Computer-Guided Approach for Placement of Zygomatic Implants: Novel Protocol and Surgical Guide

Since the inception of modern dentistry a solution has been sought to help the world's population who are missing one or more teeth, or to stabilize existing teeth. The first evidence of the use of dental implants to replace missing teeth was with the Mayan population (around 600 AD) in their use of shell fragments for lost mandibular teeth,¹ which were similar in appearance to that of more modern blade implants developed by Linkow.² A stone implant was discovered to have been used in early Honduran culture (around 800 AD) also in the mandible.¹ Dental implantology continued to evolve over hundreds of years with varying designs and differing levels of success. From Greenfield's attempts around 1913 with iridio-platinum artificial root replacements, to the Stock brothers in the 1930s who mimicked orthopedic surgeons' use of hip implants to develop a Vitallium screw to restore single missing teeth, researchers and clinicians worldwide searched for the ideal biocompatible material to be used intraorally.^3,4 Dental implant designs have been plentiful: spiral-shaped implants, double-helical spiral implants with threads into the bone (endosseous), subperiosteal implants that sat on top of the bone, flat-plate blade implants that were wedged into the bone, implants that used the ramus bone for support, and transosseous implants (staple implants) for severely atrophic mandibles.

It was not until Dr. Per-Ingvar Brånemark noticed the attachment of bone to pure titanium, a concept now known as "osseointegration," that the design and biomaterial choice of a dental implant that could actually "bond" to the host bone was fully realized. Brånemark's success in the mid to late 1960s with a two-piece pure titanium endosseous external-hex dental implant design was noted worldwide.⁵ Many newer designs followed in an effort to improve long-term results for patients who required dental implants. During the past five decades the most popular tooth replacement method has been the use of endosseous or root-form dental implants.

Innovation has continued among global implant manufacturers who have strived to improve the grade of titanium used, the surface treatment of the metal, and the connection of the transgingival component abutments to the implant. Other developments have included the addition of restorative alternatives to improve esthetics and function, reduced time for osseointegration, immediate loading, and guided surgical protocols.

Not all patients are candidates for root-form dental implants when there is a deficiency of bone volume, height, and width. Therefore, ancillary procedures also have been developed that utilize bone and soft-tissue grafting to improve potential dental implant receptor sites. These procedures include but are not limited to ridge splitting, block bone grafting, particulate bone grafting, sinus augmentation, and use of concomitant biological membranes or autologous blood products. When patients are severely atrophic in the maxillary arch, however, bone grafting procedures alone may not be sufficient to provide anchorage for conventional root-form implants, or such procedures may require several sequential surgical interventions over many months or even years without a guarantee of successful outcomes. Thus, new treatment modalities were sought for patients with minimal available bone in the maxillary arch. Again, it was Professor Brånemark's research that helped provide a solution.

In the late 1980s, concerned that there were still patients with severely resorbed maxillas being left untreated, Brånemark further researched the maxillary craniofacial bone, focusing on its ability to support dental implants. He found that the zygomatic bone demonstrated the required load-bearing qualities, density, and sufficient volume, and subsequently developed a protocol for the placement of implants in this bone. In 1998 Nobel Biocare began the production of zygomatic implants.⁶ These specially designed, elongated implants were first produced and utilized for the rehabilitation of oncological patients who had undergone maxillary resections, meaning that there was no other bone available to anchor support for a prosthesis. Once proof of concept showed evidence of success, zygomatic implants were then used to rehabilitate partial and fully edentulous patients with advanced degrees of atrophic bone. For more than 20 years the use of zygomatic implants has demonstrated a predictable and safe alternative treatment modality for the complex reconstruction of bones in the maxillary jaw, exhibiting a high percentage of success.^7-12

For an implant to traverse from the maxillary alveolar crestal region to the zygoma requires modifications of existing root-form implant designs to extended lengths ranging from 30 mm to more than 50 mm. In addition, to accommodate longer implant lengths, new and longer sequential drills are needed to engage the zygomatic bone. The surgical intervention, including flap design and sufficient tissue reflection, demands thorough knowledge of the local vital anatomic structures, including the infraorbital foramen and nerve, the medial part of the zygomatic body, and the zygomatic arch. The classic and original surgical protocol provided a straight-line access to point of entry into the zygomatic bone, a path in which the implant went through the maxillary sinus. A lateral sinus wall fenestration was required to visualize and control the direction of the drills. The fenestration is critical to avoid damaging anatomical structures such as the orbital floor, infraorbital nerve, or the vascular content of the infratemporal fossa.

Existing variations in maxillary bone morphologies made it difficult to advocate only one surgical protocol for zygomatic implants. The intrasinusal position of the implants carried from the zygoma to a point of prosthetic anchorage resulted in an intraoral position that often was too palatal, causing discomfort and speech/phonetic problems for patients. Therefore, various modifications to the original protocol have subsequently been proposed that place the prosthetic connection in a more centered position within the alveolar crest bone.

According to a review by Chrcanovic and Abreu, five different surgical approaches were identified: (1) the classic approach, (2) the sinus slot technique, (3) the exteriorized approach, (4) the minimally invasive approach by use of custom-made drill guides, and (5) the computer-aided surgical navigation system approach.¹³ In 2000 Stella and Warner described the sinus slot technique, which provided the path for the zygomatic implants in a furrow prepared on the lateral surface of the maxillary sinus; in this way the implant was only partially positioned internally into the sinus and the prosthetic connection could be placed in a more favorable site.^14,15 Subsequent authors reported excellent results using the Stella-Warner technique.^16-18 Then in 2011 Aparicio proposed a classification of zygomatic implant patients based on the anatomy of the maxilla (ie, "ZAGA": zygomatic anatomy guided approach) by establishing five different paths (ZAGA 0 through ZAGA 4) for zygomatic implants in relation to their position, be it partially or totally intra- or extra-sinus.¹⁹ Therefore, depending on the topography and morphology of the maxillary arch, the receptor site of the zygomatic implant should be chosen after a careful and accurate anatomical study.

As use of zygomatic implants became an acceptable treatment alternative additional authors published on different approaches and protocols reporting short- and long-term results.^20-22 Despite the high success rates of zygomatic implant rehabilitation, these procedures were not free of complications. Incidence of rhino-sinus (rhino-sinusitis) complications were found in a significant population (up to 37.5%) following the insertion of trans-sinus zygomatic implants.^13,23 To reduce the risk of maxillary sinusitis, Chow et al in 2010 published a variation of the surgical technique to extend the sinus fenestration, whereby the fragmented bone remained attached to the Schneiderian membrane and the fragmented bone was pushed toward the interior of the maxillary sinus after delicate and careful detachment from the membrane.²⁴ With this technique, the authors attempted to avoid extensive lesions (injuries) to the membrane and minimize the risk of oral-sinus communications.

As with the placement of any dental implant, the "goal" is not the implant itself, but rather the restoration that will replace the tooth. Therefore, although originally a surgical approach only, the use of zygomatic implants also should be restoratively driven with the intent of providing both a functional and esthetic outcome. Boyes-Varley et al utilized a device to evaluate the best insertion axis of the zygomatic implant,²⁵ and recently Fortin reported positive results inserting zygomatic implants in correspondence with the infra-zygomatic crest.²⁶ As proposed, the prosthetic connection can be situated in a more favorable site in correspondence with the first molar. These contributions to the literature demonstrate that besides the intra-sinus paths of the original protocol, other variants can be used. In fact, implants can be inserted in various positions that are derived from accurate presurgical anatomic and pre-prosthetic evaluation.

The presurgical anatomic assessment to define the desired zygomatic implant positions cannot be performed using panoramic radiograph or computed tomography (CT) 2-dimensional (2D) images, but requires a 3-dimensional (3D) view of the maxillary arch. Ideally, planning for the insertion of zygomatic implants can be accomplished with either a CT, or even more appropriate, a cone-beam computed tomography (CBCT) study and then analyzed with interactive treatment planning software. The operational challenge is to perform accurate presurgical diagnostics to clearly define the entrance point, trajectory path, and exit point of the zygomatic implant combined with successful transfer of the implant planning to the surgical phase, as can be accomplished with conventional guided solutions for conventional implants. Surgical guides used in computer-guided implantology are the instruments used to transfer the 3D implant planning done on the computer to the patient during the surgical intervention.^27,28 The ability to simulate the implant positions on a computer screen with data from the CT/CBCT provides clinicians with increased diagnostic accuracy for fabrication of CT-derived surgical guides. Many guided-surgery systems have received scientific validation for clinical use after errors in axial and angular deviations have been correlated.^29-32 These static guides, whether "template-assisted" or "full-template guidance" as defined by the authors,^33,34 produced improvements when compared with "freehand" surgical intervention.

Zygomatic implantology, however, differs because, due to the great lengths of these implants, even small errors in angular deviation could significantly alter the trajectory and, therefore, the positions of the apex, risking potential grave complications to the patient.^35,36 Additionally, surgical guides that have been used for the placement of zygomatic implants fabricated in the same manner of standard-length implants do not provide any trans-sinus control during execution of the osteotomy. Thus, surgeons often prefer to carry out the preparation of the implant sites in a freehand method to facilitate visual control of the drilling process.

In zygomatic implant surgery performed freehand, the presurgical studies are often limited with regard to the use of software or simulation on 3D models. The surgeon certainly could be "inspired" by these preliminary assessments but lack the means to transfer these evaluations to the intraoral surgical phase with precision; thus, decisions relating to the entry point, path, and exit point of the implants will need to be made during surgery. Because the analysis of every possible option requires a thorough pre-prosthetic and presurgical clinical study with computer simulation and a stereolithographic model, the extemporaneous choice made directly during the surgery may possibly not represent the best option.

Essentially, a surgical guide for the placement of zygomatic implants fabricated in the same manner as standard implants may be considered less reliable in terms of angular deviation, stability, and not permitting direct control of the drills. In contrast, freehand surgery does not allow the direct transfer of the 3D CT/CBCT-derived implant planning to the surgical phase, forcing surgeons to make too many extemporaneous decisions based on their training and the existing anatomic reality.

To aid in the definitive planning of zygomatic implants, the authors strongly advocate the use of modern 3D printing to fabricate a realistic, accurate, life-size surface model of the maxillary bone to scale. Until recently, however, the ability to produce a 3D model from a CT/CBCT dataset was limited to most clinicians due to cost and a lack of local stereolithographic production facilities. Prosthetic and surgical planning with 3D CT/CBCT diagnostic technologies, CAD CAM software, and merging of datasets, along with advances in 3D printing are vital for use in providing accurate solutions in zygomatic implantology. This article, therefore, introduces a novel protocol in collaboration with the development of a specially designed, specific surgical guide to assist clinicians in overcoming many issues associated with the placement and restoration of zygomatic implants.

The protocol first requires the diagnosis and study of the anatomy of the maxillary bone from a large field of view (FOV) CT or CBCT scan to determine the location of zygomatic implant receptor sites using 3D planning software. Once the plan has been defined, the dataset will allow for the construction of a CT/CBCT-derived surgical guide of a novel design, which will be exported as an STL (standard triangulation language) file to be fabricated using a stereolithography or 3D printing process. Concomitantly, through the use of software segmentation, an exact replica of the entire maxilla and zygomatic bone is also exported in STL format to be fabricated via 3D printing in actual scale model size.

The next step, which is critical, entails the simulation of the operation on the 3D-printed biomedical model using exact replicas of the implants to be used during the actual surgical intervention. If the simulation is successful, and the implant positions are in agreement with the planned receptor sites, the surgical guides can be used for the surgery. If any deviation from the plan materializes, the entire diagnostic procedure must be repeated until a successful validation is accomplished.

There were four patients represented in the present study, two female and two male. The females were ages 53 and 57, and the males were ages 59 and 65. All four patients were in good physical health, ASA 1 to ASA 2, and each presented with a completely edentulous maxillary arch. Each patient's CT/CBCT examination was required to have a large enough FOV extending from the alveolar process inclusive of at least half of the orbit to highlight the entire body of the zygomatic bone.

The digital imaging and communications in medicine (DICOM) data from the 3D scan first is exported from the imaging device and then imported into the implant planning software. Through the use of the diagnostic tools within the software, the entire maxillary anatomy is studied for any anatomical variations, density, topography of the maxillary sinus, and potential pathology. Consequently, the alveolar bone and zygomatic bone are evaluated for receptor sites that may be most suitable for insertion of the zygomatic implants. The simulation for the implant placement is used to define the entrance point, trajectory path (implant path), and exit point of the implants (Figure 1 and Figure 2).

As with conventional guided implant planning, there is not one view that contains enough information for definitive planning. Therefore, clinicians must utilize both the 3D and 2D views in the various spatial planes, depending on the software application used. These include coronal, axial, panoramic, cross-sectional, and sagittal oblique.

As previously stated, a critical requirement for this protocol is that a realistic, accurate, life-size surface anatomical model of the maxillary bone to proper scale is fabricated using 3D printing technology with the surgical guide(s) based on the 3D plan. It is important to note that conventional bone-supported surgical guides as fabricated for standard-length implants have several characteristics that render them unsuitable for the zygomatic surgery. The authors have identified potential weak points related to conventional bone-supported computer-guided techniques that are applicable to zygomatic implants³⁰ and submit the following four considerations: (1) Bone-supported surgical guides may be affected by any irregularity of the support base on the alveolar crest. (2) The surface of the alveolar bone crest, which provides support to the guide, often is limited in ensuring stability to the guide. (3) Surgical guides may lean or deflect with the longer drills being used, which are required to accommodate the unusual extraoral positions necessary for zygomatic implants. (4) Surgical guides could hamper the necessary intra-sinusal control of the drill's path.

To overcome these drawbacks the present authors have designed a specific, novel surgical guide for the insertion of zygomatic implants that allows for visualization of the maxillary sinusal fenestration (Figure 3 and Figure 4). The surgical guides as presented were constructed using rapid prototyping, 3D printing (Objet Eden260V, Stratasys, stratasys.com), or a stereolithographic technique (Materalise, materialise.com) with biocompatible material (Med610, Stratasys) based on the 3D computerized implant planning from the CT/CBCT scan. Metal sleeves or guide cylinders used to guide the drills 4 mm in vertical height were manually inserted into the pre-defined holes in the resin during the finishing phase of the surgical guide.

The surgical protocol has been adopted from the NobelZygoma^™ implant system (Nobel Biocare, nobelbiocare.com), which involves sequential preparation of the implant receptor sites using first, a round drill, followed by a 2.9-mm diameter drill. For this protocol the drill diameter of 3.5 mm was used either for the Brånemark Zygoma RP System (Nobel Biocare) or for the NobelZygoma 45° implants, while 4-mm diameter and 4.4-mm diameter drills are required for the NobelZygoma 0° implants. The zygoma implant lengths range between 30 mm and 52.5 mm. Two or more surgical guides were constructed with metal sleeves suitable for diverse drill diameters. For the NobelZygoma RP implants and NobelZygoma 45° implants the authors first utilize the guide with sleeves for the 2.9-mm diameter drill (Guide I), followed by a second guide with sleeves that are appropriate for a 3.5-mm diameter drill (Guide II). To use the initial round drill, a novel adapter created from polyether ether ketone (PEEK) material was inserted inside of the 3.5 mm sleeve (Guide II). The protocol can then be used with all zygomatic implant systems presently on the market.

The perimeter of the guide is specifically designed so that it extends on the lateral bone surface of the maxillary sinus and has an elongated section that follows and embraces the zygomatic process. This extended form (Figure 5) provides an prolonged area of support on the bone surface for improved stability and accompanies the drill until it reaches the zygomatic body, arriving very close to the exit point.

The support of the guide on both the zygomatic process and the lateral surface of the sinus wall is an essential component of the protocol, especially in comparison to the support provided by commonly fabricated surgical guides, which is limited by the alveolar bone crest. Importantly, in the interior of the guide a rectangular space oriented upward enables visualization of the sinus fenestration as well as lateral access to it, which, in accordance with the Chow technique,²⁴ will be pushed inside of the maxillary sinus, protecting the Schneiderian membrane from the movement of the drills, and diminishing risks of sinusitis and oro-antral communication. Therefore, the retained bony window functions as a shield to protect the sinus membrane from direct damage by the drills. The sinus fenestration window follows the direction of the zygomatic implant and extends from the infra-zygomatic crest toward the back and upward, reaching the zygomatic body, and allows visual control of the drills during every phase of the preparation of the implant site on the zygomatic body. In case of two zygomatic implants as in the quad technique,³⁷ the fenestration must be much wider to accommodate visualization of both implants (Figure 6).

To design the external perimeter of the guide and the sinus fenestration, a modeling software is used with which the operator, with direction from the surgeon, carries out simulated cuts to achieve the proper positioning of the sinus fenestration and trace the external perimeter of the guide as a function of retention zones and raised support on the anatomy. This technique was derived from the previously defined design of the novel "sinus-lift guide" by the authors using either rotary or piezosurgery inserts to access the lateral sinus wall.³⁴ In summary, the zygomatic surgical guides permit the accurate transfer of 3D planning to the patient, guiding the drills during the preparation of the implant receptor sites.

Furthermore, the extension of the guide directs the piezoelectric inserts or rotary instruments to create the outline of the lateral wall to complete the sinus fenestration. The authors recommend that two separate guides be fabricated, one for each side of the maxillary arch, right and left. A single surgical guide crossing over the midline could create an obstacle in the use of the drills, which due to their length are used from the opposite side, often from an extraoral position.

The simulation phase is critical and essential because it allows the clinician to perform all the steps of the surgery first on the stereolithographic anatomical scale model, utilizing the surgical guides up until the positioning of analogous zygomatic implants (replica implants or simulation kit). Through the simulation process it is possible to re-evaluate the accuracy of the transfer of the computerized planning to the model, identifying any issues prior to the surgical intervention (see below).

Simulation: sinus windows preparation-With the surgical guide positioned on the stereolithographic model, the perimeter of the sinus fenestration is represented as a quadrangular space within the interior of the guide itself. To realize the osteotomies, ultrasonic instruments are used directly inside the guide, or the guide can be used to design the perimeter of the sinus fenestration on the model (Figure 7). Depending on the thickness of the lateral wall, rotary instruments also may be used in conjunction with piezosurgery instrumentation.

Simulation: guided osteotomy preparation-To prepare the zygomatic implant sites, the round bur is the first drill used. It is used with the 3.5 mm sleeve incorporated into the surgical guide (Guide II) after the PEEK adapter has been inserted (Figure 8 and Figure 9). The preparation of the implant receptor sites requires that the entrance points are first marked on the alveolar crest and, with the direction of the drill being controlled through the sinus fenestration, the posterior-superior parts of the roof of the maxillary sinus are marked to allow seating of the 2.9 mm drill. The guide is then substituted with one that incorporates the guide sleeve for the 2.9 mm twist drill (Guide I), continuing with the guided preparation with this drill until arriving at the exit point on the outer face of the zygomatic bone (Figure 10 and Figure 11). Next, the guide that incorporates the 3.5 mm guide sleeve (Guide II) is used along with the pilot drill of this diameter to accurately widen the osteotomy. To complete the guided preparation, the 3.5 mm twist drill is used. The depth of the implant site preparation is controlled by positioning specific drill stops on the drills (2.9 mm and 3.5 mm) that are to be used for the surgery (Figure 12).

Simulation: zygomatic implant placements-To finish the preparation of the zygomatic implant sites, the simulations can be completed by positioning "training zygomatic implants" (NobelZygoma replicas) or other replicas of zygomatic implants that the authors are studying (simulation kit) (Figure 13).

After the simulation, accuracy of the computer-aided planning and its transfer to the stereolithographic 3D printed model is then evaluated. If the evaluation is satisfactory, the clinician may proceed to the surgical phase. For a more precise "postsurgical" evaluation, the horizontal and angular deviations may be examined in millimeters and degrees in relation to the computerized planning, as described by Testori et al.³⁸

As a second "guided" method, some surgical specialists may prefer to avoid computerized planning and perform implant placement freehand on the stereolithographic anatomical model. Freehand placement of replica implants can still accomplish a "guided" result through the optical scanning of the anatomical model and the recording of the implant positions (Figure 14). Once scanned, the digitized information is introduced into the 3D planning software and merged with the DICOM data. The position of each implant can then be translated into the digital plan for fabrication of surgical guides, which can be used at the time of surgical intervention. When there is opportunity to provide a postoperative CT/CBCT scan, a direct comparison of the proposed implant positions in the computerized project can then be made to evaluate the accuracy of the plan as transferred from the stereolithographic anatomic maxillary model compared with the actual patient placement.

The surgical procedure begins with the elevation of a wide mucogingival flap to obtain an extended exposure of the underlying maxillary bone; this highlights the alveolar crestal bone, anterolateral wall of the maxillary sinus, suborbital foramen, orbital frame, zygomatic process, and zygomatic buttress. Once a sufficient flap has been raised, the placement of the surgical guide must be accurate, stable, and adherent to the bone-bearing surface. Any interference with the gingival flap must be prevented, and the guide must not be tilting in any direction. The design of the guide incorporates a few small "windows" or "cut-outs" to help the clinician visualize that the guide is in perfect contact with the bone. Also, the quadrangular area that defines the perimeter of the sinus window allows for precise control of the surgical guided position at the level of the zygomatic body. It is important to note that the guide can have only one position of stability, and its correct, delicate placement requires careful attention. Ideally, some level of practical experience in guided surgery is needed.

After the guide is stabilized, the lateral sinus wall can be addressed. Proceeding with piezoelectric inserts, four osteotomies are performed within the guide along the perimeter of the sinus fenestration. Through use of sinus lifting instruments, the Schneiderian membrane can then be carefully detached and the bony window in-fractured or pushed within the sinus cavity, allowing enough room for the transit of the drills and subsequently the zygomatic implants. The guided preparation of the implant receptor sites within the zygomatic body can then be performed in vivofollowing the steps previously described for the simulation model surgery (Figure 15 and Figure 16), and the zygomatic implants placed as planned (Figure 17 and Figure 18). After insertion of the zygomatic implants, the space above the sinus fenestration can be grafted with bone, biomaterials, or autologous blood products, such as platelet-rich fibrin, and covered with a membrane. The soft-tissue flap must have sufficient release to allow for primary tension-free closure with carefully applied horizontal mattress sutures to complete the operation.

To evaluate whether the actual surgical placement of the zygomatic implants matched the computerized planning and simulation, the preoperative positions must be compared with the postoperative implant positions. Merging the preoperative CT scan data with the postoperative CT after implant placement makes it possible to measure the degree of accuracy of the superimposition using sophisticated software (Figure 19 through Figure 22).³¹

The procedure was accomplished as follows: First, extraction/conversion and segmentation was performed to export the STL files of the surface bone reconstructed volume analysis and the actual implants from both the original 3D planning project and the postoperative CT scan using two specific software applications (RealGUIDE 5.0, 3Diemme srl, 3diemme.it; Mimics, Materialise). Second, superimposition and registration of the common bony parts not modified by surgery was performed to acquire the simulated implant designs and those actually positioned in a comparable position. Third, the STL file of every implant extracted from the postoperative CT is often characterized by metal scattering, which determines poor quality and low resolution of the images; therefore, this file was substituted with the STL file of the corresponding implant in the software library (using Geomagic Studio 12, Geomagic Inc, 3dsystems.com). Fourth, math calculation was performed, comparing, in millimeters and degrees, the deviation in correspondence to the base and apex for each pair of implants in the two situations that were computer planned, and then post surgery (Figure 23).

Using the superimposition of the pre- and postoperative CT scans, the positions of the 10 zygomatic implants were evaluated. The results of this evaluation are reported in Table 1 and Table 2. Because two different types of software were used with two different procedures to analyze the results, two separate tables are provided to report an overall statistical calculation of the results.

The following average values were obtained: In "Table 1, Values of Apical and Coronal Deviations…"the apical deviations were recorded as 2.11 mm; the coronal deviations were recorded as 3.55 mm; and the angular deviations were 4.55 degrees. In Table 2, the values recorded were as follows: apical deviations, 2.99 mm; coronal deviations, 2.96 mm; angular deviations, 1.88 degrees. Thus, deviations from the computerized/simulated project were between 2 mm and 3 mm, and angular deviations were between 1.88 and 4.55 degrees.

These values are not directly comparable with those reported studies of accuracy relative to standard implants^29-32 for various reasons. The zygomatic implants are inserted freehand and not through a surgical guide, and thus are characterized as "template-assisted" and not "full template guidance."^33,34 The length of the drills and zygomatic implants are three to five times greater than standard implants. The training or replica implants always have the same length. Furthermore, contrary to what happens with standard implants, zygomatic implants purposely "perforate" the zygomatic bone at the exit point and the final position of the implants is determined and based on the torque and orientation of the prosthetic connection. Therefore, the values in relation to length cannot be considered in the evaluation of accuracy. In studies of guided surgery, the vertical deviations are also always considered (√ x2s + y2s + z2s) though they cannot be compared with the aforementioned deviations; only the planar of the zygomatic implants, which is not considered a vertical component of the deviation, may be compared (√ x2s + y2s). Therefore, in Table 1 and Table 2 only the x and y values should be considered.

Preliminary studies related to the use of computer-guided surgery for positioning zygomatic implants have been carried out under experimental conditions.³⁵ When zygomatic implants were placed in three formalin-fixed human cadavers using surgical drilling guides authors found that "… the use of surgical drilling guides should be encouraged for zygoma implant placement because of the lengths of the implants involved and the anatomical intricacies of the region."³⁶ Promising results have been reported and companies have been able to produce accurate and precise surgical guides.³² However, the manufacturers involved in those preliminary studies opted to abandon the ongoing multicenter analyses and not to engage in mass production of surgical guides for zygomatic implants,³⁹ as the guides actually can be produced upon specific individual request by the surgeon. The motivation behind this decision centered on problems associated with the "accuracy chain" due to small differences in the various clinical operator-dependent steps. Zygomatic implants pass quite near important anatomical structures and any deviation of the drills could cause serious injury to the patient; therefore, the use of "blind surgery," as in trusting the accuracy of the technology for the placement of zygomatic implants, is not advocated, and implant-guided surgery has been practically abandoned because of the problem with angular deviation. In fact, a small angular deviation related to the great length of these implants could significantly change the position of the apex. Thus, it seems that the problem of angular deviation may be the greatest obstacle in the use of guided implant surgery for the placement of zygomatic implants.

Vrielinck and co-workers reported the values of angular deviation of 5.14 degrees for zygoma implants in their apical portion.³⁵ Using the dynamic navigation technique, other authors have reported variable values of angular deviation of 4.1 degrees +/- 0.9 degrees³³ and 1.37 degrees +/- 0.21 degrees.⁴⁰ The values reported in these studies are of the deviation of the implant axis, which is potentially dangerous overall if techniques are used that do not permit visual control or inspection of the actions of the drills. Based on the evaluation of the results obtained with the protocol from the present authors' proposals in terms of horizontal and angular deviations in relation to the implant planning, the authors cannot refer to other works reported in literature relating to zygomatic guided implantology; because this is an original technique there are no similar cases with which to make direct comparisons. Therefore, the results of the present study will be compared with the placement method of the most diffused zygomatic implants, by a freehand approach, in response to the important question: "Can guided surgery improve safety, precision, and predictability results in zygomatic implantology?"

The answer lies within the analysis of the most common complications reported during the placement of zygomatic implants. Chrcanovic and Abreu preformed a systematic review of the literature in an attempt to determine the survival rate of zygomatic implants and the most common complications related to zygomatic implant surgery.¹³ Their review found reports of 70 cases of sinusitis, 48 of soft-tissue infection, 15 of paresthesia, and 17 of oro-antral fistulas. The authors of this present study consider these values to underestimate potential complications from zygomatic implant surgery. As there remains a controversial relationship between zygomatic implants and maxillary sinusitis, it seems that the sinusitis may be related to an oro-antral communication and to exposure of implant threads, while the soft-tissue complication at the abutment level has been interpreted as being due to the prosthetic connection being positioned too palatal.¹⁷ Also, two clinical studies described penetrations into the orbit, and in one case an intra-cerebral penetration.⁴¹ If there is an error in the trajectory, the drills may penetrate the infratemporal and pterygopalatine fossa, the nasopharyngeal, and the sphenoid sinus. There have also been reports of periorbital hematomas, nasal bleeding, and bone fractures.^42,43

The placement of zygomatic implants requires surgical experience, and there are risks of placement being in too close proximity to vital anatomical structures. Therefore, having visual control of the drills during preparation of the implant sites is essential, even when using computer-guided systems or dynamic navigation. Also, problems related to angular deviation exist due to the inherent length of zygomatic implant fixtures, and, thus, visual control of the path of the drills is required. Statistically, the most common complications associated with zygomatic implants are maxillary sinusitis, infection of the soft tissue, and oro-antral communications,¹³ Additionally, other, less frequent but grave complications, such as intra-cranial penetrations,⁴¹ penetrations of the orbit, lesions at nerve trunks, and even zygomatic bone fracture, have been reported.

Most of these complications may be related to an inadequate preoperative study and/or errors in the surgical technique executed, most often involving a freehand protocol. In a highly variable anatomical scenario with various degrees of bone atrophy the surgeon may decide the path of the drills extemporaneously and often may lose guidance as related to the prosthetic outcome. Implants positioned with an angulation that is too palatal may predispose for an infection of the soft tissue at the prosthetic connection level and could lead to continual discomfort and phonetic problems. Implants positioned with an angulation too vestibular may predispose the exposure of the implant threads and an oro-antral communication. An adequate presurgical study helps to avoid lesions on neighboring anatomical structures.

In the current protocol presented in this study, 3D computerized implant planning was used to assess the existing patient anatomy to predetermine the path of the zygomatic implants and enhance the ability to transfer the plan data to the patient during the surgical phase using a specific surgical guide.³⁵ This guide allows clinicians to replicate in vivothe design of the sinus fenestration as well as the entrance and exit points of the implants, and also to control the depth of the osteotomy while continually maintaining total visual control of the operative field. This protocol leverages the advantages of computer-guided implantology in terms of the presurgical study and implant planning, avoiding the risks associated with surgery without visual control. Additionally, the guided preparation of the sinus fenestration is very important for maintaining the integrity of the Schneiderian membrane and diminishing the risks of postoperative maxillary sinusitis.

The preliminary results have been evaluated and appear to be quite favorable with regard to angular deviation values. The values of some horizontal deviations, however, may relate to an imperfect placement of the surgical guide (Figure 19 through Figure 22; Table 1 and Table 2). Apart from the mathematical measurements, the images of the superimposition illustrate in every case good correspondence between the inserted implants and those planned, and, importantly, no implants damaged any adjacent vital anatomical structures.

Using the surgical guide design described here, the surgeon has constant visual control of the drilling protocol and the positioning of the guide in close proximity to the entry point of the zygomatic body, which aids in controlling the drills up to the vicinity of the exit point; this drastically limits problems associated with angular deviation. Underlining that, in the authors' experience this guided computer approach neither simplifies nor facilitates the surgery, but has shown in both brief and long-term results that it improves the final quality of the execution, as evaluated in the latest statistical analysis. Despite the authors' extremely positive experience and being convinced that 3D technology represents the future in zygomatic implantology for treating patients with severe bone deficiencies, further research in this area, with larger sample sizes and additional statistically significant data, is encouraged.

Marco Rinaldi, MD, DMD
Private Practice,
Bologna, Italy

Scott D. Ganz, DMD
Private Practice,
Fort Lee, New Jersey