Accuracy of digital planning in zygomatic implants

Full-arch implant rehabilitations are currently one of the most common and predictable treatments in daily clinical practise, reaching high long-term survival rates [1]. As widely known, tooth loss leads to generalized resorption of alveolar process which can produce severe resorptions at maxillae and mandibular bone, preventing implant treatments due to the lack of sufficient bone height and width [2, 3]. Furthermore, the absence of dental stimuli and the positive pressure of the maxillary sinus increase, generating a progressive pneumatization of this cavity, further compromising bone availability [4, 5]. Several techniques have been developed with the aim of achieving sufficient bone augmentation to allow dental implant placement. These procedures, however, usually require a long surgical time, multiple interventions, and are oftenly associated with severe complications and high morbidity [6].

In 1989, Brånemark et al. [6] described the use of zygomatic implants as a therapeutic alternative for patients with severe maxillary atrophy to avoid bone augmentation procedures. This technique consisted in placing a long dental implant through maxillary sinus, to be anchored in zygomatic bone. Nevertheless, to avoid functional and aesthetic complications associated with pronounced concavities in the lateral maxillary wall [7], as well as other complications such as sinusitis [8, 9], Stella y Warner et al. [10] developed the “sinus slot” technique.

One of the key factors for zygomatic implants is the implant positioning. It is essential that the apex of the implant is placed in the areas of zygomatic bone with higher bone density and reaching the greater bone to implant contact (BIC) in all its course as far as possible. Furthermore, there is a wide variety of zygomatic implant brands. In this study, the implants used in all cases were NobelZygoma®. To ensure the correct osseointegration of these implants and knowing that they have an apical diameter of 3.75 mm, the minimum amount of bone required is 5.75 mm. Takamaru et al. [11] carried on a study in which determined that the most suitable areas for implant placing were the upper posterior area and the central area of zygomatic bone. On the other hand, authors such as Hung et al. [12] have determined, by using CBCT studies, several more suitable areas for the implant apex positioning, being the upper posterior area (A3) and the central area (B1) of the zygomatic bone the most favourable, agreeing with the results of Takamaru et al. [11]. Currently, one approach of zygomatic implant treatments is based in the “Quad Zygoma” concept which seeks to achieve complete upper restorations of atrophic maxilla by placing four implants [13, 14].

Scientific advances in computerized radiology machines [15] have allowed improvements in diagnostic and therapeutic tools. Although zygomatic implants surgical guides have been demonstrated to be inaccurate, most clinicians usually perform digital planification and virtual surgery prior to the intervention on the patient [16].

Nevertheless, and to the best of our knowledge, in current scientific literature, there are no studies which evaluate the clinical relevance of these VSP or studies which compares the variations in the position and angulation of implants placed conventionally with the previous digital planning.

The aim of this study is to evaluate if there are any advantages in using three-dimensional digital planning software for zygomatic implants placed in a conventional manner.

Fourteen zygomatic implants were placed in 4 patients, 2 males, and 2 females with a mean age of 48.75 years old. Preoperative VSP took a mean time of 10 min and 20 s. Twelve of the 14 placed implants had the same length as digitally planed (85.72%) (Table 1).

Regarding implant deviations, apical TD was the highest registered value with a mean of 6.114 ± 4.28 mm. One implant in position 1.6 had a greater apical TD deviation, (more than 10 mm). This result is due to limited buccal aperture of the patient, a variable difficult to take into account in VSP. While cervical ACD was the lower levels, with a mean deviation of 1.993 ± 1.83 mm. The analysis showed grater variations in BPD and MDD at apex and neck levels respectively with mean values of 4.221 ± 4.3 mm and 3.279 ± 2.34 mm. With respect to the AD, the mean value was 8.357 ± 5.3°. The T test and the Wilcoxon-test showed significant differences for all dependent variables (p < 0.05) (Table 2).

To the best of our knowledge, this is the first study in current scientific literature that evaluates the clinical utility of VSP for the placement of zygomatic implants in a conventional manner. The present study shows higher apical mean deviations compared with cervical deviations regardless implant length or position, being the TD and AD the highest. Furthermore, we observed some differences due to the influence of the length and the position of the implants, which seems to be accentuated when the implants are placed posteriorly and with the increase in the length of the implants.

VSP has been significantly advance in implantology, allowing clinicians to visualize the surgical procedure before performing the intervention over the patient. Moreover, it provides complementary information which helps to determine the number of implants, the adequate implant length, and its proper position [16]. Nevertheless, it requires a variable learning curve depending on individual characteristics.

CAD-CAM technology [17] has helped in developing minimally invasive surgical techniques guided by intraoral splints [18] or computer-assisted procedures [19‐21]. Treatments with conventional implants performed under surgical guides have achieved survival rates similar to conventional procedures [22], providing greater accuracy and precision than freehand techniques [23]. On the other hand, due to variations in the position and angulation of implants with respect to the previous computer-assisted planification zygomatic implants, surgical guides have demonstrated to be highly imprecise [18, 24].

With respect to apical variations observed, BPD are higher than MDD and ACD probably due to midface anatomy. After implant placement, the apex is located between the orbit anteriorly and the dermis of the face laterally. These two areas are susceptible to different complications such as orbit invasion, although very rare, has been described by various authors [25, 26], or dermis perforation, which could cause cutaneous fistulas [27]. For these reasons, the posterior area was where we observed most of the variations, because it is the zone that offers more manoeuvring possibilities. Regarding cervical variations, due to the clear clinical visualization of implant neck, higher and smaller deviations were MDD and ACD respectively because of the properly position and depth of the implant at bone crest level determines the result of the prosthetic phase [28] and reduce inflammatory complications, such as mucositis and peri-implantitis, as what occurred in conventional implants [29].

In addition, we observed some differences related to implant position and implant length. It seems that the more posterior the implant position, the more variations are recorded, more specifically in those implants placed in the right first molar position. These variations are probably associated with the length of the drills and mouth opening limitation, reason of which it is strongly recommended to start the drilling sequence in the posterior region [13, 14].

On the other hand, we observed some differences depending on whether the implant was in the first or second quadrant. These variations may be related to the right-handed condition of the operator. Greater deviation in most of the measured parameters in the first quadrant is probably due to the difficulty of handling rotary instruments in a non-comfortable working position.

However, the present study has some limitations. Due to the pilot study condition of our investigation, sample size is quite reduced, preventing us to reach statistical significance in most of compared variables related to implant length and position. Nevertheless, these preliminary data have allowed us to calculate that 35 patients are the necessary sample size to perform a second study.

Taking into account the above data, although the virtual planning of this type of surgery can help to determine certain parameters related to the characteristics of the implant and its position, the absence of an effective method that transfers this virtual planning to the surgical field, causes that the surgery is carried out in a conventional way, causing certain differences between the final implant position and the virtually planned one.

Zygomatic implant surgery is a complex surgical procedure, and although VSP is a useful tool which helps the clinician determine the number and the length of zygomatic implants as well as its proper position, surgical experience is still mandatory

Although our study reports some differences between VSP and the final treatment, we strongly suggest that further developments are necessary to allow the VSP to be transferred in an accurate manner from the computer to the surgery, by static or navigation-guided surgery.