Calculation of the Real Corneal Refractive Power after Photorefractive Keratectomy Using Pentacam, When Only the Preoperative Refractive Error is Known

Purpose To check if a regression formula, IOLMaster-derived, to calculate the real corneal power after photorefractive keratectomy (PRK), can give reliable results utilizing the Pentacam. Methods Pre- and postoperative IOLMaster, Km, and Pentacam K readings were measured. Patients who had myopic PRK were divided into two groups: the first group (108 eyes) was utilized to check which of the preop Pentacam K readings (P-Kpre) better fitted with the preop IOLMaster measurements; in the second group (120 eyes), the real K (Kr), obtained adding the effective treatment to the P-Kpre, were compared with the K readings calculated with the IOLMaster-derived formula (Kc). Moreover, an attempt to find a different formula utilizing the P-Kpre was made. Results In group 1, the best correlation was found between IOLMaster Km and Pentacam equivalent K readings (r2 0.9519). In group 2, the comparison between Kr and Pentacam postop Km showed 69 eyes (57%) with differences >0.5 D and 38 eyes (31%) with differences >1 D, (P < 0.001). The comparison between Kr and Kc showed 55 eyes (45%) with differences >0.5 D and 22 eyes (18%) with differences >1 D, (P < 0.001). Moreover, a regression formula K = EKR − [ETcp + (0.8114 ∗ ETcp − 0.2031)] was obtained in order to calculate the K readings to be used with the Pentacam in the IOL power calculation in case the effective treatment is known. Conclusions K calculated with the new formula could be used in patients that underwent refractive corneal surgery in case a Pentacam device is used, pending further studies conducted in clinical practice to establish its accuracy and effectiveness. This study further proves that data obtained from different machines cannot be used interchangeably.

For this reason, in case of cataract surgery, the power of the IOL will be underestimated, and the patient will be hyperopic with the need of an IOL exchange or a piggyback lens [3,4].
Many methods have been described to calculate IOL power after refractive surgery procedures, and they can be mostly divided into two groups depending on the availability of preoperative and postoperative data.
In the literature, some papers suggest a strong correlation between IOLMaster and Pentacam K readings in nonoperated eyes [5,6,7].

Materials and Methods
is retrospective clinical study comprised consecutive patients who had PRK for myopia or for myopic astigmatism. e study was conducted in adherence to tenets of the World Medical Association's Declaration of Helsinki. Institutional review board approval was obtained, and informed consent was obtained from all individual participants included in the study. e exclusion criteria, the surgical procedure, and the postoperative treatment were the ones we routinely used in these patients, as described in previous papers [24,26].
Before and 6 months after PRK, all patients had a complete ophthalmic examination, including automatic K measurements with a rotating Scheimpflug anterior segment imaging (Pentacam, Oculus Optikgerate GmbH, Wetzlar, Germany, version 1.17r20) and an IOLMaster 500 (Zeiss, Jena, Germany, version 5.4.4.00006) evaluation. e first step was to test before surgery which of the Pentacam parameters better fitted with the IOLMaster measurements.
In the second step, in the patients that reached a sixmonth follow-up, the real K (Kr) were obtained adding the effective treatment, calculated at the corneal plane, to P-Kpre.
Before and after PRK, the effective treatment was converted at the corneal plane, with the vertex distance correction equation [25].
e Kr were then compared with the postoperative Pentacam equivalent K readings (EKR) at 4.5 mm that is considered to be the optimal zone sample size [27] and with the calculated K readings (Kc) with the previously published formula, found with an IOLMaster.
Kc were calculated adding ΔK to the P-Kpre. Moreover, an attempt was made to try to duplicate the previous work to see if it was possible to find a different formula that could be used with Pentacam data utilizing the P-Kpre [25]. e refraction and the keratometric analyses were performed by 2 independent observers (NR and MDB). e data normality was tested with Kolmogorov-Smirnov test.
e correlation between the different K values was assessed by linear regression analysis, and Bland-Altman plots were utilized to analyze the agreement of the measurements provided by the two devices.

Results
e study included for the first step 108 eyes of 54 patients (group 1) (22 women) with a mean age of 32.6 years (SD 8.77) (20-54 years) and a mean preoperative spherical equivalent refraction of −4.7 D (SD 2.35) (−14.5 to −0.5 D). e best correlation between the IOLMaster Km and the Pentacam data was obtained with the P-Kpre (Figure 1).
For the second step, 120 eyes of 60 patients (group 2) (32 women) with a mean age of 33 years (SD 8.9) (19-55 years) and a mean preoperative spherical equivalent refraction of −5 D (SD 2.23) (−14.5 to −0.5 D) were utilized to test the formula. e postop EKR, Kr, and Kc values are shown in Table 1. e comparison between Kr and postop EKR (Figures 2(a) and 2(b)) showed a statistically significant difference (P < 0.001) with 69 eyes (57%) presenting differences ≥0.5 D and 38 eyes (31%) with differences ≥1 D, leading to roughly a similar error in the IOL power calculation. e comparison between Kr and Kc (Figures 3(a) and  3(b)) showed a statistically significant difference (P < 0.001) with 55 eyes (45%) presenting differences ≥0.5 D and 22 eyes (18%) with differences ≥1 D, leading to roughly a similar error in the IOL power calculation.
ese results show that both EKR and Kc should not be utilized to calculate the IOL power after refractive surgery.
Moreover, in the attempt made to duplicate the work performed with the IOLMaster (Figure 4), we found the following regression formula to calculate the K readings to be used with the Pentacam in the IOL power calculation, in case the effective treatment is known:

Discussion
e influence of refractive surgery on the ocular parameter evaluation, such as intraocular pressure and corneal power, has been widely studied [28]. ree main reasons have been claimed to explain the overestimation of the corneal power after refractive surgery: inaccurate measurement of the anterior corneal curvature by automated and manual keratometry (K) or computerized videokeratography, inaccurate value of the keratometric index resulting from the modified relationship between the anterior and posterior corneal surface, and incorrect estimation of the effective lens position (ELP) resulting from these modifications [8,9]. Effective treatment at the corneal plane ΔEquivalent K readings In patients that underwent refractive surgery, if the achieved correction and the preoperative K readings are known, it is possible to calculate Kr.
Unfortunately, if the achieved correction is known and the preoperative K readings are unknown, it is not possible to calculate Kr because the difference in K readings does not correspond to the one detected by the machines [1,3]. Rosa et al., in 2004, studying the reliability of the IOLMaster in measuring corneal power after photorefractive keratectomy, found that this device did not accurately reflect the effective induced refractive changes, particularly in eyes that had high dioptric treatment, and proposed a regression formula which tried to overcome such a problem calculating the real refractive power, when the effective treatment is known, but the preoperative K readings are missing [25].
In the present paper, we demonstrate that this formula cannot be used with Pentacam data, further proving that, in patients that underwent refractive surgery, different devices provide different measurements, and the proposed formulas to overcome the problem of the underestimation of the corneal power after such a surgery cannot be used for all the devices.
In conclusion, K calculated with the new formula, in patients undergoing cataract surgery, could be used in patients that underwent refractive corneal surgery, in case a Pentacam device is used. Further studies conducted in clinical practice will be necessary to establish the accuracy and effectiveness of this new formula.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors report no conflicts of interest and have no proprietary interest in any of the materials mentioned in this article.