Cataract surgery is the most commonly performed elective surgery in the world, with >10 millions of cases being carried out each year (1). In addition, age-related corneal endothelial diseases (e.g., Fuchs endothelial corneal dystrophy; FECD) are common causes of visual impairment, and represent a leading indication for corneal transplantation (2–4). Therefore, with the aging global population, it is becoming increasingly common for patients to require treatment for co-existing age-related ocular diseases such as cataract and FECD.

FECD can lead to endothelial cell loss (ECL) with resultant corneal edema, ocular discomfort, and visual impairment (5). Once corneal decompensation sets in, corneal transplant serves as the mainstay of treatment for restoring the vision (6). In recent years, selective endothelial keratoplasty (EK) has been the treatment choice for managing corneal endothelial diseases (3, 4, 7). In EKs, the donor corneal tissue is inserted, and positioned against the posterior surface of the host cornea (8–10). In particular, Descemet's membrane endothelial keratoplasty (DMEK) involves the use of a manually prepared partial-thickness donor cornea containing only endothelium and Descemet membrane (11–13). DMEK has been shown to have superior postoperative visual acuity and lower graft rejection rate (14–17). Despite the established benefits, the adoption of DMEK is gaining popularity albeit slowly, owing to its steep surgical learning curve (16, 18–20).

The approach in managing a concomitant cataract with FECD can be done in various ways. One of the commonest approaches is to perform a combined DMEK and cataract surgery (i.e., “triple DMEK”). When compared to a staged DMEK procedure (i.e., cataract surgery followed by DMEK, or DMEK followed by cataract surgery), “triple DMEK” offers advantages such as improved cost-effectiveness, better intraoperative corneal clarity (due to simultaneous removal of the diseased and thickened endothelium and elimination of the risk of post-cataract surgery-induced corneal edema) and comparable clinical outcomes (8, 21). It was however also found that “triple DMEK” may be associated with a higher rate of postoperative complications such as graft detachment requiring postoperative re-bubbling (22–24). Overall, there is no consensus on whether to stage or combine DMEK with cataract surgery in patients who present with visually significant cataracts and FECD.

Thus, we performed a systematic review to appraise and compare the published evidence on the surgical outcomes of DMEK and “triple DMEK” procedures, which could help inform the future clinical practice on managing patients with co-existing corneal endothelial diseases and cataract. As graft detachment requiring postoperative re-bubbling is one of the most complications of DMEK, we have studied this as the main outcome measure of our systematic review.

In this systematic review, we aimed to compare the surgical outcomes and safety between DMEK alone and “triple DMEK”, with 36 studies and 11,401 eyes being included in this review. “Triple DMEK” demonstrated a better BCVA at 1-month postoperative (0.10 logMAR better) than DMEK, albeit non-significant at 3–6 months (0.07 logMAR better, p = 0.08). There was no significant difference in the rate of ECL and other postoperative complications such as re-bubbling rate, primary graft failure, CME, and PCR.

Our meta-analysis suggested that DMEK has a comparable rate of postoperative re-bubbling to “triple DMEK” (RD = −0.06; 95% CI: −0.13 to 0.00; p = 0.07). Whilst the difference in re-bubbling rate was statistically insignificant, it is important to highlight that there was a substantial heterogeneity (I² = 73%) among the included studies. The heterogeneity is likely ascribed to multiple confounding factors such as patient factors (e.g., age, lens status, depth of anterior chamber, and compliance to postoperative management like posturing), indication, surgeon's experience, surgical technique, choice of tamponade agent, and criteria for re-bubbling, amongst others. For instance, Dapena et al. (52) demonstrated that the graft detachment rate of DMEK reduced from 20% in the first 45 cases to 4.4% in the 91–135 cases. In addition, the use of 20% SF₆ for intraocular tamponade in DMEK has been reported to reduce the rate of partial graft detachment significantly when compared with air (69).

As direct comparative studies were lacking, we performed a meta-analysis of non-comparative DMEK studies to examine the difference in reported graft detachment comparing combined cataract surgery with DMEK and standalone DMEK. We found that in DMEK alone, the overall total and partial graft detachment rates were both 8.2%. Showail et al. reported no significant difference in graft detachment between both approaches (p = 0.78) (43) and similar observations were made by other studies (34, 39, 41). Contrary to that, Leon et al. (22) and Gundlach et al. (35) have identified triple DMEK as an independent risk factor for early graft detachment. These studies, however, do demonstrate significant heterogeneity with various confounders, e.g., age, surgeons' techniques, indications for DMEK and pre-operative lens status (phakic vs. pseudophakic) which may have led to varying outcomes of the studies. Our meta-analyses are also affected by several outliers which may reflect the learning curve of DMEK—e.g. surgeon 1 from Wubbels et al. (62) demonstrated a much higher rate of re-bubbling compared to other studies as the aim of the study was to establish the learning curve from the first 40 consecutive cases of DMEK performed.

In terms of visual outcome, our meta-analysis of existing literature suggests that “triple DMEK” offered better visual outcomes at 1 month postoperative, though non-statistically significant at 3–6 months postoperative. It is, however, important to note that the visual outcome at 1 month postoperative was based on only two studies, with significant weightage (96%) placed on one study (21). Chaurasia et al. (21) observed that “triple DMEK” resulted in a better BCVA (0.10 logMAR better) than DMEK at 1–6 months postoperative; however their finding was confounded by the higher rate of ocular co-morbidities and non-FECD cases in the latter group. Whilst there was limited long-term BCVA data available, a study by Schlogl et al. (42) evaluated the long-term outcomes of 250 eyes and found no significant difference between both approaches up to 5 years postoperatively. On the other hand, the ECL was shown to be comparable (0.8% difference) between the two approaches at 6 months postoperative, and the similarity was maintained at 5 years postoperative according to one study (42).

It is important to note that of the 17 studies that compared both approaches, four studies did not specify the preoperative lens status of DMEK eyes (39, 41–43), two studies reported a mix of pseudophakic and phakic DMEK surgeries but did not analyze them separately (37, 40). Similarly, Godin et al. (34) have reported a mix of pseudophakic and phakic DMEK surgeries and the group analyzed them independently. Four studies compared “triple DMEK” directly with pseudophakic DMEK surgeries (21, 36, 38, 44), whilst one study compared “triple DMEK” with phakic DMEK (35). These studies concluded that the surgical outcomes are comparable regardless of preoperative lens status and approaches, except for Crew et al. (38) who reported intraoperative hyphema was more common in “triple DMEK” compared to pseudophakic DMEK. Between approaches, both shared similar complication rates in terms of primary graft failure, CME and PCR.

One sequala to phakic DMEK is accelerated cataract progression, which may be secondary to surgical manipulation, air injection and postoperative topical steroid use (35). It was observed that cataract progression occurred in 72% of the phakic eyes post-DMEK and patients above the age of 50 have a higher risk of cataract progression when compared to younger patients (83 vs. 40%) (49). This differs from our meta-analysis where we observed a considerably lower (but highly variable) risk of cataract development in phakic eyes post-DMEK (mean 9.3%, ranged 0.4–72%) (49, 50, 52, 58, 59, 68). This could be attributed to several factors such as patient cohort and follow-up duration. The mean age of included studies reported cataract progression ranged from 50 to 68 years old, and the youngest patient included was 20 years old, whereas the oldest was 96 years old. Furthermore, follow-up duration was highly heterogeneous amongst studies as well, ranging from 6 to 60 months. These factors combined could lead to variable detection rates of cataract post-DMEK. Whilst doing a staged “DMEK followed by cataract surgery” offers several advantages such as more accurate biometry and potential ability to use a wider variety of intraocular lenses, anecdotally, staged “DMEK then cataract surgery” is less commonly performed due to the potential of damaging the in-situ DMEK (70, 71).

We have also attempted to further compare phakic DMEK (i.e., DMEK in phakic eyes) vs. “triple DMEK”, and pseudophakic DMEK (i.e., DMEK in pseudophakic eyes) vs. “triple DMEK”. However, this was not possible due to the lack of data and the heterogeneity in study design. Whilst we did not quantitatively evaluate the accommodation and refractive outcomes of either approach, Gundlach et al. (35) have suggested that phakic DMEK (i.e., DMEK in phakic eyes) may be beneficial in younger patients as accommodation power can be preserved. In addition, a hyperopic shift may occur following triple DMEK (65, 66), and this can be potentially avoided if cataract surgery is performed after DMEK. Given the low incidence of cataract development post-DMEK, the decision to conduct a targeted DMEK surgery or triple/sequential DMEK should consider the patient's age, preferences, refractive need, and social circumstances.

This review has several limitations. There was no RCT available in the literature that directly compared the outcome of DMEK alone and triple DMEK. In addition, the level and quality of the available evidence were mostly level 3 or 4, and low respectively, with a significant number of studies judged as having moderate to high risks of bias (Figures 2, 3). Furthermore, significant heterogeneity existed in the studies, such as study design, study population, surgical techniques, outcome measures, methods of reporting, and duration of follow-up; and we could not study other factors or important complications such as glaucoma (72), which was not routinely reported. Risk of bias is high as the indication for DMEK included not only FECD but also other causes of corneal endothelial dysfunction such as PBK, complex eyes and re-grafts (73), which have been shown to have a prognostic impact on the surgical outcome (21). There were also inadequate longitudinal studies that compared DMEK alone and triple DMEK, hence making it difficult to provide a meaningful comparison regarding the long-term clinical outcomes of both approaches. With the reasons cited above, whilst meta-analysis could be done with the limited literature available at this juncture, it is hard to make a conclusive assessment on these two approaches.

Overall, our review showed that “triple DMEK” and DMEK alone surgeries are largely comparable in surgical outcomes, sharing similar ECL and complication rates, except for re possible graft detachment rates (lower in DMEK only eyes), which are important clinical points that should be discussed with patients prior to surgery. Looking at the existing evidences, sequential DMEK surgery (cataract surgery followed by DMEK) in patients with endothelial disease who are above the age of 50 years old or have concurrent cataracts could potentially avoid graft detachment. Targeted DMEK alone may be considered in younger patients with no evidence of cataract formation. The decision should, however, be guided by other factors such as patient's preference, social circumstances, surgeon's experience, and availability of operating theaters. Finally, there exists gap in current literature and further adequately powered, randomized controlled trials specifically looking at the long-term outcomes of combined and staged DMEK (with cataract surgery) are warranted for a definitive comparison of the two approaches.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

JM and MA conceptualized and supervised the study. KT, ST, and MA conducted the literature review and curated the data. KT, ST, DT, and MA conducted the formal analysis of the data. All authors wrote, reviewed, edited and approved the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.