In this retrospective study, data from comprehensive ocular surface assessments of patients undergoing allo-HSCT were analyzed. Chronic ocular GVHD prevalence rates were similar to other studies, in which the prevalence was close to 50% . Rates of chronic ocular GVHD vary widely in the literature, partly due to different diagnostic criteria. A distinct prevalence of 33% was reported in a study of 635 patients based on the 2005 NIH diagnostic criteria that included Schirmer’s test . In a prospective multicenter study of patients with chronic GVHD (diagnosed according to the 2005 NIH criteria), the eyes were the third most commonly involved organ, they were affected in 51% of patients at the time of chronic GVHD diagnosis . Chronic ocular GVHD is a frequent and debilitating complication of allo-HSCT that causes prolonged morbidity, affecting activities of daily living and quality of life.
Input from hematologists in the ocular GVHD unit was valuable fostering the collaboration between the ophthalmologists and HSCT physicians resulting in a faster diagnosis, more accurate identification of complications, and a prompt treatment. It represents a task force to prevent vision loss and improve patients’ quality of life.
In our study, the mouth, skin, and eyes were the most frequently involved organs among patients with systemic GVHD. The occurrence of DED, a well-known manifestation of ocular GVHD, may be multifactorial as a combined of the immune effect of HSTC, meibomian gland dysfunction (MGD), immunosuppressive therapy, total body irradiation, and ocular toxicity due to chemotherapy . Systematic GVHD screening is essential for early recognition of ocular GVHD. Flowers et al  recommend routine ophthalmological screenings 3 months and at 12 months after HSCT, as well as at the time of initial chronic GVHD diagnosis at any site. Our protocol included a comprehensive ocular evaluation prior to HSCT, as well as follow-up assessments 3, 6, and 12 months after transplantation. Indeed, an additional assessment is recommended any time patients exhibit ocular symptoms or develop chronic GVHD at any site.
Ocular surface disease following HSCT varies in its clinical presentation, time of onset, and complications and thus requires close and intensive attention from heath care professionals. When distinct eye parameters were evaluated in patients with and without chronic GVHD, all assessments except for TMH (which reflects tear volume) were significantly altered among chronic GVHD patients. This finding might be a marker of reflex tearing, which occurs in the initial compensatory phase of the disease. Dry eye severity has also been found to worsen with time and in some specific periods such as when immunosuppressive drugs are tapered despite treatment among ocular GVHD patients. The ocular surface baseline evaluations revealed corneal involvement and MGD among the ocular GVHD patients included herein. This information, combined with the diagnostic methods reported herein, increased diagnostic performance. Pathak et al.  report the sensitivity and specificity of certain diagnostic tests for ocular GVHD, such as OSDI (44% and 98%, respectively), corneal staining (91% and 54%, respectively) and TBUT (80% and 67%, respectively). A prospective study showed that reflex tearing was exhibited by 86% of patients before HSCT but began to decrease approximately 3 months after HSCT, and that mean Schirmer test values decreased to ≤10 mm within 6 months . Conjunctival involvement in patients with chronic GVHD was characterized by hyperemia (signaling ocular surface inflammation) and cicatricial conjunctivitis as a sequela event. Recent studies found subtarsal fibrosis in several patients with chronic ocular GVHD and associated the condition with worsening corneal epitheliopathy [23, 24]. The status and function of the meibomian glands among ocular GVHD patients was also considered herein. The assessments revealed mucoid secretion and blepharitis, as well as meibomian gland dropout as determined by post-operative meibography. The percentage of meibomian gland area affected has been shown to reflect the severity of ocular GVHD .
Ocular GVHD is a chronic condition frequently associated with exacerbations of inflammation, ocular surface damage, DED, and vision-threatening complications such as cataract, corneal perforation and infection. Previous studies have suggested that epitheliopathy may be more severe in chronic GVHD patients when compared to other causes of DED [16, 25]. In our study, the management of corneal damage with lubricants alone appeared insufficient in cases of severe epitheliopathy and abundant mucoid secretion, and both a topical anti-inflammatory (loteprednol or prednisolone acetate) and mucolytic eyedrop were required. In cases in which these additional treatments failed to control the disease, the next step was to initiate biological substitutes, such as autologous serum tears. In this study, 48% of patients required one or more therapeutic strategies in addition to artificial tears. However, despite interventions, 3% of patients continue to worsen and may develop progressive indolent corneal ulcerations that may require surgical treatment, such as keratoplasty. Fortunately, corneal ulceration leading to perforations in ocular GVHD was uncommon, as has been seen in other studies . Two of the patients included herein required penetrating keratoplasty due to an infectious and neurotrophic corneal ulcer with progressive thinning and eventual perforation. Bandage contact lenses and tissue adhesive glues were applied with no improvement. Bacterial keratitis occurred due to poor surface integrity and intense immunosuppression. Cataract was the most common complication. The relevant risk factors herein included total body irradiation as part of the conditioning regimen and prolonged steroid therapy for the prevention and treatment of chronic GVHD. Inamoto et al. reported a prevalence of cataracts among allo-HSCT patients that ranged from 11–100% . This large variability is attributable to differences in patient populations, conditioning regimens, supportive care, and length of follow up. Cataract surgery likely contributed to some of the improvement in patients’ final visual acuity when it was compared to their worst visual acuity during the study period. Among patients who have undergone HSCT, cataract may co-occur with dry eye and other manifestations of ocular GVHD. The concurrent ocular surface disease should be controlled before any surgical procedures in order to ensure better outcomes.
By comprehensively evaluating this cohort of HSCT patients we could found that although ocular surface manifestation and DED were found in both groups, with and without chronic GHVD, presentation and risk of complications were much worse in the chronic GVHD group, raising the attention for the need of clinical and ocular close evaluation and follow-up. The GVHD unit provided such approach, improving patients’ care.