Axial elongation and the spatial distribution of white and dark without pressure in myopic eyes: insights from ultra-widefield imaging

Abstract

Background

To explore the spatial distribution and associated factors of White and Dark Without Pressure lesions in myopic patients using ultra-widefield fundus imaging.

Methods

This retrospective, observational study evaluated 2,940 eyes from 1,470 patients who underwent refractive surgery. Ultra-widefield fundus imaging was used to identify and map the distribution of white and dark without pressure lesions across 12 retinal sectors. Correlations between the incidence of these lesions, their location, axial length, and the number of affected sectors were analyzed.

Results

White and dark without pressure lesions were present in 907 eyes (30.8%), with a significant association between increased axial length and higher incidence of white and dark without pressure (P < 0.001). Lesions were predominantly located in the temporal retina, specifically in the inferotemporal and superotemporal sectors. Longer axial length and younger age were both associated with a higher number of affected sectors and a more posterior localization of lesions.

Conclusion

Axial elongation is significantly correlated with both the occurrence and posterior localization of white and dark without pressure lesions, as well as a higher number of affected sectors in myopic eyes. Ultra-widefield fundus imaging offers a comprehensive perspective on these retinal changes, providing valuable insights into the pathophysiology of white and dark without pressure.

Background

White and Dark Without Pressure (WDWP) lesions, encompassing both White Without Pressure (WWOP) and Dark Without Pressure (DWOP), are retinal abnormalities predominantly observed in myopic individuals [1, 2]. WWOP is characterized by the presence of translucent white patches on the peripheral retina, whereas DWOP is identified by homogeneous dark brown or grayish areas, more frequently documented among individuals of African descent [3, 4]. These conditions are defined by distinctive retinal discolorations that occur independently of mechanical pressure and are often associated with peripheral retinal degeneration [5, 6]. The introduction of advanced imaging modalities, particularly ultra-widefield (UWF) fundus imaging, has significantly improved the detection and comprehensive evaluation of these lesions, facilitating a more detailed assessment of the retinal periphery [7,8,9].

Research have established a strong association between WDWP lesions and high myopia, driven by structural changes due to axial elongation [10, 11]. Although WDWP lesions are typically asymptomatic, their presence may indicate peripheral retinal structural vulnerability and may occasionally precede more serious pathologies such as retinal tears. Prior studies have suggested that these lesions may involve mechanical traction, retinal pigment epithelial disturbance, or early alterations at the vitreoretinal interface.

Increased axial length has been associated with mechanical stress (e.g., vitreoretinal traction, retinal thinning) and vascular stress (e.g., peripheral hypoperfusion) [12, 13], which may be particularly pronounced in the retinal periphery due to its structural and circulatory characteristics, and may contribute to the development of WDWP lesions [9, 12]. These stresses could lead to various forms of retinal degeneration and predispose the retina to other complications. Despite these findings, the precise mechanisms underlying the formation of WDWP lesions and their exact spatial distribution across different retinal sectors remain unclear. Furthermore,there is a notable lack of comprehensive studies utilizing advanced imaging modalities to fully characterize these lesions and understand their development over time.

Ultra-widefield fundus imaging offers a broad and detailed view of the retina, enhancing the ability to detect and quantify peripheral retinal lesions that might be overlooked with traditional imaging methods [7, 14,15,16,17,18]. This technique captures an expansive field of view, enabling the identification of lesions located in the far periphery of the retina. While previous studies utilized conventional imaging techniques, ultra-widefield fundus imaging provides an additional tool for exploring WDWP lesions. This advanced technology complements the clinical exam by offering a more comprehensive assessment of the retinal periphery.

This study aims to fill these gaps by employing UWF fundus imaging to investigate the spatial distribution of WDWP lesions in a cohort of patients undergoing refractive surgery. Additionally, it seeks to analyze the factors associated with the presence and distribution of WDWP lesions, with a particular focus on varying ocular axial lengths. By addressing these research gaps, we intend to provide deeper insights into the pathophysiology of WDWP leisions and its relationship with myopic changes.

Methods

Study population

This retrospective study reviewed patients who underwent refractive surgery at the Peking University Eye Center of Peking University Third Hospital between January 2020 and March 2021. The study included patients who had undergone laser-assisted in situ keratomileusis (LASIK), small incision lenticule extraction (SMILE), photorefractive keratecotomy (PRK), and implantable collamer lens (ICL) implantation procedures. The study adhered to the tenets of the Declaration of Helsinki and was approved by the Medical Ethics Committee of Peking University Third Hospital. All participants provided written informed consent prior to refractive surgery. The standard consent form included a statement that anonymized clinical data may be used for research purposes, which covered the retrospective analysis conducted in this study. Patients with fundus diseases, including a history of ocular surgery or trauma, uveitis, or corneal diseases that could affect refractive media, were excluded.Comprehensive preoperative ocular examinations were conducted, including uncorrected visual acuity (UCVA), best-corrected visual acuity (BCVA), intraocular pressure (IOP), slit-lamp biomicroscopy, dilated fundus examination, corneal topography, axial length (AL) measurement (IOL Master 500, Carl Zeiss Meditec AG, Germany), and ultra-widefield fundus photography (Optos, Marlborough, MA, USA).

Ultra-widefield fundus photography acquisition and quantification

All patients underwent mydriasis using 0.5% tropicamide and 0.5% phenylephrine before UWF fundus photography. Imaging was performed using a scanning laser ophthalmoscopy (SLO)-based system (Optos, Marlborough, MA, USA). Retinal specialists (S.Y.C, H.P.L, and the Retinal Specialist Group at Ophthalmology Center) reviewed preoperative dilated fundus examination records to exclude other retinal lesions and identify the presence of WDWP lesions. WDWP lesions were defined as distinct white or dark bands or patches on the retina, usually located in the peripheral retina, with continuous vascular course, differentiating it from other peripheral retinal degenerations or changes in the retinal pigment epithelium, such as congenital hypertrophy of the pigmented epithelium (CHRPE). Additionally, any peripheral retinal images that were blurred or out of focus, such as images where blood vessels lost focus during capture or where the image quality affected the measurement or evaluation of WDWP, were excluded. Other peripheral lesions, such as lattice degeneration or snail track lesions, were not analyzed due to inconsistent documentation and limited visibility in retrospective imaging. Any discrepant cases were adjudicated by a panel of retinal experts, including two attending ophthalmologists, two associate chief physicians, and one chief physician from the Retinal Specialist Group at the Ophthalmology Center.

Ultra-widefield fundus images with WDWP were further analyzed using ImageJ software (National Institutes of Health; http://imagej.nih.gov/ij/). Since the macula is centrally located in the posterior pole (usually at the center of the image), the shortest distance from each WDWP lesion to the center of the fovea was measured using the shortest horizontal diameter of the optic disc (optic disc diameter, DD) as the unit of length. UWF images were divided into 12 sectors: SN1, SN2, SN3, IN4, IN5, IN6, IT7, IT8, IT9, ST10, ST11, and ST12 (Fig. 1) using the concentric rings map as a reference [19]. A sector was considered affected if any portion of a WDWP lesion overlapped with that sector. The presence of WDWP in each sector and the shortest distance from WDWP lesions to the macula were recorded. Measurements were conducted in the red channel for clearer disc boundary visualization and in the green channel for better WDWP boundary delineation.

Fig. 1

Workflow for the measurement and analysis of WDWP distribution. This workflow demonstrates the process for accurately measuring and analyzing WDWP distribution using separated color channels and a structured mapping approach. The process begins with the original fundus image showing WDWP lesions (a). The image is separated into red (b) and green (c) color channels. In the red channel, the optic disc diameter is measured (blue line in c) and used to create concentric rings (d). In the green channel, the concentric rings map serves as a reference (f-g) for measuring the distance from the macula to WDWP lesions across different retinal sectors, as indicated by the red line in (g). The ring shown in this figure is for illustrative purposes only and does not represent the full extent of the analyzed area. All lesions within the visible peripheral retina were included in the analysis

Statistical analysis

All statistical analyses were conducted using SPSS for Mac OS version 29(IBM/SPSS, Armonk, NY, USA). Means and standard deviations (SD) were calculated for key variables.The axial length and age of the participants were each divided into four quartile groups. The Chi-Square test was employed to examine the relationship between axial length and the incidence of WDWP, controlling for the confounding effect of age. By stratifying the data based on age groups, we aimed to assess whether axial length independently influences the occurrence of WDWP. A P-value of less than 0.05 was considered statistically significant.

The distribution percentage of WDWP in the 12 sectors was calculated. Correlation analysis was performed to assess the relationship between axial length and the number of sectors affected by WDWP. Additionally, the correlation of axial length and age with the shortest distance from WDWP lesions to the macula in each sector was examined (Fig. 2).

Fig. 2

Flowchart of Study Population and Image Analysis Process. This flowchart illustrates the selection and analysis of fundus images. From an initial 1,485 myopic patients, 15 were excluded due to unclear or missing images, leaving 1,470 for analysis. Images without WDWP (2,033 eyes) were excluded, while 907 eyes with WDWP were further analyzed. The images were assessed for distance to the macula and WDWP sector count, with statistical analyses exploring factors and the relationship between WDWP distribution and axial length

To account for potential correlation between paired eyes from the same individual, generalized estimating equations (GEE) were applied. An exchangeable working correlation matrix was specified to model within-subject dependence.

All measurements were conducted by two experienced observers (S.Y.C. and C.T.P.). To ensure sufficient sample size and manage the dataset effectively while maintaining precision and reliability, random samples of 30 eyes were selected from each region, totaling 360 eyes. For intraobserver variability, each observer measured the distance in each sector twice, with re-examinations conducted at intervals of 1–2 weeks. The intraclass correlation coefficient (ICC) and coefficient of variation were calculated to assess the reliability of these measurements. To evaluate interobserver consistency, Lin’s concordance correlation coefficient and Bland–Altman plots were employed.

Results

A total of 1485 patients were initially included in the study. After excluding 15 patients due to missing or unclear images, 2940 eyes of 1470 patients (708 males) with a mean age of 26.7 ± 7.2 years (range: 17–52 years) were analyzed. The mean axial length was 25.8 ± 1.2 mm (range: 21.1–31.7 mm). WDWP was identified in 907 eyes (30.8 ± 0.9%) with 203 patients having unilateral and 352 having bilateral involvement. WDWP was not found in 2033 eyes (69 ± 0.8%). The design and workflow of the study population are illustrated in Fig. 2.

A Chi-Square test of independence revealed a statistically significant association between axial length (AL) quartiles and WDWP occurrence across all age groups (P < 0.001). In each age group, higher AL quartiles corresponded to an increased occurrence of WDWP, with the fourth quartile consistently showing the highest WDWP rates (43.6% to 61.7%)(Table 1) (Fig. 3).

Table 1 Association between Axial Length and WDWP Occurrence by Age Group. This table shows the relationship between axial length (AL) quartiles and the occurrence of White and Dark Without Pressure (WDWP) lesions across age groups. Statistical significance is indicated by Pearson Chi-Square (χ²) values and p-values, highlighting a significant association between longer AL and higher WDWP occurrence (P < 0.001)

Fig. 3

Association between Axial Length (AL), Age Group, and WDWP Occurrence. This graph shows the percentage of WDWP occurrence across different age groups and AL quartiles. Higher AL quartiles are consistently linked with greater WDWP occurrence

The GEE model indicated that both longer axial length and younger age were significantly associated with the presence of WDWP lesions. Specifically, each 1-mm increase in axial length was associated with greater odds of WDWP (B = 0.818, 95% CI: 0.697 to 0.939, p < 0.001), while each additional year of age was associated with lower odds (B = –0.035, 95% CI: –0.052 to –0.018, p < 0.001), after adjusting for inter-eye correlation.

In the 907 eyes with WDWP, a total of 3624 WDWP sectors were identified. The highest frequencies were in the IT9 (19.1 ± 0.7%) and ST10 (18.4 ± 0.6%) sectors, followed by IT8 (12.8 ± 0.6%) and ST11 (10.4 ± 0.5%). This indicates that WDWP is predominantly located in the temporal retina, from the superotemporal to the inferotemporal regions (Table 2).

Table 2 WDWP Distribution and Correlation with Axial Length and Age. This table illustrates the distribution of WDWP across different retinal sectors, the mean shortest distance from WDWP to the macula measured in disc diameters (DD), and their associations with axial length and age. The table includes p-values, standardized coefficients (Beta), unstandardized coefficients (B), and 95% confidence intervals

Further analysis revealed that the number of sectors affected by WDWP was significantly associated with both younger age (P < 0.001) and longer axial length (P < 0.001). After adjusting for disc diameter, there was a significant negative correlation between axial length and the measured distance from WDWP lesions to the macula in most sectors, except SN1, SN2, and ST12. Additionally, in sectors SN3, IT8, IT9, and ST10, the distance from WDWP lesions to the macula was positively correlated with age (Table 2). This indicates that younger age and longer axial length both increase the number of affected sectors, while longer axial length tends to move WDWP lesions closer to the posterior region of the globe, especially in younger patients.

The measurement of the distance from WDWP to the macula showed an intraclass correlation coefficient (ICC) of 1.000 (P < 0.001), signifying excellent intraobserver consistency. The average coefficient of variation was 24.47% ± 0.03%. For interobserver variability, the concordance correlation coefficient was calculated as r = 0.996, indicating strong agreement between examiners. The Bland–Altman plot showed that 5.8% (21/360) of the measurements fell outside the 95% limits of agreement, reinforcing the high level of interobserver reliability.

Discussion

In this study, we investigated the incidence and characteristics of WDWP lesions in a cohort of patients undergoing refractive surgery, using UWF imaging to accurately quantify the spatial distribution and correlates of WDWP lesions. Our findings demonstrate a significant association between longer axial length and the presence of WDWP lesions, corroborating previous studies that link myopic changes with peripheral retinal abnormalities [1, 2, 11]. In our study, the incidence of WDWP in myopic patients was approximately 31%, consistent with earlier research findings [8].

Pearson Chi-square analysis revealed a positive association between axial length and WDWP prevalence, potentially reflecting cumulative mechanical and vascular stress on the peripheral retina in eyes with greater elongation [20]. Previous studies [1, 11, 21] have shown that younger age is associated with the occurrence of WDWP, indicating that younger individuals are more prone to developing these lesions. In our study, there appears to be a slight decline in WDWP incidence with increasing age, although this trend is not very pronounced (Fig. 3). This may reflect a cohort effect, in which younger individuals undergoing refractive surgery tend to have earlier-onset or more rapidly progressive myopia, resulting in longer axial length despite their age. Further analysis with larger sample sizes is required to clarify this observation.

The regional analysis indicated a predilection of WDWP lesions in the inferotemporal and superotemporal quadrants, with IT9 and ST10 sectors exhibiting the highest frequencies, which is consistent with previous research findings [1, 22,23,24]. This pattern may reflect regional anatomical and biomechanical vulnerabilities in the inferotemporal and superotemporal retina, where the retinal tissue is thinner, structural support is weaker, and the retina is more exposed to horizontal shearing forces and peripheral vitreoretinal traction during ocular movements. Our observation that WDWP lesions tend to locate posteriorly with increasing axial length (Fig. 4) further supports the notion that axial elongation exacerbates peripheral retinal stress.

Fig. 4

Ultra-widefield images of white and dark without pressure (WDWP) lesions. With increasing axial elongation, WDWP lesions are located closer to the posterior pole, extending from the peripheral retina, and involving more retinal sectors. The inset in the upper-left corner of each image highlights the location of WDWP lesions, marked by orange curved lines. The axial length (AL) of each eye is shown in the lower-left corner of each image

The etiology of WDWP remains unclear. However, our study's results suggest that as axial length increases, the vitreous may exert greater axial tractional forces on the retina. These forces appear to be more pronounced near the posterior pole compared to the peripheral retina, which could account for the more posterior localization of WDWP lesions in eyes with longer axial lengths. This suggests that in more myopic eyes, WDWP lesions are more likely to be found closer to the posterior pole (Fig. 5). This hypothesis indicates that the axial component of vitreous traction significantly influences the progression and distribution of WDWP lesions. These forces may exert stress on specific retinal regions, altering the mechanical environment of the retinal interface compared to surrounding areas, leading to changes in color and microstructure that manifest as WDWP lesions. Interestingly, age may also play a role in certain sectors, with younger individuals showing WDWP closer to the posterior pole. Younger patients may have a denser vitreous, which, along with increasing axial length, could contribute to stronger vitreous traction on the retina, potentially leading to a broader distribution of WDWP lesions.

Fig. 5

Axial Traction Changes and WDWP Lesion Variation. This schematic illustration represents a conceptual hypothesis derived from our data, showing how axial traction forces at different retinal locations may change with increasing axial length. Solid lines (red, green, blue) represent initial traction at points a, b, and c, respectively. As axial length increases, these points shift posteriorly (a′ to a, b′ to b, c′ to c). Retinal regions closer to the posterior pole (e.g., a′) are hypothesized to experience greater increases in axial traction compared to more peripheral areas (e.g., b′, c′). Dashed lines indicate the additional traction induced by elongation. This conceptual model may explain the observed posterior shift of WDWP lesions in eyes with longer axial lengths

Moreover, Orlin et al. [25] used ultra-widefield fluorescein angiography (UWFA) to evaluate eyes with WWOP and found that these eyes exhibited specific peripheral vascular patterns. The study identified three common UWFA patterns, with eyes exhibiting WWOP showing more frequent peripheral vascular abnormalities compared to those without WWOP. This aligns with our findings of higher lesion frequencies in the temporal quadrants and supports the hypothesis that fluorescein leakage or staining in regions of WWOP could be related to vitreous traction on the retina. Similar observations are documented in fluorescein angiography of patients with epiretinal membranes (ERM), where vascular leakage results from mechanical forces [26, 27].

Our findings are consistent with those reported by Yu et al. [11], who observed that longer axial length and more severe myopia are risk factors for both WWOP and DWOP. Moreover, their study noted reduced vessel density and retinal thickness in eyes with DWOP, providing morphological evidence of the structural changes associated with these conditions. Additionally, Some authors [3, 28] described changes in outer retinal reflectivity associated with WDWP lesions, highlighting the importance of advanced imaging techniques in identifying and characterizing these abnormalities. Furthermore, the study by Nagpal et al. [29] suggested a potential transient nature of DWOP lesions observed primarily in patients with systemic conditions like sickle cell hemoglobinopathies and systemic hypertension. Although the exact cause remains unknown, these observations support the hypothesis of mechanical and vascular factors playing a role in WDWP development.

To our knowledge, this is the first study to conduct a large-scale evaluation of a myopic population and systematically assess eyes with WDWP and their spatial distribution. Utilizing advanced UWF imaging systems and a statistical approach, we aim to elucidate the potential pathogenesis of WDWP.

Several limitations warrant consideration. First, the cross-sectional design precludes causal inferences, and the study's focus on a refractive surgery population may limit generalizability. The sample was derived from a single hospital and primarily consisted of patients undergoing refractive surgery, which may not be representative of the general myopic population. Future studies should include a more diverse and larger population sample to improve generalizability. Second, the retrospective nature of the study means that the data collected may be subject to biases and inaccuracies related to past records and image quality. Prospective studies are needed to verify these findings. Third, the technique and approach to evaluating WDWP lesions require further validation, as UWF images are known to have distortion near the peripheral retina, which may introduce measurement errors [16, 17]. A previous study [17] have suggested that the enlargement factor of peripheral images remains consistent despite varying axial lengths, and the newer software (V2 Vantage Pro, Optos) is claimed to correct peripheral retinal image distortion. In our study, optic disc diameter was used as a unit of length, and adjustments for optic disc diameter were made during statistical analysis. Additionally, while UWF imaging provides a detailed view of the retinal periphery, other advanced imaging modalities such as optical coherence tomography (OCT) and adaptive optics (AO) might offer additional insights into the structural and functional changes associated with WDWP lesions. Fourth, previous literature has discussed WWOP and DWOP separately, but our observations using UWF imaging often show these conditions appearing simultaneously, making them difficult to evaluate and distinguish. Therefore, we have combined them as WDWP for this study. Future research should refine the distinction between WWOP and DWOP for further analysis. Fifth, the age of myopia onset was not available for most patients, which may limit the interpretation of age-related trends. Lastly, it is important to note that the temporal quadrants are more easily imaged by the Optos system, which increases the likelihood that some lesions in other quadrants may have been missed. Long-term follow-up studies are needed to understand the natural history and progression of WDWP.

In conclusion, our study highlights that increased axial length is significantly associated with the occurrence and posterior localization of WDWP lesions. This finding suggests that axial elongation may play a crucial role in the development of these lesions, enhancing our understanding of WDWP in myopic eyes.