YIJOM-3884; No of Pages 7

Int. J. Oral Maxillofac. Surg. 2018; xxx: xxx–xxx

https://doi.org/10.1016/j.ijom.2018.02.013, available online at https://www.sciencedirect.com

Randomised Controlled Trial

Oral Surgery

Intra-alveolar epsilon- R. V. da Silva1, T. B. Gadelha2,

R. R. Luiz3, S. R. Torres4
1
Medical Clinic Programme of the Faculty of

aminocaproic acid for the Medicine, Universidade Federal do Rio de

Janeiro, Rio de Janeiro, Brazil; 2Haematology
Department, Universidade Federal do Rio de
Janeiro, Rio de Janeiro, Brazil; 3Institute for

control of post-extraction Studies in Public Health (IESC), Universidade

Federal do Rio de Janeiro, Rio de Janeiro,
Brazil; 4Department of Oral Pathology and
Diagnostics, Faculty of Dentistry,

bleeding in anticoagulated Universidade Federal do Rio de Janeiro, Rio

de Janeiro, Brazil

patients: randomized clinical

trial
R. V. da Silva, T. B. Gadelha, R. R. Luiz, S. R. Torres: Intra-alveolar epsilon-
aminocaproic acid for the control of post-extraction bleeding in anticoagulated
patients: randomized clinical trial. Int. J. Oral Maxillofac. Surg. 2018; xxx: xxx–xxx.
ã 2018 International Association of Oral and Maxillofacial Surgeons. Published by
Elsevier Ltd. All rights reserved.

Abstract. The aim of this study was to compare the effectiveness of the intra-alveolar
administration of epsilon-aminocaproic acid (EACA) and daily gentle rinsing with
EACA mouthwash with that of routine postoperative procedures for the control of
bleeding after tooth extraction in anticoagulated patients. A randomized clinical
trial was conducted involving 52 patients submitted to 140 tooth extractions,
assigned randomly to two groups. The intervention group was treated with intra-
alveolar administration of EACA immediately after surgery and gentle rinsing with
EACA mouthwash during the postoperative period. The control group received
routine postoperative recommendations. A single episode of immediate bleeding
occurred in the intervention group. Late bleeding episodes occurred in 23
procedures (16.4%): 11 (15.7%) in the intervention group and 12 (17.1%) in the
control group. Among the patients with late bleeding, 18 (78.3%) events were
classified as moderate and were controlled by the patient applying pressure to a
gauze pack placed over the extraction socket. The remaining five cases (21.7%)
required re-intervention. No statistically significant difference in the frequency of Key words: epsilon-aminocaproic acid; antic-
postoperative bleeding was observed between the groups. Thus, routine measures oagulated patients; tooth extractions.
were as effective for the control of bleeding after simple tooth extractions in
anticoagulated patients as the topical administration of EACA. Accepted for publication 23 February 2018

0901-5027/000001+07 ã 2018 International Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: da RV, et al. Intra-alveolar epsilon-aminocaproic acid for the control of post-extraction bleeding in
anticoagulated patients: randomized clinical trial, Int J Oral Maxillofac Surg (2018), https://doi.org/10.1016/j.ijom.2018.02.013
YIJOM-3884; No of Pages 7
2 da Silva et al.
Among the candidates for oral surgical
procedures, anticoagulated patients are at
greater risk of intraoperative and postoper-
ative bleeding. However, although the
maintenance of anticoagulation therapy
poses a high risk of bleeding, its discontin-
uation increases the risk of thromboembolic
complications1. Therefore, the manage-
ment of these patients should be based on
the probability of bleeding after a procedure
and on the risk of thromboembolic compli-
cations, particularly among high-risk
patients (e.g., those with prosthetic heart
valves or high venous thrombotic risk)2,3.
Warfarin sodium is an oral anticoagu-
lant used widely to reduce the risk of
thrombotic events. This drug is a vitamin
K antagonist and, therefore, interferes
with haemostasis by inhibiting factors II,
VII, IX, and X and anticoagulation pro-
teins C and S. Since the effect of warfarin
sodium is influenced by the intake of
certain foods and by other drugs (mainly
non-steroidal anti-inflammatory drugs,
antibiotics, and antifungals), it is some-
times necessary to adjust the dose to en-
sure proper anticoagulation. Routine
laboratory follow-up to assess the pro-
thrombin time (PT) and the international
normalized ratio (INR) should also be
conducted. Most medical conditions
should have an INR between 2.0 and 3.0
for effective anticoagulation4,5.
Recent studies have demonstrated a low
risk of bleeding in warfarin-treated
patients with INR in the therapeutic range
undergoing certain dental procedures, in-
cluding simple extractions (up to three
teeth), small biopsies, crown lengthening,
and root planing and scaling. When bleed-
ing occurs, it is easily controlled with local
haemostatic measures, so it is important
that a minimally invasive surgical tech-
nique is chosen and that postoperative care
is optimized6–8. Several studies advocate
this practice and claim that an INR of up to
4.0 is sufficiently safe to treat these
patients9–11. In patients with an INR
<3.0, bleeding can be controlled by mere-
ly applying local pressure, with no need
for additional haemostatic measures12.
In the case of bleeding complications,
different measures are proposed to stop the
bleeding, including fibrin glue, oxidized
cellulose, haemostatic sponges, sutures,
and compression of the surgical wound
with gauze with or without the addition of
an antifibrinolytic agent (e.g., tranexamic
acid (TXA) or epsilon-aminocaproic acid
(EACA))7,12–15.
Haemostatic mouthwashes have been
proven efficacious in the control of post-
operative bleeding in anticoagulated
patients, even among those for whom
Please cite this article in press as: da RV, et al. Intra-alveolar epsilon-aminocaproic acid for the control of post-extraction bleeding in
anticoagulated patients: randomized clinical trial, Int J Oral Maxillofac Surg (2018), https://doi.org/10.1016/j.ijom.2018.02.013
anticoagulant therapy was not modified
or discontinued9,16. TXA is the most wide-
ly used antifibrinolytic agent and also the
one whose use as a mouthwash has been
investigated most extensively. While sev-
eral treatment regimens and different con-
centrations have been proposed, the use of
this drug within the first 2 days after sur-
gery has proven efficient in the control of
bleeding17. There is little evidence of the
use of EACA as a local haemostatic agent
in these patients; however, it could be
administered as an antifibrinolytic agent
in countries where TXA is not readily
available14,18.
Antifibrinolytics act by competitively
inhibiting the activation of plasminogen
and by inhibiting plasmin, which reduces
the fibrinolytic activity of the latter, in-
creasing the blood clotting efficiency. In
addition to mouthwashes, crushed tablets
applied as a paste (mixed with sterile
saline or anaesthetic solution) can be used
directly on the surgical wound19,20.
Antifibrinolytic agents, such as EACA
and TXA, have been used either systemi-
cally or locally for the control of bleeding
in patients with haematological diseases
submitted to tooth extractions21,22. These
drugs reduce the fibrinolytic activity in the
saliva, which allows improved clot stabil-
ity in patients with coagulation disorders.
The local administration of antifibrinoly-
tics, especially TXA, has been proposed
for the prevention and treatment of bleed-
ing complications in anticoagulated
patients23. This drug remains at high con-
centrations in the saliva and its action lasts
up to 8 hours. On the other hand, the serum
concentration of TXA is nearly undetect-
able when this agent is used as a mouth-
wash, thus it poses a negligible risk of
systemic effects. When used as a mouth-
wash in the postoperative period for antic-
oagulated patients submitted to oral
surgeries, severe bleeding complications
are not observed; this includes patients
who have maintained oral therapy and
in whom the INR is within the therapeutic
range17,23. The same benefits are reported
when a gauze pad soaked in TXA is
applied firmly over the dental alveolus15
or when an intra-alveolar gel is used24.
The use of oral and topical EACA is
recommended for tooth extractions in
patients with bleeding disorders20–22. Al-
though the use of topical EACA has been
described in anticoagulated patients sub-
mitted to tooth extractions, no methodo-
logically robust clinical trials have
assessed its efficacy as a local haemostatic
agent in these patients14. The aim of this
study was to compare the effectiveness of
intra-alveolar EACA combined with gen-
tle daily EACA mouthwashes with the
effectiveness of routine postoperative pro-
cedures for the control of bleeding after
tooth extraction in anticoagulated patients.
Patients and methods
Patients
The patients included in this study were on
regular anticoagulant therapy and had
been referred from the anticoagulation
clinic to the outpatient clinic of the oral
health programme for tooth extraction.
Patients older than 18 years of age treated
regularly with warfarin sodium combined
or not with antiplatelet agents, referred for
simple tooth extractions and with an INR
between 2.0 and 4.0 on the day of surgery,
were included in the study. Patients with a
platelet count <50  109/l, pregnant or
lactating women, patients with sensitivity
to EACA components, patients with he-
reditary coagulation disorders, and
patients presenting ankylosed, partially
embedded or embedded teeth, or an ab-
normal root morphology that could com-
promise the procedure were excluded.
Fifty-two patients were enrolled, of
whom 29 underwent more than one surgi-
cal procedure. A total of 140 simple tooth
extractions were performed. Patients were
allocated randomly to one of two groups,
with patients from both groups receiving
routine postoperative recommendations
and those in the intervention group receiv-
ing additional intra-alveolar EACA. Each
tooth extraction was performed at a dif-
ferent appointment, and patients could be
allocated to a different group at each
intervention.
The extracted teeth were those with
pulpal or periodontal involvement that
could not be spared, either because of
the tooth structure or because the patient
could not afford to pay for the conserva-
tive dental treatment. Those patients who
met the inclusion criteria were assigned
randomly to one of the two groups in
accordance with the order established by
Random Allocation Software 1.0 (Isfahan,
Iran, 2004). The patients assigned to the
intervention group received intra-alveolar
EACA (Ipsilon; Nikkho do Brasil, Rio de
Janeiro, RJ, Brazil) post-extraction and
used this drug as a mouthwash to gently
rinse the mouth during the postoperative
period, as well as receiving routine post-
operative recommendations. The control
group received only routine postoperative
recommendations.
All patients agreed to participate in the
study and signed a free informed consent
agreement. This study was approved by
YIJOM-3884; No of Pages 7
the Research Ethics Committee of Clem-
entino Fraga Filho Teaching Hospital.
All patients underwent anamnesis and
clinical examination, and after being select-
ed, a panoramic radiograph was taken. One
week before the surgical procedure, whole-
mouth supragingival scaling was per-
formed using an ultrasound device (Cavi-
tron; Dentsply, Pirassununga, São Paulo,
Brazil) for control of gingival inflamma-
tion. Instructions for oral hygiene were
given and the surgical procedure was sched-
uled. On the day of tooth extraction, the
patients were referred to the laboratory of
haematology for the measurement of PT,
INR, activated partial thromboplastin time
(aPTT), and platelet count. This was done to
confirm that the INR met the study require-
ment (between 2.0 and 4.0) and to rule out
blood disorders that could be a contraindi-
cation to the procedure. If the INR was <2.0
or >4.0, the patient was sent to the antic-
oagulation clinic for reassessment by the
medical staff and for medication dose ad-
justment. Patients with heart diseases pre-
disposing to bacterial endocarditis received
antibiotic prophylaxis in accordance with
the 2007 American Heart Association
(AHA) recommendations.
Tooth extractions were always per-
formed by the same surgeon, who used
conventional and minimally invasive ex-
traction techniques – forceps and extrac-
tors – under local anaesthesia (2%
lidocaine and epinephrine 1:100,000),
avoiding tissue laceration and excessive
bone removal. A single extraction was
performed per surgical session. Immedi-
ately after tooth extraction, bidigital alve-
olar compression with sterile gauze was
performed for 5 minutes. The patients
were aware of the group they were
assigned to and so was the surgeon. After
tooth extraction, the patients assigned to
the intervention group received intra-alve-
olar EACA (Ipsilon; Nikkho do Brasil),
prepared from a 500-mg tablet that had
been crushed and mixed with 0.9% saline
solution, following which the surgical
wound was sutured with non-absorbable
3–0 silk thread. These patients were then
instructed to crush a 500-mg EACA tablet,
dilute it in two tablespoons of filtered
water, and rinse their mouth gently with
the solution three times a day for the first
2 days after the extraction. The patients in
the control group, after digital alveolar
compression, just had the surgical wound
sutured. All patients were prescribed para-
cetamol (750 mg, one tablet orally every
6 h) for postoperative pain control. They
also received postoperative recommenda-
tions for maintenance of the integrity of
the clot.
Please cite this article in press as: da RV, et al. Intra-alveolar epsilon-aminocaproic acid for the control of post-extraction bleeding in
anticoagulated patients: randomized clinical trial, Int J Oral Maxillofac Surg (2018), https://doi.org/10.1016/j.ijom.2018.02.013
Intra-alveolar EACA for anticoagulated patients 3
The instruments, equipment, and Results
reagents necessary for the laboratory tests
Between July 2013 and January 2016, 52
were purchased using research funds.
patients on anticoagulant treatment under-
went 140 simple tooth extractions; 28
Assessment of bleeding (53.8%) were female and 24 (46.2%) were
male, and they ranged in age from 40 to 93
Immediate bleeding was assessed by the
years (mean 61.9  12.1 years). Twenty-
surgeon at two time-points after the pro-
seven (51.9%) of the patients were white;
cedure: T1, immediately after tooth ex-
seven (13.5%) were illiterate. Indications
traction; T2, 20 min after tooth extraction.
for anticoagulant therapy, the type of an-
This was classified as no bleeding, mild
ticoagulant used, the medications taken by
bleeding (presence of blood in the saliva),
the patients, and socio-demographic data
moderate bleeding (controlled by gauze
are displayed in Table 1.
compression), or severe bleeding (when
Of the 140 extractions, 70 (50%) were
surgical re-intervention and/or hospital
performed in the intervention group and
admission was necessary). The patients
70 (50%) in the control group. The
were discharged only if no bleeding or
extracted teeth were grouped as follows:
mild to moderate bleeding was observed.
maxillary molars, maxillary non-molars,
Late bleeding was recorded every day
mandibular molars, and mandibular non-
by the patients on a specific chart for
molars. The frequency of extraction of
7 days postoperatively and classified as
teeth in these groups is shown in Table 2.
no bleeding, mild bleeding (presence of
blood in the saliva), moderate bleeding
(controlled by gauze compression), or se- Table 1. Baseline clinical characteristics of
vere bleeding (when surgical re-interven- the 52 study patients.
tion and/or hospital admission was Variables Number (%)
necessary). Instructions on how to fill Indications for anticoagulant therapy
out the chart were given by a second Atrial fibrillation 17 (32.7)
researcher. Upon discharge, the patients Stroke 13 (25.0)
were instructed to contact the researchers Acute myocardial infarction 12 (23.1)
if any complications arose. In the case of Venous thromboembolism 10 (19.2)
haemorrhage, local haemostatic measures Prosthetic heart valve 8 (15.4)
were adopted after professional reassess- Cardiac valve disease 8 (15.4)
ment. Stent 3 (5.8)
Pacemaker 2 (3.8)
Only those events classified as moderate
Pulmonary embolism 2 (3.8)
(requiring pressure with a gauze pack) or Revascularized patients 2 (3.8)
severe (requiring re-intervention) were Other conditions 8 (15.4)
considered to be bleeding complications, Underlying diseases
in the assessment of both immediate Hypertension 48 (92.3)
bleeding and late bleeding. Diabetes 10 (19.2)
The patients returned 7 days after the Neurological disease 5 (9.6)
surgery for suture removal. On this occa- Respiratory disease 5 (9.6)
sion, they handed in the late bleeding Endocrine disease 4 (7.7)
assessment chart, which was checked by Collagen disease 3 (5.8)
Gastrointestinal disease 3 (5.8)
an investigator who was blinded to the
Liver disease 3 (5.8)
group allocation. Allergy 3 (5.8)
Kidney disease 2 (3.8)
Medication
Statistical analysis
Anticoagulants
P-values were calculated for the compar- Warfarin 52 (100.0)
isons between the groups, taking into ac- Warfarin + antiplatelet 4 (7.7)
count the internal dependency of therapy
observations of each patient using gener- Others drugs
Antihypertensive drugs 48 (92.3)
alized estimating equations (GEE). De-
Antiarrhythmic agents 40 (76.9)
scriptive analyses were performed using Diuretics 34 (65.4)
appropriate tables and graphs according to Hypoglycaemics 9 (17.3)
the measurement scale. Mean, median, Hormones 4 (7.7)
and standard deviation values were calcu- Psychotropics 4 (7.7)
lated for the numerical variables. The Anti-inflammatory, 2 (3.8)
logistic regression model was adjusted steroidal
using GEE for the assessment of late Antineoplastic drugs 2 (3.8)
bleeding, with estimation of the odds ratio Antihistamines 1 (1.9)
(OR). The significance level was set at Others drugs not listed 38 (73.1)
<5%.
YIJOM-3884; No of Pages 7

4 da Silva et al.

Table 2. Distribution of the type and number of extracted teeth in each group.
EACA group
Variables (n = 70), n (%)
Group of teeth
Maxillary molars 11 (15.7)
Maxillary non-molars 28 (40.0)
Mandibular molars 10 (14.3)
Mandibular non-molars 21 (30.0)
Number of
Number of tooth patients EACA
extractions group
1 17
2 9
3 3
4 2
5 1
6 1
7 1
EACA, epsilon-aminocaproic acid.
This table also shows the number of
extracted teeth according to the number
of patients for each group. The same pa-
tient could undergo procedures in both
groups; for instance, one patient had 14
teeth extracted, seven in each group (in-
tervention and control).
Table 3 shows the PT, INR, aPTT, and
platelet count measured on the day of
tooth extraction for each group. The aPTT
could not be estimated in seven cases, two
in the control group and five in the inter-
vention group, due to technical problems.
The same applies to the platelet count:
results were not included for five cases
in the control group and 19 cases in the
intervention group. In these cases, the
inclusion of the patients in the study
was based on previous tests performed
up to 30 days prior to the surgical proce-
dure. The intervention group had higher
INR, PT, and platelet counts than the
control group, and there was a statistically
significant difference in PT and aPTT
levels between the groups.
Only one episode of immediate bleed-
ing was observed, which was classified by
the surgeon as moderate. This episode
occurred in the intervention group at T1,
Control group Total
(n = 70), n (%) (n = 140), n (%)
20 (28.6) 31 (22.1)
17 (24.3) 45 (32.1)
15 (21.4) 25 (17.9)
18 (25.7) 39 (27.9)
Number of
patients Control
group
17
9 Discussion
4
4
0 taining anticoagulant therapy in patients
0 undergoing
1 extraction1,2,9,17,25–31. In the present
i.e., immediately after the extraction of a
mandibular left first molar, accounting for
1.4% of the procedures performed in this
group. This patient was being treated with
warfarin and had an INR of 2.19 at the
time of surgery. No event was observed at
T2.
Late bleeding episodes occurred in 23
procedures (16.4%): 11 (15.7%) in the
intervention group and 12 (17.1%) in
the control group. Among the patients
with late bleeding, 18 (78.3%) events were
classified as moderate and were controlled
by the patients with a pressure gauze pack
placed over the socket. These events were
more prevalent on the day of surgery,
followed by day 2. The remaining five
(21.7%) events – two in the intervention
group and three in the control group –
required re-intervention. No patient had
to be hospitalized, nor did any patient
require the administration of vitamin K
or prothrombin complex concentrate.
There were no adverse reactions associat-
ed with EACA. Table 4 shows the late
bleeding episodes in greater detail.
The logistic regression model was ad-
justed for INR and aPTT levels and for the
group of teeth using GEE and calculating
the OR. While there was a statistically
significant difference in PT levels between
the groups, GEE and OR were adjusted
based on INR, since PT levels are
expressed by INR. No statistically signifi-
cant difference in late bleeding was ob-
served between the groups (Table 5).
All patients with severe bleeding were
being treated with warfarin only. The five
episodes of severe bleeding are described
in detail in Table 6.
Currently, several authors advocate main-
simple tooth
study, there was only one (0.7%) immedi-
ate bleeding episode and 23 (16.4%) late
bleeding episodes. The low prevalence of
bleeding complications found in this study
is consistent with the findings of a system-
atic review on the use of haemostatic
mouthwashes in anticoagulated patients
submitted to tooth extractions16.
In cases of bleeding complications, lo-
cal haemostasis is usually effective, and
hospitalization or more invasive interven-
tions are usually not necessary19,30. For
this reason, the efficacy of several haemo-
static measures has been investigated in
recent decades8,18,32–36, including the use
of topical antifibrinolytics, especially
TXA13,28,33. When anticoagulant therapy
is maintained unchanged and INR is with-
in the therapeutic range, a local haemo-
static agent should be used (e.g.,
administration of TXA). The prevalence
of bleeding complications has been re-
duced with the use of TXA13,16,18,23,24.
The aim of the present study was to
assess the local efficacy of EACA in the
control of bleeding after tooth extraction
in warfarin-treated patients, without inter-
ference in anticoagulation. Only one study
has assessed the use of EACA mouthwash
in these patients; hence, the review pre-
sented herein focuses on articles in which
TXA was used for the same purpose14.

Table 3. Laboratory results on the day of extraction for the patients in each group.
EACA group (n = 70), n (%) Control group (n = 70), n (%)
Laboratory tests P-value
Median (min–max) Mean  SD Median (min–max) Mean  SD
INR 2.8 (2.0–4.0) 2.8  0.525 2.5 (2.0–4.0) 2.6  0.541 0.058
PT (seconds) 29.0 (21–53) 29.7  6.2 26.5 (19–40) 27.9  5.0 0.038
aPTT (seconds)a 51.0 (30–79) 50.9  12.1 54.0 (33–96) 56.1  11.9 0.009
Platelet count (cells  109/l)b 223 (83–415) 233.86  75.36 186 (63–481) 212.91  95.98 0.091
aPTT, activated partial thromboplastin time; EACA, epsilon-aminocaproic acid; INR, international normalized ratio; max, maximum; min,
minimum; PT, prothrombin time; SD, standard deviation.
a
Due to technical problems, the results for five patients in the EACA group and two in the control group are not included.
b
Due to technical problems, the results for 19 patients in the EACA group and five in the control group are not included.

Please cite this article in press as: da RV, et al. Intra-alveolar epsilon-aminocaproic acid for the control of post-extraction bleeding in
anticoagulated patients: randomized clinical trial, Int J Oral Maxillofac Surg (2018), https://doi.org/10.1016/j.ijom.2018.02.013
YIJOM-3884; No of Pages 7

Intra-alveolar EACA for anticoagulated patients 5

Table 4. Evaluation of graded late postoperative bleeding as recorded by the patients in each groupa.
EACA group (n = 70), n (%) Control group (n = 70), n (%)
No bleeding Mild bleeding Moderate bleeding Severe bleeding No bleeding Mild bleeding Moderate bleeding Severe bleeding
Day 1 47 (67.1) 15 (21.4) 8 (11.4) 0 46 (65.7) 14 (20.0) 10 (14.3) 0
Day 2 57 (81.4) 8 (11.4) 5 (7.1) 0 55 (78.6) 12 (17.1) 1 (1.4) 2 (2.9)
Day 3 67 (95.7) 1 (1.4) 2 (2.9) 0 63 (90.0) 6 (8.6) 1 (1.4) 0
Day 4 68 (97.1) 1 (1.4) 0 1 (1.4) 67 (95.7) 2 (2.9) 1 (1.4) 0
Day 5 69 (98.6) 1 (1.4) 0 0 67 (95.7) 1 (1.4) 2 (2.9) 0
Day 6 69 (98.6) 0 1 (1.4) 0 68 (97.1) 1 (1.4) 0 1 (1.4)
Day 7 68 (97.1) 0 1 (1.4) 1 (1.4) 70 (100.0) 0 0 0
EACA, epsilon-aminocaproic acid.
a
Mild bleeding = blood-stained saliva; moderate bleeding = bleeding controlled by gauze and pressure; severe bleeding = bleeding controlled
after professional intervention.

Table 5. Odds ratio evaluation of late postoperative bleeding according to the study group.
EACA group Control group
Late postoperative bleeding (n = 70), n (%) (n = 70), n (%) Total OR Adjusted ORa 95% CI P-value
Yes 11 (15.7) 12 (17.1) 23 (16.4) 0.901 1.08 0.40–2.92 0.871
No 59 (84.3) 58 (82.9) 117 (83.6) 1 1
CI, confidence interval; EACA, epsilon-aminocaproic acid; OR, odds ratio.
a
Adjusted OR = adjusted GEE (generalized estimating equations) for variables international normalized ratio (INR); activated partial
thromboplastin time (aPTT), and maxillary and mandibular molars or non-molars.

Table 6. Description of the late postoperative bleeding episodes in the five patients who required a professional intervention for bleeding control.
Patient INR aPTT (seconds) Group of teeth Day of bleeding Day of professional intervention Group
1 2.23 65 Mandibular non-molar 1 2 Control
2 3.00 82 Mandibular non-molar 1 2 Control
3 3.44 79 Maxillary molar 2 4 EACA
4 3.43 47 Maxillary molar 4 6 Control
5 2.19 44 Mandibular molar 7 7 EACA
aPTT, activated partial thromboplastin time; EACA, epsilon-aminocaproic acid; INR, international normalized ratio.

As just one immediate bleeding episode
(1.4%) was observed in the intervention
group, it was not possible to make a
statistical comparison of this outcome
with the control group. The bleeding in
this patient was easily controlled with
local pressure and without the need for
any other haemostatic measure.
Of the 23 episodes of late bleeding, 18
(78.3%) were classified as moderate. This
represents the largest rate of bleeding
complications observed in both groups,
without any statistically significant differ-
ence between them. It should be noted,
however, that most cases of bleeding oc-
curred on the day of surgery. Thus, it is
questioned whether these findings relate to
some misunderstanding that arose while
the patients were completing the late
bleeding assessment chart, since they were
instructed to keep on biting on the gauze
pack for another 20 minutes after being
discharged. This question came up be-
cause the highest prevalence of postoper-
ative bleeding was reported on the first28
and second postoperative days6,13,15. In
this study, six (26.1%) of the 23 events
occurred on day 2, and day 2 here corre-
sponds to the first postoperative day. In
this case, the intervention group had more
bleeding events.
Some conditions require an INR greater
than 3.0 for anticoagulation, thus patients
with an INR up to 4.0 on the day of surgery
were included. Although slightly higher
INR and PT levels were detected in the
intervention group on the day of surgery,
there was no statistically significant dif-
ference in the prevalence of late bleeding
when compared with the control group. Of
the five bleeding events that required re-
intervention, the INR was lower than 3.0
in three of them and greater than 3.0 in two
of them (one in each group). As the bleed-
ing frequency was low, it was not possible
to assess this relationship statistically, but
the results suggest no close relationship
with INR, as also reported by other stud-
ies10,27,37. Conversely, an INR greater
than 3.0 is strongly related to the higher
prevalence of bleeding events13, even
though postoperative bleeding has been
described even among patients with an
INR lower than 3.06,32.
There was no association with comor-
bidities or with the use of medications that
could increase the risk of bleeding in those
patients who required re-intervention for
the control of late bleeding and in the
patient who presented immediate bleed-
ing. In these cases, the placement of a
haemostatic sponge in the socket was
sufficient to stop the bleeding, and hospi-
talization or the transfusion of blood deri-
vatives was not necessary. Of the five late
bleeding events, three occurred after the
extraction of molars (two maxillary
molars and one mandibular molar). When
maxillary molar extraction sockets were
involved, the patients’ INR ranged from
3.0 to 4.0. Although maxillary molar ex-
traction sockets appear to be more prone to
bleeding13,28,38,39, there was no statisti-
cally significant relationship between late
bleeding events and the group of extracted
teeth.
Several of the previous studies asses-
sing the frequency of postoperative bleed-
ing in anticoagulated patients submitted to
invasive oral procedures failed to stan-
dardize the procedures used and compared
the procedures with different bleeding
patterns and in patients with different
INR, including ratios lower than 2.0,

Please cite this article in press as: da RV, et al. Intra-alveolar epsilon-aminocaproic acid for the control of post-extraction bleeding in
anticoagulated patients: randomized clinical trial, Int J Oral Maxillofac Surg (2018), https://doi.org/10.1016/j.ijom.2018.02.013
YIJOM-3884; No of Pages 7
6 da Silva et al.
which may be regarded as inappropriate
anticoagulation. In addition, many of these
studies lacked a control group and had a
small sample size. Consequently, the pres-
ent study was standardized to include only
simple tooth extractions, one per appoint-
ment, in patients with INR between 2.0
and 4.0. To avoid differences in bleeding
volume, the same surgeon performed all
extractions, seeking to use a technique that
was as minimally invasive as possible.
Therefore, procedures that required a
mucoperiosteal flap and osteotomy were
ruled out. More recent studies with larger
sample sizes and including a control group
have been reported, but they still included
surgical procedures with different intraop-
erative and postoperative bleeding pat-
terns6,7,10,15.
A systematic review on haemostatic
measures in anticoagulated patients sub-
mitted to oral surgeries revealed that TXA
is the local haemostatic agent most widely
used for the control of bleeding. This is
probably due to the fact that TXA is
considered to be six to 10 times more
potent than EACA40. Since TXA is not
sold as mouthwash in many countries,
injectable solutions are diluted in water
and used as mouth rinse. Given that EACA
has been used successfully for the control
of bleeding after oral surgeries in patients
with haemostatic diseases21,22, it repre-
sents an alternative to TXA for the pre-
vention and treatment of bleeding
complications in anticoagulated patients,
taking into account its cost and application
in countries where the use of TXA is not
approved. In Brazil, EACA is currently
sold only as 500-mg tablets or as an
injectable solution (1 or 4 g/ml). It was
decided to test its local application in the
form of crushed tablets. In 2003, the
American College of Cardiology (ACC)/
AHA protocol for warfarin therapy sug-
gested the use of EACA in anticoagulated
patients.
Since the prevalence of postoperative
bleeding was low even in the group in
which suture was the only haemostatic
measure used, it appears that suture alone
is sufficient for haemostasis in patients
with normal therapeutic anticoagulation
levels, as described by Bajkin et al. in
201412. In these patients, the decision to
use or not to use a suture may be based on
the extent or nature of the surgical wound
and not necessarily on the risk of bleed-
ing8.
While it has been reported that the
association of warfarin and antiplatelet
drugs could increase the bleeding risk
and volume29,30, none of the anticoagu-
lated patients treated with warfarin and
Please cite this article in press as: da RV, et al. Intra-alveolar epsilon-aminocaproic acid for the control of post-extraction bleeding in
anticoagulated patients: randomized clinical trial, Int J Oral Maxillofac Surg (2018), https://doi.org/10.1016/j.ijom.2018.02.013
acetylsalicylic acid in the present study
had postoperative bleeding even though
therapy with both drugs was maintained.
With the progressive use of new antic-
oagulants and direct thrombin and factor
Xa inhibitors, it would be relevant to
reproduce the methodology described in
the present study in anticoagulated
patients treated with these drugs.
In view of the commonly observed
fluctuation in the INR of warfarin-treated
patients, the surgical procedure had to be
postponed several times in this clinical
trial until the predefined levels could be
reached. The small sample size was there-
fore a limitation of this study. As few
bleeding events occurred, it is assumed
that more tooth extractions would have to
be performed in order to verify whether
the results obtained herein would be con-
firmed in a larger sample. On the other
hand, this study was able to confirm the
findings of other authors, who reported a
low prevalence of bleeding events among
these patients, with no need to reduce or
discontinue anticoagulation. In the recent
past, dentists and attending physicians
instructed their patients to discontinue
anticoagulant therapy without taking into
consideration the bleeding that is expected
from the surgical procedure and the risk of
thrombotic events25. It is hoped that the
present study findings will help discourage
this practice.
Accordingly, although no differences in
postoperative bleeding were observed be-
tween the study groups, it may be con-
cluded that topical EACA is an alternative
for the control of postoperative bleeding
after simple tooth extractions in anticoa-
gulated patients with INR between 2.0 and
4.0. Given the low prevalence of bleeding,
even in those patients in whom suture
alone was performed, it is suggested that
local haemostatic agents are used exclu-
sively to manage bleeding events. This
practice could reduce treatment costs for
the patient.
Funding
We thank FAPERJ (Fundação Carlos Cha-
gas Filho de Amparo à Pesquisa do Estado
do Rio de Janeiro) for the financial sup-
port, which enabled the authors to develop
this study through a research grant (APQ1)
awarded to Professor Telma Gadelha un-
der protocol number E-26/111508/2011.
Competing interests
The authors declare no conflicts of
interest.
Ethical approval
This study was approved by the Research
Ethics Committee of Clementino Fraga
Filho Teaching Hospital on March 5,
2009, under protocol number 224/08.
Patient consent
All patients agreed to participate in the
study and signed a free informed consent
form.
References
1. Raunso J, Selmer C, Olesen JB, Charlot MG,
Olsen AM, Bretler DM, Nielsen JD, Domin-
guez H, Gadsboll N, Kober L, Gislason GH,
Torp-Pedersen C, Hansen ML. Increased
short-term risk of thrombo-embolism or death
after interruption of warfarin treatment in
patients with atrial fibrillation. Eur Heart J
2012;33:1886–92.
2. Li HK, Chen FC, Rea RF, Asirvatham SJ,
Powell BD, Friedman PA, Shen WK, Brady
PA, Bradley DJ, Lee HC, Hodge DO, Slusser
JP, Hayes DL, Cha YM. No increased bleed-
ing events with continuation of oral antic-
oagulation therapy for patients undergoing
cardiac device procedure. Pacing Clin Elec-
trophysiol 2011;34:868–74.
3. Balevi B. Should warfarin be discontinued
before a dental extraction? A decision-tree
analysis. Oral Surg Oral Med Oral Pathol
Oral Radiol Endod 2010;110:691–7.
4. Ansell J, Hirsh J, Hylek E, Jacobson A,
Crowther M, Palareti G. Pharmacology and
management of the vitamin K antagonists—
American College of Chest Physicians evi-
dence-based clinical practice guidelines (8th
edition). Chest 2008;133(6 Suppl):160S–98S.
5. Buller HR, Agnelli G, Hull RD, Hyers TM,
Prins MH, Raskob GE. Antithrombotic thera-
py for venous thromboembolic disease: the
Seventh ACCP Conference on Antithrombo-
tic and Thrombolytic Therapy. Chest
2004;126(3 Suppl):401–28.
6. Sacco R, Sacco M, Carpenedo M, Mannucci
M. Oral surgery in patients on oral anticoag-
ulant therapy: a randomized comparison of
different intensity targets. Oral Surg Oral
Med Oral Pathol Oral Radiol Endod
2007;104:e18–21.
7. Ferrieri GB, Casriglioni S, Carmagnola D,
Cragnel M, Strobmenger L, Abati S. Oral
surgery in patients on anticoagulant treatment
without therapy interruption. J Oral Maxillo-
fac Surg 2007;65:1149–54.
8. Al-Mubarak S, Al-Ali N, Rass MA, Al-Sohail
A, Robert A, Al-Zoman K, Al-Zoman K, Al-
Suwyed A, Ciancio S. Evaluation of dental
extractions, suturing and INR on postopera-
tive bleeding of patients maintained on oral
anticoagulant therapy. Br Dent J 2007;203:
e15.
YIJOM-3884; No of Pages 7
9. Aframian DJ, Lalla RV, Peterson DE. Man-
agement of dental patients taking common
hemostasis altering medications. Oral Surg
Oral Med Oral Pathol Oral Radiol Endod
2007;103(Suppl 1). S45.e1–S45.e11.
10. Broekema FI, Minnen B, Jansma J, Bos RM.
Risk of bleeding after dentoalveolar surgery
in patients taking anticoagulants. Br J Oral
Maxillofac Surg 2014;52:e15–9.
11. Karsh ED, Erdogan O, Esen E, Acarturk E.
Comparison of the effects of warfarin and
heparin on bleeding caused by dental extrac-
tion: a clinical study. J Oral Maxillofac Surg
2011;69:2500–7.
12. Bajkin BV, Selakovic SD, Mirkovic SM,
arcev IN, Tadic AJ, Milekic BR. Comparison
of efficacy of local hemostatic modalities in
anticoagulated patients undergoing tooth
extractions. Vojnosanit Pregl
2014;71:1097–101.
13. Carter G, Goss A, Lloyd J, Toccibetti R.
Tranexamic acid mouthwash versus autolo-
gous fibrin glue in patients taking warfarin
undergoing dental extractions: a randomized
prospective clinical study. J Oral Maxillofac
Surg 2003;61:1432–5.
14. Souto JC, Oliver A, Zuazu-Jausoro I, Vives
A, Fontcubertas J. Oral surgery in anticoa-
gulated patients without reducing the dose of
oral anticoagulant: a prospective randomized
study. J Oral Maxillofac Surg 1996;54:27–
32.
15. Bacci C, Maglione M, Favero L, Perini A, Di
Lenarda R, Berengo M, Zanon E. Manage-
ment of dental extraction in patients under-
going anticoagulant treatment—results from
a large, multicentre, prospective, case–con-
trol study. Thromb Haemost 2010;104:972–
5.
16. Patatanian E, Fugate SE. Hemostatic
mouthwashes in anticoagulated patients un-
dergoing dental extraction. Ann Pharmac-
other 2006;40:2205–10.
17. Carter G, Goss A. Tranexamic acid mouth-
wash—a prospective randomized study of a
2-day regimen vs 5-day regimen to prevent
postoperative bleeding in anticoagulated
patients requiring dental extractions. Int J
Oral Maxillofac Surg 2003;32:504–7.
18. Morimoto Y, Niwa H, Minematsu K. Hemo-
static management of tooth extractions in
patients on oral antithrombotic therapy. J
Oral Maxillofac Surg 2008;66:51–7.
19. Carter G, Goss A, Lloyd J, Tocchetti R.
Current concepts of the management of den-
tal extractions for patients taking warfarin.
Aust Dent J 2003;48:89–96.
20. Manual de atendimento odontológico a
pacientes com coagulopatias hereditárias.
Brazil: Ministry of Health, 2008. http://
bvsms.saude.gov.br/bvs/publicacoes/
manual_odontologico_coagulopatias_
Please cite this article in press as: da RV, et al. Intra-alveolar epsilon-aminocaproic acid for the control of post-extraction bleeding in
anticoagulated patients: randomized clinical trial, Int J Oral Maxillofac Surg (2018), https://doi.org/10.1016/j.ijom.2018.02.013
Intra-alveolar EACA for anticoagulated patients
hereditarias.pdf [Accessibility verified Janu-
ary 23, 2018].
21. Ghosh K, Shetty S, Jijina F, Mohanty D. Role
of epsilon amino caproic acid in the man-
agement of haemophilic patients with inhi-
bitors. Haemophilia 2004;10:58–62.
22. Gomes MF, de Melo RM, Plens G, Pontes
EM, Silva MM, da Rocha JC. Surgical and
clinical management of a patient with Glanz-
mann thrombasthenia: a case report. Quin-
tessence Int 2004;35:617–20.
23. Wahl MJ, Pinto A, Kilham J, Lalla RV.
Dental surgery in anticoagulated patients—
stop the interruption. Oral Surg Oral Med
Oral Pathol Oral Radiol 2015;119:136–57.
http://dx.doi.org/10.1016/j.
oooo.2014.10.011.
24. Todd DW. Evidence to support an individu-
alized approach to modification of oral anti-
coagulant therapy for ambulatory oral
surgery. J Oral Maxillofac Surg
2005;63:536–9.
25. Lim W, Wang M, Crowther M, Douketis J.
The management of anticoagulated patients
requiring dental extraction: a cross-sectional
survey of oral and maxillofacial surgeons
and hematologists. J Thromb Haemost
2007;5:2157–9.
26. Zanon E, Martinelli F, Bacci C, Cordioli G,
Girolami A. Safety of dental extraction
among consecutive patients on oral antico-
agulant treatment managed using a specific
dental management protocol. Blood Coagul
Fibrinolysis 2003;14:27–30.
27. Febbo A, Cheng A, Stein B, Goss A, Sam-
brook P. Postoperative bleeding following
dental extractions in patients anticoagulated
with warfarin. J Oral Maxillofac Surg
2016;74:1518–23. http://dx.doi.org/
10.1016/j.joms.2016.04.007.
28. Yang S, Shi Q, Liu J, Li J, Xu J. Should oral
anticoagulant therapy be continued during
dental extraction? A meta-analysis. BMC
Oral Health 2016;16:81. http://dx.doi.org/
10.1186/s12903-016-0278-9.
29. Hong C, Napenas JJ, Brennan M, Furney S,
Lockhart P. Risk of postoperative bleeding
after dental procedures in patients on warfa-
rin: a retrospective study. Oral Surg Oral
Med Oral Pathol Oral Radiol
2012;114:464–8.
30. Medeiros FB, Andrade AN, Angelis GA,
Conrado VC, Timerman L, Farsky P, Dibb
LL. Bleeding evaluation during single tooth
extraction in patients with coronary artery
disease and acetylsalicylic acid therapy sus-
pension: a prospective, double-blinded, and
randomized study. J Oral Maxillofac Surg
2011;69:2949–55. http://dx.doi.org/
10.1016/j.joms.2011.02.139.
31. Cunningham Jr LL, Brandt MT, Aldridge E.
Perioperative treatment of the patient taking
7
anticoagulation medication. Oral Maxillofac
Surg Clin North Am 2006;18:151–9.
32. Bodner L, Weinstein JM, Kleiner A. Efficacy
of fibrin sealant in patients on various levels
of oral anticoagulant undergoing oral sur-
gery. Oral Surg Oral Med Oral Pathol Oral
Radiol Endod 1998;86:421–4.
33. Blinder D, Manor Y, Martinowitz U, Taicher
S. Dental extractions in patients maintained
on continued oral anticoagulant. Compari-
son of local hemostatic modalities. Oral
Surg Oral Med Oral Pathol Oral Radiol
Endod 1999;88:137–40.
34. Al-Belasy FA, Amer MZ. Hemostatic effect
of n-butyl-2-cyanoacrylate (histoacryl) glue
in warfarin-treated patients undergoing oral
surgery. J Oral Maxillofac Surg
2003;61:1405–9.
35. Halfpenny W, Fraser JS, Adlam DM. Com-
parison of 2 hemostatic agents for the pre-
vention of postextraction hemorrhage in
patients on anticoagulants. Oral Surg Oral
Med Oral Pathol Oral Radiol Endod
2001;92:257–9.
36. Wahl MJ. Dental surgery in anticoagulated
patients. Arch Intern Med 1998;158:1610–6.
37. Salam S, Yusuf H, Milosevic A. Bleeding
after dental extractions in patients taking
warfarin. Br J Oral Maxillofac Surg
2007;45:463–6.
38. Devani P, Laver KM, Howell CJ. Dental
extractions in patients on warfarin: is alter-
ation of anticoagulant regime necessary. Br J
Oral Maxillofac Surg 1998;36:107–11.
39. Beirne OR. Evidence to continue oral anti-
coagulant therapy for ambulatory oral sur-
gery. J Oral Maxillofac Surg 2005;63:540–5.
40. Soares EC, Costa FW, Bezerra TP, Nogueira
CB, de Barros Silva PG, Batista SH, Sousa
FB, Sá Roriz Fonteles C. Postoperative he-
mostatic efficacy of gauze soaked in tranexa-
mic acid, fibrin sponge, and dry gauze
compression following dental extractions
in anticoagulated patients with cardiovascu-
lar disease: a prospective, randomized study.
Oral Maxillofac Surg 2015;19:209–16.
Address:
Rosangela Varella da Silva
Private Clinic
Av. Armando Lombardi 1000/Bloco 2/sala
108
Barra da Tijuca
Rio de Janeiro
RJ – CEP 22640 000
Brazil
Tel.: +55 21 24916267
+55 21 999134130
E-mail: rosangela.varella@saojose.br