During the 19th century, Snellen (1869) (1), Schiötz (1885) (2), Bates (1894) (3), and Lans (1898) (4) described methods to reduce astigmatism. In 1939, Sato (5) described spontaneous breaks in Descemet’s membrane in keratoconus that caused flattening of the cornea and reduction of myopia. In 1953, Sato and colleagues (6) suggested that by making radial incisions in the posterior as well as the anterior cornea, astigmatism and myopia could be treated. Sato’s surgical technique required entry into the anterior chamber at three to four different sites. Incisions were then made using a “free hand” technique through the endothelial surface, from the edge of a 6-mm central optical zone through 50% of the corneal thickness. Additional anterior incisions, placed in between the posterior ones, were then made at approximately the same depth. There is no indication that pachymetry was performed. Yamaguchi and associates (7) have estimated that 40% to 50% of the endothelial surface was damaged at the time of surgery. Approximately 75% to 80% of the 200 patients who underwent Sato’s procedure developed visually significant corneal edema an average of 20 years after the operation.

In the early 1970s, in the former Soviet Union, Yenaliev (8), Fyodorov, Agranorsky, and Durnev (8, 9, 10, 11) modified Sato’s technique, performing anterior radial incisions in the peripheral cornea and sclera. In 1978, Bores, Myers, and Cowden (12,13) reported the outcomes of the first RK procedure in the United States, using Fyodorov’s techniques (8,10).

In 1980, the National Eye Institute funded the Prospective Evaluation of Radial Keratotomy (PERK) Study to assess the efficacy, safety, predictability, and stability of RK (14). The PERK study used rigid, standardized, surgical protocols, data collection by independent observers, and a multicenter design. It was a collaborative effort among university-based and private ophthalmologists in nine clinical centers. The protocol did not allow corrections for astigmatism. The surgical plan was based on the amount of myopic error and no other preoperative factors such as age, sex, or intraocular pressure were addressed (14, 15, 16).

The surgical protocol of the PERK Study required eight freehand centrifugal incisions (clear zone to limbus) made with the length of a 45-degree diamond blade set using ultrasonic pachymetry at 100% of the thinnest paracentral corneal measurement (16). Preoperatively, optical zones were limited to 3, 3.5, or 4 mm, corresponding to low-, medium-, or high-myopia levels (range −2 to −8 D). Repeat surgery for enhancement (eight additional incisions were made between the initial eight) was carried out under specific guidelines (15,17).

The RK postoperative outcome in the PERK study showed a decrease in myopia in all patients: 88% of patients achieved uncorrected visual acuity of 20/40 or better; 90% of patients with myopia of 6 D or less achieved visual acuity of 20/40 or better. Reduction of best-corrected visual acuity (20/25) was noted in 0.8% of patients.

Throughout the decade following the initiation of the PERK study, numerous advances were introduced through the work of various surgeons (Table 70-1) (18, 19, 20, 21, 22, 23, 24, 25, 26). Two principal techniques for RK evolved: (a) American style (centrifugal), or backcutting, in which incisions are made from center to periphery; and (b) Russian style (centripetal), or frontcutting, in which incisions are made from the limbus to the edge of the desired optical zone. It has been shown that centrifugal incisions inherently yield a diminished range of corneal flattening, because incision depth for a given diamond setting is less reliable than that with a centripetal incision (27, 28, 29).

From these two approaches evolved the Genesis and Duotrack (or “combined”) techniques, which combined the safety of the centrifugal incision with the efficacy of the centripetal incision (24,30). In the combined technique, the centrifugal component provided the surgeon with a relatively shallow and linear safety groove. The centripetal component further deepened the incision groove and undercut the deep stroma adjacent to the optical zone, increasing the range of corneal flattening (Fig. 70-1) (30,31). The diamond knife blade employed for the combined technique enabled the surgeon to begin the incision at the optical (clear) zone, extend out toward the limbus, and then reverse direction back toward the central optical zone without optical zone invasion. The angled edge on the back surface (centrifugal cutting component) is sharpened along the entire length, and the vertical edge of the front surface is ground sharp to a cutting edge for a distance of only 200 to 250 μm from the blade tip (Fig. 70-2). The blunt upper portion of the vertical knife edge reduced the chance of unwanted invasion of the central optical zone when the blade tip is initially inserted into the cornea at the optical zone margin (32). In a comparative study using donor eyes (31), the combined (Genesis) technique demonstrated greater corneal flattening than the centrifugal (American) method of incision and reduced variability when compared with the centripetal (Russian) method.

As experience with RK increased, a tendency for shorter incisions developed (26,33, 34, 35, 36). In a cadaver eye study, Jester and associates (34) found that incisions extending beyond the limbus reduced the efficacy of RK. Salz and colleagues (35) found a small decrease in efficacy when using a shorter incision, but a significant effect was maintained. This is consistent with the results from Lindstrom’s (26) modified technique for low to moderate myopes, known as minimally invasive RK (“mini-RK”), which reduces the length of cornea incised. In a cadaver eye study, eight short incisions were performed from the 3-mm to the 7-mm optical zone (Fig. 70-3), producing 92% of the efficacy of full-length incisions made to the 11-mm optical zone. Pinheiro and co-workers’ (36) eye bank study also showed that reducing the incision length reduced the likelihood of corneal rupture as intraocular pressure is increased.

Approximately 25% of the adult population in the United States is myopic (37,38). It has been reported that the majority of candidates seeking keratorefractive surgery state their goal as good visual function without physical dependence on spectacles or contact lenses (39,40). Occasionally, patients may have medical (e.g., contact lens intolerance),
occupational, recreational, or cosmetic reasons for seeking surgery.

RK should be limited to patients with low to moderate myopia and an otherwise normal eye examination (24,41,42). Higher degrees of myopia have variable responses to RK, and procedures such as photorefractive keratectomy and laser-in-situ keratomileusis (LASIK) may yield better and more stable results. For extreme high myopes, clear lens extraction or phakic intraocular lens implantation may be more appropriate. In the case of the RK candidate, myopia, and astigmatism must be stable and nonprogressive (43). Postoperative corneal transplant and cataract surgery patients with residual myopic astigmatism may benefit from incisional keratotomy after all sutures are completely removed and refractive stability documented. Table 70-2 outlines the appropriate surgical procedures for the levels of severity of ametropia.