Introduction
Loud snoring present on most or all nights is reported by 24–50 per cent of males and 14–30 per cent of females.Reference Fairbanks and Fujitas1, Reference Ah-See, Stewart, Banham, Robinson, Carter and Wilson2 Many procedures have been designed to treat snoring due to palatal flutter, including uvulopalatopharyngoplasty,Reference Fujita, Conway, Zorick and Roth3 laser-assisted palatoplasty,Reference Kamami4 cautery-assisted palatal stiffening,Reference Mair and Day5 radio frequency ablation of the soft palate,Reference Powell, Riley, Troell, Li, Blumen and Guilleminault6 mucosal strip uvulectomyReference Morar, Nandapalan, Lesser and Swift7 and traditional palatoplasty.Reference Ellis, Williams and Shneerson8 These procedures all work in some, but not all, snorers. When successful, they also improve patients' quality of life,Reference Reda, Ullal and Wilson9 both for those with simple snoring and those with obstructive sleep apnoea syndrome. The difficulty lies in making the definitive diagnosis of palatal flutter as the cause of snoring, and deciding which patients should undergo which treatment. In some cases, such treatments are relatively radical. It is logical that patients with palatal flutter would benefit from palatal surgery, whereas those with tongue base or other types of snoring would not benefit.
The difficulty in determining which patients have palatal flutter has prompted two areas of investigation: acoustic analysis and sleep nasendoscopy. Acoustic analysisReference Jones, Swift, Calverley, Ho and Eais10 has not found its way into routine clinical practice and remains generally a research tool. Sleep nasendoscopy has been popularised by Pringle and Croft.Reference Pringle and Croft11
However, El-Badowey et al. Reference El-Badowey, McKee, Heggie, Marshall and Wilson12 have demonstrated that sleep nasendoscopy is not of value in the management of habitual snoring, as it is a poor predictor of favourable outcome from palatal surgery. This leaves some doubt as to whether sleep nasendoscopy truly demonstrates palatal flutter, or whether the artificial, induced sleep required for the procedure is so different from real sleep that such assessment is not a reliable measure. Other measures of snoring, such as pressure monitoring, have tended to focus more on obstructive sleep apnoea. This procedure could help to localise obstruction and may be applicable in the future for simple snoring.Reference Reda, Ullal and Wilson9 It would seem logical to investigate patients by temporarily stiffening the palate to stop flutter, and then to treat with more permanent surgery only those patients in whom the temporary procedure stops snoring.
Achieving temporary palatal stiffening has been attempted by passing a cord through the nose and pulling the palate forward,Reference Ramilingam and Smith13 and by using an upper dental plate with an attachment to press on the soft palate. Another method is injection of fluids into the palate, for example, the injection snoreplasty technique of Brietzke and Mair.Reference Brietzke and Mair14 This latter technique appears to be safe and pain-free, and may constitute a treatment when used as multiple injections.
The present study addressed the question of whether this injection snoreplasty technique can be used as an investigation for palatal flutter when administered as a single injection.
Methods and materials
Sixty patients were included in this pilot study, 36 men and 24 women. Their ages ranged from 22 to 72 years and their body mass indexes (BMIs) from 23 to 36. (These BMIs were somewhat low compared with the general population, probably due to the fact that local general practitioners tended not to refer very overweight people with snoring, preferring instead to send them to a dietician.)
A full medical history was taken from all patients, as asthma was considered a contraindication by the manufacturer; none of the 60 patients had asthma. A full ENT examination, including palate and tonsils, was undertaken; gross nasal obstruction requiring surgical attention was excluded. Body mass index, palatal and tonsillar examination results, sleep apnoea, Epworth sleepiness scale results, alcohol intake and smoking were not considered as contraindications, although all patients were given general advice on how these could affect snoring.
The injection snoreplasty procedure was explained in detail and a consent form signed.
Statistical analysis was undertaken using ‘the simplest statistical test’.Reference Pockock15
Injection snoreplasty procedure
Lignocaine spray was used on the palate (xylocaine 1 per cent). Three minutes later, a 1 ml injection of 1 per cent (10 mg/ml) or 3 per cent (30 mg/ml) sodium tetradecyl sulphate (Fibro-vein; C P Pharmaecutical Products Limited, Wrexham, United Kingdom) was given into the midline of the soft palate, 1 cm above the junction of the soft palate and the uvula. The sodium tetradycil sulphate concentration used was random, as the hospital pharmacy supplied 3 per cent or 1 per cent solutions on different days; either was considered to be adequeate. After the procedure, patients were asked to wait 20 minutes in the clinic before going home. Their first review was six weeks later.
As there were no definitive, standardised, internationally accepted snoring scales reported in the literature, we reviewed the functional outcomes of a number of assessments. These were the sleep questionnaire,Reference Weaver, Laizner, Evans, Maislim, Chugh, Lyon, Smith, Schwartz, Redline, Pack and Dinges16 the symptoms of nocturnal obstruction and related events questionnaire,Reference Piccirillo, Gates, White and Schectman17 the snoring symptom index,Reference Scotts, Richardson and Wilson18 several quality of life questionnaires, and, the most commonly used, a visual analogue scale or its digital equivalent. For ease of analysis, we used a questionnaire with a simple five point scale (i.e. 1, snoring, 2, disappeared, 3, better, 4, same, 5, worse). Patients were given this simple questionnaire at the first visit and then six weeks, three months, six months and 12 months after the procedure.
Results
Forty patients received a 1 per cent sodium tetradycil sulphate injection. Of these, two patients (5 per cent) reported that their snoring had disappeared completely 12 months after the injection, and 25 patients (62.5 per cent) reported improvement (confirmed by the fact that their partners were no longer sleeping in a separate room). Twelve patients (30 per cent) reported that their snoring was the same, and one patient (2.5 per cent) reported worsened snoring (Table I).
Table I Snoring outcomes
* n = 40; †n = 20. STD = sodium tetradycil sulphate
Twenty patients received a 3 per cent sodium tetradycil sulphate injection. Of these, no patient reported that their snoring had worsened, 12 (60 per cent) reported an improvement and eight (40 per cent) reported no change (Table I). Table II shows sleep apnoea.
Table II Sleep outcomes*
At 12 month review, there appeared to be no difference between patients who had received 1 per cent and 3 per cent sodium tetradycil sulphate.
Patients were also asked whether their sleep had improved since the injection. Twenty-seven patients (45 per cent) reported that it had, 15 (25 per cent) reported that it had not, and 18 (30 per cent) did not know (Table II).
Patients' BMIs had not changed over the one year of the study; although there had been temporary improvements, patients had returned to their original BMI eventually.
The BMI results of both sodium tetradycil sulphate concentration groups were combined, and the BMIs of the ‘failed’ patients at six weeks (39 per cent) and at 12 months (52 per cent) were compared with those of the ‘successful’ patients at the same time intervals (61 and 48 per cent, respectively). There was a nonsignificant trend towards higher BMI and higher Epworth Sleepiness Scale (ESS) in the Failed group at six weeks but no statistically significant difference (p > 0.02). At one year, however, this trend had disappeared.
Ten patients (16.6 per cent) reported no side effects of the injection at all. The main side effects were a feeling of a lump in the throat for a few days afterwards, reported by 28 patients (46.7 per cent), and mild discomfort on swallowing, reported by 16 patients (26.7 per cent). Three patients (5 per cent) reported an ulcer in the throat lasting for three days, and two patients (3.3 per cent) fainted two hours after the injection, an incident attributed to the procedure. One patient (1.7 per cent) reported severe ear pain (Table III). When asked whether they would undergo the injection again, 54 patients (90 per cent) said yes, four (6.7 per cent) said no and two (3.3 per cent) did not know (Table IV).
Table IV Patients' responses to ‘would you choose to have the injection again’?
Eleven patients experienced initial resolution of their snoring, but one year later their snoring had returned to pre-injection levels (Table V). These patients were listed for laser-assisted palatoplasty or uvulopalatopharyngoplasty. The decision between the two procedures was made based on whether large tonsils were present or not. The eleven patients all underwent successful palatal surgery one year after their injection snoreplasty.
Table V Post-injection change in snoring improvement
Discussion
It is often difficult to determine which patients will benefit from palatal surgery. All palatal surgery stiffens the palate in some way and therefore reduces palatal flutter. There is currently no investigation available to differentiate patients for whom palatal surgery has a high chance of success. In the future, acoustic analysis seems likely to be the best method, but at present it remains a research tool, and sleep nasendoscopy does not stand up to rigid testing.
We used single dose injection snoreplasty as an investigation to temporarily stiffen the palate, in order to determine which patients had palatal flutter. This was done in an effort to better select patients for uvulopalatopharyngoplasty or laser-assisted uvuloplasty. Interestingly, some patients only needed a little stiffening; in these patients, the investigation became the treatment. Patients requiring substantial palatal stiffening gained some benefit from the injection but the effect was short-lived, lasting one year at most. All of these latter patients gained further benefit from more radical treatment, either uvulopalatopharyngoplasty or laser-assisted palatoplasty. A third group showed no benefit from temporary stiffening and were thought to be inappropriate for palatal surgery. These patients were further investigated for tongue base, epiglottic or global collapse snoring. Surprisingly, patients' BMI and failure did not differ between the three groups. It could be argued that if some patients gained temporary benefit from a single dose of injection snoreplasty, then further doses should be given to the failures. This theory is currently under investigation, with second and third injections of 3 per cent sodium tetradycil sulphate being used, and will be the subject of a further report.
Although there was no deliberate randomisation of patients to receive either 3 per cent or 1 per cent sodium tetradycil sulphate, we assessed the two groups separately with respect to side effects and efficacy (Tables I and III). Interestingly, the 1 per cent dosage (67.5 per cent failure) was slightly more efficacious than the 3 per cent dosage (60 per cent failure). However, this difference was not statistically significant.
• This paper examined the role of injections to the soft palate (snoreplasty) in separating palatal causes from other causes of snoring
• The technique used sodium tetradycil sulphate injected under local anaesthetic; results were assessed using a visual analogue scale over the succeeding 12 months
• Single dose injections seemed to differentiate palatal snorers from others and to provide a safe, simple, ambulatory treatment for one-third of the patients treated: those who relapsed went on to have successful conventional surgical treatment
A number of questions remain to be answered. If injection snoreplasty improves snoring but does not eliminate it, should more radical treatment be undertaken? If injection snoreplasty improves snoring but the snoring later returns, it appears that one such treatment in these patients could achieve what from the patient's perspective is a good, albeit only year-long, result. How long the improvement may be maintained is not known, nor whether any other factors (e.g. tonsil size or palatal anatomy) could influence the outcome.
Conclusion
In this investigation, our original intent was to use single dose injection snoreplasty to decide which patients should be listed for more radical surgery. However, a significant number of patients (62 per cent) demonstrated significant improvement in the short term. This technique seems not only to constitute an investigation but also, perhaps, a safe, simple, clinic-based treatment for many patients. At the very least, its use prevented patients without palate-related snoring from undergoing painful, unpleasant surgery. Injection snoreplasty appears to be useful, both in investigation and, in some patients, as a treatment for habitual snoring.
Acknowledgement
We should like to thank Mrs A McCabe for her help in the presentation and typing of this paper.
