Effect of Latanoprost on Outflow Facility in the Mouse

Abstract

purpose. To assess the early effect of latanoprost on outflow facility and aqueous humor dynamics in the mouse.

methods. Aqueous humor dynamics in NIH Swiss White mice were assessed with an injection and aspiration system, using fine glass microneedles. A single 200-ng (4 μL) dose of latanoprost was applied to one eye 2 hours before measurement. The fellow eye served as a control. Intraocular pressure (IOP) was measured by using an established microneedle procedure. Outflow facility (C) was determined by constant-pressure perfusion measurements obtained at two different IOPs. Aqueous humor flow (Fa) was determined by a dilution method using rhodamine-dextran. Conventional and uveoscleral outflow (Fc and Fu) were calculated by the Goldmann equation.

results. Average IOP, Fa, and C of control eyes were 15.7 ± 1.0 mm Hg, 0.144 ± 0.04 μL/min (mean ± SD, n = 8), and 0.0053 ± 0.0014 μL/min per mm Hg (n = 21), respectively. Average IOP, Fa, and C of treated eyes were 14.0 ± 0.8 mm Hg, 0.138 ± 0.04 μL/min (n = 8 for each), and 0.0074 ± 0.0016 μL/min per mm Hg (n = 21), respectively. The differences between treated and control eyes were significant for IOP and total outflow facility only.

conclusions. These data indicate that the early hypotensive effect of latanoprost in the mouse eye is associated with a significant increase in total outflow facility. Alterations in the aqueous dynamics induced by latanoprost can be measured reproducibly in the mouse and may provide a useful model for further determining the mechanism by which latanoprost reduces IOP and alters outflow facility.

Topical application of latanoprost, a prostaglandin (PG) F_2α analogue, is a treatment used widely to lower intraocular pressure (IOP) in patients with glaucoma. Accumulating evidence indicates that PGs, including latanoprost, increase uveoscleral outflow in monkey ^{1 2 3} and human eyes, ⁴ although, the precise mechanism of action is not known. ⁵ As passage of aqueous humor through the uveoscleral outflow pathway is generally considered to be pressure insensitive when IOP is near normal, an increase in uveoscleral outflow would not be expected to alter facility. However, the effect of PGs on outflow facility is controversial. 

In two studies that demonstrated increased uveoscleral outflow in ocular hypertensive patients after repeated latanoprost treatment, outflow facility was significantly increased in one ⁶ and marginally increased in the other. ⁷ Bimatoprost also has recently been shown to increase outflow facility by 23% in normal subjects after once-daily application for 3 days. ^{8 9} In contrast to these results, in several studies investigating the effect of a single treatment of topical PGs on outflow facility, the investigators reported no significant changes. ^{10 11} Ambiguity in these findings may reflect insufficient sensitivity in some of the techniques used to detect small changes in outflow facility as well as differences between the studies with respect to the PGs used, the species examined, differences in dosage strategies, and the duration of treatment. The latter may be of particular importance, because the immediate IOP lowering that occurs after a single application of latanoprost may be mediated by cellular mechanisms different from those that lower IOP after repeat applications. ^{12 13} Further investigation to clarify the effect of PGs on outflow facility are clearly warranted. 

We recently demonstrated that a single dose (4 μL) of latanoprost induces a significant IOP reduction in the mouse eye by 2 hours after instillation of drops. ¹⁴ This suggests that the mouse may provide a particularly useful model for investigating the early effect of latanoprost on outflow facility. The present study was undertaken to assess outflow facility and other aqueous dynamics parameters in the mouse eye after a single dose of latanoprost. 

Methods

Animal Husbandry and Measurement Time

All experiments were performed in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. NIH Swiss White mice were obtained from Harlan Sprague-Dawley (San Diego, CA). Mice were bred and housed in clear cages covered loosely with air filters and containing white pine shavings for bedding. The environment was kept at 21°C with a 12-hour (6:00 AM to 6:00 PM) light–dark cycle. All mice were fed ad libitum. Animal age ranged from 8 to 12 weeks. All measurements were performed between 12:00 and 7:00 PM to minimize the influence of diurnal IOP rhythm, ^{15 16} and hence possible diurnal variation of mouse aqueous dynamics. 

Drug Application

Four μL of 0.005% latanoprost solution (200 ng total; Pfizer, New York, NY), phosphate buffered saline (PBS) solution (vehicle control), or 0.02% benzalkonium chloride in PBS (a second vehicle control), was applied to one eye of each mouse 2 hours before the measurements for assessment of aqueous dynamics. A 200-ng dose of latanoprost had been found to be the minimum needed to achieve a maximum IOP response in the Swiss White mouse. ¹⁴ Eight eyes of previously untreated mice were used for each measurement of IOP, episcleral venous pressure (EVP), and a further eight mice were used for measurement of aqueous humor flow (Fa). Forty-two mice were used to determine outflow facility. 

Anesthesia

The mice were anesthetized by intraperitoneal injection of a mixture of ketamine (100 mg/kg; Ketaset; Fort Dodge Animal Health, Fort Dodge, IA) and xylazine (9 mg/kg, TranquiVed; Vedco, Inc., St. Joseph, MO) prepared at room temperature. Animals were gently restrained in a tapered plastic film tube (Decapicone; Braintree Scientific Inc., Braintree, MA) to avoid stress, and anesthesia was administered with a 30-gauge needle. Each mouse was monitored carefully to assess the state of anesthesia. If the mouse did not respond to pinching of the back skin, it was placed on the platform to undergo IOP measurement and procedures for evaluating aqueous humor dynamics. 

Instrument System to Measure IOP, EVP, and C

Microneedles made of borosilicate glass (100-μm tip diameter and 1.0-mm outer diameter; World Precision Instruments [WPI], Sarasota, FL) were used for aqueous dynamics measurements, as described previously. ¹⁴ Each microneedle was mounted on a micromanipulator to facilitate positioning. A single microneedle connected to a pressure transducer (Model BLPR; WPI) was used to measure IOP. To measure EVP and outflow facility (C), the transducer was also connected to a fluid reservoir filled with physiological saline (BSS-Plus; Alcon, Fort Worth, TX) by silicon tubing with 0.38-mm inner diameter, as previously described. ¹⁴ The height of the reservoir could be varied to modulate intracameral pressure. The system pressure detected by the transducer was recorded by data acquisition computer software (Chart; ADInstruments, Colorado Springs, CO), as described previously. ¹⁴  

To measure EVP, the reservoir was raised while watching an aqueous collector channel through a dissecting microscope until blood cleared from the collector channel. Then the reservoir was gradually lowered while the investigator watched the junction of the collector channel and Schlemm’s canal. The intracameral pressure at which blood began to flow from the collector channel into Schlemm’s canal was recorded. ¹⁷ This procedure was repeated two more times and the mean of these measurements was designated EVP. 

To measure C, the distal end of the tubing was disconnected from the reservoir. The measurement of C was based on measuring total outflow volume (V_t; in microliters) during a specific time interval (in this study, 10 minutes) at two different levels of IOP (t), as described previously. ¹⁸ After insertion of the infusion needle into the anterior chamber, the height of the open end of the tubing was adjusted to alter IOP to 35 mm Hg. After verification of IOP stability, the initial fluid level in the silicon tube was marked. After a 10-minute interval, the new fluid level was marked. The total amount of infused fluid was designated the infused volume at 35 mm Hg (V_t=35). Next, IOP was decreased to 25 mm Hg and the 10-minute infused volume (V_t=25) was also measured in the same eye. The measurements of V_t=25 and V_t=35 were used to elucidate C according to the following logic. First, total outflow is equivalent to Fa, and Fa is equivalent to the sum of conventional outflow (Fc_t) and uveoscleral outflow (Fu) (all expressed in microliters per minute), which can be written as  

\[\mathrm{Fa}\ {=}\ \mathrm{Fc}_{\mathrm{t}}\ {+}\ \mathrm{Fu}.\]

Hence,  

\[\mathrm{Fa}\ {=}\ \mathrm{Fc}_{\mathrm{t}{=}25}\ {+}\ \mathrm{Fu}\ {=}\ \mathrm{V}_{\mathrm{t}{=}25}/10\]

, and  

\[\mathrm{Fa}\ {=}\ \mathrm{Fc}_{\mathrm{t}{=}35}\ {+}\ \mathrm{Fu}\ {=}\ \mathrm{V}_{\mathrm{t}{=}35}/10.\]

Subtracting equation 2 from equation 3  

\[\mathrm{Fc}_{\mathrm{t}{=}35}\ {-}\ \mathrm{Fc}_{\mathrm{t}\ {=}\ 25}\ {=}\ 0.1(\mathrm{V}_{\mathrm{t}{=}35}\ {-}\ \mathrm{V}_{\mathrm{t}{=}25}).\]

By the Goldmann equation,  

\[\mathrm{Fc}_{\mathrm{t}}\ {=}\ \mathrm{C(IOP}\ {-}\ \mathrm{EVP}).\]

Thus,  

\[\mathrm{Fc}_{\mathrm{t}{=}35}\ {-}\ \mathrm{Fc}_{\mathrm{t}{=}25}\ {=}\ \mathrm{C}(35\ {-}\ \mathrm{EVP})\ {-}\ \mathrm{C}(25\ {-}\ \mathrm{EVP})\ {=}\ \mathrm{C}\ {\times}\ 10\]

Combining equations 4 and 6 ,  

\[\mathrm{C}\ {=}\ 0.01(\mathrm{V}_{\mathrm{t}{=}35}\ {-}\ \mathrm{V}_{\mathrm{t}{=}25})\ {\mu}\mathrm{L/min\ per\ mm\ Hg}.\]

 

Aqueous Humor Flow

A perfusion method was used to measure aqueous humor flow (Fa) with minor modifications of a previously described technique. ¹⁸ For the infusion, a microneedle connected to a 25-μL syringe-equipped (Hamilton, Reno, NV) micropump (Micro4; WPI) was mounted on a micromanipulator. The micropump could perfuse specified volumes at programmed speeds. This infusion system was filled with 0.5 μg/mL rhodamine conjugated to 70-kDa dextran (RD; Molecular Probes, Eugene, OR) diluted in saline (BSS-plus; Alcon). A second needle mounted to another manipulator was connected to a water-filled narrow silicon tube that was connected to a pressure transducer positioned at the same height as the eye. Another segment of water-filled tubing connected the pressure transducer to a water-containing reservoir. As the second microneedle was to receive the fluid flowing out of the anterior chamber, the tip of this microneedle was filled with a small amount of mineral oil to prevent mixing of the perfused fluid with the water in the tubing. To conduct the measurement, the infusion microneedle was inserted into the anterior chamber and then the second microneedle was inserted. Next, the second microneedle was withdrawn, and 5 μL of RD solution was injected through the first microneedle to replace the aqueous humor with the RD solution. As this injection was made, the anterior chamber contents drained out of the hole that remained after removal of the second microneedle. After the aqueous humor had been replaced with RD solution, the second microneedle was reinserted into its original hole. The pressure in the anterior chamber was adjusted to EVP by repositioning the height of the water reservoir. This step reduced the pressure-dependent outflow to zero. These preparative procedures were completed within 5 minutes. This was unlikely to have altered the blood–aqueous barrier, as previous experiments found that insertion of a microneedle into the anterior chamber followed by raising and lowering intracameral pressure did not change aqueous humor protein concentration. ¹⁸  

After completion of the preparative procedures, the anterior chamber was perfused with RD solution at a rate (v) of 2 μL/min. The effluent of the anterior chamber was collected within the tubing linking the second microneedle to the pressure transducer. After 10 minutes of perfusion, the last 10 μL of the perfused fluid was retrieved from the tubing. Thus, this sample included the aspirated fluid that had been collected during the last 5 minutes of the perfusion. This strategy avoided collecting fluid during the first 5 minutes because of the possibility that the ratio of tracer to aqueous humor had not reached a steady state. The concentration of RD in the perfusate then was determined as described in the following sections. In pilot studies, separate analysis of RD concentration within 2 μL aliquots from each of the final 5 minutes of the perfusion confirmed that the dilution of the infused fluid by physiological aqueous outflow had reached a steady state by 5 minutes after initiation of perfusion. 

RD concentration within the samples was determined using a spectrofluorometer (model SFM25; Kontron, Zürich, Switzerland) as described previously. ¹⁸ The excitation and emission wavelengths were 555 and 580 nm, respectively. The concentration of RD in the injection fluid (C_i) was measured twice and averaged. The concentration of RD in the collected samples (C_o) was measured twice. The mean value of C_i/C_o in each eye (n = 8) was used to calculate Fa according to the following rationale: The total amount of RD in the injected fluid and the recovered fluid in a specific time are equal. Hence, the concentration of RD in the injected fluid (C_i) multiplied by flow rate of the injected fluid (v_i) and the concentration of RD in the recovered fluid (C_o) multiplied by flow rate of the recovered fluid (v_o) are equal. Thus,  

\[\mathrm{C}_{\mathrm{i}}\ {\times}\ \mathrm{v}_{\mathrm{i}}\ {=}\ \mathrm{C}_{\mathrm{o}}\ {\times}\ \mathrm{v}_{\mathrm{o}}.\]

 

During the perfusion, the injected fluid was diluted by the secreted aqueous humor and the magnitude of dilution was a measure of Fa.  

\[\mathrm{Vi}\ {+}\ \mathrm{Fa}\ {=}\ \mathrm{v}_{\mathrm{o}}.\]

Combining equations 8 and 9 9 ,  

\[\mathrm{C}_{\mathrm{i}}\ {\times}\ \mathrm{v}_{\mathrm{i}}\ {=}\ \mathrm{C}_{\mathrm{o}}\ {\times}\ (\mathrm{v}_{\mathrm{i}}\ {+}\ \mathrm{Fa}).\]

 

In the present measurement system, the perfusion rate (v) was controlled by the injection of physiological saline. Hence,  

\[\mathrm{v}\ {=}\ \mathrm{v}_{\mathrm{i}}.\]

Combining equations 10 and 11 ,  

\[\mathrm{C}_{\mathrm{i}}\ {\times}\ \mathrm{v}\ {=}\ \mathrm{C}_{\mathrm{o}}\ {\times}\ (\mathrm{v}\ {+}\ \mathrm{Fa}).\]

Hence,  

\[\mathrm{Fa}\ {=}\ \mathrm{v}\ {\times}\ (\mathrm{C}_{\mathrm{i}}/\mathrm{C}_{\mathrm{o}}\ {-}\ 1).\]

 

Determination of Conventional and Uveoscleral Outflow

Because each of the experimental measurements were from different eyes, IOP, EVP, Fa, and C were averaged. Using these values, the trabecular meshwork (or conventional) Fc was calculated for control and treated eyes as  

\[\mathrm{Fc}_{\mathrm{t}}\ {=}\ \mathrm{C(IOP}\ {-}\ \mathrm{EVP}).\]

Fu was calculated using the Goldmann equation  

\[\mathrm{Fa}\ {=}\ \mathrm{Fc}_{\mathrm{t}}\ {+}\ \mathrm{Fu}\ {=}\ \mathrm{C(IOP}\ {-}\ \mathrm{EVP})\ {+}\ \mathrm{Fu}.\]

Thus,  

\[\mathrm{Fu}\ {=}\ \mathrm{Fa}\ {-}\ \mathrm{C(IOP}\ {-}\ \mathrm{EVP}).\]

Also, the proportions of Fc and Fu within Fa were calculated, and each was expressed as a percentage of total outflow. 

Results

Aqueous humor dynamics of control and latanoprost-treated eyes in NIH Swiss White mice are shown in Table 1 . Mean (±SD) IOP, EVP, C_i/C_o, Fa, and C of control (PBS) eyes were 15.7 ± 1.0 mm Hg, 10.2 ± 0.8 mm Hg, 1.068 ± 0.014, 0.144 ± 0.04 μL/min (n = 8), and 0.0053 ± 0.0014 μL/min per mm Hg (n = 21), respectively. Mean IOP, EVP, C_i/C_o, Fa, and C of latanoprost-treated eyes were 14.0 ± 0.8 mm Hg, 10.4 ± 1.0 mm Hg, 1.072 ± 0.025, 0.138 ± 0.04 μL/min (n = 8), and 0.0074 ± 0.0016 μL/min per mm Hg (n = 21), respectively. Mean IOP in treated eyes was significantly lower than that in control eyes, with an average reduction of 1.7 mm Hg (P < 0.05, paired t-test). In addition, there was an increase in total outflow facility of 0.0021 μL/min per mm Hg (P < 0.001, unpaired t-test). In contrast, there was no significant difference of EVP and Fa between control and treated eyes. Determination of the 95% confidence intervals of the means for these data sets revealed that for IOP, the interval encompassed IOP reductions of 0.66 to 2.79 mm Hg. The 95% confidence intervals of the mean increase in C constant were 0.00092 to 0.00034 μL/min per mm Hg. In contrast, the 95% confidence intervals for the other data sets included both negative and positive effects—that is, the range crossed zero. This observation is consistent with nonsignificant differences of EVP, C_i/C_o, or Fa between control and treated eyes. To determine whether the increase in outflow facility observed was due to latanoprost and not a consequence of the drops instilled or the benzalkonium chloride preservative present in latanoprost, the experiment was repeated with control eyes receiving a 4-μL drop of 0.02% benzalkonium chloride dissolved in PBS. The C constant measurements were similar in benzalkonium chloride–treated eyes (mean = 0.0051 μL/min per mm Hg) and untreated eyes (mean = 0.0053 μL/min per mm Hg), confirming that the increase in total outflow facility was due to the latanoprost (Fig. 1 ; Table 2 ). 

Estimated Fc and Fu in control eyes derived from the Goldmann equation, which assumes that facility is present only in trabecular meshwork outflow, were 0.028 and 0.113 μL/min, respectively (Table 3) . Treatment with latanoprost did not significantly change these parameters, with Fc and Fu being 0.0278 and 0.113, respectively. Thus, according to the Goldmann equation, uveoscleral flow accounted for 80.5% and 80.3% of total outflow in untreated and treated eyes, respectively. 

Discussion

These data demonstrate the presence of early changes in aqueous humor dynamics in the mouse eye after a single application of latanoprost. A mean reduction in IOP of 1.7 mm Hg is associated with a substantial (45%) increase in outflow facility. Aqueous flow and episcleral venous pressure were not significantly altered by treatment. 

The existing literature regarding the effect of latanoprost on outflow facility is not conclusive. The increase in facility observed in this study is consistent with the 40% to 60% increase in C documented in cynomolgus monkeys ¹ and the 30% increase in facility reported in patients with ocular hypertension. ⁶ In both of these studies, increased facility was measured after twice-daily doses of latanoprost for 5 days. Our findings are significant, in that we demonstrated an increase in outflow facility after a single application of latanoprost. 

The mechanism by which latanoprost increases facility cannot be deduced from these data because the two-level, constant-pressure infusion method used in this study measures total outflow facility but is not specific for trabecular meshwork outflow. Total outflow facility is a combination of trabecular facility, uveoscleral facility and pseudofacility (IOP-dependence of aqueous humor formation). Although uveoscleral outflow facility is often considered to be negligible in the untreated monkey, it is not known whether uveoscleral outflow in the mouse eye is pressure sensitive or whether PG treatment induces facility in the uveoscleral pathways. In isotope accumulation studies in the monkey, PGF_2α administered to cynomolgus monkeys induced a 60% increase in total outflow facility, but there was no increase in trabecular outflow facility. This suggests that latanoprost has the capacity to induce a primary increase in uveoscleral facility and/or pseudofacility. ¹  

Uveoscleral outflow in this study was calculated indirectly with the Goldmann equation by use of the measurements of IOP, EVP, C. An assumption of the Goldmann equation is that facility is only present in trabecular outflow. The results of these calculations suggest that the increase in outflow facility is sufficient to account fully for the reduction in IOP observed, as there is no change in the proportion of uveoscleral outflow to total outflow between control and treated eyes. However, the Goldmann equation does not provide an accurate estimate of the proportion of uveoscleral outflow in situations in which uveoscleral outflow has facility. This situation may be present in the mouse as the absence of changes in trabecular outflow or uveoscleral outflow, as calculated by the Goldman equation, is not consistent with the observation of increased total outflow facility. Direct measures of the relative rates of uveoscleral and trabecular outflow are needed to resolve these questions. 

A 200-ng dose of latanoprost was selected because this has been shown to be the lowest dose that achieves maximum IOP reduction in the mouse. ³ This dose produced maximum lowering of IOP 2 hours after instillation of drops. Levels of IOP, EVP, and C (total outflow facility) in the control eyes of this study were the same as had been previously reported. ^{14 17 18} In contrast, we report a level of Fa that differs from the previously reported value. ¹⁸ The method used to measure aqueous humor flow was changed from the previous study protocol to be more comparable to well-established protocols for assessing aqueous dynamics in primates. ^{19 20} The main difference was that the flow rate in the present study was controlled on the inflow side of the perfusion apparatus, rather than the outflow side. This alteration of the method yielded a value for Fa that was 22% less than previously reported. ¹⁸  

There are several limitations to these results associated with the methods used. First, anesthesia was required for the collection of the measurements. As discussed previously, there is evidence indirectly supporting the view that IOP and EVP measured within 8 minutes after administration of anesthesia, as observed in the present study, closely reflects IOP in the nonanesthetized mouse. ¹⁸ Briefly, both systolic and diastolic blood pressure were observed to be the same during the first 8 minutes after anesthesia as in the untreated awake mouse. Other investigators also have discussed the potential effects of anesthesia during IOP measurement in the mouse. ^{21 22} A second limitation reflects the use of 25 and 35 mm Hg to determine total outflow facility. Aqueous outflow facility in the mouse eye is difficult to measure because the volume of outflow is small when the IOP is within the normal range. By using higher intraocular pressures, we were able to obtain reproducible measurement of flow. It is possible, however, that these higher pressures may injure the outflow pathways and alter aqueous outflow. Structural changes have been reported in monkey aqueous outflow tissues exposed to 50 mm Hg that were not seen at 15, 22, and 30 mm Hg. ^{23 24} In addition, minor changes in the lysosomal system of monkey trabecular meshwork cells were seen at 30 mm Hg that were not observed at 15 and 22 mm Hg. ^{24 25} The IOPs used to measure outflow facility in this model are, however, similar to pressures commonly encountered in the human glaucomatous eyes as well as mouse models of glaucoma. ^{18 26}  

The mechanism by which latanoprost reduces intraocular pressure remains controversial. Several days after treatment with PG, increased matrix metalloproteinases (MMPs) ^{12 13 27 28} and associated remodeling of extracellular matrix ^{27 29 30} have been observed within uveoscleral outflow tissues. It is unlikely, however, that alterations of extracellular matrix or MMPs occurred within the 2-hour period analyzed in the present study, as this time appears to be too brief for the induction of increased MMP release. ^{12 30 31} The mechanism responsible for the early increase of uveoscleral outflow and also the increase in outflow facility is not known. 

The present study demonstrates that drug-induced changes in aqueous dynamics can be measured reproducibly in the mouse. The acute reduction in IOP induced by a single drop of latanoprost is associated with a 45% increase in outflow facility. Because long-term treatment with PG analogues may reduce IOP through several different mechanisms, further investigation of this acute response may help to clarify time-dependent changes in the specific mechanism by which latanoprost increases outflow facility. 

 Supported in part by National Eye Institute Grant EY05990 (RNW), Research to Prevent Blindness (RNW), and the Margaret and Robert Boemer Research Fund of the Foundation for Eye Research (MA).

 Submitted for publication September 8, 2003; revised January 7, 2004; accepted January 26, 2004.

 Disclosure: J.G. Crowston, None; M. Aihara, None; J.D. Lindsey, None; R.N. Weinreb, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Aqueous Humor Dynamics of Controls Eyes and Eyes Treated with Latanoprost

Mouse	IOP (mm Hg)			EVP (mm Hg)			Ci/Co			Fa (μL/min)
		C	L	C	L	C	L	C	L
1	16.9	15.2	11.1	9.1	1.075	1.066	0.149	0.132
2	16.4	13.1	11.2	12.5	1.036	1.055	0.073	0.111
3	14.6	14.7	9.5	10	1.074	1.078	0.148	0.156
4	15.8	13.2	9.7	10.2	1.067	1.043	0.135	0.086
5	16.9	14	10.8	10.2	1.082	1.042	0.164	0.084
6	14.7	14.6	10.3	10.3	1.074	1.103	0.206	0.147
7	14.8	13.8	9.7	10.1	1.061	1.094	0.123	0.187
8	15.6	13.3	9.3	10.6	1.077	1.098	0.154	0.196
Mean ± SD	15.7 ± 1.0	14.0 ± 0.8	10.2 ± 0.8	10.4 ± 1.0	1.068 ± 0.014	1.072 ± 0.025	0.144 ± 0.038	0.138 ± 0.043
Difference	−1.7			+0.2			+0.004			−0.006
P	0.0016			0.692			0.690			0.746
95% CI	−0.66 to −2.79			+1.08 to −0.73			+0.026 to −0.018			+0.0376 to −0.0508

  P is significance by the SNK test with the 95% confidence interval for the difference between treated and control groups. C, control eyes; L, latanoprost-treated eyes.

Figure 1.

 

Effect of latanoprost and benzalkonium chloride on outflow facility (C). Data points for C are plotted for each mouse tested. Group mean values and 95% CI of the mean are shown by the middle horizontal bar and vertical extent of the diamond, respectively. Horizontal overlap lines are drawn above and below the group mean. Overlapping of the range demarcated by these lines in two groups indicates that the two group means are not significantly different at the 95% confidence level.

Table 2.

 

Effect of Latanoprost and Benzalkonium Chloride on Outflow Facility

	Benzalkonium Chloride 0.02 %	Untreated	Latanoprost 0.05%
Mean C constant	0.0051	0.0053	0.0074
95% CI range	0.0041–0.0060	0.0043–0.0063	0.0067–0.0081
Eyes (n)	11	10	21

 Mean C and 95% CI of the mean for control (PBS or benzalkonium 0.02%) or latanoprost-treated mice. C was measured in the right eye of all mice.

Table 3.

 

Effect of Latanoprost on Trabecular Outflow and Uveoscleral Outflow as Determined by the Goldmann Equation

	Fc (μL/min)	Fu (μL/min)	Fc (%)	Fu (%)
Control (untreated)	0.0292	0.115	20.3	79.7
Latanoprost	0.0266	0.114	19.3	80.7

 Fc = C(IOP − EVP); Fu = Fa − [C(IOP − EVP)]; Fc (%) = 100 × Fc/(Fc + Fu); Fu (%) = 100 × Fu/(Fc + Fu).

The authors thank Richard F. Brubaker (Mayo Clinic and Mayo Foundation, Rochester, MN) for valuable advice and Charles Berry (University of California San Diego) for helpful comments on the use of confidence intervals for post hoc analysis.