The Journal of Arthroplasty xxx (2019) 1e8

Contents lists available at ScienceDirect

The Journal of Arthroplasty

journal homepage: www.arthroplastyjournal.org

Hemostatic and Anti-Inflammatory Effects of Carbazochrome

Sodium Sulfonate in Patients Undergoing Total Knee Arthroplasty:
A Randomized Controlled Trial
Yue Luo, MM, Xin Zhao, MD, PhD, Yeersheng Releken, MM, Zhouyuan Yang, MD, PhD,
FuXing Pei, MD, Pengde Kang, MD, PhD *
Department of Orthopedic Surgery, West China Hospital, Sichuan University, Chengdu, Sichuan Province, People’s Republic of China

a r t i c l e i n f o a b s t r a c t

Article history: Background: Postoperative recovery after total knee arthroplasty (TKA) is associated with postoperative
Received 11 June 2019 anemia, allogeneic transfusion, and stress immune responses to surgery. Carbazochrome sodium sulfo-
Received in revised form nate (CSS) reduces bleeding through several mechanisms. We assessed the effect of CSS combined with
28 July 2019
tranexamic acid (TXA) on postoperative anemia, blood transfusion, and inflammatory responses.
Accepted 31 July 2019
Available online xxx
Methods: This study was designed as a randomized, placebo-controlled trial of 200 patients undergoing
unilateral primary TKA. Patients were divided into 4 groups: group A received TXA plus topical and
intravenous CSS; group B received TXA plus topical CSS only; group C received TXA plus intravenous CSS
Keywords:
blood loss
only; group D received TXA only.
inflammation Results: Total blood loss in groups A (609.92 ± 221.24 mL), B (753.16 ± 247.67 mL), and C (829.23
carbazochrome sodium sulfonate ± 297.45 mL) was lower than in group D (1158.26 ± 334.13 mL, P < .05). There was no difference in total
tranexamic acid blood loss between groups B and C. We also found that compared with group D, the postoperative
total knee arthroplasty swelling rate, biomarker level of inflammation, visual analog scale pain score, and range of motion at
discharge in groups A, B, and C were significantly improved (P < .05). No thromboembolic complications
occurred. There were no differences in transfusion rate, intraoperative blood loss, platelet count, or
average length of stay among the 4 groups (P > .05).
Conclusion: CSS combined with TXA was more effective than TXA alone in reducing perioperative blood
loss and inflammatory response and did not increase the incidence of thromboembolism complications.
© 2019 Elsevier Inc. All rights reserved.

Total knee arthroplasty (TKA) is associated with several post-
operative complications such as blood loss and anemia [1e3]. Much
research has been dedicated to reducing the incidence of these
Ethical review committee statement: This study was approved by the clinical trial
and biomedical ethics committee of West China Hospital, and all participants'
written informed consent was obtained. A copy of the letter from the ethics
committee approving the study is attached below.
Funding Source: This study was not supported by funding from any public, com-
mercial, or nonprofit funding agencies.
A statement of the location where the work was performed: The whole experiment
was carried out in West China Hospital of Sichuan University.
Yue Luo and Yeersheng Releken contributed equally to this work and should be
regarded as first co-authors.
* Reprint requests: Pengde Kang, MD, PhD, Department of Orthopedic Surgery,
West China Hospital, Sichuan University, 37# Wainan Guoxue Road, Chengdu
610041, Sichuan, People’s Republic of China.
complications and improving patient recovery [4]. For example,
treatment strategies such as controlled hypotension, regional
anesthesia, erythropoietin, and antifibrinolytic drugs can reduce
blood loss and transfusion requirements after TKA [5].
Although the application of tranexamic acid (TXA) can reduce
the amount of blood loss during the perioperative period of TKA
[3,4], some patients with TKA still receive blood transfusion during
or after the operation, with transfusion rates following TKA varying
greatly between different countries [1,2,5e7]. In addition, TXA has
been found to have certain anti-inflammatory effects, but this anti-
inflammatory ability is not strong [8]. In addition, thromboembolic
events can occur with any coagulant that reduces blood loss.
Although a large amount of level-I evidence confirms that TXA can
significantly reduce blood loss and transfusion demand without
increasing the risk of venous thrombotic events (VTEs), the hidden
blood loss (HBL) does not decrease [9,10]. Therefore, additional
measures are taken to further reduce HBL and inflammatory
response after TKA.

https://doi.org/10.1016/j.arth.2019.07.045
0883-5403/© 2019 Elsevier Inc. All rights reserved.
2 Y. Luo et al. / The Journal of Arthroplasty xxx (2019) 1e8
Carbazochrome sodium sulfonate (CSS) is a capillary stabilizer
that reduces capillary permeability and enhances contraction of
broken capillary ends [11,12]. It is used to treat bleeding diseases of
the urinary system, upper digestion, respiratory tract, and in ob-
stetrics and gynecology [13,14]. In addition, a single, randomized,
prospective study has shown that TXA combined with CSS signifi-
cantly reduced blood loss after TKA without increasing the risk of
asymptomatic deep venous thrombosis (DVT) [15]. In addition to its
use in orthopedics, CSS is also used in other disciplines [16e21]. For
example, the study by Passali et al [16] found that CSS can improve
the bleeding symptoms of hereditary hemorrhagic telangiectasia.
Studies by Oh-oka et al [17] have shown that CSS can effectively
improve pain and postmicturition symptoms in patients with re-
fractory chronic prostatitis. The study by Funahara et al [19] sug-
gests that the application of CSS may help eliminate the risk factors
for severe bleeding and shock in dengue hemorrhagic fever. How-
ever, a randomized prospective study showed that CSS did not
prevent plasma leakage or shock in patients with dengue hemor-
rhagic fever/dengue shock syndrome [18].
Reducing perioperative blood loss and excessive surgical stress
immune response without increasing risk of thromboembolic
complications is the key to reducing preoperative morbidity after
major surgical operations such as TKA [22,23]. Therefore, the pur-
pose of this randomized, controlled trial is to evaluate the addi-
tional effects of CSS combined with TXA on hemostasis, thrombosis,
and anti-inflammatory effects in patients undergoing TKA.
Materials and Methods
Patient Recruitment
The study was registered at the Chinese clinical trial registry
(ChiCTR1800015839) and approved by the ethics committee of our
hospital. The whole experiment was carried out in the joint surgery
department of our hospital. Written, informed consent was ob-
tained from all patients before surgery.
Patients diagnosed with knee osteoarthritis and scheduled for
unilateral primary TKA [24] between May and August 2018 were
eligible. Patients were excluded if they had a substantial deformity
of the knee joint on the operative side (flexion deformity of
20 ,
varus-valgus deformity of
20 ), low hemoglobin levels (<11 g/dL),
a body mass index greater than 35 kg/m2 [25], a history of renal
failure, renal transplantation, arterial thromboembolism (eg,
myocardial infarction or stroke), arterial stenting, DVT, pulmonary
embolism (PE), or previous knee surgery. We also excluded patients
with allergies to TXA, CSS, and patients who refused to receive
blood products. Demographic and preoperative patient character-
istics were recorded before surgery.
Randomization and Medication
Patients were randomly divided into 4 equal groups using a
computer-generated randomization table. A statistician not
involved in the data analysis conceals computer-generated
randomly assigned sequences in serially numbered, sealed enve-
lopes. A nurse who was not involved in patient management pre-
pared either a study drug or a placebo for the task assigned by the
team. All patients were given 1 g of TXA intravenously 5 minutes
before the skin incision. Group A was injected with 60 mL of CSS in
saline (0.67 mg/mL) around and within the joint capsule before it
was closed, and then 100 mL of CSS (0.6 mg/mL) was administered
intravenously at 3 hours postsurgery. Group B was given the same
treatment as group A but with 100 mL of placebo (0.6 mg/mL) at
3 hours postsurgery instead. Group C was injected with 60 mL of
placebo (0.67 mg/mL) around and within the joint capsule,
followed by intravenous administration of 100 mL of CSS (0.6 mg/
mL) at 3 hours postsurgery. Group D was given placebo both in the
joint capsule and at 3 hours postsurgery. No tourniquets were used,
and no drainage tubes were used postoperatively. The patients,
anesthesiologists, surgeons, surgical team members, and outcome
assessors were all blinded to patient treatment group.
Surgical Technique and Perioperative Management
TKAs were performed by 2 senior authors (P.D.K. and F.X.P.). A
midline skin incision was made using the medial parapatellar
approach. Patients were given general anesthesia and implanted
with a posterior-stabilized cemented prosthesis (DePuy Synthes).
According to the guidelines for perioperative blood transfusion
provided by the Ministry of Health of China, a blood transfusion
was given if the hemoglobin concentration fell below 70 g/L or if
the patient showed any signs of anemia-related organ dysfunction
(such as changes in mental state or heart palpitations), regardless of
hemoglobin concentration [26].
All patients underwent lower-extremity strength training
before surgery. All patients received general anesthesia and intra-
venous prophylactic antibiotics for 24 hours. Patients were given
passive and active physical therapy after recovering from anes-
thesia. An inflatable lower limb vein pump was applied immedi-
ately after surgery, and low-molecular-weight heparin (Clexane,
France) was given at a dose of 0.2 mL, or 2000 AxaIU at 12 hours
after surgery, followed by daily doses of 0.4 mL until discharge.
Patients were instructed to take 10 mg of rivaroxaban daily for 10
days after discharge to prevent VTE. Patients were given 60 mg of
diclofenac sodium orally every 12 hours and an intramuscular in-
jection of parecoxib sodium (40 mg) every 12 hours from the first
day after surgery until day of discharge.
During hospitalization, patients were checked daily for clinical
symptoms of VTE. Patients were given a bilateral lower limb
Doppler ultrasonography 2 weeks after surgery. Doppler ultra-
sound was routinely performed in all patients at 1 and 3 months
postoperatively to screen for asymptomatic DVT, and it was also
performed at any other time if a patient experienced symptoms
suggestive of DVT, while computed tomography was performed if
PE was suspected.
Outcome Measures
The primary outcome of this study was total blood loss on
postoperative day (POD) 2. Total blood loss was calculated using a
modification of the Gross formula [27,28]. When performing blood
transfusions or allogeneic transfusions, total blood loss is calculated
by adding the blood transfusion volume to the previously calcu-
lated blood loss [29]. Secondary outcomes included HBL [30,31],
platelet count, reduction in hemoglobin concentration (calculated
as preoperative hemoglobin concentrationdPOD 2 hemoglobin
concentration), intraoperative blood loss (IBL), coagulation and
fibrinolysis parameters, transfusion rates, inflammatory marker
levels, perioperative visual analog scale (VAS) pain score, swelling
ratio, length of stay(LOS), range of motion (ROM), and incidence of
thromboembolic events and other complications.
HBL was defined as the total blood loss minus IBL, as
described in previous studies [30,31]. IBL was calculated by
measuring the suction volume and gauze weight at the end of
the operation. Coagulation and fibrinolysis parameters
including activated clotting time (ACT), rate of thrombus for-
mation (K time), maximum amplitude, rate of thrombus for-
mation (a-angle), and lysis rate at 30 minutes (LY30) were
measured preoperatively, intraoperatively, and on POD 1, 2, and
3 by rapid thromboelastographic analysis (rapid-TEG) using a
 Y. Luo et al. / The Journal of Arthroplasty xxx (2019) 1e8
TEG5000 Thrombelastograph Hemostasis Analyzer (Haemo-
scope Corporation, Niles, IL). Kaolin and tissue factors were
used to activate blood samples for rapid-TEG [32]. Erythrocyte
sedimentation rate (ESR) and serum concentration of inflam-
matory markers including C-reactive protein (CRP) and
interleukin-6 (IL-6) were also measured preoperatively, intra-
operatively, and on POD 1, 2, and 3. VAS pain score was also
measured preoperatively and on POD 1, 2, and 3. The swelling
ratio was defined as the knee circumference of the surgical limb
divided by the knee circumference of the contralateral limb,
which includes the upper and lower poles of the patella and
was measured using the method described in the literature
[33,34]. The swelling evaluators were unaware of the treatment
group.
Statistical Analysis
The sample size was obtained mainly through the result of total
blood loss in the preliminary experiments. Preliminary data from
our institution showed that total blood loss on POD 2 was 982.19 ±
244.23 mL (mean ± standard deviation). According to our pre-
liminary experiment, we found that intravenous or topical injection
of CSS reduced the total blood loss on POD 2 by about 152 mL.
Assuming that intravenous or topical injection of CSS reduced total
blood loss by at least 152 mL on POD 2, we calculated that 44 pa-
tients were required in each group to detect a difference at a 2-
sided alpha significance level of 5% with a power of 85%.
Assuming a 10% loss in follow-up, we decided to include at least 49
subjects in each group.
Normally distributed, continuous data are shown as mean ±
standard deviation, while qualitative data are shown as frequencies
and percentages. Continuous variables were compared by 1-way
analysis of variance, and categorical variables were compared by
chi-squared or Fisher exact test. A P value of <.05 was considered
statistically significant. All analyses were performed using SPSS
(version 19.0, IBM).
Fig. 1. Flow diagram showing participant screening and allocation.
3
Results
Patients
A total of 275 patients scheduled for unilateral primary TKA
were screened from May to August 2018. Of these, 43 did not meet
inclusion criteria, leaving a total of 232 patients. Another 5 patients
did not consent to participate in the study, while 15 patients un-
derwent revision surgery and so were excluded. Twelve patients
withdrew for other reasons (such as simultaneous bilateral sur-
gery), leaving a total of 200 eligible patients. These were randomly
divided into 4 equal groups (Groups A, B, C, and D), with 50 patients
per group (Fig. 1). There were no differences in baseline charac-
teristics among the 4 groups (Table 1). Patients were followed up
for 3 months after surgery.
Outcome Measures
Total blood loss in groups A (609.92 ± 221.24 mL), B (753.16 ±
247.67 mL), and C (829.23 ± 297.45 mL) was lower than in group D
(972.19 ± 245.23 mL, Table 2, P < .05). Compared with other groups,
group A had the lowest total blood loss (P < .05). There was no
difference in total blood loss between groups B and C. HBL was also
lower in groups A (460.52 ± 211.36 mL), B (598.16 ± 236.63 mL), and
C (674.03 ± 297.9 mL) than in group D (818.39 ± 234.93 mL, P < .05).
And group A had the lowest HBL among all groups. There was no
difference in HBL between groups B and C. The reduction in he-
moglobin concentration on POD 2 was lower in groups A (2.04 ±
0.72 g/dL), B (2.45 ± 0.76 g/dL), and C (2.65 ± 0.82 g/dL) than in
group D (3.1 ± 0.73 g/dL, P < .05, Table 2). There was no difference in
IBL and platelet count among the 4 groups.
According to rapid-TEG analysis, the coagulation and fibrinolysis
functions of the 4 groups of patients were enhanced after operation
(Fig. 2). This result suggests that coagulation and fibrinolysis in-
crease from intraoperative and peak on the first day after surgery.
Using CSS does not increase coagulation and fibrinolysis rates, nor
4 Y. Luo et al. / The Journal of Arthroplasty xxx (2019) 1e8

Table 1
Patient Baseline Profiles.

Variable Group A (N ¼ 50) Group B (N ¼ 50) Group C (N ¼ 50) Group D (N ¼ 50) P Valuea

Age (y)b 67.3 ± 7.9 68.2 ± 6.6 64.5 ± 7.3 66.7 ± 8.3 .125
Sex (male/female)c 15/35 13/37 20/30 14/36 .472
BMI (kg/m2)b 25.55 ± 4.04 26.12 ± 3.59 25.63 ± 3.7 25.78 ± 3.21 .887
Operated side (left/right)c 25/25 22/28 18/32 24/26 .521
ASA classc .989
I 8 7 9 10
II 33 35 32 31
III 9 8 9 9
Preoperative valuesb
Hb (g/dL) 12.96 ± 1.37 13.12 ± 1.45 13.64 ± 1.16 13.16 ± 1.5 .103
Hct (L/L) 0.4 ± 0.03 0.41 ± 0.04 0.42 ± 0.03 0.41 ± 0.04 .107
PLT ( 109/L) 184.8 ± 65.2 204.6 ± 76.4 221.3 ± 67.4 198.1 ± 58.8 .08
INR 1.01 ± 0.08 0.99 ± 0.06 1.02 ± 0.07 1.01 ± 0.07 .323
D-dimer (mg/L FEU) 0.71 ± 0.53 0.86 ± 1.03 0.69 ± 0.0.71 0.89 ± 0.93 .556
Surgical timeb (min) 65.6 ± 7.3 65.8 ± 11.2 66.6 ± 8.1 66.5 ± 8.4 .925
Range of motion ( ) 92.2 ± 8.02 91.7 ± 8 92.4 ± 7.3 90.7 ± 8.8 .722

ASA, American Society of Anesthesiologists; BMI, body mass index; FEU, fibrinogen equivalent units; Hb, hemoglobin; Hct, hematocrit; INR, international normalized ratio;
PLT, platelet count.
a
P values were calculated by 1-way analysis of variance, the chi-square test, or the Fisher exact test.
b
These values are expressed as an average and standard deviation.
c
These values are presented in terms of patient numbers.

does it increase peaks. There were no differences in ACT, K time, a- vein thrombosis. The 4 groups showed no differences in throm-
angle, LY30, or maximum amplitude among the 4 groups at any of boembolic complications or mean LOS (P > .05). The LOS was 3.4 ±
the time points assessed, and all measurements were within the 0.7 days in group A, 3.4 ± 0.7 days in group B, 3.5 ± 0.6 days in group
normal range (Fig. 2). C, and 3.7 ± 0.7 days in group D. No incidence of PE, DVT, or other
CRP and ESR increased steadily after surgery until POD 2, while complications occurred (Table 3).
serum IL-6 levels peaked on POD 1 and then declined (Fig. 3). CRP,
IL-6, and ESR were also lower in groups A, B, and C than in group D
Discussion
on POD 1, 2, and 3 (P < .05). And the serum levels of ESR, IL-6, and
CRP in group A were the lowest. However, in terms of these 3 in-
In this study, we found that CSS combined with TXA reduced
flammatory indicators, there was no difference between group B
HBL and alleviated inflammation compared to placebo, without
and group C (P > .05). The knee pain gradually subsided after sur-
affecting blood coagulation or fibrinolytic function. This effect was
gery (Fig. 3). The differences in the pain scores on POD 1, 2, and 3
observed whether CSS was administered intravenously or topically
among the groups were statistically significant, and the pain was
alone. In addition, compared with the placebo group (group D),
lowest in group A at each time point within 3 days after surgery.
groups A, B, and C had less HBL, lower postoperative knee swelling,
The swelling rate of groups A, B, and C was significantly better than
lower levels of inflammatory markers, and better knee mobility in
that of group D on POD 1, 2, and 3 (P < .05), and compared with
the early postoperative period.
other groups, the swelling rate of group A was the lowest (Table 3).
CSS reduces permeability and enhances contractility of capil-
The ROM of all patients was significantly improved at discharge.
laries [11]. Only 2 previous studies have evaluated the hemostatic
The average ROM was 110.4 ± 3.5 in group A, 106.3 ± 2.5 in
effect of CSS in TKA, with both showing CSS to have a beneficial
group B, 105.9 ± 2.6 in group C, and 102.3 ± 2.7 in group D, with
effect on hemostatic, with no increased risk of asymptomatic DVT
statistically significant differences among the groups.
[12,35]. To ensure patient safety, TXA was used in addition to CSS
No anemia-related complications (including light-headedness,
because it has been shown to reduce blood loss during and after
presyncope, fatigue precluding participation in physical therapy,
major joint arthroplasties [36e39]. Therefore, this is the first study
palpitation, or shortness of breath not due to other causes)
to evaluate the perioperative hemostatic and anti-inflammatory
occurred in our study. Transfusions were performed in 1 patient in
effects of CSS combined with TXA in TKA.
group C (2%) and 3 patients in group D (6%) as a result of low
We found no difference in VTE among groups. However, any
postoperative hemoglobin concentration. Five patients in group A,
procoagulant intervention that reduces blood loss should care-
6 in group B, 6 in group C, and 7 in group D developed calf muscular
fully assess the risk of VTE, such as DVT and PE. Concerns about

Table 2
Outcomes Regarding Blood Loss.

Variable Mean and Standard Deviation P Valuea

Group A (N ¼ 50) Group B (N ¼ 50) Group C (N ¼ 50) Group D (N ¼ 50) I II III Ⅳ Ⅴ Ⅵ Ⅶ

Total blood loss (mL) 609.92 ± 221.24 753.16 ± 247.67 829.23 ± 297.45 972.19 ± 245.23 <.001 .027 <.001 <.001 .442 <.001 .028
Hidden blood loss (mL) 460.52 ± 211.36 598.16 ± 236.63 674.03 ± 297.9 818.39 ± 234.93 <.001 .03 <.001 <.001 .419 <.001 .02
Reduction in Hb (g/dL) 2.04 ± 0.72 2.45 ± 0.76 2.65 ± 0.82 3.1 ± 0.73 <.001 .036 <.001 <.001 .558 <.001 .015
Intraoperative blood 149.3 ± 38.2 155.6 ± 41.4 155.8 ± 37.1 154.2 ± 40.3 .852 .876 .864 .935 1 .999 .998
loss (mL)
PLT ( 109/L)
POD 1 181.6 ± 65.1 179.6 ± 64.7 198.8 ± 64.9 179.2 ± 54.3 .383 .999 .554 .998 .462 1 .445
POD 3 170.6 ± 59.1 171.6 ± 56.1 189.7 ± 58.5 163.4 ± 46.2 .139 1 .357 .926 .407 .895 .111

Bold values indicates statistical significance at P < .05.

POD, postoperative day; I, A vs B vs C vs D; II, A vs B; III, A vs C; Ⅳ, A vs D; Ⅴ, B vs C; Ⅵ, B vs D; Ⅶ, C vs D; Hb, hemoglobin; PLT, platelet count.
a
P values were calculated by 1-way analysis of variance and the Tukey post hoc multiple comparison test.
 Y. Luo et al. / The Journal of Arthroplasty xxx (2019) 1e8
Fig. 2. Perioperative thromboelastographic analysis showing (A) activated clotting time (ACT), (B) K time, (C) LY30, (D) a-angle, and (E) maximum amplitude (MA) value. POD,
postoperative day; pre, preoperative; intra, intraoperative.
this complication have prevented some surgical centers from
routinely using TXA [40,41]. However, a number of systematic
reviews have determined the effectiveness and safety of TXA in a
selective orthopedic setting [42,43]. In addition, in this study, we
found that CSS did not increase the incidence of early post-
operative thromboembolic complications. On the basis of rapid-
TEG analysis, the use of CSS did not increase the rate of post-
operative coagulation and fibrinolysis. In addition, there was no
increase in peak coagulation and fibrinolysis parameters in groups
A, B, and C compared with group D. According to ACT and K values,
there was no difference in the number of patients with
hypercoagulability between groups. Early postoperative use of
lower-extremity strength exercise, intermittent inflatable lower-
extremity venous pump, low-molecular-weight heparin injec-
tion, and other comprehensive VTE prevention strategies were
routinely used. These factors may explain why CSS administration
did not increase the incidence of postoperative thromboembolic
complications.
An appropriate inflammatory response will repair tissue dam-
age without complications [44]. However, persistent systemic
inflammation can lead to immune damage and adverse conse-
quences, such as pain, nausea, and other complications [45].
Therefore, drugs or other methods to inhibit this excessive in-
flammatory response are desirable. We found that intravenous or
topical administration of CSS led to lower secretion of proin-
flammatory cytokine IL-6 and of inflammation markers such as CRP
and ESR, compared with TXA alone. These results show that CSS has
anti-inflammatory effects in TKA. There are several possible reasons
for these effects. Firstly and more importantly, CSS combined with
TXA reduces total blood loss, which reduces CRP and IL-6 levels and
can in turn reduce inflammation and surgical trauma [46,47]. Sec-
ond, CSS reduces capillary permeability, which can reduce pro-
duction and secretion of inflammatory factors. Although the
correlation between inflammatory markers and clinical outcomes
5
is still controversial [46e48], significant differences in the level of
inflammatory markers among the 4 groups are also reflected in
clinical results such as postoperative swelling rate, VAS pain score,
and ROM. The ROM mean values of group A, group B, and group C
were higher than that of group D, and the difference between
groups was statistically significant. However, group A has the
highest ROM. The reason for this may be during the immediate
postoperative period, lower HBL and pain scores, reduced knee
effusion, and reduced leg swelling may relieve strain on the
quadriceps, thereby allowing patients to more easily perform a
straight leg raise and to demonstrate improved early ROM. In
addition, the trend of difference between ROM groups was also
reflected in HBL, VAS score, and swelling rate. Previous studies have
shown that the magnitude of difference required to be sure that a
difference in ROM is not attributable to variation in measurement
and performance between measurements is approximately 10
[49]. In addition, although ROM has significant statistical differ-
ences, the effect size is quite small. Therefore, it is not clear whether
this difference is clinically relevant.
In our study, total blood loss was lower in groups A, B, and C
than in group D, with no difference between groups B and C. And
group A had the lowest total blood loss among all groups. This may
be related to the half-life of CSS (2.51 ± 0.95 h) [35]. The addition
of CSS 3 hours after surgery in group A may extend the hemostasis
time of CSS to about 6 hours after surgery. However, studies have
reported that 65% of postoperative blood loss (including high
fibrinolysis) occurred within 8 hours of the operation [50,51], so
the total blood loss in group A was the lowest. Similarly, HBL was
lower in groups A, B, and C than in group D, with no difference
between groups B and C. These results show that combining TXA
with CSS is more effective than TXA alone in reducing HBL.
Furthermore, postoperative HBL accumulated in the third
anatomic space, leading to postoperative inflammation, lower
limb swelling, and pain [29]. Therefore, reducing HBL can reduce
6 Y. Luo et al. / The Journal of Arthroplasty xxx (2019) 1e8

Fig. 3. Mean serum concentration of inflammatory markers in the perioperative period, including (A) ESR, (B) CRP, and (C) IL-6. Mean longitudinal VAS pain score of each group (D).
x P < .05 between groups A and D, Fp < .05 between groups B and D, and z P < .05 between groups C and D. Jp < .05 between groups A and B, and 4 P < .05 between groups A and C.
ESR, erythrocyte sedimentation rate; CRP, C-reactive protein; IL-6, interleukin-6.

postoperative inflammation, lower limb swelling, and pain. This is
why inflammatory cytokines, VAS pain scores, and swelling rates
were lower in groups A, B, and C than in group D. However,
although there was a statistically significant difference in VAS
scores, the score difference between the 2 groups did not reach
the minimal clinically important difference of 22.6 mm previously
reported [52]. Therefore, this difference in VAS scores has no
clinical significance.
There was statistically significant difference in total blood loss,
but no significant difference in transfusion rate. It was difficult
to compare transfusion rates in the 4 groups, as only 4 of 200
patients needed a transfusion. In addition, no anemia-related
complications occurred in all patients in this study. All
blood transfusions were caused by postoperative hemoglobin
concentration <70 g/L. The reason for this is that this may reflect
the fact that in our facility, during preoperative preparation, all
anemia patients receive iron and/or erythropoietin to ensure
preoperative hemoglobin levels of >110 g/L, and if this level is not
reached, surgery will be delayed. This rigorous approach is based
on reports that preoperative hemoglobin levels are independent
risk factors for transfusion [53].
Our study has several limitations. First, we did not include pa-
tients with high risk of venous and arterial thrombosis and pre-
existing coagulopathy. Therefore, our results may not apply to these
patients. Secondly, although this study showed a low frequency of
DVT, PE, and other complications, a larger sample size would be
needed to adequately evaluate for significant group differences in
these adverse events.

Table 3
Postoperative Outcomes.

Variable Mean and Standard Deviation P Valuea

Group A (N ¼ 50) Group B (N ¼ 50) Group C (N ¼ 50) Group D (N ¼ 50) I II III Ⅳ Ⅴ Ⅵ Ⅶ

Transfusion rate (n, %)c 0 0 1 (2%) 3 (6%) .196 d 1.000 .242 1.000 .242 .617
Swelling ratioc (%)
POD 1 103.2 ± 1.1 104.2 ± 1.2 104.5 ± 1.1 106 ± 1.4 <.001 <.001 <.001 <.001 .711 <.001 <.001
POD 2 104.5 ± 0.9 105.7 ± 0.9 105.9 ± 1.0 108.1 ± 1.2 <.001 <.001 <.001 <.001 .644 <.001 <.001
POD 3 104.7 ± 0.7 106.6 ± 0.8 106.9 ± 0.9 110.4 ± 0.9 <.001 <.001 <.001 <.001 .229 <.001 <.001
b
LOS (d) 3.4 ± 0.7 3.4 ± 0.7 3.5 ± 0.6 3.7 ± 0.7 .068 .99 .99 .144 .928 .072 .26
ROM at discharge ( )b 110.4 ± 3.5 106.3 ± 2.5 105.9 ± 2.6 102.3 ± 2.7 <.001 <.001 <.001 <.001 .922 <.001 <.001
Postoperative complicationsc
DVT 0 0 0 0 d d d d d d d
PE 0 0 0 0 d d d d d d d
Superficial infection 0 0 0 0 d d d d d d d
CMVT 5 6 6 7 .99 1.000 1.000 .76 1.000 1.000 1.000

Bold values indicates statistical significance at P < .05.

POD, postoperative day; LOS, length of stay; ROM, range of motion; DVT, deep vein thrombosis; PE, pulmonary embolism; CMVT, calf muscular vein thrombosis; I, A vs B vs C vs
D; II, A vs B; III, A vs C; Ⅳ, A vs D; Ⅴ, B vs C; Ⅵ, B vs D; Ⅶ, C vs D.
a
P values were calculated by 1-way analysis of variance, the chi-square test, or the Fisher exact test.
b
These values are expressed as an average and standard deviation.
c
These values are presented in terms of patient numbers.
 Y. Luo et al. / The Journal of Arthroplasty xxx (2019) 1e8
Conclusions
In conclusion, our study shows that compared with the placebo,
the patients using CSS combined with TXA had less perioperative
blood loss, lower postoperative knee swelling, lower levels of bio-
markers of inflammation, better early knee mobility, and did not
increase in the incidence of early postoperative thromboembolism
complications. In particular, the topical combined intravenous use
of CSS plus TXA has the best effect. Therefore, CSS has hemostatic
and anti-inflammatory effects. Further research with a larger pa-
tient cohort is required to ensure the repeatability of these results.
Acknowledgments
None.
References
[1] Callaghan JJ, O’Rourke MR, Liu SS. Blood management: issues and options.
J Arthroplasty 2005;20(4 suppl 2):51e4.
[2] Keating EM, Meding JB, Faris PM, Ritter MA. Predictors of transfusion risk in
elective knee surgery. Clin Orthop Relat Res 1998;357:50e9.
[3] Mylod Jr AG, France MP, Muser DE, Parsons JR. Perioperative blood loss
associated with total knee arthroplasty: a comparison of procedures per-
formed with and without cementing. J Bone Joint Surg Am 1990;72:1010e2.
[4] Su EP, Su S. Strategies for reducing peri-operative blood loss in total knee
arthroplasty. Bone Joint J 2016;98-b:98e100.
[5] Levine BR, Haughom B, Strong B, Hellman M, Frank RM. Blood management
strategies for total knee arthroplasty. J Am Acad Orthop Surg 2014;22:
361e71.
[6] Shander A, Van Aken H, Colomina MJ, Gombotz H, Hofmann A, Krauspe R,
et al. Patient blood management in Europe. Br J Anaesth 2012;109:55e68.
[7] Gombotz H, Rehak PH, Shander A, Hofmann A. The second Austrian bench-
mark study for blood use in elective surgery: results and practice change.
Transfusion 2014;54:2646e57.
[8] Hallstrom B, Singal B, Cowen ME, Roberts KC, Hughes RE. The Michigan
experience with safety and effectiveness of tranexamic acid use in hip and
knee arthroplasty. J Bone Joint Surg Am 2016;98:1646e55.
[9] Good L, Peterson E, Lisander B. Tranexamic acid decreases external blood loss
but not hidden blood loss in total knee replacement. Br J Anaesth 2003;90:
596e9.
[10] Chen X, Zhu X, Yang S, et al. Tranexamic acid treatment decreases hidden
blood loss in total knee arthroplasty. Am J Ther 2015;23:e1397e405.
[11] Matsumoto Y, Hayashi T, Hayakawa Y, Shinbo M, Niiya K, Sakuragawa N.
Carbazochrome sodium sulfonate (CSS) decreases the accumulation of tissue-
type plasminogen activator in culture medium of human umbilical vein
endothelial cells. Blood Coagul Fibrinolysis 1995;6:233.
[12] Hayakawa M, Gando S. Systemic and local hemostatic agents. Jpn J Thromb
Hemost 2009;20:278e80.
[13] Yang F, Zhang AR. The hemostasis effect of carbazochrome sodium sulfonate
and sodium chloride injection in obstetrics and gynecology department. China
Health Stand Management 2018;9:78e9. https://doi.org/10.3969/j.issn.1674-
9316.
[14] Yang DX, Yang L, Wang F. Clinical effect of different drugs combined with Kalo
sodium in treatment of bronchiectasis with hemoptysis. Hebei Med 2018;24:
873e6. https://doi.org/10.3969/j.issn.1006-6233.
[15] Onodera T, Majima T, Sawaguchi N, Kasahara Y, Ishigaki T, Minami A. Risk of
deep venous thrombosis in drain clamping with tranexamic acid and carba-
zochrome sodium sulfonate hydrate in total knee arthroplasty. J Arthroplasty
2012;27:105e8.
[16] Passali GC, De Corso E, Bastanza G, Di Gennaro L, HHT Gemelli Study Group.
An old drug for a new application: carbazochrome-sodium-sulfonate in HHT.
J Clin Pharmacol 2015;55:601e2.
[17] Oh-oka H, Yamada T, Noto H, Umeyama T, Kadekawa K, Ashitomi K, et al.
Effect of carbazochrome sodium sulfonate on refractory chronic prostatitis. Int
J Urol 2014;21:1162e6.
[18] Tassniyom S, Vasanawathana S, Dhiensiri T, Nisalak A, Chirawatkul A. Failure
of carbazochrome sodium sulfonate (AC-17) to prevent dengue vascular
permeability or shock: a randomized, controlled trial. J Pediatr 1997;131:
525e8.
[19] Funahara Y, Sumarmo, Shirahata A, Harun SR, Setiabudy-Dharma R,
Nathin MA, et al. Protection against marked plasma leakage in dengue hae-
morrhagic fever by infusion of carbazochrome sodium sulfonate (AC-17).
Southeast Asian J Trop Med Public Health 1987;18:356e61.
[20] Moriuchi H, Kashiwada Y, Arai I, Yuizono T. Effects of carbazochrome sodium
sulfonate (AC-17) on oleic acid-induced lung injury. Pharmacol Toxicol
1995;77:238e40.
[21] Alejandria MM. Dengue haemorrhagic fever or dengue shock syndrome in
children. BMJ Clin Evid 2009;12:0917.
7
[22] Squizzato A, Lussana F, Cattaneo M. Post-operative arterial thrombosis with
non-vitamin K antagonist oral anticoagulants after total hip or knee arthro-
plasty. Thromb Haemost 2015;114:237e44.
[23] Jans O, Bandholm T, Kurbegovic S, Solgaard S, Kjaersgaard- Andersen P,
Johansson PI, et al. Lundbeck Foundation Centre for Fast-Track Hip and Knee
Replacement Collaborative Group. Postoperative anemia and early functional
outcomes after fast-track hip arthroplasty: a prospective cohort study.
Transfusion 2016;56:917e25.
[24] Schmitt J, Lange T, Günther KP, Kopkow C, Rataj E, Apfelbacher C, et al.
Indication criteria for total knee arthroplasty in patients with osteoarthritis - a
multi-perspective consensus study. Z Orthop Unfall 2017;155:539e48.
[25] Jiang Y, Sanchez-Santos MT, Judge AD, Murray DW, Arden NK. Predictors of
patient-reported pain and functional outcomes over 10 years after primary
total knee arthroplasty: a prospective cohort study. J Arthroplasty 2017;32:
92e100.e2.
[26] National Health and Family Planning Commission of the People’s Republic of
China. [Technical specifications of clinical blood transfusion]. 2001. Chinese, http://
www.nhfpc.gov.cn/zwgkzt/wsbysj/200804/18676.shtml [accessed 22.09.17].
[27] Gross JB. Estimating allowable blood loss: corrected for dilution. Anesthesi-
ology 1983;58:277e80.
[28] Nadler SB, Hidalgo JH, Bloch T. Prediction of blood volume in normal human
adults. Surgery 1962;51:224e32.
[29] Liu X, Zhang X, Chen Y, Wang Q, Jiang Y, Zeng B. Hidden blood loss after total
hip arthroplasty. J Arthroplasty 2011;26:1100e1105.e1.
[30] Wang D, Xu J, Zeng WN, Zhou K, Xie TH, Chen Z, et al. Closed suction drainage
is not associated with faster recovery after total knee arthroplasty: a pro-
spective randomized controlled study of 80 patients. Orthop Surg 2016;8:
226e33.
[31] Zhou K, Wang H, Li J, Wang D, Zhou Z, Pei F. Non-drainage versus drainage in
tourniquet-free knee arthroplasty: a prospective trial. ANZ J Surg 2017;87:
1048e52.
[32] Cotton BA, Faz G, Hatch QM, Radwan ZA, Podbielski J, Wade C, et al. Rapid
thrombelastography delivers real-time results that predict transfusion within
1 hour of admission. J Trauma 2011;71:407e14. https://doi.org/10.1097/TA.0
b013e31821e1bf0.
[33] Xie J, Ma J, Yao H, Yue C, Pei F. Multiple boluses of intravenous tranexamic acid
to reduce hidden blood loss after primary total knee arthroplasty without
tourniquet: a randomized clinical trial. J Arthroplasty 2016;31:2458e64.
[34] Huang Z, Ma J, Shen B, Pei F. Combination of intravenous and topical appli-
cation of tranexamic acid in primary total knee arthroplasty: a prospective
randomized controlled trial. J Arthroplasty 2014;29:2342e6.
[35] Akizuki S, Yasukawa Y, Takizawa T. A new method of hemostasis for
cementless total knee arthroplasty. Bull Hosp Jt Dis 1997;56:222e4.
[36] Sukeik M, Alshryda S, Haddad FS, Mason JM. Systematic review and meta-
analysis of the use of tranexamic acid in total hip replacement. J Bone Joint
Surg Br 2011;93:39e46.
[37] Imai N, Dohmae Y, Suda K, Miyasaka D, Ito T, Endo N. Tranexamic acid for
reduction of blood loss during total hip arthroplasty. J Arthroplasty 2012;27:
1838e43.
[38] MacGillivray RG, Tarabichi SB, Hawari MF, Raoof NT. Tranexamic acid to
reduce blood loss after bilateral total knee arthroplasty: a prospective, ran-
domized double blind study. J Arthroplasty 2011;26:24e8.
[39] Noordin S, Waters TS, Garbuz DS, Duncan CP, Masri BA. Tranexamic acid re-
duces allogenic transfusion in revision hip arthroplasty. Clin Orthop Relat Res
2011;469:541e6.
[40] Xie J, Ma J, Kang P, Zhou Z, Shen B, Yang J, et al. Does tranexamic acid alter the
risk of thromboembolism following primary total knee arthroplasty with
sequential earlier anticoagulation? A large, single center, prospective cohort
study of consecutive cases. Thromb Res 2015;136:234e48.
[41] Wu Q, Zhang HA, Liu SL, Meng T, Zhou X, Wang P. Is tranexamic acid clinically
effective and safe to prevent blood loss in total knee arthroplasty? A meta-
analysis of 34 randomized controlled trials. Eur J Orthop Surg Traumatol
2015;25:525e41.
[42] Zhou X, Tao L, Li J, Wu LD. Do we really need tranexamic acid in total hip
arthroplasty? A meta-analysis of nineteen randomized controlled trials. Arch
Orthop Trauma Surg 2013;133:1017e27.
[43] Chen W-P. Effectiveness and safety of tranexamic acid in reducing blood loss
in total knee arthroplasty: a meta-analysis. J Bone Joint Surg Am 2012;94:
1153e9.
[44] Oelsner WK, Engstrom SM, Benvenuti MA, An T, Jacobson R, Polkowski G, et al.
Characterizing the acute phase response in healthy patients following total
joint arthroplasty: predictable and consistent. J Arthroplasty 2017;32:
309e14.
[45] AbdelSalam H, Restrepo C, Tarity TD, Sangster W, Parvizi J. Predictors of
intensive care unit admission after total joint arthroplasty. J Arthroplasty
2012;27:720e5.
[46] Bergin PF, Doppelt JD, Kephart CJ, Benke MT, Graeter JH, Holmes AS, et al.
Comparison of minimally invasive direct anterior versus posterior total hip
arthroplasty based on inflammation and muscle damage markers. J Bone Joint
Surg Am 2011;93:1392e8.
[47] Manner P. Good start on using biochemical markers to compare surgical
trauma in total hip replacement approaches: commentary on an article by
Patrick F. Bergin, MD, et al.: “Comparison of minimally invasive direct anterior
versus posterior total hip arthroplasty based on inflammation and muscle
damage markers”. J Bone Joint Surg Am 2011;93:e89.
8 Y. Luo et al. / The Journal of Arthroplasty xxx (2019) 1e8

[48] Poehling-Monaghan KL, Taunton MJ, Kamath AF, Trousdale RT, Sierra RJ,
Pagnano MW. No correlation between serum markers and early functional
outcome after contemporary THA. Clin Orthop Relat Res 2017;475:452e62.
[49] Stratford PW, Kennedy DM, Robarts SF. Modelling knee range of motion post
arthroplasty: clinical applications. Physiother Can 2010;62:378e87.
[50] Prasad N, Padmanabhan V, Mullaji A. Comparison between two methods of
drain clamping after total knee arthroplasty. Arch Orthop Trauma Surg
2005;125:381e4.
[51] Howes JP, Sharma V, Cohen AT. Tranexamic acid reduces blood loss after knee
arthroplasty. J Bone Joint Surg Br 1996;78:995e6.
[52] Danoff JR, Goel R, Sutton R, Maltenfort MG, Austin MS. How much pain is sig-
nificant? Defining the minimal clinically important difference for the visual
analog scale for pain after total joint arthroplasty. J Arthroplasty 2018;33:S71e5.
[53] Aderinto J, Brenkel IJ. Pre-operative predictors of the requirement for blood
transfusion following total hip replacement. J Bone Joint Surg Br 2004;86-B:
970e3.