| Sign In to gain access to subscriptions and/or personal tools. |
Aspirin Use and Post-operative Bleeding from Dental Extractions
1 Department of Oral Medicine and Correspondence: * corresponding author, mike.brennan{at}carolinas.org
Aspirin is a common, chronically administered preventive treatment for cardiovascular disease, but is often discontinued prior to invasive dental procedures because of concern for bleeding complications. We hypothesized that aspirin does not cause increased bleeding following a single tooth extraction. Thirty-six healthy persons requiring a tooth extraction were randomized to receive 325 mg/day aspirin or placebo for 4 days. Cutaneous bleeding time (BT) and platelet aggregation tests were obtained prior to extraction. The primary outcome measure, oral BT, and secondary bleeding outcomes were evaluated during and following extraction. No significant baseline differences, except for diastolic blood pressure, were found between groups. There were no differences in oral BT, cutaneous BT, secondary outcome measures, or compliance. Whole-blood aggregation results were significantly different between the aspirin and placebo groups. These findings suggest that there is no indication to discontinue aspirin for persons requiring single-tooth extraction.
Key Words: aspirin • bleeding • clinical trial • dental extraction
Aspirin is commonly recommended for the prevention of thrombo-embolic events, including myocardial infarction (MI) and stroke (Stafford et al., 2005), and is recommended for individuals with diabetes, who are at risk for cardiovascular disease (CVD) (Colwell and American Diabetes Association, 2004). Prophylactic aspirin use in the United States has been estimated at 33% for high-risk individuals (e.g., coronary artery disease, MI, stroke, or peripheral vascular disease), 16% for persons with multiple CVD risk factors, and 12–49% for persons with diabetes (Persell and Baker, 2004; Stafford et al., 2005). Aspirin works by irreversibly inhibiting platelet function and cyclo-oxygenase type 1 (COX-1) through a selective acetylation of human COX-1, lasting for the life of the platelet (approximately 10 days). Aspirin irreversibly inactivates COX-1 activity by binding to the active site of the enzyme at the arginine120 residue and acetylating the serine529 residue. Acetylation prevents arachidonic acid from gaining contact with Tyr385, which is the normal first step in its cyclo-oxygenation (Patrono and Rocca, 2007). Aspirin is a 150- to 200-fold more potent inhibitor of COX-1 than COX-2, and COX-1 is sensitive to low doses of aspirin (40–80 mg daily) (Patrignani et al., 1982; FitzGerald et al., 1983). Inhibition of collagen-induced platelet aggregation is optimal at a daily aspirin dose of 160 mg (Gan et al., 2002). Case reports of bleeding following dental procedures have appeared in the literature (Thomason et al., 1997). Clinical trials of aspirin have been contradictory, showing both increased bleeding and no evidence of excessive post-operative bleeding following dental procedures (Ardekian et al., 2000; Schrodi et al., 2002). Discontinuation of aspirin is often suggested prior to invasive dental procedures, but no clear guidelines exist for dose alteration (Little et al., 2002). Given the goal for aspirin in these clinical conditions, discontinuation of aspirin must be weighed against the potential for significant bleeding during or following an invasive dental procedure if aspirin is continued (Lockhart et al., 2003b). The study’s purpose was to determine if aspirin is associated with increased bleeding during and following a tooth extraction. We hypothesized that aspirin would not lead to increased bleeding compared with a placebo following a dental extraction.
We enrolled 36 healthy persons requiring a single-tooth extraction and with an American Society of Anesthesiologists (ASA) physical status classification of  2 (Keats, 1978). We excluded persons on warfarin, heparin, steroids, or non-steroidal anti-inflammatory drugs, as well as those with systemic conditions with a potential for bleeding, including liver or kidney disease and acquired or congenital bleeding disorders (Lockhart et al., 2003a,b). Persons with a history or current use of alcohol of > 2 drinks/day for > 2 yrs were also excluded. Participants were randomized by a computer-generated table to receive either aspirin (325 mg/day) or an identical-appearing placebo for two days prior to extraction, and for two days following the extraction, for a total of four consecutive days. Institutional Review Board approval and written consent were obtained.
In addition to demographic data and vital signs (blood pressure and pulse), a pre-operative dental examination, radiographs, blood studies, and cutaneous bleeding time test were performed. The dental examination evaluated overall oral hygiene level (on a scale of 1–4) (Silness and Löe, 1964) and dental disease for the tooth to be extracted, which included: mobility (0–3) (Armitage, 1990), probing depth in mm (mean value from 6 probing depths around the tooth), recession (0–3), and gingival color (0–3) (Table 1
Whole-blood impedance aggregation tests (collagen at 2.0 µg/mL and 5.0 µg/mL and 0.5 mM of arachidonic acid) were conducted by means of a ChronoLog aggregometer (Havertown, PA, USA). The low collagen concentration of 2 µg/mL will detect an aggregation defect by aspirin, while the higher concentration of 5 µg/mL can bypass the aspirin effect. Aspirin will eliminate the electrical impedance aggregation response to arachidonic acid. An aliquot of whole blood (0.5 mL) was diluted with an equivalent volume of isotonic saline and incubated for 5 min at 37°C. The electrical impedance for a control, not exposed to non-steroidal anti-inflammatory medications for 7 days, was monitored until a stable baseline was established. The aggregation reagents were added to separately incubated 1-mL samples and monitored for 6 min, and the final increase in ohms over this period was displayed as a numeric readout. The individuals were anesthetized with a single carpule of 1.8 mL of 2% lidocaine with 1:100,000 epinephrine, and with 3% carbocaine without epinephrine if further anesthesia was required. The tooth was extracted and the extraction time recorded. To record the intra-oral bleeding time (primary outcome), we observed the extraction site for 2 min without gauze in place. At the end of 2 min, gauze was placed over the extraction site to remove blood extending beyond the tooth socket. After blotting the extraction site, we recorded as a positive test result any bleeding that extended beyond the crest of the socket (i.e., onto surrounding gingival tissues) during an observation period of 1 min. This same procedure was performed at 5, 8, 11, 14, and 20 min after the extraction. Intra-oral bleeding time was defined as the length of time for bleeding to cease to extend beyond the tooth socket. Thus, if bleeding was noted at 2, 5, 8, and 11 min, but not at the 14-minute interval, the bleeding time was recorded as 11 min. Our previous work demonstrated a correlation of the intra-oral bleeding time with other measures of post-operative bleeding (Brennan et al., 2002). Consequently, the clinicians performing the extraction and those recording the intra-oral bleeding time were blinded to the cutaneous bleeding time and the whole-blood aggregation results. Bleeding complications were also assessed by telephone with standardized questions at 3-7 hrs (first follow-up, FU1) and 40–55 hrs (second follow-up, FU2) following extraction, including the following: length of time (hrs) bleeding continued after the individual left the dental clinic; whether bleeding occurred after the individual left the dental clinic (y/n); whether the extraction site was bleeding at the time of the telephone contact (y/n); and whether the gauze needed to be replaced due to bleeding (y/n) (Brennan et al., 2002). Each person received oral and written post-operative instructions per clinic protocol. Acetaminophen plus hydrocodone was prescribed as needed for pain, and the individuals were instructed not to take any other pain medications. Compliance was assessed with a pill count the day of the extraction and at FU2.
Statistical Analysis and Sample Size Estimation
Thirty-six individuals were enrolled from 5/2003–12/2005 (mean age, 40.3 ± 10.4 yrs; 19 male), with 17 randomized to aspirin and 19 to placebo. No differences were noted between groups in baseline information (age, sex, race, tobacco and alcohol use, and oral disease parameters), except for diastolic blood pressure (Table 1  ). The aspirin group had a higher diastolic pressure than the placebo group, although both values were within the normal range (i.e., < 90 mm Hg), and the difference was judged not to be clinically significant. Additionally, there were no differences in extraction time, difficulty of the extraction, and location of extraction sites between groups (Table 1). All participants randomized had an extraction, oral bleeding time, and cutaneous bleeding time completed. For the first follow-up telephone contact (FU1), we could not reach two persons (12%) in the aspirin group, but we contacted all individuals in the placebo group. For the FU2, we could not reach three people (18%) in the aspirin group and two (11%) in the placebo group.
As expected, whole-blood aggregation results were significantly different between the aspirin and placebo groups: arachidonic acid = 0 ± 0 mM and 4.7 ± 4.0 mM, respectively (p < 0.0001); collagen 2.0 µg/mL = 4 ± 4.2 and 8.7 ± 4.9 (p = 0.004); and collagen 5.0 µg/mL = 8.6 ± 4.0 and 13.1 ± 3.9 (p = 0.001) (Table 2
No significant differences were found for the primary outcome of oral bleeding time between the aspirin and placebo groups: 7.2 ± 5.9 and 5.8 ± 6.2 min, respectively (p = 0.51). There also was no difference in duration of bleeding following extraction between the aspirin and placebo groups: 3.5 ± 2.9 and 5.2 ± 6.3 hrs, respectively (p = 0.31), or with intra-operative or post-operative bleeding outcomes at FU1 and FU2 (Table 2 ). Medication compliance was 100% for both groups at the day of extraction and at the FU2. No adverse events were reported for either group. Results with arachidonic acid, collagen 2.0 µg/mL, and collagen 5.0 µg/mL had no significant correlation with the cutaneous bleeding time. A negative correlation was found between the oral bleeding time and collagen 5.0 µg/mL (r = –0.34, p=0.04), while no correlation was identified between oral bleeding time and arachidonic acid or collagen 2.0 µg/mL.
Aspirin irreversibly interferes with platelet function and lasts for the lifetime of the platelet, approximately 8–11 days (Vane et al., 1990). As little as 81 mg of aspirin will have this effect (Ryan et al., 1990), with a greater impact on platelet aggregation at higher doses such as 325 mg (Feng et al., 1997; Malhotra et al., 2003). Since 2 doses of aspirin will cause maximal platelet inhibition, we used 2 doses pre-operatively, leading to significant aggregometer results (May et al., 1997; Coleman and Alberts, 2006). Earlier case reports suggest oral bleeding episodes from aspirin, but whether directly related to aspirin or to other factors is unclear (Thomason et al., 1997). More recently, prospective clinical trials have examined the impact of aspirin on post-extraction bleeding. In a randomized study of 39 persons on 100 mg aspirin /day, group 1 continued aspirin through dental extractions, while group 2 stopped aspirin 7 days prior to extractions and resumed the day after surgery. The cutaneous bleeding time was higher in group 1, but within the normal range for both groups. No episodes of excessive intra-or post-operative bleeding were found in either group (Ardekian et al., 2000). Similar results were found in 51 individuals taking low-dose aspirin (75–100 mg/day) prior to dental extractions (Madan et al., 2005). Two studies have evaluated bleeding on probing (BOP) for people taking aspirin. The first study randomized 46 persons to placebo, 81 mg aspirin, or 325 mg aspirin (Schrodi et al., 2002). The people with baseline gingivitis (> 20% BOP sites) randomized to 325 mg aspirin had an increase in BOP compared with the placebo group. In a similar study, significant differences in BOP between placebo and both 81 mg aspirin and 325 mg aspirin were identified (Royzman et al., 2004). Despite the absence of a clear link between aspirin use and significant bleeding following invasive dental procedures, there are still recommendations to delay invasive procedures for a minimum of 3 days following cessation of aspirin, until a time when the numbers of new, functional platelets have returned to a sufficient level (Little et al., 2002). The problems associated with this practice include a delay in treatment for emergent odontogenic problems, the loss of the anti-thrombotic benefit of aspirin when it is prescribed for cardiovascular disease, and unnecessary time and expense for unwarranted bleeding time testing (De Rossi and Glick, 1996). Although aspirin can increase the cutaneous bleeding time test result, and one study showed an increase in blood loss in persons taking aspirin (Goldman et al., 1988), the literature suggests that increased cutaneous bleeding time does not predict increased blood loss from surgery (Amrein et al., 1981; Barber et al., 1985; Brennan et al., 2002). Platelet aggregation tests are also affected by aspirin, but once again the predictive power of this test for intra-operative or post-operative bleeding has not been well-documented (Ryan et al., 1990). In the present study, the cutaneous bleeding time was not correlated with any measures of bleeding outcomes, while a weak negative correlation was found between the platelet aggregation test, with collagen 5.0 µg/mL, and the primary outcome measures of oral bleeding time. Several reports in the literature have discussed the advantages and disadvantages of different ways of managing anti-thrombotic medication before invasive dental procedures. Some articles recommend withholding medication, without consideration for the possible negative outcomes (Saour et al., 1994). There are well-documented examples of inappropriate discontinuation of anti-thrombotic therapy prior to dental procedures (Akopov et al., 2005). In particular, people taking warfarin, within the appropriate therapeutic range, are often told to discontinue therapy prior to invasive dental procedures, which can have a devastating outcome (e.g., cerebral stroke) (Wahl, 1998; Evans et al., 2002). Although the anti-thrombotic mechanism is different between aspirin and warfarin, the risk vs. benefit of discontinuing aspirin for an invasive dental procedure also greatly favors continuing anti-platelet therapy (Brennan et al., 2007). A limitation of the present study includes the small sample size of 36. Although the primary outcome of oral bleeding time was based on a previous study (Brennan et al., 2002), a type II error with cutaneous bleeding time is possible with a p value of 0.07. Statistically significant differences might have been identified with a larger sample size. However, the oral and cutaneous bleeding time results found in the current study were similar to those found in a previous study utilizing the same outcome measures from which the sample size estimate was derived (Brennan et al., 2002). A wide variance was noted in the extraction times in the present study, with an extraction for one person in the placebo group taking 2134 sec (i.e., 35.6 min). In a study with a small sample size, as in the present study, even a single outlier (as was the case in this study) can exert considerable influence. In our opinion, including this individual and making the appropriate non-parametric adjustment was the best choice to conserve sample size and produce a sound interpretation. This study is the first randomized, double-blind, placebo-controlled study to evaluate the impact of a higher dose of aspirin (325 mg) in persons requiring single-tooth extraction. The impact of aspirin use on post-operative bleeding from multiple extractions and the use of other anti-platelet medications on post-operative bleeding have yet to be evaluated. These findings suggest that there is no indication for discontinuing aspirin for individuals requiring a single-tooth extraction, or similarly invasive dental procedures.
Funding was provided by The Health Service Foundation, Carolinas HealthCare System, Charlotte, NC. ChronoLog (Havertown, PA, USA) supplied the Aggregometer and agonists required for reactions. The preliminary results of this study were presented at the Annual Meeting of the American Academy of Oral Medicine, San Juan, Puerto Rico, May, 2006. Received for publication December 21, 2006. Revision received February 26, 2008. Accepted for publication May 28, 2008.
Journal of Dental Research, Vol. 87, No. 8,
740-744 (2008) This article has been cited by other articles:
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
