Recurrent aqueous misdirection following intraluminal ripcord removal from a Paul Glaucoma Implant Authors: Silvio Semeria, Valentina Romani, Alessandro Siligato, Matteo Cacciatori, Giovanni Milano, Giacinto Triolo Published: 2026 ABSTRACT AbstractPurposeTo report a case of aqueous misdirection occurring shortly after intraluminal ripcord removal in a pseudophakic, previously vitrectomized eye implanted with a PAUL Glaucoma Implant.Case descriptionA 65-year-old Caucasian woman presented with primary angle-closure glaucoma and previous history of aqueous misdirection following a combined procedure of phacoemulsification and XEN45 gel stent implant, which was managed with pars plana vitrectomy elsewhere. Because of high IOP despite maximum tolerated medical treatment, a PAUL Glaucoma Implant was performed with positioning of a 6-0 Prolene® intraluminal ripcord. Shortly after ripcord removal 6 weeks postoperatively, she presented with significant myopic shift, uniformly shallow anterior chamber, and markedly raised IOP. Anterior-segment OCT showed patent tube in the superotemporal quadrant of the anterior chamber, suggesting the recurrence of aqueous misdirection. Revision of pars plana vitrectomy with extensive peripheral vitreous shaving and irido-zonulo-hyaloidectomy was required to restore anterior chamber depth, to re-establish forward aqueous flow, and to stabilize IOP.Conclusions and importanceThis case suggests that intraluminal ripcord removal from a PAUL glaucoma implant can elicit aqueous misdirection in eyes at risk. Furthermore, the final outcomes of the present case suggest that thorough pars plana vitrectomy with irido-zonulo-hyaloidectomy may be the most effective surgical approach to aqueous misdirection.Keywords: Aqueous misdirection, Malignant glaucoma, Paul glaucoma Implant, Ripcord removal, Primary angle-closure glaucoma Introduction Aqueous misdirection (AM) is an uncommon cause of secondary angle closure, characterized by myopic shift, elevated intraocular pressure (IOP), uniform shallowing of both central and peripheral anterior chamber (AC), in the absence of pupillary block, and usually elicited by ophthalmic surgical procedures.1,2 Classically, AM occurs in the early postoperative period following glaucoma surgery, with a reported incidence ranging from 0.6% to 4%,3,4 but it has also been described after penetrating keratoplasty, cataract surgery, and laser peripheral iridotomy.4,5 Recognized predisposing factors include short axial length, nanophthalmos, and primary angle closure (PAC) spectrum with peripheral anterior synechiae (PAS).1, 2, 3, 4, 5. Although several pathogenic mechanisms have been proposed, the precise pathophysiology of malignant glaucoma/aqueous misdirection remains incompletely understood. A commonly proposed mechanism involves an abnormal anatomical–functional relationship among the ciliary processes, lens, anterior hyaloid face and vitreous, leading to posterior diversion or impaired anterior movement of aqueous humour, a posterior-to-anterior pressure gradient, anterior displacement of the lens–iris diaphragm, secondary angle closure and sustained IOP elevation.6 Medical treatment includes cycloplegics and AH suppressants, along with vitreous dehydration with hyperosmotic agents. However, medical therapy alone may be insufficient in approximately 50% of eyes.7 In cases that are refractory to pharmacologic treatment, Nd:YAG laser disruption of the anterior hyaloid face through a pre-existing iridectomy/iridotomy may be helpful to re-establish the physiologic aqueous flow. When medical and/or laser therapy fail, definitive treatment is surgical by means of pars plana vitrectomy (PPV) with irido-zonulo-hyaloidectomy (IZH), with the goal of establishing a durable anterior–posterior communication and converting the eye into a unicameral system.8 Case presentation A 65-year-old Caucasian female patient attended the Glaucoma Clinic of Fondazione “Istituto di Ricovero e Cura a Carattere Scientifico” (IRCCS) Policlinico San Matteo, Pavia, Italy, with a diagnosis of bilateral Primary Angle Closure Glaucoma (PACG). Her previous ocular history included: trabeculectomy in the right eye nearly 40 years earlier, which led to significant post-operative complications (not well documented) and absence of residual vision. In the left only eye the patient underwent phacoemulsification with intraocular lens (IOL) implant combined with ab interno implantation of Xen 45 Gel Stent (AbbVie, Irvine, CA, USA) augmented with mitomycin C (MMC) approximately two years earlier. Shortly after this procedure, the patient developed AM characterized by myopic shift, AC shallowing, and raised IOP, for which a 25-Gauge (25G) PPV with peripheral iridectomy and AC reformation was performed elsewhere. On our first clinical examination, the best corrected visual acuity (BCVA) was no light perception in the right eye and 20/25 Snellen with −2.50 spherical equivalent (S.E.) in the left eye. The right eye presented with flat filtering bleb, corneal edema, shallow AC, superior peripheral iridectomy, mydriatic non reacting pupil, dense nuclear cataract with 360° posterior synechiae. The fundus was not visible, but the B-scan ultrasound showed unremarkable vitreous cavity, flat retina and choroid with significant optic disc cupping. The left eye presented with flat supero-nasal bleb over the Xen 45, clear cornea, centrally deep AC with some degree of shallowing in the periphery, short intra-cameral portion of Xen 45 at 10 o'clock hour, patent supero-nasal peripheral iridectomy, regular pupil, and IOL in the bag. On fundus examination, the left eye showed moderate optic disc cupping, with no obvious neuroretinal rim defects. The macula and retinal periphery were unremarkable. On Goldmann Applanation Tonometry (GAT) the IOP was 12 mmHg in the right eye without topical IOP-lowering medications, and 28 mmHg in the left eye despite maximum tolerated topical medical treatment (MTMT), consisting of dorzolamide-timolol fixed combination b.d., latanoprost 0.003% once nightly, brimonidine 0.2% b.d. On gonioscopy, the left eye showed synechial angle closure with posterior trabecular meshwork only visible from 12 to 2 o'clock hours and from 6 to 8 o'clock hours. Left peripapillary RNFL thickness, measured using Spectralis OCT (Heidelberg Engineering, Heidelberg, Germany), and 24-2 SITA standard Humphrey Visual Field test were within normal limits. The IOLMaster 700 (Carl Zeiss Meditec AG, Jena, Germany) on the left eye revealed a short axial length (20.92 mm), with a central AC depth of 2.94 mm, and central corneal thickness of 555 μm. The patient was therefore diagnosed with pseudophakic PAC, with aqueous misdirection component that was managed with 25G PPV and iridectomy, although it was not clear if a thorough peripheral vitrectomy with shaving of the vitreous base and IZH were performed. Given the high IOP despite MTMT and the previous failure of bleb-forming surgery with XEN gel stent implantation a decision was made to list the patient for left PAUL® Glaucoma Implant (PGI, Advanced Ophthalmic Innovations, Singapore) with the plan to insert the tube into the AC between 12 and 2 o'clock where the angle structures were visible. One month later, PGI implantation in the left eye was performed under local anaesthesia with sedation. Mitomycin C (0.4 mg/mL) was applied for 5 minutes using 2 Merocel® eye spears. To mitigate the risk of early postoperative hypotony, a 6-0 polypropylene (Prolene®; Ethicon, Inc., Somerville, NJ, USA) intraluminal ripcord was inserted within the tube. Furthermore, due to brisk aqueous outflow visible at the PGI plate intraoperatively, the tube was ligated with one 8-0 polyglactin 910 suture (Vicryl®, Ethicon, Inc., Somerville, NJ, USA), until only mild aqueous outflow was visible. On postoperative day 1, slit lamp examination confirmed correct tube position in the AC without endothelial or iris contact, while the IOP was 30 mmHg on GAT. Along with standard post-operative treatment protocol (dexamethasone 1.5% eye drops 6 times daily, ofloxacin eye drops 4 times daily), the patient was started on atropine 1% t.i.d., topical fixed association of dorzolamide and timolol b.i.d., and oral acetazolamide 250 mg, half tablet twice daily. At week 1, week 3, and week 4 postoperatively, the IOP remained between 24 and 30 mmHg, despite progressively increasing the dosage of oral acetazolamide. At each of these follow-up visits, the anterior chamber was normally deep, the tube was well positioned, and both the retina and choroid were flat. Six weeks after surgery, and when the Vicryl® ligation suture should have supposedly dissolved, the IOP was 28 mmHg despite dorzolamide/timolol fixed combination b.i.d., and oral acetazolamide 250 mg, 1 tablet three times daily. Therefore, the 6-0 Prolene® intraluminal ripcord was successfully removed at the slit lamp, with an immediate IOP drop to 5 mmHg, without clinical signs of hypotony (shallowing of AC, serous choroidal effusion, or hypotony maculopathy). Three days after the ripcord removal, the patient attended the ophthalmic emergency department complaining of blurred vision, redness, and discomfort in the left eye. BCVA was 20/32 Snellen with −5.50 S E., slit lamp examination showed diffuse conjunctival hyperaemia, corneal haze, diffusely shallow AC with peripheral irido-endothelial contact, the tube was merely visible as peripheral and embedded in the iris, and the IOP was 51 mmHg on GAT. The fundus examination was unremarkable. OCT of the anterior segment (AS-OCT, MS-39; CSO Srl, Florence, Italy) confirmed central AC shallowing with diffuse peripheral irido-corneal contact, while the PGI tube had a free tip in the AC, in contact with corneal endothelium and anterior iris surface (Fig. 1). Given the sudden myopic shift, diffuse AC shallowing despite significantly raised IOP, unremarkable fundus findings, and AS-OCT demonstrating that the tube was not obstructed nor retracted from the AC, recurrent AM was diagnosed. Anterior segment optical coherence tomography (AS-OCT) (MS-39; CSO Srl, Florence, Italy) three days after ripcord removal. It shows a diffusely shallow anterior chamber with peripheral iridoendothelial contact. The PAUL Glaucoma Implant tube is identifiable in the anterior chamber with a free ostium indicated by the asterisk (*). Intravenous mannitol was administered, and intensive cycloplegia (atropine 1% eye drops five times daily), brimonidine 0.2% b.d., dorzolamide/timolol fixed combination b.d., and oral acetazolamide 250 mg three times daily were prescribed. Topical dexamethasone 1.5% q.i.d. was also maintained. On the following day, and one week later, BCVA was 20/25 Snellen with −4 S E., the cornea was clear, the AC deepened slightly, the IOP decreased to 12 mmHg, and some irido-IOL synechiae started forming at the inferior pupil margin probably due to the anterior displacement of the iris-capsular bag-IOL diaphragm. As soon as the aforementioned treatment was tapered down during the following weeks, the myopic shift and IOP increased accordingly. Therefore, two weeks after the onset of AM, a revision 25G PPV was performed to remove the remaining peripheral vitreous gel through meticulous vitreous base shaving and to complete the pre-existing iridectomy by performing an IZH: using the vitrector from a pars plana approach, a limited sector of zonular fibers and the adjacent anterior hyaloid face corresponding to the iridectomy site were excised, thereby creating a direct communication between the anterior chamber and the vitreous cavity. Five days after revision surgery, BCVA was 20/25 Snellen with −0.75 S E., and slit lamp examination showed deep AC (Fig. 2A) and patent supero-nasal IZH (Fig. 2B). The tube appeared short in the AC, but its correct position was confirmed on both gonioscopy (Fig. 2C) and AS-OCT (Fig. 3). IOP was 25 mmHg on GAT, without IOP-lowering medications, and topical dorzolamide/timolol fixed combination b.i.d. was added. Slit-lamp examination five days after revision of 25G pars plana vitrectomy with irido-zonulo-hyaloidectomy. (A, Left) Anterior and posterior slit-lamp beams demonstrate a deep anterior chamber. (B, Middle) Retroillumination shows a patent full-thickness superonasal irido-zonulo-hyaloidectomy (white arrow). (C, Right) Gonioscopy shows the tube embedded by a residual peripheral anterior synechia (white asterisk *), with a patent lumen (black asterisk *). Anterior segment optical coherence tomography (AS-OCT) (MS-39; CSO Srl, Florence, Italy) five days after revision of 25-Gauge Pars Plana Vitrectomy with irido-zonulo-hyaloidectomy. (A, Left) Deep anterior chamber. (B, Right) Correct positioning of the PAUL Glaucoma Implant tube within the anterior chamber (asterisk *). At the last follow-up visit, two months after the 25G PPV, which confirmed BCVA of 20/25 Snellen with −0.75 S E., deep and quiet AC, PGI tube in the AC, and an IOP of 12 mmHg on dorzolamide/timolol fixed combination twice daily (Fig. 4). Slit-lamp examination at the last follow-up visit, two months after the 25G pars plana vitrectomy. (A, Left) Anterior and posterior slit-lamp beams confirm a deep anterior chamber. (B, Middle) Retroillumination shows a patent full-thickness superonasal irido-zonulo-hyaloidectomy (white arrow). (C, Right) Close-up view of the PAUL Glaucoma Implant tube in the anterior chamber periphery (asterisk *). Discussion AM is a rare but potentially vision-threatening complication that may occur after glaucoma surgery. It has been described following trabeculectomy, glaucoma drainage device implantation, and other intraocular procedures including phacoemulsification, in eyes with predisposing factors, such as short axial length, nanophthalmos, PAC/PACG, and previous history of AM.4, 5, 6, 7, 8. More specifically, in the setting of glaucoma drainage devices, AM has been reported in approximately 3–4% of cases, although most available data refer to Baerveldt glaucoma implants.9,10 In this context, the reported median time to diagnosis is approximately 33 days after implantation, again based on data from Baerveldt glaucoma drainage device surgery.11 We presented the case of AM occurring shortly after intraluminal ripcord removal from a PGI, indicating a strong temporal relationship with this slit lamp procedure. This may represent a novel clinical presentation of an unusual complication after ripcord removal. Notably, the patient had several risk factors for AM, including short axial length, PAC, and previous history of AM in the same eye, which required surgical management with PPV nearly 2 years before the PGI surgery. Although the risk of recurrent AM was deemed low due to the previous vitrectomy, the PGI was not only stented with a 6-0 Prolene ripcord suture, as per our standard practice, but also ligated with one 8-0 Vicryl suture to minimize the risk of early hypotony, which could potentially trigger AM recurrence. However, this additional precaution may have contributed to the high IOP values observed during the first 6 postoperative weeks and to the need for early ripcord removal. Intraluminal ripcord removal from the PGI has been shown to provide a significant additional IOP-lowering effect; nevertheless, recent real-world data suggest that removal is generally avoided before 8–12 weeks postoperatively, particularly in eyes receiving mitomycin C, unless IOP remains excessively high despite maximal tolerated medical therapy.12 Interestingly, while highly effective in reducing IOP, the early ripcord removal may have produced an abrupt IOP drop within the AC that ultimately triggered the AM mechanisms. Perhaps, a gradual IOP lowering should have been attempted by retracting the ripcord partially, in order to mitigate the risk of AM. Noticeably, our patient developed recurrent AM despite a previous vitrectomy. While thorough PPV with IZH should in theory prevent the risk of AM recurrence, limited core vitrectomy without IZH may not be sufficient to interrupt AM mechanisms, with a reported rate of relapsing AM as high as 75%.7 In fact, in the absence of accurate peripheral vitreous shaving and properly sized peripheral IZH, the surgical procedure may fail to eliminate the anterior hyaloid–zonular barrier and to establish a stable communication between the vitreous cavity and the AC, which would allow restoring forward aqueous flow and converting the eye into a functional unicameral system.13 Unfortunately, in the present case, it was challenging to understand whether peripheral vitrectomy and IZH (instead of standard surgical iridectomy) were previously performed elsewhere, and were therefore only presumed. Although an anterior approach to perform IZH from the AC through a pre-existing peripheral iridectomy in pseudophakic eyes was reported to be effective in resolving AM,14,15 we opted for a more radical approach with revision of PPV, which allows to perform extensive peripheral vitrectomy and IZH, maximizing the chances of successful AM treatment.7 Indeed, our patient benefited from this approach, as suggested by restoration of AC depth, resolution of significant myopic shift, and achievement of target IOP with the aid of 2 IOP-lowering molecules. Conclusion This case highlights the possibility of relapsing AM after ripcord removal from a PGI in a pseudophakic eye with several predisposing factors and a previous history of PPV. This case suggests that, similarly to ‘major’ surgical intervention—such as glaucoma drainage device implantation—even delayed post-surgical ‘minor’ procedures—such as intraluminal ripcord removal—may precipitate AM in susceptible eyes. Awareness of this potential complication is important, and careful postoperative surveillance should be considered after any intervention that abruptly modifies aqueous outflow, particularly in anatomically predisposed eyes with advanced angle-closure disease and complex surgical histories.6 Furthermore, the present case report emphasizes the importance of extensive peripheral PPV with full-thickness IZH to maximize the efficacy of the surgical approach to resolve AM, while minimizing the risks of recurrence. Patient consent Written consent to publish this case was not obtained, as no personal identifying information was included. Declaration of generative AI and AI-assisted technologies in the manuscript preparation process Artificial intelligence–assisted software, specifically ChatGPT (OpenAI), was used only for spelling correction and grammar checking. The authors reviewed and approved all AI-assisted edits and take full responsibility for the content of the manuscript. Funding This report did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. CRediT authorship contribution statement Silvio Semeria: Conceptualization, Writing – original draft. Valentina Romani: Conceptualization, Writing – original draft. Alessandro Siligato: Conceptualization, Writing – original draft. Matteo Cacciatori: Supervision. Giovanni Milano: Supervision. Giacinto Triolo: Conceptualization, Supervision, Writing – review & editing. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.