The influence of coughing on cerebrospinal fluid pressure in an in vitro syringomyelia model with spinal subarachnoid space stenosis

Abstract

Background: The influence of coughing, on the biomechanical environment in the spinal subarachnoid space (SAS) in the presence of a cerebrospinal fluid flow stenosis, is thought to be an important etiological factor in craniospinal disorders, including syringomyelia (SM), Chiari I malformation, and hydrocephalus. The aim of this study was to investigate SAS and syrinx pressures during simulated coughing using in vitro models and to provide information for the understanding of the craniospinal fluid system dynamics to help develop better computational models.

Methods: Four in vitro models were constructed to be simplified representations of: 1) non-communicating SM with spinal SAS stenosis; 2) non-communicating SM due to spinal SAS stenosis with a distensible spinal column; 3) non-communicating SM post surgical removal of a spinal SAS stenosis; and 4) a spinal SAS stenosis due to spinal trauma. All of the models had a flexible spinal cord. To simulate coughing conditions, an abrupt CSF pressure pulse (~ 5 ms) was imposed at the caudal end of the spinal SAS by a computer-controlled pump. Pressure measurements were obtained at 4 cm intervals along the spinal SAS and syrinx using catheter tip transducers.

Results: Pressure measurements during a simulated cough, showed that removal of the stenosis was a key factor in reducing pressure gradients in the spinal SAS. The presence of a stenosis resulted in a caudocranial pressure drop in the SAS, whereas pressure within the syrinx cavity varied little caudocranially. A stenosis in the SAS caused the syrinx to balloon outward at the rostral end and be compressed at the caudal end. A >90% SAS stenosis did not result in a significant Venturi effect. Increasing compliance of the spinal column reduced forces acting on the spinal cord. The presence of a syrinx in the cord when there was a stenosis in the SAS, reduced pressure forces in the SAS. Longitudinal pressure dissociation acted to suck fluid and tissue caudocranially in the SAS with a stenosis.

Conclusions: Pressures in the spinal SAS during a simulated cough in vitro had similar peak, transmural, and longitudinal pressures to in vivo measurements reported in the literature. The pressure wave velocities and pressure gradients during coughing (longitudinal pressure dissociation and transmural pressure) were impacted by alterations in geometry, compliance, and the presence of a syrinx and/or stenosis.

Figure 1

Schematic diagram of the in vitro models for coughing experiments indicating the location of the stenosis, syrinx, and flow input. Ports for pressure sensors were sited at 4 cm intervals in the subarachnoid space and in the syrinx. The simulated cough pressure pulse was inserted at the input (caudal end). SSE = stenosis and syrinx model, SSED = stenosis and syrinx model with distensible spinal column, SRE = stenosis removed model, and SAE = stenosis alone model.

Figure 2

Photograph of the syrinx and stenosis model with distensible spinal column (SSED). Approximate location of the syrinx, SAS stenosis, spinal cord, and spinal column are indicated. The small vertical tubes connecting to the larger tube are ports used for SAS pressure measurements. Wires extending from the ports are the catheters used for pressure measurement in the SAS (also located in syrinx cavity).

Figure 3

Pressures relative to baseline (P_r(t)) plotted over time for various axial locations for each experimental model. Symbols identify location of pressure sensors. Sampling frequency was 17 kHz. Note: SSED has a different pressure scale to the other plots. Compliance in the spinal column (SSED) resulted in a pressure waveform that exhibited greater damping and less oscillatory response than in the other situations. SSED, SAE, SSE, and SRE as for Figure 1.

Figure 4

Maximum, average, and minimum pressures (P_r(t)) relative to baseline in the SAS and syrinx plotted against location for each experimental model during simulated cough. The cough pressure pulse was applied caudally at 0.0 cm on the x-axis and travelled caudorostrally through the spinal SAS. The light grey rectangle denotes location of the syrinx cavity (when present), the dark grey vertical stripe denotes the location of the SAS stenosis. Note: SSED has a different pressure scale to SSE, SRE, and SAE. SSED, SAE, SSE, and SRE as for Figure 1.

Figure 5

Longitudinal pressure dissociation (LPD) (lumbar - cervical) measured in the SAS for each experimental model during coughing: LPD(t) = P_SAS,_4cm(t) - P_SAS,_32cm(t) plotted against time. A positive deflection in LPD indicates that a pressure gradient is available to suck fluid and/or tissue towards the cranium. Peak LPD sucks fluid and/or tissue toward the cranium in SSED, SSE, and SAE, while in SRE LPD oscillates back and forth around zero. Sampling frequency for each experiment was 17 kHz. SSED, SAE, SSE, and SRE as for Figure 1.

Figure 6

Transmural pressure (TP) measured at various locations along the syrinx in each model (SSED, SSE, and SRE) during coughing: TP(t) = P_{r, syrinx}(t) - P_r,_SAS(t) and plotted against time. Positive pressure trace indicates that pressure in the syrinx is greater than the SAS at adjacent sensor locations. Negative pressure trace indicates that syrinx pressure is less than the SAS. Legend symbols identify pressure sensor location. The presence of a SAS stenosis (SSED and SSE) could cause the syrinx to balloon outward at its rostral end and be compressed at the caudal end. When the stenosis was removed (SRE), TPs were smaller and oscillated about zero. SSED, SAE, SSE, and SRE as for Figure 1.

Figure 7

Spatial versus temporal pressure distribution for each model in the SAS and syrinx. Each black box indicates a pressure measurement obtained from a sensor located at a stenosis. Slope of the white arrows indicates the direction and approximate speed of the pressure wave propagation. The example (bottom right) provides a visual interpretation of the wave speed calculation technique. Note: full coughing pressure data set has been truncated temporally to visualize the foot of pressure wave arrival. SSED, SAE, SSE, and SRE as for Figure 1.