The most appropriate management for the lesion now known as the keratocystic odontogenic tumour (previously known as the odontogenic keratocyst) remains controversial. This article reviews the different management protocols adopted by one surgical unit over the last 30 years and the results obtained from the different treatment modalities. A current treatment protocol consisting of initial decompression followed by aggressive curettage and peripheral ostectomy with methylene blue staining appears to be successful, but our longest follow-up is only 6 years.
The lesion now called the keratocystic odontogenic tumour (KCOT) was first described by Philipsen in 1956 under the name odontogenic keratocyst. This original article was in Danish but had a summary in English. It was, however, not widely recognized as a separate entity until the articles by Browne in 1970 and 1971, which clearly delineated the clinical and histological features of the lesion, coupled with its high recurrence rate with simple enucleation. Prior to this time, it is felt that most of these lesions were identified as primordial cysts, which is a term no longer used.
Following the confirmation of the accuracy of histological diagnosis and of the high recurrence rate of these lesions with simple treatment, there was a division of philosophies on their management. These varied from the feeling of my old chief, Gordon Hardman from North Wales, who essentially stated that ‘we always knew there were some cysts that recurred, even though we did not have a special name for them, and all we did was curette them out every 5–10 years. Why shouldn’t we just continue to do that?’ all the way to the opinion of people like Paul Bramley who recommended much more aggressive treatment, consisting of either ‘en-bloc’ or segmental resection, depending on the size. With these two diametrically opposing views, an effort was made to identify a middle course that might result in an acceptable success rate in the long term, with an acceptably low morbidity. Paul Stoelinga and others are of the opinion that these lesions can arise from down-growth from the oral epithelium and that therefore the overlying oral epithelium should be excised along with the lesion. Since these lesions normally occur in the posterior mandible, this can involve incisions over on the lingual side of the crest of the ridge, so care has to be taken to identify and protect the lingual nerve, if appropriate.
In an effort to identify this middle course, treatment has consisted of techniques ranging from enucleation plus cryosurgery – a technique learned from one of my mentors, Paul Bradley, and utilized extensively over the last 30 years – to enucleation plus Carnoy’s solution, enucleation with peripheral ostectomy, and more recently marsupialization and decompression, with various combinations.
We first published an article on the use of enucleation coupled with liquid nitrogen in 1993 and at that time reported a recurrence rate of around 11.5%, which was less than the recurrence rates of 20–60% that were being reported for simple enucleation. With the extended follow-up of these patients for a longer period (up to 25 years now), it does appear that the recurrence rate has remained around 10%. The problem with the technique, however, has been access to the equipment, which is being used less and less in other medical specialties, so hospitals often do not have access to it. Also, if one is not meticulous in protecting the soft tissues, there will be necrosis of the soft tissues leading to wound breakdown and a prolonged recovery, and at least partial loss of any bone graft that has been used. At that time we were recommending bone grafting of any lesion over 4 cm in size, but ultimately reduced this to 3 cm in size in order to accelerate the healing of smaller lesions.
Around 1995, following the final retirement of my mentor, Gordon Hardman, in North Wales, I spent some time sorting out his slide collection which consisted of many thousands of 35-mm Kodachrome slides. In this I found a number of cases of cysts that he had obviously marsupialized and decompressed in the early 1950s or possibly even earlier. From examining these slides, it did appear to be a practical technique, which had been largely forgotten following the widespread use of enucleation and primary closure, coupled with the liberal use of antibiotics.
It was therefore decided to apply this as a definitive technique for the management of odontogenic keratocysts, and we published our first results in 2004. These seemed to show very satisfactory outcomes for a relatively small number of patients, with a recurrence rate around 5%. However, during the follow-up of these patients, we started to see more recurrences, and in 2007 I published a partial retraction stating a recurrence rate of around 12% over a relatively short period of time. Since that time, the number of patients that we have treated with this technique has risen to 73, and the recurrence rate at the present time has risen to 27% (20 recurrences). The longer we follow these patients, the more recurrences we seem to see, and one can only speculate what the final recurrence rate may be when these patients have been followed up for 20 years or longer.
Once we realized that the recurrence rate was going to be as high as it had turned out to be, we looked around for an alternative technique that would combine the best of these techniques. In 2006, we introduced the technique whereby we decompressed the lesions until they appeared radiographically to have resolved down to about 2–3 cm in size, and we then surgically enucleated the lesions and carried out a peripheral ostectomy to remove any cyst remnants that might be in the surrounding bone. The advantages of this technique are outlined below.
The lesions do decompress very satisfactorily. By this time we had modified the technique to utilize a 14 French gauge paediatric feeding tube as the decompression tube and to wire it around the adjacent first molar tooth, passing the wire through the wall of the tube so as not to compress it. We could then bring the drainage tube forward to the bicuspid region where patients found it easy to irrigate ( Fig. 1 ). This meant that patients could keep the area clean and irrigate it more easily. If the lesion was anywhere else except the posterior mandible, the technique was modified accordingly. For extremely large lesions that crossed the midline, a drainage tube could be placed on each side ( Fig. 2 ).