The results of the current study demonstrated that XEN45 implant, either alone or in combination with cataract surgery, significantly lowered the IOP and reduced the need of postoperative ocular hypotensive medication in patients with OAG in a real-life scenario.
There were no significant differences in IOP lowering between the first-cohort (the first-30 eyes who underwent a XEN device implant) and the second-cohort (The first-30 eyes who underwent surgery from January 2018).
Additionally, it should be highlighted the relatively high proportion of patients achieving low target IOPs, with 42.4% and 65.6% of patients achieving an IOP ≤ 14 mm Hg and ≤ 16 mm Hg without treatment, respectively. The proportion of eyes who achieved an IOP ≤ 14 mm Hg without treatment and the proportion of eyes who achieved an IOP ≤ 16 mm Hg, both with and without treatment, were significantly greater in the second-cohort.
From a clinical point of view, different studies have reported the mid- and long-term efficacy, in terms of IOP-lowering and the amount of ocular-hypotensive medications reduction, and safety of XEN45 implant, either alone or in combination with phacoemulsification surgery, in OAG patients [11–21].
In this respect, the results of the study do not differ significantly from the published evidence.
Our study found significant differences, throughout study follow-up, between the eyes who underwent XEN-alone and those who underwent XEN + Phaco. Interestingly, these differences were not always in favor of the same group throughout follow-up. Preoperative IOP was significantly greater in the XEN-alone group, which can be justified by the fact that the surgery was indicated for purely hypotensive purposes.
The mean IOP was significantly lower in the XEN group at day-1 and week-1, while was significantly lower in the combined group at months 1 and 6. Nevertheless, in terms of percentage, there were no significant differences in IOP lowering between XEN-alone and XEN + Phaco beyond the week-1.
The effectiveness of XEN45 in combination with cataract surgery have been analyzed in some papers. However, there have been conflicting results regarding the superiority of the solo procedure over the combined procedure with cataract surgery [12, 15, 16, 26, 27].
While some authors did not find significant differences in IOP lowering between XEN-alone and the XEN + Phaco groups [12, 15, 16], other ones reported higher success rates in the XEN-alone group [26, 27].
Our study has evaluated the effect of the learning curve, not only on the results but also on the indication of the procedure. Our study found a significant IOP lowering regardless of the surgeon's experience with the device, which agrees with the evidence suggesting that XEN devices has been associated with a fast-learning curve [28].
In our study, after adjusting by different covariates, there were no significant differences in IOP lowering between the first and the second cohorts of XEN devices, with the exception of the month-24 IOP lowering (greater in the second cohort).
Although the proportion of eyes achieving an IOP ≤ 14 mmHg and ≤ 16 mmHg was significantly greater in the second-cohort, it may due to the fact that the preoperative IOP was significantly lower in this cohort.
In our opinion, the paradigm shift that the experience with the XEN45 device has brought to our practice is noteworthy.
Although XEN45 devices were initially indicated in mild-to-moderate primary OAG, as reflected in the first-cohort (see Table 2), as experience with the device has been gained and its efficacy/safety profile has been seen, the indications have been expanded. For example, in the first 30 devices, the most common diagnosis was primary-OAG (89.3%; 25/30), while in the second cohort, primary-OAG only represented 60% (18/30) (p = 0.0097).
We have also found differences in preoperative IOP, being significantly lower in the second-cohort than in the first-cohort (p = 0.0033). In our opinion, this suggest clearly that as we have gained experience with the XEN45 device, we have broadened the indications for surgery.
Regarding safety, the incidence and type of complications was similar to that previously published [11–21, 26, 27]. Thirty-six (23.4%) eyes underwent a needling procedure and 16 (10.5%) eyes required a surgical revision. No differences were observed between the two study cohorts.
The main limitation of the current study is its retrospective design. Selection bias and potential confounders are inherent to retrospective studies. Nevertheless, the selection of strict inclusion / exclusion criteria, as well as the inclusion of a large number of eyes, may minimized these issues.