advertisement

Editors Selection IGR 11-3

Surgical Therapy: The Ahmed versus Baerveldt Study

Comment by Keith Barton & William Feuer on:

46649 The Ahmed Versus Baerveldt Study One-Year Treatment Outcomes, Christakis PG; Kalenak JW; Zurakowski D et al., Ophthalmology, 2011;

Find related abstracts

Potential conflict: Authors of similar study (ABC study)

Glaucoma surgery is experiencing a resurgence and it is encouraging that 12 months after online publication of the Ahmed Baerveldt Comparison (ABC) study^1,2 a second randomized surgical trial, the Ahmed versus Baerveldt (AVB) study, comparing two aqueous shunts that are commonly used, i.e., The Baerveldt 101-350 (BGI) and the Ahmed FP7 silicone implant (AGV) has reported baseline characteristics and one year outcomes.^3,4 It is therefore difficult to comment on the latter study without referring to the former. Christakis et al. (1598) in the AVB study randomized 238 patients, between 2005 and 2009, 124 to the AGV and 114 to the BGI with similar but slightly broader patient eligibility to the ABC study. There were minor demographic differences between the two studies, with a slight female preponderance in the Baerveldt group and more white patients in the AVB study, with slightly more POAG (50% versus 40%) and less neovascular glaucoma (21% versus 29%) compared with the ABC study. The AVB study differed from ABC in another respect, in that all surgeons were experienced users of both types of implants. Patient retention at one year was similar in the two studies. The primary outcome measure in both was success in terms of IOP control within a predefined target range, though the upper and lower IOP limits differ between the two studies. The AVB study used a more stringent cut-off of > 18 mmHg for high pressure failure and a slightly less stringent limit of 5 mmHg for hypotony (cf. > 21 mmHg and < 6 mmHg in the ABC study). In addition, the AVB study includes sight-threatening complications in its definition of failure. The reported cumulative probability of failure at one year was higher in the AGV group than the BGI group in both studies, but not significantly so in the ABC study (AVB: p = 0.049, ABC: p = 0.52). Both studies reported similar differences between groups in IOP's and medications at one year. AGV IOPs were 3 mmHg higher than BGI in both studies (both p < 0.01) and the AGV group averaged 0.4 (AVB, p = 0.03) and 0.3 (ABC, p = 0.07) more medications than the BGI group. The reported failure rate of 43% in the AGV group compared with 28% in the BGI group are higher than the corresponding failure rates of 16% and 14% respectively in the ABC study. Potential explanations for this could include the more stringent upper IOP limit and inclusion of sight-threatening complications in the definition of failure in the AVB study. However, when the two studies are compared using the alternative IOP end-point of ≥ 18 mmHg that was also reported by the ABC study group, and which is close to the definition of > 18 mmHg used by the AVB study, and when sight-threatening complications are removed from the definition of failure in the AVB study, the corresponding failure rates for the BGI are 24% (AVB) and 16% (ABC), and for the AGV remain at 41% (AVB) and 22% (ABC). When an even lower IOP limit for failure was examined in the ABC study (< 15 mmHg), patients receiving the AGV were significantly more likely to fail than the BGI group (p = 0.006). The AVB study group did not observe a significant difference between the two implants at their lower IOP target of &se; 14 mmHg. In contrast to the ABC study, that reported more complications in the BGI than AGV group, no significant difference in overall number of complications between the two groups was observed in the AVB study, though significantly more patients required re-intervention in the BGI group. A comparison of individual complications reveals some similarities between the two studies: 10% more corneal edema in the BGI group than in the AGV group (both studies p < 0.05) and more tube obstruction in BGI group (AVB 6%, p = 0.12 and ABC 7%, p = 0.02) than in the AGV group. The AVB study reported an 8% higher rate of encapsulated blebs in the AGV group than in the BGI group (p = 0.01). Motility disturbances with the AGV were similar to the ABC study at 6% in the AGV group in both studies. The rate of motility disturbance listed in the AVB study with the BGI was 3% at one year, compared with 5% reporting diplopia in the ABC study.

One year is very little time in the life of a glaucoma patient

One year is very little time in the life of a glaucoma patient and it is likely that some of the differences reported between these two implants and these two studies may change with longer follow-up. In the mean-time the AVB study provides further useful information in the differences between two of the most commonly used implants at a time when aqueous shunt usage is increasing steadily. The study investigators are to be commended for their hard work.

• ← Previous
• Next →