103 and other components discussed herein. The device 100 can be configured with different magnetic polarity configurations, e.g., all vertebrae 101 having the same magnetic polarity, vertebrae 101 with alternating magnetic polarities, vertebrae 101 with alternating pairs of magnetic polarities, vertebrae 101, etc. The present invention is not limited to any particular magnetic polarity configurations.

Each pair of adjacent interconnected vertebra 101 is rotatably connected by a cylindrical roller 104. For purposes of this discussion, a cylindrical roller is generally a hollow cylinder (e.g., tubular) although in some embodiments can be a solid cylinder. The roller 104 may be used to provide radial constraint and limit the degrees of freedom thus strengthening the assembly from a torsional standpoint. The roller 104 may be composed of a metal alloy, polymer, and/or composite. The roller 104 is preferably configured to allow rotation of the vertebrae 101 in the device plane, but otherwise to limit other degrees of freedom. The roller 104 may include features such as protrusions or recesses to help secure the roller 104 in between the two vertebra 101 and/or to control the amount of rotation that can occur between the two vertebra 101, e.g., to provide radial constraint and limit the degrees of freedom thus strengthening the assembly from a torsional standpoint. The roller 104 may be composed of any appropriate material such as a metal alloy, polymer, and/or composite. The roller 104 may be a separate component, adapted to fit between two or more magnets 103. Each magnet 103 may include a notch on each end adapted to fit a roller allowing for motion in one plane while restricting motion in the plane 90 degrees opposing.

Each pair of adjacent interconnected vertebra 101 typically also includes a flex element 106 (an example of which is shown in FIG. 23) that helps to move the vertebrae 101 from the delivery configuration to the assembled configuration. For example, the flex element 106 may operate as a spring (e.g., the flex element may include a spring or may be formed from a shape memory material) and may impose a bias toward the assembled configuration so that the vertebrae 101 are biased toward the assembled configuration upon delivery of the device 100. The flex element 106 may be composed of any appropriate material, e.g., a metal alloy, polymer, and/or composite and in particular a shape memory material. In various embodiments, the flex segment is positioned between two vertebrae skins on the outside perimeter of the array. It should be noted, however, that the flex segment may also be positioned on the internal perimeter of the array. The flex segment serves to strengthen the array as well as limit degrees of freedom. The flex member may also serve to aid the array to deploy into the proper assembled geometry.

The vertebra skin 102 may be formed of an appropriate material such as metal alloy, polymer, and/or composite and in particular may be formed of or include a shape memory material. The vertebra skin 102 encapsulates the magnet(s) 103 and helps to secure the roller 104 and flex element 106 within the device 100, forming a protective layer while also restricting degrees of freedom. The vertebra skin 102 may be secured onto the magnet(s) 103 with or without additional fasteners such as screws, pins, adhesive, interlocking elements, etc. Thus, for example, with a magnet 103, a roller 104, and a flex element 106 positioned for assembly, a vertebra skin 102 can be installed over the magnet 103, the roller 104, and the flex element 106 such as to encapsulate the components. As shown, the vertebra skin 102 may incorporate male nodes 108 and female nodes 110 on opposing ends of the vertebra capable of interacting with the opposing gender node on another vertebra. The male node 108 of one segment meshes or joins with the female node 110 of another segment. The incorporated nodes support the movement in one plane while restricting other degrees of freedom. The male nodes 108 may assist with securing the roller 104 and/or the flex element 106 within the device 100. The incorporated nodes may include stops to limit additional degrees of freedom, e.g., to prevent “backwards bending” or excess “forward bending” of the array, as represented in view 112. Stops can provide interference of the vertebrae while allowing a prescribed amount of rotation around the axis of the male and female node interaction.

FIG. 23 is an enlarged view of the various components shown in FIG. 22 and also shows an example of a flex element 106, which, in this example, is in the form of a bar having a central spring section, e.g., formed of a shape memory material. Without limitation, the flex element 106 can be configured to fit within channels 114 of the magnets 103 or in any other configuration (e.g., over or under the roller 104, which may include a flex element support 105 on which the flex element 106 rests to provide a fulcrum for the flex element 106) allowing the flex element 106 to control the movement of the vertebrae 101 such as biasing the vertebrae 101 toward the assembled configuration. Embodiments can include one, two, or more flex elements 106 and related elements such as elements 105 and 114, e.g., a flex element 106 placed on both sides of the vertebrae 101. A flex element 106 may be a unitary device or may include multiple components.

FIG. 24 shows an alternative example of a flex element 106, which, in this example, includes two oppositely-oriented U-brackets that, together with the roller 104 and magnets 103 as depicted in view 116, operate as biased springs in a manner similar to the flex element 106 shown and described above with reference to FIG. 23. Specifically, a U-bracket may be placed on each end of a magnetic segment. The U-brackets hold the roller 104 in place while allowing for rotation in one plane but restricting out-of-plane motion. The U-brackets may be composed of any appropriate material, e.g., a metal alloy, polymer, and/or composite and in particular may be formed of or include a shape memory material. The U-brackets allow for movement in a plane while restricting torsional motion and motion in the plane 90 degrees opposing. The roller and U-brackets may fit directly on the magnetic segments, and within the vertebrae skins, e.g., the U-brackets may fit through and be secured by the roller 104, as depicted in view 116, and extend along the length of the magnetic segment, e.g., in channels 114. The roller 104 may include flex element supports such as to secure the U-brackets in a position that allows bias toward the assembled configuration. By interlocking the roller 104 with the magnets 103, the roller 104 and U-brackets 106 serve to provide structure and shape to the array while also strengthening and preventing torsional motion in the array. This allows for more control over the final placement of the magnetic compression anastomosis device and prevents an undesired geometry from forming.

Various profiles of various geometries of magnetic arrays are possible with the vertebrae 101 of the present invention, as is shown in FIG. 25. For example, the vertebra skin 102 can include smooth and/or patterned features and can be configured with different outer geometries such as to accomplish different anastomosis goals. For example, geometry (a) has concave sides, geometry (b) has convex sides with notches, geometry (c) has pointed side protrusions that could assist with cutting tissue, and geometry (d) has flat sides that could compress and necrose a larger area of tissue. The vertebrae 101 can include other features such as various types of protrusions or recesses, e.g., to help secure the device 100 to tissue. All vertebrae could use the same geometry, or different vertebrae could use different geometries.

It should be noted that, while the flex element 106 is encapsulated underneath the vertebra skin in the embodiments described above, alternative embodiments could place the flex element(s) 106 on the outside of the vertebra skin.

It should be noted that, while two example flex elements 106 are shown and described herein, the present invention is not limited to these or to any particular flex element(s). For example, a flex element could include a coil spring. In some embodiments, a spring (e.g., a coil spring or a torsion spring, could act as both the roller and the flex element, for example, as depicted schematically in FIG. 26.

It should be noted that, in certain embodiments, the device 100 can be provided within a delivery device such as a catheter, endoscope, laparoscope, trocar, needle, or other delivery device and therefore the delivery device in combination with the device 100 can be considered an embodiment of the invention.

Embodiments also can include methods of manufacturing the device 100 such as by providing the various components (e.g., magnets, rollers, flex elements, and vertebra skins), positioning a roller and flex element between two magnets and securing them as needed (e.g., securing the flex element within magnet channels), and placing a vertebra skin over the magnet, roller, and flex element.

It should be noted that kits can be provided with magnets, rollers, flex elements, and different types of vertebra skins having different configurations that can be used for different types of anastomosis procedures such that different devices 100 can be prepared, e.g., depending on the amount and type of pressure needed for a particular procedure, the location of the procedure, and the size of the anastomosis to be created.

Certain alternative embodiments utilize internal vertebrae support structures such that each magnetic segment is essentially formed as a sandwich structure having two outer magnets separated and supported by an internal vertebra. The internal vertebra can be formed of one, two, or more pieces. The vertebrae can be interconnected to form a self-assembling magnetic compression anastomosis device, and elements of the types described above with reference to FIGS. 22-26 can be employed to restrict unwanted degrees of freedom.

FIG. 27 depicts portions of two multipiece internal vertebrae support structures that can be used to form sandwich structures, in accordance with certain embodiments. As shown in FIG. 28, the left-hand vertebrae support structure shown in FIG. 27 can support a first set of magnets and the right-hand vertebrae support structure shown in FIG. 27 can support a second set of magnets. Importantly, each of the vertebrae shown in FIG. 28 can support magnets having the same polarity facing outward such that when the two sets of vertebrae are interconnected to form a magnetic compression anastomosis device having an internal support, as shown in FIG. 29, both sides of the resulting magnetic compression anastomosis device can have the same magnetic polarity. In the examples shown in FIGS. 28-29, segments of the magnetic compression anastomosis device have north polarity on both sides (referred to herein as north-north segments). A similar magnetic compression anastomosis device can be formed having segments of south polarity on both sides (referred to herein as south-south segments). Among other things, having a first magnetic compression anastomosis device with all north polarity outer-facing magnets and a second magnetic compression anastomosis device with all south polarity outer-facing magnets can facilitate mating of the first and second magnetic compression anastomosis devices because the devices can mate on either side and the devices may attract each other from a further distance. Of course, segments can be configured with other polarities, e.g., north on the left side and south on the right side (referred to herein as a north-south segment) or south on the left side and north on the right side (referred to herein as a south-north segment). Also, a magnetic compression anastomosis device can have all the same type of segments (e.g., all north-north segments, all south-south segments, all north-south segments, all south-north segments, etc.) or alternatively a magnetic compression anastomosis device can have different types of segments, e.g., alternating north-north and south-south segments or other segment arrangements.

The magnets can be secured to the vertebrae, e.g., using an adhesive, mechanical fastener(s), clamp(s), interlocking element(s), etc. The magnetic compression anastomosis device can include one or more flex elements to bias the device toward the assembled configuration, e.g., one or more flex elements coupled between the two internal vertebrae support structures.

Additionally, during the coupling of two magnetic arrays, the ability to sense the mating array is more easily done with a single magnetic pole face. In various embodiments of the invention, an internal support is utilized to force the same pole faces together. The magnetic arrays may be mechanically held to the internal skeleton or adhered. The individual internal skeletons may be joined by various methods including, but not limited to, welding, bonding, swaging, and/or stamping. Individual and/or linked sections of internal skeletons provide an internal frame that resists the stresses of the magnetic segments being forced against their pole faces. Aligning the magnetic poles of the magnetic segments to have a full north face and a full south face on the array provides many benefits, such as greater sensing distance of the opposing pole magnetic array and allowing the many possible orientations of the array during coupling.

Individual internal vertebra pieces may be fabricated, e.g., from a metal alloy, polymer, and/or composite. As depicted in FIGS. 27-29, internal vertebra pieces generally have a similar shape to that of the magnetic segments but may be shaped differently than the magnetic segments. The individual internal vertebra pieces may have a male end opposite a female end, with ends of opposing genders capable of interlocking and joining pieces together. The individual internal vertebra pieces may also be connected to one another by bolts, rivets, and/or other fasteners or the like. The point of connection between internal vertebra pieces allows for rotation along an axis while also restricting tortional motion. This allows the internal vertebra pieces, when bound to magnetic segments, to move in a plane and convert from a delivery configuration to a deployed configuration.

The internal vertebra pieces may be mechanically held onto the magnetic array or may be adhered. The individual internal vertebra can be joined with a multitude of methods, such as but not limited to, welding, bonding, swaging, and/or stamping. As shown in FIG. 7, the magnetic segments attach to the internal vertebra pieces, with the magnetic poles facing against each other on the inside, allowing a full magnetic north pole facing the outside. The mating array has the same orientation, presenting a full north pole on the outside with a full south pole on the inside. By presenting full magnetic poles on the outside, the arrays will have a greater sensing distance and the rotational orientation of the arrays will not need to be specific. The internal skeletons may be used alone or in conjunction with external vertebrae supports to add structure and support while allowing motion in select degrees of freedom.

After the magnetic arrays mate, the internal skeletons join on the inside of the coupled system. As is shown in FIG. 28, full magnetic poles of the arrays oppose, such as two full norths and two full souths on the inside of the anastomosis system. These opposing poles are mated using the internal vertebrae. This results in two north poles on the outside, with two south poles on the inside of the anastomosis system. The arrays may also be configured with two south poles on the outside and two north poles on the inside of the anastomosis system. By utilizing full magnetic poles, the sensing distance between the arrays is greater and when connecting two separate magnets the geometric shapes align to produce a compression region with high enough pressure to shut down fluidic exchange to the tissue in the inner periphery of the geometric shape created by the self-forming array. This leads to more robust and safer anastomoses with a higher success rate.

It should be noted that self-assembling magnetic anastomosis addresses several of the historical disadvantages of traditional anastomosis such as allowing a surgical-quality anastomosis in a minimally-invasive fashion using devices that reproducibly re-assemble into a larger magnet structure of a predetermined shape in vivo. The constraints imposed by the described embodiments are designed to allow the devices to consistently self-assemble into the correct shape upon deployment, which greatly reduces the risks of surgical complications due to misshapen devices or premature detachment and also reduces the risks associated with surgical access and ensure that the anastomosis is formed with the correct geometric attributes. Overall, this ensures the patency of the anastomosis.

Thus, as described herein, embodiments include flexible linear magnetic devices comprising linked magnetic multipole segments that, when extruded from the end of a deployment channel or lumen, self-assemble to form a rigid, multipolar polygonal ring magnet (PRM; generally “magnetic device”). The self-assembly is directed by the configuration of magnets, rollers, flex elements, vertebra skins, and multipiece internal vertebrae support structures that is capable of returning to a pre-determined shape. Generally speaking, the physical and magnetic structure of the deployed magnetic devices is such that when two magnetic devices approach one another, there is a rapidly strengthening attractive magnetic interaction, which creates a coupling between the magnetic devices. In some instances, it is necessary to pre-align the complimentary devices, however, in other instances the devices self-align by undergoing fast in-plane rotation with respect to one another, as discussed in detail below. As described in detail below, systems including the magnetic devices may include an endoscope having sensors that allow the endoscope to sense the position of a mating magnetic device or another endoscope that will deploy the mating device.

When deployed in adjacent tissues, for example adjacent organs or different regions of the same organ, the coupled magnetic devices create a compressive ring that can be surgically opened, or allowed to form an anastomosis without further intervention. When paired devices are left alone, the compressive force against the tissues collapse the vasculature and extrude fluids in the tissues, further reducing the distance between the devices and increasing the magnetic attraction. With time, the coupled devices eventually couple completely and fall away, leaving a formed anastomosis. This cascade begins when the devices approach within “capture range,” whereby their mutually-attractive forces are sufficient to align the devices, trap the intervening tissue, and resist the natural pliancy of the tissues as well as the motion of the tissue under normal physiologic function.

Overall, the design specifications of the devices depend on the patient and the intended anastomosis. The design specifications may include: required capture range, desired effective inner and outer diameters of the deployed polygonal rings (e.g., as defined by the desired anastomosis size and instrument passage), thickness of the target tissue, and the inner diameter of guiding channel and the smallest radius of curvature to which the guiding channel may be bent and through which the magnets must pass. Once the design specifications are chosen, corresponding magnetic device designs can be determined, such as polygon-side-count and length, and the maximum lateral dimensions of the flexible linear magnetic structure that will be deployed through the delivery instrument.

Deployment of a device 100 is generally illustrated in FIG. 1. When used with the techniques described herein, the devices allow for the delivery of a larger magnetic structures than would otherwise be possible via a small delivery conduit, such as in a standard endoscope, if the devices were deployed as a completed assembly. Larger magnet structures, in turn, allow for the creation of larger anastomoses that are more robust, and achieve greater surgical success. Because the magnetic devices are generally radiopaque and echogenic, the devices generally can be positioned using fluoroscopy, direct visualization (trans-illumination or tissue indentation), and ultrasound, e.g., endoscopic ultrasound. The devices can also be ornamented with radiopaque paint or other markers to help identify the polarity of the devices during placement. In some embodiments, the devices can be positioned by use of sensors located in proximity to the delivery lumen and able to sense the position of a mating device, e.g., using a Reed switch or a Hall-effect sensor.

In general, as shown in FIG. 2A, a magnetic anastomosis procedure involves placing a first and a second magnetic structure adjacent to targeted tissues, thus causing the tissues to come together. The magnetic devices are generally deployed so that that opposite poles of the magnets will attract and bring the tissues together. The two devices may both be deployed inside the body, or one may be deployed inside the body and the other outside the body. Once the magnets have been deployed, the tissues circumscribed by the magnetic structures can be cut to provide an immediate anastomosis, as shown in FIG. 2B. In other embodiments, the tissues circumscribed by the devices will be allowed to necrose and degrade, providing an opening between the tissues. While the figures and structures of the disclosure are primarily concerned with annular or polygonal structures, it is to be understood that the delivery and construction techniques described herein can be used to make a variety of deployable magnetic structures. For example, self-assembling magnets can re-assemble into a polygonal structure such as a circle, ellipse, square, hexagon, octagon, decagon, or other geometric structure creating a closed loop. The devices may additionally include handles, suture loops, barbs, and protrusions, as needed to achieve the desired performance and to make delivery (and removal) easier.

As described with respect to the figures, a self-assembling magnetic anastomosis device can be placed with a number of techniques, such as endoscopy, laparoscopy, or with a catheter (e.g., not with direct visualization, fluoro, etc.). Regardless of method of device delivery, it is important to note that the procedure for creating the anastomosis can be terminated without perforation of tissue after confirmation of magnet coupling. As described previously, the compression anastomosis process can be allowed to proceed over the ensuing days, resulting in the natural formation of an opening between the tissues. The fused magnets can either be allowed to expel naturally or the magnets can be retrieved in a follow-up surgical procedure. Alternatively, if immediate bypass is required, the tissues circumscribed by the magnets can be cut or perforated. Perforation can be accomplished with a variety of techniques, such as cautery, microscalpel, or balloon dilation of tissue following needle and guidewire access.

In some embodiments, the self-assembling magnetic devices are used to create a bypass in the gastrointestinal tract. Such bypasses can be used for the treatment of a cancerous obstruction, weight loss or bariatrics, or even treatment of diabetes and metabolic disease (i.e. metabolic surgery). Such a bypass could be created endoscopically, laparoscopically, or a combination of both. FIG. 3 illustrates the variety of gastrointestinal anastomotic targets that may be addressed with the devices of the invention: stomach to small intestine (A), stomach to large intestine (E), small intestine to small intestine (C), small intestine to large intestine (B), and large intestine to large intestine (D). In an endoscopic procedure, the self-assembling magnetic devices can be delivered using two simultaneous endoscopes, e.g., an upper endoscope or enteroscope residing in the upper small intestine, and a colonoscope residing in the lower small intestine, as shown in FIG. 4A. Alternatively, as shown in FIG. 4B, two simultaneous upper endoscopes (e.g., one residing in the stomach and the second in the small intestine) can be used to place the devices. In other embodiments, the self-assembling magnets can be delivered sequentially through the same endoscope, which has been moved between a first deployment position and a second deployment position. For example, in FIG. 4A, a single per-oral endoscope could deliver and deploy one self-assembling magnet in the small intestine, withdraw, and then deploy the second reciprocal magnet in the stomach. Again, magnet coupling could be confirmed using fluoroscopy. FIG. 5 illustrates removal of a single endoscope after placement of two magnetic devices.

A variety of techniques can be used to detect the first deployed magnetic device to assist placement of the second mating structure. Once the first device is deployed at the desired anastomotic location, the two deployed magnetic devices need to find one another's magnetic field so that they can mate and provide the compressional force needed to prompt formation of an anastomosis. Ideally, the devices can be roughly located within several cm of one another (e.g., using ultrasound), at which point the magnets should self-capture and self-align. Where this is not possible, other techniques such as one of the following techniques can be used. A first location technique involves a direct contact method using two endoscopes. Here an endoscope's displacement in an adjacent lumen creates a displacement seen by another endoscope in the adjacent lumen. The displacement identifies a potential intersection point for an anastomosis location. For example, a magnetic deployment tool (described below) will be deflected by the presence of a deployed device on the other side of a tissue wall.

The second location technique involves trans-illumination, whereby high intensity light from one endoscope is directed at the lumen wall of the proposed anastomosis site. Using this technique, another endoscope in the adjacent lumen looks for the light, which diffuses through the lumen wall and projects onto the wall of the adjacent lumen. This light represents the potential intersection anastomosis point. A cap or lens can also be placed over the light emitting endoscope to further intensify and pinpoint the proposed intersection point. A similar technique could use radio-wave- or ultrasound-transducers and receivers to collocate the endoscope tips. In some embodiments, a system may include an endoscope having a sensor and a magnetic anastomosis device for deployment using the endoscope.

A third location technique involves magnetic sensing techniques to determine the proximity of the deployed ring magnet in the adjacent lumen. By maximizing the magnetic field being sensed, the minimum distance between the adjacent channels can be identified. The magnetic sensor can be carried on a probe inserted down the working channel of the endoscope and utilize common magnetic sensing technology such as a Hall Effect Sensor or Reed switch.

With trans-illumination and magnetic sensing, an additional accessory may also assist with delivering magnetic devises to a precise anastomosis site. A radially expanding ring structure can be deployed with the endoscope or laparoscope that can press fit and seat itself on the scope's outer diameter. The outer diameter of this expanding element is sized to allow the deployed device to seat itself on this expanding element (again likely a press fit). With this expanding element and magnetic device radially seated about the endoscope axis, the endoscope can be directed to the ideal anastomotic location through direct contact, trans-illumination, or magnetic sensing, and then the mating magnet device released when the anastomosis site is identified.

In other embodiments, the self-assembling magnet devices could be delivered using ultrasound guidance, e.g., endoscopic ultrasound. For example, using an echoendoscope in the stomach, a suitable small intestine target could be identified. As shown in FIG. 6, a delivery needle 600 (e.g., an aspiration needle) or catheter can be used to access to the small intestine target and deliver the self-assembling magnets into the small intestine lumen. The delivery can be guided with fluoroscopy or endoscopic ultrasound. Following self-assembly, these small intestine magnets would couple with a second set of magnets deployed in the stomach. The two devices can be delivered with the same needle or with different needles. It is also possible to deliver the first device with an endoscope and the second device with a needle or vice versa.

In another embodiment, illustrated in FIG. 7, a first anastomosis, created in an initial procedure, can be used to provide access for the creation of a second anastomosis. This process could theoretically be repeated multiple times to create additional anastomoses. For example, a gastrojejunal anastomosis (stomach to mid-small intestine) could serve as a conduit for the creation of a second, more distal gastrojejunal anastomosis. Ultimately, in this particular scenario, the stomach would have several bypasses to the small intestine. Additionally, in some instances, more anastomoses could be added to “titrate” to a specific clinical effect (e.g., lower glycosylated hemoglobin in type 2 diabetes). In alternative embodiments, an anastomosis may be placed to give access for a different type of surgery, e.g., tumor removal.

In another embodiment of delivery, the self-assembling magnets could be delivered laparoscopically through a surgical incision into the target organs (e.g., stomach and small intestine) and allowed to couple to create an anastomosis, as shown in FIG. 8. Again, this procedure could be directed with fluoroscopy or ultrasound and the procedure can be purely laparoscopic, or a combination of endoscopic and/or laparoscopic and/or needle procedures.

Gastrointestinal anastomoses can be used to address a number of conditions. An anastomosis or series of anastomoses between the proximal bowel and distal bowel may be used for treatment of obesity and metabolic conditions, such as Type II diabetes and dyslipidemia. The procedure can also be used to induce weight loss and to improve metabolic profiles, e.g., lipid profiles. The bowel includes any segment of the alimentary canal extending from the pyloric sphincter of the stomach to the anus. In some embodiments, an anastomosis is formed to bypass diseased, mal-formed, or dysfunctional tissues. In some embodiments, an anastomosis is formed to alter the “normal” digestive process in an effort to diminish or prevent other diseases, such as diabetes, hypertension, autoimmune, or musculoskeletal disease.

Using the self-assembling magnetic devices as discussed herein, it is possible to create a side-to-side anastomosis that does not require exclusion of the intermediate tissues, as is common with state-of-the-art bariatric procedures. That is, using the devices of the invention (or other means for creating an anastomosis) it is possible to create an alternate pathway that is a partial bypass for fluids (e.g., gastric fluids) and nutrients (e.g., food), while at least a portion of the old pathway is maintained. This design allows the ratio of “normal” to “modified” digestion to be tuned based upon the goals of the procedure. In other words, using the described procedure, a doctor can choose the ratio of food/fluids shunted down the new (partial) bypass versus food/fluids shunted down the old pathway. In most instances, the fraction shunted down the bypass limb will drive the patient toward the desired clinical endpoint (e.g., weight loss, improvement in glycosylated hemoglobin, improvement in lipid profile, etc.) The mechanism by which the endpoints are achieved may involve early macronutrient delivery to the ileum with stimulation of L-cells and increase in GLP-1 production, for example. The mechanism may also involve loss of efficiency of nutrient absorption, especially glucose, thereby reducing blood glucose levels. At the same time, however, the fraction shunted down the old pathway protects against known metabolic complications that can be associated with bariatric surgery such as excessive weight loss, malabsorptive diarrhea, electrolyte derangements, malnutrition, etc.

To achieve a desired ratio of bypass (e.g., re-routing food and secretions to flow down the new pathway, say, 70% or 80% or 90% or 100% of the time), the size, location, and possibly number of anastomoses will be important. For example, for a gastrojejunal anastomosis, it may be critical to place the anastomosis in a dependent fashion to take advantage of the effects of gravity. Also, instead of a round anastomosis, it may be better to create a long, oval-shaped anastomosis to maximize anastomotic size. Alternatively, multiple gastrojejunal anastomoses may be used to titrate to a certain clinical endpoint (e.g., glycosylated hemoglobin in Type II diabetes). Most of the procedures described herein may be used to place one or more anastomoses, as needed, to achieve the desired clinical endpoint. For example, the two endoscope procedures illustrated in FIGS. 4A and 4B can be used to create a partial bypass of a portion of the bowel. Based upon the desired ratio of bypassed and non-bypassed nutrients, the anastomoses shown in FIGS. 4A and 4B can be made larger, e.g., greater than 1 cm in open diameter, or several smaller anastomoses can be placed to achieve the desired ratio.

The procedure is also adjustable. For example, a first anastomosis may be formed and then, based upon clinical tests performed after the procedure, one or more anastomoses can be added to improve the results of the clinical tests. Based upon later clinical results, it may be necessary to add yet another anastomosis. Alternatively, it is possible to partially reverse the condition by closing one or more anastomosis. Because the partially bypassed tissues were not removed, they can return to near normal functionality with the passage of greater amounts of nutrients, etc. The anastomoses may be closed with clips, sutures, staples, etc. In other embodiments, a plug may be placed in one or more anastomoses to limit the ratio of nutrients that traverse the “normal” pathway. Furthermore, it is possible to close an anastomosis in one location in the bowel and then place a new anastomosis at a different location. Thus, is possible to generally and tunably create partial bypasses, or a series of partial bypasses, between portions of the bowel to achieve clinical endpoints, e.g., as described in FIG. 3.

The described procedures may also be used with procedures that remove or block the bypassed tissues, as is common with bariatric procedures. For example, a gastrojejunal anastomosis may be coupled with a pyloric plug (gastric obstruction) or another closure of the pylorus (e.g., sutured closure) to shunt food completely down the new bypass. Such procedures can be used, for example, to bypass tissue that is diseased, e.g., because of cancer.

In another category of procedures, endoscopic ultrasound (EUS) can be used to facilitate guided transgastric or transduodenal access into the gallbladder for placement of a self-assembling magnetic anastomosis device. Once gallbladder access is obtained, various strategies can be employed to maintain a patent portal between the stomach and the gallbladder or the duodenum and the gallbladder. In another embodiment, gallstones can be endoscopically retrieved and fluid drained. For example, using the described methods, an anastomosis can be created between the gallbladder and the stomach. Once the gallbladder is accessed in a transgastric or transduodenal fashion, the gallstones can be removed. Furthermore, the gallbladder mucosa can be ablated using any number of modalities, including but not limited to argon plasma coagulation (APC), photodynamic therapy (PDT), sclerosant (e.g., ethanolamine or ethanol).

One strategy for creation of a portal is to deploy self-assembling magnets via an endoscopic needle under ultrasound guidance into the gallbladder and also into the stomach or duodenum. These magnets will mate and form a compression anastomosis or fistula. A second strategy for creation of a portal is to deploy self-assembling magnets via an endoscopic needle 600 as shown in FIGS. 9A and 9B. While the coupled magnetic assemblies are shown as octagons, the closed frame could take the shape of any polygonal structure, e.g., a square, a circle, a triangle, hexagon, heptagon, nonagon, decagon, dodecagon, etc. One such device would be deployed into the gallbladder, and the mating device would be deployed into the stomach or duodenum. In the same fashion as discussed above with respect to gastrointestinal deployment, the tissue circumscribed by the two magnetic devices can be cut with cautery, microscalpel, needle-knife, or other deployable cutting mechanism. In another embodiment, the coupled tissues can be left to necrose and form the anastomosis.

The devices need not be limited to forming holes, however. Other structures can be coupled to one or more mating magnetic devices to created additional functionality. For example, a stent could be deployed between tissues, such as the gallbladder and the stomach, as shown in FIG. 10. Alternatively, the gallbladder magnet could be coupled to a balloon-based device that fills with air, fluid, magnetic pieces or magnetic particles. Upon inflation, the balloon would serve as an anchor in the bile duct following placement. The balloon could also have an annular configuration to allow for immediate access after coupling with the second magnet. See, e.g., FIG. 11. Regardless of embodiment, however, it is critical to contain the original access pathway within the confines of the coupled magnets, i.e., not leaving a pathway for the escape of bile. Otherwise, the opening will allow bile leakage that can result in peritonitis.

Another medical application for self-assembling magnets is direct biliary access. Currently, to achieve decompression for a malignant biliary stricture, endoscopic retrograde cholangiopancreatography (ERCP) is performed. The biliary tract is accessed endoscopically through the papilla in retrograde fashion and a stent is deployed across the stricture over a guidewire. These stents frequently require subsequent procedures for exchange, clean-out, or placement of additional overlapping stents. The need for exchange and cleaning is required to counteract the high rate of infection of the biliary tree (i.e. cholangitis) when using an ERCP procedure. Because of the high rate of morbidity, ERCP is typically limited to patients that have no other option to address pancreatic disease.

Using devices of the invention, however, it is possible to easily form an anastomosis between the bile duct (preferably the main bile duct) and either the duodenum or the stomach (choledocho-gastric and choledocho-duodenal anastomoses, respectively). This anastomosis is permanent and typically does not require intervention if located apart from the diseased tissue. In an embodiment, a biliary magnetic device is delivered directly into the bile duct under endoscopic ultrasound guidance. As described below, the self-assembling magnetic device is extruded through a needle or catheter, whereupon it deploys in the correct configuration. Using fluoroscopy or ultrasound, it is then possible to confirm that the device has self-assembled and is in the correct location. In some embodiments, the magnetic device may be tethered to the delivery needle or catheter by means of a detachable wire or suture to enable mechanical retraction until optimal positioning is confirmed.

In one embodiment, the magnetic device can be delivered endoscopically to the bile duct via wall of the duodenum, as shown in FIG. 12. In another embodiment, the biliary magnet can be delivered in conventional retrograde fashion through the ampulla into the bile duct, as shown in FIG. 13. One benefit of retrograde delivery is that it avoids needle punctures across tissue planes, as is the case with the deployment method shown in FIG. 12. Regardless of the method for delivering the biliary magnets, however, a second magnetic device is required in either the gastric (A) or duodenal (B) lumen, as shown in FIG. 14. Typically, this decision is dependent upon the patient's anatomy (e.g., size of the duodenal lumen) and the location of the initial biliary magnet. In scenarios based on endoscopic ultrasound needle delivery, the second magnetic device can be connected to the biliary magnet via the aforementioned detachable wire, and therefore extruded through the same delivery needle/catheter. Alternatively, the second device can be pre-attached to the exterior of the endoscope and slid into position for coupling after biliary magnet deployment. The latter procedure may be more applicable to forward-viewing echoendoscopes but may be used with endoscopes, generally.

In another embodiment, the biliary magnet is a balloon-based device that fills with air, fluid, magnetic pieces or magnetic particles, similar to previously described with respect to gallbladder procedures. Upon inflation, the balloon would serve as an anchor in the bile duct following placement. In an embodiment, the balloon could have an annular configuration to allow for immediate access after coupling with the second magnet. Additionally, like the gallbladder procedures described above, a biliary magnetic device can be used with a stent form-factor. In an embodiment, the stent has an internal biliary magnet and a hinged external magnet. The stent can be inserted in retrograde fashion through the ampulla into the bile duct. The hinged external magnet can then be swung around and coupled with the internal biliary magnet to form a fistula between the bile duct and the duodenum, as shown in FIG. 15.

The magnetic devices of the invention can also be used to treat pancreatic diseases. For example, the pancreatic duct requires decompression in certain disease states, such as chronic pancreatitis. Currently, extensive pancreatic duct decompression requires surgery (e.g., Peustow surgery in which the pancreas is filleted along the axis of the pancreatic duct and connected to a loop of small intestine for improved pancreatic drainage). As an alternative to Peustow surgery, extensive pancreatic duct decompression can be accomplished via creation of a large magnetic compression anastomosis between the pancreatic duct and either the stomach or duodenum using a magnetic pancreatic catheter, as shown in FIG. 16. The catheter can be magnetic along its entire length or only at certain intervals. The catheter can be in the form of a stent or straw. The pancreatic duct can be accessed using conventional ERCP methods (retrograde cannulation through the ampulla) or by direct needle access using endoscopic ultrasound (EUS). The magnetic pancreatic catheter can be delivered into the pancreatic duct and coupled with a second magnetic device in either the stomach or duodenum. As in the biliary scenario described above, the magnetic pancreatic catheter could be hinged to the second magnetic device.

Self-assembling magnetic devices can also be used to access and drain fluid collections located adjacent to the gastrointestinal tract, as shown in FIG. 17. For example, following a bout of pancreatitis, pancreatic fluid collections can form that require drainage. While drainage can be accomplished using surgery or a percutaneous catheter, endoscopic drainage has been found to be more clinically and cost-effective, but can be complicated by bleeding, perforation, and/or inadequate drainage. As an alternative to surgical draining, magnetic devices of the invention can be delivered through a needle or sharpened catheter into the collection under endoscopic ultrasound (EUS) guidance, as shown in FIG. 17. Following assembly, the first magnetic device is coupled to the second magnetic device that has been placed in the gastrointestinal lumen (e.g. stomach). In order to speed removal after drainage, the first magnet may be tethered by a connecting wire as previously described. As described previously, the intervening tissue can be cut using electrocautery or dilation followed by needle and wire access. Additional devices, such as magnetic coupling clamps can be used to control blood flows to allow for “blood-less” endoscopic entry into the collection.

Self-assembling magnets can also be used for urological applications such as forming bypasses to treat an obstructed urogenital tract, as shown in FIG. 18. For example, a magnetic anastomosis could be created between the renal calyx and bowel (A), between the ureter and bowel (B), or between the bladder and bowel (C). Self-assembling magnetic devices of the invention can be delivered into the urological tract using an endoscope, laparoscope, or needle, as described above. The reciprocal magnetic device could be delivered into the gastrointestinal tract using an endoscope, laparoscope, or needle as previously described. In other embodiments, the devices can be used for reproductive procedures, such as bypassing a portion of obstructed fallopian tube or bypassing a vasectomy.