Article Text

Download PDFPDF

Quality of diurnal intraocular pressure control in primary open-angle patients treated with latanoprost compared with surgically treated glaucoma patients: a prospective trial
  1. K Mansouri1,2,
  2. S Orguel1,
  3. A Mermoud2,
  4. I Haefliger1,
  5. J Flammer1,
  6. E Ravinet2,
  7. T Shaarawy3
  1. 1
    University Eye Clinic Basel, Basel, Switzerland
  2. 2
    Jules Gonin Eye Hospital, University of Lausanne, Lausanne, Switzerland
  3. 3
    Glaucoma Sector, Ophthalmology Clinic, University of Geneva, Geneva, Switzerland
  1. Dr T Shaarawy, Glaucoma Sector, Ophthalmology Clinic, University of Geneva, 22, rue Alcide Jentzer, 1211 Genève, Switzerland; tarek.shaarawy{at}


Purpose: To compare the intraocular pressure (IOP) diurnal fluctuations of glaucoma patients treated with latanoprost 0.005% once a day with patients with controlled IOP after deep sclerectomy or trabeculectomy.

Methods: The trial included 60 prospectively recruited subjects with primary open-angle glaucoma. The medical group consisted of 20 patients with controlled IOP (<18 mm Hg) under latanoprost 0.005% monotherapy and with no history of previous intraocular surgery or argon laser trabeculoplasty; the surgical groups included 20 patients after trabeculectomy, and 20 patients after deep sclerectomy with collagen implant (DSCI). The patients in the surgical groups had a controlled IOP without any ocular hypotensive medications. All patients underwent a diurnal tension curve (08:00–17:00/three-hour intervals), followed by a water-drinking test (WDT) with the last IOP measurement taken at 21:00 hours. The between-group differences were tested for significance by means of analysis of variance (ANOVA).

Results: Baseline IOP was significantly different between the trabeculectomy group (10.1 mm Hg (3.4 SD)), the DSCI group (13.9 mm Hg (2.8)) and the latanoprost group (15.5 mm Hg (2.0); p = 0.005). The average IOP during the diurnal tension curve (10.1, 13.7, and 15.7 mm Hg, respectively, for the trabeculectomy, DSCI, and latanoprost groups) differed significantly between groups (ANOVA; p<0.0001), but the variation was comparable in the three groups (ANOVA; p = 0.13). After the WDT, elevation of IOP was significantly greater among patients treated with latanoprost (p = 0.003).

Conclusion: Trabeculectomy patients had a statistically significant lower average IOP in the diurnal tension curve compared with the other two groups. No wider variation in diurnal IOP with latanoprost compared with the surgical procedures was found. The IOP increase during the WDT was most marked in patients under latanoprost therapy.

Statistics from

Intraocular pressure (IOP) has been recognised as a major risk factor for the development of glaucoma and is subject to cyclic fluctuations throughout the day. Diurnal variation in glaucoma was first reported in 1898.1 IOP diurnal fluctuations have been identified as a significant and independent risk factor in glaucomatous progression.25 It has been suggested that the progressive damage in some cases could be caused by peaks of IOP or diurnal IOP variability not detected by tonometry during office hours. There is evidence that patients controlled after filtering surgery have fewer IOP fluctuations during the diurnal tension curve and after a water-drinking provocative test (WDT) than medically controlled patients.6 The WDT, popular in the 1950s and 1960s, was later discarded as a diagnostic tool for open-angle glaucoma because of low sensitivity and specificity.7 8 In recent years, the WDT has been identified as a reliable and safe tool to predict maximum IOP values during a diurnal tension curve (DTC) and to assess the patency of surgical interventions.911

The prostaglandin analogue latanoprost 0.005% seems to lead to a fairly uniform circadian reduction in IOP without peaks, compared with other IOP-reducing medications, such as timolol and dorzolamide.12 13 To date, and to the best of our knowledge, however, there are no studies in the literature comparing IOP fluctuations in latanoprost medically controlled versus surgically controlled glaucoma patients.

The purpose of this study was to evaluate the IOP fluctuation rate during the DTC and the response to the WDT in patients controlled with latanoprost compared with patients controlled after trabeculectomy and deep sclerectomy with collagen implant (DSCI).


This multicentre trial was conducted at the Glaucoma Surgery Research Unit, Department of Ophthalmology, University of Basel and the Glaucoma Unit of the Jules Gonin Eye Clinic, University of Lausanne. Sixty eyes of 60 patients with primary open-angle glaucoma (POAG) were enrolled between the two centres.

Patient selection

All patients were enrolled consecutively from a population of glaucoma patients already under one of the three treatments. The medical group consisted of 20 patients with a controlled IOP under latanoprost 0.005%, administered as a single drop in the evening, as the sole ocular hypotensive medication and with no history of previous intraocular filtering surgery. The surgical group consisted of 40 patients previously submitted to filtering surgery and controlled postoperatively with no ocular hypotensive medication. This group was subdivided into 20 patients who had previously been submitted to trabeculectomy and 20 patients who had previously been submitted to DSCI. Controlled IOP was defined as IOP less than 18 mm Hg during the previous two visits in office hours. In the medical group, the patients had been on a stable ocular hypotensive medication regimen for at least three months. All trabeculectomies were performed using antimetabolites by two experienced surgeons. Deep sclerectomies were performed without the use of antimetabolites by two experienced surgeons. Operations had been performed at least one year before inclusion in the study.

All patients had typical glaucomatous optic nerve atrophy and glaucomatous visual field defects, as defined by previously published criteria.14 The exclusion criteria were (1) baseline untreated IOP (surgery group) or treated IOP equal to or higher than 18 mm Hg; (2) pseudoexfoliative and pigmentary glaucomas; (3) previous treatment with argon laser trabeculoplasty or refractive surgery; (4) corneal abnormalities preventing reliable IOP measurement; and (5) women who were pregnant or of child-bearing potential and not using adequate contraception. Informed written consent was obtained from the subjects after explanation of the nature and possible consequences of the study. The study adhered to the tenets of the Declaration of Helsinki.

Immediately after admission, all patients were submitted to a DTC, which consisted of four IOP measurements at three-hour intervals (08:00 to 17:00). Patients in the medical group were advised not to take any latanoprost during the study hours. After the 17:00 hour IOP measurement, the patients were submitted to the WDT. The patients were instructed to fast during the four-hour period preceding the test. The test was carried out in a standardised manner: the patient was required to drink one litre of tap water over 15 minutes. After that, IOP was measured a total of three times at 15-minute intervals. The last IOP measurement was taken at 21:00 hours, after which the patient was discharged. All IOP measurements were performed using the same Goldmann applanation tonometer (Haag–Streit, Bern, Switzerland) by the same experienced examiner. The mean of three measurements was used for each timepoint.

The following parameters were used for data analysis: the mean diurnal IOP was obtained by averaging all of the IOP readings of the diurnal tension curve (excluding the WDT). The difference between the highest and the lowest value was taken as the diurnal range. The difference in IOP between the peak of the three measurements after the WDT and the baseline (IOP immediately before WDT) was considered as the IOP fluctuation during the WDT.

Statistical analysis

The sample size calculation was based on the assumption that a difference in mean IOP of 2.5 mm Hg is clinically relevant. In order to reach a power of 1-β  =  0.90, given α  =  0.05 and a SD of 2 mm Hg, 15 patients were needed. We included 20 patients in each group to achieve a power of over 90%. A 3 × 4 analysis of variance (ANOVA) model was used to analyse the IOP variation. All calculations were performed using SPSS version 11.5 (SPSS Inc, Chicago, Illinois, USA). A p value of 0.05 or less was considered significant.


Sixty patients were included in the trial, 40 patients in the surgery group and 20 in the medical group. No relevant statistically significant differences were found between the treatment groups with respect to demographic characteristics (table 1). All patients completed all evaluation phases and no adverse effects related to the study methodology (particularly the WDT) were recorded. The mean defect (MD) on visual fields was 9.18 (7.71; trabeculectomy), 6.81 (5.09; DSCI), and 3.09 (4.01; latanoprost; p<0.05).

Table 1 Patient demographics

The mean IOP at enrolment into the study was 11.4 (SD 4.4) mm Hg (trabeculectomy), 13.3 (3.4) mm Hg (DSCI), and 14.8 (2.3) mm Hg (latanoprost). The difference was significant (ANOVA p = 0.006). IOP at 08:00 hours (baseline), 11:00, 14:00, and 17:00 was 10.4 (3.4), 10.1 (4.1), 9.4 (3.8), and 10.4 (3.8) mm Hg, respectively, for the trabeculectomy group. The values were 13.7 (2.7), 14.6 (3.1), 13.1 (3.2), and 13.8 (3.3) mm Hg for the DSCI group, and 15.5 (2.0), 15.9 (1.7), 15.4 (2.1), and 15.9 (2.0) mm Hg for the latanoprost group. The mean IOP during the diurnal tension curve (10.1, 13.7, and 15.7 mm Hg, respectively, for the trabeculectomy, DSCI, and latanoprost groups) differed significantly between the treatment groups (ANOVA p<0.0001), and variation in IOP throughout the day was significant (ANOVA 0.0002), but this variation was comparable in the three groups (ANOVA p = 0.13). Post-hoc comparison using the Tukey honest significant test disclosed a lower average IOP during DTC in the trabeculectomy group compared with the other two groups (p<0.001), but the difference between DSCI and latanoprost was not significant (p = 0.1; fig 1).

Figure 1 Mean intraocular pressure (IOP) levels for the medical compared with the surgical groups. DSCI, deep sclerectomy with collagen implant; TE, trabeculectomy; WDT, water-drinking provocative test.

Changes in intraocular pressure after the WDT are shown in table 2. The average value for IOP during the WDT was significantly different (ANOVA p<0.0001). Peak values were reached after 30 minutes and IOP varied significantly during the 45 minutes (ANOVA p<0.0001) and was statistically different between the three treatment groups (ANOVA p = 0.003). Post-hoc analysis using the Tukey honest significant test disclosed statistically comparable variability between the DSCI group and the trabeculectomy group (p = 0.31), but the difference between the trabeculectomy group and the latanoprost group reached statistical significance (p = 0.0002) and was of borderline significance between the DSCI group and the latanoprost group (p = 0.054), indicating a higher variability of IOP during the WDT among patients treated with latanoprost. IOP values at 21:00 hours reached pre-WDT levels (9.4 (3.9), 14.5 (2.9), 16.9 (2.9) mm Hg, respectively, for the trabeculectomy, DSCI, and latanoprost groups) and seemed to be unaffected by the WDT.

Table 2 Mean IOP (mm Hg) preceding the WDT and at each timepoint (15, 30, and 45 minutes) after the WDT for the medical compared with the surgical groups


The aim of the present study was to show if fluctuations in IOP differ significantly between patients controlled by trabeculectomy, DSCI, or latanoprost. Our results show that IOP varies significantly throughout the day, but this variation is not different between the groups. We found a significantly more marked increase in IOP with latanoprost during the WDT.

To the best of our knowledge, this issue has not been addressed before. Previous studies always compared mixed medical groups with surgery and were mostly conducted before the introduction of the prostaglandin analogues. Migdal and co-workers15 analysed the functional outcome after early surgery compared with laser and medicine but that study was conducted before the introduction of newer classes of ocular hypotensive medications. A study by Medeiros et al6 demonstrated that patients previously submitted to trabeculectomy showed less variability in IOP than patients under a mixed group of ocular hypotensive treatments. A recent study by Konstas et al16 confirmed this outcome and found the mean 24-hour pressure control to be more stable in a surgical (n  =  30) compared with a medically controlled group (n  =  30).

Despite their conclusions, these pertained only to a mixed group of medications and might not be the result achieved with every single component of the mixed group. Furthermore, the comparability between the study of Konstas et al16 and ours is rendered more difficult because of methodological differences. Our study spans over 13 daytime hours compared with the 24-hour pressure measurements of Konstas et al;16 the latter’s medical group was under maximum tolerated therapy (ie two to four medications) compared with latanoprost monotherapy in our group. Contrary to the study of Konstas et al,16 we excluded patients with pseudoexfoliative glaucoma as a result of demonstrated higher diurnal IOP fluctuations.17 18

Furthermore, the statistical powers of the two studies were different from one another. The study of Konstas et al16 had a power of 1-β  =  0.80 compared with 1-β  =  0.90 in this study, assuming an IOP of 1.5 mm Hg versus IOP of 2.5 mm Hg and a SD of 2.8 mm Hg versus SD 2 mm Hg between treatments as clinically relevant. Their patients were matched by IOP (1 mm Hg SD), whereas we decided not to match our subjects.

Prostaglandin analogues, with their reported IOP reduction and safety profile, have become an important element of medical antiglaucomatous therapy in industrialised countries. Here resides the specific aim of this study and its potential value. Latanoprost was chosen because it has shown less significant fluctuations compared with timolol and dorzolamide.12 In another study, Susanna and co-workers19 reported that latanoprost had significantly less IOP fluctuations in response to the WDT than unoprostone. It has also been reported that topical medications that enhance outflow, such as prostaglandins, may provide better IOP stabilisation under stressful conditions than those that decrease aqueous humour production.20 The WDT is thought to simulate one of these stressful situations. We employed a similar definition of clinically relevant fluctuations (2.5 mm Hg) in order to make our results comparable to previously published papers.

We investigated the diurnal IOP at five timepoints covering a period of 13 daytime hours and conducted the WDT to help establish the IOP stability of patients controlled with any of the three management options. Our raw, unadjusted data showed statistical significance in favour of trabeculectomy in terms of absolute IOP levels, and response to the WDT.

After the WDT, the IOP level varied significantly during the experiment. Post-hoc analysis showed that the response of patients treated with latanoprost was significantly more pronounced. We conducted the WDT at the time of the trough effect of latanoprost (17:00 hours).21 Therefore, it could be speculated that the response to WDT might be different between the three treatment groups depending on the timepoint of the day. The 24-hour fluctuation of latanoprost has been reported to be in the range of 2.0 to 4.4 mm Hg.12 22 Konstas et al13 found a diurnal IOP of 2.4 mm Hg in a series of 103 patients with pseudoexfoliative glaucoma under latanoprost monotherapy who were submitted to four daytime measurements (08:00 to 20:00 hours). Their data compared well with our results that show a diurnal fluctuation of 1.7 mm Hg in POAG patients. The higher WDT fluctuation of 5.7 mm Hg that was obtained in our series is situated close to the upper range of 24-hour measurements.23

Our results are in agreement with Paletta Guedes et al24 who specifically looked at the IOP response after WDT of 15 patients under a fixed timolol 0.5%–dorzolamide combination compared with 21 patients submitted to deep sclerectomy (without collagen implant) and 20 patients submitted to trabeculectomy. A fourth group included 20 non-glaucomatous individuals. They showed that IOP variations after the WDT were similar in non-glaucomatous individuals and in patients submitted to either deep sclerectomy or trabeculectomy.

One limitation of the present study was that, as a result of logistical constraints, we could not conduct a 24-hour observation and therefore night time IOP were not assessed. Earlier reports concluded that the WDT is a reliable tool to detect IOP peaks of 24-hour IOP fluctuations.6 8 2528 Like most provocative tests, the WDT is a raw method and its findings should be interpreted carefully, particularly so because at present there is no standardised way of conducting this test. Depending on the study centre, the amount and duration of fluid intake can vary substantially. One more factor that is usually not taken into consideration and that might influence test outcomes is the weight of the patient, for which no adjustment factor has been proposed in the literature. This is of particular importance because the WDT is thought to predict the IOP peak of the diurnal tension curve by causing osmotic variations. Recent research suggests that there is no relationship between body mass index and WDT results.29 More research is needed to elucidate these and other confounding factors for the WDT.

Our results show that the visual field MD were significantly lower in the medical group, whereas the difference between the surgical groups was not significant. These groups had a MD of more than –6.0 dB in the majority of patients, which falls into the category of “moderate to advanced” glaucoma as defined by Hodapp et al.30 This result can be explained by the fact that generally patients are operated when in a more advanced stage of disease.

This study only looked at diurnal fluctuations on a single day. There is, however, no physiological reason to assume that fluctuations in the course of one day are more relevant than fluctuations between different days. As Niesel and Flammer31 and Flammer and co-workers32 have shown, there are numerous reasons to believe that the total fluctuations between different days are also important. To the best of our knowledge, a clinically important difference in diurnal or 24-hour fluctuation has not been well established. This value might be 1 mm Hg, but no consensus exists here. Furthermore, daily IOP fluctuations have not yet clearly been linked to visual field progression. The Early Manifest Glaucoma Trial has recently been analysed in that sense.33 It could be shown that with increasing IOP levels, IOP variability also increases. For this reason, we had included the WDT in order to provoke those higher levels of IOP and we were able to demonstrate a significant difference between latanoprost and surgical interventions.

In this study all deep sclerectomy procedures were performed without the use of antimetabolites, whereas the trabeculectomies were carried out with antimetabolites. This fact could explain the difference in baseline IOP levels between these two groups because the use of antimetabolites such as mitomycin C and 5-fluorouracil has been associated with lower IOP levels after non-penetrating glaucoma surgery.34


This study shows that trabeculectomy patients had a statistically significant lower average IOP in the diurnal tension curve compared with the other two groups. No wider variation in diurnal IOP with latanoprost compared with the surgical procedures was found. The increase in IOP after the WDT was, however, significantly different between the groups; with the lowest increase in trabeculectomy eyes, followed by DSCI and latanoprost. This increase might indicate potentially higher diurnal intraocular pressures in patients under latanoprost monotherapy, particularly in situations under physiological stress (as stimulated by the WDT).

Our results, however, pertain to only one member of the commercially available prostaglandin family. More studies are needed to compare other prostaglandin analogues as well as different classes of ocular hypotensive medications with surgical methods in terms of IOP fluctuations. Furthermore, the selection of only well-controlled patients precludes any generalisation of the present results to the entire glaucoma population. Also worthy of note is the fact that this study addresses one aspect of the glaucoma continuum, namely diurnal pressure control. In the framework of a comprehensive management of the glaucoma patient the risk/benefit ratio should also be taken into consideration as it significantly differs for each treatment regimen and should be individualised for each patient.


The authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. They would like to acknowledge Dr Andreas Schoetzau for assistance in statistical analysis.



  • Funding: This trial was supported by an unrestricted research grant from Pfizer AG, Switzerland.

  • Competing interests: None.

  • The article was presented in part at the International Glaucoma Symposium, Cape Town, South Africa, April 2005.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.