Hormonal Modulation of Keratoconus: A Systematic Review and Screening Strategy for At-Risk Populations

Abstract

Topic

Keratoconus (KCN) is a progressive corneal ectasia with poorly understood etiology and risk factors leading to inadequate screening methods for at-risk populations. One potentially vulnerable population includes individuals experiencing fluctuations in hypothalamic–pituitary–gonadal (HPG) axis hormones. This systematic review aims to evaluate the association between HPG-axis hormonal fluctuations and the development or progression of KCN.

Clinical Relevance

Elucidating the relationship between HPG-axis hormone fluctuations and KCN may enhance screening strategies, facilitate early detection, and inform preventive measures in at-risk populations. Improved awareness among clinicians could lead to targeted monitoring and timely interventions.

Methods

A comprehensive literature search was conducted in PubMed, Scopus, Web of Science, Embase, and Google Scholar databases up to July 2025, adhering to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Studies were included if they provided clinical evidence of associations between HPG-axis hormone fluctuations and KCN onset or progression. Two reviewers independently extracted relevant data on hormone changes and KCN. Methodological quality was assessed using the Joanna Briggs Institute (JBI) appraisal tools.

Results

Twenty-six studies met inclusion criteria, comprising 16 descriptive studies and 10 case-control studies, with an aggregate of 4201 participants (∼8000 eyes). Endogenous hormonal shifts related to pregnancy and congenital hormonal abnormalities accounted for 56.2% of identified triggers (9 of 16 studies; 26 patients), while exogenous hormone exposure from hormone replacement therapy and antiandrogen treatment represented 31.2% (5 of 16 studies; 7 patients). Ten case-control studies included 3834 participants (1623 KCN patients and 2211 controls). Of these 10, 8 studies (80%) demonstrated significant alterations in sex steroid hormones (elevated dehydroepiandrosterone sulfate [DHEAS] and estradiol; decreased estrone and estriol; mixed results for testosterone) or gonadotropins (altered luteinizing hormone/follicle-stimulating hormone ratios, reduced gonadotropin-releasing hormone) in association with KCN. Methodological quality assessment indicated that 22 of the 26 studies had high reporting quality per JBI items.

Conclusion

This systematic review highlights consistent associations between HPG-axis hormone fluctuations and KCN development and progression, as evidenced across observational studies. Specifically, altered levels of DHEAS, estrogens, and gonadotropins emerged as key hormones linked to KCN pathology. These findings support a systemic component in the etiology of KCN, underscoring the importance of hormonal influences in clinical management and screening, particularly during periods of significant hormonal fluctuation, such as pregnancy, congenital endocrine disorders, or hormonal therapy.

Financial Disclosure(s)

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Keratoconus (KCN) is an ectatic corneal disease characterized by progressive stromal thinning and steepening of the cornea, leading to significant visual impairment if left untreated.^1,2 Keratoconus typically manifests during adolescence or early adulthood, affecting visual function during key developmental phases of life.^3,4 The prevalence of KCN is estimated at approximately 0.04% in the US population and as high as 3% to 5% in certain Middle Eastern cohorts, with prevalence showing an upward trend with improved diagnostics and early detection.5, 6, 7, 8, 9 Despite significant advancements in diagnostic and therapeutic approaches, the underlying etiology of KCN remains incompletely understood, limiting efforts toward comprehensive risk identification and effective prevention.^10,11

Traditionally, KCN was categorized as a localized corneal condition influenced by genetic susceptibility, environmental triggers, and mechanical factors like eye rubbing.^11,12 However, recent research highlights more intricate systemic mechanisms involving hormonal dysregulation, inflammatory pathways, and oxidative stress, informing our current understanding of KCN pathogenesis.^13,14 Among systemic factors, sex hormones exert considerable influence on ocular physiology and pathophysiology, mediated through specific hormonal receptors located within ocular tissues.^13,15,16

Various studies have demonstrated significant hormonal impacts on ocular structures including the lacrimal glands, ocular surface, tear film, and conjunctival tissues.^15,17, 18, 19 Physiological fluctuations in sex hormone levels during critical periods such as pregnancy and menopause have been associated with notable alterations in corneal structure and biomechanics.^20,21 Receptors for estrogen, progesterone, and testosterone have been specifically identified within corneal tissue, further supporting hormonal influence on corneal physiology.22, 23, 24 Observational studies have correlated reproductive hormonal changes with variations in corneal topography and function under various physiological conditions.25, 26, 27 For instance, fluctuations in corneal thickness and curvature associated with hormonal shifts during pregnancy, lactation, and menstrual cycle have been shown to result in observable visual alterations.^25,28, 29, 30

Additionally, sex hormones modulate corneal extracellular matrix composition through regulatory effects on matrix metalloproteinases and the activation of proteinase and collagenolytic enzymes within the stromal matrix, thus providing a plausible mechanistic link between hormonal variability and corneal structural weakening in KCN.^13,31, 32, 33

Despite accumulating evidence, the current literature lacks a rigorous, systematic synthesis of clinical data evaluating the role of hypothalamic–pituitary–gonadal (HPG) axis hormones in KCN pathogenesis. Prior reviews have acknowledged potential hormonal influences but have not systematically assessed all available clinical evidence to inform targeted screening approaches.^34,35 Keratoconus diagnosis frequently occurs incidentally or after visual impairment has already occurred. Consequently, early identification of asymptomatic or subclinical KCN is paramount for improving long-term patient outcomes.36, 37, 38 Although recent technological and computational advancements hold promise for enhancing early predictive capabilities, definitive screening strategies for high-risk populations remain underdeveloped.^39,40 Accurately defining these at-risk groups continues to pose a significant challenge, necessitating comprehensive epidemiological insights to inform effective public health interventions.^41,42 This issue demands particular attention in regions where KCN prevalence is notably elevated.

Therefore, this systematic review aims to address the existing knowledge gap by systematically evaluating and synthesizing available evidence on the associations between fluctuations in HPG-axis hormones, including gonadotropin-releasing hormone (GnRH), luteinizing hormone (LH), follicle-stimulating hormone (FSH), estrogen, progesterone, testosterone, and their metabolites, and the onset or progression of KCN. By clearly delineating hormonal risk factors, the findings from this review may enable earlier identification of at-risk populations, guiding the development of targeted, evidence-based clinical screening programs and preventive strategies to manage KCN at its subclinical stages, ultimately improving patient prognosis and outcomes.

Methods

Protocol and Registration

This review was developed and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines.⁴³ This study did not meet the criteria for human subjects research as defined by our institution, as it did not include patient data. Therefore, it did not require institutional review board approval or informed consent. Our study adhered to the Declaration of Helsinki.

Eligibility Criteria

Studies were included if they described clinical cases of KCN onset or progression and described or quantified fluctuations in HPG-axis hormones, including GnRH, LH, FSH, estrogen, progesterone, testosterone, or their metabolites. Exclusion criteria consisted of in vitro studies, nonoriginal research, conference abstracts without full publications, non-English articles, and studies lacking relevant hormonal or clinical data.

PICOD

Population

Patients experiencing the development or progression of KCN linked to fluctuations in HPG-axis hormones.

Intervention/Exposure

Fluctuations in HPG-axis hormones (GnRH, LH, FSH, estrogen, progesterone, and testosterone).

Comparison

For comparative studies, KCN patients versus controls; not applicable for descriptive studies.

Outcome

Occurrence or progression of KCN, or differences in HPG-axis hormone levels.

Design

Case reports, case series, cross-sectional, cohort, case-control, and descriptive studies.

Information Sources and Search Strategy

A comprehensive electronic search was conducted across PubMed, Google Scholar, Scopus, Web of Science, and Embase, covering the period from January 1963 to July 2025 without language restrictions. The search combined terms related to KCN and HPG-axis hormones. Animal studies were excluded using appropriate filters. The full detailed search strategy is provided in supplementary Appendix A (available at www.ophthalmologyscience.org).

All search results were managed using EndNote 2025 (Clarivate Analytics). Additionally, reference lists of included studies and relevant systematic reviews were manually screened to identify further eligible articles.

Study Selection

Duplicate records were removed, and titles, abstracts, and full texts were screened by two independent reviewers (M.A. and E.S.) according to the predefined inclusion and exclusion criteria. In cases of disagreement, a third reviewer (H.N.S.) was consulted to resolve discrepancies through discussion and consensus.

Data Collection Process and Data Items

Two independent reviewers (M.A. and E.S.) systematically extracted and compiled all relevant data using a standardized checklist developed specifically for this review. Any discrepancies were resolved through discussion and consensus among the authors, with a third reviewer providing independent verification to ensure accuracy and consistency (H.N.S.). For each study, detailed information was gathered on study characteristics, including the study title, publication year, country of origin, study design, sample size, patient demographics, and follow-up duration, as well as on the hormonal context, encompassing HPG-axis hormone triggers, methods of hormone measurement, criteria for KCN diagnosis and progression, clinical outcomes, observed disease progression, and key findings.

Quality Appraisal

All included studies were independently assessed for methodological quality by 2 reviewers (S.G. and A.A.) using the Joanna Briggs Institute (JBI) Critical Appraisal Checklists specific to each study design.⁴⁴ This design-specific approach ensured rigorous and standardized quality assessment, especially given the diversity of study types and the lack of a single validated tool for both descriptive and analytical observational studies. The JBI checklist for case reports consists of 8 items; for case series, 10 items; for cohort studies, 11 items; and for case-control studies, 10 items. Each item was scored as “Yes” (1 point) or “No,” “Unclear,” and “Not Applicable” (all 0 points). Any disagreements between reviewers were resolved through consultation with a third reviewer (H.N.S.).

Although JBI does not prescribe a formal numeric scoring system, a pragmatic approach was applied to categorize risk of bias based on total scores. For case reports (max score 8), scores of 0 to 2, 3 to 5, and 6 to 8 indicated high, moderate, and low risk, respectively. For case series (max score 10), scores of 0 to 3, 4 to 6, and 7 to 10 corresponded to high, moderate, and low risk. For cohort and case-control studies, percentage-based thresholds were used: scores below 50% were considered high risk, 50% to 70% moderate risk, and above 70% low risk.

Synthesis of the Results

Due to the potential heterogeneity among the included studies, the results were synthesized narratively and organized into thematic categories, as presented in Tables 1 and 2.

Table 1.

Summary of Observational Studies on HPG-Axis Hormonal Influences and Keratoconus Onset/Progression

First Author, Yr	Country	Study Design	Participants & Demographics	Follow-Up (Time Points)	Hormonal Context and Assay	KCN Diagnosis and Progression Criteria	Observations and Outcome	Key Findings
Van46, 2023	Denmark	Prospective cohort	28 patients with KCN undergoing CXL -Mean age: 25 ± 5.6 yrs; Gender: 79% male (n = 22), 21% female (n = 6)	Baseline (pre-CXL) and 2–3 mos post-CXL	Specimen type: Plasma Measured Hormones: DHEA-S, estrone, estriol Assay platform: ELISA	-KCN diagnosis: topography -Progression: increase in Kmax ≥1.5 D over 3–6 mos	Baseline (pre-CXL): DHEA-S was markedly higher than estrone and estriol (both P < 0.001) when compared with levels reported in the literature. While DHEA-S and estrone showed no correlation with KCN severity, estriol correlated with higher Kmax (P = 0.01).	Baseline estriol was significantly correlated with higher Kmax in KCN patients.
Naderan61, 2017	Iran	Prospective cohort	22 KCN patients (100% female): -11 pregnant (bilateral, stable ≥2 yrs; mean 24 ± 5 yrs) -11 age- and severity-matched non-pregnant (mean 25 ± 3 yrs).	Pre-pregnancy, 34th wk of gestation, and 6 mos postpartum	Hormonal Context: Pregnancy; no direct hormone assays	-KCN diagnosis: topography and clinical signs. -Progression: increase in K max of ≥1 D, or max-min K ≥1 D, or median K ≥0.75 D; or ≥2% decrease in central corneal thickness, or ≥0.5 D MRSE change.	K (flat, steep, mean) increased during pregnancy and postpartum compared to controls (P < 0.001).	KCN may progress during pregnancy; No progression occurred in the non-pregnant group.
Fink68, 2010	USA	Prospective cohort (CLEK Study)	295 participants (aged 48–59 yrs) -151 males (mean age 52.68 ± 2.87 yrs) -115 hormonally active females (mean age 52.36 ± 2.87 yrs) -29 hormonally inactive females (mean age 53.64 ± 2.30 yrs)	4 yrs of follow-up (data from CLEK study yrs 5–8)	Hormonal context: Sex and menopausal/HRT status assessed by questionnaire; no direct hormone assays.	-KCN diagnosis: topography, K indices, and clinical signs -Progression: change in BCVA, steep K, and scarring over 4 yrs	KCN progressed slowly and similarly across men, hormonally active females, and postmenopausal females (with or without HRT)	Sex and hormone status did not significantly affect KCN progression in midlife.
Bouhouche52, 2021	Morocco	Case series (family-based genetic study)	Two boys (ages 7 and 3 yrs) from a consanguineous family of Moroccan origin	≥3 mos clinical follow-up for probands; cross-sectional endocrine/genetic workup	Hormonal context: Steroidogenesis defect (SDR42E1 mutation) Specimen type: serum Measured hormones: testosterone, estradiol, progesterone, 17-OHP, DHEA-S, androstenedione, FSH, LH, cortisol, ACTH	-KCN diagnosis: clinical examination, OCT/biomicroscopy	Both boys had severe, early, progressive KCN. Hormonal assays showed low testosterone and estradiol, elevated androstenedione (delta 4-androstenedione), and abnormal 17-hydroxyprogesterone	SDR42E1-mediated steroidogenesis links steroid hormone biosynthesis defects to KCN and associated syndromic manifestations.
Yuksel62, 2016	Turkey	Case series	Three white female patients with KCN - Mean age 32.3 ± 3.6 yrs (range 28–36 yrs); all underwent IVF treatment. None achieved pregnancy.	Mean follow-up 15.6 ± 3.2 mos (range 12–18 mos)	Hormonal context: IVF ovarian stimulation; no direct hormone assays	-KCN diagnosis: topography. -Progression: increase in Kmax >1.0 D, Kmean >0.75 D, corneal apex power >1.0 D, MRSE >0.5 D, thinning in the thinnest part of the cornea by 2%, or decreased visual acuity	All 6 eyes of 3 patients had progression of KCN with increased Kmax and Kmean and decreased visual acuity after IVF. Decreased vision occurred 1–2 mos after IVF treatments. None achieved pregnancy before or after IVF.	IVF treatment poses a risk for KCN progression
Bilgihan67, 2011	Turkey	Case series	Four female patients with KCN (7 eyes), mean age 29.3 yrs (range 22–43 yrs)	Mean follow-up 39 mos (range 4–108 mos)	Hormonal context: Pregnancy; no direct hormone assays	-Progression: increase in steep or mean SimK ≥1 D, astigmatism ≥1 D, MRSE ≥0.5 D, and changes in RGP lens fitting (base curve radius decrease ≥0.1 mm, increased apical contact)	KCN progression was observed in 7 eyes of the 4 stable KCN patients, with 3 patients progressing within 6 mos of pregnancy and 1 progressing 3 mos postpartum	KCN may progress during pregnancy.
Soeters66, 2012	Netherlands	Case report	Two females aged 32 and 29 yrs, both diagnosed with KCN after their second pregnancy.	Case 1 followed for 4 yrs postpartum Case 2 followed for 3.5 yrs postpartum	Hormonal context: Pregnancy; no direct hormone assays	-KCN diagnosis: topography, clinical examination, refraction, and pachymetry. Progression: increase in Kmax, astigmatism, and refraction; decrease in pachymetry.	Both cases were diagnosed with KCN after second pregnancy with significant increases in corneal steepening and thinning, worsening visual acuity, and increased astigmatism. Onset occurred in case 1 during the third trimester and in case 2 postpartum.	KCN may occur and progress during pregnancy.
Hoogewoud64, 2013	Switzerland	Case report	Two healthy females, aged 32 and 26 yrs, diagnosed with bilateral KCN during pregnancy	Case 1 followed for 1 year after the second pregnancy Case 2 followed for 3 mos after the first pregnancy	Hormonal context: Pregnancy; no direct hormone assays	-KCN diagnosis: topography, decreased VA -Progression: increase in Kmax, astigmatism, decrease in visual acuity, and topographic changes	Both cases were diagnosed with KCN during pregnancy, one in the second trimester and the other in the third trimester. Both cases showed significant variations in Ks and VA during pregnancy. Postpartum, K values stabilized in case 1, and returned to pregestational values in case 2.	KCN may occur during pregnancy.
Incorvaia69, 2003	Italy	Case report	Two 25-yr-old dizygotic female twins diagnosed with nonclassical CAH (21-hydroxylase deficiency) at age 21	15 mos	Hormonal context: CAH (21-hydroxylase deficiency) Specimen type: serum Measured Hormones and markers: DHEAS, prolactin, androstenedione (ASD), 17-hydroxyprogesterone (17-OHP)	-KCN diagnosis: confirmed with topography, slitlamp examination, ultrasonic pachymetry -Progression: confirmed with topography, refraction, and clinical examination	IN this first report of KCN associated with CAH, one twin had bilateral asymmetric KCN and the other had progressive fruste central KCN. Serum hormonal abnormalities (elevated ASD and 17-OHP were correlated with severity of corneal changes. Progression of KCN confirmed during follow-up period, especially asymmetric progression in twin with more severe CAH).	CAH secondary to 21-hydroxylase deficiency can cause overproduction of androgens, which may be associated with KCN.
Deitel49, 2023	USA	Case report	28-yr-old male-to-female transgender patient with subclinical KCN	12-mo follow-up on expanded hormone therapy prior to CXL.	Hormonal context: Gender-affirming hormone therapy (including estradiol, spironolactone, and progesterone); no direct hormone assays.	-KCN diagnosis: slitlamp examination, refraction, decrease VA, tomography, and topography. -Progression: based on slitlamp examination, refraction, tomography, and topography.	KCN progression occurred 4 mos after initiation of hormone therapy. Kmax was 58.3 D OD and 77.7 D OS, and pachymetry 440 μm and 397 μm, respectively. Patient underwent CXL.	Gender-affirming hormone therapy may be associated with KCN progression.
Torres-Netto54, 2019	Switzerland	Case report	49-yr-old female with endometriosis and bilateral KCN previously stable for 10 yrs	Follow-up at baseline, 3, 6, and 13 mos post-STEAR and post-CXL	Hormonal context: Selective tissue estrogenic activity regulator (STEAR) therapy; no hormone assays	-KCN diagnosis: refraction, tomography and topography -Progression: increase in Ks; reduced pachymetry and VA.	Patient had rapid bilateral KCN progression after 4 mos of STEAR therapy, with Kmax increase to 2.7 D (OD) and 3.8 D (OS) despite 10 yrs of prior stability.	KCN can progress in previously stable eyes after hormone therapy.
Coco57, 2019	USA	Case report	51-yr-old female with BRCA mutation; stable KCN for 17 yrs	Follow-up at baseline, 6 mos, 10 mos, and 14 mos after starting HRT	Hormonal context: HRT initiated after hysterectomy and bilateral oophorectomy; no hormone assays	-KCN diagnosis: tomography and topography Progression: increase in steep K values and tomographic changes; BCVA monitored	Patient had stable KCN for 17 yrs with minimal progression prior to HRT. After HRT initiation, rapid progression observed over 14 mos. Steepest K increased from 63.7D to 71.5D (OD) and 65.8D to 78.1D (OS). Patient underwent CXL.	HRT can accelerate KCN progression
Scott53, 2020	New Zealand	Case report	36-yr-old New Zealand European female	Follow-up for 10 mos postpartum	Hormonal context: Pregnancy; no direct hormone assays.	-KCN diagnosis: tomography and clinical signs -Progression: increasing astigmatism and changes in K indices.	New diagnosis of KCN during pregnancy. Rapid progression with increase in astigmatism (2.0 D right eye, 1.25 D left eye) and maximum K up to 70.5 D OD and 54.1 D OS. Patient uinderwent CXL post-partum.	KCN may occur and progress during pregnancy.
Stock59, 2017	Brazil	Case report	26-yr-old pregnant white female with bilateral KCN and prior CXL	Follow-up was 20 mos after the first CXL, including 9 mos after her delivery	Hormonal context: Pregnancy; no direct hormone assays.	-KCN diagnosis: topography, ultrasonic pachymetry -Progression: change in topography, decreased VA	During the fifth month of pregnancy (and 7 mos after CXL), KCN progressed to acute corneal hydrops in the left eye (OS VA 20/500) despite prior CXL.	KCN may progress rapidly during pregnancy.
Glicéria65, 2013	Brazil	Case report	A 37-yr-old female, stable KCN >10 yrs in left eye	Follow-up more than 5 yrs, including the pregnancy period	Hormonal context: Pregnancy; no direct hormone assays.	-KCN diagnosis: by corneal topography -Progression: decreased VA, topography changes, and changing corneal biomechanics	KCN progressed significantly in her left eye during pregnancy despite more than 10 yrs of stability. Progression was marked by an increase in maximum K from 63.9 D to 68.5 D, a thinnest corneal thickness of ∼340 μm, increased corneal protrusion, and decreased corneal hysteresis and corneal resistance factor.	The study highlights KCN progression during the pregnancy period in an eye that had been stable for more than 10 yrs.
Natarajan60, 2017	India	Case report	35-yr-old female with bilateral KCN stable ≥5 yrs pre-HRT	Total follow-up duration: 13 yrs	Hormonal context: HRT; no direct hormone assays.	-KCN diagnosis: By corneal topography, refraction, and ultrasound pachymetry -Progression: increase in Sim K values and decrease in pachymetry	Both eyes were stable for 5 yrs pre-HRT. Six mos after HRT initiation, both eyes progressed with increased SimK, pachymetry decreased to 346 μm OD and 387 μm OS, and biomechanical parameters worsened. The patient underwent CXL OD.	KCN progression occurred in previously stable eyes 6 mos after initiation of HRT.

ASD = androstenedione; BAD-D = Belin/Ambrosio Enhanced Ectasia Display; BCVA = best-corrected visual acuity; CLEK = Collaborative Longitudinal Evaluation of Keratoconus; CLEK Study = Collaborative Longitudinal Evaluation of Keratoconus; CXL = corneal collagen cross-linking; D = diopters; DHEA-S = dehydroepiandrosterone sulfate; ELISA = enzyme-linked immunosorbent assay; HPG = hypothalamic-pituitary-gonadal; HRT = hormone replacement therapy; IVF = in vitro fertilization; K= keratometry; KCN = keratoconus; Kmax = maximum keratometry; Kmean = mean keratometry; MRSE = manifest refractive spherical equivalent; OD = oculus dexter (right eye); OHP = 17-hydroxyprogesterone; OS = oculus sinister (left eye); RGP = rigid gas-permeable (contact lenses); SDR42E1 = steroidogenesis enzyme gene; Sim K = simulated keratometry; 17-STEAR = selective tissue estrogenic activity regulator; TIMPs = tissue inhibitors of metalloproteinases; VA = visual acuity; yrs = years.

Table 2.

Summary of Case-Control Studies Comparing Quantitative HPG-Axis Hormone Levels in Keratoconus and Control Groups

First Author, Yr	Country	Study Design	Participants & Demographics	Hormonal Assays	KCN Diagnosis Criteria	Results & Outcomes	Key Findings
Stanescu45, 2025	Israel	Case-control (unmatched)	96 participants (100% female) in 3 groups: -36 premenopausal females with KCN (no refractive surgery) (mean age 29.8 ± 3.2 yrs) -29 postrefractive surgery ectasia (PRSE) (mean age 31.9 ± 2.6 yrs) -31 healthy controls (mean age 30.7 ± 3.5 yrs)	Specimen type: venous blood (collected on 2nd day of menstrual cycle) Measured hormone(s): estradiol (E2)	-KCN diagnosis: topography, tomography, clinical signs	Mean estradiol levels were significantly higher in KCN group (38.0 ± 2.4 pg/mL) compared to controls (28.6 ± 3.9 pg/mL) (P < 0.001). Adjusted logistic regression indicated elevated estradiol increased ectasia risk (OR 2.44–2.71, P < 0.001). No significant differences were found with OCP use (P = 0.52) or menstrual cycle regularity (P = 0.43) between groups.	Elevated estradiol may play a role in KCN etiopathogenesis.
Beatty48, 2024	USA	Case–control (retrospective, matched)	2288 participants (55% female): -572 KCN (mean age 58.7 ± 15.6 yrs; 55% female) -1716 age-, ethnicity-, and gender-matched controls (mean age 58.7 ± 15.6 yrs; 55% female)	Hormonal context proxied by estrogen-containing medication use and pregnancy status (collected from electronic health records (EHR) and survey data). No direct hormonal assays were performed.	-KCN diagnosis: confirmed by ICD/SNOMED codes from EHR	Multivariable logistic regression showed significant association of KCN diagnosis with estrogen-containing medication (OR = 1.77, P = 0.0014) Other variables such as pregnancy showed positive but non-significant association (OR = 1.30, P = 0.20).	Increased consumption of estrogen-containing medications was associated with KCN etiopathogenesis.
Escandon47, 2024	USA	Case-control (matched)	Participants were stratified by specimen type: -For plasma samples: 227 KCN (26.87% female), 58 controls (50% female) -For saliva samples: 274 KCN (26.64% female), 101 controls (53.46% female) All controls age- and sex-matched	Specimen type: plasma and saliva Measured hormone(s): gonadotropin-releasing hormone (GnRH) Assay platform: ELISA	-KCN diagnosis: based on clinical grading (ABCD system)	Plasma GnRH is significantly lower in KCN v. controls (median 67.4 vs 100.1 ng/mL, P < 0.0001), consistent across sex- and age-matched groups; saliva GnRH also lower in KCN (median 77.7 vs 118.6 pg/mL, P = 0.0007).	GnRH was reduced in those with KCN
Zhao31, 2022	China	Case-control (unmatched)	182 participants: -62 KCN patients with mean age of 23.73 ± 5.16 yrs (19.35% female) -120 controls with myopia/astigmatism with mean age of 23.68 ± 6.10 yrs (17.5% female)	Specimen type: plasma Measured hormone(s): estriol (E3), estradiol (E2), progesterone (P), testosterone (T) Assay platform: E chemiluminescence immunoassay	-KCN diagnosis: by clinical signs (corneal ectasia, Fleischer ring, Vogt striae), corneal topography/tomography (Pentacam, Sirius) using the criteria: Kmax increase >1 D, mean corneal refractive power increased by >1 D, spherical equivalent of manifest refraction increased by >1 D, and BCVA worsened more than one line.	Hormonal comparisons were stratified based on gender. Among females: E2 (P = 0.17), E3 (P = 0.45), and P (P = 0.56) levels were similar among KCN and control cases. However, female KCN patients had significantly lower T than controls (0.86 ± 0.33 vs 1.18 ± 0.58 nmol/L; P = 0.044), with positive correlations between T and central corneal thickness (r = 0.395, P = 0.023) and thinnest corneal thickness (r = 0.378, P = 0.030). Among males: E3 (P = 0.18) and P (P = 0.44) levels were the same for KCN and controls. Male KCN patients showed higher E2 (143.75 ± 34.82 vs 124.80 ± 43.56 pmol/L; P = 0.013) and lower T (11.59 ± 2.85 vs 13.58 ± 4.77 nmol/L; P = 0.026), with E2 positively correlated with maximum K (r = 0.222, P = 0.007).	Elevated estradiol may contribute to KCN etiopathogenesis in males, whereas reduced testosterone may be implicated in both sexes.
Jamali50, 2022	Iran	Case-control (unmatched)	102 participants: -76 KCN patients with mean age of 22.79 ± 5.71 yrs (50% female) -26 controls with mean age of 21.42 ± 5.21 yrs (50% female)	Specimen type: serum (female blood samples were taken during follicular phase [days 4–8 of menstrual cycle]) Measured hormone(s): testosterone (T), dehydroepiandrosterone sulfate (DHEAS), prolactin (PRL), luteinizing hormone (LH), follicle-stimulating hormone (FSH) Assay platform: radioimmunoassay and immunoradiometric assays	-KCN diagnosis: topography with Belin/Ambrósio Enhanced Ectasia Display (BAD-D), clinical signs	Males with KCN showed significantly higher serum T (6.18 ± 3.80 vs 1.57 ± 1.76 ng/mL; P < 0.001) and DHEAS (3.71 ± 2.23 vs 2.53 ± 1.77 μg/mL; P = 0.005) than controls. Females with KCN had significant elevation of T (0.78 ± 0.96 vs 0.32 ± 0.13 ng/mL; P < 0.001) and insignificant elevation of DHEAS levels (2.40 ± 1.57 vs 2.18 ± 0.72 μg/mL; P = 0.988). LH levels showed no significant difference, while FSH was significantly lower in males with KCN (P = 0.015) and trended lower in females (P = 0.080).	Elevated DHEAS and testosterone are associated with KCN in both sexes, whereas reduced FSH was observed only in males.
Daphne Teh51, 2021	Malaysia	Case-control (matched)	40 participants (age range 16–47 yrs): -20 KCN (35% female) -20 matched controls by age, gender, and race	Specimen type: serum Measured hormone(s): DHEAS Assay platform: untargeted metabolomics using LC-Q-ToF/MS	-KCN diagnosis: topography, slitlamp examination, and pachymetry	Significant upregulation of DHEAS and eicosanoids (5-HETE, PGA2, PGE2, PGF2α) in KCN serum compared to controls (P < 0.05); glycerophospholipid PS (17:2/20:4) upregulated (P < 0.05) in control cases. DHEAS showed significant elevation in KCN males (P = 0.026).	Elevated DHEAS may be associated with KCN
Karamichos56, 2019	USA, Denmark	Multi-center case-control (unmatched)	131 participants: -86 KCN patients (26.74% female) -45 healthy controls (51.11% female) Patients were categorized into 3 age groups: 15–29, 30–45, >46 yrs	Specimen type: Plasma Measured hormone(s): LH and FSH Assay platform: ELISA	-KCN diagnosis: topography, slitlamp examination, and refraction	LH levels showed no significant difference between KCN patients and controls; FSH level was slightly elevated overall in the KCN group but was not statistically significant; LH/FSH ratio was higher among the control group (P < 0.05). LH and FSH levels were higher among females than males for both KCN and normal groups with a more prominent difference among KCN patients. FSH was significantly increased in female KCN patients compared to female controls (P < 0.05); LH/FSH ratio was more significantly decreased in male KCN patients. KCN severity had significant effects on reducing LH/FSH ratios group with the lowest ratio was the most severe group (P < 0.05)	FSH was increased and LH/FSH ratio decreased in KCN
Ayan58, 2019	Turkey	Case-control (unmatched)	50 participants: -30 KCN patients (50% female; mean age 21.5 ± 4.2 yrs, range 18–30) -20 controls (50% female; mean age 24.6 ± 3.4 yrs, range 19–32)	Specimen type: cultured corneal epithelium collected from each group Measured hormone(s): estrogen α and β, progesterone, and androgen receptors Assay platform: Immunohistochemical detection and mRNA expression by qPCR	-KCN diagnosis: topography, clinical signs	There was no estrogen α or β receptor staining in both KCN and control groups; progesterone receptor was positive in 83.3% KCN vs 100% controls (P = 0.075); androgen receptor was positive in 13 KCN cases (43.3%) vs 8 (40%) controls (P = 0.815); no significant differences based on gender. qPCR showed significantly higher mRNA expression of estrogen α and androgen receptors in KCN group (P < 0.001); no significant difference for estrogen β and progesterone receptors mRNA; no differences in receptor expression by gender or KCN stage.	Estrogen α receptors are more highly expressed in KCN.
Sharif55, 2019	USA, Denmark	Case-control (unmatched)	207 participants: -147 KCN patients (28.6% female) -60 healthy controls (55% female); both groups were categorized into 4 age subgroups: 15–24 (13%), 25–34 (30%), 35–44 (28.5%), ≥45 (28.5%)	Specimen type: Plasma and saliva Measured hormone(s): estrone, estriol, 17β-estradiol, DHEAS Assay platform: ELISA	-KCN diagnosis: topography, tomography, clinical signs, and refraction	Estrone and estriol levels were significantly decreased in both plasma and saliva (P < 0.0001) in KCN group; 17β-estradiol levels showed no significant difference (P > 0.05); DHEAS levels were significantly elevated in plasma (P < 0.0001) and saliva (P = 0.0001).	Reduced estrone and estriol and elevated DHEAS are associated with KCN.
McKay63, 2016	USA, Denmark	Multicenter case-control (unmatched)	78 participants: -64 KCN patients (31.25% female) (mean age: males 26 ± 1.14 yrs, females 34 ± 1.75 yrs); -14 healthy controls (age/gender data are not presented for the control group in the study)	Specimen type: Saliva Measured hormone(s): estrone, estriol, 17β-estradiol, and DHEAS Assay platform: ELISA	-KCN diagnosis: topography, clinical signs	Results showed a significant 1.3-fold reduction in estrone levels in KCN group compared to healthy controls (P = 0.0222), while DHEAS was notably elevated 2.5-fold in KCN patients (P = 0.0359). Estriol (P = 0.4079) and 17β-estradiol (P = 0.7283) levels did not differ significantly. No significant association was found between disease severity and levels of estrone, estriol, or DHEAS.	Elevated DHEAS and reduced estrone are associated with KCN

BAD-D = Belin/Ambrósio Enhanced Ectasia Display; CXL = corneal collagen crosslinking; D = diopters; DHEAS = dehydroepiandrosterone sulfate; E1 = estrone; E2 = estradiol; E3 = estriol; EHR = electronic health record; ELISA = enzyme-linked immunosorbent assay; FSH = follicle-stimulating hormone; glycerophospholipid PS = glycerophospholipid phosphatidylserine; GnRH = gonadotropin-releasing hormone; GnRHR = GnRH receptor; HPG = hypothalamic-pituitary-gonadal; ICD = International Classification of Diseases; K = keratometry; KCN = keratoconus; Kmax = maximum keratometry; LC-Q-ToF/MS = liquid chromatography–quadrupole time-of-flight mass spectrometry; LH = luteinizing hormone; OCP = oral contraceptive pill; OR = odds ratio; P = progesterone; PRL = prolactin; PRSE = postrefractive surgery ectasia; qPCR = quantitative polymerase chain reaction; rGnRH = recombinant GnRH; SNOMED = Systematized Nomenclature of Medicine; T = testosterone.

Results

Literature Search

The comprehensive search across major databases initially identified 323 articles. After removal of duplicates, 268 unique records were screened by title and abstract, narrowing the pool to 42 articles for full-text evaluation. Upon detailed assessment against eligibility criteria, 26 studies comprising 16 descriptive studies and 10 case-control studies were included in the review. Sixteen studies were excluded due to incomplete data, irrelevant focus, or conference abstracts without full manuscripts (Fig 1).

Figure 1.

PRISMA flowchart. PRISMA = Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Study Characteristics and Demographic Data

A total of 4201 participants (approximately 8000 eyes) were included across 26 studies published between 2003 and 2025. These comprised 16 descriptive, observational studies, and 10 case-control studies.^31,45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69 Sample sizes ranged from isolated case reports to large-scale database analyses, such as Beatty et al,⁴⁸ which included over 2000 participants.

Descriptive Studies

A total of 16 descriptive, observational studies published between 2003 and 2023 included 367 participants (∼750 eyes), comprising 176 males (48%) and 191 females (52%), with a weighted mean age of 32.2 ± 11.8 (range: 5–51) years.

Geographically, 3 studies originated from the United States (18.7%), followed by Switzerland, Turkey, and Brazil with 2 studies each (12.5% each), and 1 study each from Denmark, Iran, Morocco, the Netherlands, Italy, New Zealand, and India (6.2% each).

Of the 16 studies, 1 study did not report any association between sex hormone status and KCN progression in middle-aged patients based on questionnaire data.⁶⁸ However, the remaining 15 studies primarily reported KCN onset or progression in the context of endogenous or exogenous hormonal changes. Among these 15 studies, 5 reported new diagnoses of KCN following hormonal changes, including 3 studies (5 patients) during pregnancy, 1 study (2 patients) due to congenital adrenal hyperplasia from 21-hydroxylase deficiency, and 1 study (1 patient) following male-to-female hormone therapy. In the remaining 10 studies, 64 patients had a known history of KCN that progressed in association with hormonal fluctuations affecting the HPG axis.

The most common endogenous trigger of hormonal change was pregnancy, reported in 7 of the 16 studies (22 patients), followed by congenital hormonal abnormalities in 2 studies (4 patients). Exogenous hormonal exposures were reported in 5 studies (7 patients), including hormone replacement therapy (HRT; 2 studies; 2 patients), in vitro fertilization (IVF) (1 study; 3 patients), antiandrogen therapy (1 study; 1 patient), and selective tissue estrogenic activity regulators (agents; 1 study; 1 patient).

One study (28 patients) reported a positive correlation between estriol and KCN progression in patients undergoing corneal collagen cross-linking without any natural or exogenous triggers of hormonal change.⁴⁶

Overall, endogenous hormonal changes accounted for 56.2% (9 of 16 studies; 26 patients) of the reported triggers, while exogenous hormone use accounted for 31.2% (5 of 16 studies; 7 patients) (Table 1).

Case-Control Studies

Ten case-control studies published between 2016 and 2025 included a total of 3834 participants, including 1623 cases with KCN and 2211 controls, representing approximately 7500 eyes. The cohort comprised 1832 females (47.8%) and 2002 males (52.2%), with a weighted mean age of 31.1 ± 13.9 years.

These studies originated from 8 countries: the United States (five studies, including 3 conducted in collaboration with Denmark), Denmark (3 studies, all in collaboration with the United States), and 1 study each from China, Israel, Iran, Malaysia, and Turkey. Most employed unmatched cross-sectional designs (7/10), while 3 used matched case-control methods: Beatty et al⁴⁸ with a large retrospective cohort from the All of Us database, and Escandon et al⁴⁷ and Daphne Teh et al,⁵¹ which were matched by age, sex, and ethnicity. Three were multicenter studies (Karamichos et al,⁵⁶ Sharif et al,⁵⁵ and McKay et al⁶³). Sample sizes ranged from 40 to 2288 participants, with patient ages spanning from the early 20s to late 50s and female representation ranging from 12.6% to 100% across included studies.

Hormonal assessments of sex steroids and HPG-axis hormones were reported in 8 studies. Estrogen levels were evaluated in 4 studies, including estrone (2 studies), estradiol (4 studies), and estriol (3 studies). Testosterone was measured in 2 studies, dehydroepiandrosterone sulfate (DHEAS) in 4 studies, LH and FSH in 2 studies each, GnRH in 1 study, and progesterone in 1 study. One study evaluated the expression of estrogen, progesterone, and androgen receptors, while another investigated the association between KCN and estrogen-containing medications. Assay methods included enzyme-linked immunosorbent assay, chemiluminescence, radioimmunoassay, Western blotting, quantitative polymerase chain reaction, and untargeted metabolomics. Beatty et al⁴⁸ uniquely used electronic health record-derived medication history as a proxy for hormone exposure.

Keratoconus diagnosis and staging were determined using a combination of clinical signs and imaging, with 9 out of the 10 studies employing corneal topography using Orbscan, Pentacam, or Sirius devices. Three studies employed standardized grading systems such as Belin/Ambrosio enhanced ectasia display, anterior, posterior, corneal thickness, and distance visual acuity, or Amsler-Krumeich. Four studies identified cases using topography criteria, and 1 study used diagnostic codes based on electronic health record. In 3 studies, KCN staging was not reported.

Across these 10 case-control studies, differences in HPG-axis hormones in the setting of KCN were consistently reported. Six studies identified alterations in sex steroid levels, including differences in estrogen levels (4 studies; increased estradiol in 2, decreased estrone and estriol in 2), testosterone levels (2 studies; increased in 1, decreased in 1), and DHEAS (4 studies; increased DHEAS reported across all 4).^{31,45,50,51,55,63} Three studies linked disruptions in the gonadotropin pathway, such as reduced GnRH (1 study), reduced FSH (1 study), and reduced LH/FSH ratios (1 study) to KCN.^47,50,56 Two articles did not measure hormonal levels directly, including 1 study that highlighted a direct association between consuming estrogen-containing medications (572 KCN cases) and KCN and another study that reported an increase in both estrogen and androgen receptor expression in KCN (30 KCN cases)^48,58 (Table 2).

Among quantitative hormonal measurement studies, the most frequently reported changes were an increase in DHEAS, observed in 4 of 4 studies (307 KCN cases),^50,51,55,63 followed by a decrease in estrone (E1) in 2 of 2 studies (211 cases),^55,63 and an increase in estradiol (E2) in 2 of 2 studies (86 cases).^31,45 Weaker associations were observed for lower GnRH (1/1 study, 501 patients) and lower estriol (E3) (1/1 study, 147 patients).^47,55 A summary of the weighted directional shifts for all hormones measured in case-control studies is presented in Table 3.

Table 3.

Summary of Hormonal Changes in Keratoconus

Hormone	Number of Case-Control Studies	Accumulated Sample Size (KCN Cases/Controls)	Absolute Values for Hormonal Changes by Studies (Ns)	Weighted Direction of Hormonal Change by Cases (N)
Estrone (E1)	255,63	211 KCN/74 Controls	Decrease (2 Studies)	Decrease (211 KCN)
Estradiol (E2)	431,45,55,63	309 KCN/254 Controls	Increase (2 Studies) Unchanged (2 Studies)	Increase (86 KCN) Unchanged (223 KCN)
Estriol (E3)	331,55,63	273 KCN/194 Controls	Decrease (1 Study) Unchanged (2 Studies)	Decrease (147 KCN) Unchanged (126 KCN)
Progesterone (P)	131	62 KCN/120 Controls	Unchanged (1 Study)	Unchanged (62 KCN)
Testosterone (T)	231,50	138 KCN/146 Controls	Increase (1 Study) Decrease (1 Study)	Increase (76 KCN) Decrease (62 KCN)
DHEAS	450,51,55,63	307 KCN/120 Controls	Increase (4 Studies)	Increase (307 KCN)
GnRH	147	Plasma Sample: 227 KCN/58 Controls Saliva Sample: 274 KCN/101 Controls	Decrease (1 Study)	Decrease in plasma (227 KCN) Decrease in saliva (274 KCN)
LH & FSH	250,56	162 KCN/71 Controls	LH Unchanged (2 Studies) FSH unchanged (1 Study) FSH decrease (1 Study) LH/FSH ratio decrease (1 Study)	LH Unchanged (162 KCN) LH/FSH ratio decrease (63 KCN) FSH unchanged (86 KCN) FSH decrease (38 KCN)

DHEAS = dehydroepiandrosterone sulfate; E1 = estrone; E2 = estradiol; E3 = estriol; FSH = follicle-stimulating hormone; GnRH = gonadotropin-releasing hormone; KCN = keratoconus; LH = luteinizing hormone; No. = number; P = progesterone; T = testosterone.

Two studies stratified their results by sex but reported conflicting findings on testosterone changes in KCN: one observed higher testosterone in both males (N = 38) and females (N = 38),⁵⁰ while the other found lower testosterone in both sexes (males: N = 50, females: N = 12) along with elevated estradiol in male patients.³¹

Some of the included studies also evaluated other biomarkers, with 1 study noting elevated prolactin levels in addition to HPG-axis changes and 4 studies reporting increased proinflammatory and matrix remodeling markers, such as matrix metalloproteinase, prolactin-induced protein, eicosanoids, myoinositol, and 1-methyl-histidine.

Quality Assessment

The methodological quality of all included studies was appraised using the JBI critical appraisal tools, tailored to the specific design of each study. Among the case-control studies (n = 10), JBI scores ranged from 6 to 10 out of 10, with a mean score of 8. Based on the risk-of-bias classification, 8 studies were considered at low risk, and 2 studies at moderate risk. No studies were rated as high risk of bias.

For the cohort studies (n = 3), scores ranged from 8 to 10 out of 11, with a mean score of 9. All 3 studies were rated as having a low risk of bias, with none falling into the moderate- or high-risk categories.

In the group of case series (n = 3), scores ranged from 8 to 10 out of 10, with a mean score of 8.7. Applying the established thresholds (scores of 7–10 indicating low risk), all studies were classified as low risk of bias.

For the case reports (n = 10), scores ranged from 5 to 8 out of 8, with a mean score of 7.1. Of these, 8 studies were classified as low risk, and 2 studies as moderate risk; no studies were rated as high risk. A detailed breakdown of the quality assessment results for each study is provided in supplementary Appendices B to E (available at www.ophthalmologyscience.org).

Discussion

The present systematic review demonstrates consistent associations between systemic sex hormone levels and the pathogenesis of KCN. Case-control studies link DHEAS, estradiol (E2), and estrone (E1) alterations to KCN, while descriptive reports suggest that fluctuations in estrogen and progesterone—particularly during pregnancy or hormone therapy—may contribute to progression. These findings support the concept that sex-hormone fluctuations can modulate KCN and highlight systemic hormonal dysregulation as a contributor to disease pathogenesis.

Peripartum and postpartum hormonal changes have been associated with KCN onset or worsening, consistent with the presence of estrogen, progesterone, and androgen receptors throughout the corneal layers.^22,70,71 While our analysis focused on clinical evidence, in vitro and ex vivo work further support these clinical observations and provide insight into the biological basis of these findings. These studies have demonstrated differential receptor expression in KCN corneas and receptor upregulation after exposure to estrogenic ligands, suggesting a plausible biological mechanism.72, 73, 74 Additional ex vivo animal data show estrogen-induced reductions in corneal biomechanical stiffness.^75,76

Descriptive studies in this review evaluated hormonal imbalance in the context of HRT, IVF, gender-affirming therapy, pregnancy, and congenital adrenal hyperplasia. Among the 20 included pregnant patients who had specified timing documented for the development or progression of KCN, 18 progressed in the second/third trimester, while 2 progressed postpartum, suggesting late-gestational or cumulative hormonal effects on disease progression, consistent with in vitro evidence demonstrating the presence of estrogen and androgen receptors in the cornea and that hormone stimulation increases receptor expression more in KCN tissue than in controls.^22,70, 71, 72, 73, 74 Reports of progression during HRT and IVF, both associated with substantial estrogen surges, further support this relationship.^60,68,77,78 However, whether hormonal fluctuations alone can trigger or accelerate KCN remains uncertain.⁶⁸ The available data suggest that hormonal effects may be more clinically significant in biomechanically compromised corneas.

Across case-control studies, the most consistent hormonal differences were elevated DHEAS, reduced estrone, and increased estradiol relative to controls. Associations with lower GnRH or estriol were weaker, and findings for testosterone, progesterone, LH, and FSH were inconsistent. Dehydroepiandrosterone sulfate can undergo local conversion to androgens or estrogens, contingent on tissue-specific enzyme expression, steroid sulfatase activity, cofactor availability, and binding proteins.^82,83 However, despite the consistently elevated DHEAS observed across our included studies, downstream sex steroids did not show parallel, directionally consistent changes, with findings for testosterone and estrogens (estrone, estradiol, and estriol) being mixed. Variation in sample types, assay methods, timing of collection, and unmeasured confounders (i.e., body mass index, thyroid/androgen status, and medications) likely contribute to these discrepancies, emphasizing the need for standardized measurement protocols.84, 85, 86

Endocrine and inflammatory pathways appear to intersect in KCN, with several studies reporting elevated tear cytokines and matrix metalloproteinase activity as well as increased prolactin or prolactin-induced protein.^55,87,88 Cortisol fluctuations have also been investigated as a potential contributor to KCN progression.^79,80,81 These data support a multifactorial model in which sex-hormone alterations interact with mechanical, genetic, endocrine, and oxidative pathways to influence disease progression.

Clinically, our findings reinforce that KCN is not solely a localized corneal disorder but one that may be hormonally modulated in susceptible eyes. Risk assessment should incorporate systemic hormonal context, particularly during periods of endocrine volatility, to guide management and timing of interventions such as corneal cross-linking.

Current evidence justifies targeted surveillance of populations at heightened risk of hormonally mediated progression, which includes individuals who are pregnant, are exposed to exogenous hormone therapy (HRT, gender-affirming hormonal treatments, or IVF), or have a history of congenital endocrine disorders. In particular, patients who have a confirmed or suspected history of ectasia and who are (1) pregnant or (2) initiating or undergoing exogenous sex-hormone exposure should be monitored closely. For these patients, a pragmatic monitoring framework should incorporate ophthalmic assessment and interval observation. For pregnant individuals, evaluation should occur at confirmation of pregnancy and again during the second and third trimesters with extended follow-up maintained through the postpartum period to capture late peripartum changes. In patients undergoing exogenous hormone therapy, evaluation should occur before treatment and at 3- to 6-month intervals during treatment depending on clinical suspicion and judgment. While this surveillance strategy targets those with pre-existing suspicion of corneal ectasia, any vision changes in the setting of pregnancy, exogenous hormone therapy, or congenital endocrine disorders should prompt an ophthalmic consultation.

Incorporating targeted surveillance strategies for these groups may enable earlier detection of KCN onset and progression and timely initiation of interventions such as corneal cross-linking. Equally important is counseling patients about the potential impact of hormonal changes on visual health and the importance of seeking ophthalmic care if new symptoms arise. Specialists who see these patients before an ophthalmologist (i.e., obstetricians, gynecologists, endocrinologists, etc.) should be aware of these risks and counsel patients accordingly.

This study has certain limitations inherent to systematic reviews, including publication bias and potential overrepresentation of rare or severe presentations. Additionally, as clinically uneventful presentations are less frequently documented and therefore underrepresented in the literature, cases of KCN that remain stable and do not progress with hormonal fluctuations are likely underreported. The retrospective nature of many included studies, variability in clinical documentation, and inconsistent diagnostic criteria limited our ability to conduct meta-analyses or establish causality. Hormone level measurements were not uniformly reported, and no randomized clinical trials were identified. Most studies did not assess cortisol or other relevant hormones such as thyroid hormones or prolactin alongside HPG-axis hormones, even though these hormones can influence circulating HPG-axis hormone levels and may therefore confound the observed associations.^50,89, 90, 91 These limitations highlight the need for larger, prospective studies that systematically evaluate sex hormone fluctuations alongside other hormonal changes to better understand their combined effects on KCN development and progression.

Despite these limitations, this review provides the most comprehensive and largest evidence to date on the association between KCN and HPG-axis hormonal changes.

Conclusion

Our findings suggest that hormonal fluctuations, both natural and exogenous, may contribute to corneal instability in susceptible individuals, with pregnancy emerging as a particularly high-risk period. These findings can aid in screening strategies and increase physician awareness when managing patients with a history of KCN undergoing systemic hormonal changes.

Declaration of Generative AI and AI-Assisted Technologies in the Writing Process

During the preparation of this work, the authors used ChatGPT for minor language editing. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the published article.

Manuscript no. XOPS-D-25-00793.