To study daytime or nighttime variability of mean arterial pressure and ocular perfusion pressure in untreated normal-tension glaucoma (NTG) patients and determine whether increased short-term mean arterial pressure and/or ocular perfusion pressure variability are associated with greater risk of visual field (VF) progression.
Longitudinal, retrospective, observational study.
This study enrolled 237 eyes of 237 untreated NTG patients who underwent 24-hour intraocular pressure and ambulatory blood pressure monitoring in the habitual position, and had ≥5 reliable VF tests during follow-up. Kaplan-Meier analyses were performed to compare outcomes with reference to the level of short-term mean arterial pressure and ocular perfusion pressure standard deviation for VF deterioration. Hazard ratios for the association between clinical factors, including short-term mean arterial pressure and ocular perfusion pressure standard deviation, and VF progression were obtained using Cox proportional hazards models.
Over-dipper NTG patients showed significantly larger daytime and nighttime mean arterial pressure and ocular perfusion pressure standard deviation than non-dippers or dippers. Both increased daytime and nighttime mean arterial pressure or ocular perfusion pressure standard deviation were associated with greater VF progression probabilities. Increased daytime mean arterial pressure or ocular perfusion pressure standard deviation was a significant predictor of subsequent VF progression ( P = .023 and P < .001, respectively).
Over-dipper NTG eyes showed significantly higher daytime or nighttime mean arterial pressure and ocular perfusion pressure variabilities than non-dipper and dipper NTG eyes. Increased daytime mean arterial pressure and ocular perfusion pressure standard deviation at baseline were significant predictors of future VF progression in NTG.
The ambulatory blood pressure monitoring device has enabled clinicians to noninvasively estimate 24-hour blood pressure and its variability. The 24-hour blood pressure data obtained using ambulatory blood pressure monitoring devices is highly reproducible. Blood pressure variability, as detected by an ambulatory blood pressure monitoring device, is a risk factor for end-organ damage, such as that of the heart, kidney, or brain. Although the mechanism involved remains to be clearly elucidated, direct vascular changes such as arterial stiffness or autonomic/vascular dysregulation that leads to abnormal responses to various internal and external conditions may result in end-organ vascular insufficiency. Both short-term (daytime or nighttime) and long-term (24-hour or visit-to-visit) blood pressure variability independently contribute to end-organ damage. For example, several studies have shown that increased daytime or nighttime blood pressure variability predicts the incidence and/or progression of coronary artery disease and cerebrovascular accidents, including stroke.
Short-term blood pressure and/or ocular perfusion pressure variability may be increased in some glaucoma patients, particularly in over-dippers, because they exhibit signs of vascular dysregulation in the form of an excessive nocturnal blood pressure dip over 24 hours. Kario and associates reported that over-dippers without glaucoma had higher daytime blood pressure variability (standard deviation) than non-dippers or dippers, with increased incidence of silent cerebrovascular damage. However, to the best of our knowledge, no information is currently available in the ophthalmic literature regarding short-term blood pressure and/or ocular perfusion pressure variability in over-dippers with glaucoma despite the knowledge that over-dippers are associated with increased risk of glaucoma progression.
Normal-tension glaucoma (NTG) is a progressive end-organ disease of the optic nerve head with a component of vascular abnormality. As a result, cerebrovascular accidents, silent myocardial infarct, and migraine are more frequently observed in patients with NTG than in controls. Our group previously showed that increased long-term (24-hour) variability of mean arterial pressure or mean ocular perfusion pressure was a consistent risk factor for NTG development and progression. We hypothesized that increased short-term blood pressure and/or ocular perfusion pressure variability might also play a significant role in disease progression in some NTG individuals, given that increased daytime or nighttime blood pressure variability is an independent predictor of disease progression for other end-organ damages. Thus, we aimed to (1) study the short-term (daytime or nighttime) blood pressure and ocular perfusion pressure variability in over-dippers with newly diagnosed NTG by comparing the variabilities among 3 groups—non-dippers, dippers, and over-dippers—in the habitual position; and (2) evaluate the relationships between daytime or nighttime blood pressure and ocular perfusion pressure variability and glaucoma progression in untreated NTG patients during follow-up.
This retrospective study was approved by the Institutional Review Board of Asan Medical Center, Seoul, South Korea. Informed consent was waived as the design of this study was retrospective. The design of this study followed the principles of the Declaration of Helsinki. The medical records of 550 consecutive patients with NTG who were seen by a glaucoma specialist (M.S.K.) between November 2009 and February 2014 at the glaucoma service of Asan Medical Center were retrospectively reviewed for potential eligibility.
At the initial glaucoma evaluation, each patient underwent a comprehensive ophthalmologic examination, including a review of the patient’s medical history, measurement of best-corrected visual acuity (BCVA), slit-lamp biomicroscopy, Goldmann applanation tonometry (GAT), gonioscopy, central corneal thickness (CCT) measurement, dilated funduscopic examination using a 90 diopter lens, stereoscopic optic disc photography, retinal nerve fiber layer (RNFL) photography, a standard automated perimetry (SAP) using a 24-2 Swedish Interactive Threshold Algorithm (SITA; Carl Zeiss Meditec, Dublin, California, USA) test, and spectral-domain optical coherence tomography (SD OCT, Cirrus HD-OCT; Carl Zeiss Meditec).
To diagnose NTG, patients had to have an optic disc of glaucomatous appearance, including a vertical cup-to-disc ratio >0.7; a difference in the vertical cup-to-disc ratio of >0.2 between eyes not explained by differences in disc size; diffuse or focal neural rim thinning; disc hemorrhage; RNFL defects indicative of glaucoma along with compatible glaucomatous visual field (VF) loss; a BCVA better than 20/30; maximum bilateral untreated intraocular pressure (IOP) less than 22 mm Hg in the outpatient clinic using GAT on multiple occasions; a normal anterior chamber; and an open angle on gonioscopic examination. CCT was measured 3 times by ultrasonic pachymetry (DGH-550; DGH Technology Inc, Exton, Pennsylvania, USA) at the initial visit and an average was calculated. Eyes were considered to have glaucomatous VF loss if the glaucoma hemifield test result was outside normal limits or the pattern standard deviation (PSD) had a P value <5%. All initial and follow-up VF tests had to be reliable, defined as a false-positive error <15%, false-negative error <15%, and fixation loss <20%. All newly diagnosed NTG patients were subsequently admitted for 24-hour IOP and blood pressure monitoring.
All patients had to meet the following additional inclusion criteria to be enrolled in the current retrospective study: newly diagnosed NTG without previous treatment; age ≥40 years; in-hospital 24-hour monitoring of IOP and blood pressure in the habitual position (sitting during daytime and supine during nighttime); regular follow-up without treatment at our clinic for more than 2 years with visits at 4- to 6-month intervals; and availability of at least 5 reliable 24-2 VF datasets during follow-up whose baseline mean deviation (MD) exceeded −20.00 dB without threat to fixation, as measured by the SAP SITA program. Antiglaucoma medical treatment was not initiated owing to relatively low IOP level during follow-up in all patients until confirmation of VF progression by the attending ophthalmologist (M.S.K.). The affected eye was selected in patients with unilateral disease. If both eyes of a patient had NTG and met the inclusion criteria, 1 eye was randomly selected and included in the analyses.
We excluded patients if at least 1 of the following conditions was present: fewer than 5 reliable 24-2 SITA tests; outpatient IOP >21 mm Hg; evidence of intracranial or otolaryngologic lesions; history of massive hemorrhage or hemodynamic crisis; previous or current use of antiglaucoma medications or systemic or topical steroids; presence of any other ophthalmic disease that could result in optic nerve head and VF defects; any ophthalmic disease other than glaucoma and mild cataracts; or a history of diabetic retinopathy. Individuals who smoked or had an irregular daily sleep schedule were also excluded, as were patients who had previous ocular laser procedures or surgeries or had corneal abnormalities that prevented reliable IOP measurements. Patients with progressive lens opacity and BCVA less than 20/30 and other ocular diseases that could affect VF during follow-up were also excluded from the study. Individuals on systemic antihypertensive or other hemodynamically active medications were not excluded.
Measurement of In-Hospital Intraocular Pressure Over 24 Hours
All IOP measurements were performed by a single, well-trained ophthalmology resident (D.W.J.). First, a separate study was performed (by D.W.J.) to test the accuracy of the TonoPen XL (Mentor Ophthalmics, Santa Barbara, California, USA) against GAT: we compared the TonoPen XL and GAT readings by performing a cross-sectional study of 52 consecutive patients (104 eyes) with glaucoma or suspected glaucoma. There was an excellent correlation between the IOP readings obtained by the TonoPen XL and GAT (r = 0.93, P < .001). The difference between the GAT and TonoPen XL readings was less than 2 mm Hg in 95% of the measurements.
All eligible patients in our main study were instructed to abstain from alcohol and caffeine for 3 days prior to hospital admission. Although the length and times of nocturnal periods at home may have differed among the patients enrolled, all measurements of IOP were obtained with the TonoPen XL at 8 AM, 10 AM, 12 noon, 2 PM, 4 PM, 6 PM, 8 PM, and 10 PM (diurnal IOP), and at 12 midnight, 3 AM, and 6 AM (nocturnal IOP) in both eyes of each patient. This schedule has been used in our previous studies and was chosen to provide the best trade-off between maximal number of IOP readings over 24 hours and minimal nonphysiological responses during in-hospital IOP measurement. Three measurements were taken for each eye, and the average value was used for analysis in both sitting and supine positions. Subjects were instructed to continue normal indoor activities during the daytime period, and daytime IOP was measured when patients were seated. During the nighttime period, lights in individual rooms were turned off by the nurse after 10 PM following the last sitting IOP measurement, and patients were instructed to sleep with their head at the same level as their body. Subjects were awakened (if necessary) and IOP measurements were taken with the TonoPen XL under dim light, with patients in the supine position (because activation of the sympathetic nervous system by changing body position at night could be nonphysiological). IOP obtained with the TonoPen XL was used in the data analysis without correction for CCT.
Calculation of In-Hospital Intraocular Pressure Variability
Calculations of peak, trough, and mean IOP were based on the 8 daytime measurements and 3 nighttime measurements. Short-term IOP variabilities were calculated by 2 methods: (1) fluctuation, and (2) standard deviation. Fluctuation was defined as the difference between the peak and trough IOP value recorded.
Measurement of In-Hospital Blood Pressure Over 24 Hours
The ambulatory 24-hour systolic and diastolic blood pressures, mean arterial pressure, and heart rate were monitored every 30 minutes with a fully automatic device (ambulatory blood pressure monitoring; Spacelabs Healthcare, Issaquah, Washington, USA). The automated device was used to minimize the variability between blood pressure examiners and to measure blood pressure at the most physiological environment while patients continued their routine 24-hour activities. During daytime hours (8 AM to 10 PM), patients stayed indoors and were encouraged to continue their normal indoor activities. During nighttime, the 8-hour period of darkness in the patient’s room was maintained while patients were instructed to sleep in a supine position. Patients were asked to refrain from any physical activities that could affect blood pressure. Meals were provided at 12:00 noon, 6:30 PM, and 7:30AM and did not include any alcohol or caffeine.
Calculation of Mean Ocular Perfusion Pressure
Mean arterial pressure was calculated for a specified time during 24 hours using the following equation:
Mean Arterial Pressure = Diastolic Blood Pressure + [1/3 × (Systolic Blood Pressure − Diastolic Blood Pressure)]
Mean ocular perfusion pressure was calculated at a specified time from the difference between the corresponding mean arterial pressure and IOP (substituting for venous pressure), following the formula of Bill :
Mean Ocular Perfusion Pressure (at sitting position during daytime) = 95/140 × Mean Arterial Pressure − IOP
Mean Ocular Perfusion Pressure (at supine position during nighttime) = 115/130 × Mean Arterial Pressure − IOP
Calculation of In-Hospital Systolic Blood Pressure, Diastolic Blood Pressure, Mean Arterial Pressure, and Mean Ocular Perfusion Pressure Variability
Calculations of peak, trough, and mean systolic blood pressure, diastolic blood pressure, and mean arterial pressure were performed for short-term periods (daytime or nighttime) based on ambulatory blood pressure monitoring readings every 30 minutes during 24 hours. For the calculation of the peak, trough, and mean ocular perfusion pressure, mean ocular perfusion pressure was calculated from the difference between mean arterial pressure and IOP at 8 AM, 10 AM, 12 noon, 2 PM, 4 PM, 6 PM, 8 PM, 10 PM, 12 midnight, 3 AM, and 6 AM. Calculations of short-term peak, trough, and mean ocular perfusion pressure were based on 8 daytime measurements and 3 nighttime measurements. Short-term systolic blood pressure, diastolic blood pressure, mean arterial pressure, and mean ocular perfusion pressure variabilities were evaluated by 2 methods: (1) fluctuation, and (2) standard deviation. Fluctuation was defined as the difference between the peak and trough blood pressure or the mean ocular perfusion pressure value recorded.
Definitions of Non-dippers, Dippers, and Over-Dippers
The nocturnal blood pressure reduction was calculated as [(diurnal averaged mean arterial pressure − nocturnal lowest mean arterial pressure)/diurnal averaged mean arterial pressure] × 100. Patients were then classified into 3 groups based on the degree of nocturnal blood pressure reduction, as follows: non-dippers, <10% nocturnal blood pressure reduction (or higher nocturnal than diurnal blood pressure) (n = 125); dippers, ≥10% but <20% reduction (n = 83); and over-dippers, ≥20% reduction (n = 29).
Definition of Visual Field Progression
All patients had undergone at least 5 or more reliable VF tests. Early Manifest Glaucoma Trial (EMGT) criteria were used to confirm VF progression using glaucoma change probability analysis (Carl Zeiss Meditec) during follow-up. VF progression was defined as a significant deterioration ( P < .05) from the baseline at 3 or more of the same test points evaluated on 3 consecutive examinations. These progression criteria have been used and validated in previous studies.
All statistical tests were performed using the SPSS for Windows statistical software package (version 18.0; SPSS Inc, Chicago, Illinois, USA). Descriptive statistics (number and percentage for categorical variables and mean ± standard deviation for continuous variables) were initially evaluated. χ 2 tests were subsequently used to detect the differences among 3 groups (non-dippers, dippers, and over-dippers) for categorical variables. The Student paired t test or Mann-Whitney U test was used for the analysis of continuous variables when appropriate (eg, daytime vs nighttime). One-way analysis of variance (ANOVA) was performed to detect differences among 3 groups, and Tukey tests were performed for post hoc comparisons. Bonferroni correction was performed for the multiple comparisons.
VF progression analysis included all follow-up measurements from baseline to time of progression or last follow-up visit. Kaplan-Meier life table analyses were constructed to compare the survival experience (time to confirmed VF progression) in the presence or absence of each potential risk factor for VF deterioration within 10 categories: daytime mean arterial pressure and mean ocular perfusion pressure variability (high [highest tertile range] vs stable [lowest tertile range]), daytime mean arterial pressure and mean ocular perfusion pressure minimum value (high [highest tertile range] vs low [lowest tertile range]), nighttime mean arterial pressure and mean ocular perfusion pressure variability (high [highest tertile range] vs stable [lowest tertile range]), nighttime mean arterial pressure and mean ocular perfusion pressure minimum value (high [highest tertile range] vs low [lowest tertile range]), mean follow-up IOP level (high [IOP >15 mm Hg] vs low [IOP ≤15 mm Hg]), and age based on the median value (younger [≤55 years] vs older [>55 years]).
Hazard ratios (HRs) for the association between potential predictive factors and glaucoma progression were obtained with Cox proportional hazards models. Univariate analyses were performed separately for each variable. Variables with P ≤ .20 in univariate analyses were included in the multivariate Cox proportional hazards models. A backward elimination process was used to develop the final multivariate model, and adjusted HRs with 95% confidence intervals (CIs) were calculated. Our current study found collinearity between mean arterial pressure and mean ocular perfusion pressure standard deviation data (r = 0.899). To avoid the effect of collinearity between the mean arterial pressure and mean ocular perfusion pressure standard deviations, 2 different multivariate models were separately built using each of these predictors while adjusting for the same confounders in each model. Schoenfeld residuals and the log [−log (survival rate)] test were used to verify that the proportional hazards assumptions were not violated. Model fit was assessed using residual analyses. P < .05 was considered statistically significant.
In total, 237 eyes of 237 patients with NTG who met the inclusion criteria were analyzed. Table 1 summarizes the overall descriptive statistics for the demographic and background variables of 3 groups. All patients were Korean: 116 men (48.95%) and 121 women (51.05%). The mean age was 55.83 ± 9.33 years (range, 40–86 years), and 60 of the patients (25.37%) had a history of hypertension, of whom 100% were taking oral medication to reduce systemic blood pressure. Thirty-one patients (13.08%) had an IOP ≥22 mm Hg at nighttime in the habitual position during the 24-hour IOP monitoring. Overall, the mean blood pressure dip was 10.36% ± 8.66%.