Usefulness of Intraocular Pressure With the Ocular Response Analyzer to Predict Postoperative Hypotony Complications in Glaucoma

Abstract

Purpose: To investigate the usefulness of intraocular pressure (IOP) using the ocular response analyzer to predict the occurrence of hypotony complications following trabeculectomy or bleb needling revision with mitomycin C.

Methods: This study included 66 eyes of 66 patients who underwent trabeculectomy (58 eyes of 58 patients) or bleb needling (8 eyes of 8 patients) with mitomycin C. A significant predictor of hypotony complications was identified from (1) operation (trabeculectomy or bleb needling), (2) age, (3) sex, (4) disease type (primary open-angle glaucoma, primary angle closure glaucoma, or exfoliation glaucoma), (5) lens status (phakia or pseudophakia), (6) preoperative Goldmann applanation tonometry IOP, (7) preoperative central corneal thickness, (8) preoperative axial length, (9) preoperative anterior chamber depth, (10) preoperative corneal hysteresis, (11) preoperative corneal resistance factor, (12) preoperative corneal compensated IOP (IOPcc), and (13) minimum IOP (IOP value when hypotony complications occurred, otherwise the minimum IOP during 3 months from trabeculectomy or bleb needling) using multivariate logistic regression.

Results: The probability of the occurrence of hypotony complications tended to increase by applying higher cutoff values to preoperative Goldmann applanation tonometry IOP and IOPcc, but not lower cutoff values to the minimum IOP. Multivariate logistic regression suggested that higher preoperative IOPcc and pseudophakia were significant predictors of the occurrence of hypotony complications (P = 0.0062 and 0.0069, respectively).

Conclusions: Higher preoperative IOPcc and pseudophakia were significant predictors of the occurrence of hypotony complications.

Translational Relevance: It is useful to measure IOP using the ocular response analyzer before trabeculectomy.

Introduction

Glaucoma is one of the leading causes of blindness worldwide.¹ The only modifiable established factor to halt the progression of glaucoma is the intraocular pressure (IOP), which should be adequately controlled to avoid visual field deterioration.^2–10 Since its proposal in the 1960s by Cairns,¹¹ trabeculectomy has been considered as one of the gold standard surgical methods for glaucoma. Although this procedure is usually highly effective in lowering the IOP, postoperative complications, such as hypotony maculopathy owing to overfiltration, for which intensive postoperative care is needed, may occur. When the IOP decreases below the episcleral venous pressure and/or is accompanied by an inflammatory reaction, leakage from the choriocapillaris develops, which subsequently leads to a suprachoroidal effusion (choroidal detachment [CD]) and/or hypotony maculopathy. The same is true after bleb revision surgeries. 

Hypotony complications are often associated with severe visual acuity loss, which can be irreversible.^12–14 Previous studies have suggested hypotony that adjunctive use of antifibrotic agents, such as mitomycin C (MMC), lower postoperative IOP, and greater IOP difference between preoperative and lowest postoperative IOPs are risk factors for hypotony complications.^15–17 Other studies suggested younger age, myopia, male sex, pseudophakia, and greater central corneal thickness (CCT) as the risk factors; however, the results are largely different among studies.^15,18–23 

At present, it has become possible to measure corneal biomechanical properties using the ocular response analyzer (ORA, Reichert Ophthalmic Instruments, Depew, NY).^24–30 Previous studies have reported the usefulness of this aspect in analyzing the mechanism of glaucoma. ORA measures corneal hysteresis (CH) by measuring the difference of pressures at the inward and outward corneal movement after applying air pulse and represents the dampening capacity of the cornea,^31–33 and it has been reported that CH is closely related to the development and progression of glaucoma.^34–36 

The present study investigated the usefulness of measuring IOP using the ORA to predict the occurrence of hypotony complications after trabeculectomy or bleb needling revision with MMC.^37–39 

Methods

This retrospective multicenter study was approved by the Research Ethics Committee of Seirei Hamamatsu General Hospital (#3835) and was conducted in accordance with the tenets of the Declaration of Helsinki. All participants signed a written informed consent form for their clinical information to be stored in the hospital database and used for research. 

Participants

The study initially included 56 eyes of 56 patients who underwent trabeculectomy with MMC and 10 eyes of 10 patients who underwent bleb needling revision with MMC at the Department of Ophthalmology in Seirei Hamamatsu General Hospital between October 2020 and July 2023. 

Surgical Technique

Trabeculectomy

After a sub-Tenon injection of 1% xylocaine, a fornix-based peritomy was performed on the superotemporal or superonasal meridian. Subconjunctival 0.02% MMC was applied, and then a 3.5 × 3.5-mm square half of the full-thickness scleral flap was created and extended 1 mm into the clear cornea. Beneath the first scleral flap, a second scleral flap with a size of 1.5 × 3.5 mm and one-quarter of the scleral thickness was created, extended to the limbus, and then removed by excision at the limbus. The anterior chamber (AC) was entered from the surgical limbus, and peripheral iridectomy was performed. The first flap was repositioned and secured by four 10-0 nylon sutures. These sutures were also used to close the conjunctival wound. Thereafter, 0.8-mg dexamethasone was injected subconjunctivally, and the wound was dressed with 0.3% ofloxacin ointment. Topical 0.3% gatifloxacin and 0.1% betamethasone eye drops were applied for 3 months. All the operations were performed by a single surgeon (R.A.). After the decision of the surgeon, postsurgical laser suture lysis was performed. 

Bleb Needling

The bleb needling group comprised patients who underwent bleb needling ≥5 years after trabeculectomy. Patients in the trabeculectomy group were not included. Under topical anesthesia, 0.2 to 0.3 mL of 0.017% MMC (0.05% MMC was diluted three times in 1% xylocaine) was injected into the superior subconjuctival space. A cellulose spear swab was used to disperse the MMC subconjunctivally in the area of the intended bleb. The subconjunctival scar tissue was dissected using a 25 G needle 1 hour after injection. If there was no drainage through the scleral flap, the flap was lifted, and the needle was inserted into the AC. Postoperatively, all glaucoma drops were stopped, and topical 0.3% gatifloxacin and 0.1% betamethasone eye drops were applied for ≥1 month. All operations were performed by a single surgeon (R.A.). 

Hypotony Complications

There is no standard definition of hypotony complications in the literature, with variations in the numerical definitions used, in the definitions of the onset and duration of hypotony complications, and in whether it is associated with clinical signs such as maculopathy and CD.⁴⁰ The present study adopted the definition of clinically significant hypotony complications: any patient requiring treatment to correct or prevent hypotony-related complications of CD, hypotony maculopathy, or shallow AC, irrespective of the IOP. CD was diagnosed funduscopically or using ultra-widefield imaging (Optos California; Optos PLC, Dunfermline, UK). In addition, hypotony maculopathy was diagnosed funduscopically or using Spectralis OCT (Heidelberg Engineering, Heidelberg, Germany). 

Clinical Data Acquisition

The patients’ baseline demographic data, such as age and sex, were collected from their medical charts. The axial length (AL) and AC depth were measured preoperatively using OA-2000 (Tomey, Aichi, Japan). IOP was measured using the Goldmann applanation tonometer (GAT). From the medical chart, the IOP value when hypotony complications occurred was identified; otherwise, the minimum IOP during 3 months from trabeculectomy or bleb needling was recorded (minimum IOP). The change in IOP was calculated as (IOP before trabeculectomy or bleb needling – minimum IOP). CCT was measured using Corvis ST (OCULUS, Wetzlar, Germany). CH and corneal resistance factor (CRF) were calculated using the ORA. 

ORA Measurement

Details of the ORA measurement were thoroughly described elsewhere.⁴¹ The ORA records two applanation pressures: during the inward and outward movements of the cornea after the application of a rapid air jet pressure. Owing to its viscoelastic property, the cornea dissipates a part of the energy given off by the air jet, resulting in a delay in the outward corneal movement and thus causing the difference in the pressures at the inward and outward applanation. This difference is called CH.³³ The ORA measurement was performed three times, with a 5-minute interval between each measurement. Only the measurements with a quality index of >7.0 were used, and the average value was used in the analysis. CRF is a parameter indicating corneal stiffness, calculated from CH. The corneal compensated IOP (IOPcc) was also measured using ORA. 

Statistical Analysis

First, various variables were compared between eyes with hypotony complications and those without using the Wilcoxon test for the numerical variable and χ² test for categorical variables. Then, the odds ratio of (1) operation (trabeculectomy or bleb needling), (2) age, (3) sex, (4) disease type (primary open-angle glaucoma, primary angle closure glaucoma, or exfoliation glaucoma), (5) lens status (phakia or pseudophakia), (6) preoperative GAT IOP, (7) preoperative CCT, (8) preoperative AL, (9) preoperative AC depth, (10) preoperative CH, (11) preoperative CRF, (12) preoperative IOPcc, and (13) minimum IOP for the occurrence of hypotony complications was determined via univariate logistic regression analysis. Subsequently, multivariate logistic regression analysis was conducted, with the occurrence of hypotony complications as the target variable and the aforementioned 13 variables as the independent variables. Then, model selection was performed using the second-order bias-corrected Akaike information criterion (AICc) index; the optimal model for the occurrence of hypotony complications was identified from all 2¹³ patterns using the 13 candidate variables. The AIC is a well-known statistical measurement used in model selection, and the AICc is a corrected version of the AIC, which provides an accurate estimation even when the sample size is small.⁴² It is recommended to use model selection methods, instead of multivariate regression, to improve the model fit by removing redundant variable, because the degrees of freedom decreases as the number of variables increases.^43,44 

Moreover, to further validate the usefulness of the result, we conducted a leave-one-out cross validation. In this analysis, the whole data was divided to training dataset (n = 75) and testing dataset (n = 1). Then, a model to predict the occurrence of hypotony complication using various variables was constructed using the training dataset. Thereafter, the occurrence of the hypotony complication was predicted for the testing dataset using the constructed model. This process was iterated 76 times, so that each patient was used as a testing dataset once. Finally, the area under the precision recall curve was calculated using the predicted value. 

All data processing and analyses were conducted using the statistical programming language R (The R Foundation for Statistical Computing, Vienna, Austria). 

Results

The present study included 66 eyes of 66 patients who underwent trabeculectomy (58 eyes of 58 patients) or bleb needling (8 eyes of 8 patients) with MMC. The patients’ baseline characteristics are presented in Table 1. Their mean preoperative ages were 66.5 ± 9.8 years, and 38 of them were men. There was one eye with a shallow AC, CD, and hypotony maculopathy and six eyes with CD and hypotony maculopathy, one eye with CD, and two eyes with CD and hypotony maculopathy. Table 2 presents the comparisons of various variables between eyes with and without hypotony complications (Wilcoxon test for the m\numerical variable and χ² test for categorical variables). Eyes with hypotony complications had deeper AC, higher preoperative IOP and IOPcc, and lower CH than those without (P = 0.041, P = 0.0033, P = 0.0016, and P = 0.044, respectively). Pseudophakia was predominant in eyes with hypotony complications compared with those without (P = 0.0038). No significant difference was observed in the remaining variables of sex, right or left eye, AL, CCT, minimum IOP, and CRF. 

Table 1.

 

Patient Demographics

 

Table 2.

 

Comparisons of Variables Between Eyes With and Without Hypotony Complications

 

Table 3 presents the result of multivariate logistic regression analysis and AICc model selection for the occurrence of hypotony complications. Higher preoperative IOPcc and pseudophakia were significant predictors of the occurrence of hypotony complications (P = 0.0062 and P = 0.0069, respectively). Other variables of sex, eye (right or left), AC depth, CCT, minimum IOP, preoperative GAT IOP, CH, CRF, disease type, and operation (trabeculectomy or bleb needling) were not included in the optimum model, as a result of the AICc model selection. 

Table 3.

 

Result of Multivariate Logistic Regression Analysis and AICc Model Selection for the Occurrence of Hypotony Complications

 

Table 4 demonstrates that the probability of the occurrence of hypotony complications did not increase by applying stricter (lower) cutoff value to minimum IOP, preoperative GAT IOP, and preoperative IOPcc. The probability tended to increase by applying higher cutoff values to preoperative GAT IOP and IOPcc but not lower cutoff values to minimum IOP. 

Table 4.

 

The Probability of the Occurrence of Hypotony Complications did not Increase by Applying a Stricter (Lower) Cutoff Value to the Minimum IOP, Preoperative GAT IOP, and Preoperative IOPcc

 

As a result of the leave-one-out cross-validation, the area under the precision recall curve value only with nonsignificant variables (age and AL; see Table 3) was 0.164, which significantly increased to 0.548 by including lens status (phakia or IOL) and also to 0.500 by including preoperative IOPcc (P < 0.05, bootstrap test). 

Discussion

In the present study, 66 eyes of 66 patients who underwent trabeculectomy or bleb needling with MMC were followed up, and the parameters associated with the occurrence of hypotony complications were identified. As a result, hypotony complications were observed in 10 eyes, and higher preoperative IOPcc and pseudophakia were found to be significant predictors of the occurrence of hypotony complications. This result was also supported by the leave-one-out cross-validation. Other variables, such as minimum IOP, CCT, and disease type, were not significantly associated with the occurrence of hypotony complications. 

The incidence rates of CD and hypotony maculopathy have been reported to be between 7.9% and 50.0%^45–47 and between 8.9% and 29.5%,^48,49 respectively, after trabeculectomy. In the present study, hypotony complications were observed in 10 of 66 eyes (15%). Various predictors of the occurrence of hypotony complications have been proposed, including lower postoperative IOP, greater IOP difference between preoperative and lowest postoperative IOPs,^15–17 younger age, myopia, and male sex.^18–20 No study has investigated the association between IOPcc and hypotony complications. In the present study, as a result of multivariate logistic regression analysis, preoperative IOPcc and pseudophakia were identified as the only two significant variables (Table 3). Consistent with the present study, pseudophakia has been reported to be a significant predictor of hypotony complications.^15,23 This is probably because intraocular surgery, even the recent small incision cataract surgery, causes the blood–aqueous barrier to break down.^50,51 Indeed, it has been reported that the levels of cytokines, such as monocyte chemoattractant protein-1 and interleukin-8, remain elevated in the aqueous humor >1 year after phacoemulsifcation.^50,51 

Preoperative IOPcc was a significant predictor in multivariate logistic regression analysis, but preoperative GAT IOP was not (Table 3), suggesting that preoperative IOPcc was more useful than preoperative GAT IOP. The relative likelihood replacing preoperative IOPcc with preoperative GAT IOP better describes the occurrence of hypotony complications was 83.2% (calculated as exp ([smaller AICc – larger AICc]/2) (see reference⁵²); data not shown in Results. IOPcc is the IOP corrected for CH. Specifically, IOPcc decreases in eyes with a large CH, but increases in those with a small CH. As a result, the preoperative IOP increases with IOPcc than with GAT in eyes with low CH; high IOPcc is a good predictor of the occurrence of hypotony complications. CH represents hysteresis of an eye. CH is thought to decrease with high cross-linking within the cornea, such as in aged eyes.⁵³ This finding suggests that IOPcc increases in eyes with low cross-linking. The same is true for the sclera because the cornea and sclera are continuous collagenous structures of an eye, and their biomechanical characteristics may be similar.^54,55 Embryologically, the sclera and Bruch's membrane are both derived from the neural crest.⁵⁶ Dawson et al.⁵⁷ and Pillunat et al.⁵⁸ have reported that the GAT IOP tends to be lower than IOPcc after trabeculectomy owing to the change in corneal biomechanics, and it was more beneficial to use IOPcc than GAT IOP when managing low IOP after trabeculectomy. In contrast, reliable ORA measurement cannot be performed in such eyes. The present study suggested the usefulness of preoperative ORA measurement in predicting the occurrence of hypotony after trabeculectomy. 

It is noteworthy that low scleral rigidity is an important element in the pathogenesis of hypotony maculopathy,^19,59 because low scleral rigidity facilitates the inward collapse of the scleral wall during hypotony.¹⁴ The reason why young age and myopic eyes are considered to be predictors of the occurrence of hypotony complications can be thought to be related to the thinner and less rigid sclera.⁵⁹ This result indicates that a high preoperative ORA-IOP adjusted with CH is more useful than preoperative GAT IOP, age, and AL. CH was a significant predictor in the univariate logistic regression analysis (data not shown in the Results), but not in the multivariate analysis (Table 3). This finding suggests that IOP adjusted by CH (IOPcc) is more useful than CH alone. 

In the present study, the preoperative IOP was used in the analysis instead of the magnitude of the decrease of GAT, because it is more clinically straightforward to use preoperative variables rather than postoperative change. Furthermore, replacing the change in the GAT IOP with the preoperative IOPcc resulted in poorer explanation of the occurrence of hypotony complications (AICc increased from 40.6 to 45.8; data not shown in the Results). These results suggest that IOPcc is more clinically useful than the change in GAT IOP. However, this does not deny the significance of minimum IOP; the current findings are obtained when the IOP is decreased to a relatively narrow range of minimum IOP (4.8 ± 1.9 mm Hg); hence, this result would not be valid with different levels of postoperative IOP. This factor, however, does not spoil the clinical relevance of the results of this study, because the current minimum IOP is thought to be a good postoperative target IOP after trabeculectomy and bleb needling. 

In the real-world clinical setting, efforts may be made to avoid hypotony complications using a minimum IOP. However, results from the present study suggest that the occurrence of hypotony complications cannot be explained by merely using the parameter with both the univariate and multivariate analyses. Furthermore, the probability of the occurrence of hypotony complications did not increase by applying a stricter (lower) cutoff value to the minimum IOP (Table 4). This finding is indeed contrary to the tendency of the probability of the hypotony complications to dramatically increase with the increase in preoperative GAT and IOPcc. This result agrees well with Gass,⁵⁹ who showed that the occurrence of hypotony complications with low IOP is decided by the rigidity of the sclera. Nicolela et al.²² have reported that the IOP at the time of hypotony complications was not significantly different between eyes with and without hypotony maculopathy in eyes with an IOP of ≤6 mm Hg after trabeculectomy; however, greater CCT was a significant risk factor. The results of the present study indicated that neither the minimum IOP nor CCT was a significant risk factor and that it was more beneficial to conduct ORA measurement preoperatively. 

The present study has limitations. First, the postoperative IOPcc or CH, which may be closely associated with the occurrence of hypotony complications, was not measured. However, these measurements may not be relevant clinically because it is often not realistic to conduct and obtain reliable results of the ORA measurement in eyes shortly after trabeculectomy. Second, this study was retrospective and included a relatively small sample size. This factor may have limited the robustness of the obtained results. We conducted a leave-one-out cross-validation to get rid of this possible problem; however, a further prospective study with a larger sample size is warranted to validate the results of this study. Finally, there is a huge variation in published definitions of hypotony complications, as detailed in.⁴⁰ Such variations in defining postoperative hypotony can have a large impact on the reported success and failure rates among studies, and there is a need for a more robust universal definition, focusing on clinically important signs, to allow better comparison between different treatment modalities.⁴⁰ 

In conclusion, eyes that underwent trabeculectomy or bleb needling with MMC were followed up, and the parameters associated with the occurrence of hypotony complications were identified. As a result, it was suggested that higher preoperative IOPcc and pseudophakia were significant predictors of the occurrence of hypotony complications. 

Acknowledgments

Supported by grants (numbers 19H01114, 18KK0253, 20K09784, and 20K18337) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan; the Translational Research program and Japan Glaucoma Society Research Project Support Program. 

Disclosure: J. Ryu, None; R. Asaoka, None; S. Nakakura, None; H. Murata, None; Y. Nakaniida, None; K. Ishii, None; A. Obana, None; Y. Kiuchi, None