The Effect of Breastfeeding on Photorefractive Keratectomy Outcomes: A Retrospective Cohort Study

Purpose To assess the refractive results, visual function, and safety of photorefractive keratectomy (PRK) in myopic breastfeeding women and compare them to healthy matched female controls. Methods Twelve mothers (23 eyes) who underwent PRK while they were still breastfeeding (B group) and continued it for at least three months participated in this retrospective matched cohort study. Twelve women (23 eyes) who were matched for age, refractive error, and operation data (NB group) were selected as the control group. Post-PRK results and complications were compared between the two groups. Results Twenty four breastfeeding mothers with a mean age of 29.70±1.8 (Standard Deviation) (range from 27 to 33years) were included. The average age in the study and the control group were 29.83 ± 1.80 and, 29.58 ± 1.98, prospectively, (P-value=0.749, T-test, poverty 0.1). The mean duration of breastfeeding was 47 weeks before surgery and 35 weeks after surgery. Conclusion Our study results did not indicate the adverse effects of breastfeeding on the results of PRK surgery. It seems that performing PRK is safe in the breastfeeding period.


Introduction
Photorefractive keratectomy (PRK) is believed to be a safe and e cacious method of correcting refractive errors. However, all patients who request surgery are not able to have keratorefractive surgery and not recommend in many cases. Many breastfeeding mothers are told that they cannot undergo refractive surgery until their breastfeeding is completed. (1) In previous studies,, nursing is considered as one of exclusion criteria. (2) Mothers are usually recommended to wait 3-6 months or longer after termination.
Even though a number of women experience visual changes during lactation, we could not nd evidence to support the idea that all mothers will experience visual changes or that vision changes that do occur during pregnancy and will usually remain throughout lactation. (3,4) However, many mothers request to have refractive surgery during this period, as they may be in leave due to breastfeeding. In this study, we evaluated the e cacy and safety in the long term follow up of PRK in breastfeeding mothers.

Study Population
Breastfeeding women affected with myopia and myopic astigmatism presenting to the Khatam-al-Anbia Eye Hospital for refractive surgery participated in this study. The inclusion criteria were age of  years, spherical equivalent (SE) refraction ranging from − 1.00 to − 8.00 diopters (D) with 3.00 D or less astigmatic error, stable refraction for at least 3 months, and preoperative best-corrected visual acuity (BCVA) of 20/20 or better.
The exclusion criteria were: any corneal dystrophies or abnormalities, the presence of any ocular pathologic condition that impaired visual function, any previous ocular surgery, keratoconus or keratoconus suspect, glaucoma or glaucoma suspect, auto-immune diseases, diabetes mellitus, moderate to severe dry eye. All patients discontinued contact lens wear at least one month before refraction and topography evaluation.

Surgical Procedure
After sterile draping, the cornea was anesthetized with tetracaine 1% eye drops and an eyelid speculum was inserted. Ethyl alcohol 20% was subsequently applied in a 9 mm well for 20 seconds and the epithelium was excised with a hockey stick spatula. Multidimensional rotational eye tracking was applied during the ablation in both groups. The minimum optical zone was 6 mm and equal optical zone was chosen for both eyes of each patient. Two surgeons (A. E. and S .Z. G.) carried out all surgeries by the use of a ying-spot 193-nm excimer laser (Technolas217z, Bausch & Lomb, Irvine, USA) with a xed pulse repetition rate of 100 Hz and a spot diameter of 1-2 mm. A sponge soaked with mitomycin C 0.02% was applied for all patients over the ablated area for 5 seconds per each diopter of treatment. A bandage contact lens was placed after copious balanced salt solution irrigation of the ocular surface.
Postoperatively, the patients received preservative-free levo oxacin 5 mg/mL and betamethasone 0.1% eye drops every 6 h. After complete reepithelialization (usually on the fth day), the bandage contact lens was taken away. Levo oxacin was discontinued after 1 week. Betamethasone was used for 1 month, and then uorometholone 0.1% eye drops were started every 6 h and gradually tapered over 2 months.
Preservative-free arti cial tears were administered frequently in the rst month and then tapered based on the ocular surface condition.

Postoperative Evaluation
At the 3-year postoperative visit as preoperative, HOAs, was measured with abberometer under mesopic conditions with a minimum pupil diameter of 6.0 mm. Moreover, root mean square (RMS) values were calculated from Zernike coe cient and reported in micrometers. Additionally, corneal topography parameters, scanning-slit corneal tomography, and refractive and visual outcomes were measured.

Data Analysis
All statistical analyses were done using SPSS18 (SPSS Inc. Chicago, Illinois, USA). The differences between the case and control groups were compared using Student t-tests. Scanning-slit corneal tomography, the independent T-test, and the Chi-square statistic were used to analyze and determine continuous variables (age) and categorical variables in the two groups, respectively (P < 0.05).

Ethical Considerations
With respect to the Declaration of Helsinki, the experimental procedures were clari ed for the participants and the informed written consent was obtained from each patient. This study was approved by Committee of Ethics in Human Research at Mashhad University of Medical Sciences.

Results
A total of 24 women with a mean age of 29.70 ± 1.8 years (Standard Deviation) (range from 27 to 33 years) took part in the study. The average age in the study and the control group was 29.83 ± 1.80 and, 29.58 ± 1.98 years, respectively, (P-value = 0.749, T-test, poverty 0.1). The mean duration of breastfeeding was 47 weeks before surgery and 35 weeks after surgery. Table 1 displays the baseline characteristics. There were no statistically signi cant differences in UCVA, BCVA, the manifest, and cycloplegic refractions. The manifest refractions showed no statistically signi cant difference in sphere, cylinder, also SE between groups; Furthermore, all patients were slightly myopic (Table 3). MRSE was moderately Table 3 Comparison of Orbscan topographic parameters of cases before and after photorephractive keratectomy. The biomicroscopy evaluation showed clear corneas in all patients. Seven women (58.3%) in the study group and 8 women (66.7%) in the control group had dry eye complaints. Consequently, preservative-free hydroxypropyl methylcellulose were prescribed and administered (P = 1.0, poverty = 0.12).
The results of corneal topography such as Surface Regularity Index (SRI) and Surface Asymmetry Index (SAI) are presented in Table 2. The best t surfaces, their minimum and maximum, the mean of Sim K's, the mean of irregular astigmatism, and central corneal thickness (CCT) were measured and provided in Table 3.
The distribution of various HOAs, RMS of total HAOs, and spherical aberrations were computed and given in Table 4).

Discussion
During normal pregnancy, hormonal, metabolic, hematological, vascular, and immunological changes can be observed (9), which can be shown to be associated with ocular changes during pregnancy. These changes include increased curvature as well as steepness of the cornea that begin during pregnancy, continue during breastfeeding, and returns to the original status with the end of breastfeeding (13). A slight increase in the thickness of the cornea at the end of the pregnancy period associated with changes in refraction and compliance can continue during breastfeeding (14,15). Another change is the reversible reduction in tear lm as a result of the prolactin secretion increase during pregnancy and breastfeeding (16). Therefore, refractive surgery is not usually recommended during pregnancy and a year after childbirth (17,18).
However, pregnancy and breastfeeding have some differences in physiological changes, which could affect results of the laser vision correction (LVC). One of these differences is the physiological changes associated with the role of placenta during pregnancy. During pregnancy, as an endocrine system, the placenta is independently capable of producing cytokines and hormones. The placenta cytokines are tumor necrosis factor alpha-TNFa, Resistin, and leptin. The important hormones of the placenta are human chorionic somatomammotropin-HCS, cortisol, estrogen, progesterone, and human placental growth hormone -HPG (19)(20)(21). TNFα potentially plays a role in physiological dosages in causing corneal scars followed by corneal opacity after PRK (22). Meanwhile, in the normal people (non-pregnant and non-breastfeeding women), a natural increase in TNFa is observed at physiological levels on the rst two days after PRK, which plays an important role in improving the corneal ulcer followed by laser ablation (23). In a study by Jiang, it was suggested that Resistin can increase the local cytokines and exacerbate in ammatory reaction by inducing extracellular leukocytosis (24). Prolactin effects has been examined on the production of tears and it has been speci ed that there is a negative relationship between serum prolactin and tear secretion, but a positive relationship has been observed between estradiol and tear secretion among women aged 30-39 years (25). It can be noted that during breastfeeding, the most important effective factor for the visual organ, which probably affects the results of PRK surgery is the prolactin hormone. Hence, the hormonal pro le of pregnancy period is distinguished from breastfeeding with placenta and further production of prolactin and all types of cytokines as well as other hormonal changes in other glands of the body; thus, classifying pregnancy and breastfeeding in one group regarding the possibility of PRK surgery is not very reasonably justi able.
In our study, postoperatively, comparison of the two groups showed no statistically signi cant differences in terms of corneal opacity, dry eye, UCVA, BCVA, dry and cycloplegic refraction (except for cylinder). In this regard, the coe cient of surgical effect in the two groups was in the range of the previous studies (26).

Visual Acuity
In comparing the uncorrected visual acuity, the two groups of breastfeeding and non-breastfeeding had no statistically signi cant differences (Table 1) In the study by Hashemi et al. (2015), after performing PRK using MMC on 30 eyes, the surgery impact factor was 1.01 (26). In the present study, although the MMC was not used in all of the patients, the results of the surgery impact factor were lower than the previous results.

Spherical Equivalent
In most of the studies, the values of manifest refraction spherical equivalent (MRSE) are in the range of mild myopia after PRK surgery (29,30), but the mean of MRSE can also be found in the range of mild hyperopia which, for instance, can be seen in the report by Randelman et al. (2009) (31). In the upcoming study, in both groups, the general overview of the average post-operative results of MRSE was shifted towards hyperopia; nevertheless, their comparison showed no statistically signi cant difference.

Dry Eye
Dry eye naturally occurs during pregnancy. It is improved by the end of breastfeeding as the level of prolactin decreases (8, 20). Besides, the incidence of dry eye could remain on the mild to moderate myopic eyes for 12 years after PRK surgery (32).
In our study, the absence of statistically signi cant difference in the incidence of dry eye among breastfeeding and non-breastfeeding women was not consistent with the comments of Umti et al. (8). Therefore, along with the effect of prolactin on dry eye, the other cases including meibomian gland dysfunction (MGD) have been also probably effective.
A slight increase in surface regularity index (SRI) and surface asymmetry index (SAI), higher level of best spherical anterior and posterior surfaces, more severe astigmatism in 5 mm, and a slight increase in HOAs in 5.0 mm were found before surgery in the breastfeeding group. These changes can be justi ed despite the edema remaining in the cornea from the pregnancy period (14,15) one result of which can be an increase in the curvature of the cornea (13). However, edema occurs further in marginal areas with thicker stroma (36). The comparison of breastfeeding and non-breastfeeding groups showed no statistically signi cant difference in terms of the above mentioned items after surgery. Since the laser ablation was performed in the corneal center for myopia correction in which it had smaller diameter than the margin and, consequently, less edema, therefore, physiological changes after pregnancy might have a lower effect on the post-operative results. Furthermore, during the ablation of the corneal center, the laser beams collided at the corneal surface at the 90° angle, which had a greater penetration impact, while the marginal rays that reached the marginal points of the cornea collided the points with a greater angle to cornea surface and would have less tissue evaporation when the optic area was about 6.5 mm (37).
Probably, fewer changes in the cornea caused by ablation in these areas associated with further edema had less inappropriate results over time with the improvement of edema after pregnancy that continued up to one year later (9). The absence of a signi cant difference in the comparison of the mean of the lowest corneal thickness between the two groups of breastfeeding and non-breastfeeding women after surgery could justify these cases. Anyway, it was in the range of previous studies (29).
Comparing subjective results and other evaluations including lowest corneal thickness and best anterior spherical surface to the best posterior spherical surface ratio between the two groups three years after the surgery demonstrated no statistically signi cant difference and were in an optimal level. From the ectasia risk factors mentioned in the previous study (38) and according to the results of Orbscan, there was probably an increase in the slope of cornea in the breastfeeding group in a stage before surgery, but these changes did not result in an even mild keratoconus.
Other factors include the ectasia risk, severe myopia correction, remaining post-ablation thin stromal thickness (39), extensive surface of ablation, irregular corneal thickness, and abnormal corneal topography (40,41). The surgery was carried out in a time interval of 11 months and 1 week after childbirth, and the average duration of breastfeeding was 8 months and 1 week. This can be associated with a decline in the natural complications of pregnancy in the visual organ, especially the cornea, (13)(14)(15), which is among other reasons that can be effective in preventing ectasia in the breastfeeding women.

Conclusion
Th results of this study did not indicate any adverse effects of breastfeeding on the results of PRK surgery. Since the average interval from the childbirth was 11 months and 1 week (at least 6 weeks) before PRK surgery, an earlier PRK surgery after childbirth requires further studies and cautious aspects. It should be noted that the success of the surgeries was probably due to the use of mitomycin C and spherical technique in the PRK surgery. Finally, considering the age range of 27-33 years, and mild/moderate myopia and regular moderate astigmatism can be the reason for the success of PRK surgery in the refractive correction of breastfeeding women.
Nevertheless, there were a number of constraints in collecting the required items for this study. Therefore, the results of this study indicate the possibility of choosing the refractive correction method of PRK surgery for women during breastfeeding period. Since no similar study has been carried out in this regard, this research has opened the way to continue the studies on the breastfeeding and laser refractive surgery. For example, the retrospective research project on the patients of this study and also a research with more individuals in the future can be advised.

Declarations
Availability of data and material: The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.
Competing interests: The authors declare that they have no competing interests Consent for publication: Not applicable, a retrospective cohort study. We had a general consent for using refractive surgeries patient's data anonymously for any research or publication. Authors' contributions: All the authors contributed signi cantly to this research, and all authors agree to be accountable for all aspects of the work. SZG, AE, SSR participated in study conceptualization, design, and in gathering the patients, and performing PRKs. SA participated in statistical analysis and interpretation. ARDE and MA participated in acquisition of clinical data and revising and nalizing the manuscript. All authors read and approved the nal manuscript.