PEI-GSL is becoming an increasingly popular treatment for PACG [11]. It has a good long-term effect of lowering IOP, even in PACG patients with failed trabeculectomy [12, 13]. In this study, the range of PAS in the included patients was 241.12 ± 100.57°, and the mean IOP was 28.6 ± 15.4 mmHg combined with three antiglaucoma medications. The overall success rate was 89.7%, and the qualified success rate was 95.12% at the last follow-up (mean 19.9 months,6–46 months).
There is no widely accepted standard for the definition of the IOP spike. A common definition of an IOP spike is the increase in IOP over 30 mmHg in the early postoperative period. Less frequent definitions include the incidence of IOP of over 28 mmHg [14] or 40 mmHg [15]. Some authors prefer to use a relative value, for example, a 50% increase in IOP [16] or an increase by 10 mmHg [17]. The IOP spike typically peaks at 3–7 h after phacoemulsification (PEI) and persists during the first 24 h. The most hazardous time for glaucoma patients is at 3–4 h postoperatively, when most of the eyes have an elevated IOP level of over 30 mmHg [14, 18]. To detect early postoperative elevated IOP, we routinely examined IOP at 2 h after surgery and defined the early IOP spike as IOP ≥ 21 mmHg at 2 h after PEI-GSL.
In our study, 32 eyes (32%, 32/100) showed an early IOP spike, including 11 eyes (11/32, 34.3%) that had IOP levels of ≥ 30 mmHg. Ahmed et al. [14] reported a significant increase in the mean IOP levels from baseline to 3 to 7 h postoperatively in both glaucomatous (10.2 mmHg) and non-glaucomatous (4.1 mmHg) eyes, with a significant number of patients having an IOP of at least 28 mmHg in the glaucomatous group (46.4%) and non-glaucomatous group (18.4%), Particularly worrisome is the finding that 18.8% of the patients with glaucoma and 3.6% without glaucoma had an IOP of above 40 mmHg. The incidence of IOP spikes was higher than our result, which may be related to the selection of cases and the mode of operation. Those cases did not reveal the type of glaucoma, and the surgical method was PEI. Our study only included PACG and PEI-GSL as the interventional procedure.
Our findings emphasize the need for vigilance in monitoring and treating postoperative IOP, especially in patients with PACG. In patients with an IOP of ≥ 30 mmHg at 2 h after surgery, anterior chamber fluid was released through the lateral incision an average of 2.82 times, which rapidly reduced the IOP and protected the further damage of optic nerves in glaucoma. IOP levels of > 21 mmHg were observed in 19% of the patients at POD1 and 4% at POD7. A meta-analysis [19] demonstrated that glaucoma was a significant risk factor for retinal vein occlusions, and there was a plausible relationship between PACG and risk of RVO (OR: 1.85; 95% CI: 0.41–8.35). Moreover, the early postoperative elevation of intraocular pressure in patients with glaucoma may lead to optic neuropathy and visual field progression. The mean deviation of VF was significantly aggravated only in the PACG group (from − 7.26 to − 8.82 dB, P < 0.001) [20]. Nowadays, day surgery for glaucoma care is trending, although the high rate of IOP spike and the potential harm may deter the application of day surgery for glaucoma patients.
The overall mechanism of the IOP spike was not clear. It was believed that PEI postoperative IOP spike of PACG was also previously seen in penetrating canaloplasty. About 40% of the patients in PACG showed an IOP spike (≥ 21 mmHg) after penetrating Schlemm canaloplasty, and the IOP dropped below 21 mmHg within 7 days to 3 months after surgery.[21] In this study, IOP decreased within 7 days after PEI-GSL. This rapid decrease of IOP may be attributed to the normal function of distal outflow pathways, such as the trabecular network, Schlemm’s canal, collector tube, and aqueous humor vein in PACG. The removal of the lens can eliminate the pupil block, substantially deepen the anterior chamber, widen the chamber angle, or separate the PAS with GSL as much as possible to resume the drainage function of the anterior chamber angle. The incidence of PACG patients with IOP spike that underwent penetrating Schlemm canaloplasty was higher, and the duration was longer than those with PEI-GSL. This difference may be attributed to the fact that penetrating Schlemm with an iTrack-illuminated microcatheter may somewhat destroy the microstructure of the Schlemm canal and affect the restoration of aqueous humor drainage function. PEI-GSL only exposes the trabecular network and removes the mechanical obstruction of the trabecular network, which may not destroy the microstructure of the Schlemm’s canal. Ahmed et al. [14] reported IOP of > 28 mmHg in 18% of the non-glaucoma patients in the early (3–7 h) postoperative PEI period, which decreased to baseline level within 4 days in most individuals. Our results agree with the study by Ahmed et al. [14], although our study defined the standard IOP spike as ≥ 21 mmHg, which was lower than their results. In our study, the IOP spike at 2 h post-operation was not associated with the intraoperative residual PAS, length of operation, postoperative cornea edema, fibrinous exudate, and postoperative hyphemia.
Previous studies have demonstrated that the postoperative IOP can be well-controlled by phacoemulsification and IOL implantation alone for patients with angle-closure glaucoma, accompanied by PAS of < 180°, whereas patients with excessive chamber angle adhesion GSL should be considered to separate these adhesions [22]. However, we believe that we should also routinely combine phacoemulsification with GSL for patients with PAS of < 180° to separate the PAS as much as possible, protect the drainage function of the angle, and avoid the progression of synechia, which may ultimately increase the IOP.
The risk factors for IOP spikes following PEI include residual viscoelastic material [23, 24], resident-performed surgery [17, 25, 26], glaucoma [14, 16], exfoliation syndrome [27], an axial length of more than 25 mm [28], and topical steroid application in steroid responders [29]. However, a few studies have reported the associations of IOP spike after PEI-GSL in PACG. In our study, patients with IOP spike had severe visual field defects compared to those without IOP spike. Also, IOP spike was more often observed in CPACG (44%) than in APACG (25%). This may indicate that the PAS was formed for a longer time in CPACG and the peripheral iris tissue adhered more closely to the trabecular meshwork. It was observed that the separation of PAS in CPACG was more difficult than that in APACG, and there was more pigmentation of the iris attached to the trabecular meshwork of CPACG. However, the situation of chamber angle pigment and angle-separation time of the two groups were not recorded, and the degree-wise classification of trabecular meshwork pigment and separation time of GSL in different types of PACG was not analyzed. In the multifactorial regression model, pre-PAS, ACD, and PSD of VF were not correlated with IOP spike.
The mean follow-up period was 19.7 ± 38.5 months (range 6 to 46 months). The overall success rate was 89.74%, and the qualified success rate was 95.12%. The mean IOP was 14.48 ± 3.66 mmHg, and the number of mean antiglaucoma medications was 0.61 ± 1.10 at the last follow-up. Four eyes (4/100, 4.00%) needed to undergo anti-glaucoma surgery again. This finding was consistent with the study by Kameda et al.[30], who reported that the probability of treatment success for all 109 eyes was 85.9%, with a mean follow-up of 40 months after Phaco-GSL, while the study by Teekhasaenee et al.[5] reported that out of the 52 eyes that underwent Phaco-GSL, 47 eyes (90.4%) and four eyes (7.69%)had an IOP of > 20 mmHg without and with the use of medications, respectively, with a follow-up of 20.8 months. A previous study [31] has demonstrated that the trabecular meshwork may be irreversibly damaged, and even the PAS is reopened to expose the trabecular network. These functions cannot be restored in PACG patients who have experienced long-term angle closure. Compared to the rate of success of conventional trabeculectomy, we believe that GSL is worth trying for patients with long-term angle closure and exposure to the functional trabecular network is vital.
There was no significant difference in the incidence of IOP of ≥ 21 mmHg between the IOP spike group and the non-IOP spike group at final follow-up. This result indicates that the early IOP spike was not related to the long-term success rate of IOP control.
In our study, the average residual PAS was 37.73 ± 54.99° at the end of the operation. However, there are no sufficient animal or human studies to determine the extent of recovery of the trabecular meshwork function after reopening the PAS, and the separation of PAS during operation depends entirely on the subjective judgment of the surgeon.