Early Experience with Descemet

1
Early Experience with Descemets Stripping
Automated Endothelial Keratoplasty Combined with
Phacoemulsification Clinical and Refractive
Outcome

• University of Texas Southwestern Medical Center
• Department of Ophthalmology
• Marvin Hsiao, MD, Pawan Prasher, MD, R. Wayne
  Bowman, MD, James P. McCulley, MD, V. Vinod
  Mootha, MD
• No financial relationships involved in this
  presentation.

2
Purpose

• Descemets stripping automated endothelial
  keratoplasty (DSAEK) has recently gained
  popularity as an alternative to the full
  thickness penetrating keratoplasty (PK) for
  treatment of endothelial dysfunction. This is
  due to the shorter healing time and the advantage
  of avoiding a full thickness graft offered by the
  DSAEK procedure.
• When a cataract coexists with the corneal
  endothelial dysfunction, either a combined or
  sequential procedure for keratoplasty and
  cataract extraction is indicated.
• For the combined procedure of PK, cataract
  extraction, and intraocular lens (IOL) implant
  (PK triple), the resultant post-operative corneal
  keratometry is difficult to predict, leading to
  problems with IOL power determination and
  post-operative refractive surprises.

3
Purpose

• In DSAEK, previous studies have shown that the
  post-op corneal keratometry is much more
  predictable than PK, although usually associated
  with a hyperopic shift in refraction.
• A pilot study has also shown that when combining
  DSAEK with phacoemulsification and IOL
  implantation, the refractive result can be
  predictable, along with rapid visual recovery. In
  this study we present our early experience with
  the combined phaco/DSAEK procedure.

4
Methods

• The study was a retrospective chart review of
  noncomparative surgical cases.
• Patients who underwent combined
  phacoemulsification, IOL implantation, and DSAEK
  for treatment of cataract and corneal endothelial
  dysfunction were included in the study.
• Patients with co-morbid conditions that preclude
  post-op best-corrected visual acuity (BCVA) of
  better than 20/60 were excluded from the study.
• Preoperative measurements of axial length,
  keratometry, anterior chamber depth, and
  white-to-white by IOL Master were recorded for
  each eye.
• Outcome measures included IOL power implanted,
  BCVA and manifest refraction at post-operative
  months 1, 3, and 6 when available.

5
Methods

• Due to the observation of hyperopic shift
  associated with DSAEK, the IOL power implanted
  was increased from the calculated emmetropic
  value to account for this observation.
• The true emmetropic IOL power based on post-op
  refraction was back-calculated and compared to
  the calculated emmetropic IOL for each eye based
  on either the SRK/T or the Holladay II formula.
  This comparison generated an IOL error for each
  eye.
• In 3 eyes, preoperative keratometry was not
  possible due to significant epithelial edema. In
  one of these eyes, keratometry from the
  contralateral eye was used. For the other two
  eyes, keratometry from the contralateral eye was
  not possible, and a keratometry value of 44.0 D
  was used.

6
Results

• A total of 13 eyes of 11 patients that underwent
  combined phaco/DSAEK were included in the initial
  stages of the study. There were 4 males and 7
  females. Average age at the time of surgery was
  64.
• Post-op month 1 BCVA ranged from 20/25 to 20/200
  (average log MAR 0.38, 20/501), with spherical
  equivalent (SE) refractive error ranging from
  -1.625 D to 3.375 D (average 0.78 ? 1.17 D).
• Post-op month 3 BCVA ranged from 20/20-1 to
  20/50-2 (average log MAR 0.21, 20/32-), with SE
  ranging from -0.75 D to 2.50 D (average 0.68 ?
  0.83 D).
• Post-op month 6 BCVA and refraction are still
  partially pending.

7
(No Transcript)
8
Results

• Based on post-op month 3 or month 6 (if
  available) refraction, the true emmetropic IOL
  power was calculated by using the formula
• IOL emmetropia IOL implanted (post-op SE x
  1.4)
• This true emmetropic IOL power was then compared
  not to the IOL power implanted, but to the
  emmetropic IOL power predicted by either the
  SRK/T or the Holladay II formula.
• IOL error IOL emmetropia (back calculation)
  IOL emmetropia (predicted)
• IOL error ranged from 0.25 D to 3.80 D (average
  2.33 ? 1.04 D), illustrating the hyperopic shift
  with DSAEK involving all eyes in this study.

9
(No Transcript)
10
Conclusion

• The advantages of DSAEK include faster visual
  recovery, avoiding complications inherent to a
  full thickness corneal button such as wound
  dehiscence and suture related problems, limited
  amount of surgically induced corneal astigmatism,
  and the potential to either repeat the DSAEK
  procedure for graft failure versus subsequent PK.
  
• In this study, visual recovery is rapid, with
  over 75 of the eyes achieving vision better than
  20/40 at post-op month 3, and 100 better than
  20/60.
• Disadvantages of DSAEK include graft dislocation,
  pupillary block, possible secondary glaucoma,
  endothelial cell loss, and a graft-host interface
  that may limited final BCVA.

11
Conclusion

• In PK, the unpredictable post-operative
  keratometry may result in refractive errors
  leading to significant anisometropia, irregular
  astigmatism, and the need for contact lens
  correction. Visual recovery with PK is a lengthy
  process, with suture and wound issues adding to
  the increased risk of infection and wound
  instability.
• The hyperopic shift in post-operative refraction
  associated with DSAEK necessitates implanting a
  higher IOL power than predicted by IOL formulas
  to avoid a hyperopic result.
• In this study, the average IOL error is 2.33
  diopters with standard deviation of 1.04
  diopters. Despite increasing the IOL power
  implanted, 10 out of 13 eyes (76.9) still
  resulted in a hyperopic refraction at post-op
  month 3.

12
Conclusion

• This hyperopic shift is thought to be from the
  donor graft decreasing the posterior radius of
  curvature of the cornea due to its concave shape,
  which in effect decreases the refractive power of
  the cornea as a whole.
• It would be advantageous to be able to predict
  the amount of hyperopic shift based on pre-op
  data. More eyes would be needed than available
  in this study to conduct a regression analysis to
  find this relationship.
• Whether DSAEK will eventually prove to be the
  preferred choice of keratoplasty for corneal
  endothelial dysfunction, the rapid recovery and
  more predictable post-operative refractive
  outcome offers an exciting new method of
  combining keratoplasty with cataract surgery.