Outcomes of Refractive Surgery Consultations at an Academic Center: Characteristics Associated with Proceeding (or Not Proceeding) with Surgery

Objective Refractive surgery volume has not rebounded despite economic recovery and literature describing safety, efficacy, and high patient satisfaction. We sought to examine characteristics of consultation seekers and status after consultation. Methods Charts of patients seeking refractive surgery at Johns Hopkins University from 2013 through 2016 were retrospectively reviewed for age, gender, refractive characteristics, and outcome: surgery (photorefractive keratectomy, laser in-situ keratomileusis, implantable collamer lens, or refractive lens exchange); no surgery—“lost candidate” (good candidates who were lost after consultation); noncandidates based on technological limitations or contraindications; or no surgery—possessing expectations that surgery would not meet. Associations between characteristics and status after consultation were examined. Results Twenty percent (142/712) of all patients were “lost candidates”; 57% (408/712) completed surgery. More women (56% or 401/712) sought consultation, but a greater percentage (63% or 195/311) of men completed surgery than women did (53% or 213/401) (p=0.02). Of consultation seekers, 60% were low myopes, 29% were high myopes (>6 diopters of myopic spherical equivalent), and 11% were hyperopes. Surgical patients' mean age was 34.2 ± 10.2 (standard deviation) years; for each additional year of age, patients were less likely to have surgery (p < 0.001). Hyperopes were ≥3 times more likely than myopes to have expectations not met by surgery or to be noncandidates than to have surgery (p < 0.005). Conclusions Most patients seeking refractive surgery had 6 diopters or less of myopia. About 20% of patients were lost after consultation; better counseling and follow-up of candidates may be warranted. Expectations and technology limit eligibility for many, especially hyperopes. Low surgery volume may affect training of future refractive surgeons.


Introduction
In recent years, the volume of refractive surgery (notably photorefractive keratectomy (PRK) and laser in-situ keratomileusis (LASIK)) has dropped from approximately 1.4 million a year in the United States in the late 1990s-early 2000s to currently 650,000 a year despite documented outcomes of improved quality of life and high patient satisfaction worldwide, comparable to or exceeding other elective procedures [1][2][3]. By some estimates, under 1% of adults with refractive error in the US undergo refractive surgery each year, implying many more adults are eligible for surgery.
While the 2008 economic recession probably had an impact, other factors, such as demographic changes and increased electronic device use, may be contributing to low volume and lack of rebound with economic recovery. In the time since a publication on trends in refractive procedures [4], small incision lenticule extraction (SMILE) and implantable collamer lenses (ICLs) have been adopted and would be expected to increase interest in refractive surgery over all. e purpose of this study was to analyze characteristics of individuals seeking refractive surgery consultation at an academic institution in order to understand patient wants and needs, to ascertain characteristics of patients who proceed (or do not proceed) to surgery, and to understand where our evaluations or procedures fall short.

Materials and Methods
e Institutional Review Board of Johns Hopkins University granted approval for the protocol under which this chart review was conducted (IRB00094392).
is review was conducted in accordance with the tenets of the Declaration of Helsinki.
e Wilmer Eye Institute of Johns Hopkins University maintains a record of candidates seeking refractive surgery consultation at the satellite clinic housing the laser suite. All candidates undergo a thorough preoperative evaluation, including medical, ophthalmic, and social history; preoperative uncorrected and best spectacle-corrected distance and near visual acuities; Scheimpflug imaging (Pentacam; Oculus, Arlington, WA); biomicroscopic and dilated examinations; pupillary examination; Schirmer testing; pachymetry; and manifest and cycloplegic refractions. Manifest refraction of myopic candidates is aimed at the least minus refraction to see the most Snellen letters; hyperopic patients are pushed with the most hyperopic refraction, often confirmed with cycloplegic refraction and/or relaxation of accommodation over time wearing an updated prescription glass in cases of latent hyperopia. In all situations, manifest refraction, cycloplegic refraction, and wavefront refraction are compared and used to determine the eventual refractive treatment. e option of glasses and/ or contact lenses is always discussed. All PRK and LASIK were wavefront-guided using the VISX Star S4 CustomVue (Johnson and Johnson Vision, Santa Ana, CA) with eye tracking and iris registration where possible. Patients seeking presbyopia-correcting lens implants were not included in this study; presbyopic inlays are not performed at our institution.
Pertinent characteristics of patients seen between January 1, 2013, and December 31, 2016, were age, gender, manifest refraction, and status after consultation; study inclusion required adequate documentation of the above. Refractive error was categorized by the mean spherical equivalent of the manifest refraction (MRSE) of a patient's two eyes: "low myopia" for 6 diopters (D) of myopia or less, "high myopia" for >6 D of myopia, and "hyperopia" for any hyperopic error.
ere were four status categories: (1) surgery, (2) no surgery-"lost candidate" (good candidates who were lost after consultation), (3) no surgery-expectations would not be met, and (4) noncandidate. Surgery type (PRK, LASIK, ICL, and refractive lens exchange) and correction (full vs. monovision variant) were recorded. e four categories were chosen prior to chart analysis. An individual was a "lost candidate" if he or she was deemed a good surgical candidate and did not express concerns about cost, risk, or other factors, yet did not proceed to surgery. Subcategories were developed retrospectively as certain patterns emerged after review of the first few hundreds of charts. "Expectations would not be met" included "seeking presbyopic correction" and "others" (e.g., patient expressed the cost of surgery exceeded perceived benefit). "Noncandidate" consisted of "ophthalmic contraindication" (e.g., ectasia risk), "technological limitation" (e.g., high hyperopia or corneal astigmatism), and "not treatable by refractive surgery" (e.g., cataract, amblyopia). At this site, most consultations are performed by optometrists who refer suitable candidates to the surgeon. e refractive coordinator attempts to arrange a brief meeting with the refractive surgery chief the same day. Otherwise, another surgeon is found, or the patient is asked to meet a surgeon another day.
For the univariate analysis, patient characteristics were compared by status after consultation. Patient age was compared using one-way analysis of variance (ANOVA). Chi-square test was used to determine whether gender and refractive categories were related to status after consultation. Multinomial logistic regression was performed to examine associations between patient characteristics (refractive category, gender, and age) and status after consultation; relative risk ratios (RRRs) with 95% confidence intervals (CIs) were presented. Potential interaction terms among age, gender, and refractive categories were examined for associations. In analyses not involving exact age, patients were grouped by 21-39, 40-59, ≥60 years of age to reflect onset of presbyopia, and other age-related eye conditions. A p value ≤0.05 was considered statistically significant for all analyses. All data were analyzed using STATA software version 14.2 (Stat-Corp, College Station, TX).

Results
Of the 868 patient charts reviewed, 712 (82%) met the inclusion criteria and were included in the analysis. e main reason for exclusion was lack of electronic charting (adopted in 2013) and lack of access to paper charts.
A higher proportion of consultation seekers were female than male (p � 0.012) ( Table 1). e majority of consultation seekers were 21-39 years of age (p � 0.001). Young women formed the largest of the 6 age-and gender-based subcategories. e smallest category was men ≥60 years.

Discussion
e downturn in refractive surgery despite high patient satisfaction [1][2][3] warrants examination of possible factors. We noted some patient characteristics and consultation outcomes, such as low myopes, forming the largest group of seeking refractive surgery yet 20% being good candidates who did not proceed to surgery and young men being the most likely to complete surgery although young women were the largest group seeking consultation. Elucidation of such characteristics and outcomes may help improve consultation quality. Although other papers have investigated reasons for surgeons not performing LASIK in certain candidates [5][6][7], our paper describes nonsurgical patients in terms of other than medical contraindications (e.g., keratoconus, high myopia, high astigmatism, and thin corneas) [5][6][7] and ascribes some consultation outcomes partly to patient, optometrist, and surgeon behavior/choice. As a substantial percentage of good candidates was lost, this review uncovered opportunities to improve the consultation process; more surgeon involvement and postconsultation survey may be needed.
Although tens of millions of patients worldwide have had successful LASIK, reports of highly dissatisfied patients led to a review by the US Food and Drug Administration [8]. Shortly thereafter, a publication on the peer-reviewed literature from 1988 to 2008 found that >95% of post-LASIK patients were satisfied, which compared as favorably as (or more favorably than) other elective procedures like rhinoplasty, botulinum toxin injections, and breast augmentation [2]. With eye-tracking, improved patient selection, femtosecond laser, and customized ablations, articles from 2008 to 2015 found fewer (1.2%) patients reported dissatisfaction [3].
One paper [9] found that 22% of patients seeking refractive surgery at a university-based center between 2005 and 2010 "changed their mind" after surgical consultation. e main reason was the perception that the risks of surgery outweighed its benefits; the authors noted that this group shrunk over time, possibly because of improved technology and perception of decreased risk. Another paper described cost or "unscientific apprehension" as the reason 15% of patients did not proceed with LASIK [10].
At first glance, our categories may appear arbitrary; one might wonder whether patients who were lost despite good candidacy were possibly the same people whose expectations would not be met. We found these two groups were different in age and type of refractive error: patients whose expectations would not be met had a mean age of 47.4 ± 12.1 years and consisted of more hyperopes, in contrast with lost candidates who were younger (38.3 ± 12 years) and myopic. More than half the patients whose expectations would not be met desired presbyopic correction. Our review showed that a monovision trial was rarely offered; only 9% of surgical patients opted for full or mini-monovision with fewer than half of 40-59-year-old patients (for whom presbyopia was the most common complaint) and 20% of those over age 60 proceeding to surgery.
A substantial proportion (20%) of consultation seekers was lost despite good candidacy and lack of stated reservations about surgery; 58% of lost candidates were low myopes. It is possible, however, that cost, though not stated, was an issue for some "lost candidates." is is particularly true for ICL and refractive lens exchange, which cost >4 times higher than PRK or LASIK (which cost the same) because of anesthesiologist and surgery center facility fees. erefore, while some ICL and refractive lens exchange candidates stated they would not proceed because cost and risk were higher than for LASIK, some "lost" candidates might just not have voiced the same concerns. After the recession, our institute reduced the price of excimer laser surgery by 20% with no subsequent volume increase. All patients who undergo refractive surgery consultations, whether determined to be good candidates or not, are reminded of the option of glasses and/or contact lenses; two patients (keratoconus and corneal ectasia, both in the noncandidate category) were referred for the first time for specialty contact lens fitting. Many "lost" candidates likely returned to contact lenses or spectacles. Hyperopic patients formed the largest proportion of noncandidates and were older than myopes, consistent with the increasing prevalence of hyperopia with age [11]. Hyperopes formed the smallest group of refractive categories seeking consultation, completing surgery, or being lost candidates. Low prevalence alone (9.9% in population-based studies) [11] may limit impetus to expand hyperopic treatment offerings.
Although more women than men sought consultation (possibly related to population-based findings that "severe myopia" (i.e., ≥5 D) is significantly more prevalent in women than in men) [11,12], male gender was associated with surgery completion. Men were less likely to be noncandidates from a medical or ophthalmic condition like dry eye. Gender differences in communication, in perception of surgery risks, and in vocational needs (e.g., more men than women serve in the military or are first responders) may be other reasons that more men complete surgery.
One might expect economic downturns to affect volume of elective procedures more so than of other surgical procedures because of high out-of-pocket cost and patient selectivity. Surprisingly, the decrease in blepharoplasty, liposuction, rhinoplasty, breast augmentation, and rhytidectomy was less than expected-about the same as the decrease in angioplasty, breast and pancreatic cancer surgery, and hip/knee arthroplasties [13]. It is unclear whether LASIK is purely cosmetic although it compares similarly inpatient satisfaction and quality of life [2]. Hand-held device use, ride-sharing services, and toric and bifocal contact lenses may be of decreasing interest in surgery. Refractive surgeons, however, could spend more effort at patient education; the public may be unaware of advances, base decisions on (negative) lay press [14] or on an outdated consultation, or rely on optometric advice.
Last, the slowdown in refractive surgery volume may have unforeseen consequences. Conclusions drawn from an academic center may not be universally applicable, but such centers train a large fraction of refractive surgeons. Without addressing points raised herein-patient education, outreach, and follow-up-low volume may adversely affect refractive surgery training [15].

Conclusions
Cost cannot wholly explain the decline in refractive surgery volume; other elective procedures with high out-of-pocket costs are not seeing declines of this magnitude and duration. Our study may lead surgeons to alter current practices regarding education, evaluation, outreach, and follow-up of refractive surgery candidates. For example, more patients might choose monovision rather than forego surgery if they had known about it although monovision alone will not address demand for presbyopic correction. Education (including updating patients on options for treating extreme refractive errors), careful examination and assessment, and communication with the patient even after the visit are key components of a refractive surgery consultation. Such emphasis may generate interest and improve the environment for training future refractive surgeons.
Data Availability e datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Disclosure
Dr. Lee is now in the Department of Ophthalmology, Lousiana State University, New Orleans, LA.

Conflicts of Interest
e authors declare that there are no conflicts of interest.