Chapter 2 ARTHROGRAPHY AND JOINT INJECTION AND ASPIRATION: Principles and Techniques
Joint injection/aspiration is a commonly performed procedure in many radiology practices, but in some practices it is uncommon and radiologists have variable experience. This chapter outlines simplified techniques for performing these procedures as well as different approaches to joints that have been described.
Conventional arthrography involves the percutaneous injection of contrast material into a joint, followed by a series of radiographs with specific views, depending on the joint being imaged. As a diagnostic technique, conventional arthrography has been replaced by other imaging modalities in nearly all cases. Conventional athrography is rarely performed today; however, it is essential for the practicing radiologist to be familiar with the techniques used because the injection methods can be applied to any advanced form of arthrographic imaging. In patients with severe claustrophobia and in centers without CT or MRI technology, it may be necessary to use conventional arthrography. Many textbooks have addressed the nuances of the art of arthrography, and these provide a useful reference guide if one of these procedures needs to be performed.
The techniques used in conventional arthrography can also be applied to joint aspiration. For joint aspiration, fluoroscopy, ultrasound, or CT guidance can be used. However, fluoroscopy is used most commonly because of its versatility, relatively low cost, and ease of use.
For aspiration of suspected septic arthritis, an 18-gauge needle or larger is recommended for joint access, since infected joint fluid usually has a higher viscosity compared with that of regular joint fluid. When performing a joint aspiration, it is essential to inject a small volume of iodinated contrast material to verify the intra-articular location of the needle tip and to assess any potential abnormal communications of the joint space. If aspiration of a suspected septic joint yields no fluid, it may be useful to instill sterile saline into the joint, followed by aspiration, thereafter sending any aspirated material for culture. When infection is suspected, it is best to avoid the initial intra-articular administration of iodinated contrast material, which has bacteriostatic properties.
CT arthrography (CTA) is useful for demonstration of cartilaginous and osseous intra-articular bodies, cartilage defects, fracture fragments, synovial abnormalities, and ligamentous disruption (Figs. 2-1 through 2-7). CTA is usually reserved for patients with contraindications to MRI, but is being performed more commonly recently owing to improved scanner technology that provides isotropic volumetric data.
Figure 2-3 CT arthrography using multidetector CT yields high resolution and can visualize small cartilage defects; however, unlike MRI it is difficult to see the cartilage unless contrast is directly adjacent. Therefore, the joint must be adequately filled with contrast. Note sharp calcification within the medullary cavity representing chronic bone infarctions. ACL, anterior cruciate ligament; PCL, posterior cruciate ligament.
Figure 2-5 CT arthrography shows excellent detail despite the presence of metal. In this case, the screws appeared loose radiographically, but CT arthrography shows that the rotator cuff repair remains water-tight.
Figure 2-6 With the advent of multidetector CT, CT arthrography has become more popular. Acquisition of thin slices yields isotropic voxels, which can be reformatted in any plane without loss of resolution.
Figure 2-7 Multidetector CT arthrography is good for imaging surface detail of cartilage. In this example, a CT arthrogram of the ankle joint shows an osteochondral lesion of the medial talar dome with only slight fissuring of overlying cartilage.
The most common contraindications to MRI are patients with claustrophobia, patients with implanted pacemaker/defibrillator devices, and patients with any other internal metallic objects that specifically exclude the possibility of MRI because of their composition or location.
CTA involves the intra-articular injection of iodinated contrast material typically followed by axial CT scanning. With multidetector CT technology, very thin sections allow for acquisition of volumetric data, thereby allowing reconstructions to be performed in any plane without loss of resolution. The newergeneration multidetector CT scanners generate a higher photon flux, thus decreasing the streak artifact that often limited CTA on older scanners. Therefore, 300 mg/mL nonionic iodinated contrast can be used without the need for intra-articular air injection when a multidetector scanner is used and excellent quality images can be acquired. With older-generation CT scanners, it is prudent to dilute contrast (typically 50:50) with sterile saline or to use a less concentrated iodinated contrast preparation to avoid streak artifact. Many authors in the past have advocated injection of variable amounts of air as a negative contrast agent. This is useful to distend the joint without creating significant artifact, and the air can be moved about the joint by putting the patient in different positions. Some authors “coat” the synovium with only a few milliliters of contrast, filling the rest of the joint with air. These techniques have fallen by the wayside with the advent of multidetector CT technology.
The techniques used for accessing the joint are the same as those used for conventional arthrography and MR arthrography (MRA). Contraindications to CTA include patients with a history of severe contrast allergy and pregnant women. Because the spatial resolution that can be achieved with CTA is high, some authors feel that CTA is excellent for the detection of subtle cartilage surface lesions. CTA has been studied in the evaluation of the postoperative knee, with visualization of meniscal re-tears indicated by contrast entering the meniscal substance. CTA is also very useful in patients who have metallic orthopedic hardware in the region of interest. Using high mAs techniques, beam-hardening artifact from implanted metal hardware can be almost completely eliminated. Such patients would typically not be candidates for MRA owing to the distortion of MR images secondary to metallic susceptibility artifact. In the near future, multidetector CTA is likely to be increasingly used for imaging postoperative joints.
Conventional MRI has enjoyed great success in imaging the musculoskeletal system and has deservedly become the “gold-standard” imaging technique for suspected internal derangement of joints. There are, however, several limitations of conventional MRI examinations, including the inability to visualize small intra-articular structures and the fact that many pathologic processes have similar signal intensity to normal anatomic structures. Postoperative findings may also be similar in signal intensity to pathologic changes. Unless there is an effusion, conventional MRI may be somewhat limited owing to nondistention of the joint.
Direct MRA involves the direct injection of dilute gadolinium, followed by MR imaging. This type of MRA leads to improved intra-articular contrast owing to the T1 shortening effect of gadolinium and also provides distention of the joint, which allows smaller intra-articular structures to be visualized (Figs. 2-8 through 2-12). The distention effect alone of saline injection has proved to be better than conventional MRI in some studies. The distention effect also forces intra-articular contrast through and around pathologic entities. With adequate distention, intraarticular gadolinium enters any pathologic entity in communication with the joint space. MRA provides excellent soft tissue contrast and demonstrates many abnormalities beyond the resolution of conventional MRI. One of the main disadvantages of MRA is the presence of artifacts seen in patients with implanted prostheses and metallic hardware.
Figure 2-10 MR arthrography showing contrast entering the posterior horn. MR arthrography visualizes communications better than noncontrast MRI owing to a distention effect as well as use of higher SNR (signal-to-noise ratio) T1-weighted technique. However, keep in mind that a horizontal cleft (top image) may not represent a re-tear; some surgeons leave these if they are stable, debriding only the unstable flaps rather than performing a more extensive resection that might accelerate joint degeneration.
Figure 2-11 Direct MR arthrogram demonstrating an undersurface partial-thickness tear of the rotator cuff extending interstitially. On the T2-weighted image, the interstitial tear is seen but communication with the undersurface component is not apparent.
To perform direct MRA, the gadolinium injected should be diluted to 2.5 mM. To achieve this concentration we add 0.1 mL of gadopentetate dimeglumine to 20 mL of normal saline. The final gadolinium dilution ratio should be 1:200 to achieve maximal signal (Fig. 2-13). It has been demonstrated that iodinated contrast reduces the T1 shortening effect of gadolinium. This effect is seen on both high and low field-strength systems. It is therefore advisable to use as little iodinated contrast as needed when performing direct MRA examinations. The quality of the examination on direct MRA does not appear to be affected by patient exercise. We therefore encourage patients to ambulate normally to the MRI suite after injection, but we do not routinely prescribe nor avoid exercise after injection for direct MRA.
Figure 2-13 A, Plot of MR signal (Y axis) versus increasing gadolinium (Gd) concentration diluted in saline (X axis); note peak T1 signal at 2.5 mM indicated by the purple shadow. T2 signal continues to diminish with higher Gd concentration. B, Same signal/Gd concentration plot with dilution in straight iodinated contrast instead of saline. Compared with saline dilution, note shift of the T1 peak to a lower concentration as well as a more rapid decrease in T2 signal with increasing concentration. The same effect is observed at low and high field strengths.
(Adapted from Montgomery et al. J Magn Reson Imaging 2002;15:334–343.)
Current indications for direct MRA include assessment of the glenoid labrum in the shoulder and acetabular labral tears in the hip. Direct MRA is also useful for the imaging of cartilage lesions. MRA can also be used for accurate determination of the congruity of the rotator cuff and for distinguishing between rotator cuff tendinosis and a full-thickness tear (when gadolinium pathologically enters the subacromial-subdeltoid bursa). Direct MRA is useful in imaging the postoperative joint (especially the shoulder and knee). It is also useful for assessment of intraosseous ligament tears in the wrist and for assessment of ligament tears about the elbow joint. Direct MRA in the ankle can determine the patency of the anterior talofibular ligament in patients with ankle sprain. Direct MRA is a safe procedure with minimal to no adverse effects reported over almost two decades.
Indirect MRA involves the injection of a standard dose of 0.1 mmol/kg of intravenous gadolinium followed by delayed imaging to create an “arthrographic effect” as the contrast diffuses into the joint (Fig. 2-14). As the gadolinium diffuses from the bloodstream into the synovial compartment of the joint being imaged, the degree of arthrographic effect depends on the volume of synovial fluid within the joint and on the degree of synovial vascularity. Although this technique is attractive from a logistical standpoint, there is an interpretive learning curve, and certain limitations should be recognized. One of the main disadvantages of indirect arthrography is the lack of joint distention (unless a preexisting effusion was present) (Fig. 2-15). Typically, patients with small to moderate joint effusions and synovial reaction achieve a better arthrographic effect with indirect MRA. A tense effusion may prevent or delay contrast uptake because of increased intra-articular pressure. Also, a large effusion or insufficient delay (usually at least 30 minutes is required between injection and imaging) may cause heterogeneous or incomplete filling of the joint with gadolinium (Fig. 2-16). Smaller joints such as the wrist and ankle and the small joints of the hands and feet achieve a more consistent arthrographic effect owing to the high synovial area/joint volume ratio. Exercise after injection of intravenous gadolinium increases diffusional flow into the patient’s joint and improves the arthrographic effect. One of the main advantages of indirect arthrography is the fact that it can be performed offsite without the need for an invasive procedure. This technique may also be more acceptable to patients because it involves an intravenous injection only. When optimal, indirect MRA can be as effective as a direct MR arthrogram (Figs. 2-17 and 2-18).
Figure 2-14 Rate of articular uptake of intravenously administered contrast depends on synovial vascularity, synovial surface area, synovial permeablity, joint fluid viscosity, and intra-articular pressure.
Figure 2-15 Indirect MR arthrography showing an anterior labral tear. Since intravenous injection merely distributes gadolinium in existing fluid compartments, there is no added distention effect. Although joint distention is desirable in arthrography, most diagnoses can be made without distention effect. However, diagnoses requiring flow of contrast from the joint through pathologic communications will be limited.
Figure 2-16 Suboptimal indirect MR arthrogram. If the delay between intravenous injection and imaging is not long enough, too little contrast will be present in the joint. Contrast that is present is distributed along the capsule and synovial recesses where more vascular tissue is present.
Figure 2-18 Indirect MR arthrography of the postoperative knee. Note that on an indirect MR arthrogram, vascular structures within and outside the joint enhance; therefore, there is a separate learning curve associated with their interpretation. In this case there is enhancement of the posterior horn of the medial meniscus. Although enhancing granulation tissue can be seen after surgery, the irregular morphology of the meniscus and irregular enhancement make it more likely to be a tear or re-tear. However, the patient is more likely to be symptomatic because of the associated cartilage loss, seen better on the T2-weighted image.
Indirect arthrography is not optimal in the setting of low signal-to-noise ratio or where fat suppression is not available (e.g., imaging with low-field MRI or deep structures such as the hip). Heterogeneous enhancement within large joints can lead to misdiagnosis, and optimally patients should be monitored and additional delay or exercise used if this is the case. Interpreter errors can occur with this technique owing to enhancement of normal intra-articular structures (such as the periphery of the triangular fibrocartilage complex) (Fig. 2-19). Indirect arthrography is also useful in the postoperative patient, but is considered less useful than direct MRA because postoperative granulation tissue can also enhance and may be confused with recurrent tear. On the other hand, this can be an advantage in the nonoperated joint. Granulation tissue is often intermediate-to-low signal on T2-weighted images; yet the tissue can exclude fluid or contrast. As a result, chronic injuries such as glenoid labral tears can potentially go undetected on routine MRI as well as with direct MRA. Although more research is needed, enhancement on indirect MRA could theoretically identify these lesions (Fig. 2-20). Intravenous injection can also be used to salvage a suboptimal direct MR arthrogram (Fig. 2-21). Advantages and disadvantages of indirect MRA compared with the direct approach are summarized in Box 2-1.
Figure 2-20 On indirect MR arthrography, scar enhances; therefore, this technique may prove to be very useful for detection of chronic injuries containing granulation tissue. Such injuries, when intra-articular such as the labral tear in Figure 2-15, may not allow contrast injected directly into the joint to enter if scar has formed. AC, acromioclavicular.