In spite of the possibility of triggering thromboembolic events, many professionals indicate the suspension of antiplatelet agents before dental surgical procedures. The aim of this study was to perform a quantitative assessment of intraoperative bleeding in patients on dual antiplatelet therapy. A case–control study was conducted in patients on dual antiplatelet therapy (APT group) and in patients who did not use these medications (control group). The following examinations were requested: complete blood cell count, blood coagulation tests, and platelet aggregation. The quantity of bleeding was measured intraoperatively by collection of aspirated blood. The mean volume of blood lost during the surgical procedure was 6.10 ml in the control group and 16.07 ml in the APT group ( P = 0.002). The mean volume of blood lost per minute was 0.60 ml/min in the control group and 1 ml/min in the APT group ( P = 0.001), with local haemostatic methods being sufficient to control the bleeding. There was no postoperative bleeding complication in any case. Patients on dual antiplatelet therapy presented a larger volume of bleeding, but this could be controlled by means of local haemostatic measures. Therefore, there is no need to stop either of the two dual antiplatelet therapy medications before dental extractions.
Cardiovascular diseases are one of the main causes of morbidity and mortality worldwide . Included in this group are patients with coronary artery disease (CAD), a pathology characterized by atherosclerotic plaque accumulation that leads to stenosis of the coronary arteries and thus represents an important risk factor for atherothrombotic accidents . Atherothrombotic events in patients with CAD are prevented with platelet aggregation inhibitors . The use of these may lead to increased bleeding in surgical procedures .
The platelet aggregation inhibitor acetylsalicylic acid (ASA) is the antiplatelet therapy most often used in patients with CAD because it is easily accessible, low cost, and is effective against platelet aggregation by irreversibly blocking the cyclooxygenase 1 enzyme (COX-1) and thromboxane A2 . Currently, therapy with ASA is usually combined with other medications, such as thienopyridine derivatives (e.g. clopidogrel, ticagrelor, and prasugrel), which are P2Y12 antagonists of the adenosine diphosphate receptors and act by blocking the activation of the glycoprotein GP II/III receptors on the surface of platelets, diminishing agglutination of fibrinogen during platelet aggregation .
The fear of bleeding complications, even with the use of single ASA therapy, led to various professionals recommending the interruption of such medications . However, over the years and with the publication of new scientific research studies, the recommendation for interrupting the use of ASA has been re-assessed, and this is no longer indicated .
The concomitant use of ASA and clopidogrel causes irreversible blockage of the two platelet aggregation pathways, potentiating the risk of bleeding during surgical procedures . Therefore, some professionals have again suggested interrupting the use of one of the two platelet aggregation inhibitors with the purpose of preventing bleeding complications . However, interrupting or changing antiplatelet therapy is known to predispose the patient to thrombotic and thromboembolic events that may lead to cardiovascular events with a high mortality rate .
It is indicated in the literature that despite the higher risk of postoperative bleeding in patients on dual antiplatelet therapy, there is no need to interrupt either of the two medications, because bleeding may be contained with local haemostatic measures . However, these studies presented limitations characterized by a retrospective design, opinions based on literature reviews, absence of direct evaluation of intraoperative bleeding, and lack of a control group.
Only three published prospective studies were designed specifically to investigate the association between bleeding and dual antiplatelet therapy. However, they used only the duration of alveolar bleeding to define the immediate bleeding event . Furthermore, Bajkin et al. and Lillis et al. also used the criteria proposed by Lockhart et al. for a late bleeding event, which is characterized by over 12 h of bleeding, need for the patient to return, presence of large haematomas or bruises, and need for a blood transfusion .
It is important to highlight that the duration of bleeding may not be directly related to the quantity of blood lost and the risk of a real bleeding complication. Therefore, the aim of this study was to quantitatively evaluate intraoperative bleeding during dental extractions in patients on dual antiplatelet therapy.
Materials and methods
This study was approved by the research ethics committees of the Dental School and the Medical School of the University of São Paulo. All participants read, understood, accepted, and signed a written informed consent agreement. This study was performed in accordance with STROBE guidelines for observational investigations.
A case–control study was conducted with CAD patients on dual antiplatelet therapy with ASA and clopidogrel (APT group) and patients with cardiovascular diseases taking no antiplatelet medication (control group), who required the extraction of up to three teeth. Patients who were alcoholics, or on therapy with warfarin, heparin, anti-inflammatory drugs (steroid or non-steroid), or oral thrombin and factor Xa inhibitors were excluded from the study.
To calculate the sample size, the statistical simulations proposed by Armitage and Berry were used . As there was no information from previous studies on the behavioural distribution of bleeding (mean and standard deviation) in the groups of patients on dual antiplatelet therapy (ASA + clopidogrel), it was assumed that the standard deviation in this group would be the same as that in the control group. Considering a two-tailed hypothesis test, 80% power, and a level of significance set at 0.05, it was determined that a sample size of 26 individuals was required for each group. To take account of possible losses during the research, two to four individuals were added to each group (8% to 15% of the sample size), resulting in a total estimated sample size of 52 (56 or 60) individuals.
In total, 348 patients with cardiovascular disease were evaluated. After applying the exclusion criteria, the number of subjects was reduced to 38 in the APT group and 35 in the control group.
At the first visit, all patients underwent a clinical examination and the following laboratory tests were requested: complete blood cell count, prothrombin time (results were expressed in seconds (PTs), as a ratio (PTr), and as the international normalized ratio (PT/INR)), activated partial thromboplastin time (aPTT), and platelet aggregation. The PT and aPTT were used as routine preoperative coagulation screening tests .
For the APT group, platelet aggregation was evaluated by means of a point-of-care test, VerifyNow (Accumetrics, San Diego, CA, USA). The VerifyNow test cartridges contain a lyophilized preparation of human fibrinogen-coated particles, platelet agonists (adenosine diphosphate (ADP) and arachidonic acid), buffer, and preservative. The results are expressed in P2Y12 reaction units (PRU) for the response to clopidogrel, and in aspirin reaction units (ARU) for the response to ASA.
For the control group, platelet aggregation was evaluated by turbidimetric assay. Platelet aggregation was measured in platelet-rich plasma by addition of ADP, arachidonic acid, and adrenaline agonists. Platelet-rich plasma was obtained by centrifugation of blood in a tube containing 3.8% sodium citrate at 1500 rpm for 6 min. The effects of stimulating agents were observed by monitoring the changes in platelet-rich plasma optical density by means of an aggregometer and an AggRAM platelet photometer (Helena Laboratories, Gateshead, Tyne and Wear, UK).
Different tests were used for platelet aggregation because the APT group subjects were on antiplatelet drugs and ADP-induced platelet aggregation frequently fails to detect patients with an impaired response to clopidogrel compared to a P2Y12-specific assay .
All of the laboratory examinations were conducted immediately before the surgical procedures. Anaesthesia was performed with 2% lidocaine without a vasoconstrictor (a maximum of four cartridges of 1.8 ml each), with the lower teeth being submitted to the technique of regional block complemented with local infiltration and the upper teeth with local infiltration only. The dental extractions were performed following the procedures recommended by Peterson et al. . During the procedure, the dental alveolus was irrigated with saline solution and a portable vacuum suction pump was used to aspirate the fluids (blood and saline solution). Bleeding was measured by subtracting the quantity of saline solution from the final quantity of aspirated fluid. The suction hand-piece of the dental equipment was used only to aspirate saliva, which was discarded and not considered .
Before suturing with 3–0 nylon thread, the dental alveolus was irrigated with abundant saline solution (0.9% sodium chloride), and a paste made of a macerated tablet of tranexamic acid (250 mg) mixed with saline solution was inserted into the dental alveolus. Additional local haemostatic measures were used in the case of persistent bleeding (paste of tranexamic acid placed onto the suture). The surgical procedure was considered concluded once it was observed that there was no bleeding at all. The surgical time was measured in minutes.
Standard postoperative recommendations were given to the patients after the surgical procedure. All of them were provided with analgesics (dipyrone or acetaminophen for a maximum of 3 days).
The criteria of Lockhart et al. were used to define the postoperative bleeding complications . All patients were re-assessed after 1 week, but they were instructed to return immediately in the case of bleeding.
The results were described using descriptive statistics of position and scale (mean and standard deviation, respectively) for quantitative variables and with absolute and relative frequencies for categorical variables. Comparisons between the mean values of continuous variables were performed using the Student t -test. When the supposition of normality was met or the number of cases in the subgroup analysis was very small, the Mann–Whitney test was used. Comparisons between categorical variables and the groups were evaluated using the χ 2 test. As the main outcome (bleeding), a generalized linear model was adjusted with gamma distribution and log link. Analyses were performed at a significance level of 5% and the results were obtained with the aid of the statistical software R 3.1.0 (R Foundation for Statistical Computing); the ggplot2 package was used to construct the graphs.
Eighty-eight dental extractions were performed in 73 patients with cardiovascular disease, of whom 26 (35.6%) were female and 47 (64.4%) were male. The mean age of the subjects was 57 years. The APT group subjects underwent 52 dental extractions (27 maxillary teeth and 25 mandibular teeth) and the control group subjects underwent 36 dental extractions (14 maxillary teeth and 22 mandibular teeth).
There was no statistically significant difference between the groups regarding sex, clinical condition of the teeth, or anatomical location of the teeth extracted; however the subjects differed significantly in age ( Table 1 ). Tooth sectioning and osteotomy were necessary in four cases in the APT group and in one case in the control group, with no statistically significant difference between the groups ( P > 0.05).
|APT group ( n = 38)||Control group ( n = 35)||Total ( n = 73)||P -value|
|Male||27 (71.1%)||20 (57.1%)||47 (64.4%)|
|Female||11 (28.9%)||15 (42.9%)||26 (35.6%)|
|Clinical condition of the teeth||0.183|
|Caries||8 (21.1%)||10 (28.6%)||18 (24.7%)|
|Periodontal disease (PD)||5 (13.2%)||10 (28.6%)||15 (20.5%)|
|Residual root (RR)||24 (63.2%)||15 (42.8%)||39 (53.4%)|
|RR + PD||1 (2.6%)||0 (0%)||1 (1.4%)|
|Anatomical location of the teeth||0.073|
|Age (years), mean ± SD||61.9 ± 10.9||51.2 ± 17.7||57 ± 15||0.002|
With regard to the complete blood cell count, there was no difference between the groups in the results, with the exception of RDW (red cell distribution width), which was found to be increased in 62.9% of the control group patients compared to 18.4% of the APT group patients ( P < 0.001).
In the blood coagulation tests, statistically significant differences were observed between the APT and control groups for PTr, PTs, PT/INR, and aPTT (all P < 0.001) ( Table 2 ). As six patients in the control group presented altered values of PT and aPTT after the dental surgery procedures, they were referred to a haematologist, who concluded that the patients had hepatic disorders.
|APT group ( n = 38)||Control group ( n = 35)|
|PLT (cells/mm 3 )||149,000||406,000||244,600||59.40||64,000||644,000||270,700||123.70|
b Used for the control group patients: the number of patients who presented with hypoaggregability (hypo), normal aggregability (normo), and hyperaggregability (hyper), according to the reference values, for each of adenosine diphosphate (ADP), arachidonic acid, and adrenaline agonists.
With regard to the main outcome, the volume of blood lost during the surgical procedure was greater in the APT group (16.07 ml) than in the control group (6.10 ml) ( P = 0.002). The duration of the surgical procedure was longer in the APT group (mean 15 min, range 5–50 min) than in the control group (mean 11 min, range 5–30 min) ( P = 0.036). The mean volume of blood lost per minute was calculated as 0.6 ml/min in the control group and 1.0 ml/min in the APT group ( P = 0.001) ( Table 3 and Fig. 1 ). Of note, the maximum bleeding in the APT group was 2 ml/min and the maximum surgical time was 50 min.
( n = 38)
( n = 35)
( n = 73)
|Mean volume of bleeding (ml)||16.07||6.10||11.3||0.002|
|Mean surgical time (min)||15||11||13||0.036|
|Mean volume of blood lost per minute (ml/min)||1.00||0.60||0.80||0.001|