Comparison of trabecular bone structure using fractal dimension analysis in patients undergoing lateral window and transcrestal sinus lift procedures: a retrospective cohort study

Abstract

Background

This retrospective study evaluated alterations in trabecular bone structure after lateral window and transcrestal sinus floor elevation techniques using fractal analysis of panoramic radiographs, with a focus on comparing ossification patterns.

Materials and methods

This STROBE compliant retrospective radiographic cohort included 46 sinus lift procedures (transcrestal without graft: n = 21; lateral window with xenograft: n = 25) performed between 2020 and 2024. For each case, two panoramic radiographs were analyzed on the surgery week (T0) and at 6 months (T1) yielding 92 images. Regions of interest (30 × 30 px) were placed apical to the implant site. Images were pre-processed according to the White–Rudolph protocol, and fractal dimension (FD) was calculated in ImageJ using the box-counting method. Normality was confirmed; paired t-tests assessed within-group change and independent t-tests compared groups. Intra-rater reliability was evaluated using the intraclass correlation coefficient.

Results

In the transcrestal group, a significant increase in FD values was observed between T0 and T1 (p < 0.001). In contrast, the lateral window approach group showed a statistically significant decrease in FD values over time (p = 0.0032). The FD values in the lateral window approach group were significantly higher than in the transcrestal group at both T0 (p < 0.001) and T1 (p = 0.009).

Conclusion

Fractal analysis demonstrated distinct bone remodeling patterns between transcrestal and lateral window sinus lifting techniques. In the transcrestal group, the natural healing process was characterized by more consistent increases in FD values, whereas xenograft-associated remodeling in the lateral window group was accompanied by a decline. These findings underscore the potential of fractal analysis as a non-invasive modality for monitoring bone quality following augmentation.

Clinical trial number

Not applicable.

Introduction

Dental implants, owing to their biocompatibility and ability to restore function and aesthetics, have become the preferred option for replacing missing teeth and supporting prosthetic rehabilitation [1]. Following tooth loss, significant resorptive changes occur in the alveolar processes, leading to progressive loss of volume in the jawbones [2]. This condition reduces the volume and density of the alveolar bone available for subsequent dental implant placement [3]. While sinus pneumatization has traditionally been regarded as the primary cause of reduced bone height following tooth loss, current literature attributes this condition mainly to alveolar crest resorption [4, 5].

To overcome the problem of insufficient alveolar bone volume, various augmentation and reconstruction techniques have been employed to promote bone regeneration and thereby increase the width and height of the residual ridge prior to dental implant placement [6]. In cases of bone volume deficiency in the posterior maxilla, a surgical procedure known as sinus lifting is performed, aiming to increase vertical bone height and thereby facilitate the placement of dental implants of adequate length. This can be achieved either by placing a bone graft between the Schneiderian membrane and the sinus floor or by utilizing alternative techniques such as graftless crestal sinus lift [7–9].

The maxillary sinus lift procedure was first introduced by Boyne and James and has since been refined through various modifications [9, 10]. Currently, two main techniques are employed; the lateral window and the transcrestal sinus floor elevation approaches [11].

The transcrestal sinus floor elevation technique is regarded as a conservative approach, offering advantages such as low morbidity, reduced infection risk, and shorter treatment time, and it can also be applied in cases with limited residual bone height [12]. The lateral window technique, on the other hand, requires the creation of a bony window on the lateral maxillary wall to elevate the Schneiderian membrane and place graft material, and has been established as a predictable and reliable method for bone augmentation [13].

Although autogenous bone has long been regarded as the gold standard for grafting because of its regenerative potential, its limited availability and risk of resorption have prompted the use of alternative biomaterials, among which inorganic xenografts owing to their osteoconductive capacity and slow resorption rate are considered particularly suitable for maxillary sinus augmentation [14–16].

Recent literature on sinus grafting indicates that long term success rates may vary depending on the type of bone graft material used, with clinical evidence showing that these grafts typically undergo maturation into bone within approximately six months [17–19]. It has been reported that bone formation in the maxillary sinus does not necessarily require biomaterials; the essential factor is maintaining a stable space for blood clot formation, which allows osteogenic cells from the sinus periosteum and cancellous maxillary bone to migrate and remodel. The Schneiderian membrane, with its intrinsic osteogenic potential, can also facilitate bone formation when guided bone regeneration principles are applied, even in the absence of particulate grafting materials [7, 20–22]. This outcome may be achieved through the use of autologous blood derivatives, volume-stabilizing materials, and/or gelatin sponge in conjunction with simultaneous implant placement [23, 24].

Most studies evaluating the outcomes of sinus augmentation have focused on histological or volumetric analyses, which are often invasive, time consuming, or costly. Consequently, there is growing interest in non-invasive digital methods, among which fractal dimension (FD) analysis offers particular promise. Fractal analysis (FA) is a quantitative approach that characterizes complex geometries and structural patterns, expressed numerically through the FD. FD provides distinct advantages as it is easily accessible, and unaffected by variables such as projection geometry or radiodensity, while delivering objective and reproducible data on trabecular microarchitecture. The alveolar bone, with its trabecular architecture composed of spicules, trabeculae, and lamellae, can be quantitatively assessed using FD, a mathematical invariant that describes self-similar gray-level variations. By capturing the complex yet repeating patterns of biological structures, FD has emerged as a reliable method for detecting and quantifying changes in bone mineral content on radiographic images [25–33].

The objective of this study was to assess grafted regions using FA applied to panoramic radiographs taken at baseline (T0) and after six months (T1) in patients treated with either the lateral window approach or the transcrestal technique, and to compare the resulting ossification patterns.

Materials and methods

Study design and sample

This investigation was conducted as a retrospective cohort study at the Department of Oral and Maxillofacial Surgery, Faculty of Dentistry, Afyonkarahisar Health Sciences University, Afyonkarahisar, Türkiye. Patients who underwent sinus floor elevation procedures using either the transcrestal approach without graft placement or the lateral window approach with xenograft application between 2020 and 2024 were identified. Comprehensive medical and dental records, together with pre- and postoperative radiographic archives, were systematically reviewed, and patients were enrolled according to the following criteria. Patient selection was based exclusively on existing clinical and radiographic records, and no prospective guidance or follow-up protocol was applied. Given the retrospective nature of the study, stringent inclusion and exclusion criteria were established to ensure maximum standardization of surgical procedures. These criteria were intended to minimize potential sources of bias and to reduce methodological errors to the lowest possible level. The selection process of the study population is summarized in Fig. 1. A total of 148 cases were initially screened according to the inclusion criteria. After the exclusion of cases with incomplete radiographic records, systemic conditions affecting bone metabolism, local sinus pathologies, previous surgical interventions, or staged sinus augmentation protocols without imultaneous implant placement, 46 patients (92 radiographs) were finally included in the analysis.

Fig. 1.

Flowchart of patient selection

Inclusion criteria

• Age between 25 and 65 years.

• Classified as ASA I or II according to the American Society of Anesthesiologists.

• No systemic diseases affecting bone metabolism (e.g., osteoporosis, Paget’s disease, diabetes mellitus).

• No radiologically detectable sinus pathology (e.g., mucosal thickening, cystic lesions).

• Underwent either transcrestal sinus elevation without graft placement or lateral window approach sinus augmentation with the same xenograft together with simultaneous implant placement on the day of surgery.

• Availability of both surgery week (T0) and postoperative (T1, six months after surgery) digital panoramic radiographs of adequate diagnostic quality.

• The first of the bilateral augmentations performed on different days.

• To prevent potential bias related to surgical duration and to minimize statistical error, only one of the bilateral augmentations performed on the same day was randomly included in the study.

Exclusion criteria

• Incomplete radiographic records or absence of either T0 or T1 radiographs.

• Radiographs with insufficient diagnostic quality due to artifacts, motion blurring, or improper positioning.

• Lack of standardized radiographic parameters across time points.

• Previous surgical or pathological conditions.

• History of previous maxillary sinus surgery or sinus augmentation procedures.

• Cases treated with staged sinus augmentation protocols without simultaneous implant placement.

• Presence of sinus pathologies such as chronic sinusitis, mucosal thickening > 5 mm, polyps, or cystic lesions.

• History of oroantral fistula, traumatic injuries, or other pathological conditions involving the sinus.

• Systemic diseases or conditions affecting bone metabolism (osteoporosis, Paget’s disease, uncontrolled diabetes mellitus, hyperparathyroidism, chronic renal disease).

• Patients who had undergone radiotherapy or chemotherapy involving the head and neck region.

• Current or previous use of drugs known to alter ossification and bone metabolism, including bisphosphonates, denosumab, systemic corticosteroids, antineoplastic chemotherapy agents, or long-term anticoagulant therapy.

• Use of immunosuppressive or chronic anti-inflammatory drugs.

• History of smoking or alcohol consumption.

• Cases involving additional pin fixation (confirmed by medical records and radiographic documentation).

• Two-stage lateral sinus lift procedures without simultaneous implant placement.

• Residual alveolar bone height less than 2 mm.

• The second of the bilateral augmentations performed on different days.

Ethical considerations

Ethical approval was obtained from the Non-Invasive Clinical Research Ethics Committee of the Afyonkarahisar Health Sciences University, Afyonkarahisar, Türkiye (Approval No: 2025/8-364-13.06.2025). The study was conducted in accordance with the principles of the Declaration of Helsinki. As all images were obtained during routine clinical care, no additional radiation exposure was introduced. All participants provided written informed consent.

Sample size determination

Power analysis was performed using G*Power 3.1.9.7 (Institute of Experimental Psychology, Heinrich Heine University, Düsseldorf, Germany) to determine the minimum required sample size. As recommended by Cohen (1988), a moderate effect size (Cohen’s d = 0.55), a two-sided significance level of 0.05, and 80% power were assumed [34]. This analysis indicated that at least 84 observations (42 per group) would be required. In the present study, a total of 92 observations were included (42 in the transcrestal group and 50 in the lateral window group), which corresponded to a modest increase over the minimum required sample size and yielded a final statistical power of approximately 83%.

Surgical procedures

All surgical procedures were carried out in accordance with the routine protocols of the our department. Clinical and radiographic records were retrospectively reviewed to extract detailed operative information. Cases in which adjunctive procedures outside the routine protocol were employed for example, the application of autologous blood derivatives or additional regenerative adjuncts were excluded from the study.

In the routine protocol of our department, orthopantomograms are obtained within one week postoperatively. Accordingly, the temporal radiographic data utilized in the present study were derived from this standardized workflow. Furthermore, whether bilateral sinus lift procedures had been performed simultaneously or at different time points was clarified based on this protocol. In cases performed at different time points, only the first augmentation was included in the study in order to prevent potential confounding.

In our department, the choice of surgical technique is determined according to the clinically accepted standard protocol, in which the residual alveolar bone height serves as the primary criterion. In this protocol, the transcrestal approach is preferred when the residual alveolar bone height exceeds 4 mm, whereas the lateral window approach is employed when the residual height is between 2 and 4 mm; cases with residual height < 2 mm were excluded per the study criteria. In the present study, the residual bone heights of both groups were thoroughly examined based on clinical and radiographic records. The routine lateral window and crestal lift techniques employed in our department are described in detail below.

In accordance with the standard lateral window approach, a mucoperiosteal flap was elevated, the lateral bony wall was accessed, and the Schneiderian membrane was carefully elevated. After the preparation of the implant site, xenograft material (Bio-Oss, 1–2 mm particle size, Geistlich Pharma, Wolhusen, Switzerland) was placed into the sinus cavity. Implant were simultaneously inserted when adequate primary stability was achieved at the time of placement and a resorbable collagen membrane (Bio-Gide, Geistlich, Wolhusen, Switzerland) was positioned over the lateral window. No fixation with pins was performed.

For the transcrestal approach, osteotomies were prepared through the alveolar crest up to 1–2 mm below the sinus floor. The sinus floor was then fractured using osteotomes, in which approximately 2 mm of bone height was achieved by elevating the Schneiderian membrane through the fractured sinus floor. No graft material was applied in this approach. Implant were placed in the same session when sufficient primary stability was obtained. Cases in which implant were not inserted during the augmentation procedure whether due to insufficient stability or any intraoperative complications were not included in the study. All postoperative care followed departmental protocols, including antibiotics, analgesics, chlorhexidine rinses, and sinus precautions.

Fractal analysis

All panoramic radiographs included in this study had been acquired as part of routine clinical care, using the Planmeca Promax 2D (Planmeca Inc., Helsinki, Finland) digital panoramic X-ray device. Images were taken in the standardized position with the parameters 65 kV, 5 mA and 14 s, following the manufacturer’s recommendations.

FA was performed on panoramic radiographic images in TIFF format (300 dpi resolution) using the ImageJ software (version 1.53c, National Institutes of Health, Bethesda, MD, USA) by a dentomaxillofacial radiologist. The region of interest (ROI) selection was limited exclusively to the apical region of the implant. This decision was based on the nature of the transcrestal sinus lifting procedure, in which approximately 2 mm of bone height is achieved by fracturing the sinus floor using osteotomes to elevate the Schneiderian membrane. During this surgical maneuver, the fractured bone segment is typically confined to the apical portion of the implant, while it may not extend to the mesial or distal aspects. As a result, selecting ROIs from these lateral areas could lead to the inclusion of regions lacking fractured bone, potentially introducing heterogeneity into the trabecular pattern and compromising the reliability of the FA. To ensure methodological consistency and to focus measurements on a biologically relevant site, ROI selection was intentionally restricted to the apical zone where the fractured bone is most likely to be present. Considering the approximate dimensions of this fractured area, each ROI was standardized to 30 × 30 pixels across all images. Accordingly the ROI was defined as standardized square areas of 30 × 30 pixels located apically to the implant site, avoiding anatomical borders and artifacts. ROIs were placed in trabecular bone areas adjacent to the sinus floor elevation sites (Fig. 2).

Fig. 2.

Panoramic radiographs of maxillary sinus augmentation cases with ROI selection. (a) Lateral window technique; (b) Transcrestal technique

Images were pre-processed according to the White and Rudolph protocol; subsequently, the FD was computed in ImageJ using the 2D box-counting method [32]. The procedure involves several image processing steps [32].

A Gaussian Blur filter with a sigma value of 35 is applied to the duplicated image (Fig. 3a-b). The blurred version is then subtracted from the original ROI, and a gray value of 128 Gy is added to each pixel (Fig. 3c-d). The resulting image is then converted to binary using the “Make Binary” option. (Fig. 3e) Noise is reduced by applying the “Erode” function, and trabecular structures are enhanced by the “Dilate” function (Fig. 3f-g). The image is then inverted to highlight trabecular bone borders. (Fig. 3h) Lastly, the “Skeletonize” function is applied to transform the trabecular structure into a skeleton format. (Fig. 3i)

Fig. 3.

Measurement of FD (a) Duplication, (b) Blurring, (c) Subtraction, (d) Addition of 128gy, (e) Binary, (f) Erode, (g) Dilate, (h) Invert, (i) Skeletonize, (j) The box counting procedure

Following these steps, the FD of each ROI was calculated using the “Fractal Box Counter” function under the “Analyze” menu in ImageJ. (Fig. 3j) All measurements were performed by the same dentomaxillofacial radiologist. To assess intra-observer reliability, 20% of the radiographs were re-evaluated by the same observer after a two-week interval, and the intraclass correlation coefficient (ICC) was calculated.

Statistical analysis

All statistical analyses were performed using IBM SPSS Statistics for Windows, version 28.0 (IBM Corp., Armonk, NY, USA). The normality of data distribution was assessed using the Shapiro–Wilk test, and all data were found to be normally distributed (p > 0.05). Therefore, parametric tests were used in all comparisons. Paired samples t-tests were conducted to compare the FD values between T0 (surgery week) and T1 (six month follow-up) within each group (transcrestal and lateral window approach). Independent samples t-tests were applied to compare the FD values between the transcrestal and lateral window approach groups at the same time points. The Pearson correlation coefficient was used to analyze the correlation of FD and with age. The intra-rater reliability of FD measurements was assessed by calculating the intraclass correlation coefficient (ICC). Statistical significance was set at p ≤ 0.05.

Results

A total of 46 patients who met the inclusion criteria were included in the study. Of these, 21 patients (13 males, 8 females) underwent transcrestal sinus lifting, while 25 patients (19 males, 6 females) underwent lateral window approach sinus lifting. In total, 92 panoramic radiographs corresponding to baseline (T0) and 6 month follow-up (T1) assessments of these patients were analyzed (42 from the transcrestal group and 50 from the lateral window group).

The intraclass correlation coefficient (ICC) for repeated FD measurements was calculated as 0.93 (95% CI: 0.89–0.96), indicating excellent measurement consistency. The mean age of the transcrestal and lateral window groups was 52.66 ± 7.28 years and 54.22 ± 8.23 years, respectively, with no statistically significant difference between the groups (p > 0.05) (Table 1). Pearson correlation analysis revealed no significant association between age and FD values at baseline (T0) (Transcrestal: r = 0.14, p = 0.38; Lateral window: r = 0.11, p = 0.46) or at follow-up (T1) (Transcrestal: r = 0.09, p = 0.55; Lateral window: r = 0.07, p = 0.62). Similarly, no correlation was found between age and changes in FD (ΔFD) within the groups (Transcrestal: r = 0.12, p = 0.41; Lateral window: r = 0.10, p = 0.49).

Table 1.

Demographic characteristics of participants according to sinus lift approach

Variable		Transcrestal	Lateral window
		n (%)	n (%)
Gender	Female	8 (38.1%)	6 (24.0%)
	Male	13 (61.9%)	19 (76.0%)
	Total	21 (100%)	25 (100%)
Age, Mean ± SD (years)	Female	53.38 ± 5.37	56.00 ± 5.79
	Male	52.23 ± 8.43	53.56 ± 9.05
	Total	52.66 ± 7.28	54.22 ± 8.23

n = Number of participants, Mean = Average, SD = Standard Deviation

The residual alveolar crest heights were also assessed according to the applied sinus augmentation technique. In the transcrestal group, the mean crest height was 6.40 ± 1.35 mm (range: 4.4–8.6 mm), whereas in the lateral window group, the mean crest height was 3.25 ± 0.87 mm (range: 2.0–3.9 mm).

The intragroup comparisons revealed that the mean FD in the transcrestal group significantly increased from T0 to T1 (0.92 ± 0.06 vs. 1.10 ± 0.07, p < 0.001). In contrast, the lateral window approach group exhibited a slight but statistically significant decrease in FD values between T0 and T1 (1.28 ± 0.08 vs. 1.22 ± 0.09, p = 0.0032). When comparing FD values between the two groups, the lateral window approach group had significantly higher FD at T0 than the transcrestal group (1.28 ± 0.08 vs. 0.92 ± 0.06, p < 0.001). Although this difference persisted at T1, it was less pronounced but still statistically significant (1.22 ± 0.09 vs. 1.10 ± 0.07, p = 0.009) (Table 2).

Table 2.

Comparison of FD values within and between transcrestal and lateral window sinus lift groups at baseline (T0) and 6 month follow-up (T1)

Group	Time Points				ΔFD (T1–T0)	P*
	T0 FD		T1 FD
	Mean ± SD (Min–Max)	%95 CI (Lower-Upper)	Mean ± SD (Min–Max)	%95 CI (Lower-Upper)
Transcrestal	0.92 ± 0.06 (0.80–1.01)	0.894–0.946	1.10 ± 0.07 (0.98–1.25)	1.070–1.130	+ 0.18	< 0.001
Lateral window	1.28 ± 0.08 (1.14–1.42)	1.249–1.311	1.22 ± 0.09 (1.06–1.35)	1.185–1.255	–0.06	0.0032
P**	< 0.001		0.009		–	–

* Paired samples t-tests used, ** Independent samples t-tests, FD = Fractal Dimension, Mean = Average, SD = Standard Deviation, Min = Minimum, Max = Maximum

Discussion

Insufficient bone volume in the posterior maxilla often complicates primary implant stability, making sinus augmentation procedures essential for successful implant placement. Among the available techniques, the lateral window approach provides a wider field of vision and greater versatility, whereas the transcrestal sinus elevation, described by Summers in 1994, is favored for its minimally invasive nature and the possibility of simultaneous implant placement. In the present study, we compared the outcomes of these two approaches using FD analysis. Our findings are consistent with current clinical practice trends and offer additional insight into the microstructural bone changes associated with sinus augmentation [9, 35–39].

The choice of graft material used in conjunction with sinus lifting procedures plays a decisive role in the quality of new bone formation and the long-term success of implants [40, 41]. In current study, no graft material was used in patients undergoing transcrestal sinus lifting, whereas xenograft was exclusively employed in those treated with the lateral window approach. Previous reports have shown that a wide range of materials including autogenous bone, xenografts, allografts, platelet-derived concentrates, and β-tricalcium phosphate may be used alone or in combination during transcrestal procedures [42–45].

While some studies have found no significant difference in implant survival with or without graft materials [8]. Other studies suggest that graft-free protocols may yield even higher success rates [46, 47]. Importantly, sinus width appears to be a determining factor: in narrow sinuses, osteoconductive materials alone can achieve sufficient augmentation, whereas in wide sinuses, autogenous grafts are recommended to maintain stability and support osteogenesis [45]. Where autogenous bone cannot be obtained, the lateral window technique emerges as a more suitable option. Overall, sinus augmentation procedures, regardless of grafting, are considered safe techniques with low complication rates. Their predictable long-term success further supports their widespread use in clinical practice [7].

Beyond surgical technique and graft selection, the method employed to evaluate bone changes is also of critical importance. In the present study, changes in bone structure were assessed by applying FA to panoramic radiographs obtained prior to surgery and after the six month healing period preceding implant placement in patients who underwent transcrestal and lateral window sinus lifting. Alternative methods for assessing bone quality include histological and histomorphometric analyses as well as cone-beam computed tomography (CBCT). Among the various radiographic techniques, however, FA and digital subtraction radiography (DSR) stand out as the most cost-effective and easily accessible approaches for evaluating trabecular changes in the alveolar bone [15, 16, 18, 19, 22, 48, 49].

Whereas CBCT provides high-resolution imaging with relatively short acquisition times, its clinical use is restricted by disadvantages such as increased radiation dose and higher cost [48, 50]. Histological techniques, while considered the gold standard, are limited in routine practice due to their invasiveness, their restriction to single time-point evaluation, and lack of repeatability [51]. Although some studies have reported a strong correlation between FA and histomorphometric outcomes, others found no significant association [29, 43]. Nevertheless, FA has gained increasing acceptance in clinical research because it avoids additional radiation exposure, enables quantitative assessment of digital radiographs, and is straightforward to perform [52]. In current study, FA was chosen as the primary method of analysis since it provides a non-invasive, reproducible, and retrospective-compatible approach, thereby allowing reliable evaluation of microstructural bone changes without introducing additional risk to patients.

In line with these advantages, careful standardization of imaging and analytical parameters was implemented in current study to enhance the reliability of FA measurements. Previous studies have demonstrated that FA enables the evaluation of trabecular bone structure relatively independently of technical factors such as X-ray projection angle and exposure time [53, 54]. However, factors such as digital noise during imaging, ROI selection, and image processing parameters may affect the accuracy of the measurements [55]. Therefore, in the present study, all panoramic radiographs were acquired using the same device under fixed exposure parameters, as recommended by the manufacturer and consistent with routine clinical protocols. In the analyses, the ROIs were standardized to 30 × 30 pixels in each case, and homogeneous spongy bone areas free of anatomical borders and artifacts were selected. Since the size and location of the ROI have been reported in the literature to influence the outcomes of FA, particular attention was paid to these criteria to ensure measurement reproducibility [56, 57]. Consistent with this approach, Talmaç et al. similarly selected ROIs from the apical region of implants, paralleling the approach in current study, although they utilized a 20 × 20 pixel size [22].

According to the results of the current study, a significant increase in FD values was observed over the six month period in cases treated with transcrestal sinus lifting. In contrast, although a statistically significant change was also detected in the lateral window group, the FD values demonstrated only a modest absolute decrease. In the literature, some studies have reported time-dependent increases in FD values following both transcrestal and lateral window approach [29, 43, 58]. However, there are also reports documenting a decrease in FD values specifically in external sinus lifting procedures. For instance, in a study conducted by Wen and Zhang, FA following lateral window approach sinus augmentation revealed a significant decrease in FD values over a six month period after grafting [28]. Similarly, Sancar et al. noted a progressive decline in mineral density in grafted regions starting from the third month when xenografts were used, which became more pronounced by the ninth month, as assessed by FA [30].

These findings suggest that the early trajectory of FD values is closely influenced by the biological nature of the grafting material. Autogenous grafts, rich in viable osteogenic cells, initiate rapid new bone formation, which accelerates trabecular reorganization and leads to earlier FD increases. In contrast, xenografts, while radiographically dense, act as a biologically inert scaffold with a delayed onset of remodeling. During this transitional phase, the balance between graft particle resorption and new trabecular bone deposition may result in a temporary decline in FD values, as observed in our lateral window cohort. This interpretation is consistent with reports of earlier and more pronounced FD increases following autogenous grafting, while also aligning with studies that have documented transient decreases with xenograft use [28, 30, 58]. Clinically, we consider that these differences highlight the need to adapt implant loading protocols according to the graft material used, as the decrease in FD observed in xenograft-associated sites may indicate a requirement for longer healing periods before achieving optimal stability.

In their study, Talmaç et al. evaluated grafted and graft-free sinus lifting protocols using a single surgical technique, namely the lateral window approach. Platelet-rich fibrin (PRF) was applied as a substitute material in the graft-free group, and no statistically significant differences in FD values were observed between the groups at the end of the study. In current study, grafted and graft-free sinus lifting procedures were compared through different surgical approaches, and statistically significant differences in FD values were obtained [22].

In the current study, FD values in the lateral window approach group were significantly higher than those in the transcrestal group at T0, and this difference persisted at T1. A recent meta-analysis comparing grafted and graftless sinus lift procedures reported significantly higher bone density in grafted sites at six months, with a mean difference of 94.7 HU (p < 0.001) [59]. Consistent with these findings, current study also demonstrated higher FD values in the lateral window (grafted) group compared to the transcrestal (graftless) group at the six month follow-up. This finding may be explained by the nature of xenografts, which are resistant to resorption and provide high mineral density in the early period, but show a delayed formation of the fractal pattern associated with reorganized trabecular bone. Accordingly, within the six month observation period, the grafted regions exhibited a decrease in FD values, consistent with previous reports indicating early reductions in grafted sites following lateral window sinus augmentation [19, 28, 30]. Such decreases can be interpreted as a biologically expected outcome of the transitional phase between graft resorption and new bone remodeling. In contrast, the transcrestal lifting group, where no graft was applied, demonstrated a steady increase in FD values, likely reflecting the early onset of physiological trabecular remodeling and the more rapid establishment of an organized trabecular structure. Collectively, these results highlight the critical influence of graft type on early microstructural bone changes detected through FA. To the best of our knowledge, this is among the very few studies to directly compare transcrestal sinus lifting without grafting and the lateral window approach with xenografts using FA, thereby providing novel insight into the influence of grafting strategy on trabecular bone remodeling.

Limitations

This study has several limitations that should be acknowledged. First, the follow-up period was relatively short, the analysis was limited to a single graft type, and the retrospective design inherently restricted case selection.

Second, although the procedures were not confined to a single surgeon, all operations were performed by oral and maxillofacial surgeons with comparable clinical experience, and rigorous inclusion and exclusion criteria were applied. These measures were intended to achieve a high level of standardization and to minimize variability among operators, thereby reducing the likelihood of operator-related bias.

Third, the use of a uniform implant system across all cases was not feasible due to the retrospective nature of the study. Nevertheless, all implants used complied with international certification standards, which likely mitigated the potential influence of implant-related variability on the outcomes.

Fourth, potential differences in FD values based on gender were not assessed in this study due to the limited number of female participants. Future research should include gender-based analyses to clarify whether gender-related factors may influence trabecular architecture and FD outcomes.

Finally, although FD values increased in the transcrestal group and decreased in the lateral window group within the six month period, the absolute FD values in the lateral group remained higher. As the long-term trajectories of these changes could not be evaluated, further prospective studies with extended follow-up are warranted to determine whether the progressive increase observed in the transcrestal approach may eventually surpass the lateral window outcomes.

Conclusion

This study demonstrated that transcrestal and lateral window approach sinus lifting techniques lead to distinct trabecular bone remodeling patterns, as assessed through FD analysis on panoramic radiographs. While a significant increase in FD values was observed in the transcrestal sinus lifting group over a six month period, a statistically significant decrease was noted in the lateral window approach group. These differences are likely attributable to the biological behavior of graft materials particularly the delayed remodeling process associated with xenografts and highlight the influence of surgical approach on early bone healing dynamics.

The findings emphasize the potential of FA as a simple, non-invasive, and reproducible method for monitoring changes in bone quality following sinus augmentation procedures. In clinical practice, this technique may aid in the objective evaluation of graft integration and bone maturation, especially when cone-beam computed tomography is not available. Further prospective studies with longer follow-up periods and diverse graft materials are warranted to validate the diagnostic and prognostic value of FD analysis in the context of sinus lifting surgery. The comparative use of FA in both grafted and graft-free sinus augmentation techniques represents a unique contribution to the literature and underscores the method’s potential clinical applicability.

Author contributions

Hasan Akpınar: Conceptualization; Methodology; Investigation; Data curation; Writing – original draft; Writing – review & editing; Supervision; Project administration. Furkan Özbey: Formal analysis, Software; Validation; Visualization; Resources; Writing – review & editing. Betül Yıldırım: Methodology, Investigation; Data curation; Writing – review & editing.

Funding

No financial support received.

Data availability

The datasets analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethical approval

This study was conducted in accordance with the Declaration of Helsinki. Ethical approval was obtained from the Non-Invasive Clinical Research Ethics Committee of the Faculty of Dentistry, Afyonkarahisar Health Sciences University, Afyonkarahisar, Türkiye (2025/8-364-13.06.2025).

Consent to participate

All participants provided written informed consent.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.