MATERIALS AND METHODS

Surgery.

Randomly bred female sheep (24–40 kg) were used in these studies. Experiments were approved by the Ethics Committee at the Sunnybrook Health Science Centre and conformed to the guidelines set by the Canadian Council on Animal Care and the Animals for Research Act of Ontario. Access to the CSF, blood, and lymph compartments was achieved as described previously (3). A minimum of 3 days before the experiment, catheters were inserted into both lateral ventricles under halothane anesthesia. On the day of the experiment, multiple cervical lymphatics were cannulated (0.58 mm ID, 0.96 mm OD; Critchley, Silverwater, Australia), and lymph was collected into a single heparinized tube. Lymphatic vessels too small to cannulate were ligated. The left jugular vein was cannulated, and a solid-state pressure transducer (model 9815-F7, Honeywell) was advanced into the superior vena cava. For access to the cisterna magna, a laminectomy of C₁ was performed. A 14-gauge angiocatheter was used to puncture the dura and arachnoid, leaving the catheter in the subarachnoid space. Free movement of CSF within the catheter assembly confirmed placement.

Tracers and solutions.

¹²⁵I-HSA (0.93 MBq/ml, 10 mg/ml) and ¹³¹I-HSA (37 MBq/ml, 10 mg/ml) were obtained from Frosst (Kirkland, PQ, Canada). All tracer solutions were purified before use by passage through a Centricon centrifugal concentrator (10,000-mol wt cutoff) to remove free¹²⁵I or¹³¹I before infusion. To ensure that the measured radioactivity in any collected sample was protein associated, a second set of aliquots was assayed after precipitation with 10% TCA. Free or non-protein-associated¹²⁵I or¹³¹I represented <1% of the total radioactivity in any sample. Macrodex saline solution (6% Dextran 70, Baxter) was purchased from Pharmacia (Quebec, PQ, Canada). Artificial CSF was made as described by Chodobski et al. (7).

Experimental design.

Tracer administration into the CSF compartment and control of CSF pressure was achieved with a ventriculocisternal perfusion system. A constant-pressure reservoir supplied artificial CSF to a lateral ventricle, and because the height of this reservoir was elevated above the outflow catheter placed in the cisterna magna, CSF flowed through the system at 0.2–0.7 ml/min. ICP was determined by the height of the reservoir and outflow catheter relative to the animal. Just before the experiment, the autologous nonradioactive CSF was exchanged with radioactive artificial CSF via a low-pressure perfusion. Attime 0 (beginning of experiment), ventriculocisternal perfusion with a flow marker (¹²⁵I-HSA) was initiated for 3 h at the chosen lower pressure. A second label of the same CSF flow marker (¹³¹I-HSA) was injected intravenously to allow correction of the data for filtration errors. After 3 h, the heights of the reservoir and outflow catheter were elevated to achieve the desired higher pressure. Two series of experiments were performed with low to high pressures set at 10 and 20 cmH₂O in one series and 20–30 cmH₂O in another. In two additional animals, pressure-drainage parameters were compared between 0 and 30 cmH₂O and between 5 and 15 cmH₂O.

The cervical lymphatics empty into the venous system at the base of the neck; therefore, cervical lymph flow is opposed by the central venous pressure (CVP). Because the cervical duct is cannulated in our experiments, the normal outflow pressure into which this vessel transports lymph would be altered. This was not true of the transport of CSF through arachnoid villi, since these structures remained connected to venous drainage. To maintain the physiological relationships as close as possible to the native state, we simulated the outflow pressure encountered by the cervical lymphatic at its normal lymphatic-venous anastomosis by adjusting the height of the lymphatic outflow catheter to create a total outflow resistance equivalent to the CVP, which was monitored continuously throughout the experiment. The pressure at the outflow end of a lymphatic was estimated as follows: P_o =Q˙R_o +h, where P_o is outflow pressure (in cmH₂O), Q˙ is flow (in μl/min), R_o is outflow catheter resistance (in cmH₂O ⋅ l⁻¹ ⋅ min), and h is height of the outflow end of the cervical lymphatic catheter (in cm above the lymphatic in the animal) (9). The total resistance of the lymphatic catheter was determined experimentally by measuring the flow of artificial CSF at different outflow heights. Flow varied linearly with perfusion pressure, and the gradient of this relationship gave a resistance factor that could be incorporated into the expression noted above along with the measured value of P_o, which was taken to be equal to CVP. From this, we were able to calculate the appropriate height of the outflow tip of the cervical lymphatic catheter to achieve the desired level of P_o.

Blood, lymph, and CSF were sampled every 30 min. Macrodex saline solution was infused into the jugular cannula to maintain hydration of the animal during anesthesia. Levels of radioactivity in the samples were determined using a multichannel gamma spectrometer (Compugamma, LKB Wallac, Turku, Finland) with appropriate window settings and background subtraction.

Mathematical model.

The CSF drainage rate through arachnoid villi and cervical lymphatics can be roughly estimated by dividing the total amount of radioactivity recovered in the plasma or lymph compartments by the concentration of tracer in the CSF. However, this approach does not take into consideration potential problems associated with tracer filtration. Because plasma recoveries in lymph-diverted animals were used to estimate arachnoid villi drainage, tracer loss through blood capillary filtration would result in this pathway being underestimated. Similarly, the transport of tracer to the plasma complicates measurements of the CSF tracer in lymph, since the HSA that was transported from the CSF into plasma by the arachnoid villi would filter back into the lymphatic compartment, resulting in an overestimate of the lymphatic contribution to CSF drainage.

To correct tracer recovery data for problems caused by filtration, we developed previously a three-compartment mathematical model. Tracer data were inserted into derived mass balance equations to provide estimates of the CSF flow rates through the arachnoid villi and lymphatic drainage pathways (2). This model integrates brain lymph and arachnoid villi drainage systems with the CSF and vascular compartments. It is assumed that there is no direct transport of tracer into CNS capillaries and that all tracer is removed from the CNS by arachnoid villi or extracranial lymphatics. All tracer concentrations (C) are dependent on time. Intercompartmental flow rates [observed lymph flow rate (L), volumetric transfer rate of CSF tracer from the plasma (F), and estimated CSF drainage rates (D)] are assumed to remain constant.

At the initiation of the experiment, a flow marker (¹²⁵I-HSA) was infused into both lateral ventricles. The flow marker transfers to the plasma via the arachnoid villi at a flow rate D_AV. Drainage also occurs to the cervical lymphatics at a flow rate D_CT. The cervical lymphatics (tracer concentration = C_CT) receive fluid and solutes from the plasma at a rate F_CT, while lymph flows to the plasma at rate L_CT. Flow marker concentrations within the plasma are designated C_P and distribute to the volume of distribution of the marker (V_P). A second label of the same flow marker injected into the CSF (¹³¹I-HSA) is infused into the plasma space to estimate the flow from the plasma into cervical lymph (F_CT). The rate of transfer of a marker protein from the plasma is taken from the observed plasma disappearance rate of ¹³¹I-HSA and is designatedK_exp. Plasma concentrations of ¹³¹I-HSA were divided by the initial concentration at time 0, andK_expwas calculated from the slope of the logarithmically transformed data. It is assumed that neither¹²⁵I-HSA nor¹³¹I-HSA crosses the blood-brain barrier from the plasma into the CSF. The compartmental concentrations of the tracer have been designated by the superscripts 125 and 131 as appropriate.

A mass balance around the cervical lymph compartment assumes that the mass of tracer into the compartment equals the mass flowing out. All concentrations are a function of time (t). The mass balance yields

Analysis of data.

Values are means ± SE. P < 0.05 was considered statistically significant. The mean changes in various parameters as a function of time were analyzed through repeated-measures ANOVA. To test the consistency of the expected direction of change in increased CSF clearance through cervical lymphatic vessels and arachnoid villi as a function of elevated ICP, the test of binomial distribution was used (20).

RESULTS

Distribution of tracers.

Figure 1 illustrates the distribution of the tracers in the CSF, plasma, and lymph compartments insheep 7. After the exchange of autologous CSF with artificial CSF containing¹²⁵I-HSA, tracer concentrations in the cisterna magna remained relatively stable over the 6-h duration of the experiment. No significant changes in CSF tracer concentrations were observed over this period (Fig. 2). The plasma tracer ¹³¹I-HSA was not detected in CSF. Both radioactive isotopes were detected in cervical lymph and blood. The highest concentration of intraventricularly administered ¹²⁵I-HSA outside the CNS was always observed in cervical lymph. The plasma concentration of¹³¹I-HSA injected intravenously declined exponentially, appeared in cervical lymph, but maintained its highest concentration in plasma. The background physiological data for all animals are illustrated in Table 1.

Cervical lymph flow rates.

Average cervical lymph flow rates (i.e., lymph flow from all tissues, not just lymph originating as CSF) tended to increase with elevation of ICP (Fig. 1). At 10, 20, and 30 cmH₂O ICP, cervical flow rates averaged 3.1 ± 0.5, 4.8 ± 0.6, and 6.9 ± 1.5 ml/h, respectively. However, over the course of the 3-h measurement at each ICP, the lymph flows remained relatively stable, indicating that a steady state had been achieved for each 3-h monitoring period (Fig. 3).

Estimates of volumetric CSF drainage via arachnoid villi and extracranial lymphatic vessels before and after elevations of ICP.

Tables 2 and 3illustrate the calculated CSF drainage rates through arachnoid villi and cervical lymphatics, respectively, in each animal at initial and raised pressures. Two values have been illustrated for each pressure:1) estimates of CSF drainage rates based on raw data uncorrected for filtration errors and2) values derived from the mass balance equations that have been corrected for filtration anomalies. Calculated as flow rates (ml/h), the differences in the values estimated by the two methods seem quite small. However, the mass balance correction of the data becomes more significant when CSF clearance volumes are determined over longer periods. Table4 illustrates the total CSF clearance for each animal (sum of arachnoid villi and lymphatic contributions estimated from the mass balance equations).

No CSF drainage by arachnoid villi or lymphatic pathways was measured when ICP was set at 0 cmH₂O. At an initial ICP of 10 cmH₂O, arachnoid villi drainage ranged from 0.88 to 1.52 ml/h (derived from mass balance equations), whereas the removal of CSF via cannulated cervical lymphatics varied from 0.10 to 0.56 ml/h. Total CSF drainage was 1.5 ± 0.2 ml/h. With elevation of ICP to 20 cmH₂O, arachnoid villi drainage range increased (from 2.30 to 6.44 ml/h), as did lymphatic drainage (from 0.12 to 3.18 ml/h), to yield a total CSF drainage rate of 5.6 ± 2.1 ml/h. With 20 cmH₂O as the starting ICP, arachnoid villi drainage ranged from 0.48 to 5.62 ml/h, lymphatic removal of CSF varied from 0.02 to 5.82 ml/h, and total CSF drainage was 4.4 ± 1.9 ml/h. Raising ICP to 30 cmH₂O increased total CSF drainage to 12.6 ± 5.0 ml/h, where arachnoid villi and lymphatic pathways contributed 0.68–12.08 and 0.04–18.04 ml/h, respectively. CSF clearance data at 10, 20, and 30 cmH₂O are summarized in Fig.4. On average, a 10-cmH₂O increase in ICP elevated arachnoid villi and lymphatic clearance 2.7- and 3.9-fold, respectively. Extrapolation of the curves to thex-intercept gives 8.1 cmH₂O for arachnoid villi and 9.9 cmH₂O for cervical lymphatics. These pressures likely represent the opening pressures that initiate CSF clearance through the two pathways.

Considerable variability between animals was observed in the rates of CSF drainage into cervical lymphatics. For example, in the animals in which CSF pressures were raised from 20 to 30 cmH₂O, lymphatic CSF clearance increased from 0.02 to 0.04 ml/h in sheep 10 and from 5.82 to 18.04 ml/h insheep 7. This variability precluded the usual statistical comparisons. Nonetheless, it is important to note that increments of ICP increased CSF drainage by the arachnoid villi and lymphatic pathways in every animal tested. The probability that all eight animals demonstrated a consistent increase by chance alone is quite small and is therefore unlikely. If sheep 1 and 2, in which the initial pressures were below the estimated pressures that initiated clearance, are excluded, the hypothesis that increased ICP produces a corresponding increase in arachnoid villi and lymphatic drainage is therefore significant according to the binomial distribution (P = 0.004,n = 8).

DISCUSSION

The demonstration that extracranial lymphatic pathways remove about one-half of the draining CSF in normal conscious sheep leaves little doubt that lymphatics contribute to the regulation of extracellular fluid in the CNS of sheep (2, 3). What remains unclear is the response of the lymphatic pathways to conditions of raised ICP. By diverting known lymphatic pathways from emptying into the blood and monitoring the appearance of a CSF tracer in plasma and lymph compartments, we were able to separate the arachnoid villi and lymphatic contributions to total CSF clearance. Combining this approach with ventriculocisternal perfusion allowed assessment of CSF drainage through the two pathways at several levels of ICP. The major conclusion from this study is that elevations of ICP increase CSF clearance into cervical lymphatic vessels. However, several key parameters incorporated into the experimental design require some comment before the implications of this result are discussed.

Physiological requirements for the use of the mathematical model.

The suitability of HSA as the CSF tracer and the derivation of the mass balance equations used in this study have been described in detail previously (2). In deriving the equations, we assumed that the volumetric flow rates defined by the letters L, D, and F remained constant for the duration of the experiment. Although the observed cervical lymphatic flow rates tended to be greater at the higher of the two pressures tested in each animal, the flow rates observed at 10, 20, and 30 cmH₂O were relatively stable (Fig. 3). Therefore, a steady-state condition was achieved for all the monitoring periods, and as a consequence, we believed that this requirement had been met. The slope of the plasma disappearance curve of intravenously injected ¹³¹I-HSA was used to calculate a coefficient of elimination for labeled HSA and to permit correction for the plasma tracer and accompanying volume that refiltered into the lymphatics. The disappearance of tracer followed a typical exponential decline. The amount of tracer entering the lymphatic compartment from the plasma would be defined by the filtration coefficient for each tissue compartment. Changes in ICP over the range of pressures used in this study would not be expected to alter the volumetric transfer of¹³¹I-HSA (F).

Experimental factors that could adversely affect CSF drainage.

In previous reports, we estimated the relative roles of arachnoid villi and extracranial lymphatics in CSF clearance in conscious sheep (2, 3). To simplify the study under consideration here, we chose to use a ventilated-anesthetized preparation. This approach resulted in CSF clearance parameters that were reduced compared with our earlier study. Contractions of lymphatics provide a major source of the energy required to transport lymph from its collection at the interstitial level to delivery into the plasma. Anesthetic agents are known to depress active lymphatic contractility (19). In addition, the motion of the head and neck, which could potentially produce passive compression of lymphatic vessel segments, would be abolished in our experimental preparation. Cervical lymph flow rates at all three pressures in the present study were below previously documented values of 9.1 ± 3.3 ml/h in conscious sheep (2). Higher cervical flow rates would be expected to have directly increased lymphatic drainage values.

The ventilated-anesthetized preparation would also have an impact on CSF drainage through arachnoid villi. CSF absorption across the arachnoid villi is a pressure-driven process that does not begin until CSF pressure is greater than sagittal sinus pressure (P_ss). Any increases in P_ss would be expected to decrease CSF absorption by diminishing the hydrostatic pressure gradient. In our study, CVP was measured continuously at the level of the superior vena cava (7.0 ± 1.0 cmH₂O,n = 10). As expected, the CVP closely paralleled end-expiratory pressure during mechanical ventilation and was assumed to estimate closely P_ss values (15). With experimental CSF pressure set at 10 cmH₂O and P_ss approximating 7 cmH₂O, the CSF pressure-blood hydrostatic pressure gradient was quite small, ∼3 cmH₂O. This driving force likely produced the low total CSF drainage rates observed under these conditions (sum of arachnoid villi and lymphatic = 1.5 ± 0.2 ml/h). Elevations of ICP to 20 and 30 cmH₂O increased the CSF pressure-P_ss gradients and yielded increased total CSF drainage rates at each increment (4.8 ± 1.3 and 12.6 ± 5.0 ml/h, respectively). Using a similar model in resting, conscious sheep, we reported earlier an average total CSF clearance in resting animals of ∼4.0 ml/h (2). In the anesthetized-ventilated group of animals studied in the present report, we had to use an ICP of ∼20 cmH₂O to achieve CSF clearances of similar magnitude.

CVP may also have an important role in determining the clearance of CSF by extracranial lymphatics. Unlike the case with arachnoid villi, which were intact with regard to their drainage target (veins), determination of a lymphatic drainage rate required the direct cannulation of multiple lymphatic vessels in the neck and collection of lymph exterior to the animal. Therefore, the normal outflow pressure sensed by these vessels would be altered. To account for this, we assumed that the outflow pressure would be equivalent to the CVP. The tip of the outflow end of the catheter was raised, such that the total resistance experienced by the cannulated vessels matched CVP. Therefore, arachnoid villi and extracranial lymphatics would be provided with the appropriate pressure gradient for CSF clearance.

Effects of ICP changes on CSF drainage into extracranial cervical lymphatics.

In every animal subjected to an increase in ICP, an increase in lymphatic drainage followed (n = 8). The most striking example was sheep 7. At 20 cmH₂O, total CSF drainage was 11.44 ml/h, with arachnoid villi and lymphatic drainage contributing in approximately equal amounts (5.62 and 5.82 ml/h, respectively). With an increase to 30 cmH₂O, arachnoid villi and lymphatic drainage increased to 12.08 and 18.04 ml/h, respectively, to yield a total CSF drainage of 30.12 ml/h. This example illustrates the magnitude of lymphatic drainage that is possible. However, considerable variability existed between animals in terms of the magnitude of clearance through arachnoid villi and lymphatics. In some sheep, CSF clearance into cervical lymph was small but, as noted earlier, always increased when ICP was elevated. This pattern of individual variability appeared in our earlier studies of CSF drainage under normal pressures in conscious sheep (2-4). This suggests that some adults are more dependent on the lymphatic drainage pathways for CSF removal, whereas others rely primarily on the arachnoid villi. Pathological consequences of a dependence on one pathway over the other remain unknown.

From the data illustrated in Figs. 1 and 3 it is also apparent that total cervical lymph flow (i.e., flow from all tissues, not just from the CSF compartment) tended to increase as ICP were elevated. This is to be expected if a portion of cervical lymph has its origins as CSF. However, in a given experiment we did not always observe an increase in the directly measured cervical lymph flow rate. This was no doubt due to the fact that even though the estimated CSF clearance into cervical lymph increased in all sheep, in six of eight animals the increase was considerably <1 ml/h. Therefore, as a percentage of the total lymph formed in the drainage basin of the cervical lymphatic vessels, in some cases the contribution from the CSF compartment was minor and the changes induced with elevations in ICP were likely lost within the normal variation in lymph flow. Incremental ICP changes of >10 cmH₂O would probably elicit greater changes in directly measured cervical lymph flow rates.

Bradbury and Westrop (6) cannulated a single cervical lymphatic in rabbits and concluded that absolute lymphatic recovery of an intraventricularly administered albumin tracer increased but that the fractional recovery compared with blood decreased. An increase in total cervical lymph flow has been observed after elevations of ICP in cats (16) and dogs (10). In addition, the increased distribution of radiolabeled albumin in postmortem peripheral tissues of rabbits with raised ICP compared with normal-pressure controls suggests that more CSF tracer was being distributed outside the CNS under conditions of elevated CSF pressures, where it would be accessible to lymphatic absorption (18). Each of these studies provides indirect evidence to support the importance of lymphatics in the volumetric removal of CSF. However, cervical lymphatics collect flow from a wide variety of tissues and not just fluid originating as CSF. Furthermore, tracer recovery data can be misleading because of the “cross-contamination” of tracer. Our model permits the separate calculation of CSF drainage via arachnoid villi and lymphatic pathways from tracer recovery data and provides the first direct evidence that CSF transport into extracranial lymphatics is enhanced by incremental changes in ICP.

One must be cautious in commenting on the magnitude of the ICP-induced changes in CSF clearance because of the large variation between animals. With this in mind, the data in Fig. 4 suggest that for each 1.0-cmH₂O elevation in ICP, arachnoid villi drainage increased ∼0.36 ml/h and lymphatic clearance increased ∼0.20 ml/h. On average, an ICP increase of 10 cmH₂O elevated arachnoid villi and lymphatic CSF clearance 2.7- and 3.9-fold, respectively. It is of interest to note that by plotting the average values for arachnoid villi drainage at each pressure, the extrapolated line at thex-intercept approximates CVP and therefore P_ss (8.1 cmH₂O; Fig. 4). CSF pressures greater than this value would be expected to result in CSF drainage that increases linearly with increases in ICP, a result that was suggested from the data in Fig. 4. Similarly, the extrapolatedx-intercept (9.9 cmH₂O) for cervical flow may represent the average opening pressure to facilitate cervical clearance. This suggests that the pressure gradients responsible for transport of CSF via arachnoid villi and extracranial lymphatic pathways may be similar, although additional experiments at numerous ICP levels are required to address this issue appropriately.

It is difficult to interpret the concept of a break point pressure that initiates CSF transport to cervical lymphatics. In rats, CSF seems to pass directly through subarachnoid channels into the lymphatics of the nasal submucosa (14). A CSF pressure gradient of ∼10 cmH₂O may drive CSF through the cribriform plate into the nasal submucosa and, ultimately, into the initial cervical lymphatics. Alternatively, CSF may not transport directly to cervical lymph but, rather, may mix in the fluid volume of the nasal submucosa. Unfortunately, the exact anatomic relationships between CSF and cervical lymph in sheep are unknown. For the mathematical analysis of the data, we have used a relatively simple three-compartment model (CSF, plasma, and extracranial lymph). It was assumed that fluid transfer occurred directly from one compartment to another. However, this may not be the case, and CSF may first empty into the nasal submucosal interstitial space. The mechanisms that control the uptake of interstitial fluid by lymphatics are still being debated. In some tissue compartments, interstitial pressures are subatmospheric, and a pressure gradient between the interstitium and lymphatic appears to be created partly by contractions of the initial lymphatic (12, 13). During the diastolic phase of contractions, a suction force may be generated to draw tissue fluid into the vessel. A more detailed analysis of the CSF pressure-clearance relationships in conscious sheep at multiple pressures may permit more accurate determination of the opening pressures for both pathways. Additionally, the development of a more complex four-compartment model may facilitate our understanding of this process.

In this study a determination of the relative roles of arachnoid villi and cervical lymphatics in the quantitative clearance of CSF was not an important objective. Reports to this end have been published previously (2-4). In this study, arachnoid villi appeared to be responsible for the majority of CSF drainage at all pressures investigated. However, lymphatic drainage was clearly underestimated. Small lymphatic vessels that were difficult to cannulate were ligated and not monitored for the appearance of CSF tracer, and unidentified vessels that contained tracer would have emptied into the blood, adding to the arachnoid villi rates. Additionally, increased CSF pressures would have distributed tracer along the spinal canal (6), to be drained ultimately by the thoracic duct (4). The thoracic duct was left undisturbed and potentially could have delivered tracer to the blood. In our previous reports, we estimated the contribution of uncannulated cervical ducts and determined conservatively that lymphatics were responsible for about one-half of all CSF absorbed from the cranial vault. Rather than revisit this issue in this report, we focused our attention on testing the hypothesis that elevations in ICP resulted in increases in lymphatic CSF clearance.

Many issues related to the relationship between ICP and CSF clearance into extracranial lymphatics remain to be elucidated. The proportional distribution of CSF drainage into arachnoid villi and lymphatic vessels over a wide range of ICP levels, including levels of pressure associated with intracranial hypertension and pathology, needs to be investigated. Other investigators have speculated that the potential resistance offered by the cribriform plate as CSF moves from the subarachnoid space into the nasal submucosa may limit substantially the transport of CSF into cervical lymphatics (6). Additionally, at a high ICP, some CSF may be shunted into the spinal canal and may be taken up by lymphatics that connect with lumbar and intercostal nodes. In earlier studies of sheep we demonstrated high levels of labeled HSA in these lymph nodes after the injection of the tracer into spinal CSF (4).

Perspectives

Current understanding of the factors that regulate ICP can be summarized by the following expression: ICP = I_f × R_out + P_ss, where I_f is the CSF formation rate and R_out is the resistance to CSF outflow (17). Because CSF formation is known to remain constant throughout a wide ICP range (8, 11), the R_out and P_ss terms assume the greatest importance. However, as originally conceived, this expression assumes that all CSF drainage occurs via the arachnoid villi, an assumption that ignores the important contribution of extracranial lymphatics in this process.

First, the R_out term is meant to represent the total resistance imparted by the anatomy of all CSF outflow pathways, but it has been generally assumed that resistance is determined by the arachnoid villi, and investigation of perceived anomalies in CSF drainage has quite naturally focused primarily on these structures. It now appears likely that a lymphatic component figures prominently in the physiological parameters defined by the R_out designation. Second, P_ss provides not only a potential impediment to CSF transport through arachnoid villi, but the equivalent pressure in the veins located in the base of the neck also provide an outflow pressure into which cervical lymphatics are forced to pump. Clearly, we have to rethink ICP regulation and begin to test hypotheses that integrate ICP with CSF drainage into lymphatics.