ATOPY

Over the past half-century, much has been written linking keratoconus to atopic disease. The Dunedin University Scottish Keratoconus Study of 200 keratoconus patients found that the prevalence of asthma (23% of keratoconus vs. 6% controls) and hay fever (30% keratoconus vs. 16% controls) was higher than in the control cohort. These results were further strengthened by the Collaborative Longitudinal Evaluation of Keratoconus (CLEK) study, which noted the prevalence of hay fever to be 53% among the over 1200 keratoconus patients studied. ^, Similarly, Rahi et al. found a significantly higher rate of atopy in keratoconus than seen in controls (35% vs. 12%). A significant correlation in the severity of atopy and keratoconus has also been observed. ^, Notably, Tuft et al. reported that severe allergic eye disease among keratoconus patients tended to reduce the time to penetrating keratoplasty (PKP) ( P = 0.06).

A recent meta-analysis of 29 studies encompassing over 50 million subjects by Hashemi et al. found allergy, asthma, and eczema to be significant risk factors for keratoconus with odds ratios (ORs) of 1.42 (95% confidence interval [CI]: 1.06–1.79), 1.94 (95% CI: 1.30–2.58), and 2.95 (95% CI: 1.30–4.59), respectively. Although atopy was positively associated with keratoconus, the linkage was not statistically significant (OR: 1.12; 95% CI: 0.40–1.85). It was postulated that inconsistent definition of “atopy” contributed to the conflicting results across the referenced reports.

Atopic disease likely contributes to keratoconus through eye rubbing associated with allergic symptoms and atopic pruritus. Atopy is one of the primary causes of eye rubbing, which in turn is one of the most well-established environmental risk factors for the development and progression of keratoconus. It is even plausible that eye rubbing itself, and not the atopy, is the culprit. However, it would be difficult to establish causality. Indeed, there are several reports of unilateral keratoconus secondary to chronic unilateral eye rubbing. ^{, ,}

The pathogenesis of eye rubbing in keratoconus is controversial. It likely serves as the second hit in a two-hit hypothesis in genetically susceptible individuals. Chronic eye rubbing can cause mechanical wear on the cornea that leads to progressive ectasia. ^, Moreover, recurrent epithelial trauma because of eye rubbing can perpetuate a chronic inflammatory event. In turn, this activates inflammatory mediators such as interleukin (IL)-6, tumor necrosis factor alpha (TNF-α), and matrix metalloproteinase-9 (MMP-9), which are overexpressed in the tears of keratoconus patients. ^, Eye rubbing in keratoconus induces IL-1-mediated apoptosis through the binding of Fas ligand by keratocytes. Keratocytes in keratoconus patients have four times the density of IL-1 receptors compared with healthy controls. Notably, keratoconic eyes have been observed to respond differently to eye rubbing than normal eyes, exhibiting higher degrees of posterior astigmatism, intraocular pressure change, and anterior chamber volume change. Taken together, eye rubbing may represent the common pathway in the progression of keratoconus across a range of conditions such as atopy, contact lens wear, or oculodigital stimulation secondary to Down syndrome or Leber congenital amaurosis (LCA).

OBSTRUCTIVE SLEEP APNEA

Obstructive sleep apnea (OSA) is a clinical syndrome characterized by recurrent episodes of apnea and hypopnea during sleep and is often associated with daytime somnolence. OSA is an independent risk factor for stroke and death. ^, Floppy eyelid syndrome (FES) typically presents with lax eyelids that evert with minimal traction and chronic papillary conjunctivitis of the upper palpebral conjunctiva. Virtually all patients with FES have OSA, but only a subset of OSA patients have FES. The association between keratoconus and FES is well established, and laterality in both tends to coincide with sleeping side preference. Frequent eye rubbing has also been suggested as a common causative factor in both conditions, as supported by asymmetric disease progression owing to same-side sleeping or asymmetric eye rubbing.

Multiple studies have observed that keratoconus patients are more likely to suffer from sleep apnea or be at high risk for it based on the Berlin questionnaire. Ezra et al. found a strong association between keratoconus and FES (OR: 19.3), and even more markedly so in patients with unilateral keratoconus (OR: 35.0). Gupta et al. observed that of the patients who had undergone prior PKP for keratoconus but were not yet diagnosed with OSA, 69% scored as high risk for developing OSA on the Berlin questionnaire compared with only 21% of milder keratoconus patients who had not undergone surgery (OR: 8.6; 95% CI: 2.8–26.6). Moreover, the average body mass index (BMI) of those postkeratoplasty patients was significantly higher than those who had not undergone PKP (33.9 vs. 26.4 kg/m ² ; P < 0.0001).

Nocturnal eyelid eversion in FES is thought to result in chronic microtrauma to the cornea and conjunctiva, especially with pillow-ocular surface interface. Upregulation of MMP activity has been associated with both FES and keratoconus and may be the pathophysiologic link between the two conditions. ^, Immunohistochemistry studies in patients with keratoconus and FES have shown an increased immunoreactivity for MMP-7 and MMP-9 compared with controls, potentially pointing to a shared common pathway in perpetuating connective tissue degradation.

Ophthalmologists should refer patients with FES for nocturnal polysomnography (“sleep study”) to evaluate for occult OSA. Treatment of OSA with continuous positive airway pressure (CPAP) therapy or surgical uvulopalatoplasty has been shown to reduce the signs and symptoms of FES and reduce systemic sequelae.

CONNECTIVE TISSUE DISORDERS

Keratoconus has long been associated with noninflammatory connective tissue disorders, such as Marfan syndrome, Ehlers-Danlos syndrome, ^, osteogenesis imperfecta, and mitral valve prolapse (MVP). Robertson found the prevalence of hypermobility in joints to be 50% in 44 consecutive keratoconus patients. Subsequent studies also observed increased joint hypermobility, particularly in the metacarpophalangeal and wrist joints. ^, However, more recent studies have failed to observe an association of joint hypermobility among keratoconus patients. ^,

Biochemically, keratoconus may reflect both a decrease in collagen content, as well as localized dysfunction in corneal collagen elasticity owing to altered enzymatic metabolism. It has been proposed that the corneal tissue of such individuals with a genetic predisposition is more easily weakened by oxidative corneal stress caused by sustained mechanical trauma, eventually leading to corneal ectasia. In fact, studies on Marfan syndrome demonstrate a reduction in corneal hysteresis and progressive thinning.

MVP is a connective tissue abnormality of the leaflets, chordate tendinea, and annulus of the mitral valve, diagnosed through cardiac echocardiography. An increased prevalence (22.6%–58%) of MVP has been observed in keratoconus patients, with a positive correlation between disease severity and that of corneal disease. ^, In a case-control study, Rabbanikhah et al. found the prevalence of MVP in patients with corneal hydrops secondary to keratoconus to be 65.6% compared with 9% in controls, and the risk of developing MVP in corneal hydrops patients was also much higher (OR: 26.7; 95% CI: 9.5–75.2). However, in a study of 95 keratoconus patients, Street et al. did not find a significant association between keratoconus, MVP, and joint hypermobility. Moreover, a Danish national registry did not find a significant association between keratoconus and mitral valve disorder (OR: 1.00; 95% CI: 0.52–1.92). Future prospective studies are thus warranted to better delineate this association.

IMMUNE DISORDERS

Keratoconus has historically been considered a non-inflammatory condition. However, more recently, the immune system has been implicated in its pathogenesis. The tears of keratoconus patients have been observed to exhibit an overexpression of inflammatory mediators, including IL-6, TNF-α, and MMP-9, when compared with healthy controls. ^, The increased activity of these intracorneal inflammatory mediators has been proposed as a possible mechanism for stromal degradation and reduced biomechanical stability in keratoconus.

Rheumatoid arthritis (RA) is most commonly associated with keratoconjunctivitis sicca, anterior scleritis, or peripheral ulcerative keratitis (PUK). However, in a retrospective case-control study, Nemet et al. found an increased risk of keratoconus in RA patients (OR: 8.1; 95% CI: 1.5–44.2), although the numbers were so small that these findings need to be replicated in a larger study. The chronic epitheliopathy secondary to the upregulation of inflammatory mediators in RA was proposed as the predisposing factor for the development of keratoconus.

Other immune disorders were also linked to keratoconus by Nemet et al., including ulcerative colitis (OR: 12.1; 95% CI: 1.3–116), autoimmune chronic active hepatitis (OR: 6; 95% CI: 1.01–36), Hashimoto thyroiditis (OR: 2.0; 95% CI: 1.2–3.3), and irritable bowel syndrome (OR: 5; 95% CI: 2.1–12.1). Notably, Kahan et al. found an excess of thyroxin in tears of patients with keratoconus, independent of their thyroid function, at 2 to 50 times higher concentration than that of healthy subjects. It has been proposed that excess levels of thyroxin in tears may affect corneal metabolism and, in turn, corneal structural integrity. These findings are still preliminary, however, and need to be replicated with larger numbers of participants.

DOWN SYNDROME

A range of chromosomal abnormalities have been associated with keratoconus, most commonly Down syndrome and Turner syndrome. Most series have reported a 5.5% to 15% prevalence of keratoconus among people with Down syndrome. ^{, ,} A retrospective longitudinal cohort study by Woodward et al. also noted increased odds of keratoconus in patients with Down syndrome (OR: 6.22; 95% CI: 2.08–18.66). Corneal hydrops in keratoconus patients has also been observed at higher rates in those with concomitant Down syndrome than in isolated keratoconus patients.

The role of genetic abnormalities in trisomy 21, which may underlie the structural or biochemical changes of keratoconus, has been evaluated. The isoenzyme superoxide dismutase SOD1 gene on chromosome 21, which plays a role in the oxidative stress response, demonstrated higher levels of expression in keratoconic corneas compared with healthy corneas. An intronic sequence deletion was found to segregate with disease in two families with keratoconus. However, owing to the overall rarity of this mutation in keratoconus patients, the existence and strength of a true pathogenic association is still unclear.

Keratoconus also occurs with higher frequency among developmentally delayed individuals without Down syndrome. In fact, the prevalence of unilateral disease is considerably higher in this group compared with the general population, plausibly because of oculodigital stimulation in low mentally functioning patients.

DIABETES

The role of diabetes mellitus (DM) in the development and progression of keratoconus has been conflicting. Although most studies have found DM to be protective against keratoconus, ^, these findings have not been universal. A retrospective longitudinal cohort study by Woodward et al. observed that patients with uncomplicated DM had 20% lower odds of keratoconus (OR: 0.80; 95% CI: 0.71–0.90), and patients with severe DM complicated by end-organ damage had 52% lower odds of having keratoconus (OR: 0.48; 95% CI: 0.40–0.58) in comparison with nondiabetic controls. Seiler et al. found that the prevalence of DM was significantly less in keratoconus patients than in controls ( P = 0.037). Kuo et al. noted that although the incidence of keratoconus was not different between diabetic and control groups, the severity of keratoconus, as measured by best-corrected visual acuity (BCVA), was less severe in diabetic patients. Conversely, a retrospective study by Kosker et al., which differentiated between type I and type II DM and stratified by age, observed a positive association in the presence and severity of keratoconus in type II DM, but not in type I. This study found that the odds of developing keratoconus were 1.4 times higher in diabetic versus healthy subjects. Lastly, a meta-analysis of six studies (two reporting protective effects of diabetes, one reporting it as a risk factor, and three reporting a nonsignificant relationship) showed that the odds of developing keratoconus were 23% lower in patients with type II diabetes, albeit not statistically significant (95% CI: 0.50–1.21).

The aforementioned studies have mostly been limited by the lack of differentiation between type I and type II DM, absent documentation of the duration and temporal relationship between keratoconus and DM, and the lack of an objective evaluation of the severity of both diabetes types, as measured by hemoglobin A1c (Hb Alc), and keratoconus, via corneal tomography. Furthermore, given that it would not be ethically possible to evaluate prospectively the effect of increasing glycosylation status and severity of diabetes in stabilizing keratoconus, all studies have been retrospective in nature. Therefore conclusions about the association of diabetes and keratoconus are not necessarily definitive.

Diabetes exerts histological changes on all layers of the human cornea. It has been postulated that prolonged diabetes-mediated hyperglycemia leads to a cascade of non-enzymatic glycation reactions that result in the formation of advanced glycation end products (AGEs). ^, The uninhibited propagation of these chemical substances produces intermolecular cross-links in the ultrastructure of corneal collagen fibers, leading to corneal stiffening. This non–age-related form of collagen cross-linking induced by DM is thought to have a protective effect on the development of keratoconus in humans. In fact, it has been shown that people with DM have increased central corneal thickness, corneal hysteresis, and corneal resistance factor (CRF), possibly reflecting the increased rigidity of diabetic corneas. Animal models have also documented abnormally polymorphous collagen fibril aggregates in the corneal stroma of diabetic monkeys. However, the impact of glucose concentration and duration of exposure in regulating the degree of cross-linking in the human corneal stroma have yet to be elucidated.

GENETIC OCULAR SYNDROMES

Another group of disorders associated with keratoconus include those with abnormal retinal function, leading to oculodigital stimulation. Such disorders include albinism, congenital rubella, Bardet-Biedl syndrome, LCA, neurofibromatosis, retinitis pigmentosa, Laurence-Moon-Bardet-Biedl syndrome, cone dystrophy, tapetoretinal degeneration, and Kurz syndrome. Oculodigital stimulation has been proposed as the shared pathway for the development of keratoconus among these patients, but this remains theoretical.

Many genetically inherited corneal dystrophies have also been observed to occur in association with keratoconus, including lattice dystrophy, granular dystrophy, posterior polymorphous corneal dystrophy, and Fuchs endothelial dystrophy. The association of keratoconus with such a wide range of corneal dystrophies and ocular abnormalities may reflect either a shared underlying genetic defect or a tightly linked network of interacting proteins, with a final common developmental pathway. Further prospective controlled studies are warranted in this area.