Drug-induced Bilateral Secondary Angle-closure Glaucoma

A Literature Synthesis

Rory M. Murphy, MD; Belal Bakir, MD; Colm O'Brien, MD; Janey L. Wiggs, MD, PhD; Louis R. Pasquale, MD


J Glaucoma. 2016;25(2):e99-e105. 

In This Article


An Increase in Recognition of the Drugs That can Cause Bilateral 2° ACG

This study is the first to systematically examine the full spectrum of drugs implicated in bilateral 2° ACG. We used a novel combination of the Naranjo scale and a bilateral 2° ACG score to assess the validity of association between these drugs and bilateral 2° ACG. Many of the drugs highlighted in the review were reported after year 2000, indicating an increased awareness of this condition. Although the number of topiramate case reports is larger than those of all other drugs combined, it is important to note that there are a number of other commonly used drugs implicated in bilateral 2° ACG (Table 2).

Pathophysiological Mechanism

The pathophysiological mechanism responsible for drug-induced bilateral 2° ACG remains unknown. Sen et al[55] suggested an osmotic disturbance in the lens with subsequent lens edema contributed to anterior chamber shallowing. Using ultrasound, Craig et al[57] measured the lens thickness in 2 bilateral 2° ACG cases and concluded that lens edema could only account for 9% to 16% of the observed anterior chamber depth decrease. They concluded the chamber shallowing was likely due to ciliochoroidal effusion, a hypothesis that has been supported by others.[54] The drug-induced ciliochoroidal effusion and ciliary body swelling leads to an anterior rotation of the ciliary process, and a forward displacement of the lens-iris diaphragm causing bilateral 2° ACG. As the lens-iris diaphragm is displaced forward, zonular tension decreases and the lens thickens. Both the forward anterior movement of the lens-iris diaphragm and lens thickening due to reduced zonular tension could contribute to the induced myopia.[56]

How the implicated drugs produce choroidal swelling and bilateral 2° ACG is currently unknown. One theory is that these drugs bind to choroidal tissue and act like a foreign antigen to incite an immune reaction; however, root chemical analysis finds no common epitope that could induce such a reaction. It is noteworthy that many drugs with probable Naranjo and high 2° ACG scores are involved in serotonin and dopamine metabolism. Several drugs (Table 2) implicated in bilateral 2° ACG directly or indirectly act to enhance serum serotonin levels, including topiramate.[68] Serotonin is present in aqueous humor[69] and receptors for serotonin are present in the ciliary body[70] and choroid.[71] Perhaps altered serotonin metabolism after drug exposure contributes to uveal effusion after drug exposure and raises the possibility that rare genetic variants involved in serotonin metabolism represents pharmacogenetic mechanisms that contribute to drug-induced bilateral 2° ACG.


Drug-induced bilateral 2° ACG is largely believed to be idiosyncratic in nature. As Abtahi et al[10] astutely pointed out in their review concerning topiramate there are 3 supportive reasons. First, a very small proportion of the population of patients who took topiramate went on to develop drug-induced bilateral 2° ACG. Second, low or even subtherapeutic levels of the drug could cause the condition. Third, there are instances of topiramate drug overdose that did not result in bilateral 2° ACG development. In fact, with the exception of the acetylsalicylic acid case[13] and the recreational drug use cases,[36,37] all cases of bilateral 2° ACG studied here involved use of the recommended doses of the drug in question. Thus one can continue to expect isolated cases of drug-induced bilateral 2° ACG that have high Naranjo and bilateral 2° ACG scores but are so rare that they are not reproduced by other investigators.

The Fine Line Between 1° ACG and 2° ACG

While relative pupillary block in anatomically predisposed eyes contributes to 1° ACG and diffuse uveal tract swelling with forward rotation of the iris-lens diaphragm contributes to 2° ACG, there is some overlap of these features. Arora et al[72] have pointed out that choroidal expansion may play a role in 1° ACG and Sakai et al[73] reported that 58% of eyes with acute 1° ACG had uveal effusion demonstrated by ultrasound biomicroscopy. Uveal effusion in 1° ACG both before and after laser iridotomy was confirmed by researchers from Singapore.[74] However, the uveal effusion associated with 1° ACG is generally smaller than that seen in 2° ACG, does not extend posteriorly, and typically is not sufficient to produce a myopic shift. None of the drugs listed as likely candidates for bilateral 2° ACG here (Table 2) have been implicated in 1° ACG to the best of our knowledge, but it would not be surprising to find future cases where the distinction between 1° ACG and 2° ACG is not so obvious.

Limitations of the Review

There are several limitations of this study that are related to the choice of the scoring systems used for the analysis. While the Naranjo Algorithm is useful in establishing whether a real cause-and-effect is present between drug exposure and reported side effect, it is perhaps more suited to adverse drug reactions in controlled clinical trials, as some parameters, such as whether the reaction was present when a placebo was given, did not apply to any of the cases reviewed. Furthermore, the severity of the side effect (loss of vision in both eyes) does not warrant a rechallenge. Finally, the Naranjo scale credits reports where there is documentation that a drug achieved toxic levels in the blood stream and this situation is not typically relevant to the situation of drug-induced bilateral 2° ACG. A limitation of the bilateral 2° ACG scoring system was a product of the strength of reporting by the authors of the individual case reports. Some low scores could result from incomplete data reporting rather than the absolute absence of the positive scoring parameter in the patient. The comparison of patient demographics and chemical structure of the compounds implicated in drug-induced bilateral 2° ACG was reported but not powered for statistical analysis. The bilateral 2° ACG score was externally validated to topiramate-induced cases but the cutoff used for evidence supporting bilateral 2° ACG was somewhat arbitrary. Moreover, none of the drugs with high Naranjo and bilateral 2° ACG scores are suspected of producing 1° ACG. The inciting ingredients in the Anorexiant mix and Basic Detox Nutrient producing bilateral 2° ACG are not entirely clear but knowledge of the association is still useful.