LAUSR

Dear Editor, We read with interest Edafe’s paper on haematoma in neck endocrine surgery [1]. Retrospective analysis of our data from the Royal North Shore Hospital in Sydney, Australia, over the preceding 3 years, shows 12 significant haematomas (defined as requiring reintervention or resulting in death) in 4663 patients (0.3%) undergoing thyroidectomy or parathyroidectomy (3193 and 1470, respectively). Nine occurred after thyroidectomy (75.0%) and 3 after parathyroidectomy (25%) (p = 0.63). Reintervention was required within 12 h of surgery in 3 patients (25.0%), 24 h in another 3 (25.0%), at 3 days in 1 (8.3%), and 5 days or more in 5 (41.7%, range 5–11). There was one mortality due to a haematoma which occurred out of hospital at day 10 and resulted in cardiorespiratory arrest. Two of the 12 patients (16.7%), including the mortality, usually took anticoagulants (one warfarin and one novel direct oral anticoagulant, DOAC). These were ceased in advance of surgery and restarted in accordance with local guidance and in conjunction with the patients’ cardiologist/haematologist. These haematomas both occurred late, at 6 days in the patient on warfarin (evacuated) and 10 days after operation in the patient on a DOAC, who died before reaching hospital. Edafe’s data show no significant difference in the risk of haematoma formation for those on anticoagulants, in agreement with a previous study of 70 haematomas in nearly 7000 thyroidectomies [2]. However, other studies have demonstrated a significantly increased risk [3, 4], including one of 22 haematomas in 4500 thyroid and parathyroidectomies (0.5%) in which the risk of haematoma in those on oral anticoagulation was 2.2%, injected agents 10.7% compared to those not on any of 0.4% (p\ 0.01) [4]. The decision-making involved in the peri-operative management of anticoagulation involves balancing the risk of bleeding with that of thromboembolic events due to cessation of the anticoagulant, to determine the duration of anticoagulant omission and whether bridging therapy is required. This balance can be extremely difficult: we applaud the authors’ whose approach has not resulted in an elevated rate of haematoma in anticoagulated patients and would be interested to know if they have observed any thromboembolic events. One in our cohort of 4663 sustained a thrombotic cerebrovascular accident after thyroidectomy, during the short period of cessation of a DOAC. Although good general guidance exists on quantifying the risks of bleeding and thromboembolism to inform perioperative planning, this has not been specifically or satisfactorily addressed in neck endocrine surgery and wide variation in practice in those undertaking neck endocrine surgery has been reported [5]. We agree with Edafe that further study is required to stratify the risks of haematoma in neck endocrine surgery and would add that a specific focus on those who are anticoagulated is required, addressing not only their bleed risk but also that of thromboembolic events. In the meantime, recognition of these risks in anticoagulated patients is essential in order to allow the surgeon to modify their threshold for surgery and to allow a full and frank preoperative discussion with the patient. & Aimee N. Di Marco [email protected]