advertisement
Editors Selection IGR 10-3
Comment
Comment by Tony Realini on:
68907 Validity of the Monocular Trial of Intraocular Pressure-Lowering at Different Time Points in Patients Starting Topical Glaucoma Medication, King AJ; Rotchford AP, JAMA ophthalmology, 2016; 134: 742-747
See also comment(s) by Kaweh Mansouri • Luciano Quaranta • Rotchfold & Anthony King •King and Rotchford are to be commended for their tenacious defense of the monocular drug trial, a clinical tool that a vast preponderance of studies has established as being generally unhelpful in the clinical management of glaucoma patients.
The monocular drug trial was proposed as a solution to a significant clinical problem: when initiating topical intraocular pressure (IOP)-lowering therapy for glaucoma, how can we distinguish between the therapeutic effect of the medication and the spontaneous IOP changes known to occur in eyes over time? We tend to initiate therapy when IOP is at its peak (i.e., higher than normal and thus deserving of reduction). Assuming that IOP in a given eye fluctuates within some range between a minimum value and a maximum value, any measurement following an extreme measurement (a near-maximum value or a near-minimum value) will likely be less extreme (the so-called regression to the mean). So if we see a patient whose IOP is unusually high (near-maximum), any subsequent IOP measurement is likely to be lower whether or not we initiate therapy. How can we distinguish between therapeutic and spontaneous IOP changes?
Per the monocular trial, we would measure IOP in both eyes, treat one eye, and after a reasonable course of therapy, re-measure IOP in both eyes. In theory, the IOP change in the untreated eye should represent purely spontaneous IOP fluctuation, while the IOP change in the treated eye should represent a mixture of both therapeutic and spontaneous IOP fluctuation. Subtracting the untreated eye's IOP change from the treated eye's IOP change should isolate the therapeutic IOP component.
Clever as it sounds, the monocular trial does not work. It does not work because essentially all of the assumptions underlying the monocular trial are false assumptions. Our group established this in a series of studies that have since all been replicated by independent groups. We also reported (in two separate cohorts of patients) that because of these false assumptions, the monocular trial itself does not work, and this finding has also been confirmed by numerous independent investigators, including the Ocular Hypertension Treatment Study (OHTS) research team which demonstrated the monocular trial's shortcomings in a post hoc analysis of the OHTS data set.
So how can King and Rotchford come up with such disparate results? One possibility is a simple Type 2 statistical error: ask the same question over and over again and eventually you will get the wrong answer by chance alone. An alternate explanation is that their data are in fact consistent with the larger body of research on this topic. A closer look at their data suggest this latter explanation as most likely. In their trial, 30 subjects underwent IOP assessment at a series of pretreatment visits, then were treated first in one eye and then both eyes with travoprost, and then underwent IOP assessment at a series of on-treatment visits. As they report, the mean error attributable to the monocular drug trial represented 24-42% of the true IOP reduction observed. Consequently, their data demonstrate that the monocular trial's estimate of IOP reduction will differ from the true IOP reduction (estimated at 7.0-8.6 mmHg on average) by approximately 2.5 to 4 mmHg or more in half of patients undergoing the trial.
All of us who have studied the monocular trial over the past decade have conceded that the trial almost certainly works well for some people ‒ but unfortunately we have no way of knowing who those people are. Imagine that you have just conducted a monocular trial and observed a 7-mmHg IOP reduction. Was this therapeutic or spontaneous IOP change? Based on the findings of King and Rotchford, there is a 50% chance that the drug really lowered IOP 3.0-4.5 mmHg or less and the remaining IOP change was spontaneous and not therapeutic. The only way you will know for sure how much your treatment lowered IOP is by continuing therapy and reassessing IOP over the next several visits, then comparing these values to those obtained over several prior pre-treatment visits. Given that this is necessary, the monocular drug trial has failed in its goal of isolating therapeutic IOP change and is of little if any real clinical value, and we should simply treat both eyes from the outset.
Comments
The comment section on the IGR website is restricted to WGA#One members only. Please log-in through your WGA#One account to continue.Log-in through WGA#One
