Insertion torque recordings for the diagnosis of contact between orthodontic mini-implants and dental roots: a systematic review

Most orthodontic mini-implants (OMIs) are inserted between dental roots. The prevalence of contacting these structures is high. Such contacts can cause permanent root damage and implant instability. Increased torque levels during implant insertion (the index test) could be a more accurate and immediate measure for diagnosing implant-root contact (the target condition) than radiographs (the reference standard) and could ultimately lead to a reduction or elimination of X-ray exposure. To address this issue, we asked three questions: (1) whether OMIs with root contact had higher insertion torque values than those without, (2) what is the accuracy of the index test compared with the reference standard to diagnose the target condition and what are the adverse effects of the index test, and (3) whether intermediate torque values have clinical diagnostic utility.

Methods were conducted according to our published protocol, which was based on the PRISMA-P 2015 statement. We applied broad spectrum eligibility criteria that included randomized and non-randomized studies on clinical, animal, and cadaver models. Not including such models would be unethical because it could slow down knowledge creation on the adverse effects of implant insertion. We conducted searches in more than 40 electronic databases including MEDLINE and 10 journals were hand-searched. Grey literature and reference lists were also searched. All research procedures were conducted independently by three reviewers. Authors of selected studies were contacted to obtain additional information. Outcomes on the three different research models were analysed separately. Systematic error was assessed with the Cochrane ‘Risk of bias tool’ for non-randomized studies.

One clinical, two animal, and two cadaver studies fulfilled the eligibility criteria of the first research question. All studies and subgroups demonstrated higher insertion torque values for OMIs with the target condition than those without. Mean differences (MD) between these effect estimates were statistically significant in one beagle model (MD, 4.64; 95 % CI, 3.50 to 5.79) and three subgroups of cadaver studies (MD, 2.70; 95 % CI, 1.42 to 3.98) (MD, 3.97; 95 % CI, 2.17 to 5.78) (MD, 0.93; 95 % CI, 0.67 to 1.20). Highest mean differences were identified in most self-drilling compared with pre-drilling groups. Clinical heterogeneity between studies was high, and many items were underreported. All studies except one cadaver study scored at least one domain as ‘serious risk’ of bias. No studies addressed the second research question. One cadaver study addressed the third question which showed the importance of recording torque levels during the entire implant insertion process. Responses of contacted authors were helpful, but often difficult to obtain. Implants fractured in one animal and in one cadaver model.

All eligible studies scored higher insertion torque values for implants with root contact than those without, but none of these studies assessed the diagnostic accuracy of the index test. The inclusion of non-randomized and animal and cadaver models in this systematic review provided key findings that otherwise would have been wasted. Such studies are important in the context of the wide applicability of this test, the high prevalence of the target condition, and the underreporting of adverse effects of interventions. A protocol for a potential new diagnostic pathway was presented, and the importance of contacting authors was addressed. The applicability of the findings should be interpreted in the context of underreporting and the many limitations of the included studies.