Transient Ciliochoroidal Detachment After Ab Interno Trabeculotomy for Open-Angle Glaucoma: A Prospective Anterior-Segment Optical Coherence Tomography Study

Importance  Although trabeculotomy is a glaucoma surgical procedure for modest intraocular pressure (IOP) reduction, some eyes exhibit very low IOP during the early postoperative period. To our knowledge, the reason and its effect have not been investigated.

Objectives  To investigate ciliochoroidal detachment (CCD) immediately after ab interno trabeculotomy and evaluate its effect on IOP immediately after surgery.

Design, Setting, and Participants  This prospective, observational, case series was conducted at Kyoto University Hospital, Kyoto, Japan. Patients with open-angle glaucoma who underwent ab interno trabeculotomy using a microsurgical device by a single surgeon between July 1, 2014, and May 31, 2015, were included. Thirty-seven consecutive patients were enrolled, 33 of whom were included in the analysis. The dates of the analysis were August 1 to August 15, 2015.

Intervention  Imaging of CCD using anterior-segment optical coherence tomography (AS-OCT).

Main Outcomes and Measures  The incidence of CCD immediately after ab interno trabeculotomy and its effect on IOP in the early postoperative period.

Results  The study cohort comprised 33 patients. Their mean (SD) age was 69.4 (13.2) years, and 19 (58%) were male. At postoperative day 3, CCD was detected in 14 of 33 eyes (42%) (CCD group) using AS-OCT. The CCD group had shorter axial length (mean [SD], 23.66 [1.67] vs 25.16 [1.59] mm) and thinner central corneal thickness (mean [SD], 505.9 [35.8] vs 533.9 [39.1] μm) than the non-CCD group. Only 5 eyes had CCD at postoperative day 10, and 4 of these eyes had CCD at 1 month after surgery. The postoperative IOPs at all follow-up periods were lower in the CCD group than in the non-CCD group, but the difference in the postoperative IOPs between the groups decreased as time passed. The mean (SD) IOPs for the CCD group vs the non-CCD group were 9.1 (3.0) vs 14.2 (5.8) mm Hg at day 1, 8.4 (2.4) vs 13.4 (5.0) mm Hg at day 3, 11.0 (3.0) vs 15.5 (6.3) mm Hg at day 10, 13.4 (2.4) vs 15.5 (3.3) mm Hg at 1 month, and 13.9 (3.4) vs 15.5 (4.0) mm Hg at 3 months. In several eyes in the CCD group, the AS-OCT images revealed a connection between the CCD and the anterior chamber via the trabeculotomy site.

Conclusions and Relevance  Ciliochoroidal detachment after ab interno trabeculotomy was not rare and was associated with low IOP immediately after surgery. Postoperative CCD may be partially attributed to the transient increase in uveoscleral aqueous outflow via the trabeculotomy site.

Two major routes for aqueous humor outflow are the conventional (trabecular) outflow pathway and the unconventional (uveoscleral) outflow pathway. In the trabecular outflow pathway, aqueous humor drains through the trabecular meshwork and the collector channels into the episcleral venous system. In the uveoscleral outflow pathway, aqueous humor drains through the face of the ciliary body and iris root to the ciliary muscle and suprachoroidal space, and then subsequently through a vein in the choroid and sclera or through scleral pores to the episcleral tissue.^1,2 In humans, 75% of the resistance to aqueous humor outflow is localized to the trabecular meshwork, and 25% occurs beyond the Schlemm canal.³

The basis of trabeculotomy, which is a surgical procedure performed for glaucoma, is reduction of abnormal outflow resistance by removing the inner wall of the Schlemm canal and the trabecular meshwork.^4,5 A microsurgical device (Trabectome; NeoMedix Corporation) is used for ab interno trabeculotomy.^4-6 The collector channels in the scleral wall of the Schlemm canal are exposed to the anterior chamber and are recruited for aqueous drainage into the scleral and episcleral venous plexus after ab interno trabeculotomy. Although this procedure can achieve a modest intraocular pressure (IOP) reduction, it cannot be expected to lower IOP beyond the episcleral venous pressure (EVP) level.

Although prolonged hypotony or choroidal effusion has not been reported after surgery with the microsurgical device,^5,6 approximately 1% (7 of 679 eyes and 24 of 1878 eyes) developed early transient hypotony (IOP, <5 mm Hg) on the first postoperative day.^7,8 In seated individuals, the EVP has been reported to be high, ranging from 7.6 to 11.4 mm Hg.⁹ Ocular hypotension values less than the EVP cannot be explained solely by an increase in aqueous outflow via the transcanalicular conventional route because the trabecular outflow is restricted by the EVP. The large SDs of the IOP at 1 day after ab interno trabeculotomy (mean [SD], 16.6 [8.8] mm Hg for trabeculotomy only and 17.8 [9.0] mm Hg for combination with phacoemulsification) also indicate that the IOP varies widely, with some eyes exhibiting very low IOP during the early postoperative period.⁷ To our knowledge, no study has investigated the reason for the unexpectedly low IOP in the early postoperative period and its effect on the subsequent outcome of ab interno trabeculotomy.

Ultrasonographic biomicroscopy and anterior-segment optical coherence tomography (AS-OCT) have been used for in vivo observation of the angle structures and ciliochoroidal space.^10-13 The advantages of AS-OCT for postoperative ciliochoroidal space assessment are that it uses a noncontact approach and provides images of high axial resolution. In contrast to the trabecular pathway, the outflow through the uveoscleral pathway is thought to be independent of IOP.¹⁴ Several previous studies reported that postoperative ciliochoroidal detachment (CCD) after some types of surgery (eg, deep sclerectomy,^15,16 trabeculectomy,^13,17,18 and ab interno microstent implantation¹⁹) is owing to an acceleration of the uveoscleral outflow of aqueous humor.

In the present study, we prospectively investigated the presence and extent of CCD before and immediately after ab interno trabeculotomy using AS-OCT. We also examined the effects of postoperative CCD on the early postoperative IOPs.

Box Section Ref ID

This prospective observational study adhered to the tenets of the Declaration of Helsinki,²⁰ was approved by the institutional review board and ethics committee of the Kyoto University Graduate School of Medicine (Kyoto, Japan), and was registered with the University Hospital Medical Information (UMIN) Network Clinical Trials Registry of Japan.^21,22 Written informed consent was obtained from all participants. This study included patients with open-angle glaucoma who underwent ab interno trabeculotomy using the microsurgical device at Kyoto University Hospital between July 1, 2014, and May 31, 2015, performed by one of us (T.A.). Patients who had undergone previous ocular surgical procedures were excluded from this analysis except for those who underwent cataract surgery before ab interno trabeculotomy. When both eyes met the inclusion criteria, the eye that was treated first was included in the analysis.

Ab interno trabeculotomy using the microsurgical device was performed through a 1.70-mm temporal clear corneal incision. The nasal trabecular meshwork was visualized with a Swan-Jacob lens (Autoclavable Gonioprism; Ocular Instruments). The microsurgical device hand piece was used to ablate the trabecular meshwork and the inner wall of the Schlemm canal nasally to form a 100° to 120° arc. Viscoelastic was not used until ab interno trabeculotomy was completed. For patients undergoing ab interno trabeculotomy combined with cataract surgery, the corneal incision was expanded to 2.40 to 2.75 mm after ab interno trabeculotomy, and we subsequently performed cataract extraction and intraocular lens implantation by phacoemulsification. The wound was closed using a single 10-0 nylon suture only when leakage of aqueous humor was observed from the wound.

All patients received a treatment regimen consisting of corticosteroid, antibiotic, and pilocarpine hydrochloride (2%) eyedrops beginning at day 1 after ab interno trabeculotomy. Patients undergoing combined surgery also received diclofenac eyedrops. Administration of corticosteroid and pilocarpine eyedrops was tapered over 8 and 12 weeks after surgery, respectively. Patients continued to use antibiotic or diclofenac eyedrops until 8 and 12 weeks after surgery, respectively. The IOP-lowering eyedrops were given at the same rate as before surgery and were subsequently reduced according to the individual postoperative IOP levels. All patients using oral carbonic anhydrase inhibitors abstained from use beginning 1 day before surgery.

All patients underwent a comprehensive ophthalmic examination before ab interno trabeculotomy. Included were the following: gonioscopy, slitlamp examination, IOP measurement with a Goldmann applanation tonometer, uncorrected and best-corrected visual acuity with a Landolt ring chart at 5 m, axial length measurement by partial laser interferometry (IOLMaster; Carl Zeiss Meditec), AS-OCT examination (Visante OCT, model 1000; Carl Zeiss Meditec), central corneal thickness (CCT) measurement with an ultrasonic pachymeter (SP-3000; Tomay), and standard automated perimetry (Humphrey Visual Field Analyzer [HFA] using the 24-2 Swedish Interactive Thresholding Algorithm [SITA] testing protocol [HFA plus 24-2 SITA]; Carl Zeiss Meditec).

Follow-up examinations were scheduled at 1 day, 3 days, 10 days, 1 month, and 3 months. Each follow-up included IOP measurement, slitlamp examination, and ophthalmoscopy. Visual acuity was assessed at 10 days and 3 months after surgery. All patients underwent AS-OCT examination at preoperative day 1 and postoperative days 3 and 10. In addition, AS-OCT examination was performed at 1 month if the AS-OCT images showed any CCD at postoperative day 10.

Anterior-segment optical coherence tomography was used to investigate CCD before and after surgery. In a seated position, the patient was asked to look to the opposite side during imaging. Taking care to avoid applying pressure to the globe, imaging was performed independently in each of the 4 quadrants (superior, inferior, nasal, and temporal), as well as 30° superior and inferior at the nasal area (Figure 1A). All participants underwent AS-OCT imaging in a well-lit room. The examination was repeated if motion or image artifacts resulting from eyelid movement were observed. Two independent observers (T.A. and E.N.) who were masked to the clinical data assessed the AS-OCT images. The CCD severity was classified based on the maximum CCD among the AS-OCT images, as previously reported.^10,11 Classification included grade 0 (no sign of CCD) (Figure 1B-D, left), grade 1 (slitlike, with CCD less than half of the ciliary body thickness) (Figure 1B, right), grade 2 (bandlike, with CCD at least half of the ciliary body thickness) (Figure 1C, right), and grade 3 (obvious, with CCD greater than the ciliary body thickness) (Figure 1D, right). The eyes with CCD (grades 1-3) and without CCD (grade 0) at postoperative day 3 comprised the CCD group and the non-CCD group, respectively.

Multiple-scan averaging was applied to swept-source OCT (SS-OCT) images (DRI OCT-1; Topcon) to generate high-resolution images, allowing for more detailed observation.²³ Ninety-six SS-OCT B-scans were averaged to evaluate CCD detail in some patients (using the same imaging protocol as AS-OCT).

All statistical evaluations were performed using commercially available software (SPSS, version 20; IBM Corporation). Unpaired t test or χ² test was used to compare age, axial length, CCT, visual field mean deviation, surgical procedure (combined or single), preoperative lens status (phakia or intraocular lens), preoperative and postoperative IOPs, and preoperative and postoperative glaucoma medications between the groups. A stepwise multiple regression analysis was used to determine the factors associated with the postoperative IOP levels. The data are presented as the mean (SD). Data analysis was conducted from August 1 to August 15, 2015.

Thirty-seven consecutive patients (37 eyes) were enrolled in the study between July 1, 2014, and May 31, 2015. Three eyes were excluded from the analysis because they did not meet the required 3 months of postoperative follow-up. Slight CCD was observed before surgery in 1 primary open-angle glaucoma, which received a grade of 1 and was excluded from the analysis because the objective of this study was to investigate the incidence of postoperative CCD. In total, 33 of 37 eyes (89%) were analyzed in this study (Table 1).

The 33 included eyes showed no obvious CCD before surgery. At postoperative day 3, a total of 14 eyes (42%) (CCD group) had grade 1 to 3 CCD, while 19 eyes (58%) (non-CCD group) had no CCD (grade 0) (Table 1 and Figure 1). The postoperative IOPs decreased in both the CCD and non-CCD groups, but no eye showed postoperative hypotony less than 5 mm Hg throughout the follow-up period among our patients (Figure 2). In the CCD group, 10 of 14 eyes showed CCD in all 6 images at 3 days after surgery. In the other 4 eyes, CCD was observed in some of the 6 images. Only 5 of 33 eyes (15%) had CCD at postoperative day 10, with 4 eyes classified as grade 1 and one eye classified as grade 2. Of these 5 eyes, 4 had grade 1 CCD at 1 month after surgery. The CCD group had shorter axial length and thinner CCT than the non-CCD group, with a mean (SD) axial length of 23.66 (1.67) vs 25.16 (1.59) mm and a mean (SD) CCT of 505.9 (35.8) vs 533.9 (39.1) μm. The postoperative IOPs at all follow-up periods were lower in the CCD group than in the non-CCD group, but the difference in the postoperative IOPs between the groups decreased as time passed. The mean (SD) IOPs for the CCD group vs the non-CCD group were 9.1 (3.0) vs 14.2 (5.8) mm Hg at day 1, 8.4 (2.4) vs 13.4 (5.0) mm Hg at day 3, 11.0 (3.0) vs 15.5 (6.3) mm Hg at day 10, 13.4 (2.4) vs 15.5 (3.3) mm Hg at 1 month, and 13.9 (3.4) vs 15.5 (4.0) mm Hg at 3 months.

To determine the factors associated with the postoperative IOP levels, a stepwise multiple regression analysis was performed using age, sex, diagnosis, axial length, CCT, visual field mean deviation, surgical procedure, preoperative IOP, preoperative glaucoma medication score, and CCD at postoperative day 3 as independent variables (Table 2). Higher CCD grades at postoperative day 3 were associated with lower IOPs at postoperative days 1, 3, and 10. The other associations with postoperative CCD are summarized in Table 2.

Although the AS-OCT images clearly revealed the site of the ablated internal wall of the Schlemm canal and postoperative CCD, the image quality was limited by the time-domain resolution limits inherent in the OCT technology (Figure 1). We observed that a low-reflective lesion seemed to be connected to the CCD and the anterior chamber at the ablated nasal side in some eyes within the CCD group (Figure 1C). To examine this finding further, a detailed evaluation was performed using high-resolution AS-OCT with multiple-scan averaging SS-OCT. This imaging method generated high-resolution AS-OCT images and revealed that CCD was likely to be directly connected to the anterior chamber via the Schlemm canal ablation area in some eyes with postoperative CCD. This result was not observed in the nonablated area (Figure 3).

In the present study, we found that CCD was common after ab interno trabeculotomy and was associated with lower IOP levels measured immediately after surgery. Postoperative CCD was mostly transient, and its effect on the postoperative IOPs lessened over time.

It is unclear how CCD occurs after ab interno trabeculotomy. Ciliochoroidal detachment is often described as uveal effusion, a well-known diagnostic feature of nanophthalmos, but may also be induced by ocular inflammation such as scleritis or Vogt-Koyanagi-Harada syndrome.²⁴ Scleral buckle surgery²⁵ and panretinal photocoagulation²⁶ have also been known to cause CCD. These conditions may be related to uveal congestion owing to venous obstruction or massive transudation in the choroid due to choriocapillaris disruption. Although ab interno trabeculotomy is unlikely to affect these mechanisms, inflammation induced by surgery could not be ruled out as a cause of CCD.

Postoperative hypotony is another possible cause of postoperative CCD because choroidal detachment frequently occurs after filtering surgery.²⁷ Sabti et al²⁸ reported that postoperative uveal effusion was observed in 5.8% (12 of 207) of patients who underwent cataract surgery. However, larger incisions were used for cataract surgery at the time of their study, and the scleral tunnel and clear corneal incisions were 5.0 to 6.5 mm long and 3.0 mm long, respectively. They reported that 8 of 12 eyes with postoperative uveal effusion showed wound leaks or subconjunctival blebs. In the present study, corneal incisions 1.70 mm long and 2.40 to 2.75 mm long were made for single and combined procedures, respectively, which are smaller than the incisions used in the study by Sabti et al. Furthermore, no wound leakage was observed in our study during any postoperative period, and there was no difference in postoperative CCD rates between single and combined procedures. Our data do not support postoperative wound leakage as the main cause of CCD after ab interno trabeculotomy. However, because intraoperative transient ocular hypotension usually occurs during this procedure and self-sealing of the corneal incision is not always reliable immediately after surgery, we cannot completely rule out the effect of transient ocular hypotension on postoperative CCD.

The main cause of CCD after trabeculectomy is posited to be postoperative hypotony, as mentioned in the previous paragraph. Other potential mechanisms such as increased uveoscleral outflow have also been suggested.^17,18 Trabeculectomy may enhance the uveoscleral outflow via a newly created route through the supraciliary spaces from the lake under the scleral flap to the uvea. Similarly, deep sclerectomy alone reportedly enhanced aqueous outflow to the supraciliary space.¹⁵ In this case, thinning of the sclera possibly induced a pressure gradient between the lake and the supraciliary space, which may have been the driving force of the uveoscleral outflow. However, a lake was not created during ab interno trabeculotomy. Amari et al²⁹ investigated trabeculectomy specimens from eyes with a history of trabeculotomy and noted that ab externo trabeculotomy may create a uveoscleral outflow pathway. In the present study, we observed that the trabecular meshwork cleft developed by ab interno trabeculotomy seemed to be connected to the CCD in some patients. It cannot be ruled out that a cyclodialysis cleft was formed during ablation of the trabecular meshwork and contributed to the CCD. Regardless, we believe that transient enhancement of the uveoscleral outflow pathway contributes to CCD after ab interno trabeculotomy and transiently reduces IOP.

Sabti et al²⁸ proposed that the combined use of topical pilocarpine and oral carbonic anhydrase inhibitors is a possible risk factor for CCD after cataract surgery, although this mechanism has not been clarified. All our patients received topical pilocarpine, but not oral carbonic anhydrase inhibitors, after surgery. During the first month after surgery, pilocarpine was used 3 or 4 times per day by all our patients, and postoperative CCD gradually decreased over time. We believe that pilocarpine use itself is not a major risk factor for postoperative CCD. However, we could not examine the association between the presence of CCD and pilocarpine use after tapering.

In the present study, we found that the eyes in the CCD group had thinner CCT and shorter axial length. Short axial length might be a risk factor for CCD after ab interno trabeculotomy because it is a purported risk factor for CCD after panretinal photocoagulation³⁰ and cataract surgery.²⁸ However, we found that the mean axial length increased from grade 1 to grade 3 (Table 2). We postulate that different mechanisms cause different types of postoperative CCD, which would explain this discrepancy in the association between CCD and axial length. On the other hand, CCT was decreased from grade 0 to grade 3. Further studies with larger sample sizes will be necessary to clarify this relationship.

When a patient has extremely low IOP and CCD as determined using AS-OCT immediately after ab interno trabeculotomy, the physician can anticipate that the IOP will gradually increase over 3 months. In clinical practice, physicians should note that prompt cessation of preoperative IOP-lowering eyedrops may accelerate the subsequent IOP increase.

The present study has some limitations. First, the sample size was small, and the follow-up period was short. Small but definitive correlations were potentially undetectable, and the statistical power of most factors was weak in our study owing to the small sample size. Because of the short follow-up period, we could not determine whether CCD in the early postoperative period affected the postoperative results after long-term periods or whether some postoperative CCD remained and contributed to IOP reduction until the late postoperative period. Further investigation with larger sample sizes and longer follow-up periods is necessary to elucidate the importance and long-term effects of postoperative CCD. Second, CCD was primarily evaluated using time-domain AS-OCT, and the imaging position was limited. High-resolution AS-OCT such as SS-OCT, coupled with more locations, would provide more detailed data regarding CCD. Further investigation will be necessary to elucidate the effects of CCD after ab interno trabeculotomy.

Ciliochoroidal detachment after ab interno trabeculotomy was not a rare phenomenon and was associated with low IOP immediately after surgery in our patients. These results suggest that postoperative CCD can be partially attributed to the transient increase in uveoscleral aqueous outflow via the trabeculotomy site. Because the IOP-lowering effect of CCD immediately after ab interno trabeculotomy is mostly transient, a subsequent increase in IOP would presumably occur in clinical practice.

Submitted for Publication: August 16, 2015; final revision received November 30, 2015; accepted December 1, 2015.

Corresponding Author: Tadamichi Akagi, MD, PhD, Department of Ophthalmology and Visual Sciences, Kyoto University Graduate School of Medicine, 54 Shogoin Kawahara-cho, Sakyo-ku, Kyoto 606-8507, Japan (akagi@kuhp.kyoto-u.ac.jp).

Published Online: January 28, 2016. doi:10.1001/jamaophthalmol.2015.5765.

Author Contributions: Dr Akagi had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Akagi, Nakanishi.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Akagi.

Obtained funding: Akagi, Yoshimura.

Administrative, technical, or material support: Akagi, Nakano, Uji.

Study supervision: Nakanishi, Uji, Yoshimura.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest, and none were reported.

Funding/Support: This research was supported by grant-in-aid for scientific research 25462713 from the Japan Society for the Promotion of Science and the Innovative Techno-Hub for Integrated Medical Bio-Imaging of the Project for Developing Innovation Systems from the Ministry of Education, Culture, Sports, Science and Technology in Japan.

Role of the Funder/Sponsor: The funding sources had no role in the design or conduct of this research; collection, management, analysis, or interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication.

Additional Contributions: Takafumi Miyazawa and Tatsuya Yamada (orthoptists employed by Kyoto University Hospital) obtained the anterior-segment optical coherence tomography images. No compensation was received outside of their usual salary.