Methods
We identified all patients who underwent a tonsillectomy (+/- other concurrent ENT procedures) by the main author at our local hospital. Informed consent was obtained from each participant prior to their procedure. The SQUIRE framework was used as our reporting guideline.
A two-cycle design was used for data collection, which involved the following: (1) a pre-intervention retrospective cycle covering two years and (2) a post-intervention prospective cycle covering the subsequent five-years. For each patient included in the study, the following parameters were noted:
  1. Type of procedure performed,
  2. Indication for surgery,
  3. Post-operative complications including,
  4. Primary haemorrhage,
  5. Secondary haemorrhage,
  6. Post operative infection.
We studied the management of all complications, particularly highlighting those who required further surgical intervention. Following this, we calculated the bleeding rate per annum for primary and secondary bleeds, with the intention to compare this to our subsequent data collection following implementation of our proposed change in practice for achieving and checking haemostasis.
The inclusion criteria included all dissection tonsillectomy procedures in children and adults by the main author. The exclusions included patients who had a tonsillectomy for histology due to suspected tonsillar cancer, and patients had a coblation intracapsular tonsillotomy. Patients with suspected tonsillar cancer have a potentially challenging vascular surgical field that might affect the haemorrhagic tendency and therefore skew the results. Similarly, intracapsular coblation tonsillotomy is a different procedure to dissection tonsillectomy and cannot be included into the sample.