Article Text

Download PDFPDF

Changes in donor corneal lenticule thickness after Descemet’s stripping automated endothelial keratoplasty (DSAEK) with organ-cultured corneas
  1. P Pogorelov,
  2. C Cursiefen,
  3. B O Bachmann,
  4. F E Kruse
  1. Department of Ophthalmology, University of Erlangen-Nuremberg, Erlangen, Germany
  1. Dr P Pogorelov, Department of Opthalmology, University of Erlangen-Nuremberg, Schwabachanlage 6, D-91054 Erlangen, Germany; pavel.pogorelov{at}


Aims: The aim of the study was to quantify changes in donor and host corneal tissue after Descemet’s stripping automated endothelial keratoplasty (DSAEK) using organ-cultured corneas and to correlate these findings with the visual acuity during intermediate-term follow-up.

Methods: Fifteen eyes of 15 consecutive patients, with Fuchs’ endothelial dystrophy treated with DSAEK using organ-cultured corneas, underwent ophthalmological examination, including slit lamp-adapted optical coherence tomography, at 1, 3 and 7 days, and 4 weeks, 8 weeks and 6 months after the surgery.

Results: The mean best spectacle-corrected visual acuity (BSCVA) improved from 20/100 pre-operatively to 20/40 at 6-months post-operatively (p<0.0001). A continuous decrease of thickness of the grafted lenticule was observed during the follow-up (mean thickness immediately after surgery 191(SD 56) μm, compared with 100 (SD 38) μm 6 months after surgery, p<0.001). The central corneal thickness decreased from 1057 (SD 86) μm at the first post-operative day to 661 (SD 74) μm after 6 months. Both central corneal thickness and the thickness of the posterior donor lamella correlated with the 6-month BSCVA (Pearson correlations −0.745 and −0.589, respectively, p<0.05).

Conclusions: Organ-cultured corneas can be used successfully for DSAEK. The thickness of the grafted corneal lenticule correlated with the BSCVA 6 months after the surgery. It decreased continuously during the follow-up period.

View Full Text

Statistics from

In contrast to penetrating keratoplasty (PK), posterior lamellar keratoplasty allows for selective replacement of diseased host endothelium.13 This is a more specific procedure than PK, and it has been successfully used in patients with endothelial corneal disorders such as Fuchs’ corneal dystrophy and pseudophakic bullous keratopathy. The most common technique for posterior lamellar keratoplasty is Descemet’s stripping automated endothelial keratoplasty (DSAEK).48 It employs mechanical stripping of the host endothelium and replacement with a healthy homograft of a posterior corneal lenticule harvested using an automated microkeratome.

Slit lamp-adapted optical coherence tomography (SL-OCT) enables non-contact in-depth measurements of the cornea and tissues lining the anterior chamber of the eye.912

The purpose of this clinical study was to quantify changes in donor and host corneal tissue after DSAEK using organ-cultured corneas and to correlate these findings with the visual acuity (VA) during intermediate-term follow-up.


Fifteen phakic eyes of 15 consecutive patients, nine women and six men, with a mean age of 72.8 (SD 5.0) years, undergoing DSAEK between August 2006 and September 2007 were enrolled in this study. Institutional review board approval was obtained. All patients had clinically significant corneal oedema from Fuchs’ endothelial dystrophy. In three eyes (20%), simultaneous phakoemulsification with implantation of a posterior chamber intraocular lens in the bag was performed (so-called triple procedure). Post-operatively, all patients underwent quantification of uncorrected VA, best spectacle-corrected VA (BSCVA), slit-lamp examination, measurement of intraocular pressure using applanation tonometry, and a SL-OCT examination of the anterior eye segment at 1, 3 and 7 days, and 4 weeks, 8 weeks and 6 months after the surgery.

All surgeries were performed by two surgeons (FEK and CC) under general anaesthesia. Donor corneas were cultured in the eye bank of the Department of Ophthalmology of the University of Erlangen-Nuremberg, Germany in minimal essential medium (MEM) for 307.7 (SD 118.9) h at 37°C. All donor corneas were subjected to 5% dextran-containing medium for 6 days prior to surgery. The post mortem time of corneal donor tissue was 8.0 (SD 6.0) h with a mean donor age of 78.1 (SD 9.7) years. The mean endothelial cell density of donor corneas was 2309 (SD 135) cells/mm2. The preparation of donor corneal lenticules was performed using a Moria ALTK microkeratome (Moria, Doylestown, PA, USA) with the blade set to 350 μm depth, as previously described.5 In patients undergoing a triple procedure (keratoplasty, cataract extraction and intraocular lens placement), phakoemulsification was performed through the same incision. The remaining steps of the surgical procedure have also been described previously.1 2 6 7 13

OCT measurements were performed with a commercially available anterior segment OCT (SL-OCT, Heidelberg Engineering, Luebeck, Germany) that uses a wavelength of 1310 nm. This SL-OCT system uses a scanning module with a lateral scan range of 6 mm (Haag-Streit BD 900, Wedel, Germany). The optical resolution is between 20 and 100 μm. The OCT image was sampled with an axial digital resolution of 3 μm. The SL-OCT examination of the anterior segment was performed with the subject in the sitting position. The scanning line was positioned onto the corneal vertex in the 180° axis, allowing cross-sectional imaging of the cornea and the entire anterior segment. The measurements of the central corneal thickness were performed manually at the corneal vertex within of 1 mm central zone (fig 5). The peripheral corneal thickness was also measured manually at a distance of 3 mm nasally and temporally of the corneal vertex.

Figure 5

Slit lamp-adapted optical coherence tomography and slit lamp imaging of the cornea before (top, corneal thickness 778 μm) and 6 months after Descemet’s stripping automated endothelial keratoplasty (bottom, corneal thickness 555 μm, posterior donor lenticule 77 μm).

Statistical analysis was performed with SPSS (version 15.0, SPSS Inc., Chicago, IL, USA) using t test for paired samples and Pearson correlation; statistical significance was set at p = 0.05.


No intra- or post-operative complications occurred in 14 of 15 eyes (93.3%). Except for one patient who needed a second air injection on day 3 because of partial detachment of the donor lamella, all transplants remained centred and attached on SL-OCT examination. No graft rejection was observed during the follow-up in all our patients. At slit-lamp biomicroscopy, the corneal oedema resolved completely within first 4 weeks.

The mean pre-operative BSCVA was 20/100 (1–logMAR 0.84 (SD 0.63)). The mean BSCVA at 6 months post-operatively was 20/40 (1–logMAR 0.37 (SD 0.18)), and was significantly different from pre-operative BSCVA (p<0.0001, paired samples t test). After DSAEK, all patients (100%) demonstrated improvement of BSCVA, three patients achieved a BSCVA of 20/25 (fig 1).

Figure 1

Pre-operative (preop) and post-operative best spectacle-corrected visual acuity (BSCVA) for all patients in the study (n = 15).

On the first post-operative day, the total central corneal thickness was 1057 (SD 86) μm. During the follow-up, corneal thickness continuously decreased to 661 (SD 74) μm 6 months after surgery (p<0.0001). However, alterations in corneal thickness were only significant within the first 8 weeks and remained stable thereafter (fig 2A). The mean peripheral corneal thickness (3 mm nasally and temporally of the vertex) at the first post-operative day was 1222 (SD 78) μm, compared with 756 (SD 74) μm 6 months after DSAEK (fig 3). The decrease in the peripheral corneal thickness was significant at all follow-up visits (p<0.05).

Figure 2

Changes in central corneal thickness (A) and donor corneal lenticule thickness (B) after Descemet’s stripping automated endothelial keratoplasty (n = 15). Values are means (SD).

Figure 3

Changes in peripheral corneal thickness during follow-up (n = 15). Values are means (SD).

Patients with co-existing ocular pathology such as age-related macular degeneration (AMD), primary open angle glaucoma or mild secondary cataract (n = 5, 33.3%) were not excluded from our analysis. The Pearson correlation coefficient for the 6-month BSCVA and central corneal thickness of patients without co-existing ocular pathology was −0.745 (n = 10, p = 0.021, fig 4A).

Figure 4

Correlation between central corneal thickness and best spectacle-corrected visual acuity (BSCVA) 6 months after Descemet’s stripping automated endothelial keratoplasty (A; p = 0.021) and between grafted corneal lenticule thickness and BSCVA (B; p = 0.05).

At 6 months, the mean endothelial cell density was 1640 (SD 390) cells/mm2, compared with pre-operative cell density of donor corneas of 2309 (SD 135) cells/mm2.

We found no significant correlation between the duration of organ culture and the BSCVA (p = 0.3). There was a significant correlation neither between the duration of organ culture and post-operative central corneal thickness at any examination time point, nor between organ culture duration and the thickness of grafted donor tissue at any examination time point. We found also no significant correlation between the thickness of prepared and grafted corneal lenticule immediately after the surgery and the organ culture duration (p = 0.92).

The SL-OCT allows for selective non-contact measurement of the thickness of the grafted corneal lenticule (fig 5). The average thickness of the donor lamella decreased over time as well. At 6 months after DSAEK the thickness was 100 (SD 38) μm, compared with 191 (SD 56) μm at the first post-operative day (p<0.0001; fig 2B). The correlation coefficient for BSCVA and posterior corneal lamella at 6 months after surgery was −0.589 (p = 0.05, fig 4B).


DSAEK represents one of the most current developments of posterior lamellar corneal surgery. It differs from previously reported techniques by introducing the use of a microkeratome to avoid manual lamellar dissection of donor cornea. As with other forms of posterior lamellar surgery, the advantages of DSAEK include rapid visual rehabilitation, minor subjective discomfort after the surgery, predictable post-operative corneal power and low regular post-operative astigmatism. Although a VA of 20/25 and even 20/20 can be achieved after DSAEK, most of our patients reached 20/40 post-operatively, which is similar to previous reports after posterior lamellar keratoplasty.4 5 In our series, all donor corneal lenticules were clear at 6 months, all eyes showed improved BSCVA, and three eyes even reached a BSCVA of 20/25.

There are many factors that determine the final VA after DSAEK, such as regular or irregular corneal astigmatism, higher-order optical aberrations, corneal haze, endothelial cell density and the regularity of the posterior donor lenticule. Another factor that might influence the DSAEK outcome is the type of storage of the donor corneas. Worldwide, the most common techniques for storing corneas use hypothermic storage media, such as McCarey–Kaufman (M-K; Aurora Biologicals Ltd, Williamsville, New York, USA) or Optisol (Plus, GS; Bausch and Lomb Surgical, Irvine, California, USA) at 2–6°C. In contrast, the majority of European eye banks uses organ culture conservation in for example MEM medium at 30–37°C with addition of 4–8% dextran for dehydration prior to surgery.14 The technique of hypothermic storage is simple and requires no complex equipment. The storage solutions are commercially available. During storage the corneal tissue remains thin and it is directly available for surgery. The modern hypothermic storage systems, such as the modified M-K medium, K-sol (Ciclo Inc., Huntingdon, West Virginia, USA) or Optisol claim a storage period of 14–16 days.14 15 To reduce the risk of primary graft failure, the recommended storage periods are kept far below the claimed maxima (up to 7–10 days for Optisol).16 In addition, the time interval between death of the donor and storage of the cornea (so-called post mortem time) is generally kept within 12 h or shorter.14

The technique of organ culture is relatively complicated. The corneas are stored in an incubator at 30–37°C, supplemented with antibiotics and antimycotics. Dehydrating macromolecules, necessary to maintain normal hydration in vitro, accumulate in the cells and layers of the cornea, leading to swelling of the donor cornea to about twice its normal thickness.17 18 The swelling should be reversed prior to grafting, performed by placement the cornea in the so-called transport medium containing 4–8% dextran.18 The dehydration time varies between cornea banks from less than 1 day up to 7 days. Organ culture allows for an extended storage period of up to 4–5 weeks and yields excellent preservation of endothelial viability.19 We recently demonstrated that a prolonged organ culture storage improves graft survival after low-risk keratoplasty.20

It is known from penetrating grafts that the thickness of organ-cultured corneal tissue continuously decreases following the first 8 weeks after the grafting, although a constant hydration is achieved by the addition of dextran to the culture medium.21 Since it is known from studies investigating penetrating grafts that BSCVA correlates with the graft thickness, we hypothesised that this might be the case in DSAEK as well. Although the patient sample size as well the follow-up period are limited, our study shows that post-operative BSCVA after DSAEK significantly correlates with the central corneal thickness and with the thickness of the grafted lamella 6 months after the surgery. In addition, our study shows that the thickness of the donor corneal lenticule decreases rapidly after DSAEK in an almost linear fashion (fig 2B). During the first 6 months after surgery, the peripheral corneal thickness behaves in similar manner (fig 3). We did not detect a significant correlation between the peripheral corneal thickness and the BSCVA at any follow-up visit. In our study, the mean post-operative endothelial cell density was comparable with previous reported results after DSAEK.5

Since it is not possible to quantify corneal thickness by slit lamp biomicroscopy, SL-OCT is important for the follow-up of patients after DSAEK. With the SL-OCT, the anterior segment of the eye can be easily examined even on the first post-operative day. In a recent study, Li et al12 showed that SL-OCT measurements were reliable and comparable with ultrasound pachymetry. Furthermore, the best agreement was reported between ultrasound pachymetry and manual SL-OCT measurement, which was also used in our study. With SL-OCT it is possible to measure not only the central corneal thickness, but also the peripheral thickness at any required point of the cornea. The graft–host interface can be visualised, as well as irregularities, fibrosis, debris, graft dislocations, Descemet’s detachment or lenticule asymmetry.

There are only few studies available comparing clinical outcome after grafting between organ-cultured and hypothermic storage tissues.22 A recent study by Rijneveld et al23 compared the outcome of corneal grafts preserved in M-K medium versus organ-cultured grafts 14 years after penetrating keratoplasty. No significant differences were observed in VA, endothelial cell density and cell loss. Which storage system optimises the survival and transparency of the lamellar grafted corneal tissue needs to be examined in further studies.24 25


View Abstract


  • Competing interests: None declared (for all authors).

  • Funding: None.

  • Ethics approval: Obtained.

  • Patient consent: Obtained.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.