CLINICAL SCIENCE
Tear Osmolarity in Ocular Graft-Versus-Host Disease Luigi Berchicci, MD, Lorenzo Iuliano, MD, Elisabetta Miserocchi, MD, Francesco Bandello, MD, and Giulio Modorati, MD
Purpose: The aim of this study was to evaluate tear osmolarity in patients with chronic graft-versus-host disease (cGVHD) with ocular involvement.
Methods: In this observational cross-sectional study of 56 patients with ocular cGVHD referred to the tertiary-care Ocular Immunology and Uveitis Service at the San Raffaele Scientific Institute, Milan, from May 2010 to November 2013, we evaluated the following clinical parameters: Ocular Surface Disease Index (OSDI) symptoms questionnaire, tear osmolarity, Schirmer test, tear film break-up time (TBUT), corneal and conjunctival staining. Results: All patients developed systemic GVHD after undergoing allogeneic hematologic stem cell transplantation. Mean osmolarity was 314.0 6 22.1 mOsm/L, mean OSDI score was 26.4 6 21.2, mean TBUT was 6.50 6 4.75 seconds, and mean Schirmer test value was 3.8 6 3.3 mm. Tear osmolarity significantly inversely correlated with TBUT (r2 = 0.681; P , 0.001). Statistically significant inverse correlation was present with the Schirmer test (r2 = 0.203; P , 0.001), and positive correlation with the OSDI score (r2 = 0.188; P , 0.001), but both with low correlation strength. Osmolarity was statistically different in the subgroups according to the Oxford corneal staining scale (P = 0.0006) and to the van Bijsterveld conjunctival staining score (P = 0.006). Conclusions: Tear osmolarity increased in patients with ocular cGVHD, significantly correlated with TBUT and, to a lesser extent, with the Schirmer test value and OSDI. These results emphasize the role of aqueous-deficient and evaporative dry eye disease in patients with cGVHD after undergoing allogeneic hematologic stem cell transplantation. Tear osmolarity may be considered a useful test in diagnostic assessment of dry eye disease associated with cGVHD. Key Words: tear osmolarity, graft-versus-host disease, dry eye disease (Cornea 2014;33:1252–1256)
O
cular chronic graft-versus-host disease (cGVHD) affects 40% to 60% of patients receiving allogeneic hematologic stem cell transplantation (allo-HSCT), and ocular complications
Received for publication June 11, 2014; revision received September 1, 2014; accepted September 3, 2014. Published online ahead of print October 24, 2014. From the Department of Ophthalmology, San Raffaele Scientific Institute, VitaSalute University, Milan, Italy. The authors have no funding or conflicts of interest to disclose Reprints: Luigi Berchicci, MD, Department of Ophthalmology, San Raffaele Scientific Institute, Vita-Salute University, Via Olgettina, 60, Milan 20132, Italy (e-mail: [email protected]). Copyright © 2014 by Lippincott Williams & Wilkins
1252
| www.corneajrnl.com
occur in 60% to 90% of transplant recipients.1–3 Allo-HSCT from human leukocyte antigen-matched donors is an established and potentially curative therapy for a variety of hematologic malignancies, autoimmune diseases, inherited disorders of metabolism, histiocytic disorders, and other malignant solid tumors.4,5 Allo-HSCT, however, has been restricted from broad use by GVHD, a major cause of morbidity and mortality mediated by donor-derived T-cell recognition of host antigens as foreign.6 Ocular cGVHD tends to present within 3 years after allo-HSCT, usually with clinical features of an autoimmune disease (eg, Sjögren syndrome).7–9 Although ocular involvement has distinctive characteristics, it is not considered a unique diagnostic criterion because its isolated occurrence does not confirm the diagnosis of GVHD.4 Ocular cGVHD includes a spectrum of clinical manifestations affecting all layers of the eye, including lids, lacrimal glands, conjunctiva, cornea, and even the posterior segment.1 Dry eye disease (DED), the most common ocular cGVHD manifestation, has been reported in 69% to 77% of patients with systemic cGVHD, and may result in significant morbidity, with a decrease in quality of life.3 The pathological process may lead to secondary epithelial changes such as superficial punctate keratopathy, filamentary keratitis, corneal erosion, and infected or noninfected stromal ulceration.10 Eyelid keratinization of the tarsal conjunctiva may progress to entropion, trichiasis, and atrophy of meibomian glands. Severe ocular surface disease may also worsen to corneal perforation.11 DED is known to be a multifactorial disease of tears and ocular surface, which is characterized by increased osmolarity of the tear film and inflammation of the ocular surface.12 Elevated tear osmolarity has been reported to be a global marker of dry eye, present in both subtypes of the disease (aqueous-deficient dry eye and evaporative dry eye).13,14 Recent research suggested that tear hyperosmolarity may lead to surface epithelium impairment, by activation of proinflammatory cytokines released into tears, and eventually causing cell death through apoptosis.13–17 These surface alterations enhance and support the inflammatory process in patients with dry eye conditions such as ocular cGVHD. Recently, new tear osmometers have become available, resulting in faster and easier measurement of tear osmolarity in various clinical settings.18 Increased tear osmolarity has already been described in several other ocular conditions, including autoimmune diseases.19,20 To the best of our knowledge, this is the first study to selectively evaluate tear osmolarity in patients with ocular cGVHD, as well as its relationship with ocular surface parameters, such as tear film Cornea Volume 33, Number 12, December 2014
Cornea Volume 33, Number 12, December 2014
break-up time (TBUT), Schirmer test value, ocular surface disease index (OSDI), and corneal and conjunctival staining.
METHODS We enrolled all patients affected by ocular cGVHD who consecutively presented to our tertiary-care Ocular Immunology and Uveitis Service at the San Raffaele Scientific Institute, Milan, from May 2010 to November 2013. The study was performed in accordance with the recommendations of the Declaration of Helsinki, and was approved by the institutional Ethics Committee of the San Raffaele University Hospital. Each patient provided written informed consent regarding the use of personal data. Inclusion criteria were age $18 years and, according to the National Institutes of Health (NIH) consensus criteria definition, a new ocular sicca syndrome documented by low Schirmer test values with a mean value of both eyes #5 mm at 5 minutes or a new onset of keratoconjunctivitis sicca by slit-lamp examination with mean values of 6 to 10 mm on the Schirmer test accompanied by distinctive manifestations in at least one other organ.4 All 56 enrolled patients had developed systemic cGVHD after undergoing allo-HSCT, classified by an internal physician of the Hematology Department of our hospital according to the consensus criteria defined by NIH.4 Exclusion criteria were inability to complete the proposed questionnaire, any ocular surgical procedure in the study eye during the previous 6 months, presence or history of any other ocular surface disorders other than ocular cGVHD (eg, corneal dystrophy, pterygium), history of contact lens wear during 7 days before tear osmolarity assessment, the presence of any other systemic disease that could affect the ocular surface, presence of any medications that could affect the ocular surface (orally administered analgesic drugs, antidepressants, neuroleptics, anti-Parkinson drugs, antihistamines, antihypertensive drugs, estrogen derivatives, topical glaucoma agents, topical nonsteroidal antiinflammatory drugs), except systemic and topical immunosuppressive and/or corticosteroid therapy for GVHD. Demographic characteristics, primary hematopoietic disease, systemic treatment received after allo-HSCT (immunosuppressants and/or corticosteroids), and topical medications (lubricants, antiinflammatory) were recorded. A comprehensive medical and ophthalmologic history was obtained from each patient. In all subjects, visual acuity measurement, anterior segment slit-lamp evaluation, Goldmann applanation tonometry, and dilated fundus ophthalmoscopy were assessed. Dry eye tests were performed in the following order to ensure that the ocular surface had recovered from the previous test: OSDI questionnaire, tear osmolarity, Schirmer test without topical anesthesia, TBUT, corneal staining with fluorescein using the Oxford scale, and conjunctival staining with lissamine green using the van Bijsterveld score. The interval between Schirmer test and TBUT was 5 minutes, and the interval between instillation of fluorescein and assessment of corneal and conjunctival staining was 2 minutes. Objective tests for DED evaluation were performed under standard constant conditions of room temperature, light, airflow, and humidity. Ó 2014 Lippincott Williams & Wilkins
Tear Osmolarity in Ocular GVHD
Tear osmolarity was evaluated with the recently developed noninvasive TearLab Osmolarity System (TearLab Corp, San Diego, CA), which is a “lab-on a-chip” system that collects and analyzes a 50-nL tear sample from the inferior lateral meniscus with a single-use test card.16 After instrument calibration, tears are collected directly from the eye, thus avoiding the use of a standard glass capillary tube. The embedded nanofluidic channels move the tear sample to the measuring electrodes by the core of the TearLab system. A desktop unit converts electrical signals generated from the laboratory card into a quantitative measurement and displays it to the user. Tear osmolarity is measured with a temperaturecorrected electrical impedance measurement (mOsm/L). An osmolarity level of $308 mOsm/L has been suggested as the most sensitive cutoff value to identify DED.17 Subjects were required to discontinue artificial teardrops at least 8 hours before examination. All measurements were performed between 9:00 and 11:00 AM, avoiding diurnal fluctuation of tear osmolarity.21 To perform the TBUT test, 1 drop of saline solution was placed on a fluorescein strip and the dye placed on the eye before receiving any other topical medication. For each test, the time to the first break-up was measured in seconds. The test was performed 3 times and then averaged to provide the final measurement. Severity grades for TBUT were the following: $10 seconds, normal; 5 to 9 seconds, mild to moderate; and ,5 seconds, severe.12,22 The Schirmer test was performed without anesthesia using a standardized filter strip for 5 minutes with the patient’s eyes closed. Schirmer test grading was as follows: .10 mm, normal; 6 to 10 mm, mild to moderate; and #5 mm, severe.12,23 Corneal staining was evaluated by slit-lamp examination after instillation of fluorescein and under a yellow filter. According to the Oxford scale score, severity of corneal staining was graded as: 1 as mild; 2 as moderate; and $3 as severe. Conjunctival staining was evaluated by slit-lamp examination after lissamine green instillation using the van Bijsterveld scale. Severity grades of conjunctival staining were: mild, if 1 to 3; moderate, if 4 to 6; and severe, if 7 to 9.12,24
Statistical Analysis Statistical analyses were performed using GraphPad Prism version 5.00 for Mac (GraphPad Software, San Diego, CA). The Spearman rank correlation test was used to assess correlation between variables. The Kruskal–Wallis 1-way test was used to compare mean values among different subgroups. Nonlinear correlation tests were assessed, and their correlation power was compared with standard linear correlation using the Akaike Informative Criteria (AICc) test, which selects the model that is most likely to have generated the data. In all analyses, P values ,0.05 were considered significant.
RESULTS We screened a total of 71 patients with ocular cGVHD. Fifty-six eyes of 56 patients (23 females, 33 males; mean age, 47.2 6 15.9; range, 21–76) fulfilled the inclusion criteria and www.corneajrnl.com |
1253
Cornea Volume 33, Number 12, December 2014
Berchicci et al
were finally enrolled. The eye showing worse osmolarity value was selected as the study eye, in accordance with the most recent literature.17 Fifteen patients were not enrolled due cataract surgery during the previous 6 months (11 eyes) or other immunologic disorders (4 patients treated for undifferentiated connective-tissue disease). Twenty-five patients (44.6%) had a history of acute myeloid leukemia, 10 patients (17.9%) had acute lymphoblastic leukemia, 3 patients (5.3%) had chronic lymphatic leukemia, 6 patients (10.7%) had Hodgkin lymphoma, 3 patients (5.3%) had non-Hodgkin lymphoma, 7 patients (12.5%) had myelodysplastic syndrome, and 2 patients (3.6%) had acute aplastic anemia. All patients developed systemic GVHD after undergoing allo-HSCT. The study visit was conducted 3.0 6 2.8 years from the transplantation time (range, 1–10 years; median, 1 year). Forty-five enrolled patients (80.3%) were using an immunosuppressive agent to control systemic GVHD and 24 of these (42.8%) were also receiving systemic corticosteroids. Eleven patients (19.6%) were not taking any systemic medication. Each of the 56 patients was treated from 4 to 10 times a day with topical lubricants. Other treatments were: topical corticosteroids (25 patients, 44.6%); topical 0.05% cyclosporine A (12 patients, 21.4%); therapeutic contact lens (11 patients, 19.6%); topical autologous serum (8 patients, 14.2%); punctum plugs (2 patients, 3.5%). Mean intraocular pressure was 13.4 6 1.9 mm Hg. GVHD-related dry eye test results are summarized in Table 1. Other signs of ocular GVHD were: meibomian gland dysfunction (17 patients, 30.4%), conjunctival hyperemia (30 patients, 53.6%), and cicatricial conjunctival changes (10 patients, 17.9%). According to ocular scoring of GVHD defined by NIH,4 subjects were categorized in mild (10 patients, 17.9%), moderate (24 patients, 42.8%) and severe (22 patients, 39.3%) ocular cGVHD. Tear osmolarity turned out to be significantly inversely correlated with TBUT (r2 = 0.681; correlation coefficient = 20.728; P , 0.001). A statistically significant inverse correlation was present with the Schirmer test (r2 = 0.203; correlation coefficient = 20.476; P , 0.001), and positive correlation with the OSDI score (r2 = 0.188; correlation coefficient = 0.455; P , 0.001), but both with low correlation strength (Fig. 1). Osmolarity was statistically different (P = 0.0006) among the Oxford scale subgroups: 305.4 6 17.7 mOsm/L in the mild group (36 eyes), 330.7 6 13.6 mOsm/L in the moderate group (8 eyes), and 329.1 6 23.9 mOsm/L in the severe group (12 eyes). Dunn multiple comparison post hoc analysis
TABLE 1. Results of Dry Eye Tests in Patients With Ocular cGVHD Dry Eye Test TBUT, s OSDI score Schirmer test value, mm Osmolarity, mOsm/L
1254
Mean 6 SD
Mean 95% Confidence Interval
Range
6.5 6 4.7 26.4 6 21.2 3.8 6 3.3
5.2–7.8 20.7–32.1 2.9–4.7
1–20 2–85 0–10
314.0 6 22.1
308.1–319.9
278–370
| www.corneajrnl.com
FIGURE 1. Correlation of tear osmolarity with various clinical parameters of dry eye tear osmolarity shows significant correlation (r2 = 0.681; P , 0.001) with TBUT (A). Moderate statistical correlation is present with the Schirmer test [(B) r2 = 0.203; P , 0.001] and with the OSDI score [(C) r2 = 0.188; P , 0.001].
pointed out that this result was driven by a significant difference between mild versus moderate groups, and mild versus severe group difference. Similarly, osmolarity turned out to be statistically different (P = 0.006) according to the van Bijsterveld score subgroups: 307.5 6 18.6 mOsm/L in the mild Ó 2014 Lippincott Williams & Wilkins
Cornea Volume 33, Number 12, December 2014
group (39 eyes), 327.8 6 26.5 mOsm/L in the moderate group (6 eyes), and 329.9 6 21.4 mOsm/L in the severe group (11 eyes). Dunn multiple comparison post hoc analysis pointed out that this result was driven by a significant difference just between mild versus severe groups. Low statistically significant correlation was found between the OSDI score and TBUT [r2 = 0.27; correlation coefficient = 20.563 (negative correlation)] and between the OSDI score and Schirmer test [r2 = 0.35; correlation coefficient = 20.646 (negative correlation)]. TBUT also showed weak correlation with the Schirmer test [r2 = 0.14; correlation coefficient = 0.44 (positive correlation)]. Other nonlinear regression models (second-order and third-order polynomial, exponential equations) were assessed to exclude possibility of a nonlinear relationship between each coupled variable. These regression models did not show any superiority compared with the standard linear Spearman rank correlation model.
DISCUSSION
Tear film hyperosmolarity is considered a pathophysiological factor both in aqueous-deficient dry eye and in evaporative dry eye.17 The purpose of this study was to evaluate tear osmolarity in patients with ocular cGVHD, and to analyze correlation between osmolarity and the clinically available diagnostic tests for dry eye (OSDI questionnaire, Schirmer test, TBUT, and conjunctival and corneal staining). In our group of patients with ocular cGVHD, we found an average tear osmolarity value (314.0 6 22.1 mOsm/L) consistent with DED, according to previously reported threshold tear osmolarity values.17 Because most of our subjects were on systemic and/or topical immunosuppressive therapy, mean tear osmolarity may be underestimated compared with those people as of yet undiagnosed and untreated for the disease. Other works have also shown increased osmolarity in various ocular diseases, such as Sjögren syndrome,19,20 and pterygium25; in patients undergoing antiglaucoma topical treatment26; in association with air pollution27; in thyroid ophthalmopathy28; and ocular mucous membrane pemphigoid.29 In our population of patients with ocular cGVHD, every diagnostic test result for dry eye turned out to be abnormal, confirming the severe impact of DED in ocular cGVHD. Our findings are consistent with quantitative and qualitative dysfunction of tear production in ocular cGVHD. Furthermore, our results may also suggest that dry eye symptoms may persist despite systemic treatment. However, conjunctival hyperemia (53.6% of our subjects), meibomian gland dysfunction (30.4%), and cicatricial conjunctival changes (17.9%) may have influenced dry eye test results. In this study, the correlation analysis of tear osmolarity with dry eye diagnostic tests found statistically significant correlation between tear osmolarity and TBUT, suggesting the presence of severe qualitative dry eye dysfunction in patients with ocular cGVHD. Regarding evaluation of tear osmolarity in patients treated for glaucoma or ocular hypertension, Labbé et al26 found that tear osmolarity disclosed the highest correlation with TBUT, in agreement with our findings. However, no correlation was observed between osmolarity and Schirmer Ó 2014 Lippincott Williams & Wilkins
Tear Osmolarity in Ocular GVHD
test, or between corneal and conjunctival staining. Likewise, Chen et al30 found similar osmolarity correlations in women using oral contraception and contact lenses. We also found a moderate significant negative correlation between tear osmolarity and Schirmer test scores. This is consistent with the results of Bunya et al19 and Utine et al,20 in patients with Sjögren syndrome. This was an expected finding because low Schirmer test values are evocative of decreased tear film secretion, and aqueous-deficient dry eye is known to result in increased tear osmolarity. Our results suggest that in ocular cGVHD, all aspects of tear physiology are affected, unlike in meibomian gland dysfunction or Sjögren syndrome, where individual aspects of tear physiology are deficient.31,32 In our group of patients, the OSDI score was moderately affected. We found a moderate significant positive correlation between the OSDI score and tear osmolarity and, to a lesser extent, between corneal fluorescein staining and tear osmolarity. However, Amparo et al33 reported that changes in tear osmolarity do not significantly correlate with changes in patient symptoms or corneal fluorescein staining in DED. There are conflicting results in the literature regarding correlation between osmolarity and the OSDI score in autoimmune diseases, such as Sjögren syndrome. Some authors20 found a positive relationship, whereas others19 highlighted negative correlation in a group of patients reporting nonsevere symptoms, despite high tear osmolarity. This study has some limitations. First, the majority of our subjects (80.3%) were taking at least one systemic immunosuppressive agent, and almost half (42.8%) were also receiving systemic corticosteroids. Moreover, 44.6% of our patients have been treated with topical corticosteroids, and 21.4% with topical 0.05% cyclosporine A. Additionally, total-body irradiation, infections, ocular toxicity of chemotherapy, and other conditioning therapies may contribute to the sicca syndrome and influence tear osmolarity values. Regarding TearLab reliability, recent studies determined diagnostic performance and revealed a sensitivity of 81% and a specificity of 80% of the threshold value of 308 mOsm/L.34 Eperjesi et al35 reported that the coefficient of reproducibility was 39 mOsm/L and the coefficient of repeatability was 33 mOsm/L. Tear osmolarity has long been regarded as a gold standard for the diagnosis and the best single predictor of DED.14 However, our data could not confirm this presumption in ocular cGVHD. Another factor that may have affected the results of our study is that tear osmolarity was measured only 1 time per eye of each subject. Khanal and Millar36 have indeed shown that consecutive tear osmolarity readings in an individual varied up to 35 mOsm/L and 3 consecutive readings are required with the TearLab osmometer to obtain a reliable measure of tear osmolarity. In conclusion, our study shows an increased mean tear osmolarity in patients with ocular cGVHD and emphasizes the relationship between tear osmolarity with TBUT and, to a lesser extent, with the Schirmer test. These results highlight the central role of both evaporative and aqueous deficiency dry eye in ocular cGVHD. The use of specific types of topical lubricants, or agents able to improve TBUT and to eventually reduce tear osmolarity, is important to control ocular inflammation and to improve quality of life of patients with cGVHD. www.corneajrnl.com |
1255
Cornea Volume 33, Number 12, December 2014
Berchicci et al
Hence, communication between hematologist and ophthalmologist is of peculiar importance, particularly regarding the dose and regimen of systemic and local immunosuppressant. Further powered prospective studies are required to assess clinical utility of tear osmolarity in monitoring patients with ocular cGVHD and targeting therapeutic decisions. REFERENCES 1. Shikari H, Antin JH, Dana R. Ocular graft-versus-host disease: a review. Surv Ophthalmol. 2013;58:233–251. 2. Choi SW, Levine JE, Ferrara JL. Pathogenesis and management of graftversus-host disease. Immunol Allergy Clin North Am. 2010;30:75–101. 3. Ivanir Y, Shimoni A, Ezra-Nimni O, et al. Prevalence of dry eye syndrome after allogenic hematopoietic stem cell transplantation. Cornea. 2013;32:e97–e101. 4. Filipovich AH, Weisdorf D, Pavletic S, et al. National Institutes of Health consensus development project on criteria for clinical trials in chronic graft-versus-host disease: I. Diagnosis and staging working group report. Biol Blood Marrow Transplant. 2005;11:945–956. 5. Copelan EA. Hematopoietic stem-cell transplantation. N Engl J Med. 2006;354:1813–1826. 6. Ferrara JL, Levine JE, Reddy P, et al. Graft-versus-host disease. Lancet. 2009;373:1550–1561. 7. Mencucci R, Rossi Ferrini C, Bosi A, et al. Ophthalmological aspects in allogenic bone marrow transplantation: Sjögren-like syndrome in graft versus-host disease. Eur J Ophthalmol. 1997;7:13–18. 8. Hassan AS, Clouthier SG, Ferrara JL, et al. Lacrimal gland involvement in graft-versus-host disease: a murine model. Invest Ophthalmol Vis Sci. 2005;46:2692–2697. 9. Westeneng AC, Hettinga Y, Lokhorst H, et al. Ocular graft-versus-host disease after allogeneic stem cell transplantation. Cornea. 2010;29: 758–763. 10. Hirst LW, Jabs DA, Tutschka PJ, et al. The eye in bone marrow transplantation. I. Clinical study. Arch Ophthalmol. 1983;101:580–584. 11. Balaram M, Rashid S, Dana R. Chronic ocular surface disease after allogeneic bone marrow transplantation. Ocul Surf. 2005;3:203–211. 12. The definition and classification of dry eye disease: report of the definition and classification subcommittee of the international dry eye workshop (2007). Ocul Surf. 2007;5:75–92. 13. Suzuki M, Massingale ML, Ye F, et al. Tear osmolarity as a biomarker for dry eye disease severity. Invest Ophthalmol Vis Sci. 2010;51:4557– 4561. 14. Tomlinson A, Khanal S, Ramaesh K, et al. Tear film osmolarity: determination of a referent for dry eye diagnosis. Invest Ophthalmol Vis Sci. 2006;47:4309–4315. 15. Luo L, Li DQ, Doshi A, et al. Experimental dry eye stimulates production of inflammatory cytokines and MMP-9 and activates MAPK signaling pathways on the ocular surface. Invest Ophthalmol Vis Sci. 2004;45: 4293–4301. 16. Sullivan BD, Whitmer D, Nichols KK, et al. An objective approach to dry eye disease severity. Invest Ophthalmol Vis Sci. 2010;51:6125–6130.
1256
| www.corneajrnl.com
17. Lemp MA, Bron AJ, Baudouin C, et al. Tear osmolarity in the diagnosis and management of dry eye disease. Am J Ophthalmol. 2011;151: 792–798. 18. Versura P, Profazio V, Campos EC. Performance of tear osmolarity compared to previous diagnostic tests for dry eye diseases. Curr Eye Res. 2010;35:553–564. 19. Bunya VY, Langelier N, Chen S, et al. Tear osmolarity in Sjögren syndrome. Cornea. 2013;32:922–927. 20. Utine CA, Biçakçigil M, Yavuz S, et al. Tear osmolarity measurements in dry eye related to primary Sjögren syndrome. Curr Eye Res. 2011;36: 683–690. 21. Farris RL, Stuchell RN, Mandel ID. Tear osmolarity variation in the dry eye. Trans Am Ophthalmol Soc. 1986;84:250–268. 22. Khanal S, Tomlinson A, McFadyen A, et al. Dry eye diagnosis. Invest Ophthalmol Vis Sci. 2008;49:1407–1414. 23. Methodologies to diagnose and monitor dry eye disease: report of the Diagnostic Methodology Subcommittee of the International Dry Eye WorkShop (2007). Ocul Surf. 2007;5:108–152. 24. Bron AJ, Evans VE, Smith JA. Grading of corneal and conjunctival staining in the context of other dry eye tests. Cornea. 2003;22: 640–650. 25. Julio G, Lluch S, Pujol P, et al. Tear osmolarity and ocular changes in pterygium. Cornea. 2012;31:1417–1421. 26. Labbé A, Terry O, Brasnu E, et al. Tear film osmolarity in patients treated for glaucoma or ocular hypertension. Cornea. 2012;31:994–999. 27. Torricelli AM, Novaes P, Matsuda M, et al. Correlation between signs and symptoms of ocular surface dysfunction and tear osmolarity with ambient levels of air pollution in a large metropolitan area. Cornea. 2013;32:e11–e15. 28. Iskeleli G, Karakoc Y, Abdula A. Tear film osmolarity in patients with thyroid ophthalmopathy. Jpn J Ophthalmol. 2008;52:323–326. 29. Miserocchi E, Iuliano L, Berchicci L, et al. Tear film osmolarity in ocular mucous membrane pemphigoid. Cornea. 2014;33:668–672. 30. Chen SP, Massaro-Giordano G, Pistilli M, et al. Tear osmolarity and dry eye symptoms in women using oral contraception and contact lenses. Cornea. 2013;32:423–428. 31. Ogawa Y, Kuwana M. Dry eye as a major complication associated with chronic graft-versus-host disease after hematopoietic stem cell transplantation. Cornea. 2003;22(7 suppl):S19–S27. 32. Khanal S, Tomlinson A. Tear physiology in dry eye associated with chronic GVHD. Bone Marrow Transplant. 2012;47:115–119. 33. Amparo F, Jin Y, Hamrah P, et al. What is the value of incorporating tear osmolarity measurements in assessing patient response to therapy in dry eye disease? Am J Ophthalmol. 2014;157:69–77.e2. 34. Sullivan BD, Eldridge DC, Berg M, et al. Diagnostic performance of osmolarity combined with subset markers of dry eye disease in an unstratified patient population. Invest Ophthalmol Vis Sci. 2010;51: E-Abstract 3380. 35. Eperjesi F, Aujla M, Bartlett H. Reproducibility and repeatability of the OcuSenseTearLab osmometer. Graefes Arch Clin Exp Ophthalmol. 2012;250:1201–1205. 36. Khanal S, Millar TJ. Barriers to clinical uptake of tear osmolarity measurements. Br J Ophthalmol. 2012;96:341–344.
Ó 2014 Lippincott Williams & Wilkins
Tear Osmolarity in Ocular Graft-Versus-Host Disease Luigi Berchicci, MD, Lorenzo Iuliano, MD, Elisabetta Miserocchi, MD, Francesco Bandello, MD, and Giulio Modorati, MD
Purpose: The aim of this study was to evaluate tear osmolarity in patients with chronic graft-versus-host disease (cGVHD) with ocular involvement.
Methods: In this observational cross-sectional study of 56 patients with ocular cGVHD referred to the tertiary-care Ocular Immunology and Uveitis Service at the San Raffaele Scientific Institute, Milan, from May 2010 to November 2013, we evaluated the following clinical parameters: Ocular Surface Disease Index (OSDI) symptoms questionnaire, tear osmolarity, Schirmer test, tear film break-up time (TBUT), corneal and conjunctival staining. Results: All patients developed systemic GVHD after undergoing allogeneic hematologic stem cell transplantation. Mean osmolarity was 314.0 6 22.1 mOsm/L, mean OSDI score was 26.4 6 21.2, mean TBUT was 6.50 6 4.75 seconds, and mean Schirmer test value was 3.8 6 3.3 mm. Tear osmolarity significantly inversely correlated with TBUT (r2 = 0.681; P , 0.001). Statistically significant inverse correlation was present with the Schirmer test (r2 = 0.203; P , 0.001), and positive correlation with the OSDI score (r2 = 0.188; P , 0.001), but both with low correlation strength. Osmolarity was statistically different in the subgroups according to the Oxford corneal staining scale (P = 0.0006) and to the van Bijsterveld conjunctival staining score (P = 0.006). Conclusions: Tear osmolarity increased in patients with ocular cGVHD, significantly correlated with TBUT and, to a lesser extent, with the Schirmer test value and OSDI. These results emphasize the role of aqueous-deficient and evaporative dry eye disease in patients with cGVHD after undergoing allogeneic hematologic stem cell transplantation. Tear osmolarity may be considered a useful test in diagnostic assessment of dry eye disease associated with cGVHD. Key Words: tear osmolarity, graft-versus-host disease, dry eye disease (Cornea 2014;33:1252–1256)
O
cular chronic graft-versus-host disease (cGVHD) affects 40% to 60% of patients receiving allogeneic hematologic stem cell transplantation (allo-HSCT), and ocular complications
Received for publication June 11, 2014; revision received September 1, 2014; accepted September 3, 2014. Published online ahead of print October 24, 2014. From the Department of Ophthalmology, San Raffaele Scientific Institute, VitaSalute University, Milan, Italy. The authors have no funding or conflicts of interest to disclose Reprints: Luigi Berchicci, MD, Department of Ophthalmology, San Raffaele Scientific Institute, Vita-Salute University, Via Olgettina, 60, Milan 20132, Italy (e-mail: [email protected]). Copyright © 2014 by Lippincott Williams & Wilkins
1252
| www.corneajrnl.com
occur in 60% to 90% of transplant recipients.1–3 Allo-HSCT from human leukocyte antigen-matched donors is an established and potentially curative therapy for a variety of hematologic malignancies, autoimmune diseases, inherited disorders of metabolism, histiocytic disorders, and other malignant solid tumors.4,5 Allo-HSCT, however, has been restricted from broad use by GVHD, a major cause of morbidity and mortality mediated by donor-derived T-cell recognition of host antigens as foreign.6 Ocular cGVHD tends to present within 3 years after allo-HSCT, usually with clinical features of an autoimmune disease (eg, Sjögren syndrome).7–9 Although ocular involvement has distinctive characteristics, it is not considered a unique diagnostic criterion because its isolated occurrence does not confirm the diagnosis of GVHD.4 Ocular cGVHD includes a spectrum of clinical manifestations affecting all layers of the eye, including lids, lacrimal glands, conjunctiva, cornea, and even the posterior segment.1 Dry eye disease (DED), the most common ocular cGVHD manifestation, has been reported in 69% to 77% of patients with systemic cGVHD, and may result in significant morbidity, with a decrease in quality of life.3 The pathological process may lead to secondary epithelial changes such as superficial punctate keratopathy, filamentary keratitis, corneal erosion, and infected or noninfected stromal ulceration.10 Eyelid keratinization of the tarsal conjunctiva may progress to entropion, trichiasis, and atrophy of meibomian glands. Severe ocular surface disease may also worsen to corneal perforation.11 DED is known to be a multifactorial disease of tears and ocular surface, which is characterized by increased osmolarity of the tear film and inflammation of the ocular surface.12 Elevated tear osmolarity has been reported to be a global marker of dry eye, present in both subtypes of the disease (aqueous-deficient dry eye and evaporative dry eye).13,14 Recent research suggested that tear hyperosmolarity may lead to surface epithelium impairment, by activation of proinflammatory cytokines released into tears, and eventually causing cell death through apoptosis.13–17 These surface alterations enhance and support the inflammatory process in patients with dry eye conditions such as ocular cGVHD. Recently, new tear osmometers have become available, resulting in faster and easier measurement of tear osmolarity in various clinical settings.18 Increased tear osmolarity has already been described in several other ocular conditions, including autoimmune diseases.19,20 To the best of our knowledge, this is the first study to selectively evaluate tear osmolarity in patients with ocular cGVHD, as well as its relationship with ocular surface parameters, such as tear film Cornea Volume 33, Number 12, December 2014
Cornea Volume 33, Number 12, December 2014
break-up time (TBUT), Schirmer test value, ocular surface disease index (OSDI), and corneal and conjunctival staining.
METHODS We enrolled all patients affected by ocular cGVHD who consecutively presented to our tertiary-care Ocular Immunology and Uveitis Service at the San Raffaele Scientific Institute, Milan, from May 2010 to November 2013. The study was performed in accordance with the recommendations of the Declaration of Helsinki, and was approved by the institutional Ethics Committee of the San Raffaele University Hospital. Each patient provided written informed consent regarding the use of personal data. Inclusion criteria were age $18 years and, according to the National Institutes of Health (NIH) consensus criteria definition, a new ocular sicca syndrome documented by low Schirmer test values with a mean value of both eyes #5 mm at 5 minutes or a new onset of keratoconjunctivitis sicca by slit-lamp examination with mean values of 6 to 10 mm on the Schirmer test accompanied by distinctive manifestations in at least one other organ.4 All 56 enrolled patients had developed systemic cGVHD after undergoing allo-HSCT, classified by an internal physician of the Hematology Department of our hospital according to the consensus criteria defined by NIH.4 Exclusion criteria were inability to complete the proposed questionnaire, any ocular surgical procedure in the study eye during the previous 6 months, presence or history of any other ocular surface disorders other than ocular cGVHD (eg, corneal dystrophy, pterygium), history of contact lens wear during 7 days before tear osmolarity assessment, the presence of any other systemic disease that could affect the ocular surface, presence of any medications that could affect the ocular surface (orally administered analgesic drugs, antidepressants, neuroleptics, anti-Parkinson drugs, antihistamines, antihypertensive drugs, estrogen derivatives, topical glaucoma agents, topical nonsteroidal antiinflammatory drugs), except systemic and topical immunosuppressive and/or corticosteroid therapy for GVHD. Demographic characteristics, primary hematopoietic disease, systemic treatment received after allo-HSCT (immunosuppressants and/or corticosteroids), and topical medications (lubricants, antiinflammatory) were recorded. A comprehensive medical and ophthalmologic history was obtained from each patient. In all subjects, visual acuity measurement, anterior segment slit-lamp evaluation, Goldmann applanation tonometry, and dilated fundus ophthalmoscopy were assessed. Dry eye tests were performed in the following order to ensure that the ocular surface had recovered from the previous test: OSDI questionnaire, tear osmolarity, Schirmer test without topical anesthesia, TBUT, corneal staining with fluorescein using the Oxford scale, and conjunctival staining with lissamine green using the van Bijsterveld score. The interval between Schirmer test and TBUT was 5 minutes, and the interval between instillation of fluorescein and assessment of corneal and conjunctival staining was 2 minutes. Objective tests for DED evaluation were performed under standard constant conditions of room temperature, light, airflow, and humidity. Ó 2014 Lippincott Williams & Wilkins
Tear Osmolarity in Ocular GVHD
Tear osmolarity was evaluated with the recently developed noninvasive TearLab Osmolarity System (TearLab Corp, San Diego, CA), which is a “lab-on a-chip” system that collects and analyzes a 50-nL tear sample from the inferior lateral meniscus with a single-use test card.16 After instrument calibration, tears are collected directly from the eye, thus avoiding the use of a standard glass capillary tube. The embedded nanofluidic channels move the tear sample to the measuring electrodes by the core of the TearLab system. A desktop unit converts electrical signals generated from the laboratory card into a quantitative measurement and displays it to the user. Tear osmolarity is measured with a temperaturecorrected electrical impedance measurement (mOsm/L). An osmolarity level of $308 mOsm/L has been suggested as the most sensitive cutoff value to identify DED.17 Subjects were required to discontinue artificial teardrops at least 8 hours before examination. All measurements were performed between 9:00 and 11:00 AM, avoiding diurnal fluctuation of tear osmolarity.21 To perform the TBUT test, 1 drop of saline solution was placed on a fluorescein strip and the dye placed on the eye before receiving any other topical medication. For each test, the time to the first break-up was measured in seconds. The test was performed 3 times and then averaged to provide the final measurement. Severity grades for TBUT were the following: $10 seconds, normal; 5 to 9 seconds, mild to moderate; and ,5 seconds, severe.12,22 The Schirmer test was performed without anesthesia using a standardized filter strip for 5 minutes with the patient’s eyes closed. Schirmer test grading was as follows: .10 mm, normal; 6 to 10 mm, mild to moderate; and #5 mm, severe.12,23 Corneal staining was evaluated by slit-lamp examination after instillation of fluorescein and under a yellow filter. According to the Oxford scale score, severity of corneal staining was graded as: 1 as mild; 2 as moderate; and $3 as severe. Conjunctival staining was evaluated by slit-lamp examination after lissamine green instillation using the van Bijsterveld scale. Severity grades of conjunctival staining were: mild, if 1 to 3; moderate, if 4 to 6; and severe, if 7 to 9.12,24
Statistical Analysis Statistical analyses were performed using GraphPad Prism version 5.00 for Mac (GraphPad Software, San Diego, CA). The Spearman rank correlation test was used to assess correlation between variables. The Kruskal–Wallis 1-way test was used to compare mean values among different subgroups. Nonlinear correlation tests were assessed, and their correlation power was compared with standard linear correlation using the Akaike Informative Criteria (AICc) test, which selects the model that is most likely to have generated the data. In all analyses, P values ,0.05 were considered significant.
RESULTS We screened a total of 71 patients with ocular cGVHD. Fifty-six eyes of 56 patients (23 females, 33 males; mean age, 47.2 6 15.9; range, 21–76) fulfilled the inclusion criteria and www.corneajrnl.com |
1253
Cornea Volume 33, Number 12, December 2014
Berchicci et al
were finally enrolled. The eye showing worse osmolarity value was selected as the study eye, in accordance with the most recent literature.17 Fifteen patients were not enrolled due cataract surgery during the previous 6 months (11 eyes) or other immunologic disorders (4 patients treated for undifferentiated connective-tissue disease). Twenty-five patients (44.6%) had a history of acute myeloid leukemia, 10 patients (17.9%) had acute lymphoblastic leukemia, 3 patients (5.3%) had chronic lymphatic leukemia, 6 patients (10.7%) had Hodgkin lymphoma, 3 patients (5.3%) had non-Hodgkin lymphoma, 7 patients (12.5%) had myelodysplastic syndrome, and 2 patients (3.6%) had acute aplastic anemia. All patients developed systemic GVHD after undergoing allo-HSCT. The study visit was conducted 3.0 6 2.8 years from the transplantation time (range, 1–10 years; median, 1 year). Forty-five enrolled patients (80.3%) were using an immunosuppressive agent to control systemic GVHD and 24 of these (42.8%) were also receiving systemic corticosteroids. Eleven patients (19.6%) were not taking any systemic medication. Each of the 56 patients was treated from 4 to 10 times a day with topical lubricants. Other treatments were: topical corticosteroids (25 patients, 44.6%); topical 0.05% cyclosporine A (12 patients, 21.4%); therapeutic contact lens (11 patients, 19.6%); topical autologous serum (8 patients, 14.2%); punctum plugs (2 patients, 3.5%). Mean intraocular pressure was 13.4 6 1.9 mm Hg. GVHD-related dry eye test results are summarized in Table 1. Other signs of ocular GVHD were: meibomian gland dysfunction (17 patients, 30.4%), conjunctival hyperemia (30 patients, 53.6%), and cicatricial conjunctival changes (10 patients, 17.9%). According to ocular scoring of GVHD defined by NIH,4 subjects were categorized in mild (10 patients, 17.9%), moderate (24 patients, 42.8%) and severe (22 patients, 39.3%) ocular cGVHD. Tear osmolarity turned out to be significantly inversely correlated with TBUT (r2 = 0.681; correlation coefficient = 20.728; P , 0.001). A statistically significant inverse correlation was present with the Schirmer test (r2 = 0.203; correlation coefficient = 20.476; P , 0.001), and positive correlation with the OSDI score (r2 = 0.188; correlation coefficient = 0.455; P , 0.001), but both with low correlation strength (Fig. 1). Osmolarity was statistically different (P = 0.0006) among the Oxford scale subgroups: 305.4 6 17.7 mOsm/L in the mild group (36 eyes), 330.7 6 13.6 mOsm/L in the moderate group (8 eyes), and 329.1 6 23.9 mOsm/L in the severe group (12 eyes). Dunn multiple comparison post hoc analysis
TABLE 1. Results of Dry Eye Tests in Patients With Ocular cGVHD Dry Eye Test TBUT, s OSDI score Schirmer test value, mm Osmolarity, mOsm/L
1254
Mean 6 SD
Mean 95% Confidence Interval
Range
6.5 6 4.7 26.4 6 21.2 3.8 6 3.3
5.2–7.8 20.7–32.1 2.9–4.7
1–20 2–85 0–10
314.0 6 22.1
308.1–319.9
278–370
| www.corneajrnl.com
FIGURE 1. Correlation of tear osmolarity with various clinical parameters of dry eye tear osmolarity shows significant correlation (r2 = 0.681; P , 0.001) with TBUT (A). Moderate statistical correlation is present with the Schirmer test [(B) r2 = 0.203; P , 0.001] and with the OSDI score [(C) r2 = 0.188; P , 0.001].
pointed out that this result was driven by a significant difference between mild versus moderate groups, and mild versus severe group difference. Similarly, osmolarity turned out to be statistically different (P = 0.006) according to the van Bijsterveld score subgroups: 307.5 6 18.6 mOsm/L in the mild Ó 2014 Lippincott Williams & Wilkins
Cornea Volume 33, Number 12, December 2014
group (39 eyes), 327.8 6 26.5 mOsm/L in the moderate group (6 eyes), and 329.9 6 21.4 mOsm/L in the severe group (11 eyes). Dunn multiple comparison post hoc analysis pointed out that this result was driven by a significant difference just between mild versus severe groups. Low statistically significant correlation was found between the OSDI score and TBUT [r2 = 0.27; correlation coefficient = 20.563 (negative correlation)] and between the OSDI score and Schirmer test [r2 = 0.35; correlation coefficient = 20.646 (negative correlation)]. TBUT also showed weak correlation with the Schirmer test [r2 = 0.14; correlation coefficient = 0.44 (positive correlation)]. Other nonlinear regression models (second-order and third-order polynomial, exponential equations) were assessed to exclude possibility of a nonlinear relationship between each coupled variable. These regression models did not show any superiority compared with the standard linear Spearman rank correlation model.
DISCUSSION
Tear film hyperosmolarity is considered a pathophysiological factor both in aqueous-deficient dry eye and in evaporative dry eye.17 The purpose of this study was to evaluate tear osmolarity in patients with ocular cGVHD, and to analyze correlation between osmolarity and the clinically available diagnostic tests for dry eye (OSDI questionnaire, Schirmer test, TBUT, and conjunctival and corneal staining). In our group of patients with ocular cGVHD, we found an average tear osmolarity value (314.0 6 22.1 mOsm/L) consistent with DED, according to previously reported threshold tear osmolarity values.17 Because most of our subjects were on systemic and/or topical immunosuppressive therapy, mean tear osmolarity may be underestimated compared with those people as of yet undiagnosed and untreated for the disease. Other works have also shown increased osmolarity in various ocular diseases, such as Sjögren syndrome,19,20 and pterygium25; in patients undergoing antiglaucoma topical treatment26; in association with air pollution27; in thyroid ophthalmopathy28; and ocular mucous membrane pemphigoid.29 In our population of patients with ocular cGVHD, every diagnostic test result for dry eye turned out to be abnormal, confirming the severe impact of DED in ocular cGVHD. Our findings are consistent with quantitative and qualitative dysfunction of tear production in ocular cGVHD. Furthermore, our results may also suggest that dry eye symptoms may persist despite systemic treatment. However, conjunctival hyperemia (53.6% of our subjects), meibomian gland dysfunction (30.4%), and cicatricial conjunctival changes (17.9%) may have influenced dry eye test results. In this study, the correlation analysis of tear osmolarity with dry eye diagnostic tests found statistically significant correlation between tear osmolarity and TBUT, suggesting the presence of severe qualitative dry eye dysfunction in patients with ocular cGVHD. Regarding evaluation of tear osmolarity in patients treated for glaucoma or ocular hypertension, Labbé et al26 found that tear osmolarity disclosed the highest correlation with TBUT, in agreement with our findings. However, no correlation was observed between osmolarity and Schirmer Ó 2014 Lippincott Williams & Wilkins
Tear Osmolarity in Ocular GVHD
test, or between corneal and conjunctival staining. Likewise, Chen et al30 found similar osmolarity correlations in women using oral contraception and contact lenses. We also found a moderate significant negative correlation between tear osmolarity and Schirmer test scores. This is consistent with the results of Bunya et al19 and Utine et al,20 in patients with Sjögren syndrome. This was an expected finding because low Schirmer test values are evocative of decreased tear film secretion, and aqueous-deficient dry eye is known to result in increased tear osmolarity. Our results suggest that in ocular cGVHD, all aspects of tear physiology are affected, unlike in meibomian gland dysfunction or Sjögren syndrome, where individual aspects of tear physiology are deficient.31,32 In our group of patients, the OSDI score was moderately affected. We found a moderate significant positive correlation between the OSDI score and tear osmolarity and, to a lesser extent, between corneal fluorescein staining and tear osmolarity. However, Amparo et al33 reported that changes in tear osmolarity do not significantly correlate with changes in patient symptoms or corneal fluorescein staining in DED. There are conflicting results in the literature regarding correlation between osmolarity and the OSDI score in autoimmune diseases, such as Sjögren syndrome. Some authors20 found a positive relationship, whereas others19 highlighted negative correlation in a group of patients reporting nonsevere symptoms, despite high tear osmolarity. This study has some limitations. First, the majority of our subjects (80.3%) were taking at least one systemic immunosuppressive agent, and almost half (42.8%) were also receiving systemic corticosteroids. Moreover, 44.6% of our patients have been treated with topical corticosteroids, and 21.4% with topical 0.05% cyclosporine A. Additionally, total-body irradiation, infections, ocular toxicity of chemotherapy, and other conditioning therapies may contribute to the sicca syndrome and influence tear osmolarity values. Regarding TearLab reliability, recent studies determined diagnostic performance and revealed a sensitivity of 81% and a specificity of 80% of the threshold value of 308 mOsm/L.34 Eperjesi et al35 reported that the coefficient of reproducibility was 39 mOsm/L and the coefficient of repeatability was 33 mOsm/L. Tear osmolarity has long been regarded as a gold standard for the diagnosis and the best single predictor of DED.14 However, our data could not confirm this presumption in ocular cGVHD. Another factor that may have affected the results of our study is that tear osmolarity was measured only 1 time per eye of each subject. Khanal and Millar36 have indeed shown that consecutive tear osmolarity readings in an individual varied up to 35 mOsm/L and 3 consecutive readings are required with the TearLab osmometer to obtain a reliable measure of tear osmolarity. In conclusion, our study shows an increased mean tear osmolarity in patients with ocular cGVHD and emphasizes the relationship between tear osmolarity with TBUT and, to a lesser extent, with the Schirmer test. These results highlight the central role of both evaporative and aqueous deficiency dry eye in ocular cGVHD. The use of specific types of topical lubricants, or agents able to improve TBUT and to eventually reduce tear osmolarity, is important to control ocular inflammation and to improve quality of life of patients with cGVHD. www.corneajrnl.com |
1255
Cornea Volume 33, Number 12, December 2014
Berchicci et al
Hence, communication between hematologist and ophthalmologist is of peculiar importance, particularly regarding the dose and regimen of systemic and local immunosuppressant. Further powered prospective studies are required to assess clinical utility of tear osmolarity in monitoring patients with ocular cGVHD and targeting therapeutic decisions. REFERENCES 1. Shikari H, Antin JH, Dana R. Ocular graft-versus-host disease: a review. Surv Ophthalmol. 2013;58:233–251. 2. Choi SW, Levine JE, Ferrara JL. Pathogenesis and management of graftversus-host disease. Immunol Allergy Clin North Am. 2010;30:75–101. 3. Ivanir Y, Shimoni A, Ezra-Nimni O, et al. Prevalence of dry eye syndrome after allogenic hematopoietic stem cell transplantation. Cornea. 2013;32:e97–e101. 4. Filipovich AH, Weisdorf D, Pavletic S, et al. National Institutes of Health consensus development project on criteria for clinical trials in chronic graft-versus-host disease: I. Diagnosis and staging working group report. Biol Blood Marrow Transplant. 2005;11:945–956. 5. Copelan EA. Hematopoietic stem-cell transplantation. N Engl J Med. 2006;354:1813–1826. 6. Ferrara JL, Levine JE, Reddy P, et al. Graft-versus-host disease. Lancet. 2009;373:1550–1561. 7. Mencucci R, Rossi Ferrini C, Bosi A, et al. Ophthalmological aspects in allogenic bone marrow transplantation: Sjögren-like syndrome in graft versus-host disease. Eur J Ophthalmol. 1997;7:13–18. 8. Hassan AS, Clouthier SG, Ferrara JL, et al. Lacrimal gland involvement in graft-versus-host disease: a murine model. Invest Ophthalmol Vis Sci. 2005;46:2692–2697. 9. Westeneng AC, Hettinga Y, Lokhorst H, et al. Ocular graft-versus-host disease after allogeneic stem cell transplantation. Cornea. 2010;29: 758–763. 10. Hirst LW, Jabs DA, Tutschka PJ, et al. The eye in bone marrow transplantation. I. Clinical study. Arch Ophthalmol. 1983;101:580–584. 11. Balaram M, Rashid S, Dana R. Chronic ocular surface disease after allogeneic bone marrow transplantation. Ocul Surf. 2005;3:203–211. 12. The definition and classification of dry eye disease: report of the definition and classification subcommittee of the international dry eye workshop (2007). Ocul Surf. 2007;5:75–92. 13. Suzuki M, Massingale ML, Ye F, et al. Tear osmolarity as a biomarker for dry eye disease severity. Invest Ophthalmol Vis Sci. 2010;51:4557– 4561. 14. Tomlinson A, Khanal S, Ramaesh K, et al. Tear film osmolarity: determination of a referent for dry eye diagnosis. Invest Ophthalmol Vis Sci. 2006;47:4309–4315. 15. Luo L, Li DQ, Doshi A, et al. Experimental dry eye stimulates production of inflammatory cytokines and MMP-9 and activates MAPK signaling pathways on the ocular surface. Invest Ophthalmol Vis Sci. 2004;45: 4293–4301. 16. Sullivan BD, Whitmer D, Nichols KK, et al. An objective approach to dry eye disease severity. Invest Ophthalmol Vis Sci. 2010;51:6125–6130.
1256
| www.corneajrnl.com
17. Lemp MA, Bron AJ, Baudouin C, et al. Tear osmolarity in the diagnosis and management of dry eye disease. Am J Ophthalmol. 2011;151: 792–798. 18. Versura P, Profazio V, Campos EC. Performance of tear osmolarity compared to previous diagnostic tests for dry eye diseases. Curr Eye Res. 2010;35:553–564. 19. Bunya VY, Langelier N, Chen S, et al. Tear osmolarity in Sjögren syndrome. Cornea. 2013;32:922–927. 20. Utine CA, Biçakçigil M, Yavuz S, et al. Tear osmolarity measurements in dry eye related to primary Sjögren syndrome. Curr Eye Res. 2011;36: 683–690. 21. Farris RL, Stuchell RN, Mandel ID. Tear osmolarity variation in the dry eye. Trans Am Ophthalmol Soc. 1986;84:250–268. 22. Khanal S, Tomlinson A, McFadyen A, et al. Dry eye diagnosis. Invest Ophthalmol Vis Sci. 2008;49:1407–1414. 23. Methodologies to diagnose and monitor dry eye disease: report of the Diagnostic Methodology Subcommittee of the International Dry Eye WorkShop (2007). Ocul Surf. 2007;5:108–152. 24. Bron AJ, Evans VE, Smith JA. Grading of corneal and conjunctival staining in the context of other dry eye tests. Cornea. 2003;22: 640–650. 25. Julio G, Lluch S, Pujol P, et al. Tear osmolarity and ocular changes in pterygium. Cornea. 2012;31:1417–1421. 26. Labbé A, Terry O, Brasnu E, et al. Tear film osmolarity in patients treated for glaucoma or ocular hypertension. Cornea. 2012;31:994–999. 27. Torricelli AM, Novaes P, Matsuda M, et al. Correlation between signs and symptoms of ocular surface dysfunction and tear osmolarity with ambient levels of air pollution in a large metropolitan area. Cornea. 2013;32:e11–e15. 28. Iskeleli G, Karakoc Y, Abdula A. Tear film osmolarity in patients with thyroid ophthalmopathy. Jpn J Ophthalmol. 2008;52:323–326. 29. Miserocchi E, Iuliano L, Berchicci L, et al. Tear film osmolarity in ocular mucous membrane pemphigoid. Cornea. 2014;33:668–672. 30. Chen SP, Massaro-Giordano G, Pistilli M, et al. Tear osmolarity and dry eye symptoms in women using oral contraception and contact lenses. Cornea. 2013;32:423–428. 31. Ogawa Y, Kuwana M. Dry eye as a major complication associated with chronic graft-versus-host disease after hematopoietic stem cell transplantation. Cornea. 2003;22(7 suppl):S19–S27. 32. Khanal S, Tomlinson A. Tear physiology in dry eye associated with chronic GVHD. Bone Marrow Transplant. 2012;47:115–119. 33. Amparo F, Jin Y, Hamrah P, et al. What is the value of incorporating tear osmolarity measurements in assessing patient response to therapy in dry eye disease? Am J Ophthalmol. 2014;157:69–77.e2. 34. Sullivan BD, Eldridge DC, Berg M, et al. Diagnostic performance of osmolarity combined with subset markers of dry eye disease in an unstratified patient population. Invest Ophthalmol Vis Sci. 2010;51: E-Abstract 3380. 35. Eperjesi F, Aujla M, Bartlett H. Reproducibility and repeatability of the OcuSenseTearLab osmometer. Graefes Arch Clin Exp Ophthalmol. 2012;250:1201–1205. 36. Khanal S, Millar TJ. Barriers to clinical uptake of tear osmolarity measurements. Br J Ophthalmol. 2012;96:341–344.
Ó 2014 Lippincott Williams & Wilkins
