Normative tonsillectomy outcome inventory 14 values as a decision-making tool for tonsillectomy

Tonsillectomy is one of the most common surgical interventions [1, 2]. The main indications for tonsillectomy are recurrent infections, suspicion of malignant disease, and grade four tonsillar hypertrophy (kissing tonsil) with obstructive sleep apnoea [3]. Recurrent tonsillitis (RT) reduces the disease-specific quality of life (QOL) [2]. The effectiveness of tonsillectomy as a treatment for RT in adults is controversial. Previous studies have shown that tonsillectomy is associated with improved health-related quality of life (HR-QOL) [1, 2, 4, 5], reduced medication consumption, less time off work, and fewer visits to the physician [6]. In another study, 75% of tonsillectomies for RT resulted in postoperative pharyngitis and upper respiratory tract infections [7].

In 2013, the Bertelsmann Foundation reported that the frequency of tonsillectomy differed regionally in Germany by a factor of eight. To address this problem, guidelines for and classification of conservative and surgical treatments were developed. In August 2015, the AWMF (a consortium of scientific medical societies) published the new S2k guidelines on the therapy of palatine tonsillitis. A main focus of the new guidelines was defining the indications for tonsillectomy. The guidelines specified that a tonsillectomy should only be considered if purulent tonsillitis has been treated with antibiotics six times within the past 12 months [8].

The lifetime prevalence of common RT is 7–11% [9]. Only a few studies have addressed whether recurrent tonsillopharyngitis affects QOL in adults [1, 5, 6, 10, 11], although this has been well-studied in children [6, 7, 12‐24]. In 2012, Skevas et al. introduced the Tonsillectomy Outcome Inventory 14 (TOI-14), which was the first worldwide-validated instrument to measure disease-specific QOL in adults with RT [25]. However, the TOI-14 is not yet commonly used in clinical practice to decide whether tonsillectomy is necessary in cases of RT.

In the present systematic prospective study, TOI-14 scores were measured in two cohorts: a middle-European reference cohort of 1000 healthy individuals and a clinical cohort of 108 tonsillitis patients. The healthy volunteers were recruited from a non-probability panel. The subsample has been quoted to the population distribution of the German Microcensus with respect to age, gender, education and region. The tonsillitis patients were scheduled for elective tonsillectomy in our department. We compared the HR-QOL of healthy individuals and tonsillitis patients, examined the factorial structure of the TOI-14 questionnaire using principal component analyses (PCA) and exploratory factor analyses (EFA), and explored whether normative TOI-14 scores can define the level of disease burden that justifies tonsillectomy.

Quality of life (QOL) measurements are becoming more important in otolaryngology [27], especially in the case of tonsillectomy, which is a common otolaryngological operation [6]. Studies have focused on the outcome and benefit of tonsillectomy in children with obstructive sleep apnoea but studies in adult patients are still lacking. According to the published literature, tonsillectomy does not only impact QOL in children but may also affect the outcome in adults [1, 5, 6, 10, 11]. Measuring the success of a medical intervention requires analysis of patient-related outcomes. To date, only two disease-specific questionnaires relating to tonsillectomy and tonsillotomy have been validated for adult patients: the Tonsillectomy Outcome Inventory 14 (TOI-14) and the Tonsil and Adenoid Health Status Instrument (TAHSI). Adenoiditis is a paediatric disease, so the TAHSI questionnaire [10] cannot be used to assess QOL in adults with recurrent tonsillitis (RT). To determine whether patients with RT are suitable for tonsillectomy, standardized values from a sufficiently large healthy reference cohort need to be obtained from a tonsillitis-specific instrument. To address this, we measured TOI-14 scores in 1000 healthy volunteers and found that this reference cohort perceived a good QOL without RT.

To our knowledge, only a few studies have assessed the TOI-14 score before elective tonsillectomy. In the first study, suitability for tonsillectomy was determined by the SIGN 117 guidelines, and 97% of the preoperative TOI-14 scores were within two standard deviations (range 26.22–65.02; mean 45.62) [27]. Skevas et al. measured TOI-14 scores before and after tonsillectomy and compared these scores to those from 67 healthy individuals, but this study had insufficient power [25]. This study also evaluated TOI-14 subscales: throat problems (questions 1–4), overall health (questions 5–6), resources (questions 7–10), and social-psychological restrictions (questions 11–14) [25]. However, the test–retest-reliabilities of the subscales “general health” (r = 0.45) and “resources” (0.44) were only moderate. To address this, we reperformed PCA and EFA and proposed three novel TOI-14 subscales—physiological (question 1–5), psychological (questions 11–14), and socio-economic (questions 6–10)—which explained 73% of the variance in TOI-14 scores. We did choose different umbrella terms than Skevas et al. [25] because our new PCA and EFA revealed novel assignments of items to particular components (as shown in Table 1), which is more plausible. Furthermore, general headings are easier to understand.

We also observed that questions 12 and 13 were relevant to the psychological impact of chronic throat problems, question 3 to the physiological impact, and question 10 impacted the socio-economic measurement. These findings indicate that the different TOI-14 subscale scores should be considered when deciding whether tonsillectomy would give the best patient outcome.

Concerning the socio-demographic aspects, age and female gender significantly influenced the TOI-14 score, especially in the psychological (age) and socio-economic (gender) subscales. This suggests that young and female patients with RT could benefit the most from tonsillectomy.

Our comparisons between the healthy reference cohort and clinical cohort of tonsillitis patients confirmed that the adverse effects of RT have a huge impact on disease-specific QOL in adults. Our preoperative patient cohort had much higher TOI-14 scores (with less variation) than the healthy reference cohort, indicating higher levels of disease burden. Absence from work and lack of concentration affected productivity and consequently the socio-economic status. This may lead to job insecurity, which negatively affects QOL and health [30]. Interestingly, TOI-14 scores were still higher in the postoperative patient cohorts than in the healthy cohort. The postoperative cohorts had the highest psychological and socio-economic TOI-14 sub-index values, and the preoperative cohort had the highest physiological values. These results raise the question as to whether adult patients really benefit psychologically and socio-economically from tonsillectomy.

The lack of benefits after tonsillectomy suggests that stricter indication criteria are needed. We started measuring our patient TOI-14 scores in 2012. Since then, the indications for tonsillectomy have been tightened thanks to the 2015 AWMF guidelines. Now, tonsillectomy should only be performed after six cases of purulent tonsillitis have been treated with antibiotics within the past 12 months. Our postoperative data need to be interpreted with caution because the drop-out rate was high, indicating potential bias.

A QOL assessment tool is needed that encompassed functional, psychological, and socio-economic properties of patients undergoing tonsillectomy. Close examination of our novel TOI-14 sub-index scores may help to select those patients who will benefit most from tonsillectomy.

We have developed a set of reference values that, together with the TOI-14, can determine the disease burden that indicates tonsillectomy, according to the AWMF guidelines.