Magnetic Resonance Visibility, Artifacts, and Overall Safety of the Self-Locating Peritoneal Dialysis Catheter with a Tungsten Tip

Background The self-locating peritoneal dialysis (PD) catheter, contains a tungsten tip. The effects of magnetic resonance (MR) on the catheter were evaluated, emphasizing its MR signal, artifacts, ferromagnetism, and possible heating production during the MR sequences. Methods The catheter was studied in an ex vivo model using a 1.5T MR system and placed into a plastic box containing saline solution. Acquisitions on coronal and axial planes were obtained on fast gradient-echo T1-weighted and fast spin-echo T2-weighted. In vivo abdominal MR exams were also carried out. Results Overall, the catheter had good visibility. In all sequences, an extensive paramagnetic blooming artifact was detected at the level of the tip tungsten ballast, with a circular artifact of 5 cm in diameter. The catheter showed no magnetic deflection, rotation, or movements during all MR sequences. After imaging, the temperature of the saline solution did not change compared to the basal measurement. Patients safely underwent abdominal MR. Conclusions The results point to the possibility of safely performing MR in PD patients carrying the self-locating catheter. The self-locating PD catheter is stable when subjected to a 1.5T MR system. However, it creates some visual interference, preventing an accurate study of the tissues surrounding the tungsten tip.


Introduction
Te self-locating peritoneal dialysis (PD) catheter was designed to avoid dislocation [1,2]. It is a classic straight Tenckhof catheter with the usual internal diameter of 2.6 mm, modifed with a heavy tip built with 12 grams of silicone-encased tungsten, which is physically and chemically inert. Te heavier catheter tip stays in the lower abdomen, while the classic Tenckhof catheter usually foats on top of the dialysis fuid, carrying a higher risk of dislocation. In addition, if the self-locating PD catheter is displaced for any reason (e.g., as in a patient who kept an upside-down position), it will return to the lower abdomen after walking for a short time [3].
Tungsten is a paramagnetic element like aluminum, oxygen, titanium, and iron oxide. Studies on the behavior of paramagnetic elements show that magnetic felds do not infuence them. On the contrary, paramagnetic elements can interfere with the magnetic feld surrounding them, leading MRI systems to artifacts in the acquired images [4]. Indeed, in the product information of the self-locating catheter, the following sentence is reported: "Te catheter ballast may interfere with MR diagnostic techniques. " Tus, it is not clear  how the catheter interferes with the magnetic feld, reducing  the visibility of the surrounding structures, or if it can be  harmful to the patient. Tis study aimed to evaluate the behavior of the selflocating PD catheter when subjected to MR, emphasizing its MR signal, artifacts, ferromagnetism, and possible heating production during the MR sequences.

Materials and Methods
A GoldSeal 1.5T Signa Excite HD magnetic resonance imaging system (General Electric Healthcare, Little Chalfont, United Kingdom) of 1.5 Tesla (T) was used to evaluate the self-locating PD catheter (Care-Cath: B. Braun Avitum Italy, Mirandola, Italy). Te catheter was tested after attachment at the external extremity of a locking adapter titanium connector (Baxter Healthcare Inc., Deerfeld, Illinois, USA) commonly used in clinical practice. Tus, both tungsten (at the catheter tip) and titanium were subjected to the MR feld. Te titanium-loaded, self-locating PD catheter was placed into a plastic box containing 150 ml of saline solution (NaCl, 0.9%), and it was oriented in parallel with the MR detector ( Figure 1). Figure 2 shows the standard Xray appearance of the self-locating PD catheter used in this experiment. Acquisitions on coronal planes ( Figure 3) were obtained on 3-mm thick fast spin-echo T1-weighted (TR/TE 400/10 msec) and fast spin-echo T2-weighted (TR/TE 3000/ 90 msec) sequences, using a 16-channel body array surface coil. Te total acquisition time of the two MR sequences was 8 minutes. Catheter signal and displacement, size of the artifacts, and possible heating production were evaluated under the same conditions by two experienced radiologists. Te heating was evaluated by measuring the temperature of the saline solution in which the self-locating PD catheter was immersed for the test. Te temperature was measured immediately before and immediately after imaging with a digital thermometer sensitive to 0.1°C.
We also report representative MR images of the abdomen of patients who underwent MR exams.

Results
In all MR sequences (Figure 3), an extensive paramagnetic blooming artifact (a "signal hole") was demonstrated at the level of the tip tungsten ballast, with a circular artifact of about 5 cm in diameter ( Figure 3). Te titanium connector generated minor artifacts (<1.5 cm), and overall the catheter had good visibility.
Catheter safety-wise, when placed into the magnetic feld, the peritoneal dialysis catheter showed no magnetic defection. Moreover, there was no evidence of rotation or movement during all MR sequences. No diference in temperature of the saline solution was detected immediately before and immediately after imaging; the temperature remained stable at 17.3°C.
Following these fndings, we obtained informed consent to perform abdominal MR scans on PD patients with weighted catheters. Tree patients with a self-locating catheter underwent MR scans of the abdomen: one for a study of the biliary tract and two for the dorsal and lumbar spines, respectively.
Although a signifcant artifact surrounded the tungsten tip in pelvic projections, afecting the visibility of the lower lumbar spine and sacrum (Figure 4(c)), the self-locating catheter was confrmed to be safe, with the patients not sufering from any side efects. Moreover, the abdomen, the retroperitoneal area, the upper lumbar region, and the dorsal spine scan were not afected by the artifact produced by the tungsten tip (Figures 4(a)-4(f )).

Discussion
Te self-locating PD catheter has the advantage of reducing dislocation from the lower abdomen. Several observational studies [5,6] and two randomized control trials [7,8] proved this feature. Complications such as cuf extrusion, tunnel infection, PD-associated peritonitis, early leakage, and obstruction were also statistically less frequent in patients with self-locating catheters than in those with classic Tenckhof catheters. In addition, the reduced malfunction rate of the self-locating catheter is associated with a lower chronic laxative burden among PD patients [9].
However, concerns have been raised regarding the performance of an MR exam in patients carrying the selflocating PD catheter because tungsten is a high-density rare metal, although chemically inert. In addition, the possible presence of ferrous impurities in the tungsten catheter tip has been hypothesized. For these reasons, despite the lack of reliable epidemiological studies on its real difusion, the use of the self-locating catheter seems to be limited worldwide. Terefore, one of the study's main aims was to dispel any doubts about the safety of the catheter in order to implement its use.
Although several studies on MR peritoneography have previously shown the safety and accuracy of MR in studying catheter-related complications in PD patients, none of them focused on self-locating catheters [10][11][12]. A recent study showed a case series of 14 PD patients who underwent MR for several reasons, mainly a central nervous system study. Notably, none of them underwent abdominal MR imaging [13]. Similar to our results, it showed a good safety profle of the self-locating catheter in the clinical setting and no defection in the ex vivo experiments. Te self-locating catheter demonstrated MR interference with the appearance of artifacts in both T1 and T2 around the catheter tip, with a greater extension with the 3T MRI (about 12 cm) than with the 1.5T MRI (about 8 cm). Only the 1.5T MRI was tested in our study, showing interference of about 5 cm. As previously reported, a more prominent interference is expected with a 3T MR machine [14]. Adding to the published ex vivo data, we confrmed that the tungsten-containing self-locating catheter produces artifacts in vivo with the 1.5T MRI.
Metallic biomedical implants and devices lead to artifacts in the magnetic resonance during patient scanning. Ferromagnetic materials are contraindicated for magnetic resonance because these materials are set into motion during magnetic resonance scanning [15]. Te following unwanted events may occur with ferromagnetic materials: attractive efects and resultant patient injury; radiofrequency interference with the MR imaging study and secondary image artifacts; radiofrequency power deposition leading to device heating and secondary patient injury [16,17].
Even with nonferromagnetic metals, such as titanium, the presence of metallic implants inside the body can determine MR susceptibility artifacts [18] and radiofrequency (RF) overfow artifacts [15,19], causing a nonuniform appearance to an image due to perturbation of RF homogeneity near the metal part. In addition, the electric conductivity of the metal allows the high-frequency electromagnetic felds during an RF pulse to induce electric currents. Tus, a metal object may prevent the RF feld from passing into a tissue, causing a signal void in an image close to the metal object.
Interestingly, an in vitro study [14] investigating the RFmetal interaction efects caused by metallic instruments and implants made of titanium or nitinol (biopsy needles, hip prostheses, vascular stents, and aneurysm clips) showed that interference is more prominent at 3T.
In the present study, we tested at 1.5T the presence of interference and instability of the self-locating peritoneal catheter carrying a tungsten tip intended to be inserted in the peritoneal cavity and the external titanium connector.
Tungsten, the chemical element with the symbol W for Wolfram and a density of 19.3 g·cm −3 , is a hard, rare metal. In other words, its density is 19.3 times that of water, comparable to that of gold, and signifcantly higher than that of lead and titanium. Metallic tungsten is hypoallergenic. Its conductive properties made it one of the primary sources for X-ray [20]. Compared to nonmagnetic materials with a relative permeability of 1, tungsten is paramagnetic with a relative permeability of 1.000068 [21]. Tus, tungstenbased devices should be MR safe, although they might show signal interferences. In addition, the nonhomogeneous composition of the metal tungsten tip could also be a source of MR interference artifacts. A previous study [22], performed in the neuro-interventional feld, tested the MR safety of endovascular microcatheters with nitinol, tungsten, and polyetheretherketone braiding at 1.5T and 3T because of the possibility that microcatheter fragments may be entrapped in patients following endovascular procedures. Subsequent diagnostic MR examinations thus pose a safety concern due to the possibility of radiofrequency heating of the metallic braid incorporated into the microcatheter. In contrast to nitinol microcatheters, the tungsten braided microcatheters did not demonstrate heating and therefore showed potential for safe use in MR imaging.
We demonstrated that the tungsten tip ballast generates signal artifacts that, even when large, do not impair the assessment of the upper abdominal organs and upper spine. However, evaluation of the pelvic structures, such as the genital organs, rectum, and sacral spine, would be critical. In addition, we showed that no movement of the catheter or generation of heat was observed at 1.5T. Moreover, to our knowledge, this is the frst study to show the efects of abdominal MRI exams in PD patients with a self-locating catheter, confrming the presence of local artifacts shown ex vivo in previous publications [13] and in our experimental setting but demonstrating the absence of detectable side efects in vivo. However, until more extensive experience is available in humans, during MR imaging in patients with the Figure 2: Conventional X-ray of the peritoneal dialysis catheter that demonstrates the radiopaque tip tungsten ballast (arrowhead) and, on the other extremity, the titanium connector (arrow). 4 International Journal of Nephrology self-locating catheter, we suggest flling the peritoneal cavity with 0.5 liters of PD fuid, as we did in our patients, to 2 liters, as is usually done in MR peritoneography. Te fuid will disperse any heat produced by the catheter tip, thus decreasing the already very low risk of a clinically signifcant thermal injury to adjacent tissues.
Te present study has some limitations. First of all, the ex vivo experiments and in vivo analyses were performed only with a 1.5T MRI because it was the only one available to us, whereas higher intensity MRIs are commercially available (3, 4, 5, or even 8 Tesla). Only by testing the catheter with these more potent MRI scanners will it be possible to International Journal of Nephrology understand if the self-locating catheter will confrm its compatibility. We suggest testing the catheter ex vivo, as we did in our study, before using a 3T or higher MRI machine on a patient carrying this kind of catheter. In addition, as a single-center pilot study, only three patients were accurately studied in the in vivo analysis. However, 12 additional patients with the tungsten-containing self-locating PD catheter underwent MRI imaging of the central nervous system or the abdomen without clinical consequences. Future studies with larger case series and higher-intensity MRI are needed to confrm our fndings.
In conclusion, this study demonstrates that the selflocating PD catheter is reasonably safe when subjected to a 1.5 Tesla MR system, although it creates some visual interference, preventing an accurate study of the tissues surrounding the tungsten tip. Tese fndings point to the possibility of performing MR imaging studies in PD patients carrying the self-locating catheter.

Data Availability
Te data used to support the fndings of this study are included within the article.

Conflicts of Interest
Te authors declare that they have no conficts of interest to disclose.