I2 = 37% at 20 to 30 minutes (z = 0.53; P = .60) and I2 = 0% at 60 minutes or longer (z = 2.34; P = .02) (raw data are shown in the eTable in the Supplement).
Heterogeneity for 20 to 30 minutes (I2 = 82%; z = 1.18; P = .24) and 60 minutes or longer (I2 = 41%; z = 1.73; P = .08) (raw data are shown in the eTable in the Supplement).
No heterogeneity detected (raw data are shown in the eTable in the Supplement).
eTable. Raw data used for forest plots by outcomes assessed, ie, blood pressure (BP) or heart rate (HR)
eFigure 1. Funnel plots with pseudo 95% confidence limits for all studies reporting change in blood pressure (BP) for 2.5% phenylephrine at 20-30 mins (a) and at =60 mins (b), and 10% phenylephrine at 10 mins (c), at 20-30 mins (d), at 40-45 mins (e) and at =60 mins (f)
eFigure 2. Funnel plots with pseudo 95% confidence limits for all studies reporting change in heart rate (HR) for 2.5% phenylephrine at 20-30 mins (a) and 10% phenylephrine at 20-30 mins (b) and at =60 mins (c)
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Stavert B, McGuinness MB, Harper CA, Guymer RH, Finger RP. Cardiovascular Adverse Effects of Phenylephrine Eyedrops: A Systematic Review and Meta-analysis. JAMA Ophthalmol. 2015;133(6):647–652. doi:10.1001/jamaophthalmol.2015.0325
Topical phenylephrine hydrochloride is routinely administered with few safety precautions, but evidence regarding its systemic safety to date is controversial. As even short-term variations in 24-hour blood pressure (BP) and heart rate (HR) can adversely affect cardiovascular health, better evidence on phenylephrine’s effects on HR and BP is required.
To perform a meta-analysis of available evidence regarding cardiovascular adverse effects of topical phenylephrine.
PubMed, MEDLINE, and the Cochrane Database of Systematic Reviews and Clinical Trials were searched for relevant literature from January 1, 1970, to January 1, 2014, using a combination of the following search terms: topical, ocular, ophthalmic, phenylephrine, tropicamide, cardiovascular effect, side effect, blood pressure, heart rate, mydriatic, and eye drops. A total of 70 articles related to the topic were identified and all full texts were retrieved.
Randomized clinical trials reporting change in BP and HR for adults were included in this review. All studies reporting results for neonates or infants, not reporting standard deviations, or not specifying the time of measurement or the concentration of phenylephrine used were excluded.
Data Extraction and Synthesis
Data from randomized clinical trials that reported BP and/or HR as well as the time following administration of topical phenylephrine at which measurements were obtained by concentration of phenylephrine as a mean change and its standard deviation were extracted. Data were synthesized by concentration of phenylephrine and time of measurement following topical application using random-effects models with inverse variance weighting to account for heterogeneity across studies.
Main Outcomes and Measures
Difference in BP and HR after topical administration of phenylephrine.
Eight RCTs with a total of 916 participants were included. Data were available for phenylephrine, 2.5%, at 20 to 30 minutes and 60 minutes or longer after administration, and neither BP nor HR changed at either time. Following application of phenylephrine, 10%, BP increased at 5 and 10 minutes (mean difference for both, +15 mm Hg; 95% CI, 11.94-18.54; P < .001) but decreased at 20 to 30 minutes and 60 minutes or longer with no changes detected against baseline. A mean increase in HR by 4.48 beats/min (95% CI, 1.09-7.88; P = .01) was present at 20 to 30 minutes following application of phenylephrine, 10%, and HR decreased by 60 minutes or longer with no changes detected compared with baseline.
Conclusions and Relevance
Phenylephrine, 2.5%, leads to no clinically relevant change in BP or HR, and the changes in BP and HR seen with phenylephrine, 10%, are short lived. Thus, phenylephrine, 2.5%, is safe to use in clinical routine.
Most clinical ophthalmic examinations as well as lens, retinal, and disc imaging involve the administration of topical mydriatic eyedrops to allow for optimal visibility. To achieve good and rapid dilation, phenylephrine hydrochloride (an α-receptor agonist) and tropicamide (muscarinic receptor antagonist) are often combined. Phenylephrine can cause pronounced cardiovascular adverse effects, including increases in both systolic and diastolic blood pressure (BP) and change in heart rate (HR), when given systemically.1,2 In contrast, topical tropicamide is thought to be safe and not cause any cardiovascular adverse effects in adults.3
To date, there is no conclusive evidence regarding the cardiovascular adverse effects of phenylephrine, 2.5% or 10%, eyedrops alone or in combination with tropicamide. The largest randomized clinical trial (RCT) that compared phenylephrine, 10% (n = 100) against tropicamide, 1% (n = 50) did not find an increase in BP or HR up to 30 minutes after administration in either group.3 In contrast, a number of case series reported an increase in both BP and HR following topical administration of phenylephrine, 2.5% and 10%, eyedrops.2,4 As mounting evidence has indicated that not only both short- and long-term increases in BP but also short-term BP variation in, for example, 24-hour ambulatory BP measurement are associated with the development, progression, and severity of cardiovascular events, target organ damage, and mortality,5 it is important to assess the potential adverse effects of topically administered phenylephrine on BP and HR. For example, carotid intima media thickness as well as left ventricular mass index, both measures of target organ damage related to arterial hypertension, could be shown to be significantly elevated in persons with increased systolic 24-hour BP variation.6 Thus, it is important to know whether topical phenylephrine leads to any increase in systolic BP, even if it is short lived. In ophthalmic or optometric practice, neither BP nor HR is routinely monitored before administration of topical phenylephrine, which further increases the necessity for it to be safe.
Against this background, we conducted a systematic review and meta-analysis of the literature to synthesize available evidence on the extent of cardiovascular adverse effects of topical phenylephrine in an effort to determine its safety.
PubMed, MEDLINE, and the Cochrane Database of Systematic Reviews and Clinical Trials were searched from January 1, 1970, to January 1, 2014, using a combination of the following search terms: topical, ocular, ophthalmic, phenylephrine, tropicamide, cardiovascular effect, side effect, blood pressure, heart rate, mydriatic, and eye drops. The searches yielded 2498 articles of which the abstracts were screened. Full texts were obtained for abstracts related to the topic of interest, and their reference lists were cross-checked for additional publications. A total of 70 articles related to the topic were identified and all full texts were retrieved (Figure 1).
Only articles reporting results for adults were included in this review, and all studies reporting results for neonates or infants were excluded. Obtained studies were categorized by study type (retrospective or prospective case series or RCT) and by assessed outcomes (BP, HR, or other). Data from RCTs that reported BP and/or HR as well as the time following administration of topical phenylephrine at which measurements were obtained by concentration of phenylephrine were extracted. Articles that did not report standard deviations and those that did not specify the time of measurement or the concentration of phenylephrine used were excluded (Figure 1). Only data on systolic BP were used for this meta-analysis because systolic BP was consistently reported across studies, with a number of studies not reporting diastolic BP, and has been shown to be a better marker of cardiovascular risk than diastolic BP.7 Publication bias in studies was assessed using funnel plots, and selection, ascertainment, and attrition biases were assessed based on study design.
Meta-analysis was performed using Stata version 12.1 statistical software (StataCorp LP), entering HR and BP before topical application of phenylephrine and HR and BP at different times after application as continuous data. The main outcome was mean difference in BP and HR at 20 to 30 minutes and at 60 minutes after topical administration of phenylephrine. Random-effects models with inverse variance weighting were used to account for high degrees of heterogeneity across studies as calculated by I2, indicating the percentage of variance in a meta-analysis that is attributable to study heterogeneity.8
A total of 8 RCTs that provided data for both phenylephrine, 2.5%, and phenylephrine, 10%, at both 20 to 30 minutes and 60 minutes or longer were included in the meta-analysis (eTable in the Supplement). Three studies tested phenylephrine in combination with tropicamide. All studies are described in detail in the Table. For the 8 RCTs, a total of 916 participants were included; the mean age was 56 years (range, 20-87 years), the sex proportions were approximately equal (54% female), and 3% had known cardiovascular comorbidities such as hypertension. Almost all studies recruited participants at routine eye clinics, either patients prior to ocular surgery or volunteers who most often were patients’ escorts. All studies excluded persons with a known history of cerebrovascular or cardiovascular events. Using funnel plots, no evidence of any publication bias was detected (eFigure 1 and eFigure 2 in the Supplement). Selection bias and ascertainment bias have been addressed in most of the included studies by either randomization or masking (4 of 8 and 5 of 8, respectively), and attrition bias is very unlikely owing to the very short follow-up with all measurements taken within 1 to 3 hours. Thus, all studies were included in the analysis.
Studies were divided by concentration (2.5% vs 10%), and results were split by time of measurement following topical application (Figure 2 and Figure 3). Phenylephrine, 2.5%, did not cause an increase in BP at either 20 to 30 minutes or 60 minutes or longer after application; in fact, there was a slight decrease in BP at the second time (mean difference, −4.65 mm Hg; 95% CI, −8.54 to −0.76; P = .02) (Figure 2).
As the effect of phenylephrine, 10%, may vary over time following application, data were analyzed across all different times for which information was available (Figure 3). Topical ocular application of phenylephrine, 10%, led to an increase in BP at 5 and 10 minutes after application (mean difference, +15 mm Hg from a weighted baseline BP of 130 mm Hg; 95% CI, 127-133; P < .001), which decreased again to baseline BP at 20 to 30 minutes and remained at this level at 40 to 45 minutes and at 60 minutes and longer (changes between −4.3 and +4.8 mm Hg; all P > .08) (Figure 3). For both concentrations and times, no studies except that by Kumar et al11 reported any extreme outliers. In the study by Kumar et al, a viscous solution of phenylephrine, 10%, was used, which likely led to greater systemic absorption (Table). As this study provided only 10 cases for each of the concentrations and times (overall weighting <5%) and as viscous phenylephrine is not commonly used in clinical practice, the study’s results are not representative of BP and HR changes following phenylephrine application in clinical routine and are only retained as a worst-case scenario. No differences were found between studies that used phenylephrine alone and those that combined it with tropicamide, suggesting that concurrent application of tropicamide has no additional effect on BP.
Fewer data were available for changes in HR at either time. No change was seen with phenylephrine, 2.5%, at 20 to 30 minutes or at 60 minutes or longer after topical application (Figure 4). At 20 to 30 minutes, mean HR was elevated by 4.48 beats/min (from a weighted baseline of 80 beats/min; 95% CI, 73-87; P = .01) following application of phenylephrine, 10%, and it decreased at 60 minutes or longer and was no different from baseline (mean difference, +2.6 beats/min; P = .31) (Figure 4). No differences were found between studies that used phenylephrine alone and those that combined it with tropicamide, suggesting that concurrent application of tropicamide has no additional effect on HR.
Collating all high-quality data available on the cardiovascular adverse effects of topical ocular phenylephrine, we found no effect of phenylephrine, 2.5%, on BP or HR and only a very short-lived effect of phenylephrine, 10%, which was present at 5 and 10 minutes after application for BP and up to 20 to 30 minutes for HR. This confirms that phenylephrine, 2.5%, is a safe drug if applied topically. Available data for phenylephrine, 2.5%, are heterogeneous; however, any effect is likely to not be clinically meaningful. As phenylephrine, 2.5%, is very effective in dilating most irises, there is little need to revert to phenylephrine, 10%, in clinical routine.
Previous studies have reported conflicting results, with the only RCT specifically set up to investigate the effect of phenylephrine, 10%, on HR and BP reporting no effect.3 Based on our results, the hazardous effect of phenylephrine, 10%, reported in the literature may have been overstated. The mean change in BP is transient and reverted fully in a short interval of less than 20 minutes. This change also is within a normal physiological range such as changes experienced with circadian or postural BP changes.5,17 Almost all systolic BP changes observed were of a magnitude less than 5 mm Hg. A mean systolic BP variation of 12 to 17 mm Hg while awake during the day has been shown to be associated with increased carotid intima media thickness and left ventricular mass index compared with a mean BP variation of 9 mm Hg or less.6 In a sample of persons with hypertension with a baseline BP variability of 10 mm Hg per half hour, BP variability at baseline was shown to be associated with target organ damage assessed by presence of arrhythmia, left ventricular mass index, cardiac volume on chest radiography, and retinal hypertensive changes 7 years later, independent of the existing hypertension.18 In another study, systolic BP variability of 15 mm Hg or less was defined as normal, and an increase more than this was associated with an increased prevalence of ischemic heart disease, severe carotid artery stenosis, and carotid intima media thickness increases.19 Thus, a very short-term BP change of less than 5 mm Hg is not likely to be of any clinical relevance.
Similarly, the observed change in HR following topical application of phenylephrine is likely not clinically relevant. The HR increases by 15 beats/min in untrained, healthy individuals 1 minute after standing from a sitting position.20 The changes in HR following topical application of phenylephrine were all less than 10 beats/min for either concentration.
Studies that reported BP and HR changes following application of a combination of topical phenylephrine and tropicamide did not differ from those that applied topical phenylephrine only, which is in keeping with a previously reported lack of effect of topical tropicamide on BP or HR in adults.2
Strengths of this review include the systematic search of literature and meta-analysis. Limitations are the dearth of high-quality evidence, particularly regarding cardiovascular adverse effects of phenylephrine, 2.5%, eyedrops, and the broad range of study designs, small sample sizes, heterogeneous samples, and concurrent application of tropicamide in some studies. Almost no data are available on cardiovascular adverse effects of topical phenylephrine in persons with cardiovascular risk factors or a history of cardiovascular or cerebrovascular events. The considerable differences between studies are reflected in the high heterogeneity coefficient in our analyses, and we chose to address this by using a random-effects model rather than the more common fixed-effects model. Our overall conclusions are limited by the suboptimal quality of available evidence. Until better evidence becomes available, however, this review constitutes the best evidence available to date.
Phenylephrine, 2.5%, leads to no clinically meaningful change in BP or HR and can be considered safe to use in clinical routine. The changes in BP and HR seen with phenylephrine, 10%, are short lived and of uncertain clinical relevance.
Corresponding Author: Robert P. Finger, PhD, Centre for Eye Research Australia, University of Melbourne, Royal Victorian Eye and Ear Hospital, Level 1, 32 Gisborne St, East Melbourne, Victoria 3002, Australia (email@example.com).
Submitted for Publication: November 8, 2014; final revision received January 14, 2015; accepted January 23, 2015.
Published Online: March 19, 2015. doi:10.1001/jamaophthalmol.2015.0325.
Author Contributions: Ms McGuinness and Dr Finger had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Finger.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Stavert, McGuinness, Finger.
Critical revision of the manuscript for important intellectual content: Harper, Guymer, Finger.
Statistical analysis: Stavert, McGuinness, Finger.
Administrative, technical, or material support: Stavert.
Study supervision: Harper, Guymer, Finger.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Funding/Support: This work was supported by the Charles Viertel Charitable Foundation, the Lloyd and Kathleen Ansell Ophthalmology Foundation, and the Mankiewicz-Zelkin Fellowship of the University of Melbourne. Ms Stavert is supported by the Hector Maclean Scholarship from the Centre for Eye Research Australia. The Centre for Eye Research Australia receives operational infrastructure support from the Victorian government.
Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Create a personal account or sign in to: