Hedner J, Everts B, Ström Möller C. Latanoprost and Respiratory Function in Asthmatic PatientsRandomized, Double-Masked, Placebo-Controlled Crossover Evaluation. Arch Ophthalmol. 1999;117(10):1305-1309. doi:10.1001/archopht.117.10.1305
To evaluate the effects of single and multiple administrations of the ocular hypotensive drug latanoprost on respiratory function, asthma symptoms, and use of asthma medication in patients with bronchial asthma.
Twenty-four stable patients with asthma (forced expiratory volume in 1 second: 70% to 90% of predicted and a minimum of 10% reversibility after inhalation of albuterol sulfate) with no previous exposure to inhaled corticosteroids participated in this randomized, double-masked crossover trial. Patients received latanoprost, 0.005%, or placebo, 1 drop per day, in each eye during two 6-day treatment periods separated by a 2-week washout period. Acute latanoprost or placebo provocation, methacholine chloride airway reactivity, and β2-stimulator reversibility tests were performed.
Main Outcome Measures
Morning and evening peak expiratory flow, spirometric performance throughout treatment periods and during different provocation tests, asthma symptoms, and use of asthma medications were evaluated.
There were no statistically significant differences between treatments in morning and evening peak expiratory flow, scored daytime and nocturnal asthma symptoms, or daily consumption of asthma medication. During placebo provocation, there was a small increase in forced expiratory volume in 1 second that was not seen during latanoprost provocation. This small difference (−0.09 L) was statistically significant but without clinical importance.
Resting and provoked airway function and asthma symptoms were unaffected by latanoprost treatment in patients with asthma with no previous exposure to corticosteroids. Latanoprost can be used in patients with glaucoma and concomitant bronchial asthma.
OVERALL prevalence of glaucoma in the general population is estimated to be 2%. Of people aged 75 years and older, approximately 4.7% are affected.1 Population surveys2 have demonstrated that as many as 40% of elderly people may have obstructive airway disease. Thus, there is a likely overlap of these conditions. Results of several studies3- 6 demonstrated that β-receptor antagonists used for glaucoma treatment may provoke or aggravate obstructive airway disease. Because of nasolacrimal duct drainage after ocular administration, drugs will undergo systemic absorption via the nasopharyngeal mucosa. Systemic absorption of timolol maleate, the most commonly used β-receptor antagonist for glaucoma therapy, has been demonstrated in some patients.7 Results of other studies8,9 show a high prevalence of unsuspected impairment of respiratory function and exercise tolerance induced by topical timolol use.
The synthetic prostaglandin (PG) F2I analogue latanoprost is widely used in the treatment of elevated intraocular pressure (IOP) in glaucoma and ocular hypertension. The IOP-reducing effect of latanoprost treatment is mainly caused by an increase in aqueous humor outflow through the uveoscleral route.10 Although some systemic absorption of latanoprost occurs after topical administration, the drug and its metabolites are rapidly eliminated with a systemic half-life of 17 minutes.11 The maximum plasma concentration of the acid of latanoprost is 53 pg/mL, obtained 5 to 15 minutes after a single topical administration of radiolabeled latanoprost (B. Sjöquist, PhD, unpublished data, 1994). Consequently, systemic effects are unlikely to occur after topical administration of latanoprost. This has been confirmed in 2 long-term studies12,13 with latanoprost, totaling approximately 500 patients with open-angle glaucoma or ocular hypertension.
The well-known bronchoconstrictive effects of PGD2 and PGF2I in the human lung seem to be mediated by thromboxane receptors.14- 16 However, latanoprost differs from PGF2 because latanoprost has been shown10,17 to be a relatively selective PGF2α receptor agonist with minimal effect on the thromboxane receptor. In a recent placebo-controlled trial,18 latanoprost administered in up to 7 times the clinical dose had no bronchoconstrictive effect in patients with asthma. However, this trial involved patients treated with corticosteroids, and provocation of a bronchoconstriction might have been partly suppressed. The present randomized, double-masked, placebo-controlled crossover study was performed to evaluate the effect of latanoprost on respiratory function in patients with asthma with no previous exposure to corticosteroids.
Twenty-four stable asthmatic patients attending the pulmonary unit of Stradins Hospital, Riga, Latvia, were included in the study according to the following criteria: forced expiratory volume in 1 second (FEV1) 70% to 90% of predicted, 10% reversibility of FEV1 after inhalation of albuterol sulfate, and at least 18 years of age.
Exclusion criteria were as follows: use of contact lenses; treatment with IOP-lowering agents; ocular inflammation; infection; or treatment with topical corticosteroids, nonsteroidal anti-inflammatory drugs, or systemic corticosteroids 3 months before the study. Patients were also excluded if they had a history of acute exacerbation of their asthma necessitating hospitalization during the 3 months preceding the first prestudy visit.
The protocol comprised a prestudy period of up to 4 weeks and two 6-day treatment periods. One drop of latanoprost (50 µg/mL) (Xalatan; Pharmacia & Upjohn, Stockholm, Sweden) or corresponding placebo was administered once daily in the late afternoon. All eyedrop instillations were performed by a registered nurse to ensure compliance with the study protocol. Treatment periods were separated by a washout period of 14 days; the second treatment period was followed by 2 to 4 weeks of follow-up. Each treatment period included 3 visits, on days 0, 5, 6, and 21, 26, and 27 (the study design is shown in Figure 1). Protocol also required an eligibility control during the prestudy period and a follow-up visit at the end of the trial. The prestudy visit for eligibility included a review of inclusion and exclusion criteria, review of concurrent asthma medication use, and ascertainment of the absence of concomitant corticosteroid treatment. Full medical histories were obtained, and physical and ocular examinations were performed. The follow-up visit included a physical examination, careful assessment of adverse events, and an additional evaluation of concurrent asthma medication use. After obtaining approval from the Swedish National Drug Agency and Human Ethics Review Boards of the Medical Faculty, University of Gothenburg, Sweden, and the Research Ethics Review Board, University of Riga, a signed informed consent form was obtained from all patients before study enrollment. The study was performed according to the Declaration of Helsinki.
Forced ventilatory capacity, FEV1, and peak expiratory flow (PEF) were recorded using a spirometer (Vicatest 3; Mijnhardt, the Netherlands) connected to a personal computer (Macintosh II; Apple Computer Inc, Cupertino, Calif) using a specialized software program (Maclab Spirometry; AD Instruments, Castle Hill, New South Wales, Australia). All spirometric testing was performed with the patient in a semirecumbent position. The best of 3 consecutive FEV1 assessments was taken for subsequent calculation.
Assessments of PEF were performed by the patient at home each morning and evening during the study periods. A standard PEF reader (Mini-Wright; Clement & Clarke, Levimed AB, Hoganas, Sweden) was used, and patients were instructed to select and record the best value obtained on 3 consecutive attempts.
The latanoprost provocation test was performed during the baseline visits—day 0 or day 21, depending on the treatment period. After spirometry, 1 drop of latanoprost (50 µg/mL) or corresponding placebo was instilled in each eye. Spirometry was repeated 30 minutes later, followed by administration of 1 drop of latanoprost (200 µg/mL) or corresponding placebo. Final spirometry was carried out 60 minutes after the start of the test (30 minutes after administration of the second dose).
This test was performed on days 5 and 26 of the first and second treatment periods, respectively. No more than a 2-hour deviation regarding the time of day at the particular visit was allowed. Spirometry was carried out before and 15 minutes after 2 consecutive inhalations of 400 µg of albuterol (Ventodiscus; Glaxo Wellcome, Greenford, UK).
The methacholine provocation test was applied to assess airway reactivity. Increasing concentrations of methacholine were inhaled (Pad Inhaler Boy; Paul Ritzau, Padwerk, Germany) until FEV1 was reduced by at least 20%. The threshold concentration, defined as the concentration of methacholine that produced at least a 20% reduction of FEV1, was determined. In addition, the concentration of methacholine producing exactly 20% FEV1 reduction was calculated by linear regression. Methacholine was dissolved in isotonic sodium chloride solution to a total volume of 4 mL to be inhaled. The methacholine concentrations applied for testing were 0.03, 0.06, 0.125, 0.25, 0.5, 1.0, 2.0, 4.0, 8.0, and 16.0 mg/mL. Inhalations at each concentration step were maintained for 2 minutes.
At least 2 weeks before study initiation (ie, before day 0), all asthma therapy was standardized. Permitted treatment included 200 µg of albuterol (Ventodiscus) and 600 mg of theophylline (Retophylline; Orion Pharma AB, Sollentuma, Sweden). Maintenance doses were individualized and kept constant throughout the study. All additional doses of albuterol used were recorded and evaluated as an outcome variable.
Patients were instructed how to use the PEF reader by a trained study nurse. Patients recorded in a separate diary their PEF on awakening and at bedtime before their morning and evening doses of asthma medication. Patients were also instructed to note how often they awoke with asthma symptoms, and to register the severity of asthma symptoms using a scale ranging from 0 to 2 (0 indicates no awakening with asthma symptoms; 1, awakening with asthma symptoms but no medication used; and 2, awakening with asthma symptoms and medication used). In addition, all patients carried out daily overall scoring of asthma symptoms on a scale ranging from 0 to 3 (0 indicates no symptoms; 1, mild symptoms not requiring medication use; 2, moderate symptoms necessitating medication use; and 3, severe symptoms interfering with normal daily activities). Adverse events were recorded throughout the study by passive reporting and active questioning.
The primary study objective was to assess the difference between latanoprost and placebo therapy in morning PEF, calculated as the mean of measurements across days 1 to 5. Secondary objectives were assessments of the corresponding difference in evening PEF, provocation tests, the β2-receptor responsiveness test, and the methacholine test; subjectively recorded data from the diary; and consumption of β2-receptor agonist medication (albuterol). All data, except for the last 3 assessments, were evaluated by analysis of variance with treatment sequence, patient within sequence, period, and treatment as factors. For the last 3 assessments, this statistical evaluation was performed using the Wilcoxon rank sum test. No adjustment of the significance level for multiple statistical tests was used.
Twenty-four patients were recruited. However, 3 patients withdrew consent: 1 before baseline, 1 during washout, and 1 on day 25. Patient characteristics and details of the asthma disease are described in Table 1.
Significant differences between treatment and period interaction effects were not found for any end point. The primary study end point, mean±SD morning PEF during the 5 study days, was almost identical in the latanoprost and placebo groups, 387.4±15.1 and 388.8±15.1 L/min (range, 238.0-518.0 and 266.0-520.0 L/min), respectively. The differences between means (−1.4 L/min; 95% confidence interval, −11.2 to 8.3 L/min) and range (−36 to 32 L/min) during the treatment periods were not significant (P=.76). No significant (slope analysis P=.12) trend was found in the differences between latanoprost and placebo therapy during the study periods (Table 2). Similarly, mean±SD evening PEF during the 5 study days was almost identical, 406.0±17.2 and 404.0±17.2 L/min, after latanoprost and placebo administration, respectively (difference, 1.9 L/min; 95% confidence interval, −9.2 to 13.0 L/min).
During the acute provocation test with latanoprost, 50 µg/mL, and then 200 µg/mL given in consecutive doses separated by 30 minutes, forced ventilatory capacity and spirometrically assessed PEF were not systematically changed compared with the baseline measurement. However, there was a marginal increase in FEV1 (Table 3). The average FEV1 increase was statistically significantly smaller after latanoprost treatment, 50 and 200 µg/mL, compared with the corresponding placebo treatment (mean differences, 0.09 [P=.005] and 0.10 [P=.01], respectively). However, the magnitude of this difference was not dose dependent and was judged clinically irrelevant.
Administration of albuterol (the β2-receptor responsiveness test) induced a similar increase in all spirometric test results (approximately 0.4 L/sec in FEV1 after both latanoprost and placebo therapy). Concentrations of methacholine producing exactly 20% FEV1 reduction in the methacholine test were almost identical after latanoprost therapy (0.23±0.12 mg/mL) compared with placebo (0.20±0.12 mg/mL); the difference was not significant (P=.58).
In general, no or only mild-to-moderate daytime asthma symptoms were reported. Neither incidence nor severity differed between the 2 treatments. Similarly, the frequency of nightly awakenings with a demand for extra albuterol was between 4 and 6 among 21 analyzable patients on the different study nights; there were no significant differences between treatments. The number of additional albuterol (200 µg) inhalations during the study period did not differ significantly, and ranged from 0 to 8 inhalations in both groups (median, 1 for each day).
Adverse events were few and were evenly distributed between treatments, including prestudy and washout periods, and included respiratory tract infection and headache. Conjunctival hyperemia was recorded in 4 patients during latanoprost treatment and in 2 during placebo treatment. One patient reported pain in the eye during latanoprost treatment, and 1 additional patient experienced worsening of asthma symptoms during latanoprost treatment. However, the worsening appeared after the methacholine provocation test and was interpreted to be associated with the provocation procedure, per se.
The present study aimed to examine the possible effect of latanoprost treatment on airway function in patients with asthma. Studies in the older population indicate that substantial overlap may exist between glaucoma and obstructive airway disease because both are highly prevalent in the elderly. Recently, new IOP-lowering agents have been introduced into glaucoma therapy. Synthetic PG latanoprost represents a new class of glaucoma medications that, in comparative trials13 with timolol, showed no apparent systemic adverse effects. Latanoprost reduces IOP by increasing uveoscleral outflow, and its effect is mediated through the PGF2α receptor subtype.10,17 Prostaglandin therapy has been shown19 to elicit bronchoconstriction, bronchial hyperresponsiveness, and airway microvascular leakage. Experiments14- 16 in isolated human bronchial smooth muscle demonstrated that contractile responses to the prostanoids thromboxane A2, PGF2, and PGE2 were predominantly mediated through the thromboxane receptor. Also, evidence20 suggests that augmentation of vascular permeability with plasma leakage in airway smooth muscle is caused by thromboxane receptor activation. Thromboxane A2 and the bronchoconstrictor PGs PGD2 and PGF2I are generated in greater amounts in asthmatic patients than in healthy subjects.21 Indeed, histamine in the allergenic reaction has been found to selectively stimulate the generation of PGF2I from human airway tissue, a response that amplifies the contractile responses of human airway smooth muscle to histamine.22
Thus, considerable experimental and theoretical evidence suggests that some PGs may have potential bronchoconstrictive effects in patients with asthma. Such effects may theoretically have escaped detection in a previous study,18 which included patients with asthma who received corticosteroid treatment, a potential inhibitor of endogenous PG formation. However, the present study did not identify deterioration of airway function in patients with asthma without corticosteroid treatment. The validity of this finding is supported by the different evaluation techniques used, including baseline spirometry and multiple PEF recordings, as well as assessment of airway hyperresponsiveness by methacholine provocation and bronchial smooth muscle responsiveness to β2-receptor stimulation (albuterol). Subjective symptom assessment and consumption of a short-acting escape medication (albuterol) did not differ between the 2 treatment periods, further supporting the finding of a lack of difference in the different pulmonary function tests. The only significant difference identified between latanoprost and placebo treatment in this study was a marginally attenuated increase in FEV1 during the acute latanoprost exposure. Provocation with placebo resulted in a small increase of FEV1, whereas latanoprost provocation did not change FEV1 from baseline. The possible clinical significance of this difference was rejected for several reasons. First, the magnitude of the FEV1 differences during provocation did not exceed 0.1 L/min and was thus, irrelevant from a clinical perspective.23 Second, there was no apparent dose dependency of the latanoprost effect. In other words, the numerical difference between treatments was similar after administration of 50 and 200 µg/mL.
It may be argued that the chance to identify and quantify a potential negative impact of latanoprost treatment is small in a limited provocation study, and that it does not permit us to exclude the possibility that use of latanoprost may induce potentially adverse effects on pulmonary function in particular subgroups of patients with asthma. This argument emphasizes the importance of continuous postmarketing surveillance. However, in this randomized, double-masked, placebo-controlled crossover study, no signs of bronchoconstriction or deterioration of asthma were observed after ocular administration of latanoprost to patients with asthma who have never taken corticosteroids. Results of this study support previous results from clinical trials in asthmatic and nonasthmatic patients. Therefore, latanoprost offers an attractive alternative for the treatment of glaucoma in patients with concomitant bronchial asthma.
Accepted for publication June 30, 1999.
Supported by grants from the Swedish Heart and Lung Foundation, Stockholm, and Pharmacia & Upjohn AB, Clinical Research, Uppsala, Sweden.
We thank Nils-Gunnar Åsenblad, MSc, for statistical analyses.
Reprints: Jan Hedner, MD, PhD, Department of Clinical Pharmacology, Sahlgrenska University Hospital, S-41345 Göteborg, Sweden (e-mail: email@example.com).