Indications for Tonsillectomy: Are We Documenting Them?

Abstract

INTRODUCTION

Tonsillectomy is one of the most frequently performed operations in the UK. Documentation of the indications for tonsillectomy is vital, and should fulfil evidence-based guidelines where possible. We present a completed audit, evaluating the documentation of our department's practice in meeting the recommendations made by the Scottish Intercollegiate Guideline Network (SIGN) on indications for tonsillectomy.

PATIENTS AND METHODS

A prospective audit of 100 children undergoing tonsillectomy for recurrent tonsillitis at a university hospital during two time periods: October 2007 to January 2008 and March to September 2008. Interventions including the production of posters and rubber stamps were agreed and implemented between the two audit periods.

RESULTS

Following the implementation of simple changes, significant improvements were seen in documentation relating to the SIGN guidelines for tonsillectomy. Overall, the number of children meeting all four SIGN criteria for tonsillectomy rose from 12% to 44% (χ² = 57.8; P < 0.001). Furthermore, a significant reduction was seen in the number of children below the age of 5 years undergoing tonsillectomy for recurrent tonsillitis (χ² = 14.66; P < 0.001).

CONCLUSIONS

With increasing scrutiny on tonsillectomy, it is important to ensure that the reasons for performing tonsillectomy are documented clearly and adhere to evidence-based guidance where possible. We have demonstrated that, with only simple and low-cost interventions, significant improvements in the documentation of tonsillectomy indications can be achieved.

Tonsillectomy is one of the most frequent operations performed by general otolaryngologists. In 2005–2006, over 50,000 tonsillectomies were carried out in England and Wales; most are for recurrent tonsillitis.¹ Current evidence suggests that tonsillectomy is an effective treatment for recurrent tonsillitis in severe cases; however, in mild-to-moderate cases, a conservative approach may be indicated.^2,3

Recently, concerns over the number of tonsillectomies being performed has been raised, in part relating to the variation in rates of tonsillectomy both nationally and internationally.⁴ The otolaryngology community generally advocates tonsillectomy as a worthwhile operation in those with severe recurrent tonsillitis. However, in order to demonstrate that appropriate patients are being listed for surgery, good documentation of its indications is essential.

In 1999, the Scottish Intercollegiate Guideline Network (SIGN), which produces evidence-based clinical practice guidelines, published guidance on the management of sore throats and indications for tonsillectomy.² The indications recommended by the SIGN group are that patients should meet all of the following criteria prior to tonsillectomy: (i) sore throats should be due to tonsillitis; (ii) five or more episodes per year; (iii) symptoms for at least a year; and (iv) the episodes should be disabling and prevent normal function.² The guideline forms a basis for clinical audit.

The aims of this audit were to: (i) determine if the SIGN guidelines were being followed and documented for children undergoing tonsillectomy for recurrent tonsillitis; and (ii) implement change in order to facilitate an improvement in practice.

Patients and Methods

Data collection and analysis

We performed a prospective audit of 50 children undergoing tonsillectomy for recurrent tonsillitis between October 2007 and January 2008 at a university teaching hospital. Basic demographic details including age, along with information relating to the SIGN guidelines were recorded on a standardised data collection form. Data were obtained from out-patient notes, clinic letters and GP letters. Patients undergoing concomitant procedures such as adenoidectomy and those undergoing tonsillectomy for other indications were excluded.

Following the implementation of the changes outlined below, the audit was repeated on a further 50 consecutive children undergoing tonsillectomy for recurrent tonsillitis between March and September 2008. Data were analysed using chi-squared analysis.

Implementation of changes

The findings of the first audit cycle were presented to the department, along with three proposed interventions. The interventions were: (i) dissemination of audit findings to all clinical staff involved in listing patients for tonsillectomy; (ii) the production of posters of the SIGN Quick Reference Guide⁵ for clinical areas; and (iii) the introduction of a rubber stamp for use in the patient case notes (Fig. 1) in order to streamline documentation and act as an aide memoire.

Figure 1.

Rubber stamp introduced for use in clinical notes.

Results

Patient demographics

The age range of children in the first cycle was 2–16 years (median, 5.0 years) and in the second cycle 3–15 years (median, 8.5 years; Fig. 2). A significant reduction in the proportion of children in the 0–4 years age group was seen in the second audit cycle (χ² = 14.66; P < 0.001).

Figure 2.

Age distribution for children undergoing tonsillectomy.

Adherence to SIGN guidelines

A 6-month period of ‘watchful waiting’ was performed in 2% of patients in the first cycle and 18% of patients in the second. The average time from decision to perform tonsillectomy to surgery was approximately 2 months in both audit groups.

In both groups all patients listed for tonsillectomy had documented evidence of recurrent tonsillitis as the cause of their sore throats. The number of children listed for tonsillectomy with documented evidence of five or more episodes of tonsillitis rose from 24% in the first cycle to 44% in the second cycle (χ² = 4.46; P < 0.05). Documentation relating to duration of symptoms improved from 22% to 52% (χ² = 9.65; P < 0.005) and documentation relating to the impact of tonsillitis on daily activities improved from 63% to 90% (χ² = 9.54; P < 0.005; Fig. 3).

Figure 3.

Documented indications for tonsillectomy and overall adherence to SIGN guidelines.²

Overall, only 12% of children listed for tonsillectomy had documented evidence meeting all four SIGN criteria in the first cycle. Following the interventions, this increased to 44% (χ² = 57.8; P < 0.001; Fig. 3).

Discussion

Tonsillectomy is one of the most frequent operations performed by otolaryngologists. In general, children recover well, but tonsillectomy is not risk-free with up to 3.5% of patients experiencing potentially serious postoperative haemorrhage.⁶ The SIGN guidelines² recommend which patients are likely to benefit from tonsillectomy in recurrent tonsillitis based on the available evidence. A period of ‘watchful waiting’ is also recommended to allow the clinician time to establish the pattern of disease and to give the patient the opportunity to evaluate the potential risks. The overall aim is to ensure that the risks of tonsillectomy are balanced by the benefits to the patient. Therefore, patients not fulfilling the SIGN criteria may be being exposed to more risk than necessary. Furthermore, with increasing scrutiny on tonsillectomy,⁴ it is important to be able to justify reasons for listing patients for the operation. Good documentation following current evidence-based guidance is important to ensure that patients are managed in the most appropriate way and accurate local and national audit can be performed.

Following the first audit cycle, three simple changes were introduced to our practice. The second cycle demonstrated that the implementation of these changes resulted in a significant improvement in patient documentation. Increased awareness of evidence-based guidance for tonsillectomy is likely to have occurred as a result of dissemination of the initial audit results. This increased awareness is likely to have been augmented by the reproduction of the SIGN Quick Reference Guide and its display in clinical areas. To help surgeons further in a busy out-patient clinic, we have designed an easy-to-use rubber stamp for quick documentation in patient clinical notes.

The interventions also appear to have reduced the number of children under the age of 5 years being listed for tonsillectomy. These children, given their size, may have an increased mortality associated with complications. Furthermore, whilst ‘age over five’ does not constitute one of the SIGN guidelines, it is difficult to see how children under 5 years of age can satisfactorily fulfil all of the criteria.

We acknowledge that further improvements in our documentation could still be made. Both the number of children with appropriate indications and those undergoing a period of ‘watchful waiting’ could be increased. Other measures such as involving the out-patient nursing staff to pre-stamp the clinical notes to encourage compliance and updating surgeons at audit meetings annually may lead to further improvement in the quality of documentation.

Study limitations

There are some limitations to our audit. We only included children undergoing tonsillectomy for recurrent tonsillitis. A significant number of children will be listed for tonsillectomy because of co-existing concerns, for example, obstructive sleep apnoea. In these cases, strict adherence to the SIGN criteria is unnecessary and may be detrimental. The SIGN guidelines are only recommendations and a degree of clinical judgement remains important. It is for this reason that 100% compliance with the SIGN guidelines is unlikely to be achieved. The rubber stamp was readily accepted but does not provide scope for justifying tonsillectomy in cases where the SIGN guidelines are not met. However, we believe that the stamp acts as an aide memoire and any additional details could be added in the patient clinical notes or, indeed, the stamp modified according to local needs. Finally, the two time periods over which our audit was conducted differed in length. In part, this may have been caused by fewer children being listed for tonsillectomy as a result of the intervention. As a result, some of the doctors listing patients changed between the two audit periods. Whilst this may have a potential impact on our results, we believe that this effect is likely to be small given the experience level of the doctors and the numbers involved.

Conclusions

This audit demonstrates how small changes to our practice have made a significant improvement on the quality of documentation of the indications for tonsillectomy. Furthermore, the interventions are simple and effective, and they can be easily adopted by other departments with minimal cost.