Clin Oral Invest (2015) 19:171–180

DOI 10.1007/s00784-014-1366-3

REVIEW

Oral surgery during therapy with anticoagulants—a systematic

review
Peer W. Kämmerer & Bernhard Frerich & Jan Liese &
Eik Schiegnitz & Bilal Al-Nawas

Received: 23 September 2014 / Accepted: 13 November 2014 / Published online: 22 November 2014
# Springer-Verlag Berlin Heidelberg 2014

Abstract Neither the international normalized ratio (INR), within the

Objectives Oral anticoagulation therapy (OAT) with vitamin therapeutic range (2–4), nor the extent of the minor oral
K inhibitors protects the patients from thromboembolic surgery procedure had an influence on postoperative bleeding
events. It may however lead to excessive hemorrhage during episodes.
and after an oral surgery procedure. The aim of this systematic Discussion There is strong evidence that OAT patients under-
review was to evaluate the justifications to reduce, withdraw, going minor oral surgery should not discontinue their medi-
or alter OATs prior to minor oral surgery procedures to man- cation in order to prevent thromboembolic complications.
age bleeding events. Clinical relevance Nonetheless, INR should be less than 4,
Materials and methods A systematic MEDLINE search was local hemostatic measures are of high importance and patients
conducted for clinical studies in English or German language need to be instructed and closely monitored as minor bleed-
from 1994 to 2014 comparing patients treated with OAT, ings might occur more often in OAT patients.
without OAT, as well as patients with altered OAT for oral
surgery purposes. Relevant outcome parameters were: post- Keywords Oral minor surgery . Extractions . Osteotomy .
operative local hemostasis, bleeding episodes, occurrence of Oral anticoagulation . Warfarin . Pheprocoumon .
thromboembolic events, and other complications due to the Acenocoumarol . Bridging . Review
anticoagulation medication. A hand search for references cited
in the identified publications completed the review.
Results After screening of 1755 abstracts, 16 clinical studies Introduction
were identified according to the selection criteria. Due to the
heterogeneity of the obtained data, aggregation and synthesis Warfarin, acenocoumarol, and phenprocoumon are deriva-
were not possible. There was no significant difference in tives of 4-hydroxycoumarin. It acts as a competitive inhibitor
bleeding events comparing patients under continued OAT to to vitamin K (inhibition of coagulation factors II, VII, IX, X,
those with reduced, altered, and/or discontinued OAT medi- and protein C and S). They are usually administered for
cations. Minor bleeding events in the test and control groups prevention of thromboembolic events in patients with hyper-
were successfully stopped with local measures. However, no coagulable conditions [1]. The anticoagulative effect develops
superiority of a single hemostatic measure could be identified. after a latency of around 48 h and takes 4 to 5 days to be fully
active with a half-life of at least 48 h [2]. This means that the
anticoagulant effect persists even after discontinuation for a
P. W. Kämmerer (*) : B. Frerich : J. Liese comparatively long time [3]. Oral anticoagulant therapies
Department of Oral, Maxillofacial and Plastic Surgery, University
(OATs) with vitamin K inhibitors are very effective for pro-
Medical Centre of Rostock, Schillingallee, 35, 18057 Rostock,
Germany phylaxis of prospectively life-threatening thromboembolic
e-mail: [email protected] events [4]. A major disadvantage of OAT is the increased risk
of hemorrhage after injuries or surgical procedures. OATs are
E. Schiegnitz : B. Al-Nawas
frequently administered in older patients, who on the other
Department of Oral, Maxillofacial and Plastic Surgery, University
Medical Centre of the Johannes Gutenberg University Mainz, Mainz, hand, have a higher demand for dental and oral surgery.
Germany Consequently, postoperative bleeding events are of a
172 Clin Oral Invest (2015) 19:171–180

significant concern. Additionally, subsequent complications Materials and methods

may also occur such as oral hematoma formation leading to a
postoperative trismus or even severe upper airway obstruc- Systematic review
tion. Previous reports indicated that the fear of possible hem-
orrhages may increase the stress levels in cardiac patients A detailed protocol was developed according to the Preferred
which in turn could induce a fibrinolytic activity [5]. Reporting Items for Systematic Review and Meta-Analyses
In order to improve the management of bleeding risks, (PRISMA) statement [19]. The focused primary question
different patient management recommendations have been concerning the Patient, Intervention, Comparison, and
proposed in the last decades addressing the OAT. Those Outcome (PICO) format [20] was: “Is the incidence of bleed-
recommendations include a reduction [6, 7], a temporarily ing complications (early/late) higher in OAT patients under-
cessation of OAT [8], or a substitution (bridging) with heparin going oral surgery compared to patients without such OAT
[9, 10] prior to surgical procedure. Reduction and discontin- and with a changed anticoagulation regimen?”
uation of OAT may lead to an increased risk of thromboem- A systematic MEDLINE review was conducted using the
bolism [11]. In brief, a suspension of vitamin K inhibitors for key sentences anticoagulation dentistry, anticoagulation oral
2 days may increase the risk for a thromboembolic event from surgery, anticoagulation local anesthesia, and anticoagulation
0.02 to 1 % [12]. Garcia et al. prospectively analyzed minor surgery. English and German articles of the last
the effect of interruption of warfarin for <5 days before 20 years, between 1994 and 2014, were scanned (last update
various minor surgery procedures. The authors conclud- July 2014). Additionally, a manual search of reference lists of
ed that this interruption is associated with a risk of topic-related articles was performed. Two independent exam-
thromboembolism of 0.7 % and a 1.7 % higher risk of iners conducted the comprehensive search of studies of po-
significant clinical bleeding episodes [13]. When tential relevance (BA & PWK).
substituting vitamin K inhibitors with heparin, an im-
portant side effect could emerge which is the heparin- Study selection and data extraction
induced thrombocytopenia that may even be complicat-
ed by thromboembolism as well. Clinical prospective, retrospective, and cohort studies com-
In cases of new direct anticoagulants such as selective paring patients with full OAT to patients without OAT, as well
inhibitors of factor II, X, or platelet aggregation inhibitors, as patients with reduced/withdrawn/bridged OAT that
scientific evidence for withdrawal, reduction, or continuation underwent oral surgery procedures were included. Studies
of the drugs is limited. Wahl summarized the available data in on antiplatelet medications as well as case reports, reviews,
a recent review and stated that out of 1283 patients, with 2343 and expert recommendations were excluded. Relevant main
oral surgery procedures on antiplatelet medication, only 2.7 % outcome parameters were postoperative local hemostasis, de-
had bleeding complications that were severe in only 0.2 % of layed bleeding episodes, and other complications such as
cases [14]. Furthermore, the discontinuation of aspirin is thromboembolisms due to the anticoagulation medication or
obsolete nowadays [15]. its withdrawal or reduction. Titles and abstracts of identified
In brief, the risk of thromboembolism due to reduction, articles were checked for eligibility and relevance. Full texts
discontinuation, or substitution of OATs has to be balanced of the studies that fulfilled the inclusion criteria were assessed
with the risk of excessive bleeding. More recent publications and reviewed. In case of disagreement, inter-reviewer discus-
suggested that minor oral surgery procedures such as place- sions were conducted and consensus was obtained.
ment of dental implants and teeth extractions are to be carried
out without interruption of OATs [16, 1], mostly on the basis
of the international normalized ratio (INR) within the thera-
peutic range [1, 17]. Nevertheless, there is no generalized Results
agreement that leads to a standardized treatment protocol for
those patients. Hence, a variety of treatment approaches are Systematic review
performed by oral surgeons [18]. In Germany, there are no
evidence-based practical guidelines for oral surgery in cases of Initially, a total of 1755 abstracts were selected. They were
OAT patients available yet. screened and reviewed for the selected in- and exclusion
Therefore, the present review evaluated the scientific evi- criteria. Sixteen prospective and retrospective clinical studies
dence for the need to disrupt, reduce, and/or bridge OATs comparing patients with and without OAT (alteration, with-
before minor oral surgery procedures. A special emphasis drawal, no OAT) under oral surgery procedures were found
was on bleeding episodes (early hemostasis and delayed [21, 1, 22–27, 2, 3, 28–31, 9, 17] (Table 1). Due to the
bleeding events) and other complications such as heterogeneity of the key parameters, aggregation of statistical
thromboembolism. data was not possible. Therefore, not a meta-analysis but a
Table 1 Included clinical studies with control groups in alphabetical order

Report Number of OAT Procedures Groups Postoperative Follow-up Postoperative Delayed bleeding Others
patients measures bleeding

Al-Belasy and 40 Warfarin Dental extractions, I. Non-OAT patients; gelatin Local pressure for Contact when Complete in Group II, 5 (33 %); None
Amer, 2003 mucoperiosteal flap, sponge, multiple 20 min bleeding, all patients significant more
alveoloplasty absorbable sutures otherwise day bleeding events in
(n=10) II. OAT patients; 10 group II
gelatin sponge, multiple
sutures (n=15) III. OAT
patients; histoacryl glue
Clin Oral Invest (2015) 19:171–180

sutures (n=15)
Al-Mubarak 214 Warfarin Dental extractions I. OAT stopped 2 days prior, Local pressure for Observation 1 h, n. a. No difference Significant
et al., 2007 resume after 12 h; no at least 30 min days 1, 3, 7 correlation
sutures (n=48); II. between higher
Continue OAT no INR level (>3)
sutures (n=58); III. OAT and increased
stop 2 days prior, resume bleedings
after 12 h, sutures
(n=56); IV. Continue
OAT, sutures (n=52)
Bacci et al., 2010 900 Warfarin Dental extractions Group I: OAT (n=451). Local and Days 3 and 8 n. a. No difference. Group I, None
Group II: Non-OAT tranexamic acid 7 (1.6 %); group II, 4
(n=449). Cellulose and for 30–40 min (0.9 %)
resorbable sutures in all
patients.
Bacci et al., 2011 154 Warfarin Implant dentistry I. OAT patients (n=109). Gauze with Days 3 and 8 n. a. No difference. Group I, None
Sutures in all patients tranexamic acid 2 (4 %); group II, 3
for 30–60 min; (2.8 %)
external ice
packs 6–8 h; no
dentures for 2–
3 weeks
Bajkin et al.,2009 214 Acemocumarol Dental extractions I. Continue OAT, collagen Local pressure for 30 min, 2 h; days n. a No difference. Group I, No sutures if
(n=195), warfarin sponges (n=109) II. 30 min 1, 2, 4, 7 8 (7.3 %); group II: 5 possible
(n=18), Switched to LMWH 3– (4.8 %)
phenprocoumon 4 days before
(n=1) intervention discontinued
12 h before operation.
OAT re-started at the
evening (n=105)
Blinder et al., 249 Coumarin Dental extractions Group I, INR 1.5–1.99 n. a n. a n. a No difference. Group I, None
2001 (n=59); group II, INR 3 (5.1 %); group II,
2–2.49 (n=78); group 10 (12.8 %); group
III, INR 2.5–2.99 III, 9 (15.2 %); group
(n=59); group IV, INR IV, 5 (16.7 %); group
3–3.49 (n=30); group V, 3 (13 %)
V, INR >3.5 (n=23).
Gelatin sponge and
non-resorbable sutures
in all patients
Bublitz et al., 101 Phenprocoumon I. OAT patients with n. a n. a n. a More bleeding in the In the control group
2000 collagen (n=31) II. OAT non-OAT group. without OAT,
173
Table 1 (continued)
174

Report Number of OAT Procedures Groups Postoperative Follow-up Postoperative Delayed bleeding Others
patients measures bleeding

Dental extractions, patients with tranexamic Group I, 6 (19 %); more bleeding
osteotomies, root acid (n=32) III. group II, 2 (6 %); events
resections, others Reduction of INR to 2, group III, 15 (40 %)
flap and sutures (n=38)
Campbell et al., 35 Warfarin Dental extractions, I. stop OAT 72–96 h prior n. a Day 1 Complete in No differences; no A higher blood loss
2000 alveoloplasty, surgery (n=12); II: all patients bleeding needing measured in
limited oral surgery Continue OAT (n=13); intervention “surgical units” in
III. Never on OAT group 2
(n=10)
Cannon and 70 Warfarin Minor oral procedures I. OAT stopped 2 days prior Local pressure for Days 3 and 5 Complete in No differences. Group I, None
Dharmar, 2003 and dental and resume same day in 20 min all patients 2 (5.7 %); group II, 3
extractions the evening (n=35). (5.6 %)
Sutures and cellulose
wound dressing in cases
of bone removal/soft
tissue surgery II.
Continue OAT (n=35).
Sutures and cellulose
wound dressing always
used
Devani et al., 65 Warfarin Extractions I. Normal OAT (n=33) II. Local pressure for 30 min, days 3 and Complete in No differences. Group I, None
1998 Stop OAT 2 days prior at least 30 min 5 all patients 1 (3 %); group II, 1
surgery (n=32). Both (3 %)
groups with cellulose
dressing and sutures.
Eichhorn 922 Phenprocoumon Osteotomies, dental I. Oral surgery with OAT n. a Days 1, 7, 10, 14 n. a More minor bleedings More minor bleeding
et al.,2012 extractions, (n=285). Both groups on OAT patients events in OAT
apicoectomies, with collagen, Group I, 47 (7.4 %); patients; most in
others cellulose, sutures, group II, 0.7 % the molar region
acrylic, splint,
compression and
combinations
Evans et al., 2002 109 Warfarin Dental extractions I. OAT stopped 2 day prior Local pressure for 20 min, day 7 n. a Nonsignificant more None
and resume same day in 10 min. bleedings in OAT
the evening (n=7). II. group. Group I, 7
Continue OAT (n=52). (14 %); group II, 15
Both groups with (26 %)
cellulose dressing and
resorbable sutures.
Gaspar et al., 47 Vitamin K inhibitor Dental extractions I. Oral surgery with OAT Mouth rinse with Day 7 Group I, 2 No difference. Group I, None
1997 (n=32) II. OAT stopped tranexamic acid (6.3 %); 2 (6.3 %); group II, 1
3 days prior surgery for 2 min. group II, 1 (6.7 %)
(n=15). Both group sites Rinsing 4×d for (6.7 %)
irrigated with tranexamic 7 postoperative
acid, cellulose, days
resorbable sutures.
Sacco et al., 2007 131 Acenocoumarol/ Extractions, I. OAT stopped till INR 1, In group 2, 2 h, days 1–6 on Group I, 10 Nonsignificant more All bleeding patients
warfarin cystectomies, 5-2 and therapy tranexamic acid telephone. Day (15.2 %); bleedings in group I. returned to
implantation continued 48 h after postoperatively 7 examination. hospital
Clin Oral Invest (2015) 19:171–180
Clin Oral Invest (2015) 19:171–180 175

More bleedings in cases Statistic methods not

appropriate due to
small sample size
descriptive analysis of obtained studies was conducted.
Within the included studies, the number of patients (together
with control groups) ranged between 35 and 900 (mean 240).
Six studies were randomized clinical trials [21, 1, 24, 29, 31,
Others

No difference. Group I, None

9].

4 (1.6 %); group II, 3

Group I, 10 (15.2 %);

Group I, 15; group II,

group IV, 17; group

group II, 6 (9.2 %)

V, 31; group VI, 0

50; group III, 36;
of higher INRs. OAT medication
Postoperative Delayed bleeding

All included studies evaluated patients under oral

(1.2 %)
anticoagulation with vitamin K inhibitors; although a variety
of coumarin drugs were used. In most of the studies, OAT
patients received warfarin [21, 1, 23, 22, 27, 2, 3, 29, 17].
group II, 6
(6.2 %)

Others reported patients under phenprocoumon [26, 28],

bleeding

acemocoumarol [9], vitamin K inhibitors in general [25, 30],

n. a

n. a

or mixtures of different vitamin K inhibitors [24, 31] (Table 1).

The INR of the OAT patients was set at a therapeutic range of
1.5–4.
Days 3 and 8
Follow-up

Oral surgery procedures

n. a
tranexamic acid

6 h for 2 days).

Most of the obtained studies were principally focused on tooth

Local pressure for
and every 6 h

30–60 min in
OAT patients
(2 min every

extraction. Therefore, generalization of all data to the whole

mouthwash
Postoperative

for 2 days

EACA and

field of oral surgery is difficult. However, there were other

used as
measures

reports on more extended oral surgical procedures such as

implant dentistry [23], cystic excisions [31], mucoperiosteal
flaps, alveoloplasties [21, 27], osteotomies [26, 28], and
heparin, tranexamic acid,
sponges for both groups

(n=250). In all patients,

doses of acenocoumarol,
Continue OAT (n=65).

fibrin sponge and non-

Group II: Never OAT
protocols. Numbers of
Gelatin and collagen

Group I: OAT (n=250)

VI groups with different

EACA and irrigation

others such as biopsies [28] (for details, see Table 1).

resorbable sutures.
surgery (n=66) II.

patients/group n. a

Test and control groups

Groups

A large variety of test and control groups were identified.

Patients under continued OAT with different local hemostatic
procedures were compared to non-OAT patients [21–23, 27,
Dental extractions

Dental extractions

28, 17]. Patients under continued OAT with or without addi-

tional hemostatic measures were compared to discontinued
Procedures

OAT with or without additional hemostatic measures [1, 26,

27, 2, 3, 29–31, 9], and patients under continued OAT were
compared to bridging with low-molecular-weight heparin
n. a. not available, OAT oral anticoagulant therapy

(LMWH) [24]. Also, different levels of INR were grouped

Acenocoumarol

and compared to each other [25] (for details, see Table 1).
Warfarin

Intraoperative hemostatic measures

Number of OAT

There was a distinct heterogeneity in the use of local hemo-

patients

static measures. In brief, gelatin sponges [21, 25, 31], sutures

Zanon et al., 2003 500

(absorbable and non-absorbable) [1, 22, 23, 25, 26, 2, 28, 30,
Souto et al., 1996 92
Table 1 (continued)

17], histoacryl glue, cellulose [22, 2, 3, 29, 30], tranexamic

acid [26, 30, 9], epsilon-aminocaproic acid [9], collagen
sponges [24, 28, 31], fibrin sponges [17], acrylic splints
Report

[28], as well as combinations of those measures were applied

(for details, see Table 1).
176 Clin Oral Invest (2015) 19:171–180

Postoperative hemostatic measures OAT vs. interruption of OAT

In 12 studies, data regarding postoperative hemostatic mea- There was no significant difference between the two groups
sures were reported. Those were local compression for 10– [1, 27, 2, 3, 30]. While Evans et al. found nonsignificant more
60 min [21, 2, 3, 1, 22, 24, 29, 17], tranexamic acid [31, 22, bleeding events in OAT patients (26 vs. 14 %) [29]. On the
30, 32], external ice packs [23], abstention of dentures [23], other hand, Bublitz et al. and Sacco et al. found more bleeding
and epsilon-aminocaproic acid [9] (for details, see Table 1). events in patients under reduced OAT [26, 31] (Table 1).

OAT vs. bridging with LMWH

Follow-up time
This was examined in one study that showed more, nonsig-
There was a broad variety in follow-up appointments and nificant, minor bleedings in the OAT group only (7.3 vs.
regimens. Whereas some authors instructed their patients to 4.8 %) [24] (Table 1).
contact them prior the 10th day in cases of bleedings only [21]
or conducted telephone interviews prior the 7th postoperative Procedures and delayed bleeding episodes
day [31], others conducted frequent clinical follow-up exam-
inations [1, 24, 28]. In three studies, no information in regard Even without the use of sutures, Al-Mubarak et al. reported an
to follow-up time was documented [25, 26, 9] (for details, see insignificant increase of postoperative bleeding in extraction
Table 1). patients under continued OAT only [1]. Bajkin et al. used no
sutures when possible and found no significant difference
between the groups as well [24]. The number of extracted
Analysis of general outcome parameters teeth (ranging 1–5) could not be correlated with the bleeding
tendencies [1]. Similarly, other reports evaluated no associa-
Postoperative bleeding events tions between number of teeth extracted and bleeding events
[33, 29]. Bacci et al. could not find any difference in bleeding
In six publications, data about postoperative local bleedings events between OAT and non-OAT patients neither for dental
were reported [21, 27, 2, 3, 31, 30]. When comparing OAT extractions nor for insertion of dental implants placement with
patients with non-OAT patients, no differences were seen [21, flaps elevation and even with augmentation procedures [23,
27]. Similar results were shown when comparing OAT pa- 22]. Others reported those findings as well [24, 25, 27, 3,
tients to those with interruption of OAT [27, 2, 3, 30, 31] 29–31, 17] (Table 1).
(Table 1).
Other complications

Delayed bleeding episodes In the included literature, after a maximum follow-up time of
14 days, no case of thromboembolism was reported. Eichhorn
Data about delayed bleeding events were reported in all in- et al. reported 15 OAT patients in need for prolonged treat-
cluded publications [21, 1, 22–27, 2, 3, 28–31, 9, 17] ment in a hospital due to bleeding events [28].
(Table 1).

Comparisons within groups Discussion

OAT vs. non-OAT OAT significantly reduces the risks of arterial and venous
thromboembolism [34, 35]. Interruption of OAT in patients
Al-Belasy and Amer reported a significantly more delayed at risk leads to three times more likely to serious embolic
minor postoperative bleeding in the OAT group with gelatin complications [36]. Moreover, arterial thromboembolism
sponges and sutures compared to the non-OAT and the OAT leads in about 40 % of all cases to permanent disability and
group with histoacryl glue and sutures. All bleedings could be is lethal in about 20 % of the cases. However, 3 % of postop-
stopped with local pressure and tranexamic acid [21]. erative bleedings in general surgery due to OAT resulted in
Eichhorn et al. reported more minor bleeding episodes in death of the patients [37]. For oral surgery procedures, no case
OAT patients [28]. Other authors could not identify a signif- of lethal postoperative bleedings under continuation of OAT
icant difference between OAT and non-OAT patients [22, 23, was reported [29, 36], whereas several fatal thromboembolic
27, 17] (Table 1). events after stopping the OAT for dental extractions are
Clin Oral Invest (2015) 19:171–180 177

known [38, 39]. Nevertheless, none such event was reported available in oral surgery—not only for OAT patients; those
in the included studies. include sutures [50], physical methods such as local compres-
There is a lack for comprehensive consensus in regard of sion [21, 24] and adjuvants like fibrin and histoacryl glue [51,
preoperative alteration of the anticoagulant medication to 21, 52, 40], local antifibrinolytic solutions [32, 23, 30, 53],
prepare OAT patients for minor oral surgery. Some of studies acrylic splints [28], collagen fleeces [24], gelatin sponges
were carried out with limited patient numbers (<100) [30, 40, [47], and cyanoacrylate [15]. When comparing the efficacy
21, 27, 3, 9], without respective control groups [41, 42, 40, 33] of different hemostatic methods, Blinder et al. concluded that
or without randomization [23, 27, 2]. Nearly all of those gelatin sponges and sutures are sufficient in patients with OAT
studies had different treatment modalities in the test and under minor dental surgery [54], whereas others suggested
control groups. Besides, there was no standard treatment deciding case-by-case on suturing as it will lead to further
protocol per case-need reported. Factors of influence such as trauma to the soft tissue [1, 24]. The combination of gelatin
gingival health, level of difficulty of teeth extractions, or other sponges with sutures and fibrin sealant led to a negligible
surgical interventions, type of local anesthesia, material and number of minor bleedings in OAT patients even after com-
suture technique, local hemostatic wound covering, and the plicated surgical removal of several teeth. In this study, no
use of analgesics differed from one study to another or were correlation to the degree of surgical trauma as well as the INR
not reported. This heterogeneity limits the scientific evidence intensity was detected [52]. Elevation of flaps with sutures in
obtained by this review. OAT patients with reduced phenprocoumon has shown to
In their broad review, Kosyfaki et al. recommended to produce more bleedings compared to those with unaltered
adopt individualized treatment strategies to OAT patients OAT and installation of collagen/tranexamic acid [26].
[43]. Nevertheless, similar to the review by Madrid and Sanz Cannon and Dharmar used cellulose wound dressings in pa-
[16], this systematic review on OAT patients undergoing tients under OAT and occasionally in patients under with-
dental surgery highlights a strong evidence to continue OAT drawn OAT without seeing differences in bleeding episodes
therapy in cases of minor oral surgery as defined below, given [2]. Local complications like delayed wound closure, de-
that certain precaution measures are considered. Nevertheless, creased healing capacity, increased inflammatory response,
it has to be kept in mind that minor, self-limiting bleedings and foreign-body reactions when using hemostasis-
might occur more often in patients under OAT [29, 24]. promoting materials could not be detected in the reviewed
Therefore, the major concern is a postoperative oozing rather studies. For an INR of ≤3, Campbell et al. reported that there
than profound bleeding [30]. might be no need for additional hemostatic interventions [27].
For vitamin K inhibitors, the international normalized ratio It was assumed that the use of absorbable sutures might
(INR) has to be within a safe range before surgical procedures. provoke thromboembolic events due to migration of absorb-
This “safe range” is controversial as some recommended an able particles into the blood circulation [55, 56, 1]. Similarly,
INR of ≤3 [1, 27], whereas others reported an INR of ≤4 [44, others reject the use of absorbable sutures because of their
29, 45, 27, 46, 24, 2, 47, 3, 22, 25] to be safe for dental variability of absorption and duration [23]. Evidence for such
extractions. This INR should be determined within 24–48 h dilemma is low and should be discussed critically as some
prior to the planned surgery. There seems to be no correlation groups used absorbable sutures without any complications,
between the occurrence of bleeding episodes and INR values and the removal of nonabsorbable sutures may be more trau-
within the range of 1–4 [47, 33], whereas Souto et al. reported matic than leaving the absorbable sutures in place [29, 30, 22].
a correlation between increased bleeding events in cases of Certain additional medications such as nonsteroid anti-
higher INRs [9]. It has to be noted that the usual therapeutic inflammatory drugs and antibiotics (f.e., cephalosporins,
range lies between 2 and 4, depending on the primary indica- macrolides, and quinolones) could potentially interfere with
tion for OAT [48]. An INR above 5 has shown to bear an the coagulation cascade [23, 57]. Additionally, administration
unacceptable risk for postoperative bleeding [49]. Next to of post-surgical antibiotics for more than 5 days has been
monitoring the INR, it is recommended to take a special care described to reduce intestinal vitamin K resorption leading
of patients with liver diseases and/or drugs affecting liver to increased INR values [15]. Nevertheless, there was no
function and postoperative hemostasis [3]. evidence for an effect of a macrolide antibiotic [41] or
In order to prevent postoperative bleeding events, this amoxicillin/clavulanic acid [58] on vitamin K inhibitor
systematic review underlines the efficiency of local hemostat- efficiency.
ic procedures in OAT patients [33]. The oral cavity is acces- From the data of this review, minor oral surgery procedures
sible and effective local measures can be conducted easily. are ranging from extractions of a single tooth to single dental
Wahl reported that out of 2400 dental operations in OAT implant to extractions of multiple teeth (reported up to 17
patients, local measures were not sufficient in 12 cases only. [47]), mucoperiosteal flaps, alveoloplasties, limited oral soft
In seven of these cases, OAT was shown to be above thera- tissue surgery, complicated osteotomies, and augmentation
peutic limits [36]. A variety of local hemostatic measures are procedures such as elevation of the maxillary sinus. Ward
178 Clin Oral Invest (2015) 19:171–180

and Smith classified the extent of oral surgery form low risk 4. Mani H, Lindhoff-Last E (2014) New oral anticoagulants in patients
with nonvalvular atrial fibrillation: a review of pharmacokinetics,
(1–5 simple extractions) to moderate (6–10 simple
safety, efficacy, quality of life, and cost effectiveness. Drug Des
extractions/1 impacted tooth/alveolectomy for one quadrant) Devel Ther 8:789–798. doi:10.2147/DDDT.S45644
to high risk (>10 simple extractions/>1 impacted teeth/ 5. Bump RL, Kolodny SC (1973) Fibrinolysis: a possible factor in the
alveolectomy for >1 quadrant/tori removal) [18]. More ex- control of postoperative hemorrhage in the patient with hemophilia.
Oral Surg Oral Med Oral Pathol 36(2):195–200
tended interventions such as repair of facial fractures, facial
6. Johnson WT, Leary JM (1988) Management of dental patients with
osteotomies, and bone grafts [59] should be classified as major bleeding disorders: review and update. Oral Surg Oral Med Oral
surgical procedures. For those major procedures, the authors Pathol 66(3):297–303
of this review and several other working groups recommend a 7. DeClerck D, Vinckier F, Vermylen J (1992) Influence of
anticoagulation on blood loss following dental extractions. J Dent
bridging therapy with low-molecular-weight heparin
Res 71(2):387–390
(LMWH) [24, 60] in cases of OAT with vitamin K inhibitors. 8. Ziffer AM, Scopp IW, Beck J, Baum J, Berger AR (1957)
Even though LMWHs are approved for prophylaxis of venous Profound bleeding after dental extractions during dicumarol
thromboembolism only, they are considered to be an alterna- therapy. N Engl J Med 256(8):351–353. doi:10.1056/
NEJM195702212560806
tive to IV heparin [10].
9. Souto JC, Oliver A, Zuazu-Jausoro I, Vives A, Fontcuberta J (1996)
For direct anticoagulants and platelet inhibitors, each of Oral surgery in anticoagulated patients without reducing the dose of
those cases should be discussed with the consulting primary oral anticoagulant: a prospective randomized study. J Oral Maxillofac
care physician individually informing him of differences of Surg 54(1):27–32, discussion 323
10. Johnson-Leong C, Rada RE (2002) The use of low-molecular-weight
oral surgery when compared to general surgery. It has to be
heparins in outpatient oral surgery for patients receiving
kept in mind that combination of antiplatelet drugs such as anticoagulation therapy. J Am Dent Assoc 133(8):1083–1087
aspirin and clopidogrel cause a synergistic antiplatelet action 11. Poller L, Thomson J (1964) Evidence for ′′rebound′′ hypercoagula-
with increased bleeding complications [61, 15]. bility after stopping anticoagulants. Lancet 2(7350):62–64
12. Antonio N, Castro G, Ramos D, Machado A, Goncalves L, Macedo
T, Providencia LA (2008) The debate concerning oral
anticoagulation: whether to suspend oral anticoagulants during dental
Conclusion treatment. Rev Port Cardiol 27(4):531–544
13. Garcia DA, Regan S, Henault LE, Upadhyay A, Baker J, Othman M,
Hylek EM (2008) Risk of thromboembolism with short-term inter-
The risk of potential fatal thromboembolism due to anticoag- ruption of warfarin therapy. Arch Intern Med 168(1):63–69. doi:10.
ulant withdrawal outweighs the risk of postoperative bleeding 1001/archinternmed.2007.23
episodes. Minor oral surgery procedures such as extractions of 14. Wahl MJ (2014) Dental surgery and antiplatelet agents: bleed or die.
teeth and placement of dental implants in OAT patients with Am J Med 127(4):260–267. doi:10.1016/j.amjmed.2013.11.013
15. Aldridge E, Cunningham LL Jr (2010) Current thoughts on treatment
vitamin K inhibitors can be conducted safely if the of patients receiving anticoagulation therapy. J Oral Maxillofac Surg
anticoagulation is within the therapeutic range, and local 68(11):2879–2887. doi:10.1016/j.joms.2010.04.007
hemostatic measures are used. This may also lead to a care 16. Madrid C, Sanz M (2009) What influence do anticoagulants have on
cost reduction even with a possibly decreased comfort for the oral implant therapy. A Syst Rev Clin Oral Implants Res 20(Suppl 4):
96–106. doi:10.1111/j.1600-0501.2009.01770.x
patients. Nevertheless, more minor bleedings might occur in 17. Zanon E, Martinelli F, Bacci C, Cordioli G, Girolami A (2003) Safety
OAT patients and the use of local hemostatic measures as well of dental extraction among consecutive patients on oral anticoagulant
as a close and extended postoperative monitoring is vital. treatment managed using a specific dental management protocol.
Blood Coagul Fibrinolysis 14(1):27–30. doi:10.1097/01.mbc.
0000046178.72384.16
18. Ward BB, Smith MH (2007) Dentoalveolar procedures for the
Conflict of interest The authors have no conflicts of interest in regard
anticoagulated patient: Literature recommendations versus current
of this publication.
practice. J Oral Maxillofac Surg 65(8):1454–1460. doi:10.1016/j.
joms.2007.03.003
19. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC,
References Ioannidis JP, Clarke M, Devereaux PJ, Kleijnen J, Moher D (2009)
The PRISMA statement for reporting systematic reviews and meta-
analyses of studies that evaluate health care interventions: explana-
1. Al-Mubarak S, Al-Ali N, Abou-Rass M, Al-Sohail A, Robert A, Al- tion and elaboration. J Clin Epidemiol 62(10):e1–34. doi:10.1016/j.
Zoman K, Al-Suwyed A, Ciancio S (2007) Evaluation of dental jclinepi.2009.06.006
extractions, suturing and INR on postoperative bleeding of patients 20. Miller SA, Forrest JL (2001) Enhancing your practice through
maintained on oral anticoagulant therapy. Br Dent J 203(7):E15. doi: evidence-based decision making: PICO, learning how to ask good
10.1038/bdj.2007.725, discussion 410-411 questions. J Evid Based Dent Prac 1:136–141
2. Cannon PD, Dharmar VT (2003) Minor oral surgical procedures in 21. Al-Belasy FA, Amer MZ (2003) Hemostatic effect of n-butyl-2-
patients on oral anticoagulants—a controlled study. Aust Dent J cyanoacrylate (histoacryl) glue in warfarin-treated patients undergo-
48(2):115–118 ing oral surgery. J Oral Maxillofac Surg 61(12):1405–1409
3. Devani P, Lavery KM, Howell CJ (1998) Dental extractions in 22. Bacci C, Maglione M, Favero L, Perini A, Di Lenarda R, Berengo M,
patients on warfarin: is alteration of anticoagulant regime necessary. Zanon E (2010) Management of dental extraction in patients under-
Br J Oral Maxillofac Surg 36(2):107–111 going anticoagulant treatment. Results from a large, multicentre,
Clin Oral Invest (2015) 19:171–180 179

prospective, case-control study. Thromb Haemost 104(5):972–975. patients on anticoagulants. Oral Surg Oral Med Oral Pathol Oral
doi:10.1160/TH10-02-0139 Radiol Endod 92(3):257–259. doi:10.1067/moe.2001.115463
23. Bacci C, Berengo M, Favero L, Zanon E (2011) Safety of dental 41. Kusafuka Y, Kurita H, Sakurai S, Suzuki S, Nakanishi Y, Katsuyama
implant surgery in patients undergoing anticoagulation therapy: a Y, Ohmori S (2013) Effect of single-dose extended-release oral
prospective case-control study. Clin Oral Implants Res 22(2):151– azithromycin on anticoagulation status in warfarinized patients.
156. doi:10.1111/j.1600-0501.2010.01963.x Oral Surg Oral Med Oral Pathol Oral Radiol 115(2):148–151. doi:
24. Bajkin BV, Popovic SL, Selakovic SD (2009) Randomized, prospec- 10.1016/j.oooo.2012.08.449
tive trial comparing bridging therapy using low-molecular-weight 42. Russo G, Corso LD, Biasiolo A, Berengo M, Pengo V (2000)
heparin with maintenance of oral anticoagulation during extraction Simple and safe method to prepare patients with prosthetic
of teeth. J Oral Maxillofac Surg 67(5):990–995. doi:10.1016/j.joms. heart valves for surgical dental procedures. Clin Appl Thromb
2008.12.027 Hemost 6(2):90–93
25. Blinder D, Manor Y, Martinowitz U, Taicher S (2001) Dental extrac- 43. Kosyfaki P, Att W, Strub JR (2011) The dental patient on oral
tions in patients maintained on oral anticoagulant therapy: anticoagulant medication: a literature review. J Oral Rehabil 38(8):
comparison of INR value with occurrence of postoperative 615–633. doi:10.1111/j.1365-2842.2010.02184.x
bleeding. Int J Oral Maxillofac Surg 30(6):518–521. doi:10. 44. Lippert S, Gutschik E (1994) Views of cardiac-valve prosthesis
1054/ijom.2001.0172 patients and their dentists on anticoagulation therapy. Scand J Dent
26. Bublitz R, Sommer S, Weingart D, Bauerle K, Both A (2000) Res 102(3):168–171
Hemostatic wound management in marcumar patients. Collagen 45. Weibert RT (1992) Oral anticoagulant therapy in patients undergoing
fleece vs. tranexamic acid. Mund Kiefer Gesichtschir 4(4):240–244 dental surgery. Clin Pharm 11(10):857–864
27. Campbell JH, Alvarado F, Murray RA (2000) Anticoagulation and 46. Webster K, Wilde J (2000) Management of anticoagulation in pa-
minor oral surgery: should the anticoagulation regimen be altered? J tients with prosthetic heart valves undergoing oral and maxillofacial
Oral Maxillofac Surg 58(2):131–135, discussion 135-136 operations. Br J Oral Maxillofac Surg 38(2):124–126. doi:10.1054/
28. Eichhorn W, Burkert J, Vorwig O, Blessmann M, Cachovan G, Zeuch bjom.1999.0176
J, Eichhorn M, Heiland M (2012) Bleeding incidence after oral 47. Cieslik-Bielewska A, Pelc R, Cieslik T (2005) Oral surgery proce-
surgery with continued oral anticoagulation. Clin Oral Investig dures in patients on anticoagulants. Preliminary report. Kardiol Pol
16(5):1371–1376. doi:10.1007/s00784-011-0649-1 63(2):137–140, discussion 141
29. Evans IL, Sayers MS, Gibbons AJ, Price G, Snooks H, Sugar AW 48. Hirsh J, Fuster V, Ansell J, Halperin JL, American Heart Association/
(2002) Can warfarin be continued during dental extraction? Results American College of Cardiology F (2003) American heart
of a randomized controlled trial. Br J Oral Maxillofac Surg 40(3): association/american college of cardiology foundation guide to war-
248–252. doi:10.1054/bjom.2001.0773 farin therapy. J Am Coll Cardiol 41(9):1633–1652
30. Gaspar R, Brenner B, Ardekian L, Peled M, Laufer D (1997) Use of 49. Beirne OR, Koehler JR (1996) Surgical management of patients on
tranexamic acid mouthwash to prevent postoperative bleeding in oral warfarin sodium. J Oral Maxillofac Surg 54(9):1115–1118
surgery patients on oral anticoagulant medication. Quintessence Int 50. Mulligan R, Weitzel KG (1988) Pretreatment management of the
28(6):375–379 patient receiving anticoagulant drugs. J Am Dent Assoc 117(3):479–
31. Sacco R, Sacco M, Carpenedo M, Mannucci PM (2007) Oral surgery 483
in patients on oral anticoagulant therapy: a randomized comparison 51. Rakocz M, Mazar A, Varon D, Spierer S, Blinder D,
of different intensity targets. Oral Surg Oral Med Oral Pathol Oral Martinowitz U (1993) Dental extractions in patients with
Radiol Endod 104(1):e18–21. doi:10.1016/j.tripleo.2006.12.035 bleeding disorders. The use of fibrin glue. Oral Surg Oral
32. Ramstrom G, Sindet-Pedersen S, Hall G, Blomback M, Alander U Med Oral Pathol 75(3):280–282
(1993) Prevention of postsurgical bleeding in oral surgery using 52. Bodner L, Weinstein JM, Baumgarten AK (1998) Efficacy of fibrin
tranexamic acid without dose modification of oral anticoagulants. J sealant in patients on various levels of oral anticoagulant undergoing
Oral Maxillofac Surg 51(11):1211–1216 oral surgery. Oral Surg Oral Med Oral Pathol Oral Radiol Endod
33. Blinder D, Manor Y, Martinowitz U (1999) Dental extractions in 86(4):421–424
patients maintained on continued oral anticoagulant. Oral Surg Oral 53. Sindet-Pedersen S, Ramstrom G, Bernvil S, Blomback M (1989)
Med Oral Pathol Oral Radiol Endod 88:137 Hemostatic effect of tranexamic acid mouthwash in anticoagulant-
34. Agnelli G, Becattini C (2013) Risk assessment for recurrence and treated patients undergoing oral surgery. N Engl J Med 320(13):840–
optimal agents for extended treatment of venous thromboembolism. 843. doi:10.1056/NEJM198903303201305
Hematol Am Soc Hematol Educ Program 2013:471–477. doi:10. 54. Blinder D, Manor Y, Martinowitz U, Taicher S, Hashomer T (1999)
1182/asheducation-2013.1.471 Dental extractions in patients maintained on continued oral anticoag-
35. Doraiswamy VA, Slepian MJ, Gesheff MG, Tantry US, Gurbel PA ulant: comparison of local hemostatic modalities. Oral Surg Oral Med
(2013) Potential role of oral anticoagulants in the treatment of pa- Oral Pathol Oral Radiol Endod 88(2):137–140
tients with coronary artery disease: focus on dabigatran. Expert Rev 55. Matthew IR, Browne RM, Frame JW, Millar BG (1993) Tissue
Cardiovasc Ther 11(9):1259–1267. doi:10.1586/14779072.2013. response to a haemostatic alginate wound dressing in tooth extraction
827469 sockets. Br J Oral Maxillofac Surg 31(3):165–169
36. Wahl MJ (2000) Myths of dental surgery in patients receiving anti- 56. Olson RA, Roberts DL, Osbon DB (1982) A comparative study of
coagulant therapy. J Am Dent Assoc 131(1):77–81 polylactic acid, Gelfoam, and Surgicel in healing extraction sites.
37. Kearon C, Hirsh J (1997) Management of anticoagulation before and Oral Surg Oral Med Oral Pathol 53(5):441–449
after elective surgery. N Engl J Med 336(21):1506–1511. doi:10. 57. Choi KH, Kim AJ, Son IJ, Kim KH, Kim KB, Ahn H, Lee EB (2010)
1056/NEJM199705223362107 Risk factors of drug interaction between warfarin and nonsteroidal
38. Wahl MJ (1998) Dental surgery in anticoagulated patients. Arch anti-inflammatory drugs in practical setting. J Korean Med Sci 25(3):
Intern Med 158(15):1610–1616 337–341. doi:10.3346/jkms.2010.25.3.337
39. Yasaka M, Naritomi H, Minematsu K (2006) Ischemic stroke asso- 58. Zhang Q, Simoneau G, Verstuyft C, Drouet L, Bal-dit-Sollier C,
ciated with brief cessation of warfarin. Thromb Res 118(2):290–293. Alvarez JC, Rizzo-Padoin N, Bergmann JF, Becquemont L, Mouly
doi:10.1016/j.thromres.2005.08.009 S (2011) Amoxicillin/clavulanic acid-warfarin drug interaction: a
40. Halfpenny W, Fraser JS, Adlam DM (2001) Comparison of 2 hemo- randomized controlled trial. Br J Clin Pharmacol 71(2):232–236.
static agents for the prevention of postextraction hemorrhage in doi:10.1111/j.1365-2125.2010.03824.x
180 Clin Oral Invest (2015) 19:171–180

59. Doonquah L, Mitchell AD (2012) Oral surgery for patients on anticoagulation therapy. Curr Med Res Opin 22(6):1109–1122. doi:
anticoagulant therapy: current thoughts on patient manage- 10.1185/030079906X104858
ment. Dent Clin North Am 56(1):25–41. doi:10.1016/j.cden. 61. Payne DA, Hayes PD, Jones CI, Belham P, Naylor AR, Goodall AH
2011.06.002, vii (2002) Combined therapy with clopidogrel and aspirin significantly
60. Spyropoulos AC, Bauersachs RM, Omran H, Cohen M (2006) increases the bleeding time through a synergistic antiplatelet action. J
Periprocedural bridging therapy in patients receiving chronic oral Vasc Surg 35(6):1204–1209