Customize your JAMA Network experience by selecting one or more topics from the list below.
Raggio BS, Barton BM, Kandil E, Friedlander PL. Association of Continued Preoperative Aspirin Use and Bleeding Complications in Patients Undergoing Thyroid Surgery. JAMA Otolaryngol Head Neck Surg. 2018;144(4):335–341. doi:10.1001/jamaoto.2017.3262
Do patients receiving aspirin before undergoing thyroid surgery experience increased bleeding complications?
In this cohort study of 570 adults undergoing thyroid surgery, aspirin use was not associated with an increased risk of blood loss. The association between aspirin use and complications of hematoma or recurrent laryngeal nerve injury was inconclusive because of the small sample size and infrequent events.
The results suggest that patients receiving aspirin may expect to undergo thyroid surgery without experiencing clinically meaningful increased bleeding complications, although further investigation is warranted.
No evidence exists to direct the management of preoperative aspirin (acetylsalicylic acid) use in patients undergoing thyroid surgery. Nevertheless, a considerable number of patients interrupt receiving aspirin therapy during the preoperative period to minimize bleeding complications despite the increased risk of experiencing major adverse cardiac events.
To determine whether aspirin therapy continued preoperatively increases bleeding complications in patients undergoing thyroid surgery.
Design, Setting, and Participants
Retrospective analysis of a consecutive sample of 570 patients, aged 18 to 100 years, who underwent thyroid surgery for benign and malignant disease from January 1, 2010, to December 31, 2015, by a single surgeon at a tertiary referral hospital center in New Orleans, Louisiana.
Patients receiving aspirin therapy and patients not receiving aspirin therapy (aspirin naive) preoperatively.
Main Outcomes and Measures
Comparison of estimated blood loss, substantial blood loss, operative hematoma, nonoperative hematoma, and recurrent laryngeal nerve injury.
Of 570 patients who underwent thyroid surgery, 106 (18.6%) were performed in patients receiving aspirin; of these, 23 (21.7%) were men and 105 (99.1%) were older than 45 years. Those receiving aspirin therapy displayed a 14.4-year difference in age (95% CI, 11.6-17.1). The aspirin group displayed a 20.3% absolute increase (95% CI, 9.3-30.7) in African American patients. Aspirin therapy was not associated with a statistically significant or clinically meaningful increase in intraoperative blood loss (2.5 mL; 95% CI, –0.4 to 5.3). Aspirin therapy was associated with a statistically significant increase in total hematoma formation (3.3%; 95% CI, 0.4-9.0), but the results were inconclusive. Aspirin therapy was not associated with a statistically significant increase in recurrent laryngeal nerve injury (2.6%; 95% CI, –1.1 to 8.6), but the results were inconclusive.
Conclusions and Relevance
These results suggest that aspirin therapy can be maintained prior to thyroid surgery without increased intraoperative bleeding. Further research with a larger sample size and more outcome events are required to make definitive conclusions regarding the association between aspirin use and complications, including hematoma and recurrent laryngeal nerve injury.
When patients receiving antiplatelet medication undergo elective otolaryngologic surgery, both patients and physicians must decide whether to stop therapy preoperatively and risk a possible life-threatening thromboembolic event1-6 or to continue therapy and risk an intraoperative hemorrhage and associated complications.1,7-12 Multiple guidelines outline the management of preoperative antiplatelet therapy based on a stratification risk of hemorrhage vs thrombosis complications, but the literature examining the effects of cessation or continuation of antiplatelet therapy for specific noncardiac procedures is lacking.13-21 There is no universal consensus about how antiplatelet therapy for elective otolaryngologic procedures should be treated. Nevertheless, the preoperative cessation of antiplatelet therapy is still routinely practiced for these surgical procedures.20,22 The same holds true for thyroid surgery, where a successful procedure relies on the surgeon performing precise tissue dissection through narrow operating fields. Those surgeons who routinely withhold aspirin and other antiplatelet and/or anticoagulant medications preoperatively often fear increased bleeding, which may limit exposure of tissue during the operation and result in complications, such as recurrent laryngeal nerve (RLN) injury and hematoma formation.23,24
The aim of this retrospective cohort study was to compare the incidence of bleeding complications between patients who underwent thyroid surgery while receiving aspirin therapy with those patients naive to aspirin therapy. We hypothesized that aspirin therapy would have no significant association with bleeding complications during thyroid surgery, including intraoperative blood loss, postoperative hematoma formation, and RLN injury.
This retrospective cohort study included consecutive patients 18 years or older who underwent thyroid surgery performed by a single surgeon (E.K.) at a tertiary referral hospital center between January 1, 2010, and December 31, 2015. The present study was approved by the institutional review board at the Tulane University School of Medicine in New Orleans, Louisiana. Patient informed consent was waived by the Tulane University School of Medicine given the absence of patient-identifying information.
Thyroid surgery procedures included total thyroidectomy, thyroid lobectomy, completion thyroidectomy, parathyroidectomy, and central neck dissection. No patient was excluded on the basis of thyroid disease. Diseases treated included nodular hyperplasia, multinodular goiter, Hashimito thyroiditis, Graves disease, follicular adenoma, Hurthle cell neoplasm, papillary carcinoma, follicular carcinoma, medullary carcinoma, and hyperparathyroidism. Patients undergoing either robotic or endoscopic procedures were excluded.
Intraoperative blood loss, hematoma formation, and RLN injury were measured as primary outcomes. Estimated blood loss was defined as the intraoperative blood loss as determined from a review of the operative report of the primary surgeon. A substantial blood loss was defined as more than 10 mL of intraoperative blood loss. The presence or absence of RLN injury was measured postoperatively for every patient (with or without dysphonia) by assessing vocal fold mobility via physician-directed, flexible fiberoptic laryngoscopy.
Temporary RLN injury was defined as vocal fold immobility that resolved spontaneously within 1 year from the date of surgery. Permanent RLN injury was defined as vocal fold immobility lasting longer than 1 year postoperatively.
Hematoma was defined as the development of a tender, tense, and discolored neck swelling. Clotted or collected blood was expressed or aspirated when the diagnosis was unclear. Nonoperative hematoma was defined as a hematoma treated conservatively (by observation or the expression or aspiration of blood from the wound), whereas operative hematoma described a hematoma necessitating surgical exploration. Total postoperative hematoma includes operative and nonoperative hematomas.
Patients were divided into 2 groups: patients receiving aspirin therapy (aspirin group) and patients not receiving aspirin therapy (aspirin-naive group). Patients receiving aspirin and other antiplatelet and/or anticoagulant medications were included in the aspirin group.
All collected information was recorded in a secure electronic database with strict control and supervision for data entry and access. All statistical analysis was performed using SAS, version 9.3 (SAS Institute Inc). The Fisher exact test was used for comparing categorical variables and the unpaired, 2-tailed t test was used for comparing continuous variables. Raw group differences, including comparison of means and comparison of proportions, were calculated for effect sizes and 95% CIs where appropriate. Logistic regression modeling was applied in testing for confounders and in estimating the odds ratio (OR) and 95% CI, where appropriate. The level of statistical significance was set as 2-sided α = .05.
Of 570 patients undergoing thyroid surgery who met the inclusion criteria, 464 (81.4%) were aspirin naive, while 106 (18.6%) were receiving aspirin. Of patients receiving aspirin, 23 (21.7%) were men and 105 (99.1%) were older than 45 years. Regarding the dosage of aspirin, 95 (89.6%) were receiving aspirin, 81 mg/d, and the remaining 11 patients (10.4%) were receiving aspirin, 325 mg/d. Most of the patients in the aspirin group (97 of 106 [91.5%]) were receiving aspirin alone, whereas other patients (9 [8.5%]) were receiving aspirin in combination with other antiplatelets and/or anticoagulants (aspirin and clopidogrel bisulfate [4 (3.8%)], aspirin and warfarin sodium [4 (3.8%)], and aspirin, clopidogrel, and warfarin [1 (0.9%)]) (Table 1).
Those receiving aspirin therapy displayed a 14.4-year difference in age (95% CI, 11.6-17.1). The aspirin group displayed a 20.3% absolute increase (95% CI, 9.3-30.7) in African American patients. Patients receiving aspirin underwent total thyroidectomy less frequently, corresponding to a 13.9% absolute difference (95% CI, 6.0-20.0). Sex, body mass index, thyroid-stimulating hormone levels, smoking history, alcohol use, anticoagulant use, dominant nodule size, and pathology results (benign vs malignant) were similar between the 2 groups (Table 1).
Based on the difference of means, patients taking aspirin experienced a 2.5-mL difference in estimated blood loss (95% CI, −0.4 to 5.3) (Table 1). Quiz Ref IDThe absolute difference of 2.5 mL is not statistically significant; the upper bound result of the 95% CI suggests there is no clinically meaningful difference. This is a conclusively negative finding.
The rate of substantial blood loss in the aspirin-naive group was 54 of 464 patients (11.6%) vs 14 of 106 (13.2%) in the aspirin group for a difference of 1.6% (95% CI, −5.1 to 10.0). The absolute difference of 1.6% is not statistically significant; the upper bound results of the 95% CI suggest there is no clinically meaningful difference. This is a conclusively negative finding. On logistic regression, aspirin therapy increased the risk of substantial blood loss (OR, 1.2; 95% CI, 0.6-2.2).
The rate of temporary RLN injury in the aspirin-naive group was 10 of 464 (2.2%) vs 5 of 106 (4.7%) in the aspirin group for a difference of 2.6% (95% CI, −1.1% to 8.6%). Quiz Ref IDThe absolute difference of 2.6% is not statistically significant; the upper bound result of the 95% CI suggests a clinically meaningful difference. This is an inconclusive finding. On logistic regression, aspirin therapy increased the risk of temporary RLN injury (OR, 2.2; 95% CI, 0.7-6.7). There were no cases of permanent RLN injury in either group (Table 1).
The rate of a total postoperative hematoma in the aspirin-naive group was 2 of 464 patients (0.4%) vs 4 of 106 (3.8%) in the aspirin group for a difference of 3.3% (95% CI, 0.4-9.0). The difference of 3.3% is statistically significant, but the small number of events leads to imprecision of the estimate. This is an inconclusive finding. On logistic regression, aspirin use significantly increased the risk of a total postoperative hematoma (OR, 9.1; 95% CI, 1.6-50.1).
The rate of a nonoperative hematoma in the aspirin-naive group was 2 of 464 patients (0.4%) vs 3 of 106 (2.8%) in the aspirin group for a difference of 2.4% (95% CI, –0.1 to 7.6). The difference of 2.4% is not statistically significant, and the upper bound of the CI was not clinically significant. On logistic regression, however, aspirin use significantly increased the risk of a total nonoperative hematoma (OR, 6.7; 95% CI, 1.1-40.8).
Only 1 of 106 patients (0.9%) receiving aspirin therapy developed an operative hematoma. None of the patients developed an operative hematoma in the aspirin-naive group (Table 1).
Quiz Ref IDThe small number of patients with a total postoperative hematoma (n = 6) precluded statistical analysis to identify hematoma risk factors (ie, univariate analysis).
Of the 6 patients who developed hematomas, 6 (100%) were older than 45 years, 5 (83.3%) were female, 4 (66.7%) were African American, and 4 (66.7%) were receiving aspirin therapy. Of the 6 patients who developed total postoperative hematoma, all had undergone thyroid lobectomy, the mean (SD) nodule size was 3.4 (1.2) cm, and 5 patients (83.3%) had benign pathology results (Table 2).
Given that 25% of middle-aged and elderly adults receive aspirin (18.6% in our study population), otolaryngologists frequently perform surgical procedures for patients receiving aspirin therapy.25Quiz Ref ID No evidence-based guidelines exist regarding preoperative use of aspirin therapy, yet most otolaryngologists routinely advise patients to stop receiving aspirin therapy before surgical procedures.22,26 Some physicians cancel surgical procedures if patients have received aspirin preoperatively, which jeopardizes health care efficiency and incurs undue medical costs.22,27,28 We hypothesized that aspirin use does not increase major bleeding complications in patients undergoing thyroid surgery.
In this retrospective cohort study of aspirin use and bleeding complications during thyroid surgery, our results suggest neither a statistically nor clinically significant association between aspirin use and intraoperative blood loss. However, the results for the association of aspirin with RLN injury and hematoma are inconclusive because of the small sample size and number of outcome events and do not rule out the possibility of increased rates of RLN injury and hematoma with aspirin use. Additional studies with larger sample sizes are required to answer these questions.
The literature examining the association of antiplatelet therapy with noncardiac procedures is limited.13-21 Authors supporting the safe continuation of aspirin during noncardiac procedures cite a reduced risk of major adverse cardiac events without increased bleeding complications in those patients receiving aspirin.26 On the other hand, the Burger et al1 meta-analysis of 49 590 patients reported an increased frequency of bleeding complications (median, 1.5; interquartile range, 1.0-2.5) in patients receiving aspirin during noncardiac procedures. However, they emphasized that aspirin therapy did not affect the risk of postoperative morbidity or mortality. They concluded that aspirin should be held preoperatively “only if low-dose aspirin may cause bleeding risks with increased mortality or sequels comparable with the observed cardiovascular risks after aspirin withdrawal.”1(p399) Similarly, patients receiving aspirin therapy in our study who experienced a nonoperative hematoma did not experience any increased morbidity or mortality.
Several studies support the safe continuation of aspirin therapy during various otolaryngologic and other surgical procedures anatomically related to the head and neck, including cataract surgery,29 vitreoretinal surgery,30 dental osteotomy,31 cutaneous surgery,32,33 in-office laryngologic procedures,34 and microlaryngeal surgery.35 Nevertheless, scarce evidence and contradictory guidelines offer no general consensus for preoperative management of aspirin therapy in otolaryngologic surgery.1,2
The association between aspirin and hematoma in patients undergoing thyroid surgery remains controversial. Few studies report an association between antiplatelet therapy and hematoma, including the review by Campbell et al24 of 207 postthyroidectomy patients with hematomas that reported antiplatelet use as an independent risk factor for operative hematoma (OR, 2.1; 95% CI, 1.1-4.1). The grouping of all patients receiving aspirin, warfarin, clopidogrel, nonsteroidal anti-inflammatory medications, or low-molecular-weight heparin limits the use of their findings. Moreover, the increased rate of operative hematoma in the study was largely driven by patients actively receiving clopidogrel, warfarin, or low-molecular-weight heparin more than aspirin use.24
Some authors dispute the association of aspirin therapy with hematoma formation. Lang et al36 found no correlation between antiplatelet therapy and operative hematoma, but all 22 patients in their study stopped antiplatelet therapy preoperatively. Likewise, Leyre et al37 reported on 70 patients and Burkey et al38 reported on 42 patients and found no increased risk of operative hematoma with preoperative antiplatelet use. However, it is unclear how many of their patients were actively using these medications at the time of the operation.
The findings of our study may prove useful for both patient and physician when deciding whether to withhold or to continue aspirin therapy before thyroid surgery. Regarding the association between aspirin and operative hematoma, no conclusion could be made from our study because of the small sample size (n = 1). The 0.9% incidence of operative hematoma in the aspirin group compares favorably with the documented 0.6% to 2.0% incidence of operative hematoma after thyroid surgery,23,36,39-41 supporting our hypothesis that aspirin does not increase major bleeding complications in patients undergoing thyroid surgery. Further research with a larger sample size and more outcome events are needed to investigate our findings and the safety of continuing aspirin and other antiplatelet and/or anticoagulation therapy in patients undergoing thyroid surgery. Ultimately, the management of perioperative aspirin therapy should involve a multidisciplinary approach until conclusive evidence-based guidelines are created.
Limitations of the study include a small sample size and a retrospective study design. In lieu of the recently established surgeon volume–outcome association for thyroidectomy,42 it may be inappropriate to extrapolate our findings to those surgeons who do not routinely perform thyroidectomies.
Quiz Ref IDThe difference in mean age between the aspirin group and the aspirin-naive group (mean, 64.1 vs 49.7 years) warrants discussion, particularly because multiple studies have reported older age as an independent risk factor for hematoma. Dehal et al43 report age 41 to 65 years (reference group, 18-40 years) as a risk factor for developing hematoma after thyroid surgery (OR, 1.4; 95% CI, 1.1-1.6). Godballe et al44 similarly report age greater than 50 years (reference group, <50 years) as a risk factor for postthyroidectomy hemorrhage (OR, 1.5; 95% CI, 1.2-2.0). Weiss et al23 report an increased risk of hematoma starting at age 45 years (OR, 1.4; 95% CI, 1.1-2.0) compared with the reference group of patients aged 18 to 29 years. Although most authors would agree that increasing age is a risk factor for hematoma, the lack of consensus on which age constitutes such a risk precludes any meaningful conclusions related to our study.
A higher percentage of African American patients in the aspirin group (60.4%) compared with the aspirin-naive group (40.1%) may confound bleeding complication rates. This finding is especially true because previous studies have identified African American race as an independent risk factor for hematoma formation following thyroid surgery, including Dehal et al43 (OR, 1.5; 95% CI, 1.2-1.7) and Weiss et al23 (OR, 1.4; 95% CI, 1.2-1.6), who both use the term white as the reference population. Although the etiology is not understood, one can postulate that health care disparities involving racial minorities may be a contributing factor for increased complications in this population.
The difference in procedures performed between the 2 groups is difficult to interpret since the extent of thyroid resection (hemithyroidectomy vs total thyroidectomy) and its effect on neck hematoma is inconsistent. Several studies showed an increased risk of bleeding when total thyroidectomy is performed compared with partial thyroidectomy24,35 However, other studies report that the extent of thyroidectomy was not associated with the rate of hematoma formation.45
Our results suggest that aspirin therapy can be continued before thyroid surgery without increasing intraoperative bleeding. Our study findings are inconclusive regarding the association of aspirin use with hematoma formation and RLN injury. Preoperative aspirin therapy management should involve a multidisciplinary approach until conclusive evidence-based guidelines are created. Further research involving larger sample sizes is required.
Accepted for Publication: December 23, 2017.
Corresponding Author: Blake S. Raggio, MD, Department of Otolaryngology, Tulane University Medical Center, 1430 Tulane Ave, Mailbox 8659, New Orleans, LA 70112 (firstname.lastname@example.org).
Published Online: March 1, 2018. doi:10.1001/jamaoto.2017.3262
Author Contributions: Dr Raggio had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Raggio, Barton, Kandil.
Acquisition, analysis, or interpretation of data: Raggio, Barton, Freidlander.
Drafting of the manuscript: Raggio, Barton, Freidlander.
Critical revision of the manuscript for important intellectual content: Raggio, Barton, Kandil.
Statistical analysis: Raggio, Barton.
Administrative, technical, or material support: Barton, Freidlander.
Study supervision: All authors.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Meeting Presentation: The results of this study were presented as a poster at the American Head and Neck Society Ninth International Conference on Head and Neck Cancer; July 16-20, 2016; Seattle, Washington.
Additional Contributions: Taha Hassoon, MD, assisted with data mining and Zaid Al-Qurayshi, MD, assisted with statistical analysis. Dr Hassoon and Dr Al-Qurayshi did not receive financial compensation for their work.
Create a personal account or sign in to: