Skip Navigation LinksHome > March 2014 - Volume 23 - Issue 3 > Incidence of Deepening of the Upper Eyelid Sulcus After Topi...
Journal of Glaucoma:
doi: 10.1097/IJG.0b013e31826a7e09
Original Studies

Incidence of Deepening of the Upper Eyelid Sulcus After Topical Use of Travoprost Ophthalmic Solution in Japanese

Maruyama, Katsuhiko MD, PhD*; Shirato, Shiroaki MD, PhD*,†; Tsuchisaka, Asako MD*

Free Access
Article Outline
Collapse Box

Author Information

*Department of Ophthalmology, Tokyo Medical University

Yotsuya Shirato Eye Clinic, Tokyo, Japan

Disclosure: The authors declare no conflict of interest.

Reprints: Katsuhiko Maruyama, MD, PhD, Department of Ophthalmology, Tokyo Medical University, 6-7-1, Nishi-shinjuku, Shinjuku, Tokyo 160-0023, Japan (e-mail:

Received January 11, 2012

Accepted June 12, 2012

Collapse Box


Purpose: To investigate the incidence of deepening of the upper eyelid sulcus (DUES) with topical use of travoprost in Japanese glaucoma patients.

Patients and Methods: This prospective study enrolled 32 primary open-angle glaucoma Japanese patients who had been treated topically with travoprost unilaterally for 6 months at baseline, and started treatment in both eyes. The patients were observed during outpatient visits at 2, 4, and 6 months. At every visit, the photographs of both eyes and forehead were displayed randomly and the presence of DUES was diagnosed when 3 examiners unanimously rated the case as positive. The patients were also asked if they noticed any subjective symptom of DUES. Sex, refraction, and intraocular pressure (IOP) were evaluated as potential risk factors.

Results: DUES was identified objectively in 34% (11/32) of the patients after 2 months of treatment, and in 53% (17/32) after 4 and 6 months of treatment. The incidence was significantly higher in older patients (P<0.05), but not related to sex, refraction, pretreatment and posttreatment IOP, and IOP reduction. Finally, 41% (7/17) of patients with objectively diagnosed DUES noticed the presence of DUES by themselves. No patient dropped out of the study because of DUES.

Conclusions: DUES is a common side effect of topical travoprost treatment in Japanese glaucoma patients.

Prostaglandin analogues have been approved as the first-line drugs for glaucoma because of their potent intraocular pressure (IOP)-lowering activity with few systemic adverse effects. However, local adverse events such as iris pigmentation, increasing eyelash growth, and conjunctival hyperemia are well recognized.1 Recently, deepening of the upper eyelid sulcus (DUES) associated with the use of bimatoprost,2–7 travoprost,6–9 or latanoprost10 has been reported as a local side effect. Although almost all the reports so far were case reports, 1 study examined the incidence of DUES in a prospective manner. Aihara et al11 investigated the incidence of eyelid change caused by bimatoprost, and examined the latency of onset and background factors related to onset. However, all subjects in their study were previously treated with latanoprost before switching to bimatoprost, and previous treatment with latanoprost might have affected the incidence or latency.

The purpose of this study was to investigate prospectively the incidence of DUES in patients receiving travoprost ophthalmic solution as initial treatment for primary open-angle glaucoma, and analyze the background factors related to this complication in Japanese patients.

Back to Top | Article Outline


Thirty-two consecutive patients with primary open-angle glaucoma who had been treated unilaterally with 0.004% travoprost (Travatan Z; Alcon Laboratories Inc., Fort Worth, TX) once daily as initial treatment for at least 6 months at Yotsuya Shirato Eye Clinic, and started to receive bilateral treatment between November 2009 and May 2010 were enrolled in this study. The eyes newly treated with travoprost were the subject of this study. All patients were Japanese with no history of intraocular surgeries including laser treatment. This observer-masked, open-label prospective study was conducted in accordance with the Declaration of Helsinki and the study protocol was approved by the ethical committee of Tokyo Medical University. Informed consent to participate in this study was obtained from all the patients.

Before the beginning of bilateral treatment with travoprost (baseline) and 2, 4, and 6 months after starting treatment, IOP and ocular surface conditions were examined routinely by slit lamp. No additional medication was prescribed if adverse events did not occur. Facial photographs, including eyebrows and lower eyelids, were taken under identical conditions by one of the authors (S.S.) using a 9.1 mega-pixel digital camera (EX-FC 100; Casio, Tokyo, Japan) without flash. The average of 3 serial IOP measurements before the trial was regarded as pretreatment IOP. Posttreatment IOP was obtained from IOP measurements at the 3 visits during treatment. IOP reduction was calculated by subtracting posttreatment IOP from pretreatment IOP. At each visit, patients were asked whether they noticed any local and systemic symptom.

To identify DUES, all authors independently evaluated a series of 4 pictures (at baseline, and after 2, 4, and 6 mo of treatment) of each patient displayed simultaneously on a 21-inch liquid crystal display (FlexScan L997; Eizo, Tokyo, Japan) and assessed the status of DUES. The 3 pictures taken after treatment were displayed in random order. When all observers agreed unanimously on obvious DUES in the eye being studied in at least 1 of the 3 photographs compared with that at baseline, and no change in the fellow eye that had started treatment before the study, the patient was diagnosed as having DUES. The time of onset of DUES was determined as the date when the sign were first detected by all observers. At each visit, patients were asked if they noticed any sign of DUES subjectively.

All data are presented as mean±SD and analyzed using JMP 8.0 software (SAS Institute Inc.). Age, refraction, pretreatment IOP, posttreatment IOP, and IOP reduction were compared between the groups with and without DUES by t test, and sex was compared using the Fisher exact test.

Back to Top | Article Outline


Demographic Data

All 32 patients completed 6 months of examination. The patient group comprised 10 men and 22 women. Demographic data of 32 patients are shown in Table 1. The eye evaluated in the present study was the right eye in 18 patients and the left eye in 14 patients. No patient discontinued using travoprost ophthalmic solution before the completion of the study at 6 months because of subjective DUES symptom, or other complication.

Table 1
Table 1
Image Tools
Back to Top | Article Outline
Incidence of DUES and Subjective Symptoms

DUES was detected objectively in 34% (11/32), 53% (17/32), and 53% (17/32) of the patients after 2, 4, and 6 months, respectively, of treatment. There was no new onset of DUES after 4 months. At 2, 4, and 6 months of treatment, 2 of 11 (18%), 4 of 17 (24%), and 7 of 17 (41%) patients, respectively, with objectively detected DUES noticed DUES subjectively. The rate of concordance between subjective symptoms and objective signs increased gradually in 2, 4, and 6 months (Fig. 1). Figure 2 is a typical case of DUES diagnosed 4 months after starting travoprost in the right eye (left eye had been treated with travoprost before this study).

Figure 1
Figure 1
Image Tools
Figure 2
Figure 2
Image Tools
Back to Top | Article Outline
Factors Contributing to DUES

Analysis of factors affecting the incidence of DUES is shown in Table 2. The incidence of DUES was significantly higher in older patients (P=0.046 by t test). The incidence of DUES was not related to sex, refraction, pretreatment and posttreatment IOP, and IOP reduction.

Table 2
Table 2
Image Tools
Back to Top | Article Outline


This is the first report of a prospective study investigating the incidence of DUES associated with topical use of travoprost evaluated by objective examination of serial photographs. In addition, the present study is the first to examine the incidence of DUES when a prostaglandin analog is used as the initial treatment for glaucoma, and not after switching from another drug. In 32 patients treated with travoprost in our study, DUES was detected in 34% of the patients after 2 months of treatment, increasing to 53% after 4 months, and remained unchanged at 53% after 6 months. Furthermore, the group with DUES was older compared with the group without DUES.

In case reports of DUES associated with prostaglandin analogs, the causative drugs were mostly bimatoprost2–6 and travoprost.6,8,9 In contrast, few reports of DUES associated with latanoprost have been reported, despite the fact that the drug was launched more than 10 years ago. A recent report from the United Kingdom described 3 cases of DUES after the use of latanoprost for 1 to 4 years.10

Although many case reports have been published, there is only 1 prospective study on the incidence of DUES in patients who switched from latanoprost to bimatoprost. Aihara et al11 studied 25 Japanese patients with open-angle glaucoma already treated with latanoprost for at least 6 months, and latanoprost was discontinued and switched to bimatoprost aiming to further lower IOP. The prevalence of DUES in these patients was evaluated by comparing photographs of the eyelid region taken before and after switching. The rates of DUES were 44% and 60% after 1 and 3 months, respectively, of treatment, and remained unchanged at 6 months.

In most of the case reports published so far, DUES was diagnosed based on self-reporting by the patient. In contrast, DUES was evaluated by comparing eyelid photographs before and after travoprost treatment in the present study. The rate of subjective awareness of DUES among objectively evaluated DUES-positive cases was 41% after 6 months of treatment. Aihara et al11 who conducted the same analysis reported a rate of 53%. Therefore, approximately one-half of the patients were not aware of morphologic change of the upper eyelid even though DUES had occurred. A possible reason is that the upper eyelid change is too mild to be noticed subjectively. Another reason may be that even when DUES occurs, the patient may perceive the change as swelling due to aging or other physiological changes and does not pay special attention. Therefore, assessment of DUES from the patient’s self-report alone may under-evaluate the incidence of DUES.

The duration of treatment with prostaglandin analogs until onset of DUES described in case reports ranged widely from 1 month to 4 years, with a relatively long treatment history reported in most cases.2–6,8–10 However, in the present study that identified DUES objectively by evaluating photographs, the rate of DUES was unchanged from 4 to 6 months of treatment. Aihara et al11 also reported no change at 3 and 6 months. Although it remains unknown whether the rate changes in long-term treatment, the findings so far suggest that upper eyelid sulcus changes become constant at 3 to 4 months of treatment.

Our study showed that the group that developed DUES was relatively older than the group that did not develop DUES. In a previous study, Aihara et al11 compared clinical factors between a group with DUES (15 cases) and a group without DUES (10 cases) after switching from latanoprost to bimatoprost, and reported a significantly higher age in the group with DUES. Although the reason remains unknown, it is possible that atrophy of adipose tissue in the upper eyelid due to aging may promote the development of DUES. From experience, the morphology of the upper eyelid has great influence on the development of DUES. In future, the effect of age on DUES development should be studied upon matching the type of eyelid morphology.

The precise mechanism of development of DUES remains to be elucidated, but the lipolytic effect of prostaglandin analogs is suspected to have an important contribution. In a diagnostic imaging study, Jayaprakasam and Ghazi-Nouri6 conducted magnetic resonance imaging on a 56-year-old female who developed DUES during bimatoprost treatment in 1 eye, and reported periorbital fat atrophy in the treated eye compared with the untreated eye. In a histologic study, Park et al12 examined the mean adipocyte density in preaponeurotic orbital fat biopsies obtained from 11 patients who developed DUES (bilateral use in 7 eyes, unilateral use in 4 eyes; mean age, 58.2 years; mean duration of drug use, 4.8 years) out of 18 patients treated with bimatoprost, latanoprost, or travoprost ophthalmic solution. In 4 patients using prostaglandin analog unilaterally (bimatoprost in 1, latanoprost in 2, travoprost in 1), the mean adipocyte density was 1758.21±158.15 cells/mm2 in the treated eyes, and 1258.73±127.54 cells/mm2 in the untreated eyes, showing an increase in density due to adipocyte atrophy in the treated eyes. Their results suggest that fat atrophy may be associated with the development of DUES caused by prostaglandin analogs. Furthermore, basic research also showed that latanoprost, travoprost, tafluprost, and bimatoprost all had high affinity to prostaglandin F2 α (FP) receptor, and inhibited differentiation of adipocyte precursors through stimulating FP receptors.13,14 These drugs have also been shown to bind prostaglandin E receptor 3 (EP3 receptor) and is involved in lipolysis mediated by EP3 receptor.15

There is insufficient evidence to conclude whether the causative drug should be actively discontinued in patients who developed DUES. If DUES is accompanied by functional abnormality, then the causative drug should be discontinued actively. However, Altieri and Ferrari16 reported that the levator muscle of upper eyelid in patients treated with latanoprost, travoprost, or bimatoprost ophthalmic solution was functionally similar to that of normal controls. Moreover, Yang et al9 also reported that in a patient treated unilaterally with travoprost, there was no difference in levator function between the treated eye and the fellow eye. Therefore, in the absence of adequate evidence indicating functional abnormality in DUES, consideration of discontinuing the causative drug in a patient who has developed DUES is based solely on the patient’s unacceptability of the esthetic changes. In the present study, none of the patients who developed DUES requested discontinuation of medication, and only 1 patient in the study of Aihara et al11 requested discontinuation. Therefore, DUES is not necessarily an adverse side effect. In addition, the reversibility of DUES is recognized. Only 1 case report of DUES caused by bimatoprost described no improvement of DUES even after discontinuation of bimatoprost,17 whereas many reports confirmed improvement upon discontinuation of the causative drug.2,3,5,6,7,9 Before initiation of prostaglandin analogs, physicians should explain to the patients the possibility of the onset of DUES, and that even if DUES occurs, the symptoms will improve by discontinuing the medication.

There are some limitations in this study. First, the study is uncontrolled and conducted on a small number of patients from a single center. Second, the follow-up period was only 6 months, and the long-term DUES rate is not known. In addition, only those patients in whom response to travoprost was confirmed by preliminary treatment in 1 eye were selected. Therefore, there might have been selection bias. Moreover, as the DUES signs assessed on photographs may also be caused by physiological changes, it is not absolutely certain that the changes of upper eyelid sulcus observed after travoprost treatment is definitely a side effect of the drug. Despite the above issues, this is the first prospective study that examined the incidence of DUES during the use of travoprost as the initial treatment. The results of this study are expected to have great impact on the choice of medication and consideration of side effects in glaucoma pharmacotherapy.

The results of the present study revealed that the prevalence of DUES caused by topical instillation of travoprost was 34% after 2 months of treatment, 53% after 4 months, and remained unchanged at 6 months. Furthermore, analysis showed that patients who developed DUES were relatively older than those who did not develop DUES. Cosmetic change may lead to diminished trust in doctors and lowered adherence to treatment, eventually affecting the efficacy of glaucoma therapy. Further studies that investigate the prevalence of DUES associated with other prostaglandin analogs, compare the incidence between drugs or different upper eyelid morphologies, and examine the incidence in long-term treatment are required to accumulate evidence for educating patients on DUES as a side effect of prostaglandin analogs.

Back to Top | Article Outline


1. Hollo G.The side effects of the prostaglandin analogues.Expert Opin Drug Saf.2007;6:45–52.

2. Peplinski LS, Albiani SK.Deepening of lid sulcus from topical bimatoprost therapy.Optom Vis Sci.2004;81:574–577.

3. Filippopoulos T, Paula JS, Torun N, et al..Periorbital changes associated with topical bimatoprost.Ophthal Plast Reconstr Surg.2008;24:302–307.

4. Tappeiner C, Perren B, Iliev ME, et al..Orbital fat atrophy in glaucoma patients treated with topical bimatoprost—can bimatoprost cause enophthalmos?Klin Monbl Augenheilkd.2008;225:443–445.

5. Yam JC, Yuen NS, Chan CW.Bilateral deepening of upper lid sulcus from topical bimatoprost therapy.J Ocul Pharmacol Ther.2009;25:471–472.

6. Jayaprakasam A, Ghazi-Nouri S.Periorbital fat atrophy—an unfamiliar side effect of prostaglandin analogues.Orbit.2010;29:357–359.

7. Nakakura S, Tabuchi H, Kiuchi Y.Latanoprost therapy after sunken eyes caused by travoprost or bimatoprost.Optom Vis Sci.2011;88:1140–1144.

8. Suh MH, Park KH, Kim DM.Effect of travoprost on intraocular pressure during 12 months of treatment for normal-tension glaucoma.Jpn J Ophthalmol.2009;53:18–23.

9. Yang HK, Park KH, Kim TW, et al..Deepening of eyelid superior sulcus during topical travoprost treatment.Jpn J Ophthalmol.2009;53:176–179.

10. Ung T, Currie ZI.Periocular changes following long-term administration of latanoprost 0.005%.Ophthal Plast Reconstr Surg.2012;28:e24–e44.

11. Aihara M, Shirato S, Sakata R.Incidence of deepening of the upper eyelid sulcus after switching from latanoprost to bimatoprost.Jpn J Ophthalmol.2011;55:600–604.

12. Park J, Cho HK, Moon JI.Changes to upper eyelid orbital fat from use of topical bimatoprost, travoprost, and latanoprost.Jpn J Ophthalmol.2011;55:22–27.

13. Serrero G, Lepak NM.Prostaglandin F2 alpha receptor (FP receptor) agonists are potent adipose differentiation inhibitors for primary culture of adipocyte precursors in defined medium.Biochem Biophys Res Commun.1997;233:200–202.

14. Liu L, Clipstone NA.Prostaglandin F2 alpha inhibits adipocyte differentiation via a G alpha q-calcium-calcineurin-dependent signaling pathway.J Cell Biochem.2007;100:161–173.

15. Strong P, Coleman RA, Humphrey PP.Prostanoid-induced inhibition of lipolysis in rat isolared adipocytes: probable involvement of EP3 receptors.Prostaglandins.1992;43:559–566.

16. Altieri M, Ferrari E.Do prostaglandin analogs affect eyelid position and motility?J Ocul Pharmacol Ther.2011;27:511–517.

17. Aydin S, Işikligil I, Tekşen YA, et al..Recovery of orbital fat pad prolapsus and deepening of the lid sulcus from topical bimatoprost therapy: 2 case reports and review of the literature.Cutan Ocul Toxicol.2010;29:212–216.


travoprost; upper eyelid sulcus; side effect

© 2014 by Lippincott Williams & Wilkins.


Article Level Metrics

Search for Similar Articles
You may search for similar articles that contain these same keywords or you may modify the keyword list to augment your search.