Body position and eerebrospinal fluid pressure Part 2' Clinical studies on orthostatic pressure and the hydrostatic indifferent point BJORN MAGNAES, M.D. Department of Neurosurgery, Rikshospitalet, Oslo University Hospital, Oslo, Norway v" Lumbar cerebrospinal fluid (CSF) pressure was recorded in 116 adult neurosurgical patients in the lateral and sitting positions. The level of zero CSF pressure while in the sitting position (ZPS) and hydrostatic indifferent point (HIP) for lateral and sitting positions were determined and referred to the craniospinal axis. In control patients ZPS was located mainly at the upper cervical region, and showed nearly the same variation and frequency distribution as CSF pressure in the lateral position when efforts were made to reduce sources of error and there was no orthostatic change in CSF filling pressure. Under these circumstances ZPS may be used as a variable comparable from one subject to another. In control patients the HIP was located between C-6 and T-5. In 25 hydrocephalic patients, shunting resulted in a mean caudal shift of ZPS of 244 mm, and a mean pressure fall of 126 mm H20 in the lateral position. This difference was due to a caudal shift of HIP on shunting. A caudally located ZPS was found in patients with complete cervical subarachnoid block. Prevention and treatment of CSF leakage cranial to HIP is discussed. KEY WORDS 9 postural changes 9 cerebrospinal fluid pressure 9 orthostatic pressure 9 hydrostatic indifferent point T HE recognition of complications due to lowered cerebrospinal fluid (CSF) pressure secondary to CSF shunting has led to an increasing interest in recording the ventricular CSF pressure in the erect body position? '6 Recording of lumbar CSF pressure in the sitting position has not been of great value in clinical practice. The reason for this may be 1) problems in reproducing identical and stable conditions for the pressure recording, 2) difficulty in defining a valid zero reference level, and 3) a varying hydrostatic pressure component due to differences in body length. Yet, the level of zero CSF pressure with the patient in the sitting position (ZPS) has been of some interest? When the body is tilted from the lateral to the sitting position, the pressure in the lower part of the body rises while the pressure in the upper part falls. There must obviously exist a transition zone where the fluid pressure is equal in the lateral and sitting positions. This so-called hydrostatic indifferent point (HIP) has been determined for the venous and 698 J. Neurosurg. / Volume 44 / June, 1976
Body position and CSF statics arterial system and has been stated to represent a useful reference point of postural pressure changes? There seems to be no report on the HIP in the CSF space. This paper deals with the ZPS and the HIP for lateral and sitting positions determined by lumbar CSF pressure recording. Emphasis was placed on evaluating these variables for clinical neurosurgery. Clinical Material and Methods Clinical Material The lumbar CSF pressure was recorded in 116 adult neurosurgical patients in the lateral and sitting positions. Group 1: Control. The control group consisted of 72 patients with neck and arm pains due to cervical spondylosis. A spinal subarachnoid block test was performed on these patients. All patients had normal CSF protein and no hindrance to CSF flow on jugular compression. In these patients, the distance from the occipital protuberance to C-7 was also measured to compare the variation in ZPS with the variation in CSF pressure in the lateral position. Group 2." Hydrocephalus. This group comprised 25 patients with hydrocephalus of whom seven had noncommunicating hydrocephalus. The CSF pressure was recorded before and after ventriculoatrial shunting with a Pudenz medium pressure system. These were selected patients in whom a functioning shunt could be determined fairly accurately by their clinical improvement. The shunts were tested before insertion by measuring opening and closing pressure. Three of these patients had a continuous intraarterial blood pressure recording. Group 3." CSF Leakage. The CSF pressure was recorded during a period of CSF leakage and at a later stage with no leakage in four patients with CSF rhinorrhea and two patients with CSF leakage after neck surgery. Group 4." Skull Defect. In five patients with large, flaccid skull defects the CSF pressure was recorded before and after cranioplasty. Group 5." Cervical Block. In eight patients with complete subarachnoid block due to cervical spondylosis the CSF pressure was recorded before and after decompressive laminectomy resulting in a normal square wave response on jugular compression. Group 6." One-Hour Recording. In 10 control patients and five hydrocephalic patients who were judged to have a high degree of mental and cardiovascular stability, the pressure recording in the sitting position was extended to a period of 1 hour. Pressure Recording and Tilting Procedure The lumbar CSF pressure and arterial blood pressure were recorded and the tilting performed as described? In patients with skull defects lying in the lateral position, the CSF pressure was given as the mean of the pressure recorded in the right and left lateral position. During the 1- hour pressure recording in the sitting position, the Group 6 patients were slightly supported and constantly reminded to keep erect and relaxed, and to keep their eyes fixed on a target placed level with their heads. Determination of ZPS and HIP The CSF pressure in the sitting position was read in millimeters of water, measured vertically from the transducer along the body axis and referred to the spinous process of a vertebra or to the head. The spinous process was identified by palpation. This reference point was the ZPS, or the level of zero CSF pressure in the sitting position. At this point the CSF pressure is equal to atmospheric pressure. Cranial to the ZPS the CSF pressure is negative. From the ZPS the CSF pressure recorded in the lateral position was measured caudally to find the HIP as shown in Fig. 1. At this point the CSF pressure would be equal in the lateral and sitting positions, and this is the HIP for the lateral and sitting positions. Pressure Fall on CSF Shunting The pressure fall in the lateral position was recorded in the usual way with the mid-spine as zero reference level. The pressure fall with the patient in the sitting position was recorded as the distance between the ZPS before and after shunting. Results Lateral to Sitting Position When changing to the sitting position, 112 patients had either a transient or no change in the filling pressure, while four hydrocephalic J. Neurosurg. / Volume 44 / June, 1976 699
B. Magnaes FIG. 1. Drawing shows that the distance between ZPS and HIP corresponds to the CSF pressure in the lateral position (mm H20). ZPS = Level of zero CSF pressure while in the sitting position. HIP = Hydrostatic indifferent point for lateral and sitting positions. patients had a stationary increased filling pressure in the sitting position? In these four patients the primary pressure rise was used when determining the ZPS and HIP. In 108 patients the pressure level at 5 minutes after changing was remarkably stable and was used for determining the ZPS and HIP (Fig. 2 left). Three hydrocephalic patients and one control patient felt faint in the sitting position and had a fall in CSF pressure. The tracing from one of the hydrocephalic patients who also had an intraarterial blood pressure recording running is shown in Fig. 2 right. In these four patients the first stable part of the tracing was used in determining the ZPS and HIP. ZPS and HIP Group 1: Control. The CSF pressure in the sitting position ranged from 320 to 630 mm H~O, with a mean pressure of 490 mm H20. This wide range of CSF pressure was markedly narrowed when recorded at the ZPS, which was usually located at the level of the cervical spine with the corresponding HIP's at the level of the upper thoracic spine (Fig. 3). The ZPS and the CSF pressure FIG. 2. Left: Typical stable CSF pressure level while in the sitting position. Right: Orthostatic intolerance. Fainting symptoms and unstable CSF pressure level while in the sitting position, concomitant with a fall in arterial blood pressure (BP). 700 J. Neurosurg. / Volume 44 /June, 1976
Body position and CSI ~" statics recorded in the lateral position are compared in Fig. 4. All except four patients had the ZPS located by reference to the occipital protuberance and to C-7, a mean distance of 140 mm in the 72 control patients. The CSF pressure in patients in the lateral position varied within 130 mm H20, ranging from 50 to 180 mm H20. Group 2." Hydrocephalus. In these patients the ZPS before shunting was at or above the level of the cervical spine, while after shunting, between the levels of T-2 and T-9. Shunt- FIG. 3. Frequency distribution of ZPS and HIP referred to craniospinal axis in 72 control patients. The star (,) = the occipital protuberance. Fie. 5. Measurements of caudal shift of ZPS and HIP in 25 hydrocephalic patients on CSF shunting. The star (,)= the occipital protuberance. FIG. 4. Measurements in the 72 control patients. Upper: Variation in ZPS and CSF pressure in the lateral position. Distance from occipital protuberance (OP) to C-7 was 140 mm. Lower. Frequency distribution of CSF pressure in the lateral position. FIG. 6. Graph showing CSF shunting resulting in a larger pressure fall in the sitting than in the lateral position. The difference was due to a caudal shift of HIP on shunting. Broken line indicates line of identity (equal pressure fall in lateral and sitting position). J. Neurosurg. / Volume 44 / June, 1976 701
B. Magnaes FIG. 7. Measurements in patients with CSF leakage indicate that the ZPS and HIP shifted caudally. The star (*) = the occipital protuberance. FIG. 8. Measurements in patients with large, flaccid skull defects show that the ZPS and HIP shifted cranially. The star (,) = the occipital protuberance. ing also resulted in a marked caudal shift of the HIP to below the HIP's in the control group (Fig. 5). The pressure fall on shunting was larger when measured in the sitting position than in the lateral position. The mean pressure fall in the lateral position was 126 mm H20 compared with 244 mm H20 in the sitting position (Fig. 6). Group 3: CSF Leakage. During periods of verified CSF leakage the ZPS and HIP were found caudally displaced compared with the levels at a later stage with no leakage. The ZPS and HIP were then within the levels found in the control group (Fig. 7). Group 4: Skull Defect. In the five patients with skull defects, the ZPS and HIP were cranially shifted, while after cranioplasty they were located within the levels found in the control group (Fig. 8). Group 5: Cervical Block. In the patients with complete subarachnoid block, the ZPS and HIP were caudally shifted compared with the control group. After decompressive laminectomy the ZPS and HIp were located within the levels in the control group (Fig. 9). Group 6: One-Hour Recording. Control patients had a slight pressure rise ranging from 20 to 60 mm H~O during the l-hour pressure recording in the sitting position. This pressure rise always leveled out before the recording was ended. When changing back to the lateral position, the pressure fell to less than the pre-tilt level, which, however, was reached within 5 minutes or less. Hydrocephalic patients had pressure changes before the shunting similar to those in the control group, with a pressure rise ranging from 30 to 60 mm H20. After shunt- FIG. 9. Measurements in patients with complete cervical subarachnoid block show that the ZPS and HIP were located caudally. Before laminectomy these patients had a double set of ZPS and HIP, and by lumbar CSF pressure recording only the lower set was determined. The star (,) = the occipital protuberance. 702 J. Neurosurg. / Volume 44 / June, 1976
Body position and CSt: statics FIG. 10. Tracings of lumbar CSF pressure recorded for 1 hour with patients in the sitting position. Upper Tracing." Typical finding in control patients and hydrocephalic patients before shunting: a CSF pressure rise of 50 mm H20. Lower Tracing. Remarkably stable CSF pressure level after shunting in hydrocephalic patients. ing the pressure level was remarkably stable in all patients with no or only a few millimeters rise or fall in pressure. When changing back to the lateral position, pressure fell to the pre-tilt level in two patients. In three patients pressure fell to below the pre-tilt level, which was then reached within 15 minutes or less. A typical pressure recording before and after shunting is shown in Fig. 10. None of these 15 selected patients felt faint or complained of any discomfort during the pressure recording. Discussion Comparison of Lumbar CSF Pressures When comparing the lumbar CSF pressure in the lateral and sitting positions in the same patient, attention must be paid to measurement errors and orthostatic changes in filling pressure. When comparing the lumbar CSF pressure in the sitting position between patients, attention must also be paid to the individual difference in the hydrostatic pressure component. Measurement Errors Cerebrospinal Fluid Leakage. The lumbar CSF pressure in the sitting position is about three times the pressure in the lateral position, thus increasing the risk of CSF leakage during the measurement. It is therefore important to obtain a proper placement of the spinal needle at the first attempt. A No. 19 needle was found to be the thinnest needle that was practical to use; it had great maneuverability and gave a good tactile identification of the dura and arachnoidea. Correct placement of the needle at the first attempt was best achieved by performing the lumbar puncture under local anesthesia with the patient in a sitting position. Changing Body Position. To standardize orthostatic CSF pressure recording the posture must be fairly stable and reproducible. In preliminary studies we tried to make the patient keep his head up against a bar; however, the most stable and reproducible sitting position was obtained when the patient was instructed and assisted in assuming an erect but relaxed sitting position with his eyes fixed on a target on the wall. Orthostatic Changes in CSF Filling Pressure Cerebral Vasodynamics. In patients with increased intracranial pressure and a diseased brain, rapid tilting to the sitting position may result in a stationary increased CSF filling pressure? These pressure changes were usually associated with headache and discomfort, probably reflecting unstable or impaired cerebral vasoregulation. Systemic Blood Pressure. A fall in the CSF filling pressure was found in all those patients who felt dizzy and faint in the sitting position. These symptoms called orthostatic intolerance, are caused by failing arterial blood pressure (BP) leading to cerebral ischemia. 7 Psychic stimulation due to discomfort, anxiety, and pain has been found to increase orthostatic intolerance2 Consequently, it is important to keep the patient from suffering pain by performing lumbar puncture under local anesthesia, and to have a confident attitude toward the patient. A special kind of orthostatic intolerance was found in patients experiencing craniospinal block and imminent herniation in the sitting position. ~ The rise in intracranial CSF filling pressure while in the sitting position, whether transient or stationary, is probably an important factor in this development. Long-Term Changes. The long-term change in CSF filling pressure while in the sitting position was moderate and for practical purposes did not influence the 5-minute pressure J. Neurosurg. / Volume 44 / June, 1976 703
recording. After tilting back to the lateral position, the pre-tilt pressure level was reached within 5 minutes and, after shunting, within 15 minutes. Having the patient in bed for at least 30 minutes before lumbar puncture and establishing a 5-minute stable pressure level in the lateral position before tilting should thus be considered a sufficient period of pressure stabilization when recording CSF pressure in daily practice. Difference in Hydrostatic Pressure Component When recording the lumbar CSF pressure with the patient in the lateral position, the hydrostatic pressure component is related to the breadth of the head. The individual difference is small, and negligible for practical purposes. Thus the pressure recorded is mainly a measure of the degree of filling of the CSF space and, therefore, comparable from one patient to another. When recording the lumbar CSF pressure with the patient in the sitting position, the hydrostatic pressure component, which is related to the length of the body, shows considerable individual variation and must be taken into account. For this the length of the body is used as the unit of measurement by determining the ZPS, which eliminates, for practical purposes, the individual difference in hydrostatic pressure component. This seemed to be verified by the fact that the ZPS in control patients showed nearly the same frequency distribution and variation as the CSF pressure in the lateral position (Figs. 3 and 4). Hydrostatic Indifferent Point The HIP in a fluid space must always be defined for certain positions. Model studies 1'~ have shown the position of the HIP to depend on the relative "give" of the ends of a tube. If the distensibility is equal around a horizontal line, the HIP is located in the middle of the tube and does not shift when the filling pressure is altered. This is the situation when measuring the lumbar CSF pressure in the lateral position. The HIP for right and left lateral positions is then located on the sagittal midline which is also the zero reference level. When the distensibility differs, the HIP moves toward the more distensible side, and is also shifted when the filling pressure changes. This is the situation when measuring B. Magnaes the lumbar CSF pressure in the lateral position in patients with asymmetrical skull defects. It is also the situation when measuring the lumbar CSF pressure with the patient in the sitting position and determining the HIP for the lateral and sitting positions. In accord with this fact is the finding of a caudally-shifted HIP after shunting and a cranially-shifted HIP in patients with large, flaccid skull defects. Clinical Implications Pressure Fall on CSF Shunting. The pressure fall while in the sitting position was larger than the pressure fall while in the lateral position (Fig. 6). The pressure fall recorded in the lateral position was the reduction in CSF filling pressure, while the pressure fall recorded in the sitting position was the reduction in CSF filling pressure plus the caudal shift of the HIP. This "magnification" of the pressure fall on shunting makes the ZPS a useful variable in control of CSF shunt function. If the pressure recording includes a stationary increased filling pressure while in the sitting position before shunting, then the difference between the pressure fall while in the lateral and sitting positions on shunting will be even larger. Spinal Subarachnoid Block. Patients with a complete cervical block had a double set of ZPS and HIP, and by lumbar puncture only the lower set is measured. Thus, the finding of a caudally-located ZPS in a patient with a spinal disease is strongly indicative of a complete subarachnoid block. CSF Leakage Cranial to the HIP. The CSF pressure cranial to the HIP is lowered when changing from the lateral to the sitting position. Thus, if the dura has been opened above the upper thoracic region during surgery, then there is no reason, as far as the hydrostatic pressure component is concerned, to keep the patient strictly in bed postoperatively for the purpose of preventing CSF leakage. The patient may be allowed to sit or walk for periods of 15 minutes, as there is only a slight increase in the filling pressure within this period of time (Fig. 10). If there already is a leakage, then the HIP is probably shifted caudally (Fig. 7), and the need for keeping the patient strictly in bed is probably even less. The possibility of air entering the CSF space must, however, be taken into consideration. 704 J. Neurosurg. / Volume 44 / June, 1976
Body position and CSF statics Acknowledgment I am grateful to electrical engineer Rune Aaslid for valuable discussions and advice on the biophysical aspects of this paper. References 1. Clark JH, Hooker DR, Weed LH: The hydrostatic factor in venous pressure measurements. Am J Physiol 109:166-177, 1934 2. Fox JL, McCullough DC, Green RC: Effect of cerebrospinal fluid shunts on intracranial pressure and on cerebrospinal fluid dynamics. 2. A new technique of pressure measurements, results, and concepts. 3. A concept of hydrocephalus. J Neurol Neurosurg Psychiatry 36:302-312, 1973 3. Gauer OH, Thron HL: Postural changes in the circulation, in Hamilton WF (ed): Handbook of Physiology, Section 2: Circulation. Volume 3. Baltimore: Williams & Wilkins, 1965, pp 2409-2439 4. Langfitt TW: Increased intracranial pressure. Clin Neurosurg 16:436-471, 1969 5. Magnaes B: Body position and cerebrospinal fluid pressure. Part 1: Clinical studies on the effect of rapid postural changes. J. Neurosurg 44:687-697, 1976 6. Portnoy HD, Schulte RR, Fox JL, et al: Antisiphon and reversible occlusion valves for shunting in hydrocephalus and preventing post-shunt subdural hematomas. J Neurosurg 38:729-738, 1973 7. Samnegfird H: Studies on internal carotid artery blood flow in man. Electromagnetic flowmetry after carotid artery surgery. Seand J Thorae Cardiovase Surg 8 (Suppl 13):1-29, 1974 8. Stevens PM: Cardiovascular dynamics during orthostasis and the influence of intravascular instrumentation. Am J Cardiol 17:211-218, 1966 Address reprint requests to: BjCrn Magnaes, M.D., Department of Neurosurgery, Rikshospitalet, Oslo 1, Norway. J. Neurosurg. / Volume 44 / June, 1976 705