For patients after cardiac arrest, neuroprotection provided by mild hypothermia (32–35°C ) after global brain ischaemia has become clinical standard. Cooling devices based on surface pads, cold saline infusions and intravenacaval catheters provide whole body hypothermia. A more distal intraarterial approach would allow for selective brain cooling, reducing time to target temperature and preventing deleterious effects of systemic hypothermia, though it leads to the need of device miniaturisation.
In particular for ischaemic stroke patients, who are treated with endovascular, catheter-based recanalization devices, the combination with an intra-carotid cooling device seems to be a promising technique, providing a neuroprotective effect in the critical phase of reperfusion.
We propose a coaxial system for combined selective brain hypothermia and mechanical thrombectomy . The cooling catheter presents two lumens for close loop circulation of the coolant, a distal balloon for heat exchange with the surrounding blood, and one lumen for delivery of a recanalization catheter.
The aim of this study is the development of adequate mathematical and numerical models to predict the cooling performance of a balloon catheter placed within the CCA and estimate the influence of construction parameters on heat exchange. For these purposes, four different configurations of a single-balloon catheter, varying in balloon diameter and wall thickness, are investigated. For verification, corresponding prototypes were built and tested invitro in a vasculature model.
In terms of model building, the CCA is simplified by a rigid tube with a diameter of 7 mm and is perfused with a flow rate of 400 ml/min and an inlet temperature of 37 °C.The coolant (water) enters the balloon at a temperature of 10 °C and flows in countercurrent to the blood with a flow rate of 100 ml/min.
Subject of the investigation are four 8F cooling catheter configurations, having a single balloon each with a length of 40 mm, specified in Table 1 (balloon-samples available on the market).
2.1 Mathematical model
Heat flux J [W/m2] is a function of the heat transfer coefficient k [W/m2K] and temperature difference between blood and coolant:
Heat transfer coefficient k results from the heat transfer resistance of the three zones (coolant flow, balloon wall and blood flow) following the equation:
whereas αc and αb [W/m2K] represent the convection heat transfer coefficient at the coolant and blood side respectively, s [m] the wall thickness of the balloon and [W/mK] the thermal conductivity of the balloon material. The left term in the equation above stands for the overall resistance of the heat transfer, while the three right terms for the single resistances of the three zones.
While heat transfer resistance of the balloon wall depends only on material conductivity and wall thickness, the determination of heat transfer at the blood and coolant side is more complex and results in addition to fluid conductivity and thermal capacity from flow pathway, particularly turbulences and stagnation.
The convection heat transfer coefficient is a function of the Nusselt number, which can be interpreted as the dimensionless heat transfer coefficient normalized to the characteristic length xch and the thermal conductivity of the fluids.
For determination of the nusselt number we used the following approximation formula for laminar flow in cylindrical tubes :
Integrating the heat flux over the whole balloon surface and considering the temperature change along the balloon, the whole heat exchange of the balloon as well as the temperature gradient at the blood side along the balloon were calculated.
2.2 Numerical simulation
A numerical calculation of combined fluid flow and heat transfer within the (CFD) software Solid Works Flow Simulation 2011 (Dassault Systčmes SolidWorks Corp., Waltham, Massachusetts) was performed. Therefore a simplified steady-state model, neglecting blood flow pulsatility, was used. The model neglects heat transfer with the surrounding tissue, assuming an adiabatic behaviour of the vessel wall. Rheological properties of the human blood were modelled through the power law model, using the blood fluid parameters specified in the Solid Works database.
For simplification purposes, the numerical model consists only of the catheter tip with balloon. The bulk-average temperature (area and flow-rate weighted average) of both blood and coolant was measured proximal and distal to the region of heat exchange.
2.3 In-vitro testing
The test-setup comprises a blood substitute- (fluid: mixture of 56% glycerine and 44% bi-distilled water) and a coolant-circuit, based on a system of PVC tubes (Guttasyn Kunststoff GmbH, Witten, Germany).
Blood substitute flow was provided by a pulsatile blood pump (Pulsatile Blood Pump for Large Animals and Hemodynamic Studies, Havard Apparatus, Holliston, Massachusetts) with an output phase ratio (systole/diastole) of 30/70 and a pump rate of 60 strokes per minute. To approach the physiologic pressure difference of ~80– 120 mmHg within the system, a pressure tank is integrated downstream the pump. In a heating circulator bath (Haake DC10-B3, Thermo Haake GmbH, Karlsruhe, Germany) the water/glycerine-mixture was warmed up to 37°C.
Coolant is tempered in a cooling bath, whereas temperature at catheter inlet is constantly measured by means of precision fine wire thermocouples (5TC-GG-JI-20-2M, Omega Engineering Inc., Stamford, Connecticut) and pumped through a peristaltic pump (Behrotest PLP 220 with pump head PPH 303, Behr Labor-Technik GmbH, Düsseldorf, Germany). Coolant flow rate was calibrated before the measurement and verified afterwards.
After a preconditioning phase, each catheter sample was tested over a period of 2 minutes. Data of blood substitute temperature, proximal and distal to the balloon (precision fine wire thermocouples, 5TC-GG-JI-20-2M, Omega Engineering Inc., Stamford, Connecticut), pressure (pressure transducer, HPSA-B10CV-AB-020-G, Althen GmbH, Kelkheim, Germany) and flow rate (ultrasonic flowmeter, M-1500-T21-011-009, Malema Sensors, Boca Raton, Florida) were continuously recorded and processed with an in-house programmed software (Labview 2012, National Instruments Corporation, Austin, Texas) with a rate of one measurement per second. Due to the formation of fluid layers of different temperatures, the temperature distal to the balloon was measured approximately ~40 cm distal to the catheter tip, where a mixing of layers is supposed. Temperature values were averaged over the measurement time.
First, the thermal resistance of the three different zones (blood, coolant and balloon wall) is considered. The results from the mathematical model indicate that the cooling performance is limited mostly by the thermal boundary layer of both fluids rather than thermal resistance over the balloon wall (Figure 2).
This is supported by the results of cooling performance (Figure 3) using the three different models, where the temperature decrease of the blood substitute is similar for catheter configuration #3 and #4, which have the same balloon outer diameter but different wall thickness.
As expected, the increase of the balloon outer diameter and consequently surface results in a higher heat exchange.
As depicted in Figure 4, the blood temperature gradient measured in-vitro and normalized to the heat exchange surface increases with balloon diameter, with only configuration #4 not following the trend. Thus, reduction of the cross-sectional area of blood flow seems to have a bigger impact on cooling performance than at the coolant side, regardless of the exchange surface.
Temperature distribution at the blood and water zones resulting from the numerical simulation is depicted in Figure 5 for all four configurations. An increase of the cross-sectional area leads to an enlargement of the thermal boundary layer for both blood and coolant. This is in accord with the thermal resistance within the single fluids depicted in Figure 2.
The focus of the presented study lies on model building and validation with the aim of developing a tool for the dimensioning of a new intra-carotid cooling catheter.
Results show a good agreement between mathematical, numerical and in-vitro models. This makes the theoretical approach a useful designing tool, which allows reduction of sample iterations. The results of the in-vitro tests show the highest heat exchange values compared to both theoretical models. This could be explained by the fact that only for this model blood flow is pulsatile, which possibly results in an increase of heat exchange through the formation of turbulences.
Within in-vitro testing, a maximal temperature decrease along the balloon of 1,4 °C was measured. Supposing a physiological temperature of 37 °C proximal to the balloon, this cooling performance would not allow a target brain temperature of 35°, which was demonstrated to have a neuroprotective effect in previous studies . However, in the models in-vivo recirculation effect of already cooled blood was neglected, which may add to a further effect over the treatment time.
Moreover, one could argue that distal blood temperature doesn’t necessarily correspond to brain temperature; however, heat transfer with surrounding tissues is dominant in smaller vessels, this leading to the assumption that cooling is mostly transferred to the perfused brain tissue.
In spite of several model limitations, like straight tubing with different thermal conductivity than arterial walls and simplified steady-state flow model in the mathematical and numerical models, this study provides important conclusions for the development of a cooling catheter design. The thermal resistance of the balloon wall is small compared to that in the fluid thermal boundary layers, so that the main focus for further optimization should address increase of flow convection, e.g. by surface texturing. The use of balloon array , optionally with decreasing diameter from proximal to distal for partial catheter placement into the internal carotid artery, can also be considered for surface enhancement.
For stroke patients, selective brain hypothermia by means of intra-carotid blood cooling has the potential to significantly reduce the time to target temperature, providing a stronger neuroprotective effect compared to systemic cooling devices. The combination with endovascular recanalization techniques, e.g. with stent-like devices for clot retraction or angioplasty-balloon-catheters for stenosis treatment, would allow a “cold reperfusion”, addressing the most critical phase of ischaemia management.
The work described in this paper is sponsored by the Federal Ministry of Education and Research (BMBF), grant number 13GW0015B.
Cattaneo G, Schumacher M, Wolfertz J, Jost T, Meckel S, Combined Selective Cerebral Hypothermia and Mechanical Artery Recanalization in Acute Ischemic Stroke: In-vitro Study of Cooling Performance, accepted for publication inamerican journal of neuroradiology (july 2015)
About the article
Published Online: 2015-09-12
Published in Print: 2015-09-01
Conflict of interest: Julia Wolfertz and Giorgio Cattaneo are employees of Acandis GmbH & Co.KG (Pforzheim,Germany). Stephan Meckel received from Acandis an honorarium and support for travel as a member of Scientific Advisory Board. Material and Methods: Informed consent: Informed consent has been obtained from all individuals included in this study. Ethical approval: The research related to human use has been complied with all the relevant national regulations, institutional policies and in accordance the tenets of the Helsinki Declaration, and has been approved by the authors’ institutional review board or equivalent committee.