Content uploaded by Zeyu Huang
Author content
All content in this area was uploaded by Zeyu Huang on Jul 26, 2015
Content may be subject to copyright.
1 3
Arch Orthop Trauma Surg (2014) 134:561–570
DOI 10.1007/s00402-014-1948-1
KNEE ARTHROPLASTY
Comparison of three different tourniquet application strategies
for minimally invasive total knee arthroplasty: a prospective
non‑randomized clinical trial
Ze Yu Huang · Fu Xing Pei · Jun Ma · Jing Yang ·
Zong Ke Zhou · Peng De Kang · Bin Shen
Received: 18 September 2013 / Published online: 11 February 2014
© Springer-Verlag Berlin Heidelberg 2014
preoperatively. The HSS knee score, VAS pain score, range
of motion (ROM), limb swelling and hospital stays were
also recorded.
Results The mean levels of Hb and Hct were lower in
Group C (104.2 ± 10.4 g/L, 31.8 ± 3.2 %) than those in
Groups A (111.4 ± 14.4 g/L, p = 0.035; 34.1 ± 4.1 %,
p = 0.032) and B (112.8 ± 14.3 g/L, p = 0.013;
34.5 ± 3.7 %, p = 0.011) immediately after the surgery.
Compared with Groups A and B, both serum inflammation
and muscle damage markers were lower in Group C. There
were no significant differences between the groups in terms
of HSS knee score, ROM, estimated blood loss, swelling
ratio, VAS pain score and hospital stays.
Conclusions Using a tourniquet full time in minimally
invasive TKA causes less intraoperative blood loss and
more excessive inflammation and muscle damage. How-
ever, the advantage of part-time using tourniquet did not
show in early functional outcomes.
Keywords Total knee arthroplasty · Blood loss ·
Tourniquet application strategies · Minimally invasive total
knee arthroplasty · Muscle damage markers · Inflammation
markers
Introduction
Total knee arthroplasty (TKA) has become a common
orthopedic procedure. Recently, minimally invasive TKA
(MIS-TKA) has been put under the spotlight. Reducing
blood loss and functional recovery of the patients treated
with TKA are surgeons’ main concerns [1]. Though the
application of a pneumatic tourniquet in TKA remains con-
troversial [2–4], it is still a routine practice performed in
TKA, thought to be of great benefit to the reduction of the
Abstract
Introduction It is still controversial on the optimal timing
of tourniquet used in total knee arthroplasty (TKA). Most
previous studies focused on the comparison of different
tourniquet application in controversial TKA, while the aim
of our work was to compare three strategies of tourniquet
application in minimally invasive TKA.
Materials and methods 90 patients were enrolled in this
study. Based on the different tourniquet application strate-
gies, they were divided into three groups. Group A: using
tourniquet during the whole surgery; Group B: tourniquet
inflated before incision and deflated after the hardening of
the cement; Group C: using tourniquet during the cemen-
tation. Blood loss and serum levels of C-reactive pro-
tein, IL-6, creatine kinase and myoglobin were checked
Z. Y. Huang · F. X. Pei · J. Ma · J. Yang · Z. K. Zhou · P. D. Kang ·
B. Shen (*)
Department of Orthopaedics, West China Hospital, Sichuan
University, 37# Wainan Guoxue Road, Chengdu 610041,
People’s Republic of China
e-mail: 492385233@qq.com
Z. Y. Huang
e-mail: Zey.huang@gmail.com
F. X. Pei
e-mail: peifuxing1951@yahoo.cn
J. Ma
e-mail: dr.majun@foxmail.com
J. Yang
e-mail: yangjingcd1961@126.com
Z. K. Zhou
e-mail: ZongKeZhou@126.com
P. D. Kang
e-mail: pengdekang1965@126.com
562 Arch Orthop Trauma Surg (2014) 134:561–570
1 3
blood loss, thus it can improve visualization, facilitate the
cementing technique and reduce the duration of surgery
[5]. However, ischemia and later hemodynamic changes
caused by tourniquet may lead to a higher rate of swell-
ing and stiffness of the joints [6], delayed wound healing
[7], nerve injury [8], postoperative pain and even deep vein
thrombosis [3, 9]. So it is quite essential for surgeons to
balance the benefits and complications when using tourni-
quet in TKA.
Nowadays, there are several different tourniquet applica-
tion strategies widely used for TKA. However, no consen-
sus has been achieved with regards to defining an optimal
tourniquet application strategy. The mostly used tourniquet
application strategies can be classified into three main pat-
terns: One is the tourniquet inflated before incision and
deflated after the closure of the incision; another is the tour-
niquet inflated before incision and deflated after the harden-
ing of the cement; the other is the tourniquet inflated before
prosthesis placement and deflated after the hardening of the
cement. Though several trials have compared these differ-
ent strategies, no consensuses have been reached [10–13].
What is more, to our knowledge, there were no similar tri-
als ever performed to compare these three practices in the
minimally invasive TKA.
In order to clarify the roles of tourniquet and compare
the effect of these different strategies, we conducted this
prospective clinical trial. Unlike the previous studies, firstly
we compared these practices in MIS-TKAs; secondly this
study included the serum inflammation and muscle damage
markers as one of the primary outcomes besides blood loss.
Functional outcomes were used as the secondary outcomes.
The authors hypothesized that there was no difference
between these strategies.
Materials and methods
Study design
This prospective non-randomized comparative trial,
approved by the Institutional Review Board of West China
Hospital of Sichuan University, was performed at our
center from June 2012 to July 2012.
Patients’ cohort
All patients suffering from symptomatic osteoarthritis of
the knee, not responding to a trial of the conservative treat-
ment, were screened as potential participants for this trial.
They were informed about the purpose of the study by giv-
ing a written informing consent signed before the surgery.
After that, these patients were assessed at the satisfaction of
inclusion/exclusion criteria prerequisite for enrollment in
the study (Table 1). Each patient selected the practice that
he or she preferred after the surgeon (B.S.) explained the
potential advantages and disadvantages of each practice.
According to the different practices, patients were divided
into three groups: (1) Group A: tourniquet inflated before
incision and deflated after the closure of the incision; (2)
Group B: tourniquet inflated before incision and deflated
after the hardening of the cement; (3) Group C: the tourni-
quet inflated before prosthesis placement and deflated after
the hardening of the cement.
Surgical procedure and device
All surgeries were performed by one joint surgeon (B.S.)
in the same laminar air flow operation room. The surgeon
Table 1 Exclusion criteria
Age <50 or >80 years
BMI <20 or >30 kg/m2
ASA ≥ III physical status
Using anticoagulation therapy
History of bleeding disorders
History of any thromboembolic complications or acute cardiac insufficiency
Hematological diseases
Severe deformities of the knee which might increase the complexity of the operative procedure: >30° flexion contracture, >30° valgus, >20°
varus
Previous knee operation history
Active malignancy
Operative exclusion criteria
Synovectomy
Patella resurfacing
Lateral retinacular release (LRR)
563Arch Orthop Trauma Surg (2014) 134:561–570
1 3
had performed more than 500 MIS-TKAs, using the mini-
midvastus (MMV) approach. Each tourniquet application
strategy was taken in more than 500 patients treated with
TKAs prior to the start of this study. All patients who had
a cemented TKA using Scorpio NRG (Stryker Allendale,
NJ, USA) were given 1.5 g cefuroxime within 30 min prior
to skin incision, and general anesthesia was administered
in all cases. A midline skin incision and MMV approach
were used in all cases. We applied tourniquet in 100 mmHg
above systolic blood pressure. The tourniquet system used
in this study was 500ELC tourniquet system (VBM Ger-
many) and the cuff was 76 cm long and 8.0 cm wide. A sin-
gle layer of cast padding was applied between the skin and
cuff. Vacuum wound drainage was used in every patient.
In all three groups wound was closed after wound irriga-
tion and hemostasis and then was wrapped with elastic
bandages.
Postoperative care
After the surgery, patients were first transferred to the anes-
thesia recovery unit for a 2-h period of postoperative care,
then to the in-patients unit. Immediately they arrived at the
in-patients unit, cold pack was used to the surgical sites for
12 h. The drainage was kept releasing for 24 h before being
moved. Celecoxib was administered orally with a regular
dose of 200 mg bid for the first postoperative days and then
administered as requested.
At 7:30 a.m. on the first postoperative day, bladder cath-
eter and all the monitors were removed and non-restrict
solid food intake was allowed for all the patients. A physi-
otherapist supervised and assisted the patients in taking the
daily functional training, including muscle power training,
passive and active range-of-motion training, and walk-
ing training. The criterion for a blood transfusion was set
as a hemoglobin (Hb) level of <70 g/L or 70–100 g/L with
symptomatic anemia.
Half dose of low-molecular weight heparin (LMWH)
(0.2 mL 2,000 IU) was started 6 h after the surgery
and repeated at 24-h intervals with a full dose (0.4 mL
4,000 IU) on the subsequent days if there was no contrain-
dication. Besides, an intermittent foot sole pump system
was also used as a routine practice to prevent deep vein
thrombosis (DVT). After the discharge, 10 mg rivaroxaban
was administered orally to the patients for 10 days, if no
bleeding events happened.
The criteria for discharge were as follows: (1) stable
surgical wounds; (2) patients can demonstrate a knee flex
of 90° and a smooth straight-leg raise. After the discharge,
patients were followed up in the clinic for the first 3 weeks
and then every 3 months. A surgeon and a physical thera-
pist examined the patients and made the next postoperative
reconstructive protocols for them.
Outcomes assessment
To ensure the similar cohorts, we calculated the age at the
time of surgery, body mass index (BMI), American Society
of Anesthesiologists (ASA) grade, knee range of motion
(ROM) and preoperative hospital for special surgery
(HSS) knee scores. The preoperative knee circumferences
at both the upper pole of the patella and the lower pole of
the patella were measured to ensure similar patient body
habitus. Measurements were taken when the patients were
placed in their knee extension position.
We compared operative time, tourniquet time, estimated
blood loss, duration of hospitalization, drainage volume
and transfusion requirements among the three groups. As
the primary outcomes of this study, Hb and hematocrits
(Hct) were measured immediately preoperatively, immedi-
ately postoperatively and on postoperative days 1, 2 and 3.
Estimated blood loss was calculated by the modified Gross
formula [14, 15]. C-reactive protein (CRP) and interleu-
kin-6 (IL-6) are well-recognized measures for inflamma-
tion and surgical insult [16]. According to White et al.’s
report [17] that CRP rises more slowly postoperatively than
the other indexes, we measured CRP level on immediately
preoperatively, postoperative days 1, 2 and 3. Muscle dam-
age markers such as serum creatine kinase (CK) and myo-
globin were also measured at each time point. All collected
serum samples were stored at −20 °C after being labeled
without patients’ ID in a blinded fashion. Measurements
were done by the department of laboratory medicine of
West China hospital certificated by CAP (Clinic American
Pathology).
Function was assessed by means of HSS and ROM.
HSS score and knee ROM were got at the time of charge,
discharge, each follow-up time points to find out the dif-
ferent functional outcomes between groups. The ROM of
each knee was measured twice in the supine position with a
standard (60-cm) goniometer preoperatively and at the time
of discharge by two observers (ZY.H. and J.M.), both of
them blinded to the type of tourniquet application. In order
to assess intra-observer reliability, the goniometer measure-
ment was performed thrice (with a 4-hour interval between
each measurement). Finally, we found that the chance-
corrected kappa coefficient for intra-observer agreement
ranged from 0.76 to 0.84. Perioperative visual analog scale
(VAS) pain scores were obtained by a physician assistant
also blinded to the type of tourniquet application. All clini-
cal data were compiled and collected by a separate research
associate.
Statistical analysis
Based on our previous study [1], a power of 80 % to detect
a significant difference (p < 0.05, two-sided) of serum
564 Arch Orthop Trauma Surg (2014) 134:561–570
1 3
inflammation markers (IL-1 and IL-6) level, 29 patients
were needed for each group. And according to the previ-
ous study [18] on volume loading test for the detection
of hypovolemia and dehydration, Hahn et al. found that a
sample size of almost 30 subjects per group was needed to
detect a difference of 4.1 % between group means of 13.3
and 9.2 % with known group standard deviations of 3.3 and
5.0 (α = 0.05, power 95 %, two-sided two sample student
t test). Finally we enrolled 90 patients, 30 patients in each
group.
All data management and statistical analysis were per-
formed with SPSS version 18.0 software (SPSS Inc.,
Chicago, IL, USA). If the Levene’s test for comparison
of variances did not reject hypothesis on equality of vari-
ance between groups, mean values were compared using
ANOVA with Bonferroni correction. Kruskall–Wallis test
was applied to non-normally distributed data. Analysis was
performed with significance level α = .05 (two-sided).
Results
Patient demographics
Totally 90 patients (30 in each group) were enrolled in
our study. Patient demographics showed no statistical sig-
nificant difference among the three groups in age, gen-
der, BMI, ASA grade, operated sides, preoperative ROM,
preoperative VAS pain score and knee circumference
(Table 2). As shown in Table 2, mean duration of tour-
niquet time in Group C (11.2 ± 1.0 min, p < 0.001) was
much lower than in Groups A(88.6 ± 14.4 min, p < 0.001)
and B(61.2 ± 12.6 min, p < 0.001). Duration of surgery,
postoperative VAS pain scores and length of hospital stay
in each group were not significantly different. No signifi-
cant difference between groups was found in the severity of
limb swelling (Fig. 1).
Blood loss
Both Hb and Hct levels in the three groups showed sig-
nificant decreases after the surgery and they both reached
the lowest point on POD3. The mean levels of Hb and Hct
were lower in Group C (104.2 ± 10.4 g/L, 31.8 ± 3.2 %)
than those in Groups A(111.4 ± 14.4 g/L, p = 0.035;
34.1 ± 4.1 %, p = 0.032) and B (112.8 ± 14.3 g/L,
p = 0.013; 34.5 ± 3.7 %, p = 0.011) immediately after the
surgery. However, such differences were not detected on
POD1, POD2 and POD3 (Fig. 2).
No significant differences were observed between
groups in terms of estimated blood loss (Table 3). Post-
operatively, three patients from Group A, three patients
from Group B and five patients from Group C received
a blood transfusion. The average blood transfusion was
0.3 ± 0.92 U for Group A, 0.3 ± 0.95 U for Group B and
0.47 ± 1.1 U for Group C (Table 3).
Table 2 Preoperative and intraoperative demographics
ASA American Society of Anesthesiologists, BMI body mass index, HSS Hospital for Special Surgery, ROM range of motion
† p stands for p value of Group A vs. B vs. C, p1 stands for p value of Group A vs. B, p2 stands for p value of Group A vs. C, p3 stands for p
value of Group B vs. C
‡ Significantly different
a The values are given as mean ± standard deviation
Demographic Group A Group B Group C p†p1
†p2
†p3
†
Agea (year) 66.2 ± 7.6 66.1 ± 5.8 66.3 ± 6.1 0.993 0.953 0.953 0.906
Gender (female/male) 20/10 20/10 19/11 0.952 1 1 1
BMIa (kg/m2) 26.1 ± 3.0 25.9 ± 3.3 26.5 ± 2.4 0.737 0.796 0.611 0.444
Preo hemoglobina (g/L) 130.7 ± 11.8 128.0 ± 15.6 128.2 ± 12.1 0.674 0.432 0.455 0.969
Preo hematocrita (%) 39.9 ± 3.3 39.7 ± 4.2 39.6 ± 4.3 0.954 0.820 0.770 0.948
ASA gradea2.0 ± 0.5 2.0 ± 0.6 2.0 ± 0.6 0.886 0.806 0.806 0.624
Side (right/left) 15/15 17/13 14/16 0.733 0.796 1 0.606
HSS (points)a56.3 ± 6.8 57.7 ± 6.5 57.0 ± 5.2 0.682 0.383 0.677 0.647
ROM (°)a97.5 ± 9.5 98.8 ± 9.4 99.8 ± 9.5 0.635 0.588 0.344 0.684
VAS Pain score 7.6 ± 0.81 7.8 ± 0.61 7.8 ± 0.71 0.463 0.283 0.283 1.0
Preo knee circumference (cm)a
Upper pole of patellaa37.5 ± 3.0 36.5 ± 2.9 37.1 ± 3.0 0.471 0.222 0.588 0.494
Lower pole of patellaa32.7 ± 2.7 32.9 ± 2.0 32.8 ± 2.2 0.923 0.696 0.787 0.904
Duration of tourniqueta, ‡ (min) 88.6 ± 14.4 61.2 ± 12.6 11.2 ± 1.0 <0.001 <0.001 <0.001 <0.001
Duration of surgerya (min) 88.6 ± 14.5 88.3 ± 14.7 89.2 ± 12.2 0.689 0.521 0.860 0.413
565Arch Orthop Trauma Surg (2014) 134:561–570
1 3
Inflammation markers
The extent of the serum levels of inflammation markers at
different time points was listed as one of the primary out-
comes in the study. As an acute inflammation marker, CRP
increased postoperatively in all patients. The mean levels
peaked on POD2 in all the groups (Fig. 3a). Significant dif-
ferences occurred among the three groups on POD2 and
POD3 (Fig. 3a).
The peak of serum level of IL-6 was noticed on POD1
and POD2. Multiple comparisons showed that the serum
level of IL-6 in Groups B and C was lower than Group A
at every time point except for POD3. On POD3, Group C
had the lowest level while no statistical differences were
detected between Groups A and B (Fig. 3b).
Muscle damage markers
Serum CK level peaked on POD 2. Mean serum CK
level was lower in Group C than in Groups A (p < 0.001,
p = 0.001, p < 0.001, p < 0.001) and B (p < 0.001,
p = 0.01, p = 0.006, p = 0.007) at each time point.
Fig. 1 The postoperative swelling ratio (a) and the postoperative
visual analog scale (VAS) scores (b) showed no significant difference
between the groups. The swelling ratio was defined as the postopera-
tive mean circumference of upper pole of patella and lower pole of
patella divided by the preoperative value. The error bars show the
standard deviation. POD postoperative day
Fig. 2 The perioperative levels of hemoglobin (Hb) (a) and hemato-
crits (Hct) (b). The levels of Hb and Hct decreased after the surgery
and reached the trough on postoperative day 3. The asterisks indi-
cate values that were significantly different between the three groups
immediately after the surgery. Multiple comparison showed that the
mean levels of Hb and Hct were lower in Group C (104.2 ± 10.4 g/L,
31.8 ± 3.2 %) than those in Groups A(111.4 ± 14.4 g/L, p = 0.035;
34.1 ± 4.1 %, p = 0.032) and B (112.8 ± 14.3 g/L, p = 0.013;
34.5 ± 3.7 %, p = 0.011) immediately after the surgery. However,
such differences were not detected on POD1, POD2 and POD3. The
error bars show the standard deviation. Pre OP preoperative; IM PO
immediately after the surgery; POD postoperative day
566 Arch Orthop Trauma Surg (2014) 134:561–570
1 3
However, no significant difference was observed between
Groups A and B at every time point (p = 0.385, p = 0.428,
p = 0.258, p = 0.376) (Fig. 4a).
The mean serum myoglobin level showed similar trend
immediately after the surgery as the CK level: Group
C (41.8 ± 22.5) had a lower level than that in Groups A
(62.9 ± 23.8, p = 0.001) and B (60.0 ± 21.6, p = 0.003)
while no significance was detected between Groups A and
B (p = 0.615). On POD1 the mean serum myoglobin level
was lowest in Group C (vs. Group A, p < 0.001; vs. Group
B, p = 0.002). Yet, on POD 2 and 3 no difference was
observed among the three groups (Fig. 4b).
Functional assessment
The HSS scores were used to assess patients’ knee function.
Baseline demographics were similar in the three groups
(Table 3).At the time of discharge, substantial improvement
was found in all the groups: mean HSS scores increased
to 78.6 (SD 5.8) in Group A, to 79.8 (SD 7.1) in Group B
Table 3 Postoperative parameters
† p stands for p value of Group A vs. B vs. C, p1 stands for p value of Group A vs. B, p2 stands for p value of Group A vs. C, p3 stands for p
value of Group B vs. C
a The values are given as mean ± standard deviation
Variable Group A Group B Group C p†p1
†p2
†p3
†
EBL (mL)a1,221.6 ± 476.6 1,236 ± 443.8 1,278.3 ± 396.7 0.844 0.895 0.579 0.672
Number of patients needing transfusion 3 3 5 0.672 1.0 0.706 0.706
Volume of transfusion (units)a0.30 ± 0.92 0.30 ± 0.95 0.47 ± 1.1 0.756 1.0 0.518 0.518
Length of hospital stay (days)a4.7 ± 0.9 4.6 ± 0.8 4.7 ± 0.7 0.744 0.446 0.760 0.647
HSS scorea56.3 ± 6.8 57.7 ± 6.5 57.0 ± 5.2 0.682 0.383 0.677 0.647
PO HSS scorea78.6 ± 5.8 79.8 ± 7.1 80.0 ± 6.1 0.639 0.453 0.384 0.903
Change of HSS scorea22.3 ± 2.6 22.1 ± 2.2 23.0 ± 2.6 0.301 0.795 0.233 0.147
p<0.001 <0.001 <0.001 / / / /
ROM (°)a97.5 ± 9.5 98.8 ± 9.4 99.8 ± 9.5 0.635 0.588 0.344 0.684
PO ROM (°)a107.3 ± 6.3 109.3 ± 5.4 108.8 ± 4.9 0.350 0.165 0.296 0.727
Change of ROM (°)a9.8 ± 9.4 10.5 ± 9.6 9.0 ± 8.9 0.823 0.782 0.730 0.535
p<0.001 <0.001 <0.001 / / / /
Fig. 3 The postoperative serum levels of CRP (a) and IL-6 (b). The
serum levels of CRP and IL-6 increased after the surgery. There
was no significant difference observed between the groups in terms
of serum CRP levels on POD1 (27.6 ± 14.8 ng/L for Group A,
21.7 ± 12.3 ng/L for Group B and 20.8 ± 11.6 for Group C). Mul-
tiple comparisons showed that the serum level of IL-6 in Groups B
and C was lower than Group A at every time point except for POD3.
On POD3, Group C (23.0 ± 23.4) had the lowest level while no sta-
tistical differences were detected between Groups A (43.4 ± 32.6,
p2 = 0.005) and B (29.5 ± 24.6, p3 = 0.356). The error bars show
the standard deviation. Pre OP preoperative; IM PO immediately
after the surgery; POD postoperative day. p2 stands for p value of
Group A vs. C, p3 stands for p value of Group B vs. C
567Arch Orthop Trauma Surg (2014) 134:561–570
1 3
and to 80.0 (SD 6.1) in Group C. However, no difference
approached significantly between groups in the HSS score
at discharge, though Group C showed a litter bit higher.
Preoperatively, mean ROM was 97.5° (SD 9.5) in Group
A, 98.9° (SD 9.4) in Group B and 99.8° (SD 9.5) in Group
C. No significant difference was detected between the
groups in ROM at the time of discharge (Table 3).
Complications
Doppler ultrasound was used to diagnose the deep vein
thrombosis on POD1, POD3 and the time of discharge. No
DVT was found in any patients of the three groups. One
patient in Group B suffered from superficial infection,
cured with dressing changes and antibiotics. At the time of
discharge this patient was satisfied with the knee function.
No other wound complications such as persistent drainage
or dehiscence occurred in the groups during hospitalization.
Discussion
The most important finding of the present study was that
patients undergoing TKA with full time tourniquet had a
less intraoperative blood loss and might cause more soft
tissue injury. However, the difference of inflammation and
muscle damage markers between groups did not reflect in
postoperative functional outcomes. There were no differ-
ences between the groups in terms of postoperative swell-
ing, pain, or length of hospital stay.
Which kind of tourniquet application strategies should
be taken in TKA is still a controversial topic debated in the
literatures [5, 10–12]. Many studies showed that though
using tourniquet during the whole surgery process could
reduce intraoperative blood loss, there is still a substantial
blood loss, so-called hidden blood loss after the surgery
[19–21]. Further studies showed that reperfusion of tissues
after tourniquet deflation increased hidden blood loss [22,
23]. A recent meta-analysis showed that the time of tourni-
quet application may have major impact on the hidden and
total blood loss [24]. On the other hand, several other stud-
ies reported that perioperative blood loss was significantly
greater among patients undergoing total knee arthroplasty
without cement in comparison with those undergoing fixa-
tion with cement [25–28]. Based on this phenomenon,
Loktke et al. [29] pointed out the continuing bleeding from
the cut cancellous bone was a major source of blood loss.
Using the tamponade effect and local temperatures above
48 °C caused by polymerization of cement may stanch
bleeding from the cut cancellous bone [30, 31]. Thus, some
surgeons figured out that deflating tourniquet after the hard-
ening of the cement or using tourniquet only during cement
fixation might reduce the total blood loss associated with
total knee arthroplasty. A few studies had been performed
to compare the three tourniquet application strategies [10–
13]. Mittal et al. [13] found that the total blood loss and
risk of transfusion increased with the timing of tourniquet
application. Kvederas et al. [10] discovered that tourniquet
inflated before incision and deflated after the hardening of
the cement tended to be associated with lower estimated
blood loss. In the present study, we found the significantly
lower Hb and Hct levels immediately after the surgery
compared with Groups A and B. The Hb and Hct levels
immediately after the surgery reflected intraoperative blood
loss and this finding also coincided with previous studies
[10–13], while we did not find any significant difference
between groups in terms of estimated blood loss, number
of patients needing transfusion and number of transfusions
Fig. 4 The postoperative serum levels of CK (a) and myoglobin (b). The error bars show the standard deviation. Pre OP preoperative; IM PO
immediately after the surgery; POD postoperative day
568 Arch Orthop Trauma Surg (2014) 134:561–570
1 3
per patient. Such findings might attribute to several reasons
as follows: firstly the benefit of less intraoperative blood
loss might be balanced by more hidden, postsurgical blood
loss, for the surgeons were unable to identify and cauter-
ize bleeding vessels in comparison with other groups; sec-
ondly, the minimally invasive techniques whose core idea
was to protect soft tissues and minimize the surgical injury
to the patients who were taken in our surgical practice, so
the totally estimated blood loss in three groups was lower
than that in other studies and this might cause the differ-
ence too small to be detected.
Serum levels of CRP and IL-6 were used to evaluate sur-
gical injury to the soft tissue. CRP is thought to be an acute
inflammation marker responding rapidly to the surgery. In
our study, the CRP level rose after the surgery and reached
the peak on POD2. At each time point, the CRP levels of
patients in Group A were higher than those in the other
groups. It showed that the level of IL-6 consistently ele-
vated in patients with major trauma, but its role beyond the
induction of acute-phrase proteins and pro-coagulation is
not well defined [32]. Damas et al. [33] found that there was
a correlation between the degree of injury and the serum
level of IL-6, and the increases in IL-6 during the first few
days after injury followed by a decrease in levels reflect the
amount of tissue damage followed by an uneventful course.
In our study, we also checked the serum level of IL-6 in
three groups and the change in trend was quite consistent
with the previous studies. Group A had higher serum levels
of IL-6 than Groups B and C at each time point. There are
two major coming sources of inflammation after the TKA:
the inflammation caused by surgical performance and the
reperfusion [1]. In our study, the surgery was performed
by one surgical team using the same approach in the same
operative room, which kept a well homogeneity between
patients and minimized the difference caused by the surgi-
cal performance. Reperfusion is thought to be followed by
an inflammatory response in the ischaemic tissue that can
aggravate local injury [34–36]. We believe the differences
in changed serum levels of inflammation markers were
caused by the reperfusion. With the growth timing of tour-
niquet application, the inflammation response increased. So
the findings might reflect that Group A had more invisible
injury to soft tissue than the other groups.
In terms of the serum creatine kinase levels, Group C had
much lower levels than the other two groups at each time
point, despite similar baselines in patient demographics.
A significant difference also existed between Group C and
the other groups in terms of serum myoglobin levels imme-
diately after the surgery and on POD1. These outcomes
indicated that compared with using tourniquet only during
cementing, full time using tourniquet or inflated the tourni-
quet before prosthesis placement and deflated after the hard-
ening of the cement could cause more muscle damage.
The differences of inflammation and muscle damage
markers did not reflect in functional assessment. No sig-
nificant differences were found in functional assessments
between the groups. Previous clinical and experimental
studies demonstrated that the muscle damage caused by
tourniquet was attributable to the amount of pressure and
the ischemia duration [37–40]. If we set the thigh tour-
niquet cuff pressure less than 225 mmHg and control
the ischemia duration less than 120 min the difference
between soft tissue markers might not reflect in functional
assessments.
Before the trial, we had the impression of reduction in
patients’ complaints about the thigh pain after decreas-
ing the timing of tourniquet application. However, we did
not find any differences on postoperative VAS pain score
between the groups, which may reflect that the difference
between the groups was so small that most patients’ pain
can be controlled by standard postoperative analgesics.
Tourniquet application was thought to increase the risk
of potential adverse events such as nerve compression,
thromboembolism and infection [3, 6–9]. In our study, one
of the patients experienced adverse events. However, no
DVT was detected during the in-patient period. It might lie
in the following two main reasons: Firstly, early rehabili-
tation activities which can reduce the risk of thromboem-
bolism were performed under instruction of physical thera-
pists; secondly, standard chemical prophylaxis was applied
to every patient postoperatively. One patient in Group B
suffered from superficial infection, which may have been
caused by a relative longer operative time (105 min). For-
tunately, other complications such as nerve palsy, vascular
injury, altered patellofemoral tracking, persistent drainage
and dehiscence did not occur in our study. Complications
associated with tourniquet application are believed to have
close correlation with the amount of pressure and the dura-
tion of the tourniquet application [37–40]. Jorgensen [39]
reported that controlling the duration of tourniquet within
150 min can reduce the risk of complications significantly.
A recent study showed that patients with a cuff pressure of
≤225 mmHg could lower down the risk of perioperative
complications effectively [40]. In our study, the duration of
surgery was controlled less than 110 min and the cuff pres-
sure was 100 mmHg above systolic blood pressure, which
may account for why few early surgical complications were
found.
The strengths of the current study are as follows: it was
a prospective study; the baselines of patient cohorts were
similar between groups; one surgical team using the same
techniques in the same operative room.
However, the study had several limitations. First, only
the needed sample number was calculated according to
serum creatine kinase level and blood loss. To detect signif-
icant differences in assessment outcomes between groups,
569Arch Orthop Trauma Surg (2014) 134:561–570
1 3
a larger sample size is needed. Furthermore, this study was
not randomized. Third, this is a preliminary report of the
clinical trial and long-term follow-up should be taken to
compare the long-term results between groups.
Conclusion
In this prospective, non-randomized clinical trials, we con-
cluded that using a tourniquet full time in minimally inva-
sive total knee arthroplasty causes less intraoperative blood
loss and more excessive inflammation and muscle damage.
However, the advantage of part-time using tourniquet did
not show in early functional outcomes.
Acknowledgments All authors have no financial or personal rela-
tionships with other people or organizations that could inappropri-
ately influence this work.
Conflict of interest None.
References
1. Huang Z, Shen B, Ma J, Yang J, Zhou Z, Kang P, Pei F (2012)
Mini-midvastus versus medial parapatellar approach in TKA:
muscle damage and inflammation markers. Orthopedics
35(7):e1038–e1045. doi:10.3928/01477447-20120621-17
2. Harvey EJ, Leclerc J, Brooks CE, Burke DL (1997) Effect of
tourniquet use on blood loss and incidence of deep vein thrombo-
sis in total knee arthroplasty. J Arthroplasty 12(3):291–296
3. Aglietti P, Baldini A, Vena LM, Abbate R, Fedi S, Falciani M
(2000) Effect of tourniquet use on activation of coagulation in
total knee replacement. Clin Orthop Relat Res 371:169–177
4. Alcelik I, Pollock RD, Sukeik M, Bettany-Saltikov J, Armstrong
PM, Fismer P (2012) A comparison of outcomes with and with-
out a tourniquet in total knee arthroplasty: a systematic review
and meta-analysis of randomized controlled trials. J Arthroplasty
27(3):331–340. doi:10.1016/j.arth.2011.04.046
5. Rama KR, Apsingi S, Poovali S, Jetti A (2007) Timing of tourni-
quet release in knee arthroplasty. Meta-analysis of randomized,
controlled trials. J Bone Joint Surg Am 89(4):699–705
6. Silver R, de la Garza J, Rang M, Koreska J (1986) Limb swelling
after release of a tourniquet. Clin Orthop Relat Res 206:86–89
7. Clarke MT, Longstaff L, Edwards D, Rushton N (2001) Tourni-
quet-induced wound hypoxia after total knee replacement. J Bone
Joint Surg Br 83(1):40–44
8. Horlocker TT, Hebl JR, Gali B, Jankowski CJ, Burkle CM, Berry
DJ, Zepeda FA, Stevens SR, Schroeder DR (2006) Anesthetic,
patient, and surgical risk factors for neurologic complications
after prolonged total tourniquet time during total knee arthro-
plasty. Anesth Analg 102(3):950–955
9. Irvine GB, Chan RN (1986) Arterial calcification and tourniquets.
Lancet 2(8517):1217
10. Kvederas G, Porvaneckas N, Andrijauskas A, Svensen CH, Ivask-
evicius J, Mazunaitis U, Andrijauskas P (2012) A randomized
double-blind clinical trial of tourniquet application strategies for
total knee arthroplasty. Knee Surg Sports Traumatol Arthrosc.
doi:10.1007/s00167-012-2221-1
11. Ledin H, Aspenberg P, Good L (2012) Tourniquet use in total
knee replacement does not improve fixation, but appears to
reduce final range of motion. Acta Orthop 83(5):499–503. doi:10.
3109/17453674.2012.727078
12. Steffin B, Green-Riviere E, Giori NJ (2009) Timing of tour-
niquet release in total knee arthroplasty when using a post-
operative blood salvage drain. J Arthroplasty 24(4):539–542.
doi:10.1016/j.arth.2008.01.302
13. Mittal R, Ko V, Adie S, Naylor J, Dave J et al (2012) Tourniquet
application only during cement fixation in total knee arthro-
plasty: a double-blind, randomized controlled trial. ANZ J Surg
82(6):428–433. doi:10.1111/j.1445-2197.2012.06083.x
14. Barwell J, Anderson G, Hassan A, Rawlings I (1997) The effects
of early tourniquet release during total knee arthroplasty: a pro-
spective randomized double-blind study. J Bone Joint Surg Br
79(2):265–268
15. Demey G, Servien E, Pinaroli A, Lustig S, Ait Si Selmi T et al
(2010) The influence of femoral cementing on perioperative
blood loss in total knee arthroplasty: a prospective randomized
study. J Bone Joint Surg Am 92(3):536–541. doi:10.2106/JBJ
S.H.01159
16. Andres BM, Taub DD, Gurkan I, Wenz JF (2003) Postoperative
fever after total knee arthroplasty: the role of cytokines. Clin
Orthop Relat Res 415:221–231
17. White J, Kelly M, Dunsmuir R (1998) C-reactive protein level
after total hip and total knee replacement. J Bone Joint Surg Br
80(5):909–911
18. Hahn RG, Andrijauskas A, Drobin D, Svensen C, Ivaskevicius J
(2008) A volume loading test for the detection of hypovolemia
and dehydration. Medicina (Kaunas) 44(12):953–959
19. Vandenbussche E, Duranthon LD, Couturier M, Pidhorz L,
Augereau B (2002) The effect of tourniquet use in total knee
arthroplasty. Int Orthop 26(5):306–309
20. Li B, Wen Y, Wu H, Qian Q, Lin X et al (2009) The effect of
tourniquet use on hidden blood loss in total knee arthroplasty. Int
Orthop 33(5):1263–1268. doi:10.1007/s00264-008-0647-3
21. Sehat KR, Evans RL, Newman JH (2004) Hidden blood loss fol-
lowing hip and knee arthroplasty. Correct management of blood
loss should take hidden loss into account. J Bone Joint Surg Br
86(4):561–565
22. Matziolis D, Perka C, Hube R, Matziolis G (2011) Influence
of tourniquet ischemia on perioperative blood loss after total
knee arthroplasty. Orthopade 40(2):178–182. doi:10.1007/
s00132-010-1727-9
23. Sehat KR, Evans R, Newman JH (2000) How much blood is
really lost in total knee arthroplasty? Correct blood loss manage-
ment should take hidden loss into account. Knee 7(3):151–155
24. Tai TW, Lin CJ, Jou IM, Chang CW, Lai KA et al (2011) Tour-
niquet use in total knee arthroplasty: a meta-analysis. Knee
Surg Sports Traumatol Arthrosc 19(7):1121–1130. doi:10.1007/
s00167-010-1342-7
25. Hays MB, Mayfield JF (1988) Total blood loss in major joint
arthroplasty. A comparison of cemented and noncemented hip
and knee operations. J Arthroplasty 3 suppl:S47–S49
26. Semkiw LB, Schurman DJ, Goodman SB, Woolson ST (1989)
Postoperative blood salvage using the cell saver after total joint
arthroplasty. J Bone Joint Surg Am 71(6):823–827
27. Mylod AG Jr, France MP, Muser DE, Parsons JR (1990) Periop-
erative blood loss associated with total knee arthroplasty. A com-
parison of procedures performed with and without cementing. J
Bone Joint Surg Am 72(7):1010–1012
28. Porteous AJ, Bartlett RJ (2003) Post-operative drainage after
cemented, hybrid and uncemented total knee replacement. Knee
10(4):371–374
29. Lotke PA, Faralli VJ, Orenstein EM, Ecker ML (1991)
Blood loss after total knee replacement. Effects of tourniquet
release and continuous passive motion. J Bone Joint Surg Am
73(7):1037–1040
570 Arch Orthop Trauma Surg (2014) 134:561–570
1 3
30. Charnley J (1970) The reaction of bone to self-curing acrylic
cement. A long-term histological study in man. J Bone Joint Surg
Br 52(2):340–353
31. Jefferiss CD, Lee AJ, Ling RS (1975) Thermal aspects of
self-curing polymethylmethacrylate. J Bone Joint Surg Br
57(4):511–518
32. Clementsen T, Reikeras O (2008) Cytokine patterns after tourni-
quet-induced skeletal muscle ischaemia reperfusion in total knee
replacement. Scand J Clin Lab Invest 68(2):154–159
33. Damas P, Ledoux D, Nys M, Vrindts Y, De Groote D et al (1992)
Cytokine serum level during severe sepsis in human IL-6 as a
marker of severity. Ann Surg 215(4):356–362
34. Blaisdell FW (2002) The pathophysiology of skeletal muscle
ischemia and the reperfusion syndrome: a review. Cardiovasc
Surg 10(6):620–630
35. Ayala A, Wang P, Ba ZF, Perrin MM, Ertel W et al (1991) Differ-
ential alterations in plasma IL-6 and TNF levels after trauma and
hemorrhage. Am J Physiol 260:167–171
36. Roumen RM, Hendriks T, van der Ven-Jongekrijg J, Nieuwenhui-
jzen GA, Sauerwein RW et al (1993) Cytokine patterns in patients
after major vascular surgery, hemorrhagic shock, and severe blunt
trauma. Relation with subsequent adult respiratory distress syn-
drome and multiple organ failure. Ann Surg 218(6):769–776
37. Gersoff WK, Ruwe P, Jokl P, Panjabi M (1989) The effect
of tourniquet pressure on muscle function. Am J Sports Med
17(1):123–127
38. Sapega AA, Heppenstall RB, Chance B, Park YS, Sokolow D
(1985) Optimizing tourniquet application and release times in
extremity surgery. A biochemical and ultrastructural study. J
Bone Joint Surg Am 67(2):303–314
39. Jorgensen HR (1987) Myoglobin release after tourniquet
ischemia. Acta Orthop Scand 58(5):554–556
40. Olivecrona C, Ponzer S, Hamberg P, Blomfeldt R (2012) Lower
tourniquet cuff pressure reduces postoperative wound complica-
tions after total knee arthroplasty: a randomized controlled study
of 164 patients. J Bone Joint Surg Am 94(24):2216–2221. doi:10.
2106/JBJS.K.01492
