Removal of bacterial endotoxins

Abstract

Methods of cleaning a medical device are provided via exposing the medical device to a compressed CO2-based mixture. The compressed CO2-based mixture includes carbon dioxide, a surfactant, and water in the form of water-in-CO2 microemulsions. In one particular embodiment, the ratio of water-to-surfactant mixed together in the CO2 has a range of about 5-100 molecules of water per molecule of surfactant (e.g., about 5-30 molecules of water per molecule of surfactant).

Description

BACKGROUND OF THE INVENTION

Most common cleaning methods for reusable medical devices and biomaterials rely on manual and automated washing, using brushes to dislodge soil in the presence of water and detergents, or organic solvents. However, the cleaning apparatus may compound the accumulation of residual soil by causing surface abrasion or grooving. Failure to successfully clean the medical device leads to biofilm formation and bacterial colonization, which may harbor bacterial residuals such as endotoxins.

Endotoxin contamination and its effects on biocompatibility have not yet entered into widespread consciousness among biomaterial specialists. Endotoxins, also called lipopolysacharides (LPS), are an integral part of the outer cell membrane of Gram-negative bacteria that are shed upon cell death, growth, and division. When introduced to the blood stream, they elicit an immune response, especially through monocytes and macrophages. These cells release mediators, such as tumor necrosis factor and free radicals, having potent biological activity responsible for adverse effects. Among these are affecting structure and function of organ cells, changing metabolic functions, raising body temperature (pyrogen reactions) triggering the coagulation cascade, modifying hemodynamics, causing septic shock, and in extreme cases multiple organ failure, with a high mortality rate.

Due to their ubiquitous nature, endotoxins are persistent bio-contaminants that deposit and adhere to many materials. Previous studies have revealed that significant levels (15 endotoxin units (EU)/m² of surface area) of adherent endotoxin existed on cleaned, passivated, and gamma-sterilized implant surfaces, especially on those made from titanium (Ti). Their ability to adhere to materials has been related to many factors such as material type, surface properties, and pH. However, affinity for metallic biomaterials such as Ti appears to be primarily a function of surface energy. The surface energy of the endotoxins is thought to be about 30 mJ/m² or less. Hence, for endotoxins to adhere, the biomaterial must exhibit surface energies greater than 30 mJ/m².

Eliminating endotoxins has been a major challenge to the pharmaceutical and medical industry, and is by far the greatest concern in achieving depyrogenation of medical devices. Yet, a generally applicable method for the removal of endotoxins is not available. Since endotoxins are highly heat-stable they are not destroyed by standard autoclaving conditions. However, endotoxins can be destroyed by dry heat at 250° C. for more than 30 min or at 180° C. for more than 3 h. However, there are possible complications associated with dry-heat decontamination. One is the lack of uniformity of temperature within the oven. Hot air has a tendency to stratify and may not uniformly heat a cooler material. Another complication is heat damage and oxidation of biomaterials. To remove endotoxin from metallic particles a cycle of alkali ethanol (0.1 M NaOH in 95% ethanol) at 30° C. followed by 25% nitric acid both for 18-20 h each is recommended. In reusable medical devices, a useful recommendation to minimize endotoxin contamination is to process, package, and promptly sterilize the item in order to limit the time of bacterial contamination and growth. However, conventional sterilization by steam or ethylene oxide does not destroy endotoxin, and does not alter the pyrogenic activity of endotoxic fragments.

Thus, a need still remains for techniques and processes to achieve safe endotoxin levels on medical devices (e.g., ≦20 EU/device according to the US Pharmacopeia-Standard USP27-NF22).

SUMMARY

Objects and advantages of the invention will be set forth in part in the following description, or may be obvious from the description, or may be learned through practice of the invention.

Methods of cleaning a medical device are generally provided via exposing the medical device to a compressed CO₂-based mixture. The compressed CO₂-based mixture includes carbon dioxide, a surfactant, and water in the form of water-in-CO₂ microemulsions. In one particular embodiment, the ratio of water-to-surfactant mixed together in the CO₂ has a range of about 5-100 molecules of water per molecule of surfactant (e.g., about 5-30 molecules of water per molecule of surfactant).

The compressed CO₂-based mixture can have a temperature of about 0° to about 100° C. (20° C. to about 60° C.) and a pressure of at least about 400 psi (e.g., about 400 to about 600 psi or about 800 to about 5000 psi). As such, the compressed CO₂-based mixture can be a liquid or a super critical fluid, depending on the temperature and pressure selected.

According to certain embodiments, the compressed CO₂-based mixture can remove at least 85% of bacterial endotoxin from the medical device, such as at least 95% of bacterial endotoxin from the medical device. In one particular embodiment, the compressed CO₂-based mixture can remove at least 99% of bacterial endotoxin from the medical device.

Other features and aspects of the present invention are discussed in greater detail below.

BRIEF DESCRIPTION OF THE DRAWINGS

A full and enabling disclosure of the present invention, including the best mode thereof to one skilled in the art, is set forth more particularly in the remainder of the specification, which includes reference to the accompanying figures.

FIG. 1 shows a stainless steel plate with Ti disks, according to the Examples.

FIG. 2 shows a schematic of the 1 L pressure vessel apparatus according to the Examples, in which the following components are shown: (1) CO₂ gas cylinder; (2) pump; (3) water/coolant bath; (4) pressure vessel; (5) stainless steel plate attached to the shaft; (6) cooling coil; (7) heating jacket; (8) pressure indicators.

FIG. 3 shows configurations employed within the PEM vessel according to the Examples.

FIG. 4 shows results according to the Examples of processing disks with pure CO₂ and CO₂-based mixtures in the 1 L pressure vessel: (a) percent of endotoxin removed (Mean±SD); and (b) residual endotoxin levels (Mean±SD).

FIG. 5 shows a percent of endotoxin removal (Mean±SD) from Ti disks according to the Examples using liquid CO₂+Ls-54 and water at 27.6 MPa & 2 hr; and 13.8 MPa & 4 hr in the 1 L pressure vessel.

FIG. 6 shows a percent of endotoxin removal (Mean±SD) with liquid CO₂ and mixtures of water and Ls-54 in the PEM system, configuration 3a, according to the Examples.

FIG. 7 shows a percent of endotoxin removal (Mean±SD) from Ti disks using liquid CO₂+Ls-54 and water in the PEM system with mass transfer limitations and flow restrictions, according to the Examples.

FIG. 8 shows a percent of endotoxin removed (Mean±SD) from stainless steel lumens after processing with pure liquid CO₂ and liquid CO₂ microemulsions according to the Examples.

DETAILED DESCRIPTION OF INVENTION

The following description and other modifications and variations to the present invention may be practiced by those of ordinary skill in the art, without departing from the spirit and scope of the present invention. In addition, it should be understood that aspects of the various embodiments may be interchanged both in whole or in part. Furthermore, those of ordinary skill in the art will appreciate that the following description is by way of example only, and is not intended to limit the invention.

Generally speaking, the present invention is directed towards a process that will substantially remove bacterial endotoxins from biomaterials and reusable medical devices. In particular, methods for the removal of bacterial endotoxins (e.g., Escherichia coli) are provided through the use of compressed carbon dioxide (CO₂)-based mixtures.

To enhance the water solubility in CO₂ and make it accessible for dissolving endotoxins, a surfactant can be used to form water-in-CO₂ microemulsions. A microemulsion is a thermodynamically stable dispersion of two immiscible fluids stabilized by surfactants. There are roughly three types of microemulsions; water-in-oil, bicontinuous, and oil-in-water microemulsions. Surfactants typically have very low volatility, and thus interact to a much lesser degree with the substrate. Furthermore, they often dramatically improve the solubility of polar species, well beyond that of simple modifiers.

Microemulsions containing water, surfactant and, CO₂ have been designed to achieve: (1) low interfacial tensions for favorable wetting of small features on substrate; (2) solubilization of residues into micelles in water, water droplets, or CO₂ continuous phase; and (3) prevention of redeposition. The advantages of using compressed CO₂ as the continuous phase over conventional organic solvents for cleaning are that, in addition to being nontoxic and nonflammable, CO₂ has low viscosity and high diffusion coefficient. Moreover, the stability of water-in-CO₂ microemulsions depends on the density of the compressed CO₂. Therefore, the breakdown of the microemulsions can be accomplished simply by controlling the temperature and/or pressure of the system. The temperature range for CO₂-based systems is typically 0° to 100° C., (e.g., about 20° C. to about 60° C.).

A CO₂-based microemulsion would have minimal environmental impact since, at this scale of use, the solvent is environmentally benign. In addition, the CO₂ continuous phase has a very high capacity for lower polarity organic solutes, and these materials can be easily recovered from the solvent. The very high transport rates of the SCF phase (at least an order of magnitude higher than for water) greatly enhance the cleaning rates and are especially attractive for processing of porous or intricate materials. Finally, the polarity of the CO₂ microemulsion can be adjusted either through selection of different types of surfactants or through adjustments in the amount of water that is added to the microemulsion. The ability of the microemulsion to dissolve polar solutes depends solely on the characteristic of the microemulsion droplet. Thus, the design of surfactants compatible with CO₂ is crucial for the formation of stable water-in-CO₂ microemulsions.

In general, the surfactant can be a non-ionic surfactant, such as a fatty molecule (e.g., a fatty alcohol, a fatty acid) or a derivized fatty molecule (e.g., a derivatized fatty alcohol or a derivatized fatty acid). As used herein, a derivatized fatty molecule is a fatty molecule that has been reacted with at least one other compound. For example, the derivatized fatty molecule, in one embodiment, can be alkoxylated to form an alkoxylated fatty molecule. Additionally, for instance, the alkoxylated fatty molecule can be further reacted with a phosphoric compound, such as phosphorous pentoxide, polyphosphoric acid, or the like.

In one embodiment, the surfactant of the present invention can comprise a derivatized fatty alcohol. Fatty alcohols are long chain alcohols typically having the formula ofR—OHwherein R represents a hydrocarbon chain, either saturated or unsaturated. The hydrocarbon chain of the fatty alcohol can be of any length, such as comprising from about 6 to about 26 carbons, for example from about 8 to about 22 carbons. For instance, in one particular embodiment, the hydrocarbon chain can comprise from about 10 carbons to about 14 carbons.

The hydrocarbon chain on the derivatized fatty alcohol surfactant can be either saturated or unsaturated fatty alcohols, including both monounsaturated and polyunsaturated fatty alcohols. A saturated carbon chain means that all the carbon to carbon bonds in the hydrocarbon chain are single bonds, allowing the maximum number of hydrogens to bond to each carbon, thus the chain is “saturated” with hydrogen atoms. Conversely, an unsaturated hydrocarbon chain means that the carbon chain contains at least one carbon to carbon double bond, thereby reducing the number of hydrogens present on the chain. A monounsaturated hydrocarbon chain contains one carbon to carbon double bond, while a polyunsaturated hydrocarbon chain contains at least two carbon to carbon double bonds.

Many fatty alcohols have common names, relating to their corresponding hydrocarbon chain, to describe the alcohol. The hydrocarbon chains can also be described by the number of carbon atoms present in the chain and the number and location of any double bonds present in the chain, represented by n:m^{Δp,p′,p″}, where n is the number of carbons in the hydrocarbon chain, m is the number of carbon to carbon double bonds in the chain, p is the location of the first double bond (if present), p′ is the location of the second double bond (if present), p″ is the location of the third double bond (if present), and so on. Examples of saturated fatty alcohols that can be used as an surfactant include, but are not limited to, lauryl alcohol (12:0), tridecyl alcohol (13:0), myristil alcohol (14:0), pentadecyl alcohol (15:0), cetyl alcohol (16:0, also known as palmityl alcohol), heptadecyl alcohol (17:0), stearyl alcohol (18:0), arachidyl alcohol (20:0), and behenyl alcohol (22:0). Examples of unsaturated fatty alcohols that can be used as an surfactant include, but are not limited to, palmitoleyl alcohol (16:1^Δ9), oleyl alcohol (18:1^Δ9), linoleyl alcohol (18,2^Δ9,12), conjugated linoleyl alcohol (18:2^Δ9,11), linolenyl alcohol (18:3^Δ9,12,15), γ-linolenyl alcohol (18:3^Δ6,9,12), eicosenoyl alcohol (20:1), eicosadienoyl alcohol (20:2^Δ11,14), arachidonyl alcohol (20:4^Δ5,8,11,14), cetoleyl alcohol (22:1^Δ11), and erucyl alcohol (22:1^Δ13).

Derivatives of unsaturated fatty alcohols can also be used as surfactants according to the present disclosure. For example, the hydrocarbon chain of the fatty molecule can comprise a reactive group. For instance, the hydrocarbon chain can comprise an acrylate group.

In one embodiment, the surfactant of the present disclosure can be a derivative of a fatty alcohol. For example, a fatty alcohol as described above can be alkoxylated to form an alkoxylated fatty alcohol, also known as an alcohol alkoxylate. Such as, in one embodiment, the fatty alcohol can be ethoxylated to form an ethoxylated fatty alcohol, also known as an alcohol ethoxylate. For example, the fatty alcohol can be reacted with from 1 mole to about 10 moles of ethylene oxide, such as from about 2 to about 8 moles. The resulting product of the fatty alcohol ethoxylation can generally be represented by the following formula:R—O—(CH₂CH₂O—)_nHwhere R is the carbon chain of the fatty alcohol and n is an integer from 1 to about 10, such as from about 2 to about 8. In one particular embodiment, for example, n can be about 6. Another suitable alkoxylated fatty alcohol can be propoxylated by reacting propylene oxide with the fatty alcohol to form an propoxylated fatty alcohol, also known as an alcohol propoxylate. For example, the fatty alcohol can be reacted with from 1 mole to about 10 moles of propylene oxide, such as from about 2 to about 8 moles.

In another embodiment of the present invention, the surfactant can comprise a derivatized fatty acid. Fatty acids have a similar structure to fatty alcohols described above and can be represented by the following formula: RCOOH where R represents a hydrocarbon chain, either saturated or unsaturated. In this embodiment, the fatty acid surfactants can have the same hydrocarbon chains as described above in reference to fatty alcohols. Also, as described above, fatty acids can be saturated, monounsaturated, or polyunsaturated. In one particular embodiment, the fatty acid can be comprise a conjugated hydrocarbon chain. Many fatty acids have common names, relating to their hydrocarbon chain, that describe the molecule. In fact, most the fatty alcohols listed above, either saturated or unsaturated, have a corresponding fatty acid molecule with a similar common name. Those corresponding fatty acids are included, as well as others, within the scope of this disclosure.

Also, in this embodiment, the fatty acid can be derivatized by alkoxylation as described above in reference to the derivatized fatty alcohol embodiment.

In yet another embodiment, the derivatized fatty molecule, such as a derivatized fatty alcohol or a derivatized fatty acid, can be alkoxylated with a combination of alkylene oxides. For example, the derivatized fatty molecule can include at least one ethylene ester and at least one propylene ester, as represented below:R—O—(CH₂CH₂O—)_n—(CH₂(CH₃)CH₂O—)_m—Hwhere n is about 1 to about 8 and m is about 1 to about 8. One particularly suitable surfactant is available commercially under the trade name Dehypon Ls-54 from Cognis Corporation, now part of BASF, which is believed to be a fatty alcohol (C12-C14) with approximately 5 moles ethylene oxide and approximately 4 moles propylene oxide (i.e., where R includes a fatty alcohol (i.e., R is R—CO—), n is about 5, and in is about 4). The reason for high solubility may be that the Ls-54 surfactant has low molecular weight and has four propylene oxide groups, which have been proven to be CO₂-philic,

Fluorosurfactants are also compatible with CO₂, and can include the class of alkane/fluoroalkane hybrids and perfluoropolypropylene oxide (e.g., fluorinated sodium bis(2-ethylhexyl)sulfosuccinate (AOT) analogue). The fluorinated chains represent low cohesive energy density groups thereby promoting low solubility parameters and low polarizability. Although high CO₂ compatibility can be achieved by fluorinated surfactants, the cost of fluorinated compounds is high and they are toxic. On consideration of the environmental and economical factors, hydrocarbon surfactants, hybrid fluorocarbon-hydrocarbon surfactants, and oxygen-containing surfactants formed by incorporating oxygen into the surfactant tails may be more suitable for use. Another specific example of an oxygen-containing surfactant (other than Ls-54) is octa(ethylene glycol) 2,6,8-trimethyl-4-nonyl ether. Other oxygenated surfactants include nonionic block copolymers composed of oligomers of propylene oxide or butylene oxide with branches on the polymer backbone.

CO₂ has relatively low interfacial tension, liquid-like solvating properties, and gas-like diffusion and viscosity that enable rapid penetration into complex structures for the removal of contaminants. The unique properties of compressed CO₂, coupled with those of a dispersed microemulsion phase, enables dissolution of the endotoxins and subsequently, removal from the contaminated metallic parts. Thus, such a method can out-perform traditional water-based cleaning processes, particularly for complex structures, since it is not be hampered by high surface tensions as occurs with water.

Thus, in one particular embodiment, the compressed CO₂-based mixtures include carbon dioxide, a surfactant, and water in the form of water-in-CO₂ microemulsions. In one particular embodiment, the compressed CO₂-based mixture can be substantially free from other components (i.e., consisting essentially of carbon dioxide, a surfactant, and water in the form of water-in-CO₂ microemulsions). When the surfactant mixture is applied, the contaminant (e.g., the endotoxin) becomes dissolved in the water inside the microemulsion. So, any contaminant may be incorporated or dissolved inside the surfactant structure.

The concentration of CO₂ itself is not meaningful, since there is an array of tiny droplets floating in a vast excess amount of CO₂. What is critical is the range of ratios of water-to-surfactant that are mixed together in the CO₂. This ratio, referred to as “W₀”, can have a range of about 5-100 molecules of water per molecule of surfactant (e.g., about 5-30 molecules of water per molecule of surfactant).

The compressed CO₂-based mixture generally has a pressure of at least about 400 psi (e.g., nominally about 400 to about 600 psi). However, in certain industrial applications, the compressed CO₂-based mixture can have a pressure of about 800 to about 5000 psi. Thus, in order to apply the compressed CO₂-based mixture to the substrate (e.g., a medical device) for removal of any endotoxins, the substrate can be loaded into a chamber, and the compressed CO₂-based mixture can be introduced into or formed within the chamber.

EXAMPLES

The purpose of the following illustrative example was to evaluate compressed carbon dioxide (CO₂)-based mixtures for the removal of Escherichia coli endotoxin first from smooth Ti surfaces and then from more complex geometries such as stainless steel lumens. Both Ti and stainless steel are common medical materials used in many applications that offer relatively high surface energies (33 and 70 mJ/m², respectively.), which favor endotoxin adherence. Naturally occurring E. coli endotoxin was used as the bio-contaminant because it is representative of the endotoxin type commonly found on Ti implant surfaces, catheters, wound dressings, and prosthodontic materials.

Solubilities of Ls-54 surfactant in supercritical (SC) CO₂ have been previously measured, and reported to be 0.05 M solubility of Ls-54 in CO₂ at 308.15 K and 22.0 MPa, along with a variety of molar water to surfactant ratios (W₀) for microemulsion formation at different pressures and temperatures. In this example, W₀ was selected to be 12.3 as the appropriate typical but not exclusive composition for our conditions. Thus, the amount of water and surfactant could be calculated based on the volume of the cleaning vessel. Below is a representation of the Ls-54 surfactant chemical structure.

Materials

• Chemicals and bio-contaminant. Dehypon Ls-54 surfactant was donated by Cognis Corporation, Ambler, Pa.; and bone-dry grade CO₂ (National Welders Supply Co., Durham, N.C.) with 99.8% purity was used as the main cleaning solvent. E. coli O55:B5 endotoxin (Lonza Walkersville Inc., Walkersville, Md.) was selected as the model bio-contaminant. Endotoxin-free water (HyPure™ Cell Culture Grade Water) was used for reconstitution, endotoxin recovery, and dilution processes (HyClone Laboratories Inc., Logan, Utah). The Limulus Amebocyte Lysate (LAL) Kinetic-QCL assay kit (Lonza Walkersville Inc.) was employed to determine endotoxin levels.
• Substrates. Commercially pure Ti disks with smooth surfaces measuring 12 mm in diameter and 2.5 mm in thickness were provided by Dr. Yuehuei An of the Medical University of South Carolina. Stainless steel tubes (Valco Instruments Co. Inc., Houston Tex.) of 3.175 mm (⅛ in) OD and 2.159 mm (0.085 in) ID were used to simulate lumens. Lengths of 102 mm (4 in) and 610 mm (24 in) were used to study the effect of length on cleaning efficiency.Methods
• Disk Surface Preparation. Disk surfaces were polished using sand paper (40, 15, 9, 5, and 1 μm grit) in a Multiprep polisher (Allied High Tech Products Inc., Rancho Dominguez, Calif.) for 20 minutes per grit. Subsequently, the disks were passivated using ASTM Standard F86-76. This standard requires sonication (Bransonic Ultrasonic Cleaner, model 8510R-MT, Branson Ultrasonics Corporation) in a detergent solution for 15 minutes, then acetone for 15 minutes and finally in 30% nitric acid for 30 minutes. After each step the disks were rinsed three times with DI water.
• Depyrogenation of Materials. Before each experiment, Ti disks, lumens, pipettes, and other glassware were depyrogenated in a dry heat oven (Fisher Scientific Isotemp Oven, model 725F) at 250° C. for 30 minutes. For depyrogenation, pipettes were placed in metal canisters and beakers, bottles, and disks were wrapped in aluminum foil. The LAL assay indicated no endotoxin on the depyrogenated items after evaluation.
• Endotoxin Reconstitution and Stock Solution Preparation. Vials of lyophilized E. coli endotoxin (2.5 mg/vial; nominal 7.5×10⁶ EU) were reconstituted as specified by the supplier and diluted with endotoxin-free water to obtain multiple stock solution concentrations. Subsequently, the desired substrate was contaminated using the stock solution.
• Endotoxin Detection Assay. Endotoxin levels were assayed using the chromogenic LAL Kinetic-QCL assay, which has a sensitivity range of 0.005-50 EU/mL. Samples were placed in a multi-detection microplate reader (model Synergy HT, Bio-Tek Instruments, Inc., Winooski, Vt.) and incubated for 10 minutes at 37° C. After the initial incubation, the LAL reagent was added and the samples were automatically monitored over time at 405 nm throughout the incubation period. The reaction time is inversely proportional to the endotoxin level. The concentration of endotoxin in a given sample is then calculated from the reaction time by comparison to the reaction time of solutions containing known amounts of endotoxin standard.
• Procedure for Coating and Processing Ti Disks. An aliquot of 200 μL from a stock solution of approximately 12,000 EU/mL was applied on the Ti disk surface and dried in a biohood at room temperature. This produced a film of approximately 2,000-2,500 EU/disk, depending on the stock solution concentration. Three coated disks 12 and one non-coated disk 14 were secured in a stainless steel plate 10 as shown in FIG. 1. The plate 10 was then attached to the shaft 16 of a stirrer and placed in a 1 L pressure vessel (FC series, Pressure Products Industries, Warminster, Pa.) for processing. A schematic of the 1 L pressure vessel apparatus is shown in FIG. 2. A standard CO₂ gas cylinder 1 provides CO₂ to the pump 2 (model P, Thar Design Inc., Pittsburgh Pa.), which in turn delivers compressed CO₂ to the vessel 4. An external heating jacket 7 and internal cooling coil 6 are provided at the vessel 4 to maintain the desired temperature. The shaft 16 is rotated at 400 rpm, which generates local shear forces on the surface of the disks 12, 14. However, the rotation of the flat disk 5 does not cause significant agitation of CO₂ in the vessel. After the desired time, the stirrer motor is turned off and the vessel 4 is depressurized. Temperatures and pressures for this preliminary study ranged from 5 to 40° C. and 13.8 to 27.6 MPa, respectively.

Further experiments were then conducted in a 23 mL phase equilibrium monitor (PEM) vessel (SPM 20, Thar Technologies Inc., Pittsburgh Pa.), which is a high pressure vessel having a maximum volume of 23 mL. A schematic of the apparatus is given in FIG. 3. The PEM is equipped with a video camera and sapphire windows (not shown) to allow visualization of the contents under pressure. It also has a motor-driven stirrer that allows high stirring rates, up to 3,800 rpm. Three physical configurations (as illustrated by FIG. 3) were investigated to study the effects of bulk agitation with and without mass transfer restrictions. All experiments in the PEM vessel were 2 hours long and stirring was set to 1900 rpm. A porous fit (5 μm porous size) is employed to support the disk and prevent initial direct contact between the coated surface and additives (surfactant and water) while the system is set-up. The frit also serves as an internal mass transfer barrier to create the desired flow restrictions for configurations shown in FIGS. 3b and 3c. Due to the volume limitation in this system, only one disk per experiment was processed at a time. Endotoxin recovery from Ti surfaces was achieved by sonication in an ultrasonic cleaner (model 250D, VWR, West Chester, Pa.). The disks (whether treated or untreated) were placed individually in a 40 mL depyrogenated glass bottle with 20 ml of endotoxin-free water and sonicated for 10 minutes Immediately following the recovery procedure, samples were diluted (1:200) and tested with the LAL assay.

• Procedure for Inoculation and Processing Lumens. A stock solution of approximately 30,000 EU/mL was drawn through the length of the lumens with the use of a syringe. The lumens were then capped at one end and placed vertically in a vacuum oven (VWR Vacuum Oven, model 1450M) for approximately 17 hours at 70° C. and 50.5 kPa to evaporate the water, leaving the endotoxin coated to the interior. Lumens were weighed in an analytical balance (model XS105 DualRange, Mettler-Toledo Inc., Columbus Ohio) before and after filling. On average the amount of stock solution was 0.360±0.005 g in the 102 mm lumens and 2.119±0.009 g in the 610 mm lumen. Measurements taken after drying confirmed that the water was completely evaporated. Endotoxin-contaminated lumens were processed for two hours in the same 1 L pressure vessel configuration shown in FIG. 2. Bulk agitation was provided by a flat-blade impeller rotating at 1900 rpm. Endotoxin recovery was carried out by placing the lumens separately in depyrogenated glass containers with an amount of endotoxin-free water (15 mL for the 102 mm lumens and 350 mL for the 610 mm lumen) and sonicated for 10 minutes. Immediately following the endotoxin recovery procedure, samples were diluted (1:200 and 1:100, respectively) and tested for endotoxin levels.Results and DiscussionSmooth Ti Disks

Initial experiments were carried out in the 1 L pressure vessel with non-bulk agitation in both liquid (5° C.; 13.8 and 27.6 MPa) and supercritical (40° C.; 27.6 MPa) CO₂ regions. Table I summarizes conditions (temperature, pressure, and time) and endotoxin loadings evaluated using the 1 L pressure vessel. Subsequent experiments were completed in the PEM vessel to investigate the impact of bulk agitation and mass transfer limitations at the conditions of pressure and temperature that gave the best indication of endotoxin removal in the 1 L pressure vessel.

The maximum recoverable endotoxin was defined as the endotoxin recovered from a contaminated, untreated disk immersed and sonicated in endotoxin-free water (negative controls). Treated disks were compared against the negative controls in each experiment to determine the endotoxin removal level. FIG. 4 shows the percentage endotoxin removal (FIG. 4a) and the residual endotoxin levels (FIG. 4b) for each treatment in the 1 L pressure vessel at 4 hours and 27.6 MPa. Neither pure SC CO₂ nor liquid CO₂ removed a significant fraction of endotoxin from the Ti surfaces. This is as expected, because CO₂ alone has insufficient solvent strength to dissolve the large endotoxin biomolecule. These results agree with a visual experiment previously performed in the PEM, where it was observed that compressed CO₂ did not dissolve endotoxins. For experiments employing liquid CO₂+Ls-54 and liquid CO₂+water, 80% and 78% endotoxin removal was attained, respectively. However, adding both Ls-54 and water together in CO₂ enhanced the removal to 93% as shown in FIG. 4. High levels of endotoxin removal were achieved when adding Ls-54 and water to both SC CO₂ (81%) and liquid CO₂ (93%). These results suggest that Ls-54 microemulsions are formed in either the liquid or SC CO₂ phase and the microemulsion system is effective in removing endotoxins. Nevertheless, microemulsions in liquid CO₂ removed a greater fraction of endotoxin than SC CO₂ microemulsions. The higher efficiency achieved in the liquid state (ρ=23.4 mole/L, 5° C. and 27.6 MPa) suggests that more microemulsions are formed in this state than in the SC state (ρ=20.4 mole/L, 40° C. and 27.6 MPa), because Ls-54 has higher solubility in compressed CO₂ at lower temperatures and constant pressures according to the data presented by Liu et al. (2002).

The 80% endotoxin removal for experiments with liquid CO₂+Ls-54 only, suggests formation of a microemulsion of the surfactant and endotoxin. Thus, indicating that an inverse micelle is still formed between the non-polar solvent and the surfactant allowing the endotoxin removal. However, further phase equilibrium studies should be addressed to evaluate this hypothesis.

Endotoxin removal was also appreciable when processing the Ti disks with liquid CO₂+water (78% endotoxin removal). Both the effect of water in the solvent capability of CO₂ and its affinity for endotoxins are to be examined. According to King et al. (1992) the water solubility in CO₂ at 25° C. and 20.7 MPa is approximately 0.079 M. Hence, for experiments carried out in the 1 L pressure vessel having CO₂+water, it is expected that at least 1.4 mL of water (out of the 12 mL initially added) is dissolved in CO₂. It is thus believed that water might work as a co-solvent, as previously reported by Casas et al. (2007) for the extraction of bioactive compounds using SC CO₂, increasing its polarity to enhance the endotoxin removal. In addition, LPS molecules contain long carbohydrate chains that favor its solubility in water. The LPS molecule contains two regions; the lipid chain (Lipid A) that is the hydrophobic region and the polysaccharide section (O-antigen and Core Region) that maintains the hydrophilic domain of the molecule. This suggests that the hydrophilic group in the endotoxin (which is larger than the hydrophobic region) dissolves in the mixture of liquid CO₂+water, thus explaining its removal.

Liquid and SC CO₂ microemulsions decreased the endotoxin levels in the disks to 144 and 498 EU/disk, respectively, from an initial loading of approximately 2,500 EU/disk. It is desirable to reduce endotoxin to less than 20 EU/disk. This might be feasible with a two-stage process. Hence, experiments with lower initial endotoxin loading (440±32 EU/disk) were conducted in the 1 L pressure vessel with liquid CO₂ microemulsions (i.e. CO₂ and additives at the liquid state). The cleaned disks had an average endotoxin level of 12±21 EU/disk for an average percent removal of 97%. This level is below the established USP requirements for medical devices and suggests that a two-stage process, using liquid CO₂ microemulsions with non-bulk agitation, might remove a theoretical 99.5% of endotoxin for surfaces initially coated with 2,500 EU.

The results presented in FIG. 4 were obtained with a 4 hour cleaning at 27.6 MPa. Shorter duration or lower pressures were also evaluated and the results are shown in FIG. 5. The average endotoxin level after 2 hour treatment at 27.6 MPa was 1296±189 EU/disk, corresponding to 59±2.4% removal. For the 4 hour treatment at 13.8 MPa the average endotoxin removal was 57±8%. Both of these treatments were less effective than the 4 hours treatment at 27.6 MPa. According to Liu et al. (2002) the solubility of Ls-54 in CO₂ decreases with pressure, thus, one expects less formation of microemulsions. With no bulk mixing in the 1 L pressure vessel, there is a lack of energy to form the microemulsions, thus longer time is needed for the system to reach equilibrium and achieve complete endotoxin removal. Therefore, it is expected that decreasing the duration or pressure of the treatment would reduce the efficiency of endotoxin removal.

Experiments in the PEM vessel were conducted at the temperature and pressure for the highest indication of endotoxin removal determined in the 1 L pressure vessel. Thus, all experiments were conducted in the liquid CO₂ region (25° C.; 27.6 MPa), adding either Ls-54 or water or both. The duration of these experiments was 2 hours. Three physical configurations (as shown in FIG. 3) were evaluated and Table II shows the endotoxin loadings and cleaning fluids employed for each configuration.

FIG. 3a gives the configuration for endotoxin removal with bulk agitation (1900 rpm stirring rate) and no flow or recirculation restrictions on the CO₂/microemulsion fluid. FIGS. 3b and 3c show two additional configurations of the PEM vessel. These configurations place the porous frit so as to restrict the circulation of the CO₂. Thus, it is possible to infer some effects of mass transfer restrictions by comparing results. Configuration 3b simulates of a cleaning process through a porous structure. Configuration 3c is somewhat similar to the 1 L pressure vessel, in that bulk agitation is provided directly to the CO₂ but not to the water and surfactant. The main difference between configurations is that 3b allows stirring of all cleaning fluids (CO₂ plus additives) while 3c allows stirring only of CO₂. For configuration 3b, the cleaning fluids initially lay below the contaminated disk surface, which is on top of the frit.

FIG. 6 shows percentage endotoxin removal in the PEM vessel system using configuration shown in FIG. 3a. Complete endotoxin removal (100%) was attained with both Ls-54 and water added. With vigorous bulk agitation conditions, the water-in-CO₂ microemulsion system is developed rapidly. Stronger agitation will also facilitate the mass transfer of the endotoxin into the microemulsion, making possible its complete removal in 2 hours. When Ls-54 or water were added individually to CO₂, the endotoxin removal was similar to that seen in the 1 L pressure vessel. With liquid CO₂+Ls-54, 85% of the endotoxin was removed while 83% was removed with liquid CO₂+water. However, the cleaning process in the PEM vessel was run for 2 hours at room temperatures (25° C.). This indicates that better (in the case of liquid CO₂ microemulsions) or similar (in the case of adding water or Ls-54 individually) endotoxin removal can be achieved in less time when strong stilling is provided. The 17% endotoxin removal with pure CO₂ is probably due to higher agitation and physical dislodgment, and residual water from the inlet and outlet lines.

FIG. 7 presents the results for experiments with mass transfer and flow restrictions. For configuration 3b, the average endotoxin removal was only 16%. In this configuration the porous frit initially separates the liquid additives from the contaminated disk. Hence, it is necessary that the water-in-CO₂ microemulsion phase migrates through the porous frit to dissolve and remove endotoxin from the disk surface, as would happen in a porous device. Configuration 3b models an actual porous structure where the liquid CO₂ microemulsions would be required to penetrate the porous surface in order to remove the contaminant. Because of this restriction, there is a mass transfer limitation and additional time is required to achieve complete endotoxin removal.

For configuration 3c the average endotoxin removal was 37%. In this configuration the liquid CO₂ is initially separated from the Ls-54 and water by the porous frit. Agitation is applied to the contaminated surface where liquid CO₂ is introduced. Because the contaminated surface is directly exposed to the rotating impeller, it can be inferred that some of the endotoxin removal is due to the high agitation and physical dislodgment. Surfactant and water must diffuse through the porous frit to the contaminated surface for complete removal. A similar phenomenon affected experiments carried out in the 1 L pressure vessel, where limited stirring energy meant that micelles moved to the disk surface mostly by diffusion.

Although strong stirring was provided for configurations 3b and 3c, mass transfer limitation still existed due to the porous frit. Placing the frit to support the disk in the middle of the vessel creates a barrier through which the cleaning fluids must diffuse. The data indicate that removal is dependent on the formation of microemulsion and its diffusion to the contaminated surface. In configuration 3b the microemulsion forms, but there is a diffusion limitation due to the frit barrier. Configuration 3b is more likely to occur in actual cleaning processes; therefore further work should be addressed in this scenario. Configuration 3c has both diffusion and microemulsion formation limitations. This might require more time to allow microemulsion to reach equilibrium and achieve higher removal. In contrast, for configuration 3a, there was neither diffusion limitation nor restrictions.

To take into account any endotoxin re-deposition during depressurization, a blank (non-coated) disk was processed simultaneously along with the contaminated disks for all cleaning trials in the 1 L pressure vessel. For the PEM vessel, due to space limitations, a blank disk was not processed simultaneously with the contaminated substrate. Instead, a non-coated disk was processed separately adding to the cleaning fluids the same amount of stock solution used to coat the disks. None of them exhibited endotoxin contamination when analyzed.

Stainless Steel Lumens

In this study, stainless steel lumens of two different lengths were used to determine whether endotoxin removal could be accomplished. The 102 mm lumens were first used to evaluate the effectiveness of the CO₂-based cleaning. Subsequent experiments were then conducted with 610 mm lumens, which are more representative of actual medical applications. Experiments were carried out in the liquid CO₂ region (27.6 MPa and 25° C.) employing either pure CO₂ or CO₂ with surfactant+water as the cleaning fluids.

As in the Ti disks, the maximum recoverable endotoxin for each evaluated lumen was defined as the endotoxin recovered by sonication of the contaminated, untreated lumen (negative control). Endotoxin levels detected from the processed or treated lumen were then compared to the average negative control to determine the endotoxin removal. For the 102 mm length, a total of 3 lumens were used through out the study. The 3 lumens were inoculated and processed simultaneously. On average 4274±682, 4154±398, and 4345±546 EU (n=3) was recovered from each untreated lumen as shown in Table III. Due to space limitations in the pressure vessel, only one lumen was used for the 610 mm length. The average endotoxin recovered from the inoculated, untreated 610 mm lumen was 26,932 EU±4802 (n=3). It needs to be pointed out that the loading in the longer lumen is 204 times greater than the highest endotoxin amount found in reusable angiographic catheters (450-1100 mm in length) as reported by Kundsin and Walter (1980). The high endotoxin loading, along with the fact that stainless steel supports endotoxin adherence, presents a strong cleaning challenge.

FIG. 8 shows the percentage endotoxin removal for all the evaluated lumens after treatment with both liquid CO₂ microemulsions and pure liquid CO₂. Complete endotoxin removal (100%) was attained for all lumens with liquid CO₂ microemulsions. These results, particularly for the long lumen, suggest that there was no mass transfer limitation under the experimental conditions tested in this work. However, pure liquid CO₂ did not remove a significant fraction of endotoxin from the stainless steel lumens, as expected. On average the remained EU for each lumen of 102 mm length after pure liquid CO₂ treatment was 3585±61, 3222±99, and 3678±84. For the 602 mm length, an average residual endotoxin of 24,254±760 EU remained after pure CO₂ treatment. The low removal percentage is consistent with the endotoxin removal attained from smooth Ti surfaces when only pure CO₂ is used.

These results are very promising for actual processes dealing with difficult-to-clean long narrow-lumen medical devices. As the design of new biomaterials and medical devices becomes more complex and environmentally-sensitive, new techniques to assure proper endotoxin removal must be developed as well. However, while this study was intended to determine the general efficiency and applicability of this technology, additional work should be addressed for configurations as described by FIG. 3b. This configuration presents limitations that could be found in actual cleaning process and also will be representative of porous devices such as those found in acetabular shells and femoral stems. Further work is being conducted applying this technology for porous coated Ti substrates.

Tables:

TABLE I
Experiments in the 1 L Pressure Vessel
			Time	Initial loading
Cleaning Fluid (s)	T (C)	P(MPa)	(hr)	(EU/disk)
Supercritical (SC) CO2	40	27.6	4	2900
SC CO2 + Ls-54 & water	40	27.6	4	2628
liquid CO2	5	27.6	4	2502 ± 71
liquid CO2 + Ls-54 & water	5	27.6	4	2348 ± 82
liquid CO2 + Ls-54 & water	5	27.6	4	440 ± 32
liquid CO2 + Ls-54	5	27.6	4	2970 ± 457
liquid CO2 + water	5	27.6	4	2618 ± 265
liquid CO2 + Ls-54 & water	5	13.8	4	2169 ± 810
liquid CO2 + Ls-54 & water	5	27.6	2	3145 ± 438

TABLE II
Experiments in the PEM (25° C., 27.6 MPa, and 2 hrs)
	Loading	PEM
Cleaning Fluid (s)	(EU/disk)	Configuration
liquid CO2 + Ls-54 & water	1633 ± 91	FIG. 3a
Pure liquid CO2	2225 ± 85	FIG. 3a
liquid CO2 + Ls-54	2225 ± 85	FIG. 3a
liquid CO2 + water	2225 ± 85	FIG. 3a
liquid CO2 + Ls-54 & water	1600 ± 100	FIG. 3b
liquid CO2 + Ls-54 & water	3397 ± 92	FIG. 3c

TABLE III
Endotoxin Levels Recovered from Untreated Lumens (102 mm long)
	Recovered Endotoxin Units (EU) per Lumen
Lumen	Exp 1	Exp 2	Exp 3	Mean ± SD
1	5007	3658	4156	4274 ± 682
2	4508	3723	4231	4154 ± 398
3	4900	3808	4327	4345 ± 546

Conclusions

The presence of endotoxin contamination represents a serious threat to biomaterials and medical products. The present study demonstrated that for a well mixed system, the novel water-in-CO₂ microemulsion system described in this invention can, at room temperatures and moderate pressures (25° C. and 27.6 MPa), remove 100% of the endotoxin applied on Ti surfaces and also to the endotoxin inoculated in two different lengths of stainless steel lumens.

Unexpectedly, higher endotoxin removal was achieved in the liquid region (5-25° C. and 27.6 Mpa) than in the SC region (40° C. and 4000 psi). This suggests that water-in-CO₂ microemulsions were formed in the liquid region. Higher Ls-54 solubility in compressed CO₂ at lower temperatures is shown in the data published by Liu et al (2002).

In the absence of high stirring rates (i.e. poor circulation rates or non-bulk agitation) mass transfer limitation existed; demanding more time for the mixture to reach equilibrium and achieve higher endotoxin removal from the Ti disks.

Safe endotoxin levels (≦20 EU/device as required for medical devices), were achieved after two hours when processing both Ti disks and lumens in a well mixed solution of liquid CO₂ microemulsion. At poorly mixed conditions, longer periods of time (>4 hours) were required to attain ≦20 EU/disk.

High fractions of endotoxin were removed from the Ti disks when employing mixtures of liquid CO₂ with either water or Ls-54. The endotoxin removal for both treatments was similar and ranged from 80% to 85%. Pure CO₂, either in the liquid or SC region, did not remove significant amount of endotoxins from the Ti disks and lumens because they are not soluble in CO₂.

The successful removal of endotoxins with compressed CO₂ is a promising alternative technology for the final cleaning of biomaterials and reusable medical devices. Compressed CO₂ at room temperature and relatively low pressure (25° C. and 27.6 MPa) with a small fraction of Ls-54 surfactant and water completely removed endotoxins from smooth Ti surfaces and stainless steel lumens. The use of CO₂ is favorable because CO₂ is inexpensive, non-toxic, non-flammable, and is readily available from industrial sources. In addition to complete removal of persistent contaminants such as endotoxins, this technology provides waste minimization and hazardous solvent elimination. The use of CO₂ as a cleaning solvent can reduce the need for washing in organic solvents, thus reducing their overall use in manufacturing processes.

These and other modifications and variations to the present invention may be practiced by those of ordinary skill in the art, without departing from the spirit and scope of the present invention, which is more particularly set forth in the appended claims. In addition, it should be understood the aspects of the various embodiments may be interchanged both in whole or in part. Furthermore, those of ordinary skill in the art will appreciate that the foregoing description is by way of example only, and is not intended to limit the invention so further described in the appended claims.

Claims

1. A method of cleaning a medical device, the method comprising loading the medical device into a chamber, andexposing the medical device to a compressed CO2-based mixture within the chamber, the compressed CO2-based mixture comprising carbon dioxide, a surfactant, and water in the form of water-in-CO2 microemulsions, wherein the compressed CO2-based mixture has a pressure of at least 400 psi.

2. The method as in claim 1, wherein the compressed CO2-based mixture consists essentially of carbon dioxide, a surfactant, and water in the form of water-in-CO2 microemulsions.

3. The method as in claim 1, wherein the ratio of water-to-surfactant mixed together in the CO2 has a range of about 5-100 molecules of water per molecule of surfactant.

4. The method as in claim 1, wherein the ratio of water-to-surfactant mixed together in the CO2 has a range of about 5-30 molecules of water per molecule of surfactant.

5. The method as in claim 1, wherein the compressed CO2-based mixture has a temperature of about 0° to about 100° C.

6. The method as in claim 1, wherein the compressed CO2-based mixture has a temperature of about 20° C. to about 60° C.

7. The method as in claim 1, wherein the surfactant is a non-ionic surfactant.

8. The method as in claim 1, wherein the surfactant comprises fatty molecule.

9. The method as in claim 8, wherein the surfactant comprises a derivized fatty molecule.

10. The method as in claim 9, wherein the surfactant comprises a fatty molecule derivatized by alkoxylation.

11. The method as in claim 9, wherein the surfactant comprises a fatty molecule derivatized by fluorination.

12. The method as in claim 1, wherein the compressed CO2-based mixture has a pressure of about 400 to about 600 psi.

13. The method as in claim 1, wherein the compressed CO2-based mixture has a pressure of about 800 to about 5000 psi.

14. The method as in claim 1, wherein the compressed CO2-based mixture is a liquid.

15. The method as in claim 1, wherein the compressed CO2-based mixture is a super critical fluid.

16. The method as in claim 1, wherein the compressed CO2-based mixture removes at least about 85% of bacterial endotoxin from the medical device.

17. The method as in claim 1, wherein the compressed CO2-based mixture removes at least about 95% of bacterial endotoxin from the medical device.

18. The method as in claim 1, wherein the compressed CO2-based mixture removes at least about 99% of bacterial endotoxin from the medical device.

19. The method as in claim 1, wherein exposing the exposing the medical device to a compressed CO2-based mixture within the chamber comprising introducing the compressed CO2-based mixture into the chamber.

20. The method as in claim 1, wherein exposing the medical device to a compressed CO2-based mixture within the chamber comprises forming the compressed CO2-based mixture within the chamber.