Peritonsillar abscess (PTA) has a relatively high incidence of 41 per 100,000/year in Denmark. In spite of that, there is no regional or national consensus on treatment of PTA. Abscess drainage can be done by aspiration, incision or acute tonsillectomy. Several studies show that incision and aspiration are equally successful. The aim for this study is to compare aspiration to acute tonsillectomy (tonsillectomy a chaud) in a RCT study regarding sick-leave days, days of admission, pain, consumption of antibiotics, consumption of painkillers and patients´ self-assessed quality of life.

Background Peritonsillar abscess (PTA) is a complication of acute tonsillitis. The condition occurs when the infection spreads from the tonsil to the peritonsillar tissue, resulting in accumulation of pus between the tonsil and the muscles of the throat.

PTA is a common condition with an incidence of 41 per 100.000/year in Denmark. Untreated, PTA may develop into a potentially fatal condition with respiratory problems due to para- or retrofaryngeal abscess or spread of infection to surrounding tissue with necrotizing fasciitis, sepsis and thrombophlebitis of the jugular vein (Lemierres syndrome).

In spite of the high incidence of PTA, there is no regional or national consensus on the treatment of the PTA. There are several methods used for the treating of PTA. Draining of the abscess can be done by puncture and aspiration, incision or immediate abscess tonsillectomy (tonsillectomy a chaud). Additional treatment may consist of antibiotics, analgesics and rehydration. Depending on the general condition patients are either hospitalized for treatment or receive outpatient treatment (aspiration or incision).

Acute surgical intervention (tonsillectomy a chaud) is relevant in children, adults who cannot cooperate to aspiration, if malignancy is suspected in the patient, patients who experience insufficient effect from treatment with aspiration and antibiotics, patients with recurrent PTA or history of recurrent tonsillitis. Based on these criteria, a previous published article reported a relative indication for immediate abscess tonsillectomy in approximately 30% of patients with PTA. The advantage of surgery is that the abscess cavity cannot reform and that, in many cases, antibiotic treatment may be discontinued after the operation. The risk of surgery is primarily post-operative hemorrhage. The hemorrhage rate after immediate abscess tonsillectomy is not significantly higher compared to elective tonsillectomy.

Alternative treatment options are puncture and aspiration of pus or incision and dilatation. Benefits of these treatments are that the patient usually can be treated in outpatient clinics, and avoid surgery in general anesthesia. Good patient compliance is required as well as concomitant antibiotic treatment. Incision treatment has not been found more successful compared to aspiration treatment (NNT=48).

Several studies have compared immediate abscess tonsillectomy to puncture and aspiration and reported high success rates (82-95%) for aspiration. However, the risk of aspiration or incision is that the PTA will reform, necessitating subsequent tonsillectomy. The PTA reformation risk is highest in patients with a history of recurrent tonsillitis.

A prospective study including 53 patients hospitalized with PTA, has reported significantly shorter hospital stay for patients treated with tonsillectomy compared to aspiration, but no randomized studies have been conducted investigating differences in duration of hospital stay or days absent from work between treatments. Furthermore, there is a lack of studies investigating patient reported outcomes such af quality of life and satisfactionwith treatment.

The purpose of this study is to evaluate the effect of tonsillectomy a chaud versus needle aspiration for PTA. The primary outcome is number of sick-leave days and days spent in hospital. Secondary outcomes include consumption of antibiotics, consumption of painkillers, VAS score and patients' self-assessed quality of life in relation to treatment. The results of the study will improve information and advice given to patients with PTA as well as aid clinicians in choosing the best treatment.

Materials and Methods The study is designed as a prospective randomized non-blinded controlled multicenter trial. Patients will be enrolled from Departments of ENT-HNS in Southern Denmark. Inclusion criteria are adult patients (age 18 years or older) with PTA. Patients with respiratory distress, suspected malignancy, previous PTA, deep neck infection or history of recurrent tonsillitis and patients with poor compliance to aspiration will be excluded. Criteria are chosen based on current clinical guidelines.

PTA is defined by presence of pus, confirmed by 3-point aspiration. Aspiration can be performed under local anesthesia with xylocaine spray or injection of local anesthetic. Puncture and aspiration will be done from three predefined points. The patient is eligible for inclusion in the presence of pus and is randomized to either needle aspiration or tonsillectomy a chaud.

Patients treated with aspiration will receive standard antibiotic treatment according to clinical guidelines: Penicillin and metronidazole or Clindamycin alone in case of penicillin allergy. These patients will be treated in the outpatient clinic and will be examined again the day after inclusion. Aspiration will be done if necessary. At this first control visit the clinician will schedule the next visit based on findings.

Patients treated with tonsillectomy a chaud are admitted for intra-venous treatment with penicillin and metronidazole until surgery. Antibiotic treatment is discontinued after surgery and the patient may be discharged from the hospital the day after surgery.

Data will be collected regarding gender, age, history of tonsillitis (number of sore throats per year over the past 3 years) and regarding predefined outcomes: duration of hospital stay, duration of sick leave, use of antibiotics, VAS score and use of painkillers. The patient is also asked to indicate when he feels well again and free from disease. Quality of life will be measured using the Glasgow benefit Inventory which is validated for use in ENT-surgery. In addition, in the tonsillectomy group data regarding the number of episodes postoperative bleeding and readmissions will be collected and in the aspiration group data regarding the number of aspirations (outpatient controls) and number of rescue tonsillectomies due to failure of aspiration treatment will be collected.

All patients will be follow-up after 1 week, 2 weeks, 4 weeks and 3 months using online surveys.

The project has been approved by the The Regional Committees on Health Research Ethics for Southern Denmark and The Danish Data Protection Agency. Data is stored in SharePoint or RedCap and analyzed in STATA. Randomization was done by Sealed Envelope Ldt. 2017 (https://sealedenvelope.com/simple-randomiser/v1/lists [Accessed 6 Oct 2017]).

All enrolled patients will receive written information.

I. Power calculation Power calculation was done in order to estimate the minimum number of patients to be included in the study. A default value of 0.80 was chosen. Calculations are based on the assumption of the minimum relevant clinical difference between the two types of treatment in terms of number of days of sick leave. Patients will be randomized in two groups of equal size. Based on similar studies the investigators expect normal/parametric distribution of the data with standard deviation of 6 and mean difference of 3 between the two groups. Based on these assumptions a minimum of 64 patients should be included in each group to reject the null hypothesis (that there is no difference in the number of sick leave days). Significance level is set at 0.05.

II. Statistics In case of normal distribution unpaired t-test will be applied. Non-parametric tests will be applied if data is non-normally distributed.

Quality of life data will be analyzed according to recommendations by the developers of the applied instrument.

Ethical aspects Both treatment options included in the study are well established and used in current treatment of PTA. Patients fulfilling any of the exclusion criteria will be offered tonsillectomy.

Written consent will be obtained from all patients after oral and written information about the study.

Impact on the daily function in the ENT departments Patients with PTA are usually referred to the ENT department from general practitioners, ENT doctors in private clinics, emergency departments etc. A smaller proportion is treated in ENT private clinics if the general condition of the patient is good. Therefore, ENT doctors in private clinics will be asked to refer all patients with PTA to hospital departments for inclusion in the study.

The study will not interfere significantly with daily functions in the departments. However, at first visit the clinician will inform patients orally and hand out written information on the study as well as obtain written consent.

Economy Funding may be applied for in relevant research Foundations e.g. OUH's Research Group, Independent research, Consultant Council at OUH and Maersk Foundation.

Perspectives The investigators expect to include a total of 128 patients divided into two groups during a period of 12 months. The number of included patients from the different departments will vary according to size and activity level. The investigators hope to obtain enough evidence for a clinical guideline on the treatment of PTA.

Results are expected to be published in an international journal.

Inclusion Criteria:

• Peritonsillar abscess

Exclusion Criteria:

• respiratory distress
• suspected malignancy
• previous PTA
• history of recurrent tonsillitis
• patients with poor compliance to aspiration

masking is not possible