Secondary Logo

Journal Logo

Clinical Science

Anatomical Changes in the Anterior Chamber Volume After Descemet Membrane Endothelial Keratoplasty

Onouchi, Hiromi MD, PhD*; Hayashi, Takahiko MD, PhD†,‡; Shimizu, Toshiki MD; Matsuzawa, Akiko MD, PhD§; Suzuki, Yasuyuki MD, PhD*; Kato, Naoko MD, PhD*,†

Author Information
doi: 10.1097/ICO.0000000000002535


Recent advances in endothelial keratoplasty (EK) have allowed selective corneal endothelium transplantation in place of full-thickness keratoplasty (penetrating keratoplasty) as a treatment for corneal endothelial diseases.1,2 Currently, there are 2 EK types: Descemet-stripping automated endothelial keratoplasty (DSAEK) and Descemet membrane endothelial keratoplasty (DMEK).2,3 In both procedures, implanted donor corneal grafts can be fixed to the back surface of the corneal stroma by using air/gas bubbles.2,3

Compared with penetrating keratoplasty, EK has some advantages, including quick visual outcome improvement, better visual acuity, and lower immunologic rejection.4–6 For most patients with cataracts, cataract removal and intraocular lens (IOL) implantation should be performed before or along with EK.1 However, IOL power calculation is difficult in these situations given the irregular corneal surface and post-EK hyperopic shifts. Apart from a few reports regarding post-EK refractive changes,7–10 there have been no studies regarding the precise examination of parameters such as the post-DMEK anterior chamber depth (ACD) or anterior chamber volume (ACV). Therefore, we believed that knowing the amount and mechanisms of change of the refractive power by EK might be valuable for IOL calculation before cataract surgery and for explaining the expectation of visual function after EK to the patients.

In this study, we assessed the pre- and post-DMEK anatomical changes of the anterior chamber. Moreover, we investigated the relevant factors that affect these changes and whether the post-DMEK hyperopic change was associated with anatomical changes in the anterior chamber.


This retrospective study was performed in compliance with the ethical standards of the Declaration of Helsinki and was approved by the Institutional Review Board at Yokohama Minami Kyosai Hospital (approval number: YKH_29_03_05, 92, UMIN 000027586). We obtained written informed consent from all the participants before study enrollment.


We included a consecutive case series of 25 eyes from 25 patients [21 women and 4 men; age, 72.7 ± 7.7 years (mean ± SD)] who underwent DMEK. These were all pseudophakic eyes that had undergone pre-DMEK cataract extraction and IOL implantation. We could not obtain precise information regarding previous cataract surgery. We excluded eyes with a scleral-sutured IOL or a sulcus-fixated IOL. The causes of the endothelial decompensation across the eyes were as follows: Fuchs endothelial corneal dystrophy (FECD) (8 eyes), bullous keratopathy caused by argon laser iridotomy (9 eyes), pseudoexfoliation syndrome (1 eye), endotheliitis (1 eye), and unknown (6 eyes).

Surgical Procedures

DMEK procedure was performed as previously reported.11,12 Briefly, Descemet membrane graft was prepared under sterile conditions shortly before starting surgery. We stained a donor disk held in a vacuum punch (Moria Japan, Tokyo, Japan) with 0.1% Brilliant Blue G dye. Descemet membrane was gently peeled from the stroma; moreover, 4 small asymmetric semicircular marks were made on the donor-graft edge.13,14 After creating 2 paracenteses in the corneal limbus, we created a 2.8-mm wide corneoscleral tunnel at the 12-o'clock position. We inserted an anterior chamber maintainer through 1 corneal side port. Furthermore, we performed peripheral iridectomy at the 6-o'clock position by using a 25-gauge vitreous cutter (Stellaris PC Vitrectomy system; Bausch + Lomb, St. Louis, MO). Descemet membrane was stripped using a DMEK Sinskey hook. Subsequently, the DMEK donor graft was inserted using an IOL inserter (WJ-60M; Santen, Osaka, Japan), after which all incisions were immediately sutured using 10-0 nylon (Mani, Tochigi, Japan). After performing donor-graft unfolding using the nontouch technique, filtered room air was injected underneath the unfolded graft. After confirmation of graft attachment to the host cornea, the air volume was adjusted to 80% to 90% of the ACV. At the end of surgery, 0.4 mg betamethasone (Rinderon; Shionogi, Osaka, Japan) was subconjunctivally injected; furthermore, 1.5% levofloxacin eye drops (Cravit; Santen) were instilled. Two hours after surgery, we performed slitlamp examination. All patients were asked to maintain a supine position for several days until complete air disappearance.

Postoperative medications included 1.5% levofloxacin (Cravit; Santen), betamethasone (Sanbetason; Santen), and 2% rebamipide ophthalmic solution (Mucosta; Otsuka, Tokyo, Japan), which were administered 4 times per day for 3 months and subsequently tapered. We did not include topical tropicamide in the postoperative regimen.

Ophthalmic Examinations

We performed conventional ophthalmic examinations, including slitlamp microscopy, fundoscopy, best spectacle-corrected visual acuity (BSCVA), spherical equivalent (SE), and intraocular pressure measurement. Moreover, we measured the following parameters: ACV, ACD, pupil diameter (PD), axial length (AXL), central corneal thickness (CCT), scleral spur angle (SSA), iris damage score (IDS),15,16 and iris posterior synechiae score (IPSS).17 We evaluated the BSCVA before DMEK and at 1 month and at 3, 6, and 12 months after DMEK and the SE before DMEK and at 6 to 12 months after DMEK. We evaluated the ACV, ACD, CCT, PD, and SSA through anterior segment optical coherence tomography (SS-1000; Tomey Corporation, Aichi, Japan) before DMEK and at 1 month and 3, 6, and 12 months after DMEK. We calculated ∆ACV and ∆ACD by subtracting the postoperative ACV and ACD at 1 month after DMEK from their preoperative values. We used the same protocol to calculate the ∆SSA and ∆CCT. We calculated ∆SE by subtracting the postoperative SE at 6 to 12 months after DMEK from the preoperative values. Before DMEK surgery, we evaluated the AXL using optical biometry (IOLMaster 500; Carl Zeiss Meditec AG, Oberkochen, Germany). The preoperative IDS was defined as the iris damage area and classified using 5 grades as previously reported.15,16 Briefly, grades 0, 1, 2, 3, and 4 indicated no damage or iris damage limited to only 1, 2, 3, or 4 quadrants, respectively; moreover, it was measured between cataract surgery and DMEK. The IPSS was evaluated as previously reported.17 Briefly, the iris posterior synechiae severity was defined as a grade from 0 to 8 based on the extent of the area in which it existed. The pupillary edge was divided into 8 areas of 45 degrees each. Grade 0 meant no iris posterior synechiae, whereas grades 1, 2, 3, 4, 5, 6, 7, and 8 indicated that iris posterior synechiae were observed in <45, 46 to 90, 91 to 135, 136 to 180, 181 to 225, 226 to 270, 271 to 315, and 360 degrees (over the entire circumference of the iris edge; the pupil is hardly dilated), respectively.

Statistical Analysis

All statistical analyses were performed using StatMate V for Windows statistical software (ATMS Co, Ltd, Tokyo, Japan). The Wilcoxon test or paired t test was used to compare mean preoperative and postoperative ACV, ACD, CCT, and SSA values, where appropriate. We used Pearson correlation analysis to determine the correlations between ∆ACV, ∆SE, and other variables (pre-DMEK ACD, SSA, and IDS values and post-DMEK ∆ACD, ∆SSA, ∆CCT, and IPSS values). We considered a P value <0.05 as statistically significant.


Table 1 summarizes the post-DMEK changes in the BSCVA, SE, PD, CCT, SSA, ACD, and ACV in all eyes. The data of eyes with FECD and the other eyes are shown in Supplemental Digital Content 1 (see Supplemental Table 1, Figure 1 shows the change in ACV after DMEK in all eyes. Figure 2 shows the AS-OCT images of SSA measurements.

Preoperative 1 mo 3 mo 6 mo 12 mo
BSCVA (logMAR) 0.91 ± 0.52 0.22 ± 0.24 0.10 ± 0.20 0.07 ± 0.17 0.07 ± 0.17
P <0.001 <0.001 <0.001 <0.001
SE (D) −1.87 ± 1.85 −0.87 ± 1.85
P <0.001
PD (mm) 3.18 ± 1.07 2.98 ± 0.87 3.3 ± 0.99 3.24 ± 0.86 3.23 ± 0.72
P 0.132 0.251 0.353 0.38
CCT (μm) 719 ± 112 521 ± 45 507 ± 48 515 ± 40 522 ± 42
P <0.001 <0.001 <0.001 <0.001
SSA (degrees) 50.6 ± 15.6 61.8 ± 13.4 66.2 ± 14.1 67.0 ± 11.7 65.5 ± 10.9
P <0.001 <0.001 <0.001 <0.001
ACD (mm) 2.89 ± 0.41 3.51 ± 0.4 3.51 ± 0.41 3.49 ± 0.4 3.51 ± 0.44
P <0.001 <0.001 <0.001 <0.001
ACV (mm3) 129.2 ± 24.4 153.2 ± 19.3 153.6 ± 19.3 155.7 ± 20.1 155.4 ± 18.9
P <0.001 <0.001 <0.001 <0.001

Post-DMEK changes in ACV. There was a significant post-DMEK increase in the ACV at all post-DMEK points. *P < 0.001.
Representative anterior segment optical coherence tomography images of SSA measurements before (A) and 12 months after (B) DMEK.

Iris Damage Score

Five eyes (20.0%) did not have any preexisting iris damage. Eight eyes (32.0%) had grade 1 iris damage, 9 eyes (36.0%) grade 2, and 2 eyes (8.0%) grade 3. None of the eyes exhibited grade 4 damage. One eye (4.0%) had grade 5 damage.

Iris Posterior Synechiae Score

None of the eyes had any iris posterior synechiae after phacoemulsification. After DMEK, iris posterior synechiae were detected in 16 of 25 eyes (64.0%). The IPSS were as follows: 9 eyes (36.0%) had no iris posterior synechiae (grade 0); 1 eye (4.0%) showed grade 1, 6 eyes (24.0%) grade 3, 7 eyes (28.0%) grade 4, and 2 eyes (8.0%) grade 5 iris posterior synechiae; none of the eyes exhibited grades 2, 6, 7, or 8.

Correlations Between ΔACV, ΔSE, and Other Preexisting Factors

We observed that ∆ACV was negatively correlated with preoperative ACD and SSA values (R = 0.643 and 0.555, respectively, all P < 0.001; Figs. 3A and B) and positively correlated with ∆ACD and ∆SSA (R = 0.799, P < 0.001, and R = 0.608, P < 0.001, respectively; Figs. 3C and D). There was no significant correlation of ∆CCT, IDS, and IPSS with ∆ACV (R = 0.004, 0.244, and 0.232, P = 0.985, 0.194, and 0.217, respectively). We observed that ∆SE was negatively correlated with preoperative ACV, ACD, and SSA values (R = −0.645, −0.587, and −0.649; P < 0.001, <0.01, and <0.001, respectively; Figs. 4A–C) and positively correlated with ∆ACV, ∆ACD, and ∆SSA (R = 0.680, 0.455, and 0.478; P < 0.001, <0.05, and <0.05, respectively; Figs. 4D–F). The data in eyes with FECD and the other eyes were almost similar to those in all eyes (see Supplemental Tables 2 and 3, Supplemental Digital Content 1,

Correlations of ΔACV with pre-DMEK, ACD, and SSA and with ΔACD and ΔSSA. ΔACV was negatively correlated with pre-DMEK ACD (A) and SSA (B) and was positively correlated with ΔACD (C) and ΔSSA (D).
Correlations of ΔSE with other factors. ΔSE was negatively correlated with pre-DMEK ACV (A), ACD (B), and SSA (C) and was positively correlated with ΔACV (D), ΔACD (E), and ΔSSA (F).


Our findings demonstrated several post-DMEK anatomical changes; specifically, there was a significant post-DMEK increase in the ACV, ACD, and SSA and a decrease in the CCT. These changes could be attributed to specific eye conditions, including a shallow anterior chamber because ∆ACV was negatively correlated with preoperative ACD and SSA and positively correlated with ∆ACD and ∆SSA. There was no significant correlation between ∆ACV and preoperative IDS, ∆CCT, and IPSS.

Although the IOL position, in theory, is not changed by DMEK, there was a remarkable post-DMEK increase in ACV. We observed a similar tendency in the ACD time course. Regardless of the stable IOL position and fixation, our results indicate the need to assess the mechanism underlying the observed phenomenon.

∆SE increased after DMEK, and ∆SE was positively correlated with ∆ACV, ∆ACD, and ∆SSA and was negatively correlated with preoperative ACV, ACD, and SSA. Although hyperopic shifts of the refraction have previously been reported after combining DMEK with cataract surgery (triple DMEK),10,18 the causes for refractive alteration have not been fully clarified. We speculate that a slight posterior shift of IOL could occur due to enlargement of the SSA, ACD, and ACV, causing a hyperopic shift in refractive power. Moreover, some flattening of the anterior curvature and the increase in ACD might contribute to the hyperopic shift.19 A prospective study that would analyze the refractive status of the anterior and posterior curvatures after DMEK is warranted.

This phenomenon could have been attributed to drastic SSA changes. In our study, we observed a postoperative increase in the SSA. There was a shift in the iris plane concomitant with the enlarged SSA. Consequently, we hypothesized that the ACV enlargement was caused by a posterior iris plane shift, which could have resulted from the iris change generated during the air/gas tamponade. We previously reported a high incidence of post-DMEK iris change (iris posterior synechiae) that was positively correlated with shorter AXLs of the bulbus and preexisting iris damage.17 Our hypothesis is consistent with the observed inverse correlation between ∆ACV and preoperative ACD and SSA values.

In our study, we assessed whether post-DMEK iris posterior synechiae could have been caused by short distances between the iris posterior plane and the anterior lens capsule, which further increased inflammatory cytokine release from the damage to the blood–aqueous barrier of the iris vessels.17 By contrast to our expectations, ∆ACV was not affected by preoperative IDS or IPSS, which indicates that iris posterior synechia does not affect the post-DMEK posterior iris plane shift. However, there was a positive correlation of ∆ACV with ∆ACD and ∆SSA, which indicates that the ACV increased with anterior chamber deepening, that is, with posterior iris movement.

Taken together, our findings regarding the posterior iris shift could be reflective of the anatomical shift of the iris plane. This shift does not occur because of the mechanical traction exerted by the posterior synechiae onto the anterior lens capsule. Instead, it occurs because of improved inflammation around the ciliary body and iris stroma near the corneal limbus, which is caused by DMEK-induced functional recovery of corneal endothelial cells. Consistent with our hypothesis, Nishiyama et al20 reported that keratitis-induced severe anterior segment inflammation could induce ciliochoroidal detachment, which subsequently reduces ACD due to anterior rotation. We previously reported a relatively high incidence of post-DMEK cystoid macular edema in eyes with higher preexisting iris damage, which suggests postoperative inflammation around the iris and ciliary body.21 We also speculate that these changes are caused by biometric changes in the anterior chamber from a peripheral iridectomy in DMEK because it has been reported that laser peripheral iridotomy results in an increase in the angle width in primary angle-closure suspects.22,23

Moreover, we assessed the anterior shift of the posterior corneal surface because corneal edema was resolved by the implanted healthy donor endothelial cells. However, we did not observe a correlation of ∆ACV with ∆CCT, which indicates that functional recovery of endothelial cells was not the main contributing factor to the observed ACV increase.

Our study had several limitations. We did not evaluate ciliary body edema; moreover, we used a relatively small sample size. Future studies involving quantitative evaluation of the ciliary body through tomographic analysis are needed because this could allow precise elucidation of the mechanism underlying the anatomical anterior chamber changes and enable pre-EK and post-EK global inflammation management.

In conclusion, this study reported a post-DMEK increase in the ACV and SE, especially in eyes with a narrow-angled shallow anterior chamber. This could be attributed to improvements of ciliary inflammation after DMEK and iridotomy. Therefore, it might be possible to predict ∆ACV and apply it in the pre-EK estimation of several parameters.


1. Price MO, Gupta P, Lass J, et al. EK (DLEK, DSEK, DMEK): new frontier in cornea surgery. Annu Rev Vis Sci. 2017;3:69–90.
2. Gorovoy MS. Descemet-stripping automated endothelial keratoplasty conclusion: DSAEK surgery allows rapid, excellent BSCVA visual. Cornea. 2006;33907:886–889.
3. Price MO, Price FW. Descemet membrane endothelial keratoplasty. Int Ophthalmol Clin. 2010;50:137–147.
4. Vasiliauskaitė I, Oellerich S, Ham L, et al. Descemet membrane endothelial keratoplasty: ten-year graft survival and clinical outcomes. Am J Ophthalmol. 2020 Apr 10;S0002-9394(20)30172-0. [epub ahead of print].
5. Schrittenlocher S, Bachmann B, Tiurbe AM, et al. Impact of preoperative visual acuity on Descemet membrane endothelial keratoplasty (DMEK) outcome. Graefes Arch Clin Exp Ophthalmol. 2019;257:321–329.
6. Koenig SB, Covert DJ. Early results of small-incision Descemet's stripping and automated endothelial keratoplasty. Ophthalmology. 2007;114:221–227.
7. van Dijk K, Rodriguez-Calvo-De-Mora M, van Esch H, et al. Two-year refractive outcomes after Descemet membrane endothelial keratoplasty. Cornea. 2016;35:1548–1555.
8. Ham L, Dapena I, Moutsouris K, et al. Refractive change and stability after Descemet membrane endothelial keratoplasty. Effect of corneal dehydration-induced hyperopic shift on intraocular lens power calculation. J Cataract Refract Surg. 2011;37:1455–1464.
9. Hwang RY, Gauthier DJ, Wallace D, et al. Refractive changes after Descemet stripping endothelial keratoplasty: a simplified mathematical model. Invest Ophthalmol Vis Sci. 2011;52:1043–1054.
10. Laaser K, Bachmann BO, Horn FK, et al. Descemet membrane endothelial keratoplasty combined with phacoemulsification and intraocular lens implantation: advanced triple procedure. Am J Ophthalmol. 2012;154:47–55.e2.
11. Kruse FE, Laaser K, Cursiefen C, et al. A stepwise approach to donor preparation and insertion increases safety and outcome of Descemet membrane endothelial keratoplasty. Cornea. 2011;30:580–587.
12. Hayashi T, Oyakawa I, Kato N. Techniques for learning Descemet membrane endothelial keratoplasty for eyes of Asian patients with shallow anterior chamber. Cornea. 2017;36:390–393.
13. Matsuzawa A, Hayashi T, Oyakawa I, et al. Use of four asymmetric marks to orient the donor graft during Descemet's membrane endothelial keratoplasty. BMJ Open Ophthalmol. 2017;1:e000080.
14. Bachmann BO, Laaser K, Cursiefen C, et al. A method to confirm correct orientation of Descemet membrane during Descemet membrane endothelial keratoplasty. Am J Ophthalmol. 2010;149:922–925.e2.
15. Aketa N, Yamaguchi T, Suzuki T, et al. Iris damage is associated with elevated cytokine levels in aqueous humor. Invest Ophthalmol Vis Sci. 2017;58:BIO42–BIO51.
16. Ishii N, Yamaguchi T, Yazu H, et al. Factors associated with graft survival and endothelial cell density after Descemet's stripping automated endothelial keratoplasty. Sci Rep. 2016;6:25276.
17. Shimizu T, Hayashi T, Yuda K, et al. Short axial length and iris damage are associated with iris posterior synechiae after Descemet membrane endothelial keratoplasty in Asian eyes. Cornea. 2018;37:1355–1359.
18. Price MO, Giebel AW, Fairchild KM, et al. Descemet's membrane endothelial keratoplasty: prospective multicenter study of visual and refractive outcomes and endothelial survival. Ophthalmology. 2009;116:2361–2368.
19. Clemmensen K, Ivarsen A, Hjortdal J. Changes in corneal power after Descemet stripping automated endothelial keratoplasty. J Refract Surg. 2015;31:807–812.
20. Nishiyama I, Oie Y, Matsushita K, et al. Transient extremely shallow anterior chamber caused by ciliochoroidal detachment in a patient with Mycobacterium chelonae keratitis. Am J Ophthalmol Case Rep. 2019;15:100530.
21. Inoda S, Hayashi T, Takahashi H, et al. Risk factors for cystoid macular edema after Descemet membrane endothelial keratoplasty. Cornea. 2019;38:820–824.
22. How AC, Baskaran M, Kumar RS, et al. Changes in anterior segment morphology after laser peripheral iridotomy: an anterior segment optical coherence tomography study. Ophthalmology. 2012;119:1383–1387.
23. Vryonis N, Nikita E, Vergados I, et al. Anterior chamber morphology before and after laser peripheral iridotomy determined by Scheimpflug technology in white patients with narrow angles. J Glaucoma. 2013;22:679–683.

anterior chamber volume; bullous keratopathy; Descemet membrane endothelial keratoplasty; endothelial keratoplasty

Supplemental Digital Content

Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc.