, although there was no statistical signifi- umn was changed from injection water to sodium cance. Clinicians should pay careful attention to the pyrosulfite and sodium carbonate. Yamada et al. intrafilter pressure increases. (21) reported that in the Cellsorba column with A previous study demonstrated that nafamostat new filling solution, bradykinin production was sup- mesilate and LMWH are equally effective when used pressed when heparin was used; no bradykinin- as anticoagulants during LCAP treatment for UC associated ARs were noted. They also reported that (14); however, no previous studies have compared the new filling solution may have reduced the the efficacy of LCAP between nafamostat mesilate amount of negative charge on the nonwoven polyes- and heparin. In the present analysis, the efficacy of ter fiber in the Cellsorba column. In the present LCAP in the nafamostat mesilate and heparin groups study, the Cellsorba column with new filling solution was equivalent. In addition, the efficacy in the pa- was used. Therefore, we believe that bradykinin- tients who switched from nafamostat mesilate to hep- induced decreased blood pressure should not be arin or LMWH was comparable with that of the problematic in clinical practice when heparin is used nafamostat mesilate and heparin groups. To the best for LCAP. of our knowledge, the present analysis is the first to A previous study reported that heparin causes suggest that heparin and nafamostat mesilate are heparin-induced thorombocytepenia (HIT) (22). In equally effective anticoagulants. the present analysis, three patients who received The usual heparin doses during extracorporeal heparin developed a decreased platelet count. How- circulation treatments are 1000–3000 IU for one- ever, none of these cases were serious, and all pa- shot use and 1000–2000 IU/h for continuous admin- tients recovered without treatment. Nevertheless, if istration. Nagase et al. reported that they used a decreased platelet count occurs physicians should heparin at 2000 IU for one-shot and 3000 IU/h for consider reducing the heparin dose or switching the continuous administration (13). In the present patient to another anticoagulant. study, we found that the doses of heparin varied Nagase et al. (13) reported that there was a higher greatly among patients because of differences in incidence of blood clotting with heparin compared their individual conditions, complications, and con- with nafamostat mesilate use. In the present study, comitant drug use. Hence, a randomized controlled there was no statistically significant difference in the study is required to establish standard optimal incidence of intrafilter pressure increases between doses that are safe and minimize intrafilter pressure the nafamostat mesilate and heparin groups. Yamada increases.

The present study has some limitations. For exam- ulcerative colitis. A randomized study. N Engl J Med 1987;317:1625–9. ple, because we were analyzing the results of a treat- 3. Meyers S, Sachar DB, Goldberg JD, Janowitz HD. Corticotro- ment outcome study, the numbers of patients in the pin versus hydrocortisone in the intravenous treatment of ul- nafamostat mesilate and heparin groups and their at- cerative colitis. A prospective, randomized, double-blind clinical trial. Gastroenterology 1983;85:351–7. tributes were inconsistent. A prospective, randomized, 4. Ogata H, Kato J, Hirai F et al. Double-blind, placebo- controlled study will be required to validate the results controlled trial of oral tacrolimus (FK506) in the management of the present study. Nevertheless, we believe that the of hospitalized patients with steroid-refractory ulcerative colitis. Inflamm Bowel Dis 2012;18:803–8. present analysis of the safety and tolerability of 5. Rutgeerts P, Sandborn WJ, Feagan BG et al. Infliximab for in- nafamostat mesilate and heparin based on a large duction and maintenance therapy for ulcerative colitis. N Engl amount of clinical data is clinically significant (23). J Med 2005;353:2462–76. 6. Sawada K, Muto T, Shimoyama T et al. Multicenter randomized controlled trial for the treatment of ulcerative colitis with a leukocytapheresis column. Curr Pharm Des 2003;9:307–21. CONCLUSION 7. Sawada K, Kusugami K, Suzuki Y et al. Leukocytapheresis in ulcerative colitis: results of a multicenter double-blind pro- In the large-scale treatment outcome study analyzed spective case–control study with sham apheresis as placebo here, nafamostat mesilate and heparin were used in 676 treatment. Am J Gastroenterol 2005;100:1362–9. 8. Hibi T, Sakuraba A. Is there a role for apheresis in gastrointesti- patients (81.3%) and 113 (13.6%), respectively. The in- – fi nal disorders? Nat Clin Pract Gastroenterol Hepatol 2005;2:200 1. cidence of adverse reactions and intra lter pressure in- 9. Andoh A, Ogawa A, Kitamura K et al. Suppression of creases were comparable in the nafamostat mesilate interleukin-1beta- and tumor necrosis factor-alpha-induced in- and heparin groups, and hemorrhage or blood pressure flammatory responses by leukocytapheresis therapy in patients with ulcerative colitis. J Gastroenterol 2004;39:1150–7. decreases associated with heparin use were not ob- 10. Yagi Y, Andoh A, Inatomi O et al. Modulation of platelet ag- served. Moreover, switching patients who experienced gregation responses by leukocytapheresis therapy in patients ARs due to nafamostat mesilate to heparin allowed with active ulcerative colitis. J Gastroenterol 2006;41:540–6. 11. Iwama H. Bradykinin-associated reactions in white cell- LCAP treatment to be continued without further reduction filter. J Crit Care 2001;16:74–81. ARs. In addition, there were no significant differences 12. Kanke K, Hoshino M, Tominaga K, Nakano M, Terano A, in the clinical remission rates and mucosal healing rates Hiraishi H. Selection of anticoagulants for leukocytapheresis therapy in cases of active ulcerative colitis. Blood Purif between the nafamostat mesilate and heparin groups. 2007;25:370–6. Therefore, the safety and efficacy of LCAP performed 13. Nagase K, Sawada K, Ohnishi K, Egashira A, Ohkusu K, using heparin as an anticoagulant are equivalent to Shimoyama T. Complications of leukocytapheresis. Ther Apher 1998;2:120–4. those of nafamostat mesilate. Therefore, we conclude 14. Ashizuka S, Nishiura R, Ishikawa N et al. Leukocytapheresis that both nafamostat mesilate and heparin can be well for ulcerative colitis: a comparative study of anticoagulant tolerated as anticoagulant a in LCAP for UC. (nafamostat mesilate vs. dalteparin sodium) for reducing clinical complications. Ther Apher Dial 2006;10:54–8. 15. Yokoyama Y, Matsuoka K, Kobayashi T et al. A large-scale, Acknowledgments: Asahi Kasei Medical funded this prospective, observational study of leukocytapheresis for ul- study and participated in the study design, data collection, cerative colitis: treatment outcomes of 847 patients in clinical data management, and data analysis. The authors had full practice. J Crohns Colitis 2014;8:981–91. access to all the data and take responsibility for the integ- 16. Lichtiger S, Present DH, Kornbluth A et al. Cyclosporine in rity of the data and the analysis. All the authors were in- severe ulcerative colitis refractory to steroid therapy. N Engl J Med 1994;330:1841–5. volved in critical revisions of the manuscript and fi 17. Sutherland LR, Martin F, Greer S et al. 5-Aminosalicylic acid approved the nal manuscript. enema in the treatment of distal ulcerative colitis, proctosigmoiditis, and proctitis. Gastroenterology Conflict of interest: K. Sawada and M. Ohdo have served 1987;92:1894–8. as speakers of Asahi Kasei Medical. T. Ino, T. Nakamura, 18. Akizawa T, Sato M, Kitaoka T et al. The usefulness of the T. Numata, H. Shibata, J. Sakou, and M. Kusada are multi-enzyme inhibitor, nafamstat mesilate, in high bleeding risk haemodialysis. EDTA-ERA Proc 1985;22:334–8. employees of Asahi Kasei Medical. T. Hibi has served as 19. Olsson P, Lagergren H, Ek S. The elimination from plasma of an advisory board member for Asahi Kasei Medical. intravenous heparin: an experimental study on dogs and humans. Acta Med Scand 1963;173:619–30. Financial support: Asahi Kasei Medical funded this 20. Akizawa T, Koshikawa S, Ota K, Kazama M, Mimura N, Hirasawa Y. Nafamostat mesilate: a regional anticoagulant study. for hemodialysis in patients at high risk for bleeding. Nephron 1993;64:376–81. 21. Yamada R, Kusaoi M, Murayama G et al. Reduction in bradykinin generation during leukocytapheresis using novel ™ REFERENCES cellsorba CS-180S: effects of changing the filling solution. J Clin Apher 2014;29:90–6. 1. Hendrickson BA, Gokhale R, Cho JH. Clinical aspects and 22. Napolitano LM, Warkentin TE, Almahameed A, Nasraway SA. pathophysiology of inflammatory bowel disease. Clin Heparin-induced thrombocytopenia in the critical care setting: Microbiol Rev 2002;15:79–94. diagnosis and management. Crit Care Med 2006;34:2898–911. 2. Schroeder KW, Tremaine WJ, Ilstrup DM. Coated oral 5- 23. Silverman SL. From randomized controlled trials to observa- aminosalicylic acid therapy for mildly to moderately active tional studies. Am J Med 2009;122:114–20.