Assessment of Silicone Oil Emulsification: A Comparison of Currently Applied Methods

Purpose To compare ultrasound biomicroscopy (UBM), Coulter counter, and B-scan ultrasonography in the evaluation of silicone oil (SO) emulsification. Methods Patients who underwent primary pars plana vitrectomy with SO tamponade for rhegmatogenous retinal detachment and SO removal were included. UBM images were acquired before the SO removal, and B-scan images were taken after removal. The number of droplets in the first and last 2 mL of washout fluid was analyzed using a Coulter counter. The correlations between these measurements were analyzed. Results Thirty-four eyes received both UBM and Coulter counter analysis for the first 2 mL of washout fluid, and 34 underwent B-scan and Coulter counter analysis of the last 2 mL washout fluid. The mean UBM grading was 26.41 ± 9.71 (range: 1–36); the mean SO index obtained with B-scan was 5.25 ± 5.00% (range: 0.10–16.49%), and the mean number of SO droplets was 1.26 ± 2.45 × 107/mL and 3.34 ± 4.22 × 106/mL in the first and last 2 mL of washout fluid, respectively. There were significant correlations between UBM grading and SO droplets in the first 2 mL and between B-scan grading and SO droplets in the last 2 mL (all P < 0.05). Conclusions UBM, Coulter counter, and B-scan ultrasonography could all be used in the evaluation of SO emulsification, and their findings were comparable.

Several studies had similar fndings, and the close relationship between SO emulsifcation and elevated intraocular pressure (IOP) was confrmed by researchers using UBM [16] or Coulter counter [21]. However, some diferences also emerged. For example, fuid-air exchange was found to be inefective to reduce residual SO droplets in a study using B-scan ultrasonography [18] and efective in a study using a Coulter counter [20]. Slit lamp, UBM, and gonioscopy focus on the anterior segment, whereas B-scan ultrasonography focuses on the posterior one; thus, their results might difer. Moreover, slit lamp, UBM, and gonioscopy can evaluate SO emulsifcation both before and after SO removal, whereas B-scan ultrasonography can be used only after SO removal.
Tus, each of these studies only evaluated a piece of the puzzle, and the combination of diferent modalities can help us acquire a comprehensive understanding of SO emulsifcation. However, the measurements by diferent methods should be consistent to be combined. We compared UBM, Coulter counters, and B-scan ultrasonography results for the evaluation of SO emulsifcation in a group of patients who underwent vitrectomy and SO tamponade for rhegmatogenous retinal detachment (RRD). Te relationship between these measurements was then analyzed.

Study Participants and Ethics Statement.
Tis singlecenter study followed an observational, cross-sectional design. We collected data from patients who underwent primary pars plana vitrectomy (PPV) with SO tamponade for RRD at the Eye and ENT Hospital of Fudan University between January 2019 and January 2022. Patients who completed a minimum of 8-week follow-up after SO removal with the retina properly attached were eligible. Exclusion criteria were diabetes mellitus, previous SO injection, intraocular surgery other than PPV or cataract surgery, intraocular diseases other than RRD or cataract (e.g., glaucoma, uveitis), elevated IOP (>21 mmHg) before PPV, or age <18 years at the time of primary PPV.
Te study was approved by the Institutional Review Board of the Eye and ENT Hospital of Fudan University and conformed to the tenets of the Declaration of Helsinki. All patients provided written informed consent. Patient who underwent both UBM examination before SO removal and Coulter analysis of the frst 2 mL washout fuid were enrolled in the frst group to study the possible correlations between the results of Coulter counter and the UBM grade. Patients who underwent both B-scan ultrasonography after SO removal and Coulter analysis of the last 2 mL washout fuid were enrolled in the second group to study the correlations between the results of the Coulter counter and SOI.

Surgical Procedures.
During the SO injection surgeries, a standard three-port, 23-gauge PPV was performed in all patients by a single surgeon (Chunhui Jiang) using the Alcon Constellation system (Alcon Laboratories, Inc., Geneva, Switzerland) and SO (5700 cSt; Bausch & Lomb Inc., Bridgewater, NJ, USA). During SO removal, additional procedures, such as membrane peeling, phacoemulsifcation, and/or intraocular lens implantation, were performed as necessary. Constant irrigation of the vitreous cavity for 10 minutes was adopted to ensure thorough removal of the emulsifed SO droplets.

Main Ophthalmic Measurements.
Before and after SO removal, each patient underwent a thorough ophthalmic examination, which included the assessment of the bestcorrected visual acuity, logarithm of the minimum angle of resolution (logMAR), spherical equivalent power (calculated as half of the cylindrical dioptric plus the spherical diopter), slit-lamp microscopy, dilated fundus examination with a non-contact lens (MaxField 84 Diopter; Ocular Instruments, Bellevue, WA, USA), and measurement of IOP by non-contact tonometry. Te axial length (AL) measurement by IOLMaster (version 3.01; Carl Zeiss Meditec, Jena, Germany) was performed only preoperatively. Demographic data and clinical histories were also collected.

Silicone Oil Emulsifcation Examinations.
Te UBM examination of the anterior segment (MD-300L, 50-MHz probe transducer; Meda Co., Ltd., Tianjin, China) was performed within the week before SO removal. Washout fuid samples were collected during the SO removal for the Coulter counter analysis; then, the B-scan ultrasonography (AVISO, Quantel Medical, France) was performed 8-12 weeks postoperatively, when the infammation or hemorrhage caused by SO removal had subsided.

Ultrasound Biomicroscopy Grading.
Te UBM exam and grading were performed according to the method previously described in detail [16,22]. In brief, eight signs of SO emulsifcation in the UBM images were graded as 1 (present) or 0 (not present), and the grades for all signs in each eye were summed. All UBM images were analyzed by two independent readers (Hongmei Zhao and Jian Yu). When the grades determined by both graders were identical, they were used as the fnal grades; when they difered, the fnal grade was determined by a senior specialist (Qian Chen).

Coulter
Counter. Te number of SO droplets in the washout fuid samples was assessed with a Coulter counter, using the method introduced by Chan [19,23]. In brief, the frst and last 2 mL of washout fuid extracted during the SO removal were collected. Using a Multisizer ® 3 Coulter counter (Beckman Coulter, Brea, CA, USA), the size and number of droplets in the samples were measured, and the particles ranging in diameter between 0.4 and 12 μm were included in the analysis. Te values for each sample represent the mean of three consecutive measurements. Based on the diameter, the droplets were divided into smaller (0.4-5 μm) and larger (5-12 μm) ones.

B-Scan Ultrasonography.
Finally, B-scan ultrasonography examination and analysis were performed according to the methods previously described [17,23]. Te SO index (SOI) was calculated as the percentage of signals from hyperechoic droplets in the area of the vitreous cavity.

Statistical
Analysis. All analyses were performed using SPSS software version 20.0 (IBM, Armonk, NY, USA). Te Kolmogorov-Smirnov test was used to examine the normality of the data. Spearman's correlation coefcient was used to assess the correlations between the results of the Coulter counter and the UBM grade or SOI. P values <0.05 were considered statistically signifcant.

Results
Tirty-four patients underwent both UBM examination before SO removal and Coulter analysis of the frst 2 mL washout fuid. Teir mean age was 55.32 ± 13.78 years (range: 24-86); the mean duration of SO in situ was 28.11 ± 16.24 weeks (range: 14.00-103.57), and the mean AL was 25.63 ± 2.57 mm (range: 21.61-33.16). Five patients had choroidal detachment and four patients had a history of ophthalmic trauma at the time of PPV.
Another 34 patients underwent both B-scan ultrasonography after SO removal and Coulter analysis of the last 2 mL of washout fuid. Teir mean age was 51.91 ± 16.76 years (range: 18-86); the mean duration of SO in situ was 27.46 ± 9.67 weeks (range: 14.86-55.00), and the mean AL was 25.79 ± 2.86 mm (range: 22.01-33.00). Two patients had choroidal detachment and six patients had a history of ophthalmic trauma at the time of PPV.
A similar correlation was found between the SOI obtained with B-scan ultrasonography and the total number of droplets in the last 2 mL (r � 0.342, P � 0.048; Table 1). Again, the number of smaller SO droplets (0.4-5 μm) was signifcantly correlated with the SOI (r � 0.358, P � 0.037), while for the larger droplets (5-12 μm), it was not (r � 0.052, P � 0.771; Table 1).

Discussion
In this study, the possible correlations between the SO emulsifcation results evaluated by Coulter counter and UBM or B-scan ultrasonography were explored. Since UBM refected the SO emulsifcation before SO removal and the frst washout fuid were also collected before SO removal, the possible correlation between the two was explored. In the same way, B-scan refected SO residual after SO removal, and the last washout fuid were also collected at the end of SO removal, which could also refect SO residual. Terefore, correlation between the two were also studied. A signifcant correlation was found between the number of droplets and the quantitative assessment of SO emulsifcation provided by the imaging modalities. Te main advantages of the latter are the high reproducibility and non-invasiveness; furthermore, these methods do not require any specifc technique or software. On the other hand, the Coulter counter could simultaneously and accurately count and measure the size of SO droplets, though this method needs expensive equipment and SO samples.
To the best of our knowledge, this study reports the frst direct comparison of these methods. Our fndings were consistent, as both UBM and B-scan ultrasonography results were closely correlated with the total number of SO droplets measured with Coulter counter. Tis result was not granted since the resolution of UBM and B-scan ultrasonography is approximately 40 and 500 μm, respectively, while the emulsifed SO droplets are markedly below those dimensions. Te emulsifed SO droplets appear as highly refective dots in ultrasound images; thus, these modalities can detect SO droplets of all sizes. Tis characteristic might explain the close correlation between the results of these scans and the total number of droplets identifed with the Coulter counter.
Previous studies suggested a consistency between the results obtained with these methods. A signifcant negative correlation was found between age and SO emulsifcation evaluated by both UBM [16] and Coulter counter [21]. Furthermore, a close relationship between SO emulsifcation and elevated IOP was confrmed by studies using UBM [16,24], B-scan ultrasonography [17], or Coulter counter [21]. Tese similarities between the results acquired with diferent methods support the close relationships between them. We also noticed that the UBM and B-scan ultrasonography results were signifcantly correlated with the number of small SO droplets and not with the larger ones, though the reason remains to be determined. However, this fnding may be simply due to the small droplets being the vast majority (99.09 ± 0.82% and 98.22 ± 3.26% of the total in the frst and last washout).
Several studies have been conducted to observe and evaluate SO emulsifcation, improving our understanding of this condition [19,25]. Te three methods we evaluated could all assess SO emulsifcation, from diferent perspectives. In addition to being performed at diferent time points and in diferent parts of the eye, B-scan ultrasonography and Coulter counter analysis study the droplets in the vitreous cavity, whereas UBM can also detect those infltrating other tissues [22]. Previous studies also reported the diferences and limitations of these methods [18,19,25]. Azzolini et al. [25] noted that, in UBM examinations, some SO droplets could be masked by after-ringing efects from other SO droplets. Moreover, the Coulter counter has a sensitivity between 0.4 and 30 μm, and not all droplets were considered [19]. Tus, these methods have limitations; however, they could be considered complementary since they are performed at diferent time points and identify droplets in diferent parts of the eye. With the combination of their results, we could have a more complete view of SO emulsifcation, although frst we should ensure that their results are comparable.
With this study, we verifed the strong associations between the results of these three methods; therefore, we confrmed that the results from previous studies using different methods can be compared and evaluated together. Moreover, these three methods could all be used to assess SO emulsifcation in future research. Trough a comprehensive analysis, a more thorough and complete knowledge of SO emulsifcation could be achieved, and subsequently, patient management could be improved. Among these methods, UBM could be used preoperatively and might be more relevant since the results could provide additional information before the SO removal procedure. Patients with severe emulsifcation identifed by UBM should be monitored closely and receive a thorough irrigation of the vitreous cavity.
Our study was limited by its single-center, crosssectional design and the limited number of cases.

Conclusion
UBM, Coulter counter, and B-scan ultrasonography provide reliable and highly consistent fndings; these tests could all be used in the evaluation of SO emulsifcation, and their fndings are comparable.

Data Availability
Te research data used to support the fndings of this study are included within the article.

Disclosure
Hongmei Zhao and Qian Chen are co-frst authors.

Conflicts of Interest
Te authors declare that they have no conficts of interest.