Digital implant placement accuracy: a clinical study on a fully-guided flapless single-unit immediate-loading protocol

Since the advent of implant-supported prostheses, clinicians and manufacturers have constantly strived to produce more acceptable outcomes. Considering surgical and prosthetic aspects of the implant placement, it has evolved from the two-staged delayed-loading flapped technique being the prevalent treatment choice to implementing immediate loading, flapless, and digitally-guided techniques into the routine mode of care.

Flapless implant placement has shown several merits, such as decreased morbidity, less patient discomfort, shorter clinical time, reduced post-op bone loss, and improved blood microcirculation [1‐9]. Nevertheless, this technique comes with its disadvantages, including poor real-time visualization of the implant site, which demands accurate preoperative planning to prevent the violation of vital structures and bone perforations [6]. Implant placement accuracy using the flapless technique was found unacceptable while employing conventional 2D radiographs in conjunction with a regular examination of the implant site [10].

Computer-guided surgery has been introduced in an attempt to reach a more optimized implant positioning. Recent advancements in imaging modalities (e.g., the introduction of cone-beam computed tomography (CBCT) and digital scanners) and the production of sophisticated computer programs laid the foundations for computer-guided surgery [11]. The level of guidance (partially or fully) and the surgeon’s ability to modify the implant’s position during the surgery (static or dynamic) are two features by which the guided surgery is classified [12]. In line, a meta-analysis has shown that static computer-aided implant placement’s accuracy is superior to free-handed (no guides included) and partially-guided implant placement (regardless of the flapped or flapless approach) [13].

The flapless surgery’s combination with the computer guidance was the answer to concerns about its accuracy. Comparing flapless static computer-aided with free-handed and partially-guided implant placement in terms of accuracy shows angular and linear apical deviation significantly lower in the former group [14‐18].

Immediate-loading protocols have become more popular due to their ability to improve the patient's experience, reduce edentulous periods, satisfy esthetic demands, and preserve gingival architecture [19, 20]. However, these protocols present complications; for instance, they have been linked with more implant failure compared to the delayed-loading treatment plan, which means proper case selection is vital [19].

Whereas several studies evaluated flapless techniques, computer-guided surgery, and immediate-loading procedures individually, it is still necessary to research the therapeutic procedures that combine these techniques into a whole treatment plan. Moreover, this study employed a relatively modern technique for designing surgical templates (which uses CBCT records and intraoral scans), in contrast to the conventional dual scan technique with CBCT records of radiographic templates used in the previous studies.

This clinical study primarily aimed to evaluate the implant placement accuracy of a flapless computer-guided single-unit protocol by comparing the virtually planned and the actual post-surgical position of the implant fixture (by assessing angular deviation and linear apical deviation). A 3-month follow-up was performed following immediate-loading of the implants. The difference between the virtual and the delivered provisional restorations’ occlusal level, early implant failure, bleeding on probing, and peri-implant pockets were secondary outcome variables evaluated in this study.

Referred patients to the dental clinic of Tehran University of Medical Sciences were examined and selected based on the inclusion and exclusion criteria. Nine patients — seven males and two females — with a mean age of 52 years and 14 single-unit implant sites were included in the study. An oral and maxillofacial surgeon (M.B) with over 20 years of experience carried out all surgical operations (fully-guided flapless single-unit implant placement). Subsequently, a senior dental student (P.P) familiar with the digital scanner performed intraoral scans and delivered provisional restorations under the prosthodontist’s (F.A) supervision. Tehran University of Medical Sciences ethical committee approved this study before any clinical phase started (approval ID: IR.TUMS.MEDICINE.REC.1400.1101). Informed consent was obtained before patients' enrolment in the study.

A total of 14 implants were placed for nine patients. Comparison between the virtual and actual implant positions showed a mean angular deviation of 5.07° (standard deviation [SD] = 2.06°, minimum [min] = 1.77°, maximum [max] = 9.73) and a mean linear apical deviation of 1.74 mm (SD = 0.63 mm, min = 0.94 mm, max = 2.75 mm) (Table 1).

Nine provisional restorations were analyzed for the occlusal level difference, as five were missed due to the following reasons: One implant failed before the second intraoral scan, two were unacceptable for immediate loading due to insufficient primary stability, and two restorations were not adjusted as no opposing teeth were present (Table 1).

Two implants were subject to early failure within less than a month of the surgery. As for other complications during the first 3 months of the surgery, one screw loosening and one restoration breakdown were reported. They were resolved immediately by screw tightening and cementing a duplicate of the provisional restoration, respectively. In the 3-month follow-up examination, ten remaining restorations were examined for Bleeding on Probing and peri-implant pockets (Table 1).

There are a limited number of studies evaluating the accuracy of implant placement by digitally-designed surgical templates in which CBCT and intraoral scans are the input data. Still, we need to investigate different designing and surgical protocols to form a comprehensive opinion. Regarding the present study, the accuracy measurement was successfully done without any dropouts. However, secondary outcome variables were reported for some cases due to the presented complications. A digital workflow and tools required for accuracy measurement were described, which can be drawn on in future similar studies.

Guided surgery was introduced more than three decades ago as a manner to translate implant planning into the real surgical setting. Acrylic surgical templates — designed using optimal positioning of implant-supported prosthesis on stone casts — served as the conventional mean for guided surgery [24]. CBCT radiographs and stereolithographic techniques made it possible to virtually plan and fabricate surgical guides without including casts [25]. Virtual planning became widespread with the advent of dual scan protocols, which employed CBCT imaging of radiopaque templates to locate suitable implant positions in the jaw [26].

Several studies evaluated the accuracy of surgical templates fabricated by the dual scan protocol [15, 17, 21, 27‐30]. Most recently, Magrin et al. [15] conducted a randomized clinical trial (RCT) comparing free-handed flapped and fully-guided flapless implant placement surgery, replacing single missing teeth similar to the present study. The guided group results showed 2.53 ± 1.11 mm for the mean apical linear deviation (ALD) and 2.2 ± 1.1° for the mean angular deviation (AD). De Oliveira et al. [28] conducted the study with the largest sample size (115 implants) but on fully edentulous patients; by which mean ALD of 2.41 ± 0.74 mm and mean AD of 2.41 ± 0.15° for maxillary, and mean ALD of 2.18 ± 0.43 mm and mean AD of 2.50 ± 0.43° for mandibular implants were obtained. Valente et al. [21] also included a considerable sample size of 89 implants; subsequently, a mean ALD of 1.6 ± 1.2 mm and a mean AD of 7.9 ± 4.7° resulted. This study included partially and fully edentulous patients, meaning both tissue- and tooth-supported templates were used. Different data acquisition protocols, different analyzing methods, and inclusion of both tissue- and tooth-supported templated might be the reason between the current study and the mentioned studies’ results.

The introduction of intraoral and tabletop scanners has led to the next generation of virtual planning systems. These systems rely on the superimposition of CBCT and scan records, which consider soft tissue morphology in contrast with the conventional dual scan techniques. Moreover, these modern systems decrease clinical sessions and provide more flexibility as the procedure can be modified at any time to comply with different implant systems and designing software [26].

There are a few studies which calculated accuracy of implant placement using CBCT and intraoral scans [31‐35]. The study with the largest sample size (145 fixtures) showed mean ALD of 1.06 ± 0.44 mm and mean AD of 2.72 ± 1.42°. Maximum numbers calculated for mean AD and mean ALD in the mentioned studies were 3.1 ± 2.3° [34] and 1.3 ± 0.6 mm [34], respectively. Minimum numbers were also 2.25 ± 1.41° [35] and 1.06 ± 0.44 mm [32]. Different surgical and prosthetic settings, analyzing methods, and case selection criteria (length of the edentulous space) might be responsible for the different results in the present study compared with the previous ones.

Little research has been conducted on immediate-loading protocols using prefabricated restorations in the context of fully-guided flapless implant surgery. Ko et al. [36] evaluated the success rate and the marginal bone loss of immediate and delayed loading protocols after a fully-guided flapless surgery. Even though the success rate was lower in the immediate loading group, marginal bone loss was acceptable in both groups. Oh et al. [19] have calculated the angular deviation of prefabricated screw-type provisional restorations after single-implant placement in an experimental study on typodonts, which lays grounds for further clinical studies. The present study showed that discrepancies between the virtually planned and delivered provisional restorations should be expected. Although, further studies are still needed to evaluate prefabricated restoration’s accuracy and success.

Regarding the current study’s limitations, it might be necessary to include a greater sample size in future studies to yield more solid results. Applying other surgical and prosthetic protocols, as well as using different intraoral scanners and computer designing programs, might help form a comprehensive opinion about the accuracy of digital implant placement. Finally, employing and comparing other calculation methods of the implant placement accuracy (using different software and protocols) in future studies might help understand each method’s errors and limitations.

The implant placement accuracy using the DIONAVI protocol can be described by a mean angular deviation of 5.07° and a mean linear apical deviation of 1.74 mm, which might prove helpful to the clinician using it. Although these findings are in the higher range of the deviations calculated in the previous studies with different implant placement and calculation methods, further studies are required to reach a final verdict on the DIONAVI protocol’s accuracy. Regarding immediate-loading protocols and prefabricated provisional restorations, more research is vital to fully understand their pros and cons before arriving at a conclusion to apply them in routine care.