Most techniques for endodontic treatment in primary teeth require instrumentation of root canals and filling with an absorbable paste. This filling material could be pastes based on zinc oxide and eugenol, calcium hydroxide or iodoform, or filling materials composed by mixture of some of these components. This method, however, requires a long clinical time, which can be a challenge in the Pediatric Dentistry clinical practice. Therefore, some alternative techniques in which non-instrumentation is necessary have been purposed.
The non-instrumentation technique most used in Brazil is a technique previously described, in 1964. This method involves the access to pulp chamber and removal of necrotic tissue with irrigation. After that, with no instrumentation of the root canals, a medicament containing zinc oxide and eugenol and two antibiotics, chloramphenicol and tetracycline, is placed in the entrance of the canals, and the restoration of the teeth is performed. Although the non-instrumentation methods for endodontic treatment of primary teeth are attractive and demand a short clinical time, strong evidence from well-conducted clinical trial lacks.
Therefore, this four-center parallel-arms non-inferiority randomized clinical trial was designed to evaluate the efficacy of a non-instrumentation of root canals and use of a paste containing antibiotics (chloramphenicol and tetracycline) and zinc oxide and eugenol (CTZ) technique for endodontic treatment of primary molars. The hypothesis is that the efficacy of non-instrumentation treatment is non-inferior to the standard treatment involving manual instrumentation of root canals and filling with zinc oxide and eugenol (ZOE) paste after 24 months of follow-up. Children who look for dental treatment in one of four participating centers and who have at least one primary molar requiring endodontic treatment will be eligible. If the child has more than one molar with pulp involvement, the included tooth will be determined by chance. Other teeth will be treated, but they will not be included in the study. The participating centers are the dental schools of Santa Catarina Federal University, University of Sao Paulo, Rio de Janeiro State University and Iguaçu University. Included children will be randomly allocated to one of the two groups (allocation ratio of 1:1): CTZ group (non-instrumentation technique) and ZOE group (standard treatment).
Randomization list will be generated in an appropriate website (www.sealedenvelopes.com), in permuted blocks (2, 4, and 8 samples) and stratified for the center. The sequence will be enclosed in individual opaque envelopes separated by the stratum. The disclosure of the allocated group will occur when the children will be seated in the dental chair, after the anesthesia, rubber dam isolation and opening of the tooth included for that child. Than the included tooth will be allocated to the CTZ group or to ZOE group. After the treatment, children will be followed-up for 6, 12, 18 and 24 months. Evidence of treatment failure in any period will be recorded, and the teeth will be extracted.
The primary endpoint will be the success of the dental treatment after 24 months of follow-up (teeth with no evidence of failure in all follow-ups). The sample size calculation was estimated based on a success rate of 92% obtained in pulpectomies using ZOE from a previous systematic review (Coll et al., 2020), a non-inferiority limit of 15%, a significance level of 5% and statistical power of 90%. It was added 20% to this number to contemplate possible drop-outs and 60% due to the multi-center design. The anticipated sample size was 218 teeth (109 per group), divided among the centers.
However, in March 2023, we decided to modify the original sample size from 218 to 200 participants, due to slower-than-anticipated recruitment. With this sample, considering the same parameters, we maintained a statistical power of around 90% (90.6%).
Another problem we have noticed in the last year (between March 2024 and March 2025) is that one of the study centers (Iguaçu University) was unable to include and follow up an adequate number of participants (less than 10 children in almost two years). Therefore, we decided to drop this center, and we recalculated the sample size, adjusting the correction for the multicenter design to 1.4. Our final minimum sample size to reach an adequate power, considering the prespecified parameters - 5% of significance level, 90% of statistical power, 92% of success rate for the control group, a non-inferiority limit of 15% and adding 60% to this number (20% for drop-outs and 40% due to multicenter design) - was 182 participants. As our randomization process was stratified per center, we do not expect a significant bias due to the center drop.
Masking will not be possible for participants, operators, researchers, or even outcome assessors due to the different nature of the treatments. However, the outcome assessor will not be informed of the allocated group for each tooth.
Other secondary endpoints will be the clinical time spent with the treatments, children's behavior during the treatment, discomfort immediately after the end of the treatment reported by the children, post-operative pain, short- and long-term improvement in the negative impact of Oral Health-Related Quality of Life, costs and cost-efficacy. The frequency of treatment success after 24 months will be compared between the groups in the intention-to-treat population considering the non-inferiority limit of 15% by the Miettinen-Nurminen's method. Sensitivity analysis will be conducted through non-inferiority Cox regression. For this, the period that the treatment failure occurred will be recorded. Secondary outcomes will be compared by proper statistical tests considering two-tailed hypothesis.
