Skip to main content

Zygomatic Implant Trajectory and Maxillary Sinus Clearance: A Patient-Specific Anatomical and Physiological Modeling Methodology for Predicting Postoperative Sinus Change

Paper type
Computational modeling methodology and retrospective validation protocol
Status
Original Work · Working Paper
Contents
  1. 1Introduction
  2. 2Background and Current Scientific Evidence
  3. 3Research Gap
  4. 4Research Objective and Central Hypothesis
  5. 5Scientific and Theoretical Framework
  6. 6Methodology
  7. 7Analytical Development
  8. 8Discussion
  9. 9Clinical Significance
  10. 10Limitations
  11. 11Future Validation
  12. 12Conclusion

Abstract

Background. Maxillary sinusitis is the most frequently reported complication of zygomatic implants. Pooled estimates range from 2.4% to 12% depending on technique and number of implants, and individual series report chronic sinusitis in 0% to 26.6% of patients. Outcomes are reported by surgical approach category (intrasinus, sinus slot, extrasinus or anatomy-guided). No study identified has quantified how a specific implant trajectory relates to the pathways by which the sinus clears mucus.

Objective. To define trajectory-specific measures of interference with maxillary sinus drainage, to specify a physiological model of mucociliary transport around an intrasinus obstacle, and to design a retrospective study that tests whether these measures predict postoperative sinus change better than approach category.

Framework. The sinus mucosal surface is reconstructed from computed tomography. Mucociliary transport is represented as a surface flow directed toward the natural ostium. An implant segment not covered by ciliated mucosa is an obstacle. Four quantities are computed: the transport-path interference index (the fraction of mucosal surface whose path to the ostium crosses the obstacle), the minimum distance from implant to ostium and infundibulum, the exposed intrasinus length, and the thickness of bone at the crestal entry. A thin-film model gives the conditions under which mucus pools upstream.

Proposed validation. A multicenter retrospective cohort with preoperative and late postoperative tomography, sinus-level analysis, and comparison of nested prediction models.

Central hypothesis. Trajectory-specific measures add predictive information beyond approach category, with ostial proximity carrying more weight than exposed length.

Conclusions. An analytical argument indicates that interference rises inversely with distance from the ostium, so most zygomatic trajectories, which run along the lateral wall, should obstruct little. If the retrospective study confirms this, attention should shift to the crestal seal and to pre-existing sinus disease. The methodology is proposed and no results are reported.

Keywords: zygomatic implants; maxillary sinus; mucociliary clearance; sinusitis; ostiomeatal complex; cone-beam computed tomography; computational model; atrophic maxilla

1. Introduction

The zygomatic implant was introduced to anchor a prosthesis in the severely atrophic maxilla without grafting. In the original technique, the fixture entered at the palatal aspect of the residual crest, passed through the maxillary sinus, and engaged the zygomatic bone [1]. From the outset, the intrasinus course raised a question that sinuscopy in a small early series answered reassuringly: after more than a year, the mucosa around fixtures in 14 patients showed no infection or inflammation [2]. Larger experience has been less uniform. A seven-year prospective study reported three patients with unilateral sinus infections refractory to antibiotics that required functional endoscopic sinus surgery [3], and sinusitis heads the list of complications in every systematic review [4–7].

Surgical technique evolved in response. The sinus slot approach moved the implant laterally and simplified orientation [8]. Extramaxillary anchorage placed the implant body outside the sinus wall [9]. The zygoma anatomy-guided approach (ZAGA) classified sites by the relation between the alveolar crest, the lateral sinus wall, and the zygomatic buttress and let that anatomy determine whether the path was intrasinus, partly intrasinus, or external [10, 11]. Success criteria for zygomatic rehabilitation now include rhinosinus status explicitly [12], and an international consensus has reviewed indications, techniques, and outcomes [13].

The comparative evidence is organized by these categories. A systematic review found sinusitis less frequent with the extrasinus approach and less frequent in recent than in older studies [14]. Another estimated the probability of sinusitis at 9.53% with the original technique and reported a lower figure with the anatomy-guided approach [7]. Such comparisons are confounded by era, case selection, loading protocol, and operator experience, and a category is a coarse description of what the implant does inside a particular sinus. Two intrasinus implants may differ in whether the apex passes near the infundibulum or far posterior to it, in whether the membrane was kept intact and displaced or torn, and in how much threaded surface lies bare in the cavity.

The physiology suggests which of those differences should matter. The maxillary sinus clears itself by mucociliary transport along the mucosal surface toward the natural ostium, high on the medial wall, against gravity [15–17]. A foreign body affects clearance according to where it sits relative to those paths and to the ostium, and whether it presents a non-ciliated surface. A search for studies combining zygomatic implants with mucociliary, airflow, or computational fluid dynamic modeling returned no records (Appendix A). This manuscript proposes a methodology to fill that gap: geometric measures of trajectory-specific interference, a physiological model that explains when interference leads to pooling, and a validation study against postoperative imaging.

2. Background and Current Scientific Evidence

Incidence of sinus complications

Estimates vary widely. A meta-analysis gave a sinusitis probability of 2.4% (95% CI 1.8–3.0) [4]. A review of surgical complications reported 3.9% [5]. For quadruple zygomatic implants, the pooled incidence was 12% (95% CI 4%–23%) [18]. An overview of seven systematic reviews covering 2,313 patients and 4,812 implants found a success rate of 96.72% and identified maxillary sinusitis as the most common surgical complication [6]. Sinusitis has been compared between sinus elevation and zygomatic surgery, with a pooled prevalence of 1.11% after sinus lift [19]. A long-term systematic review included sinusitis prevalence among its meta-analyzed outcomes [20]. Five-year prospective data are available for immediately loaded zygomatic implants [21].

One reason for the spread is that sinusitis is not uniformly defined. Jin and colleagues noted that reported rates of chronic sinusitis range from 0% to 26.6% and attributed the inconsistency in part to the absence of a uniform quantitative standard [22]. Clinical rhinosinusitis is a symptom-based diagnosis with endoscopic or radiographic confirmation [23]; many implant studies record only symptomatic cases requiring treatment, and others record any radiographic change.

Radiographic change after zygomatic implants

Studies with systematic imaging show that radiographic change is more common than clinical disease. The first study of sinus reactions to immediately loaded zygomatic implants used clinical criteria and computed tomography before and after surgery [24]. D’Agostino and colleagues scored mucosal thickening and opacification with the Lund–Mackay system [25, 26]. In the series by Jin and colleagues, the thickness and morphology of the Schneiderian membrane, infundibular obstruction, and the posterior sinus wall were analyzed; the number of fully opacified sinuses rose from none before surgery to six (13%) afterward [22]. Protrusion of conventional implants more than 4 mm into the sinus without membrane elevation has been examined in a small series for sinus complications [27], and sinus function after augmentation has been assessed by accepted diagnostic criteria [28]. Together these indicate that an implant in the sinus often provokes a mucosal response, that the response is usually subclinical, and that a minority of sinuses progress to complete opacification. Why some do is not explained by the category of approach.

Candidate mechanisms

Four mechanisms have been proposed in the clinical literature, and they are not mutually exclusive. The first is contamination through an oroantral communication at the crestal entry, where a long implant emerges through thin or absent alveolar bone and relies on a soft-tissue seal; recession or bone loss at this point opens a path from the mouth to the sinus, and this concern motivated both the anatomy-guided approach and the offset and soft-tissue criteria in current success codes [11, 12]. The second is obstruction of the ostium, by reactive mucosal thickening or by pre-existing disease. The third is interference with mucociliary transport by the implant body. The fourth is colonization of exposed threads acting as a nidus. Clinical studies have not been designed to apportion sinusitis among these.

Maxillary sinus clearance and its models

Cilia propel the mucus blanket across the sinus mucosa in a pattern that converges on the natural ostium, a pattern that persists when accessory openings are present, which is the physiological basis of functional endoscopic sinus surgery [15]. Ciliary function is impaired in chronic rhinosinusitis by acquired mechanisms [16]. A recent mathematical treatment modeled sinus mucus as a thin film on an active surface inside a fluid-producing cavity: a steady mucus layer develops only when the rate of ciliary clearance exceeds a threshold that depends on the rate of production, and gravity competes with cilia-driven drainage [29]. Numerical models of the maxillary sinus have addressed gas exchange and found its interaction with mucociliary transport negligible, with accessory ostia increasing ventilation [30]; computational fluid dynamics has characterized airflow in the nasal passages and sinuses of patients with chronic rhinosinusitis [31]. None of these models includes an intrasinus foreign body.

Summary

It is established that sinusitis is the leading complication of zygomatic implants, that its reported frequency depends on definition and technique, and that subclinical radiographic change is common. Mechanisms have been proposed but not separated. Mathematical models of sinus clearance exist. The link between a specific implant trajectory and clearance physiology has not been modeled or measured.

3. Research Gap

Three gaps are specific. No trajectory-level descriptor of the relation between a zygomatic implant and sinus drainage anatomy has been defined; existing descriptors classify the alveolar and lateral-wall anatomy and the resulting approach [10]. No physiological model predicts when an intrasinus implant will cause mucus retention. And no study has tested trajectory-specific measures against postoperative sinus outcomes with adjustment for the competing mechanisms, in particular the crestal seal and preoperative sinus status.

There is a further reason to address the question quantitatively. If interference with clearance is rarely significant for typical trajectories, effort directed at avoiding the sinus cavity may be better spent on the crestal entry and on preoperative management of sinus disease. A method that can show the absence of an effect is as useful here as one that shows its presence.

4. Research Objective and Central Hypothesis

The objective is to develop and validate a patient-specific method for evaluating zygomatic implant trajectories with respect to maxillary sinus clearance.

The central hypothesis is that, in a retrospective cohort with pre- and postoperative tomography, a model containing trajectory-specific measures (ostial proximity, transport-path interference, exposed intrasinus length, and crestal bone thickness at entry) predicts postoperative worsening of the maxillary sinus better than a model containing approach category and baseline sinus status alone, as judged by discrimination, calibration, and net benefit.

A secondary hypothesis, derived analytically in Section 7, is that ostial proximity is the dominant clearance-related predictor and that exposed length, once proximity is accounted for, contributes little. A second concerns mechanism: crestal bone thickness at the entry point predicts sinus worsening independently of clearance-related measures, consistent with an oroantral route.

The central hypothesis is refuted if the added measures do not improve prediction. In that event, the cohort would still provide the first adjusted estimates of the association between approach, baseline sinus status, and standardized radiographic outcome.

5. Scientific and Theoretical Framework

Geometry

From preoperative tomography, the maxillary sinus air space, the mucosal surface Σ, the natural ostium O (represented as a closed curve on Σ), and the infundibulum are segmented. From postoperative tomography registered to the preoperative volume, or from the surgical plan, the implant axis and diameter define a cylinder C. The intersection of C with the preoperative air space is the intrasinus segment. Whether that segment is covered by displaced, intact mucosa or lies exposed cannot be determined from the plan; it is assessed on postoperative imaging as soft-tissue density surrounding the implant and, where available, from the operative record. The exposed footprint Γ⊂Σ is the area of mucosal surface replaced or shadowed by non-ciliated implant surface, including a buffer of width b to represent disturbed mucosa at the margin.

Transport field

Mucociliary transport is represented by a tangential velocity field u_c on Σ. In the absence of patient-specific measurements, the simplest field consistent with the described convergence on the ostium is adopted: transport follows the surface paths of shortest length to O. Let ϕ(x) be the geodesic distance on Σ from point x to the ostium. Then

u_c(x)=−U (∇_sϕ)/(∥∇_sϕ∥)  (1)

with speed U and surface gradient ∇_s. The geodesic assumption is an approximation. Described transport routes are stereotyped and may not be shortest paths, and the assumption is one of the things the validation is intended to probe, by comparing predicted regions of retention with observed sites of mucosal thickening.

Geometric measures

Four measures are defined for each implant–sinus pair.

Transport-path interference index (TPI). Let Σ_Γ be the set of points on Σ whose transport path to the ostium, following equation (1), passes through Γ. Then

TPI=(area(Σ_Γ))/(area(Σ))  (2)

This is the fraction of the mucosal surface whose drainage is intercepted.

Ostial proximity d_O: the minimum distance (mm) from the implant surface to the ostium, and separately to the infundibulum.

Exposed length L_e: the length (mm) of the intrasinus segment not covered by mucosa on postoperative imaging.

Crestal bone thickness t_c: the thickness (mm) of alveolar bone surrounding the implant at its oral entry, measured circumferentially, with the minimum recorded. This measure belongs to the oroantral mechanism and is included so that clearance-related and seal-related effects can be distinguished.

Thin-film model of retention

For a subset of cases, a physiological model replaces the purely geometric index. Let h(x,t) be the mucus film thickness on Σ. Following the thin-film formulation for a fluid-producing ciliated cavity [29],

(∂h)/(∂t)+∇_s​⋅​(h u_c+(ρ g h^3)/3μ t_g)=q  (3)

where q is the mucus production rate per unit area (m/s), ρ density, μ viscosity, g gravitational acceleration, and t_g the tangential component of the gravity direction on the surface. On the footprint Γ, u_c=0 and q=0. The ostium is an outflow boundary. Steady solutions, where they exist, give the film thickness everywhere; regions where no bounded steady solution exists are regions of predicted pooling. Head position enters through t_g, and the model can be evaluated upright and supine.

Parameter values for U, q, and μ in the human maxillary sinus must be taken from primary physiological sources. They were not verified in this review and are not asserted here. The cited analysis reports that measured production and clearance rates in healthy sinuses fall within the regime of steady drainage [29], which implies some reserve; how much reserve remains after partial interception is what equation (3) is meant to estimate.

6. Methodology

Stage 1: pipeline development and verification

Sinus air space and bone are segmented with an open platform [32]; the mucosal surface is taken as the air–soft tissue boundary where the sinus is aerated. The ostium is marked by two trained readers on multiplanar reconstructions, and inter-reader agreement on its location is reported. Geodesic distance on the triangulated surface is computed by a standard fast-marching or heat-method algorithm. The pipeline is verified on analytic shapes (a sphere and an ellipsoid with a circular outlet) for which geodesic paths and TPI for a cylindrical obstacle can be computed in closed form. Mesh-convergence of TPI is documented [33]. Reproducibility is quantified by intraclass correlation for each measure across readers and across repeat segmentations [34].

The thin-film solver is verified against the published behavior of the model in idealized cavities [29] before patient geometries are used.

Stage 2: retrospective cohort

Design and setting. Multicenter retrospective cohort, reported to STROBE [35], with the prediction analysis reported to TRIPOD+AI [36].

Participants. Consecutive adults who received one or more zygomatic implants and have preoperative tomography including the full maxillary sinus and ostiomeatal complex and postoperative tomography at 12 months or later obtained for clinical reasons. Exclusions: prior sinus surgery on the study side; oncologic resection; scans that do not include the ostium. No imaging is acquired for research [37].

Unit of analysis. The sinus. Bilateral cases contribute two sinuses, and sinuses with two implants are analyzed with implant-level measures summarized as the maximum TPI, minimum d_O, total L_e, and minimum t_c.

Outcome. The primary outcome is radiographic worsening of the maxillary sinus, defined as an increase of at least one point in the maxillary component of the Lund–Mackay score (0, clear; 1, partial opacification; 2, complete opacification) between the preoperative and late postoperative scans [26], read by two assessors blinded to implant measures with the implant masked where feasible. Secondary outcomes: complete opacification; change in maximum mucosal thickness; ostial or infundibular obstruction; clinically diagnosed rhinosinusitis by accepted criteria [23]; sinus-related intervention; and oroantral communication.

Predictors. The four measures of Section 5.3; approach category as recorded and ZAGA type [10]; baseline Lund–Mackay score, mucosal thickness, and ostial patency; accessory ostia; number of zygomatic implants in the sinus; implant surface in the intrasinus portion (threaded or machined); smoking; immediate loading; center.

Sample size. The outcome frequency is uncertain. If radiographic worsening occurs in roughly 13%–15% of sinuses, a figure suggested by one series for complete opacification [22] and adopted here only as a planning assumption, then a model with about eight candidate parameters developed under current sample-size guidance for prediction models [38] would need on the order of 100 or more events, corresponding to about 700 sinuses. The exact number will be computed from the outcome frequency observed in the first contributing center and the anticipated model performance. Clinical sinusitis is too infrequent, at 2%–12% [4, 18], to serve as the primary outcome at a feasible sample size.

Analysis. Mixed-effects logistic regression with random intercepts for patient and center. Three nested models are compared: (A) baseline sinus status and patient factors; (B) model A plus approach category; (C) model B plus the trajectory-specific measures. Performance is assessed by the c-statistic, calibration slope and intercept [39], and decision-curve analysis [40], with optimism corrected by bootstrap and with internal–external cross-validation by center. The secondary hypothesis on the dominance of ostial proximity is examined by the standardized coefficients and by the change in performance when each measure is removed. Continuous predictors are not dichotomized.

Missing data. Exposure of the intrasinus segment may be indeterminate on imaging; this is recorded as a category, and sensitivity analyses treat indeterminate segments as exposed and as covered. Other missing covariates are handled by multiple imputation.

Stage 3: physiological simulation in a nested sample

Among cohort sinuses, a case–control sample (all sinuses progressing to complete opacification, and matched sinuses without change) undergoes thin-film simulation with literature-based parameter ranges. The outputs, pooled volume and maximum steady film thickness, are compared between cases and controls, and the spatial agreement between predicted pooling regions and observed mucosal thickening on the postoperative scan is quantified by overlap measures. A global sensitivity analysis over U, q, μ, buffer width b, and head position identifies which uncertainties dominate.

Stage 4: in silico comparison of alternative trajectories

For preoperative scans in the cohort, alternative trajectories that satisfy anchorage constraints (minimum length of engagement in zygomatic bone, acceptable prosthetic emergence, and clearance from the orbit and infratemporal fossa) are generated, and the four measures are computed for each. The output is the range of each measure attainable within the constraints for each ZAGA type. This shows how much control the surgeon actually has over clearance-related measures, and whether reducing interference necessarily worsens crestal bone thickness or emergence position.

Bias and confounding

Patients with late postoperative tomography are selected: scans are more likely to have been ordered for symptoms, which inflates outcome frequency and may distort associations. Centers that scan routinely at follow-up are prioritized, and the indication for each scan is recorded and used in a sensitivity analysis restricted to routine scans. Approach category is confounded with era and operator; calendar year and center are modeled. Baseline sinus disease is the strongest expected predictor and is measured with the same instrument as the outcome. Readers are blinded as described, although the presence of an implant on postoperative images cannot be fully concealed.

Ethics

Institutional review board approval or exemption is required at each center, with de-identified image transfer under data-use agreements.

7. Analytical Development

Interference scales inversely with distance from the ostium

Consider the neighborhood of the ostium, where transport paths converge. Approximate the surface locally as a plane and the ostium as a point sink toward which paths converge radially from a sector of opening angle Θ (radians). At radial distance r from the ostium, the arc available to transport has length Θr. An obstacle of width a measured perpendicular to the local transport direction, centered at distance r, intercepts a fraction

B(r)≈a/(Θ r),  r>a/Θ  (4)

of all paths, and therefore of all surface area draining through that sector if production is uniform. All of the area upstream of the obstacle within its angular shadow is intercepted. The index TPI is the area-weighted form of this quantity on the true surface.

The dependence on 1/r has a direct implication. An implant of 4-mm diameter crossing the transport field 20 mm from the ostium, with paths arriving from a half-plane (Θ=π), intercepts roughly 4/(20π), about 6% of paths. The same implant 4 mm from the ostium intercepts about a third. These numbers are illustrative values from equation (4) under its simplifying assumptions; they are not measurements. They indicate that position relative to the ostium should matter far more than the length of implant inside the sinus.

Condition for pooling

Upstream of an obstacle, intercepted mucus must be carried around it. In steady state, mass conservation across the gap beside the obstacle requires the film there to thicken. For a locally uniform approach flow of thickness h_0 in a channel of width W partly blocked over width a, and neglecting gravity,

(h_(gap))/h_0≈W/(W−a)  (5)

Ciliary propulsion is effective only for a film up to some thickness h_c, beyond which the upper layer is not coupled to the cilia. Pooling begins when h_(gap) exceeds h_c, that is, when the blocked fraction exceeds

(a/W)_(crit)=1−h_0/h_c  (6)

A healthy sinus, with h_0 well below h_c, tolerates a large blocked fraction. A sinus with thickened secretions, reduced ciliary speed, or pre-existing mucosal disease has h_0 close to h_c and tolerates little. Baseline sinus status and geometric interference should therefore interact: the same trajectory is innocuous in one sinus and decompensating in another. The analysis plan includes this interaction term.

Where typical trajectories lie

The natural ostium opens high on the medial wall. A zygomatic implant, whatever the approach, runs from the alveolar crest laterally and upward toward the zygoma, along or through the lateral wall. For most trajectories the implant is therefore on the opposite side of the cavity from the ostium, at a distance comparable to the sinus width. Equation (4) then predicts a small TPI for most cases, and equation (6) predicts that pooling occurs only when baseline reserve is already depleted. On this reasoning, clearance obstruction by the implant body is unlikely to be the principal cause of sinusitis in typical cases, and the observation that sinuscopy found healthy mucosa around intrasinus fixtures [2] is what the model would expect.

Exceptions should exist. Quadruple zygomatic placement puts an anterior implant closer to the medial wall, and the pooled sinusitis incidence is higher in that setting [18], although the comparison is uncontrolled. Small or septated sinuses shorten all distances. An apex or drill path that approaches the infundibulum is a rare event that would have a disproportionate effect. The retrospective cohort is designed to detect whether sinuses with unusually small d_O account for a disproportionate share of opacifications.

Gravity and the sinus floor

The crestal entry is at the floor, the most dependent part of the sinus in the upright position and the origin of the longest transport paths. Equation (3) includes a gravitational flux proportional to h^3, which opposes upward transport and becomes important as the film thickens. An exposed footprint at the floor combined with an oroantral leak, which adds fluid and bacteria at the same location, places the disturbance where clearance is already working hardest. The model thus gives a physiological reason, in addition to the microbiological one, for the clinical emphasis on the crestal seal [12], and it predicts that t_c and floor-level exposure will be associated with dependent, floor-based opacification, whereas small d_O will be associated with diffuse opacification secondary to ostial obstruction. The pattern of opacification is recorded so that this can be examined.

8. Discussion

The methodology replaces a categorical description of surgical approach with continuous, physiologically motivated measures of what an implant does in a given sinus. Its most useful result may be to narrow the field of plausible mechanisms. The analytical argument suggests that the implant body rarely obstructs clearance to a meaningful degree, because it is usually far from the ostium, and that when sinus disease follows zygomatic surgery the cause is more likely to lie at the crestal entry or in a sinus that was already compromised. If the cohort bears this out, the long-running comparison of intrasinus and extrasinus paths would be reframed: the benefit of external approaches would be attributed to better crestal bone and soft-tissue conditions at the entry point, and an intrasinus path with a sound crestal seal in a healthy sinus would not be expected to carry excess risk.

The alternative outcomes are also informative. If TPI or exposed length predicts worsening independently of ostial proximity and crestal thickness, the foreign-body and nidus mechanisms gain support, and membrane preservation and smooth intrasinus surfaces acquire a stronger rationale. If nothing but baseline sinus status predicts outcome, preoperative sinus assessment and treatment become the priority and trajectory, within the range surgeons use, is secondary.

Several assumptions could undermine the model. The geodesic transport field is a guess at a pattern that has been described qualitatively; real pathways may be more stereotyped and could, for example, run along the floor and up the medial wall in a way that makes floor-level obstacles more consequential than equation (4) implies. Ciliary function is not uniform and is degraded by the very inflammation the implant may cause, a feedback the steady model ignores. The mucosa is displaced and remodeled after surgery, so preoperative geometry is an imperfect template. And the distinction between covered and exposed implant surface, which determines the footprint, is difficult to make on tomography with metal artifact.

The outcome also deserves scrutiny. A one-point change in the Lund–Mackay maxillary score is coarse and includes mild mucosal thickening of doubtful significance; restricting to complete opacification is more specific but reduces events. Clinical sinusitis is the outcome that matters to patients and is too rare for a primary analysis. The secondary outcomes are graded to allow readers to judge whether associations seen with the radiographic outcome are carried by the clinically relevant end of the spectrum.

Airflow was deliberately omitted. Ventilation and gas exchange in the sinus have been modeled [30, 31], and an implant near the ostium could affect them, but the interaction of gas exchange with mucociliary transport was found to be negligible in the one model that examined it, and adding a second physical model would increase complexity without a clear hypothesis. It could be added if ostial proximity proves important.

9. Clinical Significance

If validated, the measures could be computed at the planning stage from the tomography already obtained and displayed alongside anchorage and prosthetic information. A plan with small ostial distance or with thin crestal bone at entry in a sinus with reduced baseline reserve would be flagged, and the in silico stage would show what alternatives exist for that anatomy. The interaction between baseline status and interference offers a principled basis for preoperative referral: the same trajectory may be acceptable after sinus disease is treated and inadvisable before.

For follow-up, the model predicts patterns: floor-based opacification with thin crestal bone points to the entry site; diffuse opacification with ostial narrowing points to drainage. Recognizing the pattern could direct management toward local repair or toward restoring ostial patency.

These uses are contingent on a validation that has not been done. The measures should not be used clinically on the strength of this manuscript.

10. Limitations

The transport field is assumed, and no patient-specific measurement of mucociliary direction or speed is available from routine imaging. Physiological parameters for the thin-film model were not verified and may be poorly constrained in the literature. Exposure of the intrasinus segment is uncertain on imaging. The retrospective design carries selection by indication for late imaging, confounding of approach with era and operator, and heterogeneity of scanners and protocols. Cone-beam scans often truncate the ostiomeatal complex, which will exclude cases and may bias the sample toward centers using wider fields of view. The sample-size reasoning rests on an outcome frequency from a single series. The analytical results in Section 7 use planar approximations and uniform production and are intended to indicate scaling, not to supply predictions for individuals. Ventilation, bacterial factors, and host immune status are not modeled. The literature was assessed from abstracts and index records.

11. Future Validation

A prospective cohort with standardized preoperative and 12-month tomography in all patients, independent of symptoms, would remove selection by indication and allow the prediction model to be validated externally. Endoscopic assessment of mucociliary transport in a subset, for example by dye or saccharin transit from the floor to the ostium before and after surgery, would test the transport-field assumption directly. If ostial proximity is confirmed as a predictor, a planning study could evaluate whether displaying it changes trajectories and whether changed trajectories preserve anchorage. Independent groups should recompute the measures from shared, de-identified image sets to establish reproducibility across software.

12. Conclusion

How a zygomatic implant affects the maxillary sinus has been studied by comparing categories of surgical approach. This manuscript proposes measures that describe each trajectory in relation to the ostium, the mucociliary transport paths, and the crestal seal, together with a thin-film model of when interception leads to retention. A scaling argument predicts that interference grows as the inverse of distance from the ostium and that most trajectories, lying along the lateral wall, intercept little, so that baseline sinus reserve and the integrity of the crestal entry should dominate. The retrospective study described would test these predictions. Until it is performed, they remain propositions derived from anatomy and a simplified physical model.

Cite this paper

Jain S, Gupta A. Zygomatic Implant Trajectory and Maxillary Sinus Clearance: A Patient-Specific Anatomical and Physiological Modeling Methodology for Predicting Postoperative Sinus Change. Working paper. Center for Implant Dentistry; 2026. Available from: https://bayareaimplantdentistry.com/clinician-research-and-publications/zygomatic-trajectory-sinus-clearance

References

  1. Brånemark PI, Gröndahl K, Ohrnell LO, Nilsson P, Petruson B, Svensson B, et al. Zygoma fixture in the management of advanced atrophy of the maxilla: technique and long-term results. Scand J Plast Reconstr Surg Hand Surg. 2004;38(2):70-85. doi:10.1080/02844310310023918 (opens in a new tab). PMID: 15202664 (opens in a new tab).
  2. Petruson B. Sinuscopy in patients with titanium implants in the nose and sinuses. Scand J Plast Reconstr Surg Hand Surg. 2004;38(2):86-93. doi:10.1080/02844310310023909 (opens in a new tab). PMID: 15202665 (opens in a new tab).
  3. Bedrossian E. Rehabilitation of the edentulous maxilla with the zygoma concept: a 7-year prospective study. Int J Oral Maxillofac Implants. 2010;25(6):1213-21. PMID: 21197500 (opens in a new tab).
  4. Chrcanovic BR, Albrektsson T, Wennerberg A. Survival and Complications of Zygomatic Implants: An Updated Systematic Review. J Oral Maxillofac Surg. 2016;74(10):1949-64. doi:10.1016/j.joms.2016.06.166 (opens in a new tab). PMID: 27422530 (opens in a new tab).
  5. Molinero-Mourelle P, Baca-Gonzalez L, Gao B, Saez-Alcaide LM, Helm A, Lopez-Quiles J. Surgical complications in zygomatic implants: A systematic review. Med Oral Patol Oral Cir Bucal. 2016;21(6):e751-e757. doi:10.4317/medoral.21357 (opens in a new tab). PMID: 27694789 (opens in a new tab).
  6. Sales PH, Gomes MV, Oliveira-Neto OB, de Lima FJ, Leão JC. Quality assessment of systematic reviews regarding the effectiveness of zygomatic implants: an overview of systematic reviews. Med Oral Patol Oral Cir Bucal. 2020;25(4):e541-e548. doi:10.4317/medoral.23569 (opens in a new tab). PMID: 32520920 (opens in a new tab).
  7. Kämmerer PW, Fan S, Aparicio C, Bedrossian E, Davó R, Morton D, et al. Evaluation of surgical techniques in survival rate and complications of zygomatic implants for the rehabilitation of the atrophic edentulous maxilla: a systematic review. Int J Implant Dent. 2023;9(1):11. doi:10.1186/s40729-023-00478-y (opens in a new tab). PMID: 37198345 (opens in a new tab).
  8. Stella JP, Warner MR. Sinus slot technique for simplification and improved orientation of zygomaticus dental implants: a technical note. Int J Oral Maxillofac Implants. 2000;15(6):889-93. PMID: 11151591 (opens in a new tab).
  9. Maló P, Nobre Mde A, Lopes I. A new approach to rehabilitate the severely atrophic maxilla using extramaxillary anchored implants in immediate function: a pilot study. J Prosthet Dent. 2008;100(5):354-66. doi:10.1016/S0022-3913(08)60237-1 (opens in a new tab). PMID: 18992569 (opens in a new tab).
  10. Aparicio C. A proposed classification for zygomatic implant patient based on the zygoma anatomy guided approach (ZAGA): a cross-sectional survey. Eur J Oral Implantol. 2011;4(3):269-75. PMID: 22043470 (opens in a new tab).
  11. Aparicio C, Manresa C, Francisco K, Claros P, Alández J, González-Martín O, et al. Zygomatic implants: indications, techniques and outcomes, and the zygomatic success code. Periodontol 2000. 2014;66(1):41-58. doi:10.1111/prd.12038 (opens in a new tab). PMID: 25123760 (opens in a new tab).
  12. Aparicio C, López-Piriz R, Albrektsson T. ORIS Criteria of Success for the Zygoma-Related Rehabilitation: The (Revisited) Zygoma Success Code. Int J Oral Maxillofac Implants. 2020;35(2):366-378. doi:10.11607/jomi.7488 (opens in a new tab). PMID: 32142574 (opens in a new tab).
  13. Al-Nawas B, Aghaloo T, Aparicio C, Bedrossian E, Brecht L, Brennand-Roper M, et al. ITI consensus report on zygomatic implants: indications, evaluation of surgical techniques and long-term treatment outcomes. Int J Implant Dent. 2023;9(1):28. doi:10.1186/s40729-023-00489-9 (opens in a new tab). PMID: 37698775 (opens in a new tab).
  14. Tavelli C, Tedesco A. Survival and complication rate of zygomatic implants: a systematic review. J Oral Implantol. 2022. doi:10.1563/aaid-joi-D-22-00008 (opens in a new tab). PMID: 36473176 (opens in a new tab).
  15. Stammberger H. Endoscopic endonasal surgery–concepts in treatment of recurring rhinosinusitis. Part I. Anatomic and pathophysiologic considerations. Otolaryngol Head Neck Surg. 1986;94(2):143-7. doi:10.1177/019459988609400202 (opens in a new tab). PMID: 3083326 (opens in a new tab).
  16. Gudis D, Zhao KQ, Cohen NA. Acquired cilia dysfunction in chronic rhinosinusitis. Am J Rhinol Allergy. 2012;26(1):1-6. doi:10.2500/ajra.2012.26.3716 (opens in a new tab). PMID: 22391065 (opens in a new tab).
  17. Sleigh MA, Blake JR, Liron N. The propulsion of mucus by cilia. Am Rev Respir Dis. 1988;137(3):726-41. doi:10.1164/ajrccm/137.3.726 (opens in a new tab). PMID: 3278666 (opens in a new tab).
  18. Lan K, Wang F, Huang W, Davó R, Wu Y. Quad Zygomatic Implants: A Systematic Review and Meta-analysis on Survival and Complications. Int J Oral Maxillofac Implants. 2021;36(1):21-29. doi:10.11607/jomi.8417 (opens in a new tab). PMID: 33600519 (opens in a new tab).
  19. Rocha RS, Vianna CP, Trojan LC, Padovan LEM, Dos Santos MCGL. Comparison of sinusitis rate after sinus lift procedure and zygomatic implant surgery: a meta-analysis. Oral Maxillofac Surg. 2024;28(1):63-77. doi:10.1007/s10006-023-01159-1 (opens in a new tab). PMID: 37266797 (opens in a new tab).
  20. Brennand Roper M, Vissink A, Dudding T, Pollard A, Gareb B, Malevez C, et al. Long-term treatment outcomes with zygomatic implants: a systematic review and meta-analysis. Int J Implant Dent. 2023;9(1):21. doi:10.1186/s40729-023-00479-x (opens in a new tab). PMID: 37405545 (opens in a new tab).
  21. Davó R, Malevez C, Pons O. Immediately loaded zygomatic implants: a 5-year prospective study. Eur J Oral Implantol. 2013;6(1):39-47. PMID: 23513201 (opens in a new tab).
  22. Jin S, Ma H, Ling X, Tao B, Wang F, Wu Y. Longitudinal reactions of maxillary sinus in patients treated with multiple zygomatic implants: A modified radiographic evaluation with clinical follow-up. Clin Implant Dent Relat Res. 2023;25(6):1091-1102. doi:10.1111/cid.13256 (opens in a new tab). PMID: 37574767 (opens in a new tab).
  23. Fokkens WJ, Lund VJ, Hopkins C, Hellings PW, Kern R, Reitsma S, et al. European Position Paper on Rhinosinusitis and Nasal Polyps 2020. Rhinology. 2020;58(Suppl S29):1-464. doi:10.4193/Rhin20.600 (opens in a new tab). PMID: 32077450 (opens in a new tab).
  24. Davó R, Malevez C, López-Orellana C, Pastor-Beviá F, Rojas J. Sinus reactions to immediately loaded zygoma implants: a clinical and radiological study. Eur J Oral Implantol. 2008;1(1):53-60. PMID: 20467644 (opens in a new tab).
  25. D’Agostino A, Trevisiol L, Favero V, Pessina M, Procacci P, Nocini PF. Are Zygomatic Implants Associated With Maxillary Sinusitis?. J Oral Maxillofac Surg. 2016;74(8):1562-73. doi:10.1016/j.joms.2016.03.014 (opens in a new tab). PMID: 27070844 (opens in a new tab).
  26. Lund VJ, Mackay IS. Staging in rhinosinusitus. Rhinology. 1993;31(4):183-4. PMID: 8140385 (opens in a new tab).
  27. Jung JH, Choi BH, Jeong SM, Li J, Lee SH, Lee HJ. A retrospective study of the effects on sinus complications of exposing dental implants to the maxillary sinus cavity. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2007;103(5):623-5. doi:10.1016/j.tripleo.2006.09.024 (opens in a new tab). PMID: 17257862 (opens in a new tab).
  28. Timmenga NM, Raghoebar GM, Boering G, van Weissenbruch R. Maxillary sinus function after sinus lifts for the insertion of dental implants. J Oral Maxillofac Surg. 1997;55(9):936-9;discussion 940. doi:10.1016/s0278-2391(97)90063-x (opens in a new tab). PMID: 9294502 (opens in a new tab).
  29. Desai N, Lauga E. Modelling Mucus Clearance in Sinuses: Thin-Film Flow Inside a Fluid-Producing Cavity Lined with an Active Surface. Bull Math Biol. 2024;86(11):134. doi:10.1007/s11538-024-01360-7 (opens in a new tab). PMID: 39367965 (opens in a new tab).
  30. Hood CM, Schroter RC, Doorly DJ, Blenke EJ, Tolley NS. Computational modeling of flow and gas exchange in models of the human maxillary sinus. J Appl Physiol (1985). 2009;107(4):1195-203. doi:10.1152/japplphysiol.91615.2008 (opens in a new tab). PMID: 19608923 (opens in a new tab).
  31. Kumar H, Jain R, Douglas RG, Tawhai MH. Airflow in the Human Nasal Passage and Sinuses of Chronic Rhinosinusitis Subjects. PLoS One. 2016;11(6):e0156379. doi:10.1371/journal.pone.0156379 (opens in a new tab). PMID: 27249219 (opens in a new tab).
  32. Fedorov A, Beichel R, Kalpathy-Cramer J, Finet J, Fillion-Robin JC, Pujol S, et al. 3D Slicer as an image computing platform for the Quantitative Imaging Network. Magn Reson Imaging. 2012;30(9):1323-41. doi:10.1016/j.mri.2012.05.001 (opens in a new tab). PMID: 22770690 (opens in a new tab).
  33. Roache PJ. Perspective: A Method for Uniform Reporting of Grid Refinement Studies. Journal of Fluids Engineering. 1994;116(3):405-413. doi:10.1115/1.2910291 (opens in a new tab).
  34. Koo TK, Li MY. A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research. J Chiropr Med. 2016;15(2):155-63. doi:10.1016/j.jcm.2016.02.012 (opens in a new tab). PMID: 27330520 (opens in a new tab).
  35. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370(9596):1453-7. doi:10.1016/S0140-6736(07)61602-X (opens in a new tab). PMID: 18064739 (opens in a new tab).
  36. Collins GS, Moons KGM, Dhiman P, Riley RD, Beam AL, Van Calster B, et al. TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ. 2024;385:e078378. doi:10.1136/bmj-2023-078378 (opens in a new tab). PMID: 38626948 (opens in a new tab).
  37. Bornstein MM, Scarfe WC, Vaughn VM, Jacobs R. Cone beam computed tomography in implant dentistry: a systematic review focusing on guidelines, indications, and radiation dose risks. Int J Oral Maxillofac Implants. 2014;29 Suppl:55-77. doi:10.11607/jomi.2014suppl.g1.4 (opens in a new tab). PMID: 24660190 (opens in a new tab).
  38. Riley RD, Ensor J, Snell KIE, Harrell FE Jr, Martin GP, Reitsma JB, et al. Calculating the sample size required for developing a clinical prediction model. BMJ. 2020;368:m441. doi:10.1136/bmj.m441 (opens in a new tab). PMID: 32188600 (opens in a new tab).
  39. Van Calster B, McLernon DJ, van Smeden M, Wynants L, Steyerberg EW. Calibration: the Achilles heel of predictive analytics. BMC Med. 2019;17(1):230. doi:10.1186/s12916-019-1466-7 (opens in a new tab). PMID: 31842878 (opens in a new tab).
  40. Vickers AJ, Elkin EB. Decision curve analysis: a novel method for evaluating prediction models. Med Decis Making. 2006;26(6):565-74. doi:10.1177/0272989X06295361 (opens in a new tab). PMID: 17099194 (opens in a new tab).