Aims. The incidence of complications associated with cerebrovascular diseases in patients who receive hemodialysis for a long-term period is higher than that of other complications. It is known that mortality due to cerebral hemorrhage is two times higher compared to non-dialysis patients. Anti-coagulants used for hemodialysis are essential. Accordingly, in cases in which the cerebral hemorrhage occurred, the selection of anti-coagulants for the prevention of further bleeding poses a great challenge to physicians. The change of hematoma and patient prognosis has a direct relationship. Many ongoing studies are conducted to examine the causative factors causing the increased hematoma and their related prognostic factors. In the current study, we examined the effect of nafamostat mesylate (a serine protease inhibitor) on the change of hematoma compared to heparin in hemodialysis patients. Methods. The current study was conducted in 17 hemodialysis patients who developed a cerebral hemorrhage. These patients were assigned to two groups based on the type of anti-coagulants that they used (i.e., nafamostat mesylate and heparin). Then, the factors affecting the change of hematoma following the onset of cerebral hemorrhage were examined. The prognosis of hematoma was assessed based on brain CT scans, which were performed two weeks after the onset of cerebral hemorrhage in four groups. Following this, groups 1 (the decreased hematoma) and 2 (the decreased delay) were merged to group A (resolving group), and groups 3 (the increased hematoma) and 4 (the death following the aggravation) were merged to group B (the expansion group) for further analysis. Results. There were no significant differences in baseline characteristics between the nafamostat group and the heparin group. A comparison between the resolving group and the expansion group also showed that there were no significant differences in baseline characteristics. In the anti-coagulants and the change of hematoma, however, there were significant differences between the two groups (p = 0.024). A comparison of the change of hematoma between the four groups was also made. This showed that platelet counts and BUN level were significant factors (Platelet; p = 0.042, BUN; p = 0.043 ANOVA with resolving group). Conclusions. Nafamostat mesylate has a similar profile of anti-coagulative activity to heparin. It is assumed, however, that nafamostat has an affirmative effect on the recovery of damaged sites following the onset of cerebral hemorrhage. It is an anti-coagulant that can be safely used for hemodialysis following the onset of cerebral hemorrhage.
Keywords: hemodialysis; nafamostat mesylate; intracerebral hemorrhage
Hemodialysis accounts for a majority of renal replacement therapy that is performed in patients with end-stage renal disease. With the increased number of patients who receive hemodialysis for a long-term period, the incidence of complications associated with hemodialysis is also increasing.[
We attempted to examine the effect of nafamostat mesylate, which has been reported to cause no further hemorrhage in patients with cerebral hemorrhage that concurrently had bleeding, because it produced a local anti-coagulator activity during the blood coagulation for a short-term period, on the clinical course of cerebral hemorrhage compared to the conventional heparin.
This study was conducted in patients who were hospitalized for the treatment of hemorrhage in the brain parenchyma due to hypertension of 31 patients who were receiving hemodialysis at Yonsei University Wonju Christian Hospital during a period ranging from January 1, 2004, to August 31, 2007. Of the total cases, those in which the treatment was abandoned due to severe bleeding, those of subarachnoid hemorrhage due to vascular anomaly, those of cerebral hemorrhage due to trauma, those in which there was a bleeding tendency due to other causes, and those in which there was a past history of cerebral lesions including cerebral infarction were all excluded from the current analysis. Of total patients, 17 were included in a follow-up observation. These patients were randomly selected for the use of anti-coagulants. During the follow-up period for two groups (the heparin group and the nafamostat mesylate group), the current study was conducted based on two weeks' brain CT scans and a retrospective review of medical records.
In 17 subject patients, the age, sex, blood pressure, presence of diabetes mellitus, period of hemodialysis, admission period following the onset of symptoms, Glasgow coma scale, hematoma volume, blood test associated with the hemorrhage, and period of the discontinuation of hemodialysis were checked. To determine whether the hematoma was aggravated, subject patients underwent brain CT two times (at the time of admission and two weeks later). Surgery was performed for general indications (i.e., in cases in which the hematoma volume exceeded 35 ml) in basal ganglia or thalamus and those in which it exceeded 50 ml in the subcortical area. Of total patients, surgery was performed in two cases.
Of total 17 patients, eight used nafamostat mesylate and nine used heparin for the hemodialysis. The prognosis was assessed based on four categories in primary endpoint (viz., decreased cerebral hemorrhage, decreased delay, progression, and death) in both groups. Then, patients were subdivided into two subgroups: the resolving group (combine patients who showed a decreased cerebral hemorrhage with those who had a decreased delay) and the expansion group (combine patients who showed a progression with those who died). Factors affecting the prognosis in two groups in which nafamostat mesylate or heparin was used as an anti-coagulant were analyzed.
At a two-month interval, a brain CT was performed, and thereby the same level was confirmed. Thus, the areas where the hematoma was maximally observed were compared. To calculate the amount of cerebral hemorrhage, the longest axis was measured using a formula A × B × C/2 and then expressed as a unit of milliliter. Cases in which the length of long axis or the calculated amount of cerebral hemorrhage were decreased on brain CT scans that were performed in the same area (the decreased extent was not defined herein), including the decreased delay, were defined as the resolving group. Cases in which the length of long axis or the calculated amount of cerebral hemorrhage were increased on brain CT scans that were performed in the same area (the increased extent was not defined herein), including death, were defined as the expansion group (see Figure 1).
Graph: Figure 1. Methods for volume of hematoma and interval change of CT: (a) admission, (b) two weeks later. On the CT slice with the largest area of ICH, the largest diameter (A) of the hematoma was measured by use of the centimeter scale on the CT film. The diameter of the hemorrhage perpendicular to the largest diameter represented the second diameter (B). The height of the hematoma was calculated by multiplying the number of slices involved by the slice thickness, providing the third diameter. The three diameters were multiplied and then divided by 2 (AxBxC/2) to obtain the volume of ICH (this formula was validated by Kwak et al.).
Hemodialysis was performed using equipment Gambro AK 95® (Gambro, Stockholm, Sweden), and the dialysis membrane used herein was polyamix® (Gambro). Dialysis solution used herein was Bicart® (Gambro). Hemodialysis was performed three times a week at 4 hours/day at a mean blood flow rate of 150–200 mL/min. A loading dose of anti-coagulant (heparin) was 1,000 IU, and this was intravenously infused. A maintenance dose was maximally 200 IU per hour. Nafamostat mesylate was intravenously infused at an hourly dose of 35 mg without a loading dose.
In the current study, statistical analysis was done using SPSS (version 12.0). All of the data were expressed as mean ± SD. Continuous variables were compared using t-test between the heparin and nafamostat groups. Discontinuous variables were compared using a chi-square test. To identify the correlations between the increased hematoma and the related factors, a multiple regression analysis was performed. A value of p < 0.05 was considered statistically significant.
Total subject patients (n = 17) were composed of ten males and seven females whose mean age was 59.9 ± 9.7 years. Of the underlying diseases, eight had a diabetes mellitus and 17 had hypertension. Mean systolic pressure was 198.7 ± 32.2 mmHg and mean diastolic pressure was 98.6 ± 11.3 mmHg. A blood test showed hemoglobin 10.4 ± 1.5 g/dL, platelet 183.8 ± 72.1 (×1000/L), PT 11.6 ± 1.3 sec, PTT 32.5 ± 4.3 sec, BUN 50.6 ± 24.2 mg/dL, and Cr 7.7 ± 2.7 mg/dL. The period of hemodialysis was 70.6 ± 51.2 months, the amount of cerebral hemorrhage was 20.3 ± 28.9 ml, the Glasgow coma scale (GCS) on admission was 13.9±1.6, the period elapsed until the outpatient visit since the onset of symptoms was 24.8 ± 52.3 hr, and the delayed hemodialysis following the onset of hemorrhage was 67.3±41.6 hr (see Table 1).
Table 1 Epidemiology and characteristics of total patients
Subjects (n = 17) Age (yrs) 59.9 ± 9.7 Male 10 (58.8%) DM 8 (47.1%) Blood pressure SBP (mmHg) 198.7 ± 32.2 DBP (mmHg) 98.6 ± 11.3 Hb (g/dl) 10.3 ± 1.5 Platelet (×1000/L) 183.8 ± 72.1 BUN (mg/dl) 50.6 ± 24.2 Cr (mg/dl) 7.7 ± 2.7 PT (sec) 11.6 ± 1.3 PTT (sec) 32.5 ± 4.3 Anticoagulant (nafamostat mesilate) 8 (47.1%) History of hemodialysis (months) 70.6 ± 51.3 Interval to admission (hrs) 24.8 ± 52.3 GCS 13.9 ± 1.6 Hematoma volumre (ml) 20.2 ± 28.9 Hemodialysis skip period (hrs) 67.2 ± 41.6
5 Abbreviations: DM = diabete mellitus, SBP = systolic blood pressure, DBP = diastolic blood pressure, Hb = hemoglobin, BUN = blood urea nitrogen, Cr = creatinine, GCS = Glasgow coma scale.
A total of 17 patients were assigned to the nafamostat (n = 8) and heparin (n = 9) groups. In these two groups, the following parameters were compared: age, period of hemodialysis, period of outpatient visit following the onset of symptoms, blood pressure, hemoglobin, platelets, PT, aPTT, BUN, Cr, amount of cerebral hemorrhage on brain CT scans, GCS on admission, and period of delayed hemodialysis. There were no significant differences in these parameters between the two groups (see Table 2).
Table 2 Comparison of risk factors between nafamostat and heparin groups
Nafamostat group(n = 8) Heparin group (n = 9) Age (yrs) 60.7 ± 11.7 59.2 ± 8.2 NS Male/female 4/4 6/3 NS DM/non-DM 3/5 5/4 NS History of hemodialysis (months) 81.7 ± 59.6 60.7 ± 43.5 NS Interval to admission (hrs) 17.9 ± 24.9 31.0 ± 69.6 NS GCS 14.5 ± 0.5 12.3 ± 2.5 NS Blood pressure (mmHg) SBP 201.0 ± 27.0 196.6 ± 37.7 NS DBP 99.5 ± 13.8 97.8 ± 9.3 NS Hematoma volume (ml) 6.3 ± 6.1 29.5 ± 35.0 NS Hb (g/dl) 10.6 ± 1.5 10.2 ± 1.5 NS Platelet (×1000/L) 177.1 ± 51.7 189.7 ± 89.2 NS BUN (mg/dl) 53.0 ± 23.7 48.5 ± 25.8 NS Cr (mg/dl) 7.5 ± 2.4 7.9 ± 3.1 NS PT (sec) 11.9 ± 1.5 11.3 ± 1.1 NS PTT (sec) 34.2 ± 2.8 30.9 ± 4.9 NS Hemodialysis skip period (hrs) 79.3 ± 48.0 56.5 ± 34.1 NS
6 Abbreviations: DM = diabetes mellitus, SBP = systolic blood pressure, DBP = diastolic blood pressure, Hb = hemoglobin, BUN = blood urea nitrogen, Cr = creatinine, GCS = Glasgow coma scale.
A total of 17 patients were assigned to the resolving (n = 10) and expansion (n = 7) groups. There were no statistically significant parameters in these two groups. The following parameters were compared: age, period of hemodialysis, period of outpatient visit following the onset of symptoms, blood pressure, hemoglobin, platelets, PT, aPTT, BUN, Cr, amount of cerebral hemorrhage on brain CT scans, GCS on admission, and period of delayed hemodialysis (see Table 3). However, in four groups, there were a difference of platelet counts in the died group (p = 0.042) and BUN in the delayed resolving group (p = 0.043) compared with resolving group in ANOVA (see Table 4).
Table 3 Comparison of risk factors between resolving and expansion groups
Resolving group (n = 10) Expansion group (n = 7) Age (yrs) 61.5 ± 10.5 57.7 ± 8.6 NS Male/female 5/5 5/2 NS DM/non-DM 5/5 3/4 NS History of hemodialysis (months) 82.9 ± 49.6 53.0 ± 51.8 NS Interval to admission (hrs) 38.2 ± 66.2 5.7 ± 4.3 NS GCS 14.0 ± 1.7 13.5 ± 2.1 NS Blood pressure (mmHg) SBP 203.8 ± 32.5 191.4 ± 32.8 NS DBP 97.0 ± 11.7 101.0 ± 11.2 NS Hematoma volume (ml) 5.4 ± 5.7 35.1 ± 36.1 NS Hb (g/dl) 10.6 ± 1.3 10.0 ± 1.8 NS Platelet (×1000/L) 211.6 ± 77.4 144.1 ± 41.6 NS BUN (mg/dl) 46.4 ± 19.6 56.7 ± 30.2 NS Cr (mg/dl) 7.4 ± 2.7 8.1 ± 2.8 NS PT (sec) 11.3 ± 1.5 12.1 ± 1.2 NS PTT (sec) 31.7 ± 5.2 33.6 ± 2.8 NS Hemodialysis skip period (hrs) 61.3 ± 41.9 75.8 ± 42.8 NS
7 Abbreviations: DM = diabetes mellitus, SBP = systolic blood pressure, DBP = diastolic blood pressure, Hb = hemoglobin, BUN = blood urea nitrogen, Cr = creatinine, GCS = Glasgow coma scale.
Table 4 Comparison of risk factors among four groups about change of hematoma size
Resolving group Delayed resolving group Expansion group Died group Age (yrs) 62.1 ± 11.0 54.0 ± 15.5 61.3 ± 4.7 56.0 ± 6.5 NS Male/female 5/4 2/0 2/1 2/1 NS DM/non-DM 4/5 0/2 1/2 2/1 NS History of hemodialysis (months) 81.2 ± 52.3 84.0 ± 101.8 63.0 ± 33.6 37.4 ± 35.9 NS Interval to admission (hrs) 39.8 ± 69.9 2.0 ± 1.4 12.3 ± 10.2 7.6 ± 5.6 NS GCS 13.8 ± 1.7 15.0 ± 0.0 15.0 ± 0.0 12.0 ± 0.0 NS Blood pressure (mmHg) SBP 204.2 ± 34.4 180.0 ± 0.0 186.6 ± 41.6 206.6 ± 32.1 NS DBP 96.6 ± 12.3 106.0 ± 8.4 91.6 ± 10.4 106.6 ± 5.7 NS Hematoma volumre (ml) 5.3 ± 5.7 1.78 ± 0.0 1.8 ± 0.0 57.4 ± 27.2 NS Hb (g/dl) 10.4 ± 1.2 9.1 ± 1.3 11.0 ± 2.0 10.3 ± 2.3 NS Platelet (×1000/L) 214.4 ± 81.6 175.5 ± 12.0 168.0 ± 26.9 113.3 ± 45.0 0.042 BUN (mg/dl) 46.4 ± 20.8 85.5 ± 19.1 40.3 ± 24.0 50.3 ± 27.3 0.043 Cr (mg/dl) 7.2 ± 2.8 10.1 ± 2.6 7.5 ± 3.8 7.7 ± 1.5 NS PT (sec) 11.2 ± 1.3 12.7 ± 1.9 11.6 ± 0.6 12.0 ± 1.2 NS PTT (sec) 31.3 ± 5.2 34.2 ± 1.8 32.4 ± 2.9 35.0 ± 3.0 NS Hemodialysis skip period (hrs) 60.8 ± 44.4 112.0 ± 42.4 51.6 ± 35.7 72.6 ± 34.5 NS
- 4001 *ANOVA: p < 0.05 compared with resolving group.
- 8 Abbreviations: DM = diabetes mellitus, SBP = systolic blood pressure, DBP = diastolic blood pressure, Hb = hemoglobin, BUN = blood urea nitrogen, Cr = creatinine, GCS = Glasgow coma scale.
Two weeks later, the changes of hematoma associated with the use of anti-coagulants seen on brain CT scans were evaluated. This showed that the nafamostat group had a significantly better prognosis than the heparin group. To put this another way, patients were assigned to two groups: the resolving group, where patients showed a decreased amount of cerebral hemorrhage or a decreased delay on brain CT scans, and the expansion group, where patients showed an increased amount of cerebral hemorrhage or died. Based on the findings seen on brain CT scans, which were performed two weeks later, it was shown that the nafamostat group had a significantly better prognosis than the heparin group (see Table 5).
Table 5 Chi-square test about CT changes and anticoagulants
CT on 2 weeks Resolving group (%) Expansion group (%) Total Anticoagulation method Nafamostat mesilate (%) 7 (41.2) 1 (5.9) 8 (47.1) Heparin (%) 3 (17.6) 6 (35.3) 9 (52.9) Total 10 (58.8) 7 (41.2) 17 (100) Fisher's exact test
To identify the correlations between the findings seen on brain CT scans, performed after two weeks following the onset of cerebral hemorrhage, and the patient prognosis, a multiple logistic regression analysis was performed for such risk factors as the age, period of hemodialysis, period of outpatient visit following the onset of symptoms, blood pressure, hemoglobin, platelets, PT, aPTT, BUN, Cr, amount of cerebral hemorrhage on brain CT scans, coma index on admission, and period of delayed blood dialysis. However, there were no statistically significant results in this series.
Hemodialysis accounts for a majority of renal replacement therapy performed in patients with end-stage renal disease. With the advancement of dialysis technology, the number of patients who receive hemodialysis for a long-term period has recently been increasing. According to this, the incidence of complications associated with hemodialysis is also increasing. Of the death causes that have been reported in dialysis patients, excluding cardiovascular diseases, cerebrovascular diseases account for the majority. Of them, the incidence of cerebral hemorrhage is two times higher in dialysis patients than normal healthy people.[
Unfractionated heparin and low-molecular weight heparin are commonly used as anti-coagulants for hemodialysis. Because heparin can cause severe thrombocytopenia and its use is somewhat restricted in patients who concurrently have bleeding, however, thrombin inhibitors such as local citrate and serine protease inhibitors are considered as the second-line measures.[
It is well known that serine protease is involved in the constriction of blood vessels that are distributed in the organs in a shock state due to the bleeding and the suppression of serine protease activity, which plays a crucial role in inhibiting the aggravation of the organ damage.[
Hemorrhage due to uremia occurs with the involvement of various pathophysiologic factors. Little is known about the exact mechanisms. The derangement of platelet aggregation plays a crucial role in the pathophysiology of hemorrhage due to uremia. In cases in which the bleeding occurred, various treatment regimens or methods have been studied to prevent further bleeding. At present, however, such measures as appropriate hemodialysis, erythropoietin, desmopressin, estrogen, and cryoprecipitate have been reported to be effective to some extent. Hemodialysis and the use of erythropoietin are demonstrated methods for the prevention of further bleeding.[
In hemodialysis patients, only a few studies have been conducted to examine the predictive factors associated with bleeding (the cerebral hemorrhage in particular) at present. According to a study enrolling 1,064 patients who received blood dialysis for more than three months, however, the cerebrovascular diseases occurred in 9.2% of total patients. Of patients who definitely had a cerebral hemorrhage, 71.4% died within three months following the onset of cerebral hemorrhage.[
In conclusion, when the anti-coagulants should be selected for hemodialysis patients who developed cerebral bleeding, nafamostat mesylate is safer and more effective in inhibiting the formation of hematoma than heparin. This deserves further study.
The authors report no external funding.
By Jae Won Yang; Byoung Geun Han; Bi Ro Kim; Yo Han Lee; Young Sub Kim; Jong Myeong Yu and Seung Ok Choi
Reported by Author; Author; Author; Author; Author; Author; Author