|Home | About | Journals | Submit | Contact Us | Français|
Post-thyroidectomy bleeding is a low frequency but potentially life threatening event that is very difficult to predict. Given the increasing drive towards thyroidectomy with same day discharge, this study was conducted with the aim of identifying patterns, timing and consequences of post-thyroidectomy bleeding to assess the feasibility of day-case thyroidectomy.
All patients who underwent a thyroidectomy between 2008 and 2015 at our institution were identified. Patterns, timing and consequences in all those who developed post-thyroidectomy bleeding were studied.
Of the 805 patients included in the study, 14 required re-exploration for bleeding; 7 (50%) of these within 8 hours of surgery, 6 (43%) between 18 and 30 hours, and 1 (7%) at 49 hours. Just under half (43%) of those with post-thyroidectomy bleeding had thyrotoxicosis.
A significant number of postoperative haemorrhages occurred beyond the immediate postoperative period. Same day discharge after thyroidectomy cannot therefore be recommended as a routine practice.
Post-thyroidectomy haemorrhage can be a life threatening complication of thyroid surgery. It has a reported incidence of 0.1–2.1%.1–3 While a variety of factors (eg postoperative hypertension, use of anticoagulants and extent of dissection) have been implicated in increasing the risk of post-thyroidectomy haemorrhage, it is generally considered a low frequency stochastic event, which is difficult to predict.4–7 Bleeding deep to the strap muscles is associated with a higher risk of life threatening airway swelling and obstruction, and the risk of post-thyroidectomy haemorrhage is highest in the early hours following surgery.8 For these reasons, patients traditionally remained in hospital for 72 hours or longer after thyroidectomy. There is an increasing drive towards delivering efficient and cost effective care, and reducing the length of hospital stay is one of the main areas for improvement.
Many centres (especially in the US) perform day-case thyroidectomy and have reported good outcomes and safe care.9–21 In the UK, the British Association of Day Surgery included thyroidectomy as a potential day-case procedure in 2001 but the 2011 consensus statement of the British Association of Endocrine and Thyroid Surgeons (BAETS) was more cautious, recommending that thyroidectomy should be a ‘day-and-stay’ procedure with a minimum length of stay of 23 hours.22 This study examined the pattern and timing of post-thyroidectomy haemorrhage in a large interdisciplinary district hospital to identify potential risk factors for this complication as well as the extent to which its impact on preventing short stay thyroid surgery may be mitigated.
All patients who underwent thyroidectomy by general and ear, nose and throat surgeons at Barking, Havering and Redbridge University Hospitals NHS Trust between 2008 and 2015 were studied retrospectively. Information about age, sex, ASA (American Society of Anesthesiologists) grade, indication for surgery, nature of surgery, bleeding risks, use of a Harmonic® scalpel (Ethicon, Somerville, NJ, US), duration of the procedure and occurrence of post-thyroidectomy haemorrhage was obtained from patient records. For those patients who had post-thyroidectomy haemorrhage, the location of the bleeding in relation to strap muscles, whether bleeding was from an identified vessel or whether it was a general ooze and whether there was any difficulty in tracheal intubation or any airway related complication at the time of neck re-exploration were recorded. The time interval from the end of the primary procedure to re-exploration was recorded. The relationship between duration of primary surgery and time to neck exploration was ascertained using simple correlation. Statistical analysis was performed with MedCalc® (MedCalc Software, Mariakerke, Belgium).
Over the study period, 805 patients underwent thyroidectomy. The mean age at surgery was 52 years (standard deviation [SD]: 13 years) and 80% of patients were female. There were 449 total thyroidectomies, 59 completion thyroidectomies and 297 hemithyroidectomies. Surgical drains were inserted in all patients. Seventeen patients required a return to theatre. The reason was post-thyroidectomy haemorrhage in 14 cases and in 3 cases, it was to drain a neck abscess.
The incidence of this complication in our series was 1.7%. There were four male and ten female patients, and the mean age of patients requiring treatment for post-thyroidectomy haemorrhage was 57 years (SD: 17 years). The sex and age distribution was not significantly different from the overall patient population.
There were no unexpected events during any of the procedures except for division of sternohyoid muscle in one patient to facilitate access. The indication for surgery was multinodular goitre (n=7, including 3 toxic multinodular goitres), toxic thyroid nodule (n=3), Hashimoto’s thyroiditis (n=2), Graves’ disease (n=1) and completion thyroidectomy for papillary thyroid cancer (n=1). Four patients had ASA grade 1, nine patients had ASA grade 2 and one patient had ASA grade 3. Only one patient had been taking aspirin and this was stopped ten days before surgery. A Harmonic® scalpel had been used in eight patients. There were eight total thyroidectomies and six hemithyroidectomies. All but one of the procedures were performed by high volume thyroid surgeons and the one operation that was undertaken by a senior trainee was performed under direct supervision of a consultant.
There were three temporal bleeding patterns. Half of the cases occurred within eight hours of surgery. None of these patients had a surge in blood pressure in the immediate postoperative period. A second bleeding cluster of six cases occurred between 18 and 30 hours, and one patient bled at 49 hours (Fig 1). The three cases of bleeding beyond 24 hours occurred in patients who were treated for thyrotoxicosis and the nature of the bleeding was superficial ooze in all of these patients.
There were eight cases of potentially dangerous bleeding deep to the strap muscles. In three of these patients, the wound needed to be opened on the ward to manage an expanding haematoma and avert acute airway obstruction. In one of these cases, significant intubation difficulty was encountered owing to acute pharyngolaryngeal oedema but no long-term harm resulted. The source of bleeding was arterial in one patient and venous in two patients. In the remaining six cases, bleeding was superficial to the strap muscles (Fig 2). There were no deaths or permanent harm resulting from post-thyroidectomy haemorrhage in this series.
Post-thyroidectomy bleeding is a potentially life threatening complication. Early detection, proactive airway management, haematoma evacuation and control of bleeding points are crucial to prevent escalation into death or permanent injury due to airway obstruction. The incidence of post-thyroidectomy bleeding in our series was 1.7%, which is in line with published reports.1–3 Half of the haemorrhages occurred within 8 hours of surgery and the majority occurred within the first 24 hours of surgery. These findings are in comfortable alignment with the BAETS recommendations to consider thyroidectomy a 23-hour procedure.22 Based on the observation that a significant number of haemorrhages occur beyond the immediate postoperative period, we have reservations about routine day-case thyroidectomy although it may be feasible in selected cases.
A number of studies have attempted to identify predictors for post-thyroidectomy haemorrhage. Revision thyroidectomies were noted to be associated with a higher risk of bleeding, and data from the Swedish thyroidectomy registry suggest that male sex and older age are risk factors for this complication.2 These findings were echoed in a study by Promberger et al, who identified male sex, older age, extent of resection, bilateral procedures, recurrent disease and surgeon experience as risk factors for post-thyroidectomy haemorrhage.23 Lee et al noted that nine of their ten cases of post-thyroidectomy haemorrhage occurred in patients with papillary thyroid cancer.24
On the other hand, a study by Burkey et al5 and subsequent works2,4 have suggested that post-thyroidectomy haemorrhage is a low frequency stochastic event that cannot be reliably predicted using preoperative and surgical variables. This observation is in keeping with our findings. There were no significant differences in age, sex, pathology or extent of surgery between patients who did and those who did not have post-thyroidectomy bleeding. All patients who bled beyond 24 hours had surgery for thyrotoxicosis and their operations took longer than 1.5 hours although given the small numbers (n=3), the credence of this finding is somewhat limited.
A surgical bed in the UK costs in the region of £350–£430 per day22 and day-case thyroidectomy would bring significant savings. However, in addition to the incalculable human cost associated with death or major injury following surgery, from a financial perspective, one excess death or major injury following day-case thyroidectomy would eliminate the monetary benefit from almost 1,000 saved bed days.22 Given the incidence, unpredictability, timing and potential consequences associated with post-thyroidectomy bleeding, our findings support the BAETS consensus statement that achieving a safe outcome requires patients to be monitored for at least 24 hours following thyroidectomy.
Post-thyroidectomy bleeding cannot be predicted using preoperative or surgical variables and the majority of these cases occur within 24 hours of surgery. We cannot recommend routine day-case thyroidectomy with same day discharge. Conversely, day surgery with an overnight stay and at least 24 hours of postoperative observation in the hospital is a practical option.