A novel technique of rhinostomy size assessment by immersion b-scan probe following dacryocystorhinostomy – A comparative study

Abstract

Purpose:

To measure and compare the ostium size following external and endoscopic dacryocystorhinostomy (DCR) using an innovative, noninvasive immersion B-scan ultrasonography technique and to correlate ostium dimensions with functional patency assessed by the fluorescein dye disappearance test (FDDT).

Methods:

This prospective comparative study included 50 patients with nasolacrimal duct obstruction who underwent either external (n = 25) or endoscopic DCR (n = 25) between September 2022 and February 2024. Intraoperative ostium size was measured using Castroviejo’s calipers for external DCR and Schirmer’s strips for endoscopic DCR. Postoperative ostium dimensions (length, width, and area) were assessed on days 15, 30, and 90 using an immersion B-scan probe. Functional patency was evaluated by FDDT. Data were analyzed with ANOVA and multilinear regression.

Results:

The mean age was 54.9 ± 13.26 years in the external DCR group and 51.76 ± 13.79 years in the endoscopic group, with female predominance (78%). At 3 months, the external DCR group showed a significantly larger ostium size (length: 11.66 ± 2.12 mm; width: 7.09 ± 1.28 mm; area: 78.91 ± 24.47 mm²) compared to the endoscopic group (length: 9.84 ± 0.88 mm; width: 6.46 ± 1.14 mm; area: 58.95 ± 16.25 mm²) (P < 0.05). External DCR demonstrated superior functional patency with an FDDT sensitivity of 91% and better F1 scores throughout follow-up.

Conclusion:

Immersion B-scan ultrasonography is a reliable, noninvasive tool for postoperative assessment of ostium dimensions. External DCR results in a larger ostium size and better functional patency than endoscopic DCR.

Dacryocystitis is an inflammation of the lacrimal sac secondary to nasolacrimal duct obstruction, clinically presenting with epiphora and swelling at the inferomedial canthus. It may be acute or chronic and congenital or acquired in origin. In India, acute dacryocystitis is commonly caused by bacterial pathogens including Staphylococcus aureus, Streptococcus species, Pseudomonas aeruginosa, and Haemophilus influenzae.[1] Chronic dacryocystitis often results from recurrent infections, dacryolith formation, or systemic conditions like sarcoidosis and granulomatosis with polyangiitis. Congenital cases typically involve membranous obstruction at the valve of Hasner, while acquired cases may arise secondary to trauma, surgery, medications, or tumors. The condition predominantly affects older women from lower socioeconomic backgrounds.[2]

Management of acute dacryocystitis primarily involves systemic antibiotics targeted toward Gram-positive and Gram-negative bacteria, supported by analgesia and warm compresses. Chronic dacryocystitis is most effectively managed by dacryocystorhinostomy (DCR), which creates an alternate tear drainage pathway into the nasal cavity. External DCR involves a skin incision to access the lacrimal sac directly, whereas endonasal DCR employs an endoscopic transnasal approach to create the osteotomy. One of the surgical challenges remains that postoperative stenosis of the osteotomy site accounts for failure in 50.6% of external and 55.5% of endonasal DCR.[3] A statistically significant decrease in ostium size was noted between 2 and 4 months of the postoperative period,[4] warranting vigilant monitoring of ostium patency.

B-scan ultrasonography is routinely employed to evaluate vitreous and retinal pathologies due to its convenience, affordability, and noninvasive nature. However, its utility is limited in certain anterior segment conditions, such as lens abnormalities and space-occupying lesions of the ciliary body and iris, where detailed visualization is often inadequate. In such cases, the Pan-immersion B-scan technique offers superior resolution and comprehensive imaging. It proves especially useful in accurately diagnosing retinal detachment (RD), persistent hyperplastic primary vitreous (PHPV), and intraocular tumors.[5]

This study explores the novel application of immersion B-scan as a noninvasive, cost-effective method to assess ostium dimensions following external and endonasal DCR compared to intranasal endoscopy and CT scan used routinely. The aim is to evaluate and compare postoperative ostium sizes using this technique to enhance anatomical and functional monitoring.

Methods

This 18-month prospective, comparative study was conducted at the ophthalmology and otorhinolaryngology department of our institution between September 2022 and February 2024. The study adhered to the tenets of the Declaration of Helsinki and was approved by the local institutional ethics committee (IEC). Participants were enrolled after receiving a detailed study briefing and signing informed consent forms. It included 50 patients with acquired nasolacrimal duct obstruction, 25 patients who underwent external DCR in the ophthalmology department, and 25 patients who underwent endonasal DCR in the ENT department. Inclusion criteria encompassed primary and secondary nasolacrimal duct obstruction, chronic dacryocystitis, and functional obstruction, while congenital anomalies, canalicular infections, tumors, and trauma were excluded.

Presurgical evaluations included clinical history, systemic assessment, slit-lamp examination, lacrimal syringing, ENT evaluation, and lab tests. Patients were prepared with cessation of anticoagulants, antibiotic eye drops, and nasal decongestants. Ultrasound (EchoVue) was used for imaging. All surgeries were performed by the same senior surgeon with extensive surgical experience.

For external DCR, local anesthesia with 2% lignocaine-adrenaline and 0.5% bupivacaine (3:2 ratio) was administered. Nasal packing with lignocaine-soaked gauze was placed in the ipsilateral nostril. A J-shaped incision near the medial canthus exposed the lacrimal fossa via blunt dissection. Kerrison’s punch created the ostium, followed by H-shaped incisions in the lacrimal sac and nasal mucosa for flap formation. Intraoperative ostium length and width measurements were performed using Castroviejo’s callipers [Fig. 1a and b]. Sutures were placed, and patency was assessed via lacrimal syringing. Endonasal DCR was performed under general anesthesia using 0° and 30° rigid endoscopes. A rectangular incision near the uncinate process was made, with mucosal cuff removal and bone exposure via Kerrison’s punch. The medial sac wall was excised with a tissue punch, and patency was ensured using saline syringing. Intraoperative ostium length and width measurements were performed using Schirmer’s strip [Fig. 1c and d]. Both techniques ensured effective ostium creation with thorough patency evaluation.

Figure 1.

(a) Intraoperative measurement of the length of the ostium using Castroviejo’s calipers in external DCR, (b) shows the intraoperative measurement of the width of the ostium using Castroviejo’s calipers in external DCR. (c) shows the measurement of intraoperative ostium length using a Schirmer’s strip in endonasal DCR, and (d) shows the measurement of intraoperative ostium width using a Schirmer’s strip in endonasal DCR

Patients underwent immersion B-scan and fluorescein dye disappearance test (FDDT) on day 15, followed by evaluations at 1 and 3 months. Each visit included lacrimal syringing, immersion B-scan for ostium dimensions, and FDDT for functional patency. Oral antibiotics, anti-inflammatory drugs, and local antibiotic ointments were administered postoperatively to ensure recovery. The B-scan probe is disinfected with sterilium and dried with a soft cloth.

Modified immersion B-scan ultrasonography includes a standard B-scan probe (10MHz), a surgical glove, and water. A finger of the sterile surgical glove was cut, prepared wearing sterile gloves and filled with tap water, and tied off to create a stand-off under aseptic conditions. The water-filled glove is placed on the lacrimal sac area, and then, an aqueous gel is applied between the probe and the glove [Fig. 2, Video 1]. The ostium is identified on the scan by observing an area of discontinuity in the lacrimal and frontal process of the maxillary bone. The size is measured using an in-built calliper in the B-scan machine [Fig. 3a and b, Video 2]. Ostium is localized by a hyperechoic line corresponding to the bony margins of the created osteotomy and a central hypoechoic (dark) area representing the patent fistula or opening, where soft tissues and fluid allow less ultrasound reflection. Pre- and post-DCR surgery functional patency was assessed using FDDT, performed using cobalt blue light to demonstrate effective dye drainage [Fig. 3c and d].

Figure 2.

Procedure of B-scan probe with the immersion stand-off placed on the area of the ostium under aseptic conditions

Figure 3.

(a and b) show the measurement of the length (vertical measurement) and width (horizontal measurement) of the ostium, respectively, using the inbuilt calipers in the B-scan machine. Ostium is localized by a hyperechoic line corresponding to the bony margins of the created osteotomy and a central hypoechoic (dark) area representing the patent fistula or opening, where soft tissues and fluid allow less ultrasound reflection. (c and d) show FDDT pre- and post-DCR surgery using cobalt blue light, hence showing effective drainage of the dye postsurgery

Statistical analysis

The statistical analysis was done on Microsoft Excel 2013 and a Python Google Colab notebook. Statistical methods used were descriptive statistics, Pearson coefficients, paired t-test, ANOVA test, Chi-square test, and a multilinear regression model equation. Of less than 0.05 was considered statistically significant. Python coding was used for ROC curve plotting.

Results

In this study, 50 patients were included with ages ranging from 36 to 87 years in the external DCR group, and patients with 20 to 71 years were recruited in the endonasal DCR group. The mean age was 54.59 (±13.26) years in the external group and 51.76 (±13.79) years in the endonasal group. There were 7 (26%) males and 18 (72%) females in external DCR and 4 (16%) males and 21 (84%) females in endonasal DCR. Table 1 summarizes the ostium length, width, and area across both the groups. Ostium areas were across all three follow-ups in both procedures [Fig. 4a].

Table 1.

Descriptive statistics of the ostium length, width, and area in external and endonasal DCR groups

n=50		Mean ostium length (±SD) in mm		Mean ostium width (±SD) in mm		Mean ostium area (±SD) in (mm2)
Intraoperative (Ext DCR)		12.48 (±2.4)		8.1 (±1.54)		101.14 (±26.63)
Intraoperative (Endo DCR)		10.58 (±2.4)		7.5 (±1.37)		79.89 (±19.12)
Post op day 15 (Ext DCR)		12.48 (±2.36)		8.06 (±1.53)		100.88 (±26.83)
Post op day 15 (Endo DCR)		9.94 (±1.01)		7.40 (±1.36)		78.61 (±26.83)
Post op 1 month (Ext DCR)		11.96 (±2.40)		7.52 (±1.44)		90.54 (±25.73)
Post op 1 month (Endo DCR)		9.84 (±0.88)		6.70 (±1.33)		67.12 (±17.19)
Post op 3 months (Ext DCR)		11.66 (±2.12)		7.09 (±1.28)		78.91 (±24.47)
Post op 3 months (Endo DCR)		9.84 (±0.88)		6.46 (±1.14)		58.95 (±16.25)

Figure 4.

(a) compares the ostium areas in both external and endonasal DCR procedures. (b) shows a comparative analysis of FDDT among both the procedures in 3 follow-ups. (c) shows a comparative analysis of receiver operating characteristic (ROC) curves for FDDT at day 15 and 1 month, and (d) shows a comparative analysis of ROC curves for FDDT at 1 month and 3 months

The paired t-test performed for ostium length showed a statistically significant difference (P < 0.05) during intraoperative and all three follow-ups, and similar results were observed for ostium area. Although the paired t-test was performed for ostium width, only post-op 1 month showed a statistically significant difference (P = 0.04). Linear regression analysis revealed a higher coefficient of determination for external DCR (R² = 0.87) compared to endonasal DCR (R² = 0.74), indicating that 87% and 74% of the variation in ostium area were explained by predictors (length and width) [Table 2]. These predictors showed statistical significance (P < 0.05) for ostium area prediction at 3 months in both procedures. ANOVA indicated statistically significant differences in ostium area across all follow-ups, leading to post hoc pairwise comparisons. Intraoperative and 3-month ostium areas (P = 0.008) and day 15 to 3-month areas (P = 0.008) showed significant differences in both groups after Bonferroni correction.

Table 2.

Predictive regression model equation for the ostium area at post-op 3 months

n=50		R 2		Regression equation		P (X1=length)		P (X2=width)
External DCR		0.87		y=7.16 X1+10.80 X2+ (-81.21)		0.000		0.000
Endoscopic DCR		0.74		y=6.68 X1+9.74 X2+ (-69.66)		0.004		0.000

Fluorescent dye disappearance test illustrated the functional efficacy and comparative performances across both the procedures at three follow-ups [Fig. 4b]. The ability to diagnose true positive cases improved significantly over time for both external and endonasal DCR groups, with excellent outcomes. The sensitivity and specificity for external DCR were 0.91 and 0.50 between day 15 and 1 month, improving to 100% for both metrics by 3 months. Similarly, for endonasal DCR, the sensitivity and specificity improved from 0.28 and 0.25 to 100% and 80%, respectively, by 3 months [Table 3].

Table 3.

Classification report of the FDDT analysis

n=50		External DCR (postop day 15 and 1 month)		External DCR (1 month and 3 months)		Endonasal DCR (postop day 15 and 1 month)		Endonasal DCR (1 month and 3 months)
Sensitivity (Recall)		0.91		1		0.28		1
Specificity		0.50		1		0.25		0.8
Precision		0.95		1		0.66		0.95
Accuracy		0.88		1		0.28		0.96
F1 score		0.93		1		0.40		0.97

Precision, or positive predictive value, was excellent for both groups, with the final accuracy improving from 0.88 to 100% for external DCR and from 0.28 to 96% for endonasal DCR [Table 3]. The F1 score, representing the harmonic mean of recall and precision, indicated high reliability and clinical relevance of FDDT as a tool for assessing functional outcomes.

ROC curve analysis for FDDT showed progressive enhancement in predictive accuracy [Fig. 4c and d]. For external DCR, the area under the curve (AUC) improved from 0.71 at day 15 to a perfect 1.0 at 3 months, reflecting minimal false positives. Endonasal DCR exhibited a lower initial AUC of 0.27 but improved to 0.90 by 3 months.

Discussion

In the present study, immersion B-scan ultrasonography was introduced as a novel, noninvasive technique to assess postoperative ostium dimensions following external and endonasal DCR. Traditionally, ostium evaluation relies on endoscopic examination and CT imaging, which are either invasive or costly. In contrast, immersion B-scan offers a simple, reproducible, and cost-effective alternative that provides clear visualization of ostium margins and enables quantitative measurement of length, width, and area. This application of immersion B-scan thus extends its established role in posterior segment evaluation to lacrimal surgery, providing a valuable tool for monitoring anatomical outcomes and correlating them with functional patency.

This prospective study included 50 patients equally divided into external and endonasal DCR groups (25 each) and followed up on postoperative day 15, at 1 month, and at 3 months. Parameters measured were ostium length (L), width (W), area (OA), and functional patency via FDDT. The immersion B-scan technique was used for precise measurements of ostium dimensions, with area calculated by multiplying length and width. The external DCR group had a mean age of 54.96 ± 13.26 years (range: 36–87), while the endonasal group had a mean age of 51.76 ± 13.79 years (range: 20–71). Both groups exhibited a normal Gaussian age distribution, with the external DCR group encompassing a broader age range.

This study observed a predominantly female population, with 72% and 84% in the external and endonasal DCR groups, respectively. This aligns with findings from Balwinder, who reported 63% female predominance in endonasal DCR,[6] and Bulent, who noted 80% in external DCR, though the present study included more middle-aged patients.[7] The higher female prevalence is attributed to the shorter, narrower, and S-shaped nasolacrimal duct in women.

Statistical analysis using paired t-tests demonstrated significant differences in ostium length (P = 0.000) over three follow-ups, indicating notable variation in ostium construction. Ostium width showed statistical significance at 1 month (P = 0.04) and clinical relevance on day 15 (P = 0.09) and at 3 months (P = 0.08). For the ostium area, paired t-tests confirmed a statistically significant difference (P = 0.01) across follow-ups in both procedures, reflecting substantial changes in ostium size.

Mean B-scan measurements revealed larger ostium lengths in external DCR compared to endonasal DCR, suggesting better structural outcomes [Table 1]. These findings align with studies like Balwinderi et al., which reported a progressive reduction in ostium length in endonasal DCR,[6] and M. Ali et al.,[8] which documented slightly higher lengths in endonasal DCR in contrast to the current study.

The mean B-scan measurements for ostium width seemed to be similar and observed a slight reduction in the ostium width in both groups during the three follow-ups in the present study [Table 1]. Balwinder et al. study on endoscopic DCR measured intraoperative ostium width using an olive-tipped suction tube as 7.2 mm, comparable to the 7.5 mm ostium width measured by Schirmers strip in the present study.[6] M Ali et al.[8] study reported postoperative ostium widths of 7.0 mm and 6.6 mm at 1 and 6 months follow-ups, similar to the study.

Ostium area measurements revealed a progressive contraction over time, though external DCR maintained a consistently larger area than endonasal DCR [Table 1, Fig. 4a]. This study recorded an intraoperative area of 101.14 ± 26.63 mm² in external DCR, reducing to 78.91 ± 24.47 mm² at 3 months, comparable to Yalcin Tok et al.’s[9] reduction from 230 ± 48.72 mm² to 52 ± 50.12 mm² over 6 months. Similarly, Al Huthail, observed a decrease from 142 mm² to 23 mm² over 33 months.[8]

For endonasal DCR, Kanchan, reported intraoperative areas between 28 and 108 mm², consistent with this study’s range of 54–115 mm². At 1 and 3 months, Kanchan et al. observed areas of 0–35 mm² and 0–24 mm², compared to this study’s larger measurements of 41.65–105.84 mm² and 36.9–97.92 mm², reflecting better patency maintenance confirmed by FDDT results. These comparisons underscore the superior ostium stability and functional outcomes of external DCR.[10]

Weng Onn Chan, Selva[11] reported a 50% shrinkage in ostium area by the fourth week and an additional 15% reduction over 12 months in endonasal DCR patients. In contrast, the present study observed a 14% reduction by the 4^th week and 11% between 4 weeks and 3 months. Similarly, Balikoglu et al.[12] found the final ostium size in external DCR (33.7 ± 17.4 mm²) to be significantly larger than in endonasal DCR (19.0 ± 8.9 mm²) at 12 months, aligning with the present study’s findings of 78.9 ± 24.47 mm² and 58.95 ± 16.25 mm², respectively, at 3 months (P = 0.001).

FDDT bar plots showed a rise in negative cases for endonasal DCR, likely due to gel foam use, which aids hemostasis and prevents adhesions [Fig. 4b]. Chi-square analysis of FDDT revealed a significant difference (P = 0.0001) between groups on the fifteenth postoperative day, indicating independence and rejecting the null hypothesis. However, at 1 month and 3 months, no significant differences were found (P = 0.38 and P = 0.22), suggesting group outcomes converged over time. Joon Ho Roh and Chi[13] similarly noted initial declines in FDDT values post-endonasal DCR, followed by improvements as inflammation subsided, reflecting maintained patency in later follow-ups.

A limitation of this study is that in the immersion B-scan probe images, the ostium appears as discontinuous bone with hazy margins, leading to possible inter-observer variability. To reduce false positives, the probe must be positioned perpendicularly over the ostium. Standardized probe placement improves measurement accuracy and reliability.

Conclusion

The present prospective study on external and endonasal DCR procedures discovered a progressive reduction in the ostium length, width, and area during the three follow-ups. Analysis of ostium length and area established a statistically significant difference, suggesting variability in constructing the dimensions. Ostium length and width were the significant predictors that explained the prediction of the final ostium area in both procedures. B-scan immersion technology provided a reliable and efficient measurement of ostium dimensions in both procedures. Therefore, this study recommends the use of the immersion B-scan method, which is a simple and quick way of determining the ostium size for analysis during follow-ups.

The fluorescein dye disappearance test, as it is a noninvasive test for patency of rhinostomy, achieved 100% sensitivity at final follow-up in both surgical procedures. Hence, we recommend the inclusion of FDDT in patients undergoing DCR.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.