Posture-Induced Changes in Intraocular, Orbital, Cranial, Jugular Vein, and Arterial Pressures in a Porcine Model

Abstract

Purpose: The purpose of this study was to determine posture-induced changes in arterial blood pressure (ABP), intraocular pressure (IOP), orbital pressure (Porb), intracranial pressure (ICP), and jugular vein pressure (JVP) at various tilt angles in an in vivo pig.

Methods: Anesthetized and ventilated pigs (n = 8) were placed prone on a tiltable operating table. ABP, IOP, Porb, ICP, and JVP were monitored while the table was tilted at various angles between 15 degrees head up tilt (HUT) and 25 degrees head down tilt (HDT) either in stepwise changes (5 degrees per step) or continuously. The mean pressure was calculated from digitized pressure waveforms from each compartment. For stepwise changes in tilt angle the pressures were plotted as a function of tilt angle. For continuous tilt changes, the pressures were plotted as a function of time.

Results: In the case of stepwise changes, ABP remained relatively stable whilst IOP, Porb, ICP, and JVP demonstrated significant differences between most angles (typically P < 0.0001). The difference was greatest for IOP (P < 0.0001) where the average IOP increased from 13.1 ± 1.23 mm Hg at 15 degrees HUT to 46.3 ± 2.03 mm Hg at 25 degrees HDT. The relationship between pressure and tilt angle was almost linear for ICP and JVP, and sigmoidal for IOP and Porb. Interestingly, the effect of changes in tilt angle occurred very rapidly, within a few seconds.

Conclusions: Our results in a pig model demonstrate that changes in posture (tilt angle) induce rapid changes in IOP, Porb, ICP, and JVP, with IOP affected most severely.

The effects of posture-induced changes in pressure in some key compartments, such as the arterial blood, jugular vein, and the intraocular space, clearly have the potential to alter ocular perfusion.^1–3 Changes in cerebrospinal fluid (CSF) pressure may also influence pressure gradients across the lamina cribrosa, an important parameter in glaucoma progression.⁴ The pressure in the orbit and its venous system also has the potential to influence ocular function.⁵ Posture-induced orbital pressure changes have not previously been reported. The present study simultaneously recorded posture (tilt angle) induced changes in pressure in each of the key compartments related to ocular function; namely the arterial blood pressure (ABP), the intraocular pressure (IOP), the orbital pressure (Porb), the intracranial pressure (ICP), and the jugular vein pressure (JVP). The study hypothesis is that changes in tilt angle can produce differentiated pressure changes in different organs and systems. The relative magnitude and approximate timing of such posture-induced changes in each of these compartments may provide new and valuable information regarding their influence on ocular perfusion and function in both physiological and pathological conditions. 

Maintaining pressure within a normal range is critical for many organs in our body. Abnormal pressure could be an important pathogenic factor. For example, elevated IOP is a major pathogenic factor for glaucoma.⁶ ICP needs to be closely monitored in some brain diseases, such as traumatic brain injury.^7–9 Posture-induced pressure changes could be an additional factor in the clinical management of such diseases. Space flight associated neuro-ocular syndrome (SANS) is an example of zero-gravity induced fluid pressure changes within the body.^10–14 However, the pathogenesis of SANS is still not fully understood. Changes in posture, such as tilting of the body, are widely used as a model for microgravity induced pressure changes.^15–17 

In the present study, an anesthetized pig model was used and the five pressures (ABP, IOP, Porb, ICP, and JVP) were simultaneously measured as a function of tilt angle. The relationship between pressure amplitude and tilt angle in these five pressures was determined, and the time course of the amplitude changes estimated. Whereas it is recognized that there are many structural differences in the skull and orbit between pig and man, the pig is accepted as a suitable bio-model for medical research purposes.^18,19 

Methods

All procedures conformed to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by the Animal Ethics Committee of the University of Western Australia. 

Animal Preparation

Eight juvenile pigs (Large white x Landrace x Duroc) weighing 29.8 (± 3.3) kg were used in this study. The pigs were sourced commercially and acclimatized to the Large Animal Facility at the University of Western Australia for at least 5 days prior to surgery. The room was maintained at 22 ± 2°C and had a 12:12 hour light/dark cycle. 

Anesthesia was induced with a combination of zolazepam and tiletamine (4 mg/kg, Zoletil 100, 100 mg/mL; Virbac Australia Pty Ltd., Milperra, NSW, Australia) and xylazine (2 mg/kg Xylazine, Ilium Xylazil 100, 100 mg/mL; Troy Laboratories Pty Ltd., Glendenning, NSW, Australia) by intramuscular injection in the trapezius muscle of the neck. A branch of the right auricular vein was cannulated and propofol (1-2 mg/kg, Propofol-Lipuro 1%, 10 mg/mL; B Braun Australia Pty Ltd., Bella Vista, NSW, Australia) was administered by intravenous injection to facilitate oral endotracheal intubation with a cuffed endotracheal tube (Portex; Smiths Medical, Plymouth, MN, USA). The pig was then placed in the prone position and secured to the operating table with duct tape. General anesthesia was maintained with a combination of intravenous propofol and fentanyl (DBL fentanyl injection, 50 mg/mL; Hospira Australia Pty Ltd., Melbourne, VIC, Australia) initially at 3.1 (± 0.5) mg/kg/h and 8.5 (± 2.1) mg/kg/h, respectively, and adjusted to maintain an adequate depth of anesthesia. Anesthesia was monitored continuously by experienced animal care staff and a veterinary anesthetist. 

Volume cycled mechanical ventilation (Datex Ohmeda ADU anesthetic machine; GE Healthcare, Sweden) was commenced immediately after tracheal intubation and adjusted to target normocapnia (end-tidal CO₂ 35 to 45 mm Hg). The FiO₂ was 0.3 using a combination of oxygen and medical air. Positive end-expiratory pressure was set at 5 cm H₂O. A triple lumen central venous catheter was placed in the external jugular vein by ultrasound guidance. This catheter was used for the administration of anesthetic drugs and measurement of central venous pressure. Intra-operative monitoring included oxyhemoglobin saturation and pulse rate from a pulse oximeter placed on the pinna, capnography, pharyngeal temperature, electrocardiography, central venous pressure, and direct arterial blood pressure from a catheter placed in an auricular artery. These parameters were measured continuously and recorded every 5 minutes on a paper anesthetic record. Hartmann's solution (compound sodium lactate; Baxter, Toongabbie, NSW, Australia) was administered at 10 mL/kg/h into the auricular vein during anesthesia. 

Pressure and Tilt-Angle Measurements

Sites for pressure measurements are illustrated in Figure 1. Cobe pressure transducers (LivaNova, Houston, TX, USA) were used in combination with a fluid filled tube and hypodermic needle tip (IOP measurement) or a catheter (femoral arterial and jugular venous pressure). Codman transducers (Integra; LifeScience, Princeton, NJ, USA) were used for ICP and Porb measurements. For ICP measurements, the tip of the Codman transducer was placed in the subdural space. A bolt was used for securing the transducer on the skull without any leakage according to the standard procedure recommended by the manufacturer. For Porb measurements the tip of the Codman transducer was placed in the temporal side of the orbit near the muscular cone in the left eye. A custom-built device was used to seal around the entry hole and allow entry of the Codman transducer. The tilt angle of the operating table was measured from the analogue output of an inclinometer and the operating table could be adjusted to lie among 15 degrees up from horizontal, that is, 15 degrees head up tilt (HUT), to 25 degrees below horizontal, that is, 25 degrees head down tilt (HDT). 

Figure 1.

 

A schematic drawing to illustrate the experimental setup. The pig was placed on a tiltable operating table. Five pressures, ABP, IOP, Porb, ICP, JVP, as well as tilt angle (TA) were measured simultaneously. All signals were amplified, and low pass filtered prior to analogue to digital conversion (A/D) and data logging. Fluid filled catheters connected with Cobe transducers were used for ABP, IOP, and JVP measurements, whereas Codman transducers were used for Porb and ICP measurements. The operating table tilt angle could be adjusted between 15 degrees head up tilt (HUT) and 25 degrees head down tilt (HDT) by manually rotating a control wheel. Insert (A) is a custom-built device for sealing skin entry and connecting bolt. (B) Is a photograph showing this device during the experiment. (C) Is a photograph showing fixation of a Codman transducer by a bolt. Insert (D) is a schematic drawing showing the placement of a Codman transducer for subdural ICP measurement.

All signals were amplified, and low pass filtered prior to analog to digital (A/D) conversion and data capture via National Instruments A/D interfaces (USB X Series) at a sampling rate of 1000 Hz. An in-house custom-written LabView program was used to calibrate each signal, graphically display all channels on-screen, and export the acquired data (including the average pressure value for each channel) to a Microsoft Excel spreadsheet for later analysis. Each channel was also recorded continuously on an 8-channel chart recorder (Yokogawa LR8100, Tokyo, Japan). 

Experimental Protocol

Baseline (Horizontal Position) Pressure Measurement

Once the transducers were in position and the animal was stable, all 5 pressures were measured (LabView and chart recorder) with the table horizontal and recorded in a 10-second burst. 

Stepwise Changes in Tilt-Angle

Stepwise changes in tilt angle were performed at 5 degrees per step from 15 degrees HUT to 25 degrees HDT. The 5 pressures were recorded (LabView and chart recorder) at each tilt angle for a duration of 10 seconds. The animals were held at each stepwise tilt angle for approximately 30 seconds before pressure recordings were collected. This was a sufficient time for the pressures to stabilize. 

The exclusion of pressure data from any tilt angle change run was only necessary when there was an identified problem with the fidelity of one of the pressure recordings. For example, an unusual IOP response was sometimes caused by the needle tip touching the iris or cornea due to movement of the eyeball position during tilting. Another example was an unusual ABP response due to a temporary thrombus blockage of the arterial blood pressure line prior to flushing. 

Continuous Changes in Tilt-Angle

The 5 pressures and tilt angle were also recorded (LabView and chart recorder) while the tilt angle was continuously varied from 15 degrees HUT to 25 degrees HDT, or from 25 degrees HDT to 15 degrees HUT, by manually turning the control wheel at a constant rate. This was done to determine the relationship between the tilt angle and the pressure amplitudes, and to estimate the time course of amplitude changes. 

At the end of the experiment, the pigs were euthanized by intravenous injection of pentobarbitone (160 mg/kg, Lethabarb euthanasia injection, 325 mg mL⁻¹; Virbac Australia Pty Ltd., NSW, Australia). 

Data Analysis and Statistical Methods

Data and statistical analyses were performed using SigmaPlot (version 15; SigmaPlot, Palo Alto, CA, USA) and R (version 4.3.1).²⁰ All statistical tests were performed at the 5% level of significance. 

For the baseline (horizontal position) pressure measurement data for each pig, the pressure values were plotted against time for each pressure compartment in a stacked plot (vertically stacked with the same horizontal time axis). 

For the stepwise changes in tilt angle data, the average pressure values, hereinafter called pressure responses, were plotted against tilt angle for each pressure compartment in a stacked plot (vertically stacked with the same horizontal tilt angle axis) for each pig and for the pooled data. The SigmaPlot regression wizard was used to fit either a straight-line or sigmoid function to pressure response versus tilt angle for each compartment for representative data from one of the pigs. R was used to fit a separate (generalized) linear mixed effects model to the data (from all pigs) for each pressure compartment, to investigate statistically significant differences in the average pressure values among the different tilt angles. The mixed effects modeling and analyses were performed using the R packages lme4,²¹ MASS,²² and emmeans.²³ Each model (one for each pressure compartment) was fitted with tilt angle as a fixed effect and pig number as a random effect. 

For the continuous changes in tilt angle data, the pressure values and the tilt angle were all plotted against time in a stacked plot (vertically stacked with the same horizontal time axis). This plot was used to identify the relationship between tilt angle and pressure value in each pressure compartment, including the timing and magnitude of tilt-related pressure changes. 

The number of recording sessions from individual animals were not equal. The number of recording sessions were dependent on animal preparations, systemic conditions, and instrument performance. Therefore, linear mixed models were used for statistical analysis. For stepwise tilt angle data, we pooled data from 5 pigs and had 6 recording sessions from pigs #4 and #7, 9 from pigs #5 and #8, and 15 from pig #6. 

Results

Baseline (Horizontal Position) Pressure Measurements

Figure 2 shows a set of representative pressure waveforms simultaneously recorded. All pressures were pulsatile, whereas each pressure waveform has differences in shape, pressure level, pulse amplitude, and number of sub-waveforms. 

Figure 2.

 

Representative pressure waveforms from one of the pigs. Three cardiac cycles are included divided by the red dashed lines and systolic and diastolic phases of the first cycle are divided by the black dashed line. Note that all five pressures were simultaneously recorded. Each pressure (A) ABP, (B) IOP, (C) Porb, (D) ICP, and (E) JVP is plotted on a different pressure scale to enhance the visibility of the waveforms.

Stepwise Changes in Tilt Angle

The results from stepwise changes in tilt angle (5 degrees per step from 15 degrees HUT to 25 degrees HDT) from one pig are shown in Figure 3. The ABP, ICP, and JVP points each appear to fall on a straight line, whereas the IOP and Porb points appear to fall on a sigmoid curve. 

Figure 3.

 

Representative set of the five compartment pressures as function of tilt angle in a representative pig. Responses of ABP (A), IOP (B), Porb (C), ICP (D), and JVP (E) to tilt angle from 15 degrees HUT to 25 degrees HDT. Each response (point) is the average of the amplitudes of the pressure waveform recorded at a specific tilt angle. IOP, Porb, ICP, and JVP responses tended to increase with tilt angle, but not ABP responses. The red dashed lines show the results of curve fitting: straight-line fits for ABP, ICP, and JVP, and sigmoid function fits for IOP and Porb (see the text for details). Note the different Y axis scales in each plot.

SigmaPlot was used to fit a straight line (y = mx + b) separately to the responses for ABP, ICP, and JVP. The coefficient of determination in each case shows this model fits the observations very well: R² = 0.965 for ABP, R² = 0.999 for ICP and R² = 1.000 for JVP. The slopes of the fitted lines were m = −0.530, m = 0.528, and m = 0.577, respectively. The data shows that the responses for ICP and JVP increased from 5.4 and -1.5 mm Hg at 15 degrees HUT, 12.5 and 7.5 mm Hg at 0 degrees, and then to 26.6 and 21.6 mm Hg at 25 degrees HDT, respectively. However, the responses for ABP reduced slightly. 

SigmaPlot was used to fit the following four-parameter sigmoid function separately to the responses for IOP and Porb: 

\begin{eqnarray*}y = {y_0} + \frac{a}{{1 + {e^{ - \left( {\frac{{x - {x_0}}}{b}} \right)}}}}\end{eqnarray*}

 

The coefficient of determination was R² = 0.998 for IOP and R² = 0.997 for Porb. The data show that the IOP response increased from 12.5 mm Hg at 15 degrees HUT, to 15.3 mm Hg at 0 degrees, to 42.5 mm Hg at 25 degrees HDT, whereas the Porb response increased from 4.9 mm Hg at 15 degrees HUT, to 7.5 mm Hg at 0 degrees, to 21.6 mm Hg at 25 degrees HDT. 

Figure 4 shows the five compartment pressures as a function of stepwise changes of tilt angle pooled across all pigs and recording sessions. Each colored symbol or response is the average of the amplitudes of the waveform recorded at a given tilt angle for a single pig in a single recording session (single set of measurements over all angles). The average ICP (Fig. 4D) and average JVP (Fig. 4E) responses (large black dots) appear to fall on straight lines, increasing as the tilt angle was altered stepwise from 15 degrees HUT to 25 degrees HDT. 

Figure 4.

 

Pooled data for the five compartment pressures as a function of stepwise changes of tilt angle. Responses of ABP (A), IOP (B), Porb (C), ICP (D), and JVP (E) to changes in tilt angle from 15 degrees HUT to 25 degrees HDT. Each colored symbol corresponds to an individual measurement. Multiple colored symbols joined by a thin line correspond to measurements from a recording session for a single pig. The large black dots and horizontal bars correspond to the average and standard deviation, respectively. Note the different Y axis scales for each plot.

Separate linear mixed effects models were fitted to the data for ICP and JVP, respectively, with the angle as a fixed effect and the pig as a random effect. To satisfy model assumptions (normally distributed residuals with constant variance), it was necessary to transform the ICP and JVP responses using the power transformation (y − β)^α with α = 1.5,  β = 0 for ICP, and α = 1.3,  β = −2.2 for JVP (Shapiro-Wilk normality test, P = 0.4659 and P = 0.4794, respectively). From the ICP model (estimated marginal means), the average ICP response increased from 4.75 ± 1.556 mm Hg at 15 degrees HUT to 26.17 ± 0.663 mm Hg at 25 degrees HDT. The results of a post hoc multiple comparison test (using Tukey P value adjustment) showed this difference, and indeed differences between all angles, to be statistically significant (P < 0.0001). From the JVP model, the average JVP response increased from 1.15 ± 0.868 mm Hg at 15 degrees HUT to 22.95 ± 0.474 mm Hg at 25 degrees HDT. This difference, and indeed all differences between all angles, were statistically significant (all P < 0.0001). 

A generalized linear mixed model (inverse Gaussian family with link = “1/mu^2”) was fitted to the ABP data. The fit satisfied model assumptions, that is, normally distributed residuals with constant variance (Shapiro-Wilk normality test, P = 0.0926). From the model, the average ABP response was 73.7 ± 2.61 mm Hg at 15 degrees HUT and 64.7 ± 1.88 mm Hg at 25 degrees HDT. The results of a post hoc multiple comparison test (using Tukey P value adjustment) showed this difference to be statistically significant (P < 0.0001). The only angles that had a statistically significant difference to 0 degrees were 15 degrees, 20 degrees, and 25 degrees HDT (P = 0.0105, P = 0.0001, and P < 0.0001, respectively). 

A generalized linear mixed model (quasi family with link = power(0.63), variance = “mu”) was fitted to the IOP data. The fit satisfied model assumptions, that is, normally distributed residuals with constant variance (Shapiro-Wilk normality test, P = 0.1026). From the model, the average IOP response was 13.1 ± 1.23 mm Hg at 15 degrees HUT, 18.8 ± 1.42 mm Hg at horizontal position and 46.3 ± 2.03 mm Hg at 25 degrees HDT; the latter is a 33.2 mm Hg increase from the 15 degrees HUT and 27.5 mm Hg from the horizontal position. Interestingly, a post hoc multiple comparison test (using Tukey P value adjustment) showed all differences among all angles were statistically significant (all P ≤ 0.0033), except for between 5 degrees HUT and 10 degrees HUT, 5 degrees HUT and 15 degrees HUT, and 10 degrees HUT and 15 degrees HUT. 

SigmaPlot was used to fit a 4-parameter sigmoid function to the average IOP response data (large black dots) yielding R² = 0.999. 

A linear mixed model was fitted to the Porb data. The fit satisfied model assumptions, that is, normally distributed residuals with constant variance (Shapiro-Wilk normality test, P = 0.3385). From the model, the average Porb response was 11.3 ± 3.09 mm Hg at 15 degrees HUT, 13.6 ± 3.09 mm Hg at the horizontal position and 21.6 ± 3.09 mm Hg at 25 degrees HDT; the latter is a 10.3 mm Hg increase from 15 degrees HUT and an 8 mm Hg increase from the horizontal position. Interestingly, a post hoc multiple comparison test (using Tukey P value adjustment) showed all differences between all angles were statistically significant (all P ≤ 0.0450), except between horizontal and 5 degrees HUT, 5 degrees HUT and 10 degrees HUT, 10 degrees HUT and 15 degrees HUT, 5 degrees HUT and 15 degrees HUT, and 5 degrees HDT and 10 degrees HDT. 

Sigma Plot was used to fit a 4-parameter sigmoid function to the average Porb response data (large black dots) yielding R² = 0.999. 

Continuous Changes of Tilt Angle

Continuous changes of tilt angle were used to determine the responses of ABP, IOP, Porb, ICP, and JVP to changes of tilt angle as a function of time. 

Figure 5 shows representative data from a continuous tilt change from 25 degrees HDT to 15 degrees HUT. In the approximately 3 second period before changing the tilt angle all the pressures were relatively stable. When starting to tilt up, IOP, ICP, Porb, and JVP almost immediately reduced. Manual turning of the control wheel to change the tilt angle from 25 degrees HDT to 15 degrees HUT usually took approximately 15 seconds. During this period, IOP, ICP, Porb, and JVP gradually reduced. When the tilt was stopped at 15 degrees HUT, IOP, ICP, Porb, and JVP became relatively stable again. 

Figure 5.

 

Representative data of measurements of the five pressures during continuous change of tilt angle from 25 degrees HDT to 15 degrees HUT. Plots (A) ABP, (B) IOP, (C) Porb, (D) ICP, (E) JVP, and (F) tilt angle as a function of time. Note that each pressure scale is different. Approximately 3 seconds was recorded at 25 degrees HDT before a continuous change in tilt angle back up to 15 degrees HUT. A dashed red line indicates the start of changing the tilt angle.

Figure 6 shows a second set of representative data, but from a continuous tilt change from 25 degrees HDT to 0 degrees. Interestingly there was a short pause and slight turning back of the tilt control wheel at approximately 15 seconds during the tilting indicated by a red arrowhead (see Fig. 6F). This induced a small pressure change in IOP, ICP, Porb, and JVP, but not in ABP. From a visual inspection of the induced peaks corresponding to the tilt angle change peak (dashed red line), it can be seen that the time delay of the pressure change after the tilt angle change was very short (<1-2 seconds). However, precise determination of time delay was problematic due to the pulsatile nature of the pressures. However, these data certainly supports the conclusion that the time delay between tilt start and changes of pressures was very short. 

Figure 6.

 

Representative data of measurements of five pressures during continuous changes of tilt angle from 25 degrees HDT to 0 degrees. Plots (A) for ABP, (B) IOP, (C) Porb, (D) ICP, (E) JVP, and (F) tilt angle. Note that each pressure scale is different to show changes in pressure as a function of time. Approximately 5 seconds was recorded at 25 degrees HDT before changing tilt angle. Manual tilting from 25 degrees to 0 degrees takes approximately 12 seconds. The blue arrows indicate starting and end of tilting the operation table (F). A red arrowhead with a dashed line indicates where the rotation of the turning control wheel was reversed for a short time before continuing with the original direction of rotation.

Figure 7 is a boxplot of the pressure values for IOP, ICP, Porb, and JVP measured during continuous change of tilt angle pooled over all data. The average magnitude was 26.3 ± 3.1 mm Hg in IOP, 22.7 ± 1.4 mm Hg in ICP, 22.2 ± 2.5 mm Hg in JVP, and 10.7 ± 2.8 mm Hg in Porb. Interestingly, the average IOP was significantly different to that in all other pressures (P = 0.040 for ICP, P = 0.017 for JVP, and P < 0.001 for Porb; repeated measures ANOVA with n = 23 recordings). These results are consistent with Figures 2 to 6 and indeed for all the data where the average IOP is in all cases greater than that for ICP, JVP, and Porb. 

Figure 7.

 

Pooled data of the measurements of IOP, Porb, ICP, and JVP during continuous changes of tilt angle. HUT to HDT marked with red symbols. HDT to HUT marked with blue symbols. Measurements of IOP, Porb, ICP, and JVP during continuous changes in tilt angle. Multiple small symbols show individual measurements.

There were no statistically significant differences in the magnitude changes for IOP, Porb, ICP, and JVP among tilt angle changes from 15 degrees HUT to 25 degrees HDT and 25 degrees HDT to 15 degrees HUT (all P > 0.05). 

Discussion

As far as we know, there have been no previous studies in which five key pressures (ABP, IOP, Porb, ICP, and JVP) have been measured simultaneously during posture (tilt angle) induced changes in a large animal model. Therefore, such a study may provide some valuable insights of posture-induced changes in different compartments that may influence ocular perfusion and function. A limitation of the present study is that only acute changes in tilt-induced compartment pressures were studied. It is certainly the case that pressure changes in SANS and glaucoma are long-term effects. 

We used tilt angles from 15 degrees HUT to 25 degrees HDT, as these were the limits of adjustability of the operating table but this provided a sufficiently large range of tilt angles to induce significant pressure changes based on the existing literature.^13,24–28 However, our range of tilt angles is significantly less than used in human studies. In selecting 5 degrees per step, we sought to obtain sufficient data points to define the relationship between pressure changes and tilt angle. 

The major findings from this study are (1) we observed different responses to tilt angle changes in the five pressure sites measured, (2) the very rapid time course of any tilt-induced changes, and (3) tilt-induced changes in IOP were more severe than in any other compartment monitored. 

Simultaneous monitoring of ABP, IOP, Porb, ICP, and JVP with tilt angle changes provides valuable information of the relationship between IOP and other pressures. It is clearly evidenced that IOP, Porb, OCP, and JVP were reduced when tilt angles changed from 25 degrees HDT to 15 degrees HUT, whereas ABP was relatively stable. Such reductions in IOP, Porb, ICP, and JVP occurred rapidly, within a few seconds after changing the tilt angle. 

It is interesting to note that IOP was most severely affected by tilt angle changes. Our results show that the IOP response (average amplitude of the recorded pressure waveform) increased 33.2 mm Hg from 15 degrees HUT to 25 degrees HDT with stepwise changes (see Fig. 4) and that this difference was statistically significant. IOP response was higher than ICP, JVP, and Porb at 25 degrees HDT (see Figs. 3, 4). It is clearly evidenced that IOP response as a function of tilt angle is not linear but more like a sigmoid curve (see Figs. 3, 4). For all pairs of tilt angles, the difference in IOP response was statistically significant, except among 5 degrees HUT and 10 degrees HUT, 5 degrees HUT and 15 degrees HUT, and 10 degrees HUT and 15 degrees HUT. We also found that IOP had a rapid response to tilt angle change with IOP changes occurring within a second or two. 

Posture-induced IOP changes have been studied in human subjects.^3,28–32 It was reported that there was significant increase in IOP up to more than 44 mm Hg or 2-fold increase during headstand posture.^3,29,33 Such an increase in IOP could potentially create further damage in glaucoma patients.^28,34 In addition, Friberg and Weinreb³³ also showed the eye lid had numerous scattered petechiae on the upper lid. We found notable eyelid swelling and eyeball movement forward at 25 degrees HDT indicating that IOP increase may be associated with orbital changes.²⁸ 

To precisely measure Porb, we had to build a custom device to allow a Codman transducer to be inserted in the temporal side of the orbital tissue space and seal the entry point with a skin clamp. A needle and optical fiber endoscope were used to determine the depth and orientation of the tip of the Codman transducer. We managed to measure pulsed Porb with some variations between individual animals. Time delay of Porb response to tilt angle changes was comparable with that of IOP. The sigmoidal relationship of Porb was also found as a function of stepwise tilt angle changes. However, the magnitude of Porb changes was smaller than that of the IOP. 

The average Porb of 13.6 mm Hg at horizontal position from our study is higher than the measured values from humans.^35–37 It could be caused by species differences between human and pig particularly in the orbit structure and anatomy. The pig has a relatively open orbit with a strong fibrous ligament stretching from the frontal bone to the zygomatic bone, effectively enclosing the orbit but with slightly less rigidity than in humans. Another reason could be the measuring techniques. We have used a custom-built device to make sure that there was a very tight seal of the skin around the Codman probe. The large scatter of Porb values from different measurements could be influenced by different locations of the tip of the pressure transducer. It was a challenge to try and determine the exact location of the tip of the pressure transducer in the orbit although multiple approaches were used, such as the use of a needle fiber optical endoscope and an ultrasonic device. The sigmoid function fits for Porb are interesting and Porb may have a critical influence for IOP. We believe that Porb could be an interesting topic for further investigation. 

As far as we know, posture-induced Porb changes have not been previously reported. The orbit is a compartment between two important compartments, the intraocular and intracranial compartments. It is likely that that the orbit will receive more attention in the near future not only because of its special anatomic location, but also because it contains the eyeball which is closely associated with ocular diseases. Porb measurements could be of interest as the orbit contains the eyeball, optic nerve with subarachnoid space, and rich blood vessels in a limited volume bone-lined structural chamber.^37,38 Mathematical modeling has been used to investigate the effects of cerebrospinal fluid system on microstructures and posture on optic nerve subarachnoid space dynamics.³⁹ The roles of optic nerve and orbital changes in SANS have received specific attention.^40,41 

ICP is derived from cerebral blood and CSF circulatory dynamics and plays critical roles in the central nervous system. Posture-induced changes in ICP have been recognized for decades.⁴² ICP measurements have been performed in the supine position in human subjects, with reported values of 5 to 15 mm Hg, whereas our mean value (estimated marginal mean) is 11.91 ± 0.982 mm Hg for 0 degrees in pigs. Our results show approximately linear changes from 15 degrees HUT (4.75 ± 1.556 mm Hg) to 25 degrees HDT (26.17 ± 0.663 mm Hg). The influence of posture on ICP in humans has also been studied.⁴² Postural transition was studied from the supine to upright position. Such postural transition is largely equivalent to continuous tilt change described in our study. However, tilt angles were different between our study from 25 degrees HDT to 15 degrees HUT and their study from the supine to the upright position (90 degrees) in human subjects. They found a biphasic decrease in ICP. At first phase (tilt angle between 0 degrees and head up 30–45 degrees) ICP was rapidly decreased and proportional to tilt angle. At the second phase, which was following the first phase and tilt angle was higher than head up 30 to 45 degrees, ICP was relatively stable.⁴² It is expected that multiple mechanisms may also be involved in posture-induced ICP changes, such as the spinal buffer the Monro-Kelly doctrine states in the different compartments of ICP.^43–45 Interestingly, the results from human subjects agree with the findings from our data for both continuous tilt and stepwise tilt, although we do not have data from head up more than 15 degrees. 

Our results for the pig data show approximately linear changes for mean JVP from 15 degrees HUT (1.15 ± 0.868 mm Hg) to 25 degrees HDT (22.95 ± 0.474 mm Hg). Interestingly, more recently a similar study on the effects of stepwise tilt angle changes on JVP was performed in human subjects.⁴⁶ Measurements were taken at a seated baseline, 45 degrees HUT to 45 degrees HDT with 15 degrees per step. This article is one of very few reports which has investigated the effects of stepwise tilt angle on JVP. Therefore, we have an opportunity to compare their results with our data. They show that JVP remains largely unchanged at approximately 10 mm Hg from 45 degrees HUT to 15 degrees HUT, but from 15 degrees HUT to 45 degrees HDT it demonstrates an almost linear increase from approximately 10 mm Hg to 50 to 60 mm Hg. Our results show that JVP demonstrates an almost linear increase from 15 degrees HUT to 25 degrees HDT. They found that the changes of the internal jugular vein area and the internal JVP are closely associated and found that common carotid artery cross-sectional area was unchanged during changes from 45 degrees HUT to 45 degrees HDT with 15 degrees per step. It has been reported that jugular vein collapse may influence the posture-induced changes in JVP.¹⁷ 

Our data clearly shows that the response of ABP to tilt angle changes was different to that of IOP, Porb, ICP, and JVP with either continuous tilt or stepwise tilt changes. It was difficult to discern ABP changes during continuous tilt, but ABP did change (with some notable variance between pigs) during stepwise tilt both for HUT and HDT. This may indicate that ABP responds to tilt angle changes and attempts to maintain stability through adaptation mechanisms. We did not observe any physical differences between animals or their condition that would explain the variations in tilt-induced pressure responses between animals. It may be that anesthesia level plays a role. Although we had the expert services of a veterinary anesthetist, and an experienced veterinary nurse, some variation between animals or during different phases of the experiment are unavoidable. The findings from a study in human subjects¹ show ABP does not initially change but, over time, does increase during the head down position. 

The capability of relatively strong adaptation of cardiovascular system during spaceflight has been described.^15,47,48 

Posture-induced changes in pressures are interesting and significant. The mechanisms involved in each pressure are different and worthy of further investigation. The advantage of using a pig model is that it provides an opportunity to simultaneously monitor multiple pressures including ABP, IOP, Porb, ICP, and JVP. The results from our posture-induced pressure change study mostly agree with those from human subjects, but we are aware there are some significant anatomical differences between pigs and human subjects. 

As far as we know, our study is the first study to simultaneously monitor changes of in ABP, IOP, Porb, ICP, and JVP with stepwise and continuous changes in tilt angle. We currently know very little about the mechanisms behind these responses. The nonlinear effects in some compartments (IOP and Porb) are particularly interesting, but we have no explanation for such effects. Further studies in which compartment structure is monitored by magnetic resonance imaging (MRI) during larger changes in tilt angle are proposed, which may shed further light on the specific mechanisms involved. 

The timing of tilt angle induced pressure responses is very interesting. It is clear that IOP, Porb, ICP, and JVP almost immediately responded to tilt angle changes (see Figs. 5, 6). Such rapid responses occurred in each experiment. Furthermore, Figure 6 shows that tiny changes in tilt angle (a few degrees) can induce quick response of IOP, Porb, ICP, and JVP. 

We attempted to quantify the time delay between tilt angle and pressure responses of IOP, Porb, ICP, and JVP. The starting point of tilt angle change could be reliably determined, but the starting point of pressure response could not despite using several different approaches. The presence of pressure pulsation and the effects of respiration gave unreliable results. However, the time delay could be a clue to define the mechanisms of tilt angle change-induced pressure responses and the inter-relationship between tilt angle change-induced IOP, Porb, ICP, and JVP. In future studies, we plan to have the tilt angle changed via an electric motor, such that the start of the tilt change can be synchronized to the heart or respiration phase. 

Combination of the data of our experimental study and the data from human subjects may provide useful information regarding how our body responds and adapts to changes in posture or gravity. Further study is certainly required to address some crucial questions, such as the mechanisms behind such pressure changes, the inter-relationship between these pressures, and possible intervention to prevent microgravity induced pressure changes. 

Acknowledgments

Grant support was provided by a National Health and Medical Research Council of Australia (Investigator Grant (APP1173403), and Project Grant (APP1162615)). 

The authors acknowledge the expert technical assistance of staff at the Large Animal Facility, The University of Western Australia. 

Supported by a National Health and Medical Research Council of Australia (Investigator Grant APP1173403, and Project Grant APP1162615). 

Disclosure: D.-Y. Yu, None; S.J. Cringle, None; D. Darcey, None; L.Y.H. Tien, None; A.J. Vukmirovic, None; P.K. Yu, None; A. Mehnert, None; W.H. Morgan, None