Hultcrantz 2004.

Study characteristics
Methods	Non‐blinded (open‐label), parallel‐group randomised controlled surgical trial with 3‐year duration of follow‐up
Participants	Setting: 3 otolaryngology clinics (1 university clinic and 2 otolaryngology clinics at local hospitals) in Sweden Sample size: Number randomised: 150 randomised (22 declined participation, 12 excluded due to exclusion criteria, 116 children included in study) Number completed: 92 children Participant (baseline) characteristics: Age: 5 to 15 years Gender: 49% males, 51% females Inclusion criteria: children on the waiting list for tonsillectomy because of obstructive problems due to tonsil hypertrophy with or without recurrent tonsillitis (approximately 60% of children in both groups had more than one episode of tonsil infection) Exclusion criteria: children who had had peri‐tonsillitis, those with small tonsils and those who no longer fulfilled the criteria for surgery
Interventions	Intervention group: tonsillectomy (TE) by cold knife and blunt dissection (n = 43) Comparator group: tonsillotomy (TT) using radiofrequency technique (n = 49) Use of additional interventions: concurrent adenoidectomy performed in 49% of children
Outcomes	Short‐term outcomes (10 days): intraoperative blood loss, postoperative bleeding events, postoperative infections, weight, anxiety, postoperative pain, dietary intake, snoring Long‐term outcomes (1 year): general health, snoring, eating difficulties, infections using the Qu1 questionnaire and the Child Behaviour Checklist (CBCL), reoperation rates, tonsil regrowth, parental satisfaction Long‐term outcomes (3 years): generic health‐related quality of life using the Glasgow Children's Benefit Inventory (GCBI), snoring, eating difficulties, number of ENT infections using the Qu1 questionnaire, reoperation rates, tonsil regrowth, parental satisfaction
Notes	Participants lost pre‐trial total: 22% pre‐trial dropout (method of Zelen; 22 declined participation, 11 excluded due to exclusion criteria) Participants lost to short‐term follow‐up: 21% dropout (24 spontaneous recovery so not operated on) Participants lost to long‐term follow‐up: 0% dropout at 1 and 3 years
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "The randomization was performed according to a modification of Zelen's method"
Allocation concealment (selection bias)	Unclear risk	Method not described
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not blinded
Blinding of outcome assessment (detection bias) All outcomes	High risk	Not blinded
Incomplete outcome data (attrition bias) All outcomes	High risk	22% pre‐trial dropout (method of Zelen; 22 declined participation, 12 excluded due to exclusion criteria) 21% loss to short‐term (10 days) follow‐up due to spontaneous recovery pre‐surgery 0% loss to long‐term (2‐year) follow‐up
Selective reporting (reporting bias)	Unclear risk	No protocol available; insufficient information to permit a judgement of low or high risk
Other bias	Unclear risk	Baseline characteristics: not balanced for gender in tonsillotomy group (M 39%:F 61%) Unclear whether ITT analysis was performed Formal sample size calculations: performed Use of co‐interventions: similar across groups; concurrent adenoidectomy performed in 44% TE and 53% TT