Potential use of a diluted high-relaxivity gadolinium-based intra-articular contrast agent for magnetic resonance arthrography: an in-vitro study

Magnetic resonance imaging (MRI) has become the most important diagnostic tool for the assessment of joint disorders throughout the body, allowing for detecting abnormalities of the joint space, cartilage, fibrocartilages, tendons, ligaments, and synovial tissue [1]. In this setting, a further advanced tool is represented by MR arthrography (MRA), which is mainly used to evaluate young and less young patients and has been demonstrated to be superior in detecting a number of intra-articular joint abnormalities compared to conventional MRI [2, 3].

MRA requires intra-articular injection of gadolinium-based diluted paramagnetic contrast material [1, 4]. Contrast solution is injected until the capsule distends [5]. Traditionally, paramagnetic contrast solution was prepared manually, diluting a certain amount of gadolinium-based contrast agent into saline solution to obtain an approximate concentration of 2 mmol/l [5]. A recent survey showed this approach is still used in about 60% of institutions [6]. On the other hand, two commercial preparations are available as pre-filled syringes ready to be injected. These solutions are based on gadoteric acid (Gd-DOTA, 2.5 mmol/l; Dotarem, Guerbet, France) or on gadopentetate dimeglumine (Gd-DTPA, 2 mmol/l; Magnevist, Bayer, Germany). These two contrast agents have different molecular structures and properties but have similar R1-relaxivity values of approximately 4.2–5.3 l/mmols^− 1 at 1.5 T and produce similar contrast enhancement when administered at the same dose [7]. Gadobenate dimeglumine (Gd-BOPTA, MultiHance, Bracco Imaging SpA, Italy) is a gadolinium-based contrast agent which has unique features compared to other agents, including very high R1-relaxivity thanks to a transient weak binding to blood proteins and is mostly used for liver imaging [8].

To our knowledge, Gd-BOPTA has never been used for intra-articular applications.

The purpose of our study was to test in vitro different concentrations of Gd-BOPTA to be potentially used for MRA.

Figure 2 shows T1-weighted axial scan of the two different sets of pipes containing different solutions of gadolinium-based contrast with and without addition of synovial fluid.

Our in vitro experience demonstrated that synovial fluid slightly but significantly increased SI of Gd-BOPTA. In pipes containing synovial fluid, Gd-BOPTA at 1 mmol/l had a + 28% SI increase in comparison to Gd-DTPA 2 mmol/l.

Gd-BOPTA has unique features that distinguish it from all other gadolinium-based contrast media. In particular, it has a dual route of elimination from the body that enables Gd-BOPTA to be used both as nonspecific agent and liver-specific agent [10]. Moreover, a higher R1-relaxivity compared to other gadolinum-based contrast agents permits lower overall doses to be used or provides increased SI enhancement at equivalent dose [7]. In fact, Gd-BOPTA has a weak and transient interaction with serum albumin which confers a partial blood-pool effect and a lower rate of contrast dilution compared to that of other gadolinium agents [11]. Indeed, Gd-BOPTA has been proven to be effective in brain [12], vascular [13], liver [14], and breast MRI [15], in which it can be used at lower doses compared to other gadolinium-based agents. At present, however, the use of Gd-BOPTA has been limited to the liver only, given the reported accumulation of Gd in patients’ tissues after repeated intravenous administrations [16, 17]. On the other hand, it should also be noted that a recent report showed how Gd does not accumulate in patients undergoing MRA [18].

Regarding MRA, the use of Gd-BOPTA has never been clearly reported before except for an animal study published in 2007 in which its effect on cartilage was tested [19]. However, in some studies dealing with MRA the type of contrast agent may be not reported [20]. Moreover, other studies used Gd-BOPTA to perform indirect MRA [21], an alternative imaging technique based on the premise that contrast material administered intravenously diffuses into the joint space, owing to high vascularization of the synovial membrane and lack of basement membrane in its capillaries. Compared with the direct technique, indirect MR arthrography has the main disadvantage of not allowing for capsular distension. On the other side, intravenous injection allows assessing disease activity in patients affected by inflammatory joint diseases [22]. At any rate, the use of direct MRA is continuously increasing, notwithstanding improved coils at 1.5 T and increased diffusion of 3 T systems.

Compared to 2 mmol/l solution of Gd-DTPA, SI of Gd-BOPTA was higher at all concentrations except 0.5 mmol/l. The highest SI was achieved at 1 mmol/l, while a similar SI was obtained at 0.5 mmol/l in pipe sets added with synovial fluid. In clinical practice, these differences may be negligible, implying that also a low concentration of gadolinium-based solution can be considered acceptable. The main practical result of these findings is that, in a simulated joint environment, the same SI of pre-diluted Gd-DTPA syringes can be obtained with up one fourth concentration of Gd-BOPTA. However, in vivo testing of this speculation should be warranted.

We acknowledge the limitations of an in vitro study in which joint environment has been reproduced only adding a small amount of synovial fluid to the contrast agent contained in sterile pipes. This implies that our data may be not directly applicable in vivo, as other factors (e.g., synovial tissue, inflammatory cytokines) may be implicated in the interaction with Gd-BOPTA. Also, body temperature may somewhat affect behavior of contrast agents in vivo, while our experiment was conducted at room temperature. Moreover, we used the synovial fluid withdrawn from the knees of patients with a mean age of 68 years and mild degenerative osteoarthritis. It is known that synovial fluid of elderly subjects may contain less glycoproteins and glycosaminoglycans. Thus, in younger subjects, in whom MRA is more likely to be performed, results may be somewhat different. We can speculate that a higher content of glycoproteins may further improve SI, although at present we do not have data to confirm such hypothesis. Finally, we did not assess separately T1 and T2 effects of the contrast media and our results are only valid for one specific parameter set TR and TE. Also, relative signal changes with noise correction should also be measured to assess reliable relaxivity values R1 and R2. However, our study was primarily aimed to measure SI, that is the most useful parameter that may affect clinical practice.

In conclusion, in vitro, Gd-BOPTA at 1 mmol/ had a + 28% SI increase in comparison to Gd-DTPA 2 mmol/l. A SI similar to Gd-DTPA can be obtained using one fourth concentration of Gd-BOPTA. Further in vivo studies are warranted to confirm clinical applicability and value of our results.