Long-term anticoagulant therapy is currently administered to more than 6 million patients for the treatment of atrial fibrillation, deep venous thrombosis, various valve diseases, and malignancy-related syndromes.^4,5 In addition, recent advances in coronary stenting procedures have rapidly increased the need for treatment with dual antiplatelet agents.⁵ However, about 10% of these patients will undergo invasive procedures every year, which will require temporary discontinuation of the antithrombotic therapy or some bridging therapy.^4,5,21,28 Evidence-based recommendations have been proposed to deal with this complicated countermeasure in some less invasive procedures with lower risk of bleeding, such as dental procedures or cataract operations.^{1–3,6–19,21–25,29,30,32–38} However, procedures without discontinuation of antithrombotic treatment have received little attention,³⁹ and the perioperative period in neurosurgical procedures has also been less well studied.^{20,26,27,31,40–42}

The present clinical study retrospectively analyzed a series of transsphenoidal surgeries performed without discontinuation of antithrombotic therapy for comparison with the same number of transsphenoidal surgeries performed in the absence of antithrombotic therapy.

Methods

Between October 2008 and January 2014, 15 consecutive patients (11 males and 4 females; age range 51–75 years [mean 68.2 years]) with sellar and parasellar tumors were treated via a transsphenoidal approach by a single surgeon (Y.O.), accounting for 2.51% of all transsphenoidal operations during the same period. The histological diagnoses were pituitary adenoma (n = 9), Rathke's cleft cyst (n = 4), clival chordoma (n = 1), and biopsy of skull base pachymeningitis (n = 1). Eleven patients were treated with antiplatelet agents and 4 with anticoagulant agents. The prothrombin time–international normalized ratio (PT-INR) was measured preoperatively before administering anticoagulant agents, which verified that all PT-INRs were within the required limits (Table 1). The patients continued to take antithrombotic agents until the morning of surgery, and, after verification of the absence of ischemic and/or hemorrhagic complications on the morning of postoperative Day 1, administration of antithrombotic agents was restarted. All operations were performed via sublabial mucosal incisions, and the closest attention was paid to preserve the arachnoid plane at the top of the lesions in expectation of postprocedural hemostasis by compressive pressure of CSF spaces. Extended transsphenoidal surgeries through the subarachnoid spaces and giant pituitary adenomas with a maximum diameter larger than 40 mm were excluded due to the presumptive high risk of after-bleeding. Clinical data were compared with the data of 15 patients who underwent transsphenoidal surgery without preoperative antithrombotic therapy between December 2013 and January 2014. The surgical policy was explained preoperatively to the patients, and written informed consent was obtained. The overall study design was approved by the ethics committee of Kohnan Hospital. Statistical comparisons used Mini Statmate software (ATMS Co., Ltd.), and p values < 0.05 were regarded as significant.

Results

Gross-total removal of the tumor or total aspiration of the content of the Rathke's cleft cyst was achieved in 13 patients, and subtotal removal was achieved in 1 patient with a small remnant in the cavernous sinus. Aggressive removal was suspended in 1 patient because intraoperative histological examination had revealed pachymeningitis. Head CT scanning performed the morning of postoperative Day 1 revealed that most cavities of the sellae were occupied with CSF, so-called empty sellae. No patient experienced massive abnormal bleeding throughout the hospital course. Comparison of the antithrombotic agent group and the control group found no difference in preoperative tumor volume (antithrombotic agent group vs control: 2.45 ml vs 3.18 ml, p = 0.48), but patients were older in the antithrombotic agent group (63.32 years vs 57.73 years, p = 0.029). Opening of the ipsilateral cavernous sinus occurred in 4 patients in the antithrombotic agent group, but hemostasis was achieved in all patients with common techniques, such as compression with cotton flakes and/or point-by-point coagulation. In the antithrombotic agent group, bleeding varied from 100 ml to 485 ml (mean 255 ml), and no patient required transfusion, even after open cavernous sinus surgeries. Comparison of the antithrombotic agent group and the control group found no difference in intraoperative bleeding (255 ml vs 215 ml, Mann-Whitney U-test, p = 0.547). Operation time varied in the antithrombotic agent group from 114 minutes to 241 minutes (mean 167.8 minutes). Comparison of the antithrombotic agent group and the control group found no difference in operation time (167.8 minutes vs 150.0 minutes, Mann-Whitney U-test, p = 0.262) (Tables 2 and 3). All patients were discharged on postoperative Day 12 without neurological deficits.

Discussion

The risk of thrombosis has been estimated for some systemic diseases or conditions.^4,5,28 Recent venous thrombosis (within 1 month), repeated arteriovenous thrombophilia, and mitral valve disease carry higher risks of thrombosis, whereas old venous thrombosis (> 3 months) and nonvalvular atrial fibrillation involve lower risks of thrombosis.²¹ Invasive procedures also involve the risk of bleeding in patients receiving antithrombotic therapy. Orthopedic hip procedures and colon polypectomy have higher risks of bleeding, whereas dental procedures, cataract surgery, dermatological procedures, and surgery for carpal tunnel syndrome have lower risks of bleeding.⁴ The factors of thrombogenesis and fibrinolysis have been specifically considered in some invasive procedures, resulting in evidence-based recommendations for discontinuation of antithrombotic therapy or introduction of bridging therapy.^4,5,21,28 A large randomized study was reported recently in patients receiving or not receiving aspirin therapy who underwent noncardiac surgery. The study revealed that there was not an increase in cardiac or cerebrovascular events in the patients who discontinued the antiplatelet therapy compared with those who did not, although there was a slight increase in the amount of hemorrhage in the group that stayed on antiplatelet therapy.¹² However, in this study more than two-thirds of the patients had received bridging anticoagulant therapy, so the true result with or without discontinuation of antiplatelet therapy in perioperative period remains unsolved.

Very few clinical studies have considered perioperative management of neurosurgical patients receiving antithrombotic therapy except for antiplatelet therapies incorporated in intravascular treatments.^{20,26,27,31,42} Temporary discontinuation of antithrombotic therapy or introduction of bridging therapy is rarely mentioned in cases of internal carotid artery dissection,⁴⁰ as is placement of cutaneousventricular drainage for patients after intravascular treatment.⁴¹ However, these cases do not provide a high level of medical evidence.

The present cohort study was performed at a single institution and by a single surgeon. The protocol was simple, as neither discontinuation of antithrombotic therapy nor introduction of bridging therapy was required. Therefore, the treatment and control groups were easily compared. However, the true implications cannot be fully clarified, and validation of this protocol is limited only to extraarachnoid, typical transsphenoidal surgeries. Large randomized clinical trials at multiple centers are essential to recommend standardization of this management for patients receiving antithrombotic therapy. Although patients receiving antithrombotic therapy should only undergo transsphenoidal surgery when absolutely necessary, various types of procedures should be investigated to establish a high level of medical evidence about this increasing problem in the neurosurgical field.

Conclusions

Transsphenoidal surgeries were performed in 15 patients without discontinuation of antithrombotic therapy. No patient required transfusion, and intraoperative bleeding, operation time, and tumor removal rate showed no significant differences between these patients and the control group. The present study suggests that discontinuation of antithrombotic therapy may be unnecessary before the typical transsphenoidal surgery. Large randomized clinical trials are needed to establish validation of this procedure.

Author Contributions

Conception and design: Ogawa. Acquisition of data: Ogawa. Analysis and interpretation of data: Ogawa. Drafting the article: Ogawa. Critically revising the article: both authors. Reviewed submitted version of manuscript: Tominaga. Approved the final version of the manuscript on behalf of both authors: Ogawa. Statistical analysis: Ogawa. Administrative/technical/material support: Ogawa. Study supervision: Tominaga.