We like to congratulate Mullany et al. for their paper on normal-tension glaucoma is associated with cognitive impairment.1 To link normal-tension glaucoma (NTG) to cognitive impairment and therefore to a neurodegenerative process opens a new vista and research approach for glaucoma research. NTG indeed is an intriguing optic neuropathy that presents with a glaucomatous optic disc appearance and visual field loss similar to that seen in primary open angle glaucoma. The main risk factor for glaucoma however, increased intraocular pressure is missing.

Unlike other cranial nerves the optic nerve is a white matter tract of the brain, enveloped in the meninges (dura, arachnoid and pia mater) and surrounded by cerebrospinal fluid (CSF) on its entire length. And CSF indeed may be the link that connects the neurodegenerative process leading to cognitive impairment and the glaucomatous optic neuropathy in NTG. Recent research demonstrated a relationship between decreased CSF flow, measured in the ventricles and the spinal cord, and cognitive deficit in the elderly.2 In NTG, impaired CSF dynamics was demonstrated with computer assisted cisternography in the subarachnoid space of the intraorbital optic nerve most pronounced in the bulbar region behind the eye globe.3 In a recent publication we found an elevated L-PGDS concentration in the subarachnoid space of the optic nerve in NTG patients with optic nerve sheath compartment syndrome that results in a reduced CSF turnover.4 Reduced CSF turnover is thought to result in malnutrition of axons, neurons and glial cells as well as in accumulation of toxic waste products on the other hand. An experimental induced optic nerve sheath compartment in an animal model demonstrated the most impressive damage to axons and glial cells not on the site where the compartment was created (close to optic canal), but behind the lamina cribrosa in an area densely packed with mitochondria.5 A toxic damage following the CSF compartmentation seems to be the most likely explanation.
We encourage to focus in further studies on CSF dynamics in both, dementia and NTG.
 
References:
1. Mullany S, Xiao L, Qassim A, Marshall H, Gharahkhani P, MacGregor S, Hassall MM, Siggs OM, Souzeau E, Craig JE. Normal-tension glaucoma is associated with cognitive impairment. Br J Ophthalmol. 2021 Mar 29:bjophthalmol-2020-317461. doi: 10.1136/bjophthalmol-2020-317461. Epub ahead of print. PMID: 33781990.
2. Attier-Zmudka J, Sérot JM, Valluy J, Saffarini M, Macaret AS, Diouf M, Dao S, Douadi Y, Malinowski KP, Balédent O. Decreased Cerebrospinal Fluid Flow Is Associated With Cognitive Deficit in Elderly Patients. Front Aging Neurosci. 2019 Apr 30;11:87. doi: 10.3389/fnagi.2019.00087. PMID: 31114494; PMCID: PMC6502902.
3. Pircher A, Montali M, Wostyn P, Pircher J, Berberat J, Remonda L, Killer HE. Impaired cerebrospinal fluid dynamics along the entire optic nerve in normal-tension glaucoma. Acta Ophthalmol. 2018 Aug;96(5):e562-e569. doi: 10.1111/aos.13647. Epub 2018 Mar 12. PMID: 29532640.
4. Pircher A, Neutzner A, Montali M, Huber A, Scholl HPN, Berberat J, Remonda L, Killer HE. Lipocalin-type Prostaglandin D Synthase Concentration Gradients in the Cerebrospinal Fluid in Normal-tension Glaucoma Patients with Optic Nerve Sheath Compartmentation. Eye Brain. 2021 Apr 14;13:89-97. doi: 10.2147/EB.S297274. PMID: 33883963; PMCID: PMC8053785.
5. Jaggi GP, Harlev M, Ziegler U, Dotan S, Miller NR, Killer HE. Cerebrospinal fluid segregation optic neuropathy: an experimental model and a hypothesis. Br J Ophthalmol. 2010 Aug;94(8):1088-93. doi: 10.1136/bjo.2009.171660. Epub 2010 May 27. PMID: 20508039

We read with interest the article by Sarker et al(1) in which they compared the outcomes of trabeculectomy versus Ahmed glaucoma valve (AGV) implantation in Sturge–Weber syndrome (SWS) patients with secondary glaucoma aged 11-62 years. As it noted in the paper, the authors found that complete success rates after 24 months were 80% and 70% in the AGV and trabeculectomy groups, respectively, and qualified success rates were 90% and 85% at same period in the AGV and trabeculectomy groups, respectively. We were delighted to get the conclusion that both AGV implant and trabeculectomy appeared to be safe and efficacious in controlling glaucoma secondary to SWS.
As it reported by Mohamed et al., the complete success rate and qualified success rate (intraocular pressure≤17mmHg) of trabeculectomy reported were 80% and 100% at 12 postoperative follow-up month, respectively(2). However, the qualified success rate (90%) of AGV implantation in SWS patients with secondary glaucoma is higher than that reported by Hamush et al. (79%)(3) and Kaushik et al. (76%)(4) at 2 years of follow-up. Meanwhile, the trabeculectomy with MMC success rate in this study was comparable to other studies about primary glaucoma(5, 6), but the success rate of tube shunt surgery was higher than in prior reports. The qualified success rate of Baerveldt implantation for patients who not had undergone previous incisional ocular surgery was 73% in Primary Tube Versus Trabeculectomy (PTVT) study(6) and 75% reported by Islamaj et al(5)at 2 years of follow-up. The qualified success rates of AGV implantation and Baerveldt implantation for patients with refractory glaucoma were 76% and 68% at 2 years of follow-up, respectively(7).
This more favorable result of AGV implantation relative to previous reports may because this small sample size study excluded 8 patients (16.7%), enrolled patients may uncomplete 2 years of review, or most of them are older than 18 years old compared with other study about SWS patients with secondary glaucoma(3, 4). The study enrolled eyes may at lower risk of surgical failure than that excluded from the study. As the authors mentioned, a total of 48 patients in glaucoma associated with SWS were surgically treated and 8 patients were excluded because of unreliable follow-ups and/or incomplete case records. Substantial differences in the success rate of cases with and without follow-up may overestimate the success rates of two surgeries and prove misleading on interpreting the results in this small sample size, retrospective study. What’s more, mean±SD follow-up in the study was 23.15±2.36 (range, 15–24) and 22.95±2.87 (range, 13–24) in the AGV and trabeculectomy groups, respectively, which may indicate incomplete 2 years of review for some patients. Patients who experienced successful surgical treatment at 15 months may subsequently experience treatment failure at 24 months. It’s better to supplement the number of patients at each follow up visit.
In conclusion, since the small sample size, it is better to supplement the outcomes of eight patients excluded from the study and the number of patients at each follow up visit so as to yield more convincing results.

Reference

1. Sarker BK, Malek MA, Mannaf SMA, Iftekhar QS, Mahatma M, Sarkar MK, et al. Outcome of trabeculectomy versus Ahmed glaucoma valve implantation in the surgical management of glaucoma in patients with Sturge-Weber syndrome. Br J Ophthalmol. 2021;105(11):1561-5.
2. Mohamed T, Salman A, Elshinawy R. Trabeculectomy with Ologen implant versus mitomycin C in congenital glaucoma secondary to Sturge Weber Syndrome. International journal of ophthalmology. 2018;11(2):251-5.
3. Hamush N, Coleman A, Wilson M. Ahmed glaucoma valve implant for management of glaucoma in Sturge-Weber syndrome. American journal of ophthalmology. 1999;128(6):758-60.
4. Kaushik J, Parihar J, Jain V, Mathur V. Ahmed valve implantation in childhood glaucoma associated with Sturge-Weber syndrome: our experience. Eye (London, England). 2019;33(3):464-8.
5. Islamaj E, Wubbels R, de Waard P. Primary baerveldt versus trabeculectomy study after 5 years of follow-up. Acta ophthalmologica. 2020;98(4):400-7.
6. Gedde SJ, Feuer WJ, Lim KS, Barton K, Goyal S, Ahmed IIK, et al. Treatment Outcomes in the Primary Tube Versus Trabeculectomy Study after 3 Years of Follow-up. Ophthalmology. 2020;127(3):333-45.
7. Christakis P, Zhang D, Budenz D, Barton K, Tsai J, Ahmed I. Five-Year Pooled Data Analysis of the Ahmed Baerveldt Comparison Study and the Ahmed Versus Baerveldt Study. American journal of ophthalmology. 2017;176:118-26.

Title: Management of Glaucoma During Pregnancy

Author: Angelo P. Tanna

Affiliations:
Department of Ophthalmology, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA
Division of Ophthalmology, Ann & Robert H. Lurie Children’s Hospital of Chicago, Chicago, Illinois, USA

Conflicts of Interest Disclosure:
APT: Consultant to Ivantis, Sandoz, and Zeiss

Acknowledgment:
APT is supported by an unrestricted departmental grant from Research to Prevent Blindness, NY, NY

Corresponding Author:
Angelo P. Tanna, M.D.
Department of Ophthalmology
Northwestern University Feinberg School of Medicine
645 N. Michigan Ave., Suite 440
Chicago, IL 60611
Telephone: 312-908-8152
Fax: 312-503-8152
E-mail: atanna@northwestern.edu

Dear Editor:

I read with interest the work of Doctor Hashimoto and colleagues on the risk of adverse neonatal outcomes (congenital anomalies, preterm birth, low birth weight) associated with maternal exposure to intraocular pressure-lowering medications during pregnancy.1 They used a large Japanese claims database and state-of-the-art statistical methodology to evaluate the frequency of adverse events in a cohort of live births of 91 women who had “at least one dispensation of IOP-lowering medications during the first trimester,” compared to that observed in 735 women with glaucoma or suspicion of glaucoma who did not have such an exposure.

The authors discuss the previously used and outdated United States Food and Drug Administration (FDA) risk classification system for drugs used during pregnancy. The FDA required the removal of the pregnancy letter categories – A, B, C, D, and X from all drug product labels in 2015. Instead, for systemically absorbed drugs (which includes all ocular hypotensive medications), the FDA requires labeling to include a summary of the risks of using a drug during pregnancy as well as “risk statements based on data from all relevant sources (human, animal, and/or pharmacologic), that describe, for the drug, the risk of adverse developmental outcomes.”2

The investigators observed any adverse outcome in 17.6% of neonates with and in 13.3% without fetal exposure to IOP-lowering medications. The authors concluded that after propensity score adjustment, IOP-lowering medications were not significantly associated with more frequent adverse events. For example, the adjusted odds ratio for congenital anomalies was 1.43 (95% CI, 0.66 to 3.12).

The investigators only evaluated live births; therefore, the potentially increased risk of spontaneous abortion or fetal demise associated with the use of these agents during pregnancy cannot be known from this analysis. It is possible some of the women in the control cohort may have been exposed to ocular hypotensive agent(s) during the first trimester, using medication already in their possession, without necessarily having been dispensed any such agent during the first trimester. This could confound the comparative analysis.

Finally, the authors report their study had a power > 80% for detecting a two-fold increase in the composite outcome (i.e., the risk of any of the adverse neonatal outcomes studied). This begs the question: How much increased risk is a pregnant woman willing to accept? I believe a much lower threshold is necessary to arrive at a meaningful conclusion. In my experience, many women would reject even a 1% increase in the risk of a congenital anomaly. So then, the concluding message in the abstract that “IOP-lowering medications during the first trimester were not significantly associated with increase in congenital anomalies, preterm birth or low birth weight” is not meaningfully supported by the data. The study only supports the concept that the use of these medications is probably not associated with a doubling of the risk. Patients and society are interested in a higher threshold of safety.

Fortunately, pregnancy is often associated with a spontaneous reduction in intraocular pressure (IOP)4; therefore, continued treatment may not be required. Selective laser trabeculoplasty is also an option to consider for some patients. Moreover, many young patients with glaucoma can tolerate nine months of higher IOP.

Glaucoma in pregnancy is a complex problem that requires complex, collaborative decision-making. Pregnant women, their ophthalmologists and obstetricians must evaluate the potential risks associated with continued use of ocular hypotensive agents during pregnancy and weigh those against the risks of modifying or stopping therapy. I congratulate the authors on exploring this important topic and encourage others to conduct similar studies. Eventually, a meta-analysis may yield evidence that can guide clinical decision-making.

REFERENCES:
1. Hashimoto Y, Michihata N, Yamana H, Shigemi D, Morita K, Matsui H, Yasunaga H, Aihara M. Intraocular pressure-lowering medications during pregnancy and risk of neonatal adverse outcomes: a propensity score analysis using a large database. Br J Ophthalmol. 2021 Oct;105(10):1390-1394. doi: 10.1136/bjophthalmol-2020-316198. Epub 2020 Sep 9. PMID: 32907812.

2. Content and Format of Labeling for Human Prescription Drug and Biological Products;
Requirements for Pregnancy and Lactation Labeling. Department of Health and Human Services. Food and Drug Administration 21 CFR Part 201 [Docket No. FDA-2006-N-0515] RIN 0910-AF11. Available online at http://federalregister.gov/a/2014-28241.

3. Mezawa H, Tomotaki A, Yamamoto-Hanada K, Ishitsuka K, Ayabe T, Konishi M, Saito M, Yang L, Suganuma N, Hirahara F, Nakayama SF, Saito H, Ohya Y. Prevalence of Congenital Anomalies in the Japan Environment and Children's Study. J Epidemiol. 2019 Jul 5;29(7):247-256. doi: 10.2188/jea.JE20180014. Epub 2018 Sep 22. PMID: 30249945; PMCID: PMC6556438.

4. Ziai N, Ory SJ, Khan AR, Brubaker RF. Beta-human chorionic gonadotropin, progesterone, and aqueous dynamics during pregnancy. Arch Ophthalmol. 1994 Jun;112(6):801-6. doi: 10.1001/archopht.1994.01090180099043. PMID: 8002840.

Dear Editor,

We read with interest the study by Silva and colleagues.[1] The authors investigate the prognostic potential of B-scan ultrasonography, visual electrophysiology and perioperative videoendoscopy (VE) for 13 patients undergoing keratoprosthesis (KPro) surgery and identified perioperative intraocular VE as a predictor of functional visual outcome at 1-year follow-up.[1] While we find this study interesting, we would like to caution against the interpretation and over-generalization of the findings reported therein.

Negative predictive value (NPV) was as defined as the number of patients with abnormal VE findings and subsequent unsatisfactory visual acuity over all patients with unfavourable VE. The authors report a NPV of 50% in 10 patients. By contrast, they report a positive predictive value (PPV) of 100% for this test.[1] Although a high PPV, as reported by the authors, is of great importance when deciding which patients are appropriate KPro candidates preoperatively, once the patient is undergoing surgery, we believe identifying patients at highest risk of poor visual outcome using NPV is more clinically relevant. The small sample size of 10 patients with a low prevalence of patients with unsatisfactory post-operative visual acuity, and NPV of 50% are important limitations of this study. From these findings, we are unable to justify VE's clinical benefit to the surgeon and their patient at the time of surgery. This is especially true given the invasive nature, cost and associated complications of this ancillary test. In addition to the risk of elevated intraocular pressure and cataract formation as outlined by the authors, the use of perioperative VE increases the risk of iatrogenic retinal tears, vitreous hemorrhage, and wound leaks,[2] not to mention the risk of retrobulbar anaesthesia,[3] all of which are morbid complications in KPro eyes. Moreover, 23% of the patient cohort (n=3) were excluded from analysis due to technical issues relating to perioperative VE. This highlights the additional challenges this ancillary test may present to the surgeon and their team.

Although prognosticating the visual outcomes of KPro and identifying which patients are at highest risk of KPro failure remain important areas of research and discussion, we believe the data presented by the authors are insufficient to position VE as a predictive perioperative ancillary test and urge readers to consider the associated risks to the patient and medical costs to the healthcare system. B-scan ultrasonography is widely accepted as the gold-standard preoperative imaging modality used to assess the posterior segment in eyes with severe and dense anterior segment opacities.[4] After reading this study, we believe that B-scan ultrasonography remains the method of choice for KPro preoperative evaluations.

References:
1. Silva, L. D. et al. B-scan ultrasound, visual electrophysiology and perioperative videoendoscopy for predicting functional results in keratoprosthesis candidates. Br. J. Ophthalmol. (2020).
2. Nagiel, A. et al. VISUAL AND ANATOMIC OUTCOMES OF PEDIATRIC ENDOSCOPIC VITRECTOMY IN 326 CASES. Retina (2020).
3. Hamilton, R. C. A discourse on the complications of retrobulbar and peribulbar blockade. Can. J. Ophthalmol. (2000).
4. Williamson, S. L. & Cortina, M. S. Boston type 1 keratoprosthesis from patient selection through postoperative management: A review for the keratoprosthetic surgeon. Clinical Ophthalmology (2016).