Effectiveness of different treatments for odontogenic keratocyst: a network meta-analysis

Odontogenic keratocysts (OKC) are benign but aggressive lesions. As there is a lack of well randomized clinical studies assessing the effectiveness of the different treatment options for OKC, a network meta-analysis (NMA) was performed to identify the best treatment option with the lowest recurrence rate. An electronic search was performed following the PRISMA guidelines to identify all clinical studies comparing treatment options against enucleation alone. The outcome variable was recurrence. The predictor variables were treatments. The eight included treatments were: enucleation with peripheral ostectomy/curettage (E + PO/curettage); enucleation with cryotherapy (E + CRYO); enucleation with/without PO followed by modified Carnoy's solution (E ± PO+MCS); enucleation with PO and with topical 5-fluorouracil (E + PO+5FU); enucleation with/without PO followed by original Carnoy's solution (E ± PO+CS); marsupialization alone (MARS); marsupialization followed by secondary enucleation with/without PO (MARS+2°E ± PO); and resection. The odds ratio was used to estimate the recurrence rate. A frequentist NMA was performed using Stata software. A total of 2989 patients in 40 studies were included. Both direct pairwise meta-analysis and NMA showed that E + 5FU+PO was significantly superior to E ± PO+MCS. However, no statistically significant difference was found between E ± PO+CS vs E + 5FU+PO, E ± PO+MCS, and resection, respectively (all very low quality evidence). The three most effective treatments in reducing the recurrence rate were E + PO+ 5FU (98.1%; very low quality evidence), resection (83.5%; very low quality evidence), and E ± PO+CS (63.8%; moderate quality evidence). The findings from this study suggest that CS remains the most effective fixative agent after enucleation and PO until proven otherwise. Additionally, 5FU appears to be an effective method with promising results that needs further research. Finally, the efficacy of MCS remains controversial; further in vivo and in vitro studies are required to determine new protocols. As this NMA included retrospective studies, the results should be interpreted with great caution (level of evidence: type III).