Answers are grounded in this paper and cite their sources.
Sir, Clinically significant iatrogenic keratectasia is a rare but increasingly recognized complication following laser in situ keratomileusis (LASIK), particularly in highly myopic eyes with thinner corneas (Kohnen 2002). In many instances this can markedly compromise visual outcome and hence is a cause for concern. Excimer laser photorefractive keratectomy (PRK) has also been shown to induce steepening of the posterior corneal surface, but without any clinically apparent keratectasia (Naroo & Charman 2000). We report a case of posterior corneal ectasia over a period of 6.5 years following repeated myopic PRK, evaluated using scanning slit-section topography and pachymetry. In January 1994, a 47-year-old man underwent excimer laser PRK under another ophthalmologist's care at a different unit, to correct a refractive error of − 9.75 dioptres (D) in the right eye, using a 5-mm ablation zone. There was no clinical evidence of corneal ectasia (keratoconus) preoperatively, although corneal topography and pachymetry were unavailable. Two months after treatment, the subject's right uncorrected visual acuity (UCVA) was 6/9, improving to 6/5 with a pinhole. Four months later, marked corneal haze was noted, reducing his right UCVA to 6/36, improving to 6/5 with a pinhole. In April 1995, corneal haze was still present, although his right VA was 6/18 with a − 5.0 D sphere. A further PRK was performed at another unit, with a target correction of − 4.5 D and a 6-mm ablation zone, resulting in a right UCVA of 6/18. Neither pachymetry nor topography were available preoperatively at this centre. The subject attended our unit for a second opinion in June 1996, complaining of a white opacity in his right cornea. On examination, his left VA was 6/4 with a − 9.5 D soft contact lens and his right UCVA 6/12, correcting to 6/6 with − 2.5 D/− 1.0 D × 160. His right cornea showed marked corneal haze. He remained stable until reviewed in March 1999, when we noted a right UCVA of 6/36, correcting to 6/6 with − 10.0 D/− 1.50 D × 180. He was fitted with gas-permeable contact lenses and remained satisfied visually. When reviewed in July 2000, the subject's right VA was 6/12 with a correction of − 9.50 D/− 2.5 D × 10. Scanning slit-section topography was performed using the Orbscan topography/pachymetry system (Orbscan II; Orbtek Inc., Bausch & Lomb, Rochester, New York, USA). The right cornea showed a mean power of 51.6 D in the 3.0 mm zone and a central thickness of 350 µm (Fig. 1). Elevation of the posterior corneal surface in comparison with a theoretical best-fit sphere revealed a marked posterior ectasia of 75 µm (Fig. 2). The left cornea was topographically normal, with a central corneal thickness of 620 µm and no evidence of keratectasia (Fig. 3). Slit-section topography/pachymetry of right cornea.From top left, clockwise:anterior and posterior corneal height profiles,pachymetry (0.35mm centrally) and keratometry (51.6D in the 3.0mm zone). Elevation map of the right posterior corneal surface, in relation to a theoretical best fit sphere, showing 0.075mm posterior ectasia. Slit-section topography/pachymetry of left cornea, showing no evidence of keratectasia and a central corneal thickness of 0.62 mm. To the best of our knowledge, this would appear to be the first report of posterior keratectasia followed over a period of 6.5 years after unilateral repeated myopic PRK. Severe corneal ectasia has been reported in one patient 18 months after bilateral myopic PRK, although this may have represented an occult case of keratoconus or forme fruste keratoconus (Shimmura et al. 1997). This may also explain the one case of keratectasia reported after phototherapeutic keratectomy (PTK) for band keratopathy (Miyata et al. 2001). Importantly, the fellow eye of our patient showed no signs of corneal ectasia, rendering the possibility of underlying corneal ectasia in the treated eye unlikely, as unilateral keratoconus is extremely rare (Holland et al. 1997) However, we cannot categorically exclude this possibility due to the inadequacy of data from the referring unit. Iatrogenic keratectasia is increasingly recognized following LASIK, especially in thinner corneas of high myopes, where the residual bed thickness is less than 250 µm (Amoils et al. 2000; Kohnen 2002). Loss of biomechanical corneal strength is considered a major factor, as the anterior flap forms a relatively weak attachment to the posterior lamella following LASIK. This may be compounded by a deregulation of keratocyte apoptosis, leading to altered proteolysis and wound healing dynamics (Comaish & Lawless 2002). Both these mechanisms are likely to have contributed to the development of keratectasia in our patient following repeated PRK treatment. Scanning slit-section topography/pachymetry takes multiple slit-section images of the cornea and reconstructs anterior and posterior corneal height profiles using 3-dimensional ray tracing (Naroo & Charman 2000). This method enables measurement of anterior and posterior corneal curvatures, corneal thickness profile, anterior chamber depth and total corneal power. We therefore advocate its use in highly myopic patients prior to either PRK or LASIK, to detect underlying ectasia and thin residual corneas that may develop greater postoperative corneal bulging and refractive regression. Patients with extremely thin corneas possibly run a higher risk of developing keratectasia and should be approached cautiously. In eyes which have previously undergone LASIK, Orbscan pachymetry tends to underestimate corneal thickness, and in these cases ultrasound pachymetry might prove more accurate (Chakrabarti et al. 2001). Other methods of measuring corneal thickness may also be incorporated, including partial coherence interferometry, confocal microscopy and optical coherence tomography (Kohnen 2002). Nonetheless, the Orbscan topography/pachymetry system clearly provides excellent imaging of the posterior corneal profile and facilitates the detection of keratectasia – as was usefully highlighted in our patient.
No takes yet. Share an insight, caveat, or question.
Parmar et al. (2004) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: