Acute corneal hydrops (ACH) is characterized by sudden-onset corneal edema, typically in the setting of a corneal ectatic disorder. It is a vision-threatening complication of keratoconus that can result in permanent scarring and may even be the presenting sign of keratoconus. A clinical description of ACH in keratoconus patients can be found as early as 1854 in the treatise by Dr. John Nottingham. The term “acute hydrops,” however, was not introduced until 1940 when Drs. Ralph Rychener and Daniel Kirby were able to reproduce acute corneal edema by surgically inducing breaks in the posterior corneas of rabbits.
Epidemiology and Risk Factors
The reported incidence of ACH varies widely, from 0.2% to 2.8% in keratoconus patients. , Notably, it may occur in up to 11% of patients with keratoglobus and pellucid marginal corneal degeneration. , Most patients present in the third or fourth decade of life. Some studies have shown that males with keratoconus have nearly double the risk of ACH compared with females. ,
Atopic disease and eye rubbing have been identified as risk factors for ACH. A recent case-control study identified nonocular atopic conditions as risk factors for the development of ACH in keratoconus patients. In addition to its association with atopy, eye rubbing may also be an independent risk factor. Not surprisingly, patients with steeper keratometry prior to hydrops were also found to be at increased risk of ACH. , There are, however, a number of patients whose first presentation to an ophthalmologist is after an acute episode of hydrops and whose baseline visual acuity and keratometry are unknown.
Patients with prior ACH in one eye have a greater than six-fold increased risk of hydrops in the other eye. ACH is found at increased rates in patients with learning disabilities and is more likely to present bilaterally in patients with Down syndrome. , A retrospective case series in New Zealand has found that those of Pacific ethnicity had a higher risk of hydrops compared with White Europeans. In a population-based survey in the United Kingdom, South Asian and Black patients had a higher prevalence of both keratoconus and ACH.
Imaging, particularly with anterior segment optical coherence tomography (AS-OCT), has revealed anatomic risk factors for the development of ACH. Corneas with epithelial thickening, stromal thinning at the keratoconus cone, hyperreflective anomalies within Bowman’s layer (BL), and an absence of corneal scarring are more likely to experience an episode of ACH.
Acute hydrops presents clinically as a marked decrease in visual acuity accompanied by pain and photophobia. Slit-lamp examination will reveal an edematous cornea, which may be Seidel positive ( Fig. 37.1 ). , In most cases, the edema resolves spontaneously over the course of weeks to months. Deturgescence is thought to occur after endothelial cells migrate to cover the exposed stroma. There is often a flattening of the cornea following an episode of ACH, which may induce a beneficial refractive change and allow for improved contact lens fitting in keratoconic eyes. ,
As the cornea heals, haze and scar formation may occur, which may be vision-impairing. A worse prognosis is seen in cases with large areas of the cornea involved, or in episodes with a prolonged duration of edema. Prognosis is particularly poor for corneas that develop neovascularization. In episodes of ACH with edema lasting longer than 11 weeks, corneal neovascularization may be seen in up to 75% of patients. Additional risk factors for corneal neovascularization include the severity of hydrops, edema located at the limbus, and the presence of intrastromal clefts. , Vision-limiting scars that cannot be aided by a contact lens may require corneal transplantation.
Classically, ACH has been thought to occur after a spontaneous break in Descemet membrane (DM) and the endothelium. Early reports using slit-lamp examination of patients with keratoconus and histopathologic samples taken from patients undergoing penetrating keratoplasty (PKP) after ACH were able to identify breaks in the DM of some patients. Since then, there have been many descriptions of DM scrolls in ACH.
Intraocular pressure (IOP) has been suggested as an etiologic factor in ACH. IOP elevated to approximately twice the baseline level has been shown to significantly increase corneal steepening in keratoconic eyes but does not create any significant change in normal corneas. Eye rubbing, a known risk factor for keratoconus and an independent risk factor in the development of ACH, also transiently increases IOP. A recent in vivo study on nonhuman primates showed that eye rubbing can cause IOP increases that exceed 300 mmHg and average 109 mmHg above baseline. These repeated increases in IOP may increase the risk of ACH development.
There is some evidence that, in addition to DM, disruption of a layer of posterior stroma may be involved in some cases of ACH. Stone et al. published the first case series that examined ACH with electron microscopy, demonstrating a tear in DM that curled anteriorly around a small amount of separated stroma. More recently, a role of posterior stromal rupture has been suggested by Parker et al. Two groups of patients with keratoconus were compared: (1) patients with coexisting Fuchs dystrophy who were undergoing Descemet membrane endothelial keratoplasty (DMEK) and (2) patients undergoing BL transplantation who had inadvertent rupture through the posterior stroma. All eyes with inadvertent rupture through the posterior stroma during BL transplantation developed ACH, but edema was not seen in those patients undergoing DMEK.
In reviewing published histopathology, we have encountered multiple examples where DM rupture appears to be accompanied by adherent posterior stroma. , , Though the scrolling of DM is thought to be due to its elastic nature, evaluation of the mechanical properties of rabbit corneas showed that DM is relatively stiff compared with the rest of the cornea, whereas the posterior stroma is the least stiff corneal tissue. Taken together, there is evidence that the posterior stroma may be ruptured in some cases of ACH and may participate in its pathogenesis. It is unclear if involvement of the posterior stroma may affect prognosis or preferred management.
The diagnosis of ACH is made via the history of present illness, focused clinical examination, and corroborating imaging.
Cases of ACH are often described as spontaneous, although they may be unknowingly precipitated by episodes of eye rubbing, coughing, sneezing, and other potential mechanisms of IOP elevation. There is evidence of bilateral involvement in up to 18% of cases, suggestive of a direct causative mechanism.
Patients present with acutely decreased vision, photophobia, and pain. They may have a known history of keratoconus, although in many cases, ACH may be the initial presentation of keratoconus. In a study of New Zealanders with ACH, the acute episode was the first visit to an eyecare provider in 30% of patients. This may be even higher in regions without a national public health service. When obtaining a history, emphasis should be placed on the presence of known risk factors of ACH, including a history of atopy and eye rubbing.
In addition to visual acuity, IOP (which may be artificially low in the presence of corneal edema), and pupil responses, particular attention should be paid to anterior segment examination. The degree of edema and extent of corneal involvement should be measured. A tear involving DM may be observed, although corneal edema may preclude detailed evaluation of the posterior cornea. Seidel testing should be performed and, when positive, may indicate transudation of aqueous humor through the cornea, rather than a perforation. An ectatic appearance may be observed in either the involved or contralateral cornea.
AS-OCT can reveal rupture of DM as well as other corneal abnormalities including possible posterior stromal rupture, and can determine the extent of detachment from the rest of the cornea. , These findings may provide useful prognostic indicators, as eyes with larger DM detachments have been shown to require more time for resolution of hydrops and are subsequently at increased risk for scarring. Likewise, the presence of a tear involving the posterior stroma may decrease the likelihood of spontaneous clearance. AS-OCT of an edematous cornea may produce shadowing that can preclude detailed imaging of the posterior stroma, and repeat AS-OCT images may be necessary to fully visualize the extent of pathology ( Fig. 37.2 ). In addition to aiding diagnosis, AS-OCT may be a useful approach for monitoring resolution after treatment. Serial AS-OCT images may be used to localize the DM (and possible posterior stromal) tear, demonstrate later reattachment of DM, follow resolution of edema, monitor intrastromal clefts, identify areas of absent DM, and visualize scar formation.
In vivo confocal microscopy (IVCM), another noninvasive imaging modality, allows for detailed evaluation of the cornea at the cellular level. Serial examination with IVCM of eyes with hydrops has been used to follow the presence of presumed inflammatory cells in the cornea. Their presence may indicate a poorer prognosis. Although they are found only in a minority of cases of ACH, their prolonged presence beyond 4 weeks is associated with subsequent development of neovascularization. Confocal microscopy requires greater technical skill than other imaging techniques and may be particularly challenging to interpret in an eye with hydrops.
Imaging during ACH by ultrasound biomicroscopy (UBM) may reveal a DM tear, seen as the absence of the normally brightly intense DM spike, even when evaluation of the posterior cornea is precluded by other methods. , This is also a useful method for evaluation of intrastromal clefts and monitoring of DM reapposition. ,
Corneal tomography provides detailed evaluation of the anterior and posterior corneal surfaces as well as assessment of the distribution of corneal thickness. Corneal tomography has characteristic findings in keratoconus and can be used to monitor progression of the disease. Tomography of the uninvolved eye may be particularly useful for diagnosing ACH in patients who present in the absence of a known history of ectasia.
Many potential treatments have been tried to decrease the duration and visual impact of ACH. In the first half of the 20th century, treatment was aimed primarily at decreasing IOP, and was accomplished by sclerectomy, keratotomy, iridectomy, and frequent paracentesis. , Despite these various treatments, it was noted that visual improvement seemed to follow the same course in most cases. Although our understanding of ACH has improved, its treatment remains quite varied.
Because many episodes of ACH resolve spontaneously within 2 to 4 months, early treatment is often conservative. There is a lack of case-control studies of topical treatments for ACH, and evidence of the efficacy of topical treatments is largely anecdotal. , Treatment should be started promptly, on the day of presentation. Initial management depends on the presence of Seidel positivity. In cases of aqueous leak, we recommend consideration of an aqueous suppressant and combination antibiotic–steroid ointment with pressure patching between instillation of medications. Patients should be followed closely, every few days, until the cornea is found to be Seidel negative. After resolution of the aqueous leak, this initial regimen may be discontinued.
Following management of Seidel positivity or in its absence, frequently used topical measures include ocular antihypertensive therapy, hypertonic saline, cycloplegia, and an antibiotic. Topical corticosteroids have been added in some cases to inhibit the development of corneal neovascularization, but previous studies of ACH complicated by neovascularization did not demonstrate an inhibitory benefit from steroids. Our preferred regimen is hypertonic saline 5% drops four times daily and hypertonic saline 5% ointment at bedtime. In patients with photophobia, cycloplegia may be helpful. In cases that do not resolve within the first few weeks, we recommend adding prednisolone acetate 1% four times a day. At this stage, we typically see patients every 1 to 3 weeks. Serial imaging is helpful in monitoring resolution. We recommend consideration of AS-OCT at each visit if possible. AS-OCT may reveal reattachment of DM and improvement in corneal edema in cases that resolve with conservative medical management. Imaging may also guide the need for a procedural intervention. In addition to AS-OCT, corneal tomography should be obtained in the fellow eye, including in cases without a known history of ectasia.
To summarize, because hydrops often improves without procedural interventions, we recommend conservative medical management with close follow-up initially, followed by evaluation every 1 to 3 weeks. Patients are evaluated for improvement based on symptoms, clinical examination, and imaging. If the edema fails to resolve, or if the level of corneal haze stabilizes without further improvement, procedural intervention may be considered.
Although patients with keratoconus tend to have a very favorable prognosis following corneal transplant, the risk of endothelial rejection is increased in eyes that have a history of hydrops. , The risk of graft rejection is further increased by the presence of neovascularization. The goal of procedural intervention is two-fold: to hasten DM reattachment and corneal deturgescence in cases with persistent edema, and to address scar formation after resolution of ACH.
If AS-OCT reveals that DM is detached and widely separated from the rest of the cornea, a pneumatic descemetopexy may be considered to aid in reattachment. The use of air or gas injection into the anterior chamber has been used to aid in reapposition of DM. Early trials of acute hydrops treated with intracameral injection of air found that treatment shortened the period of corneal edema compared with conservative medical treatment. Similar effects have been shown with perfluoropropane (C3F8) and sulfur hexafluoride (SF6) gas injection. The benefit of gas injection is its longer duration in the anterior chamber, which may reduce the need for reinjection, but likewise patients require a longer duration of supine positioning and have increased risk for complications. Gases should be used at the isoexpansile concentration of 14% C3F8 or 20% SF6 with a typical volume of 0.2 mL injected into the anterior chamber.
There are two potential mechanisms for hastening resolution after air and gas injection. One possibility is that the air or gas acts as a tamponade, preventing aqueous humor from entering the corneal stroma. The air or gas may also act by unrolling and reapproximating the DM tear to allow for faster migration of endothelial cells over the area of rupture. Though air and gas may speed edema resolution by approximately 1 month, there appears to be no difference in final visual acuity or need for transplantation. Adverse events after the use of air or gas are uncommon but include pupillary block glaucoma, intrastromal migration of gas, cataract, and endothelial cell loss. Patients should be followed closely until the bubble resorbs.
Corneal compression sutures have been described as another method for attempting reapproximation of the rupture, both in isolation and in combination with intracameral gas. , With either technique, corneal edema was found to resolve significantly faster, often within the first week, and appeared to offer particular benefits in corneas with intrastromal clefts. Both full thickness , and pre-DM sutures have been used with success. Identifying the DM tear may be improved intraoperatively with air injection. When visualized, 10-0 nylon sutures may be placed perpendicular to the tear. The number of sutures used is variable (between 2 and 13), partly depending on the extent of the tear.
New evidence suggests a potential role for intracameral injection of eye platelet-rich plasma (E-PRP) in the management of ACH. E-PRP is an autologous conditioned plasma that contains high levels of growth factors and cytokines and has been used for the treatment of ocular surface disorders. In a single case report using intracameral E-PRP during ACH, resolution of edema occurred within the first week.
Treatment of visually debilitating scar formation after ACH resolution is traditionally accomplished with PKP. Success of PKP is reduced in keratoconus patients following ACH, and further reduced in patients who have developed corneal neovascularization. , Given the relatively young age of ACH patients and the significant risks of PKP, including long-term graft failure, the use of partial thickness corneal transplantation has gained popularity in recent years. Modifications of deep anterior lamellar keratoplasty (DALK) to replace stroma up to the pre-Descemet layer have been used successfully to restore vision in patients after ACH.
Recent evidence suggests that early endothelial keratoplasty (EK) after an episode of acute hydrops may hasten deturgescence, restore vision, and may theoretically decrease the need for later PKP or DALK. Descemet stripping automated EK (DSAEK) , and DMEK , have been used successfully to resolve edema in ACH cases with large DM detachment or with suspected posterior stromal rupture. By restoring the normal posterior corneal anatomy and endothelial pump function, EK may decrease the duration of edema and reduce the formation of stromal scarring that would otherwise require more extensive transplantation. Following EK, vision may improve over the course of weeks, and visual acuity has been preserved without the need for further intervention. Patients may need a rigid contact lens to achieve best-corrected acuity.
Despite its relative rarity, ACH is a vision-threatening complication seen in the setting of corneal ectasia. It presents acutely, often without any obvious precipitating factor, with marked corneal edema, pain, and photophobia. The etiology of ACH is traditionally described as DM rupture in a susceptible cornea, but careful review of established literature as well as recently reported observations in keratoconus patients have suggested that the posterior stroma may be involved in some cases. Typically, ACH is self-limited with deturgescence occurring in weeks to months, and medical management may include the use of aqueous suppressants, hypertonic saline/ointment, cycloplegia, and topical steroids. Corneal imaging, including AS-OCT, can be helpful in disease assessment, including identifying possible posterior stromal rupture, and in documentation of resolution. Various procedural interventions, including intracameral injection of air or gas, compression sutures, and EK, have been employed to hasten recovery and reduce scar formation. Traditionally, patients have undergone PKP for vision-limiting scars, but recent case series have shown successful visual outcomes with modified DALK. Many of the current medical and surgical therapies used in the treatment of ACH have not been studied systematically, and further research is needed to determine optimal treatment strategies.