Consensual Ophthalmotonic Reaction in Fellow Eyes in the Early Postoperative Period Following Unilateral Glaucoma Surgery

Abstract

Introduction

Consensual ophthalmotonic reaction (COR) refers to consensual intraocular pressure (IOP) change in the fellow eye following alteration of IOP in one eye. The study aimed to investigate COR and the magnitude of consensual IOP change after unilateral glaucoma surgery.

Methods

We analyzed 356 eyes from 293 patients who underwent unilateral trabeculectomy, bleb needling, trabectome, or microcatheter-assisted trabeculectomy (MAT) between 2016 and 2022. In the fellow eye, baseline IOP was compared with IOP at 1 day, 1 week, 1 months, 3 months, and 6 months postoperatively. Fellow eyes with IOP change exceeding 4 mmHg or 20% increase from baseline were defined as fellow eyes with COR. Univariate and multivariate analysis were performed to identify influencing factors associated with COR and IOP change in the fellow eye.

Results

In fellow eyes, IOP was 18.3 ± 7.3 mmHg at baseline, significantly increasing to 19.3 ± 7.5 mmHg at postoperative day 1 (POD1) and remaining stable at other time points, with a COR incidence of 40.7% (145 eyes) at POD1. Univariate and multivariate analysis identified preoperative use of α₂-receptor agonists in the fellow eye as a significant protective factor against COR after trabeculectomy and bleb needling. We also identified that key influencing factors for COR varied by surgery type: pseudophakia in the fellow eye was protective in trabeculectomy, history of diabetes was a risk factor for bleb needling, and prior glaucoma surgery predicted COR after trabectome. Greater IOP reduction in the operated eye correlated with a larger consensual IOP rise following minimally invasive glaucoma surgery (MIGS) procedures.

Conclusions

Collectively, our findings suggest that COR is a common early postoperative phenomenon regardless of glaucoma surgery type, and the preoperative use of α₂-receptor agonists is associated with a lower incidence of COR. Therefore, it is essential to monitor IOP in the fellow eye perioperatively.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40123-025-01307-8.

Key Summary Points

Why carry out this study?

Existing studies on the consensual ophthalmotonic reaction (COR) following unilateral glaucoma surgery are limited, show inconsistent findings, and lack data for modern minimally invasive glaucoma surgeries (MIGS).

The fellow eye often determines a patient's overall visual function, making its intraocular pressure (IOP) stability critical. However, postoperative IOP fluctuations in the fellow eye are frequently overlooked.

What was learned from the study?

COR is a common early postoperative phenomenon occurring after both traditional filtering surgeries and MIGS procedures, with an incidence of 40.7% in the fellow eye on the first day after surgery.

Preoperative use of α₂-receptor agonists in the fellow eye was associated with a lower incidence of COR, indicating it may be a key modifiable factor.

Perioperative IOP monitoring of the fellow eye is equally vital as the operated eye.

Introduction

Consensual ophthalmotonic reaction (COR), first proposed by Weekers in 1924, describes the phenomenon whereby a corresponding pressure change in the fellow eye is induced by the alteration of intraocular pressure (IOP) in one eye [1]. COR following monocular application of medication or laser trabeculoplasty has been documented, demonstrating consensual IOP decrease in the fellow eye [2–6]. However, there are only a few COR-related studies of glaucoma surgery and they fail to demonstrate a consistent association. Published results of research regarding IOP response after unilateral glaucoma surgery, mostly focusing on unilateral trabeculectomy, remain controversial, with studies showing an increase, decrease, or no change in the fellow eye [7–11].

The use of minimally invasive glaucoma surgery (MIGS), a less traumatic surgical alternative for patients with mild to moderate glaucoma or who are intolerant to standard medical therapy, has increased during the last decade [12, 13]. In contrast to trabeculectomy, there is much less information about the effects on the fellow eye after performing MIGS in one eye.

The impact of COR in the fellow eye induced by unilateral glaucoma surgery should not be underestimated. Studies have found that after trabeculectomy in one eye, 13.6–33.8% of the fellow eyes need to receive further glaucoma medication treatment due to increased IOP, and about 10% of fellow eyes require glaucoma surgery, so the fluctuation of IOP caused by glaucoma surgery in one eye may accelerate the progression of the disease [8]. Moreover, glaucoma typically affects both eyes, and asymmetric glaucomatous optic nerve and visual field damage between eyes is common, leading some patients to undergo unilateral glaucoma surgery. The non-surgical fellow eye often exhibits better visual acuity, lower IOP, and milder glaucomatous damage, with overall visual function predominantly determined by this eye. However, perioperative IOP fluctuations in the fellow eye are more likely to be overlooked compared to the surgical eye.

Herein, we systematically investigated IOP change in the fellow eye and identified factors and characteristics in patients with glaucoma related to consensual IOP rise in fellow eyes after unilateral glaucoma surgery.

Methods

Records of patients undergoing unilateral glaucoma surgery including trabeculectomy, bleb needling, trabectome, and microcatheter assisted trabeculectomy (MAT) between January 2016 to November 2022 were reviewed at a tertiary care referral institute in northern China.

The inclusion criteria included (1) According to the Chinese Glaucoma Guidelines (2020), patients diagnosed with primary open-angle glaucoma (POAG), primary angle-closure glaucoma (PACG), juvenile open-angle glaucoma (JOAG), secondary open-angle glaucoma (SOAG), secondary angle-closure glaucoma (SACG), or glaucoma of other undetermined origins [14]; (2) Patients presenting with advanced damage of optic nerve, retinal fiber layer thickness or visual field on maximum tolerated medical therapy (MTMT) scheduled for trabeculectomy, bleb needling, trabectome, and MAT in one eye. The exclusion criteria included (1) Patients with unavailable baseline data such as preoperative IOP or severely missing early postoperative follow-up data; (2) Patients whose postoperative follow-up was less than 1 day; (3) Patients with only one functional eye; (4) Patients with the use of systemic IOP-lowering medications such as acetazolamide or systemic drugs known to affect IOP such as corticosteroids and beta-blockers within 2 weeks prior to surgery; (5) Patients undergoing concurrent bilateral glaucoma surgery. In this study, in order to create homogeneous groups, we included patients with primary or secondary glaucoma regardless of whether the fellow eye was surgically treated or non-surgically treated. To ensure consistency in surgical technique, specific procedure types were performed by designated, experienced surgeons. All trabeculectomies, bleb needling procedures, and trabectome surgeries were conducted by a single senior glaucoma surgeon, while all MAT procedures were performed by another dedicated specialist. This approach aimed to minimize technical variability within each surgical type. Finally, 356 eyes of 293 patients were enrolled in our study.

Demographic information and clinical information at baseline were collected preoperatively including age, sex, and medical history of hypertension and diabetes. Besides, standardized ophthalmic examination of both eyes was performed on the day of surgery indication and IOP, logMAR visual acuity (VA), vertical cup-disc ratio (CDR), lens status, and the types and number of preoperative topical IOP-lowering medications used in both eyes were recorded. Data were collected on all topical agents, including single-agent and fixed-combination formulations. The single-agent drugs comprised four major classes: α₂-receptor agonists, β-blockers, carbonic anhydrase inhibitors (CAIs), and prostaglandin analogues (PGAs). Each single-agent preparation was counted as one agent, whereas each fixed-combination formulation was counted as two agents.

On the basis of studies of COR in the field of glaucoma surgery over the past decade [7–9, 11, 15–17], IOP in the fellow eye frequently fluctuates within the first 6 months following unilateral glaucoma surgery. Therefore, this study focused on IOP in both eyes at postoperative day 1 (POD1), week 1 (POW1), month 1 (POM1), month 3 (POM3), and month 6 (POM6). Fellow eyes were defined as having a phenomenon of COR if they exhibited an IOP change exceeding 4 mmHg or 20% increase from baseline values. This threshold was adopted to distinguish a clinically meaningful IOP elevation from physiological fluctuation, and is consistent with the criterion used in studies of postoperative consensual responses [16]. During the follow-up period, treatment in the fellow eye was coincident with preoperative treatment after glaucoma surgery in one eye. If an intervention such as additional topical medication or glaucoma surgery was needed to control IOP in the fellow eye, it would be recorded and subsequent data was excluded from analysis. We did not exclude these patients in the beginning in order to enrich our data and investigate the outcome of the fellow eye with significantly elevated IOP following unilateral glaucoma surgery.

Given the distinct mechanisms and clinical profiles of the different glaucoma surgeries, all primary analyses were pre-specified to be stratified by surgery type (trabeculectomy, bleb needling, trabectome, and MAT). To evaluate changes of IOP in the fellow eye from baseline at each postoperative time point, a linear mixed-effects model was fitted with a subject-specific random intercept. This model provided estimates, 95% confidence intervals (95% CI), and P values for the comparison of each postoperative time point against baseline. Correlations between the IOP change in the fellow eye and other continuous variables were analyzed using Spearman’s rank correlation. To assess correlation between different variables and IOP changes within the groups, we performed monofactor analysis and Fisher’s exact test was used for categorical variables while Student’s t test was used for continuous variables. Next, variables with P values ≤ 0.1 were entered into a multiple regression analysis and by stepwise regression analysis, a regression model from a set of candidate predictor variables was built. The odds ratio (OR) with 95% CI was calculated within the logistic regression analysis. Prior sample size calculation was not feasible because of the lack of reliable effect size estimates for COR following glaucoma surgery in the existing literature, so post hoc power analyses (pwr package, using observed ORs, event rates, and actual sample sizes) were performed to confirm the robustness of significant multivariate associations within surgical subgroups. All statistical analyses were performed using R, version 3.6.2. P values < 0.05 were considered as statistically significant.

Ethical Approval

Approval was obtained from the ethics committee of Peking University Third Hospital (Approval number IRB00006761-M2021191) and the research adhered to the Declaration of Helsinki. Given the use of de-identified retrospective data, the requirement for written informed consent was waived by the ethics committee. All patient data were kept confidential and used solely for the purposes of this research.

Results

Baseline Data of 356 Operated and Fellow Eyes

A total of 293 patients with mean age 53.4 years including 171 male (58.4%) and 122 female (41.6%) were enrolled in our study. Baseline data of both eyes including baseline logMAR VA, vertical CDR, IOP, number of preoperative IOP-lowering medications, lens status, type of glaucoma, and treatment status are presented in Table 1. The majority of both eyes were phakic (operated eyes, n = 274; fellow eyes, n = 285). Over 60% of the eyes had no history of prior glaucoma surgery (operated eyes, n = 236; fellow eyes, n = 243). Paired comparisons revealed that the fellow eyes had significantly better logMAR VA, smaller vertical CDR, lower baseline IOP, and required fewer classes of topical IOP-lowering medications preoperatively compared to the operated eyes (all P < 0.001), indicating that the fellow eyes demonstrated less severe glaucomatous damage at baseline. To assess the comparability of baseline characteristics across the different surgical procedures, an intergroup comparison was performed (Table S1). While some heterogeneity related to operated eye was observed, the key fellow-eye parameters relevant to COR analysis were broadly balanced across the four surgery groups.

Table 1.

Baseline data of operated and fellow eyes

	Operated eye (n = 356)	Fellow eye (n = 356)	P
logMAR VA	0.7 ± 0.7	0.6 ± 0.8	< 0.001
Vertical CDR	0.8 ± 0.2	0.7 ± 0.2	< 0.001
IOP (mmHg)	27.9 ± 11.2	18.3 ± 7.3	< 0.001
Number of preoperative IOP-lowering medications	2.3 ± 1.0	1.5 ± 1.3	< 0.001
Lens status (%)
Phakic	274 (77.0)	285 (80.1)
Pseudophakic	73 (20.5)	66 (18.5)
Aphakic	6 (1.7)	2 (0.6)
NA	3 (0.8)	3 (0.8)
Previous glaucoma surgery (%)
No	236 (66.3)	243 (68.3)
Yes	90 (25.3)	85 (23.9)
NA	30 (8.4)	28 (7.8)
Diagnosis (%)
POAG	213 (59.8)	214 (60.1)
PACG	69 (19.3)	69 (19.3)
SOAG	49 (13.8)	18 (5.1)
SACG	10 (2.8)	2 (0.6)
JOAG	8 (2.2)	7 (2.0)
Glaucoma of undetermined etiology	7 (2.0)	0 (0.0)
Non-glaucomatous (only fellow eyes)	0 (0.0)	46 (12.9)

VA visual acuity, CDR cup-disc ratio, IOP intraocular pressure, NA not available, POAG primary open-angle glaucoma, PACG primary angle-closure glaucoma, SOAG secondary open-angle glaucoma, SACG secondary angle-closure glaucoma, JOAG juvenile onset primary open-angle glaucoma

IOP Change in Both Eyes Following Unilateral Glaucoma Surgery in All Subjects

Next, we compared preoperative and postoperative IOP in both eyes (Table 2 and Fig. 1). For operated eyes, compared with preoperative IOP, postoperative IOP significantly reduced to 18.7 ± 9.4 mmHg at POD1, 17.9 ± 8.0 mmHg at POW1, 18.1 ± 7.2 mmHg in POM1, 17.4 ± 6.1 mmHg in POM3, and 17.0 ± 5.4 mmHg in POM6 (all P < 0.001). For the fellow eye, postoperative IOP significantly increased to 19.3 ± 7.5 mmHg at POD1 (P = 0.041), however, there was not any statistically significant difference between preoperative IOP and IOP in the fellow eye at POW1, POM1, POM3 and POM6 (P = 0.796, P = 0.891, P = 0.796, P = 0.514, respectively). Besides, the incidence of COR in the fellow eye at different time points was collected and among all the fellow eyes (n = 356), the incidence of COR was 40.7% (145 eyes) at POD1. This incidence progressively decreased over time and persisted in 17.1% of the fellow eyes at POM6, indicating that COR is most likely to occur in the early postoperative period after unilateral glaucoma surgery.

Table 2.

IOP in the operated and the fellow eye at all time points

Time points	Operated eye		Fellow eye
	IOP (mean ± SD) (mmHg)	P	IOP (mean ± SD) (mmHg)	P	Incidence of COR (%)
Baseline (n = 356)	27.9 ± 11.2	–	18.3 ± 7.3	–
POD1 (n = 356)	18.7 ± 9.4	< 0.001	19.3 ± 7.5	0.041	40.7
POW1 (n = 276)	17.9 ± 8.0	< 0.001	18.1 ± 7.0	0.796	23.6
POM1 (n = 187)	18.1 ± 7.2	< 0.001	17.8 ± 6.6	0.891	22.0
POM3 (n = 182)	17.4 ± 6.1	< 0.001	16.9 ± 5.5	0.796	19.7
POM6 (n = 118)	17.0 ± 5.4	< 0.001	16.8 ± 6.0	0.514	17.1

IOP intraocular pressure, SD standard error, COR consensual ophthalmotonic reaction, POD1 postoperative day 1, POW1 postoperative week 1, POM1 postoperative month 1, POM3 postoperative month 3, POM6 postoperative month 6

Fig. 1.

Absolute change and relative change of IOP in both eyes following unilateral glaucoma surgery. IOP intraocular pressure, POD1 postoperative day 1, POW1 postoperative week 1, POM1 postoperative month 1, POM3 postoperative month 3, POM6 postoperative month 6

IOP Change in Both Eyes Following Unilateral Glaucoma Surgery in Four Surgery Groups

On the basis of the type of glaucoma surgery performed, subjects were categorized into four groups for further evaluation of postoperative IOP fluctuations: trabeculectomy group (n = 123), bleb needling group (n = 90), trabectome group (n = 106), and MAT group (n = 37). For the operated eyes, consistent with the overall findings, IOP at all postoperative time points was significantly lower than that at baseline in all four surgery groups (Table 3). For the fellow eyes, a significant IOP elevation compared to baseline was observed only at POD1 following all four surgery groups (Table 3).

Table 3.

IOP change in both eyes following different glaucoma surgery

	Time points	Trabeculectomy	Bleb needling	Trabectome	MAT
Operated eye	POD1	− 12.6 ± 13.2***	− 11.1 ± 12.5***	− 4.3 ± 9.3***	− 7.6 ± 11.6***
	POW1	− 13.8 ± 13.2***	− 11.1 ± 15.6***	− 4.6 ± 9.1***	− 9.0 ± 9.0***
	POM1	− 10.7 ± 13.0***	− 9.0 ± 13.0***	− 5.2 ± 7.2***	− 11.5 ± 9.9***
	POM3	− 10.9 ± 12.9***	− 8.7 ± 12.1**	− 4.6 ± 7.5***	− 8.7 ± 10.1*
	POM6	− 9.8 ± 11.4***	− 8.6 ± 11.1**	− 4.4 ± 8.0***	− 8.0 ± 11.1**
Fellow eye	POD1	0.9 ± 5.9*	0.9 ± 5.1*	1.0 ± 5.0*	1.6 ± 6.3*
	POW1	1.1 ± 6.3	0.2 ± 4.7	0.2 ± 3.5	0.1 ± 6.4
	POM1	1.1 ± 5.8	− 0.6 ± 6.6	0.1 ± 4.6	− 0.6 ± 3.3
	POM3	1.8 ± 3.9*	− 0.1 ± 4.3	− 0.4 ± 3.4	0.2 ± 6.4
	POM6	− 1.2 ± 4.0	− 0.3 ± 4.5	− 0.7 ± 4.1	− 0.2 ± 6.4

IOP intraocular pressure, MAT microcatheter assisted trabeculectomy, POD1 postoperative day 1, POW1 postoperative week 1, POM1 postoperative month 1, POM3 postoperative month 3, POM6 postoperative month 6

*P < 0.05; **P < 0.01; ***P < 0.001

Since the COR phenomenon in the fellow eye primarily occurred at POD1, we focused our analysis on the absolute IOP change (ΔIOP) in both operated and fellow eyes at this specific time point across the different glaucoma surgery groups (Fig. S1). Our findings indicated that trabeculectomy and bleb needling achieved a significantly greater IOP reduction in the operated eye at POD1 compared to trabectome (P < 0.001; Fig. S1a). Although the ΔIOP following MAT did not reach a statistically significant difference compared to the other surgeries (P > 0.05), a strong trend toward superior efficacy over trabectome was observed (P = 0.057). However, no significant difference was found in the magnitude of IOP elevation in the fellow eye at POD1 among the different surgery groups (Fig. S1b).

To further investigate the distribution of ΔIOP in fellow eyes at POD1, we found that over 60% of fellow eyes (n = 216) exhibited a ΔIOP > 0 mmHg, indicating a postoperative IOP increase. Among these, 145 fellow eyes met the study’s criteria for COR, and 12 eyes (3.4%) even showed a ΔIOP exceeding 10 mmHg, demonstrating COR as a common occurrence in the early period after unilateral glaucoma surgery. Although the incidence of COR at POD1 varied among the different surgical groups, the distribution pattern of ΔIOP in the fellow eye was similar across all four surgery groups (Fig. 2). Approximately half of all of the fellow eyes exhibited only minor IOP fluctuations (− 3 mmHg < ΔIOP < 3 mmHg) at POD1.

Fig. 2.

Distribution histograms of ΔIOP in the fellow eye at POD1 in four glaucoma surgery groups: a trabeculectomy, b bleb needling, c trabectome, d MAT. IOP intraocular pressure, MAT microcatheter-assisted trabeculectomy, POD1 postoperative day 1, 95% CI 95% confidence interval

Across the four glaucoma surgery types, approximately 60% of the fellow eyes exhibited a higher IOP on POD1 compared to the preoperative baseline (ΔIOP > 0 mmHg; trabeculectomy, 74/123; bleb needling, 53/90; trabectome, 65/106; MAT, 24/37; Fig. 2). The majority of fellow eyes with elevated IOP showed an increase of less than 10 mmHg, with the highest frequency of ΔIOP observed in the 0 to 5 mmHg range. Furthermore, we identified 14 fellow eyes with a substantial IOP elevation (ΔIOP ≥ 10 mmHg) at POD1. Notably, one fellow eye in the trabeculectomy group exhibited a pronounced elevation exceeding 25 mmHg, indicating that severe COR can occur in a subset of patients during the early postoperative period (Fig. 2).

Influencing Factors Related to COR in the Fellow Eye After Unilateral Glaucoma Surgery at POD1

On the basis of these findings, IOP in the fellow eye demonstrated a significant increase exclusively at POD1, coinciding with the highest incidence of COR observed at POD1. Consequently, we categorized fellow eyes into fellow eyes with COR and fellow eyes without COR based on the occurrence of COR at POD1 and conducted a retrospective analysis of COR in four glaucoma surgery groups. We performed a univariate analysis to identify factors associated with COR in the fellow eye at POD1, guided by previous literature and clinical expertise. Next, factors demonstrating a significant association (P < 0.05) in the univariate analysis were included in a multivariate logistic regression model to identify independent determinants of COR at POD1 across the different surgery groups.

In the trabeculectomy group (n = 123), fellow eyes were categorized into fellow eyes with COR (n = 47) and fellow eyes without COR (n = 76) based on the occurrence of COR at POD1. Univariate analysis indicated a significant association between COR occurrence and the lens status of the fellow eye (P = 0.001; Table 4), and multivariate analysis confirmed that COR tended to occur in phakic fellow eyes (OR 0.16, P = 0.018; Table 5). Besides, COR occurrence was significantly associated with the use of α₂-receptor agonists (P = 0.002; Table 4) but not with other kinds of topical IOP-lowering medications, and multivariate analysis indicated preoperative use of α₂-receptor agonists (OR 0.32, P = 0.016; Table 5) in the fellow eye as an independent protective factor against COR development following trabeculectomy (Table 6).

Table 4.

Univariate analysis of COR in the fellow eye at POD1 following unilateral trabeculectomy

Variables	Fellow eye without COR (n = 76)	Fellow eye with COR (n = 47)	P
Sex
Male	37 (48.7)	24 (51.1)	0.854
Female	39 (51.3)	23 (48.9)
Age (years)	58.5 ± 14.7	58.0 ± 16.0	0.917
Previous history of diabetes
No	62 (81.6)	38 (80.9)	1.000
Yes	14 (18.4)	9 (19.1)
Previous history of hypertension
No	59 (77.6)	37 (78.7)	1.000
Yes	17 (22.4)	10 (21.3)
Baseline IOP in the operated eye (mmHg)	31.7 ± 12.2	32.0 ± 11.6	0.775
ΔIOP in the operated eye at POD1 (mmHg)	− 13.5 ± 13.7	− 11.2 ± 12.4	0.430
Type of glaucoma in the fellow eye
Angle-closure	28 (36.8)	25 (53.2)	0.226
Open-angle	38 (50.0)	17 (36.2)
Healthy	10 (13.2)	5 (10.6)
Lens status in the fellow eye
Phakic	55 (73.3)	45 (95.7)	0.001
Pseudophakia	20 (26.7)	2 (4.3)
Previous glaucoma surgery in the fellow eye
No	55 (72.4)	40 (85.1)	0.124
Yes	21 (27.6)	7 (14.9)
Number of preoperative topical IOP-lowering medications in the fellow eye	1.6 ± 1.4	1.2 ± 1.3	0.129
Preoperative topical IOP-lowering medications in the fellow eye
Use of α2-receptor agonists
No	42 (55.3)	39 (83.0)	0.002
Yes	34 (44.7)	8 (17.0)
Use of beta-blockers
No	51 (67.1)	30 (63.8)	0.845
Yes	25 (32.9)	17 (36.2)
Use of CAIs
No	45 (59.2)	32 (68.1)	0.345
Yes	31 (40.8)	15 (31.9)
Use of PGAs
No	58 (76.3)	34 (72.3)	0.672
Yes	18 (23.7)	13 (27.7)

COR consensual ophthalmotonic reaction, IOP intraocular pressure, POD1 postoperative day 1, CAIs carbonic anhydrase inhibitors, PGAs prostaglandin analogues

Table 5.

Multivariate logistic regression analysis of COR in the fellow eye at POD1 following unilateral trabeculectomy

Variables	β coefficient	SE	OR	95% CI		P
				Lower bound	Upper bound
Use of α2-receptor agonists in the fellow eye (Yes vs No)	− 1.13	0.47	0.32	0.12	0.78	0.016
Lens status in the fellow eye (Pseudophakia vs Phakic)	− 1.86	0.78	0.16	0.02	0.60	0.018

COR consensual ophthalmotonic reaction, POD1 postoperative day 1, SE standard error, OR odds ratio, 95% CI 95% confidence interval

Table 6.

Univariate analysis of COR in the fellow eye at POD1 following unilateral bleb needling

Variables	Fellow eye without COR (n = 52)	Fellow eye with COR (n = 38)	P
Sex
Male	31 (59.6)	20 (52.6)	0.527
Female	21 (40.4)	18 (47.4)
Age (years)	54.9 ± 14.9	54.1 ± 14.1	0.601
Previous history of diabetes
No	51 (98.1)	28 (73.7)	0.001
Yes	1 (1.9)	10 (26.3)
Previous history of hypertension
No	44 (84.6)	28 (73.7)	0.286
Yes	8 (15.4)	10 (26.3)
Baseline IOP in the operated eye (mmHg)	32.0 ± 9.8	31.2 ± 11.8	0.445
ΔIOP in the operated eye at POD1 (mmHg)	− 11.9 ± 10.7	− 9.9 ± 14.7	0.228
Type of glaucoma in the fellow eye
Angle-closure	10 (19.2)	8 (21.1)	0.096
Open-angle	38 (73.1)	21 (55.3)
Healthy	4 (7.7)	9 (23.7)
Lens status in the fellow eye
Phakic	40 (76.9)	28 (73.7)	0.806
Pseudophakia	12 (23.1)	10 (26.3)
Previous glaucoma surgery in the fellow eye
No	21 (55.3)	25 (78.1)
Yes	17 (44.7)	7 (21.9)	0.076
Number of preoperative topical IOP-lowering medications in the fellow eye	1.8 ± 1.4	0.6 ± 1.0	< 0.001
Preoperative topical IOP-lowering medications in the fellow eye
Use of α2-receptor agonists
No	24 (46.2)	36 (94.7)	< 0.001
Yes	28 (53.8)	2 (5.3)
Use of beta-blockers
No	35 (67.3)	34 (89.5)	0.022
Yes	17 (32.7)	4 (10.5)
Use of CAIs
No	32 (61.5)	31 (81.6)	0.062
Yes	20 (38.5)	7 (18.4)
Use of PGAs
No	36 (69.2)	32 (84.2)	0.138
Yes	16 (30.8)	6 (15.8)

COR consensual ophthalmotonic reaction, IOP intraocular pressure, POD1 postoperative day 1, CAIs carbonic anhydrase inhibitors, PGAs prostaglandin analogues

In the bleb needling group (n = 90), based on COR occurrence at POD1, there were 52 fellow eyes without COR and 38 fellow eyes with COR. Univariate analysis revealed that, in contrast to trabeculectomy findings, a history of diabetes demonstrated a significant association with COR occurrence (P = 0.001). Notably, the number of topical IOP-lowering medication classes (P < 0.001), use of α₂-receptor agonists (P < 0.001), and use of β-blockers (P = 0.022) in the fellow eye were significantly associated with COR development, while the use of CAIs and PGAs showed no significant correlation (P > 0.05). Multivariate logistic regression identified preoperative use of α₂-receptor agonists in the fellow eye as an independent protective factor (OR 0.03, P = 0.013), whereas a history of diabetes emerged as an independent risk factor (OR 16.62, P = 0.017) for COR at POD1 following bleb needling (Table 7).

Table 7.

Multivariate logistic regression analysis of COR in the fellow eye at POD1 following bleb needling

Variables	β coefficient	SE	OR	95% CI		P
				Lower bound	Upper bound
Use of α2-receptor agonists in the fellow eye (Yes vs No)	− 3.53	1.43	0.03	0.00	0.33	0.013
Use of beta-blockers in the fellow eye (Yes vs No)	− 1.74	1.21	0.18	0.01	1.60	0.151
Number of preoperative topical IOP-lowering medications in the fellow eye	0.30	0.51	1.35	0.52	4.07	0.557
Previous history of diabetes (Yes vs No)	2.81	1.18	16.62	2.43	361.18	0.017

COR consensual ophthalmotonic reaction, POD1 postoperative day 1, SE standard error, OR odds ratio, 95% CI 95% confidence interval, IOP intraocular pressure

In the trabectome group (n = 106), 39 fellow eyes developed COR at POD1, while 67 fellow eyes did not. The baseline characteristics of the two groups are detailed in Table S2. Univariate analysis identified that COR occurrence was significantly associated with fellow eye-related variables, including the type of glaucoma (P = 0.004), previous glaucoma surgery (P = 0.022), number of preoperative topical IOP-lowering medications (P = 0.032), and the use of β-blockers (P = 0.036) in the fellow eye. In contrast to findings from other surgery types, preoperative use of α₂-receptor agonists in the fellow eye showed no significant correlation with COR occurrence (P > 0.05), suggesting that the protective effect of this α₂-receptor agonists may be surgery-specific. Subsequent multivariate analysis revealed that previous glaucoma surgery of the fellow eye was an independent risk factor for COR (OR 3.65, P = 0.015), whereas a diagnosis of open-angle glaucoma in the fellow eye served as an independent protective factor (OR 0.12, P = 0.011) (Table S3).

In the MAT group (n = 37), fellow eyes were categorized into fellow eyes with COR (n = 21) and fellow eyes without COR (n = 16) based on the occurrence of COR at POD1, with their baseline characteristics detailed in Table S4. Univariate analysis demonstrated no statistically significant associations between COR development and any collected and examined variables, (all P > 0.05; Table 3). This lack of significant associations may be attributable to the limited sample size of the MAT group. Notably, while no fellow eyes with COR had received preoperative α₂-receptor agonists, nearly 20% of fellow eyes without COR had been treated with this medication (P = 0.072), suggesting a potential protective trend of preoperative α₂-receptor agonists use against COR development following MAT (Table S4).

Influencing Factors Related to ΔIOP in the Fellow Eye After Unilateral Glaucoma Surgery at POD1

To further investigate the magnitude and influencing factors of consensual IOP rise in the fellow eye at POD1 when IOP decreases in the operated eye following unilateral glaucoma surgery, we selected cases of which ΔIOP in the operated eye < 0 and ΔIOP in the fellow eye > 0 from our retrospective cohort. This subcohort comprised 61 fellow eyes from the trabeculectomy group, 43 from the bleb needling group, 46 from the trabectome group, and 18 from the MAT group. We analyzed factors influencing ΔIOP in the fellow eye at POD1 across these surgery groups (Table 8) and visualized correlations between continuous variables and ΔIOP in the fellow eye at POD1 (Fig. 3).

Table 8.

Influencing factors related to ΔIOP in the fellow eye after unilateral glaucoma surgery at POD1

Variables	Trabeculectomy (n = 61)		Bleb needling (n = 43)		Trabectome (n = 46)		MAT (n = 18)
	P value	Effect size/correlation	P value	Effect size/correlation	P value	Effect size/correlation	P value	Effect size/correlation
Sex	0.193	0.168	0.697	0.061	0.182	0.198	0.51	0.169
Age	0.809	0.332	0.592	− 0.084	0.999	< 0.001	0.596	0.134
Previous history of diabetes	0.224	0.157	0.357	0.143	1	0.003	0.485	0.117
Previous history of hypertension	0.649	0.059	0.472	0.112	0.258	0.17	0.77	0.092
Baseline IOP in the operated eye	0.675	0.055	0.966	0.007	0.009	0.382	0.255	0.283
ΔIOP in the operated eye at POD1	0.635	0.062	0.324	0.154	0.016	− 0.353	0.425	− 0.2
Type of glaucoma in the fellow eye	0.442	− 0.006	0.988	− 0.067	0.245	0.008	0.811	0.061
Lens status in the fellow eye	0.248	0.006	0.215	0.013	0.126	0.05	0.381	− 0.014
Previous glaucoma surgery in the fellow eye	0.2	0.165	0.816	0.041	0.065	0.287	0.368	0.223
Preoperative topical IOP-lowering medications in the fellow eye	0.228	− 0.157	0.008	− 0.401	0.164	− 0.209	0.482	− 0.177
Use of α2-receptor agonists	0.003	0.375	< 0.001	0.532	0.896	0.021	0.205	0.322
Use of beta-blockers	0.787	0.036	0.894	0.023	0.187	0.196	0.302	0.254
Use of CAIs	0.321	0.128	0.128	0.235	0.628	0.073	0.707	0.101
Use of PGAs	0.955	0.008	0.473	0.112	0.437	0.116	0.243	0.286

IOP intraocular pressure, POD1 postoperative day 1, MAT microcatheter-assisted trabeculectomy, CAIs carbonic anhydrase inhibitors, PGAs prostaglandin analogues

Fig. 3.

Correlations of baseline IOP in the operated eye, ΔIOP in the operated eye at POD1 and number of preoperative topical IOP-lowering medications with the magnitude of IOP elevation (ΔIOP) in the fellow eye at POD1 following different glaucoma surgeries: a–d baseline IOP in the operated eye vs ΔIOP in the fellow eye at POD1; e–h ΔIOP in the operated eye vs ΔIOP in the fellow eye at POD1; i–l number of preoperative topical IOP-lowering medications in the fellow eye vs ΔIOP in the fellow eye at POD1. Surgeries: trabeculectomy (a, e, i), bleb needling (b, f, j), trabectome (c, g, k), and MAT (d, h, l). The shaded area represents the 95% CI. IOP intraocular pressure, MAT microcatheter-assisted trabeculectomy, POD1 postoperative day 1, 95% CI 95% confidence interval

In the trabeculectomy subgroup (n = 61), ΔIOP in the fellow eye at POD1 correlated significantly only with preoperative α₂-receptor agonists use (P = 0.003). Similarly, in the bleb needling (n = 43), ΔIOP in the fellow eye at POD1 showed a significant association with α₂-receptor agonists use (P < 0.001). Furthermore, we identified a significant negative correlation between number of preoperative topical IOP-lowering medications and ΔIOP in the fellow eye at POD1 (Spearman’s r = − 0.401, P = 0.008; Fig. 3), indicating that more preoperative medications were associated with smaller consensual IOP elevations.

In the trabectome subgroup (n = 46), ΔIOP in the fellow eye at POD1 demonstrated a significant positive correlation with baseline IOP in the operated eye (Spearman’s r = 0.353, P = 0.009), and a significant negative correlation with ΔIOP in the operated eye (Spearman’s r = − 0.209, P = 0.016). This indicates that higher preoperative IOP and greater IOP reduction in in the operated eye were associated with larger consensual IOP rise in in the fellow eye (Fig. 3). No significant correlations were found between ΔIOP in the fellow eye and preoperative IOP-lowering medication-related factors in this group (P > 0.05). No significant determinants of ΔIOP in the fellow eye at POD1 were identified in the MAT subgroup (n = 18). However, analogous to the findings in the trabectome subgroup, higher baseline IOP of and greater IOP reduction in the operated eye at POD1 corresponded to a tendency for increased ΔIOP in the fellow eye (Spearman’s r = 0.283, P = 0.255; r = − 0.200, P = 0.425, respectively; Fig. 3). Post hoc power analyses confirmed that the study was adequately powered to detect the significant associations identified in the multivariate logistic regression models for the trabeculectomy, bleb needling, and trabectome groups (all statistical power ≥ 0.80).

In contrast to other topical IOP-lowering medications, preoperative use of α₂-receptor agonists in the fellow eye was associated with prevention of COR phenomenon as described in the previous section. We therefore further summarized and visualized the effect of preoperative α₂-receptor agonists use on ΔIOP in the fellow eye at POD1 (Fig. 4). In the trabeculectomy and bleb needling groups, ΔIOP in the fellow eye at POD1 was significantly associated only with the use of α₂-receptor agonists, suggesting that preoperative administration of α₂-receptor agonists in the fellow eye before traditional filtering surgeries may help avoid excessive IOP elevation, thereby reducing the risk of COR. In contrast, no statistically significant association was found between ΔIOP in the fellow eye and α₂-receptor agonists use following either trabectome or MAT. Nevertheless, a non-significant protective trend was still observed in the MAT group, the lack of statistical significance likely attributable to its limited sample size.

Fig. 4.

Association between ΔIOP in the fellow eye at POD1 and preoperative α₂-receptor agonists in the fellow eye following unilateral glaucoma surgery. The center line represents the median, the box boundaries represent the IQR, and the whiskers extend to 1.5 × IQR. OP intraocular pressure, IQR interquartile range, MAT microcatheter-assisted trabeculectomy, POD1 postoperative day 1

Discussion

Given the limited number their inconsistent findings of existing studies on COR following glaucoma surgery, and the scarcity of research related to MIGS, we conducted a single-center retrospective study to systematically investigate IOP changes in the fellow eye and the occurrence of COR after unilateral glaucoma surgery [7–9, 11, 15–17]. By stratifying our analysis according to surgery type, we were able to demonstrate that the incidence, time course, and risk factors for COR are not uniform but vary meaningfully across different surgery types (trabeculectomy, bleb needling, trabectome, and MAT).

Our findings indicated a significant increase of IOP in the fellow eye in the early postoperative period, particularly at POD1, with an incidence of COR reaching 40.7% (145 fellow eyes) at this time point, underscoring the importance of vigilant monitoring of IOP in the fellow eye after unilateral glaucoma surgery. Our observations of a significant postoperative IOP increase in the fellow eye align with several previous studies. Kaushik et al. observed a significant rise in the fellow eye (peaking at 4.7 ± 3.3 mmHg at 6 weeks) after trabeculectomy or combined surgery [8]. This finding is corroborated by other studies involving trabeculectomy and ExPress shunt implantation, where notable IOP increases in the fellow eye were recorded within the early postoperative months [7, 15]. Further supporting this early consensual rise, Meshksar et al. documented a significant IOP elevation in the fellow eye during the first postoperative week, particularly following Ahmed glaucoma valve implantation [16]. Conversely, the discrepancy between our findings and those of the large prospective Collaborative Initial Glaucoma Treatment Study (CIGTS), which reported no substantial IOP increase in the fellow eye, is likely attributable to methodological differences. Critically, CIGTS had no follow-up data before the 3-month postoperative visit, thereby missing the early window (POD1 to POW1) where we and others have observed the most pronounced consensual response [9]. However, our definition of COR (> 4 mmHg or > 20% IOP increase), consistent with several prior studies, is relatively lenient. This choice directly influences the reported incidence rate and may limit direct comparability with studies employing stricter thresholds.

To our knowledge, this study represents the first systematic evaluation of COR in the fellow eye following unilateral trabectome and MAT procedures. Our results suggested that, similar to traditional filtering surgeries, both trabectome and MAT were associated with notable increase of IOP and incidence of COR in the fellow eye in the early postoperative period, particularly at POD1. Interestingly, the incidence of COR following MAT was even higher than that observed after traditional surgeries, which may be attributed to the abrupt and substantial IOP reduction resulting from its 360° Schlemm’s canal incision [18]. As MIGS adoption grows, early contralateral monitoring becomes increasingly important.

COR should not be considered as a benign physiological phenomenon, as it carries direct and substantial clinical implications. In a study of patients following unilateral glaucoma surgery (n = 131), 18 fellow eyes with consensual IOP elevation required additional glaucoma medication or surgical intervention within 6 months of follow-up, including 10 eyes that were initially defined as “healthy fellow eyes” [16]. Corroborating this finding, another investigation (n = 71) reported a significant IOP increase in the fellow eye from POD1, peaking at 6 weeks and 6 patients already required augmented topical therapy or surgery due to IOP exceeding acceptable limits [8]. Such fluctuations may accelerate glaucomatous progression in the fellow eye, which often determines the patient’s overall visual function. Therefore, vigilant monitoring of IOP in the fellow eye during the early postoperative period is imperative.

COR represents a clinically significant but frequently neglected threat to IOP stability in the fellow eye. Despite its potential harm, effective strategies to prevent and manage COR remain unexplored. Notably, in our cohort, preoperative use of α₂-receptor agonists in the fellow eye was associated with a significant reduction in the consensual IOP rise—suggesting a potential strategy for managing COR. And this observed association may be may be explained by the proposed neural reflex mechanism of COR. The neural circuitry of IOP homeostasis and COR involve distinct afferent nerve, control center, and efferent pathways. On the afferent side, studies by Zuazo [19], Belmonte [20], and Tamm [21] have established that those mechanosensitive receptors within the ciliary nerves (branches of the trigeminal nerve) in the anterior chamber are capable of detecting IOP fluctuations and transducing them into neural signals. Hypothalamus has been considered as the control center of COR reflex [22–24]. The efferent side involves the autonomic innervation of structures critical to aqueous humor dynamics. Specifically, the ciliary muscle, ciliary body, and its associated vasculature, governed by postganglionic fibers originating from superior cervical ganglion (SCG), serve as the final executors of neural regulation [22, 25]. Diestelhorst and Krieglstein, using fluorophotometry, demonstrated a significant increase in aqueous humor flow in the unoperated fellow eye (from 2.56 to 2.92 μl/min, P < 0.05) on the fifth postoperative day after trabeculectomy, proposing that filtration surgery in one eye triggers a CNS-mediated reflective increase in aqueous flow to maintain physiological stability in the anterior chamber of the surgically treated eye [26]. However, whether the signal of IOP reduction could indeed be transmitted from the operated eye to the fellow eye remained unresolved until 2020, when Yang et al. proved the efferent pathway from ipsilateral hypothalamus to the autonomic innervation in the anterior chamber structures bilaterally, providing anatomical evidence for COR [27].

Integrating prior evidence with the associations observed in our clinical data, we propose a model for a neural reflex that might underlie COR (Fig. 5a) and, on the basis of this, hypothesize a potential mechanism through which α₂-receptor agonists might reduce the consensual IOP response in the fellow eye (Fig. 5b). According to this proposed model, the sudden IOP reduction during unilateral glaucoma surgery would be sensed by intraocular mechanoreceptors, potentially triggering an afferent signal that ascends to the ipsilateral hypothalamus [21, 23, 28]. In a compensatory response to maintain anterior chamber homeostasis, this central regulator might issue a command to elevate IOP [26]. The hypothalamus, acting as the regulatory center, initiates a compensatory sympathetic outflow to raise IOP. Since this efferent command is bilaterally distributed, it could engage the contralateral SCG [27]. The subsequent release of norepinephrine (NE) from SCG postganglionic fibers might then mediate the consensual IOP elevation in the fellow eye (Fig. 5a) [22, 25]. In our proposed model, the administration of α₂-receptor agonists to the fellow eye serves as a key intervention. By stimulating presynaptic α₂-receptor, α₂-adrenergic agonists inhibit the release of NE into the synaptic cleft, effectively alleviating COR.

Fig. 5.

Proposed neural reflex model of COR (a) and hypothesized mechanism for the association of α₂-receptor agonists with reduced consensual IOP response (b). COR consensual ophthalmotonic reaction, IOP intraocular pressure, NE norepinephrine, SCG superior cervical ganglion

We also investigated the influence of other factors associated with COR and ΔIOP in the fellow eye. Pseudophakic lens status in the fellow eye was associated with a lower likelihood of COR in the trabeculectomy group, possibly due to the improved outflow anatomy which allows the fellow eye to better accommodate the consensual increase in aqueous flow [29]. Unexpectedly, we found that fellow eyes with open-angle glaucoma were less likely to develop COR after trabectome than healthy eyes, possibly because of their preoperative medication use. Conversely, a history of glaucoma surgery in the fellow eye increased COR risk, a finding consistent with Meshksar et al. in the context of trabeculectomy, potentially reflecting more advanced glaucomatous damage and compromised outflow capacity in these eyes [16]. We also observed that greater IOP reduction in the operated eye was correlated with a larger ΔIOP in the fellow eye, particularly following MIGS procedures such as trabectome and MAT. This supports the hypothesis that the magnitude of IOP change in the operated eye is a key driver of the consensual reflex [26].

​​This study has several limitations that warrant consideration. Firstly, its retrospective, observational design carries inherent risks of selection and information bias. The selection of surgical type was based on clinical indications rather than randomization, leading to heterogeneous patient groups and potential confounding by indication. Although we stratified all analyses by surgery type and standardized surgical techniques within each group to mitigate this, unmeasured confounding factors may persist. Secondly, while our study included a substantial number of patients (356 eyes from 293 patients) and post hoc power analyses suggested adequate statistical power for the main findings, the sample sizes for certain surgery subgroups (MAT, n = 37) might still limit the generalizability of some findings. Therefore, future research should aim for prospective, multicenter studies with larger cohorts to validate these findings. Thirdly, the reported higher COR incidence reflects our chosen definition (> 4 mmHg or > 20% IOP increase), and stricter criteria would likely yield lower rates. Additionally, while the observed associations are consistent with and may lend indirect support to a proposed sympathetic neural reflex mechanism for COR, this study does not provide direct evidence for it. The mechanistic model discussed remains hypothetical and requires validation in future studies. Ultimately, a more comprehensive understanding of COR will facilitate the development of more personalized and effective management strategies for patients with glaucoma undergoing unilateral surgery.

Conclusions

Our study suggests that COR is a prevalent and clinically significant phenomenon in the early period following various types of unilateral glaucoma surgery. Preoperative use of α₂-receptor agonists in the fellow eye may be associated with a lower risk of this consensual IOP rise. These findings underscore the critical importance of vigilant perioperative IOP monitoring in the fellow eye.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Chun Zhang conceived and designed the study. Chun Zhang and Jing Hong provided the resources. Zhen Yang and Weijia Zhang collected the clinical data and performed the analysis. Zhen Yang and Chun Zhang wrote the manuscript. All authors read and approved the submission of this manuscript.

Funding

This study was supported by the Changping Key Research Project of the Beijing Natural Science Foundation Program (No. L234016). The funders had no role in the study design, data collection and analysis, decision on publishing, or preparation of the manuscript. The journal’s Rapid Service fee is funded by the authors’ institution, Beijing Tsinghua Changgung Hospital, Tsinghua University.

Data Availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Conflict of Interest

Zhen Yang, Weijia Zhang, Jing Hong, and Chun Zhang declare that they have no conflicts of interest and no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Ethical Approval

Approval was obtained from the ethics committee of Peking University Third Hospital (Approval number IRB00006761-M2021191) and the research adhered to the Declaration of Helsinki. Given the use of de-identified retrospective data, the requirement for written informed consent was waived by the ethics committee. All patient data were kept confidential and used solely for the purposes of this research.