Intraarticular Local Anesthesia: Can It Reduce Pain Related to MR or CT Arthrography of the Shoulder?

Transcription

1 Musculoskeletal Imaging Original Research Choo et al. Musculoskeletal Imaging Original Research Hye Jung Choo 1 Sun Joo Lee 1 Dong Wook Kim 1 Seok Jin Choi 1 In Sook Lee 2 Choo HJ, Lee SJ, Kim DW, Choi SJ, Lee IS Keywords: arthrography, intraarticular injection, local anesthesia, pain, shoulder DOI: /AJR Received June 4, 2012; accepted after revision August 15, This work was supported by the 2012 Inje University research grant. 1 Department of Radiology, College of Medicine, Inje University Busan Paik Hospital, Gaegeum-dong Jin-gu, Busan, Republic of Korea, Address correspondence to H. J. Choo (nayaa_neo@hanmail.net). 2 Department of Radiology, Pusan National University School of Medicine, Pusan National University Hospital, Busan, Korea. AJR 2013; 200: X/13/ American Roentgen Ray Society Intraarticular Local Anesthesia: Can It Reduce Pain Related to MR or CT Arthrography of the Shoulder? OBJECTIVE. The objective of this study was to prospectively evaluate whether intraarticular anesthesia can reduce pain after MR or CT arthrography of the shoulder. SUBJECTS AND METHODS. This study included 120 patients who underwent CT or MR arthrography of the shoulder. The patients were randomized into two groups: the study group (n = 60), each administered an intraarticular injection of the contrast agent mixed with a local anesthetic (2% mepivacaine); and the control group (n = 60), each injected with the contrast agent only. Each patient s pain level was assessed at five phases baseline and immediately, 2 hours, 1 day, and 2 days after injection by using a visual analog scale or a verbal rating scale. The net pain score at each phase was also calculated. The pain course and net pain score were compared between study and control groups by repeated-measures analysis of variance. the patients were divided into subgroups according to patient- or procedure-related factors, the efficacy of the intraarticular local anesthetic in each subgroup was evaluated. RESULTS. The pain course showed a quadratic trend and was not significantly different between study and control groups. The net pain score also was not significantly different between the two groups. No subgroup showed a significantly different efficacy of the intraarticular local anesthetic between the two groups. CONCLUSION. Intraarticular local anesthesia did not reduce arthrography-related pain. D irect MR arthrography of the shoulder is a very accurate modality to evaluate the glenoid labrum and rotator cuff tendon and recently has become the primary imaging modality to detect internal derangement of the shoulder [1, 2]. CT arthrography is an alternative method used to diagnose shoulder abnormalities in particular, labral lesions [3 5]. Direct MR and CT arthrography are somewhat invasive procedures, however, and patients often complain afterward of arthrography-related pain [6 9]. For this reason, some radiologists administer an intraarticular injection of a local anesthetic along with the contrast agent to control the pain. However, an effect of intraarticular local anesthesia in relieving arthrography-related pain has seldom been reported in the medical literature [10]. Therefore, we conducted a prospective study to determine whether intraarticular injection of a local anesthetic with the routinely used dose can reduce pain related to MR or CT arthrography of the shoulder. Subjects and Methods This single-center, randomized, prospective study was approved by the institutional review board for human research of Inje University Busan Paik Hospital. Informed consent was obtained from all patients. Patients Between May 2011 and March 2012, a total of 210 consecutive patients with shoulder discomfort underwent direct MR or CT arthrography of the shoulder at our institution. Among them, patients who declined to participate in this study (n = 80), patients who were unavailable for a telephone interview (n = 6), and patients who underwent MR arthrography of both shoulders on the same day (n = 4) were excluded, resulting in the inclusion of 120 patients in this study. No patient had septic arthritis or an allergy to iodine or was younger than 16 years of age. All the patients were allocated into either the study or the control group by block randomization. Patient history of shoulder arthroscopic surgery and trauma was obtained by reviewing the medical records. 860 AJR:200, April 2013

2 Arthrographic Technique All the arthrographies were performed with fluoroscopic guidance by three senior radiology residents and one staff musculoskeletal radiologist with 7 years of experience, who also supervised the arthrographies performed by the senior residents. By use of a 25-gauge needle (a syringe-needle unit), approximately 1 ml of 2% mepivacaine hydrochloride (Mevan, Hanlim Pharm) was administered to anesthetize the skin at the puncture site. A 22-gauge spinal needle was advanced, targeting the middle portion of the anterior glenohumeral joint space, and approximately 1 ml of an iodinated contrast agent, iopromide (Ultravist 370, Bayer Schering Pharma), was injected to confirm the intraarticular needle position. Then, the physician injected the contrast mixture until resistance was encountered, to ensure joint fullness, and checked the piston syringe, to draw back in response to overdistension and high-pressure buildup in the glenohumeral joints before withdrawing the syringe. The physician who performed the arthrography recorded the volume of the injected contrast mixture and the level of difficulty of the arthrography procedure as either difficult or easy. Mixture of Contrast Agent Patients in the study group were administered an intraarticular injection of diluted contrast material with 2% mepivacaine. The contrast mixture used in MR arthrography was composed of 0.07 ml of gadobutrol (Gadovist, Bayer Schering Pharma), 1.5 ml of 2% mepivacaine, and 18.5 ml of normal saline, and in CT arthrography, it was composed of 6.5 ml of iodinated contrast material (Ultravist 370), 1.5 ml of 2% mepivacaine, and 12 ml of normal saline. Patients in the control group were injected with diluted contrast material without a local anesthetic. In MR arthrography, the contrast mixture was composed of 0.07 ml of gadobutrol and 20 ml of normal saline, and in CT arthrography, it was composed of 6.5 ml of iodinated contrast material (Ultravist 370) and 13.5 ml of normal saline. Pain Assessment The patients pain level was assessed using a visual analog scale or a verbal rating scale [11], in which a score of 0 indicated no pain and a score of 10 indicated the most severe pain ever experienced. The patients were questioned about their pain level at five phases: before the injection, immediately after the injection, 2 hours after the injection, 1 day after the injection, and 2 days after the injection. The pain levels before and immediately after the injection were obtained by the radiologist who performed the shoulder arthrography. The pain levels at the other three phases (2 hours, 1 day, and 2 days after injection) were assessed by a telephone interview conducted by another radiologist who was blinded to the patients group assignments. During the period of pain assessment, the patients were also blinded as to whether intraarticular local anesthetic had been used when they underwent arthrography. MRI and CT MR images were obtained with a 3-T MR scanner (Achieva 3.0T TX, Philips Healthcare) using an eight-element phased-array shoulder coil or a 32-element phased-array cardiac coil. The standard protocol for MR arthrography of the shoulder included axial, oblique coronal, and oblique sagittal fat-suppressed T1-weighted images and oblique coronal and oblique sagittal T2-weighted images. The total imaging time for MRI was approximately 30 minutes. CT was performed on a 128 detector row CT scanner (Definition AS+; Siemens Healthcare) with a tube voltage of 120 kvp and automatic tube current modulation. The sections were reconstructed into axial, oblique coronal, and oblique sagittal planes with 1-mm slice thickness. The total imaging time for CT was approximately 2 seconds. Imaging Findings The presence of extraarticular contrast leakage, rotator cuff (supraspinatus-infraspinatus tendon and subscapularis tendon) tears, and labral tears on CT TABLE 1: Comparison of Patient-Related and Procedure-Related Factors Between and s Factor p Patient-related factors Age years 13 (22) 17 (28) > 40 years 47 (78) 43 (72) Sex 0.36 Male 30 (50) 36 (60) Female 30 (50) 24 (40) History of previous shoulder surgery 26 (43) 28 (47) 0.86 History of shoulder trauma within 1 year 13 (22) 15 (25) 0.83 Rotator cuff tears 32 (53) 33 (55) 1.00 Supraspinatus-infraspinatus tendon tear 25 (42) 28 (47) 0.71 Subscapularis tendon tear 21 (35) 19 (32) 0.85 Labral tear 19 (32) 14 (23) 0.41 pain 1.00 < 4 30 (50) 29 (48) 4 30 (50) 31 (52) Procedure-related factors Contrast leakage 29 (48) 33 (55) 0.58 Contrast volume ml 35 (58) 40 (67) > 13 ml 25 (42) 20 (33) Level of difficulty 1.00 High 11 (18) 10 (17) Low 49 (82) 50 (83) Physician experience 0.71 Resident 39 (65) 36 (60) Specialist 21 (35) 24 (40) Modality 0.71 CT arthrography 25 (42) 28 (47) MR arthrography 35 (58) 32 (53) Note Data are no. of patients (%) except where otherwise indicated. AJR:200, April

3 Choo et al. TABLE 2: Mean Pain Score in Each Phase and Comparison of Pain Course According to Passage Between and s Factor Injection Injection Injection Injection All patients Patient-related factors Age 40 years > 40 years Sex Male Female History of previous shoulder surgery History of shoulder trauma within 1 year Rotator cuff tears Supraspinatus-infraspinatus tendon tear Subscapularis tendon tear Labral tear pain < Procedure-related factors Contrast leakage Contrast volume 13 ml > 13 ml Level of difficulty High Low Physician experience Resident Specialist Modality CT arthrography MR arthrography p Note Data are mean pain score (from 0 to 10, where 0 is no pain and 10 is most pain ever experience) except where otherwise indicated. 862 AJR:200, April 2013

4 group group Fig. 1 Graph shows mean pain score at each time phase in all patients. or MRI was determined by consensus of two radiologists who were blinded to the patients group assignments. group group Statistical Analysis The homogeneous distribution of patient- and procedure-related factors between study and control groups was compared by using chi-square analysis. The analyzed patient-related factors included age, sex, tears of the rotator cuff or labrum, baseline shoulder pain, and history of shoulder arthroscopic surgery ever or trauma within 1 year. The analyzed procedure-related factors included presence of leakage of contrast material, injected volume of the contrast mixture, level of difficulty of the arthrography procedure (difficult vs easy), physician experience (staff vs resident), and imaging modality (CT vs MRI). The continuous variables age, baseline shoulder pain, and injected volume of the contrast mixture were analyzed, after division of the patients among two groups, by using the cutoff points of 40 years of age, pain score of 4, and injection volume of 13 ml, respectively. The pain course according to time passage between study and control groups was compared by using repeated-measures analysis of variance. The net pain score (representing the amount of pain induced only by arthrography) was calculated at each phase (immediately, 2 hours, 1 day, and 2 days after injection) by utilizing the following formula: NetP A = P A P B, where NetP A is the net pain score in phase A, P A is the pain score during phase A, and P B is the pain score at baseline. The net pain score at each phase was compared between study and control groups by using the Student t test. Subgroup analysis was also performed after dividing the patients according to patient- and procedure-related factors. In each subgroup, the differences, between study and control groups, in terms of the pain course according to time passage and in terms of the pain score at each phase were reanalyzed. All statistical analyses were performed by using PASW Statistics 18 software (IBM), with the significance set at p < Results Patient s The study group consisted of 30 women (mean age, 55.4 years; range, years) and 30 men (mean age, 42.0 years; range, years). The control group consisted of 24 women (mean age, 52.3 years; range, years) and 36 men (mean age, 49.0 years; range, years). Forty-seven patients in the study group and 43 in the control group were older than 40 years of age. Twentysix patients in the study group and 28 in the control group had a history of shoulder arthroscopic surgery. History of shoulder trauma within 1 year was present in 13 patients in the study group and in 15 in the control group. Rotator cuff tears were present in 32 patients in the study group and in 33 in the control group, whereas labral tears were present in 19 patients in the study group and in 14 in the control group. pain ranged from 0 to 10 in both the study group (mean score, 3.7) and the control group (mean score, 3.7). Thirty patients in the study group and 29 in the control group had a baseline pain score of less than 4. Statistically, all the patient-related factors were homogeneously distributed between the study and the control groups (Table 1). Extraarticular leakage of the contrast agent was found in 29 patients in the study group and in 33 in the control group. The volume of the injected contrast mixture ranged from 5 group group A B Fig. 2 Mean pain score at each time phase in subgroups by age. A and B, Graphs show mean pain score in patients up to 40 years of age (A) and older than 40 years of age (B). to 17 ml (mean, 13.2 ml) in the study group and from 5 to 18 ml (mean, 1 ml) in the control group. Twenty-five patients in the study group and 20 in the control group were injected with a volume of contrast mixture of more than 13 ml. Statistically, all the procedure-related factors also were homogeneously distributed between the study and the control groups (Table 1). Pain Course According to Passage All patients The pain course according to time passage showed a quadratic trend, and it was not significantly different between the study and the control groups (Table 2). The mean pain score peaked at 2 hours after injection and returned to baseline level 2 days after injection in both the study and the control groups (Fig. 1). Subgroups The results of the subgroup analyses were similar to those found in the analysis of all patients (Table 2). In all subgroups except one, the pain course according to time passage showed a quadratic trend in both the study and the control groups, and it was not significantly different between the two groups. However, in the subgroup that had a history of shoulder trauma within 1 year, the pain course according to time passage showed a quadratic trend in the study group but a linear or cubic trend in the control group, and the pain course between the study and the control groups was significantly different. The mean pain score in a majority of subgroups peaked 2 hours after injection and in a minority of subgroups (control group of men, study and control groups of patients with shoulder trauma within 1 year, control group of patients who had baseline pain score of 4 or greater, and control group of AJR:200, April

5 Choo et al. group group Fig. 3 Graph shows mean pain score at each time phase in subgroup with history of shoulder trauma within 1 year. patients who were injected with a volume of contrast mixture of more than 13 ml) peaked immediately after injection (Figs. 2 4). Net Pain Score at Each Phase All patients The respective net pain scores of the study and control groups were 0.43 and 0.52 immediately after injection, 0.63 and 0.61 at 2 hours after injection, 0.35 and 0.16 at 1 day after injection, and 0.16 and 0.05 at 2 days after injection. The net pain score at each phase was not significantly different between the study and the control groups (Table 3). Subgroups In all the subgroups divided according to patient-related factors (age, sex, history of shoulder surgery, history of shoulder trauma within 1 year, rotator cuff tear, and baseline pain), the net pain score at each phase was not significantly different between the study and the control groups. In all the subgroups divided according to procedure-related factors (extraarticular contrast leakage, injected contrast volume, physician experience, and imaging modality), the net pain score at each phase also was not significantly different between the study and the control groups (Table 3). Discussion Thus far, only one study, conducted by Fox and colleagues [10] in 2012, has reported efficacy of intraarticular anesthesia in reducing arthrography-related pain. In their study, intraarticular local anesthesia was effective in reducing periprocedural pain associated with shoulder MR arthrography. Their result is opposite to the result evident in our study; we found that intraarticular local anesthesia did group group not reduce the pain related to MR or CT arthrography. Because the number of patients and the study design were very similar, it is uncertain why the efficacy of intraarticular local anesthesia was different between the two studies. One possible answer is the difference in the type and dose of the local anesthetic used in the two studies. Fox and colleagues injected 12 ml of contrast mixture containing 10 ml of 0.5% ropivacaine (volume and amount injected were 6 ml and 30 mg, respectively), whereas we injected approximately 13 ml of contrast agent mixed with 1.5 ml of 2% mepivacaine (volume and amount injected were 0.9 ml and 18 mg, respectively). In numerous previous studies [8, 12 16], ml of 2% mepivacaine, which was the actual intraarticularly injected dose, was used in MR arthrography of the shoulder. The intraarticularly injected dose of mepivacaine used in this study was determined on the basis of these previous studies. The use of intraarticular ropivacaine in arthrography was described in a recent report [9], and the dose used was similar to that used in the study by Fox and colleagues [10]. For a brachial nerve block, 20 ml each of 2% mepivacaine (400 mg) and of 0.5% or 1% ropivacaine (100 or 200 mg) were used in a previous study [17]. For pain relief after arthroscopic surgery of the knee, mg of ropivacaine was intraarticularly injected in another study [18]. With consideration of these doses of local anesthetics used in a nerve block or as an intraarticular analgesic after arthroscopy, the dose of mepivacaine used in our study was relatively small as compared with the dose of group group A Fig. 4 Mean pain score at each time phase in subgroups by modality. A and B, Graphs show mean pain score in patients who underwent shoulder MR arthrography (A) and CT arthrography (B). ropivacaine used in the study by Fox and colleagues [10]. Therefore, the difference in the injected dose of local anesthetic between our study and theirs may have caused the differences in local anesthetic efficacy that were observed between the two studies. The type of local anesthetic also may have affected the results of the two studies. However, until now, there has been no study on the type and dose of local anesthetic to effectively control arthrography-related pain. One detail that must be considered when choosing the type and dose of local anesthetic is a toxicity of local anesthetic to chondrocytes. This issue has been highlighted because postarthroscopy glenohumeral chondrolysis has been reported in patients who underwent continuous intraarticular infusion of bupivacaine [18]. Since then, several experimental studies have shown that local anesthetics are toxic to human and animal chondrocytes and that chondrocyte toxicity of local anesthetics is dose and time dependent [19, 20]. Several comparison studies have revealed that mepivacaine, ropivacaine, and levobupivacaine are less toxic to chondrocytes than bupivacaine and concluded that these local anesthetics may be safer than bupivacaine for intraarticular analgesia, although all local anesthetics were found to be toxic to chondrocytes [21 24]. These in vitro studies were designed to simulate continuous intraarticular infusion of local anesthetics at high concentration after arthroscopy; however, the effect of a single intraarticular injection of local anesthetic on articular cartilage has not been clearly shown. It is presumed that intraarticularly injected local anesthetics are B 864 AJR:200, April 2013

6 TABLE 3: Comparison of Net Pain Score in Each Phase Between and s Factor Injection Injection Injection Injection p p p p All patients Patient-related factors Age 40 years > 40 years Sex Male Female History of previous shoulder surgery History of shoulder trauma within 1 year Rotator cuff tears Supraspinatus-infraspinatus tendon tear Subscapularis tendon tear Labral tear pain < Procedure-related factors Contrast leakage Contrast volume 13 ml > 13 ml Level of difficulty High Low Physician experience Resident Specialist Modality CT arthrography MR arthrography Note Data are net pain score (i.e., amount of pain attributed to arthrography alone, calculated by the formula NetP A = P A P B, where NetP A is the net pain score in phase A, P A is the pain score during phase A, and P B is the pain score at baseline) except where otherwise indicated. AJR:200, April

7 Choo et al. unlikely to affect chondrocytes owing to the dilutional effects of joint fluid and active synovial absorption [25]. In one animal model, however, reduced chondrocyte density was seen 6 months after a single injection of local anesthetic [26]. Therefore, it is uncertain whether a single intraarticular injection of local anesthetic is totally safe from causing cartilage toxicity. For this reason, it is important to determine the effective and safe type of local anesthetic and its minimal effective dose in reducing arthrography-related pain. Further investigation is required. In the study by Fox and colleagues [10], the mean pain score of the study group, which used additional local anesthetic, was one unit lower on the pain scale than that of the control group. Because the maximal arthrography-related pain score increased by approximately one unit from the baseline pain score [16], an additional intraarticular injection of local anesthesia with 10 ml of 0.5% ropivacaine, as was used by Fox and colleagues [10], seems appropriate to control the pain. However, peak postarthrographic pain usually occurs 4 12 hours after arthrography [7, 8, 16], and the study by Fox and colleagues [10] assessed the pain score only until after MRI, which was approximately 30 minutes after the injection. Therefore, on the basis of the study by Fox and colleagues, it is unclear whether an additional intraarticular injection of 10 ml of 0.5% ropivacaine is effective in controlling the peak pain related to arthrography. Our study using intraarticular 2% mepivacaine at the dose routinely published in many articles assessed the pain score until 2 days after arthrography, and it was observed that the maximal pain related to arthrography was not controlled by the local anesthetic. As mentioned already, peak postarthrography pain occurs 4 12 hours after arthrography. However, in this study, this phase was not included to assess pain for several reasons. In the previous studies mentioning the phase of peak postarthrographic pain, all patients were injected with an intraarticular local anesthetic for shoulder arthrography [7, 8, 16]. The authors of one of these studies [8] explained that maximal pain was noted at 4 hours after the procedure possibly because the effect of the local anesthetic would have worn off 4 hours after it had been administered. The current study was designed to evaluate the effect of an intraarticular local anesthetic. The duration of action of mepivacaine is minutes, although the duration depends on the anesthetic dose and route of administration [27]. Therefore, to achieve the study goal, we obtained the pain score at 2 hours after the procedure, instead of at 4 hours or 12 hours after the procedure. Previous studies [28, 29] have suggested that postarthrographic pain may be due to chemical synovitis caused by contrast material. Delayed onset of postarthrographic pain can be explained by this theory because chemical synovitis is an inflammatory response that takes some time to develop. In an animal histologic examination [30], knees intraarticularly injected with an iodinated contrast agent had transient histologic changes, such as synovial hyperplasia and leukocyte infiltration, whereas those injected with a gadolinium agent had a normal appearance. Therefore, it is likely that the pain after CT arthrography is more severe than that after MR arthrography. However, in other studies of MR arthrography of the shoulder [7, 31], the level of pain or discomfort with MRI was significantly greater than that with arthrography itself owing to the noise, the feeling of confinement, and the need to remain still for a longer time. Therefore, it is likely that the level of pain or discomfort with CT is less than that during MRI because the imaging time for CT is much shorter than that for MRI and the noise during CT is less than that during MRI. This study included direct MR and CT arthrographies. Although we did not compare the difference in pain between direct MR and CT arthrographies, the pain course over time and the net pain score after the injection seemed to be similar for both imaging modalities. Another important issue is the potential impact of the anesthetic on imaging quality, particularly in the case of MRI. Montgomery and colleagues [32] studied the effects of iodinated contrast material and albumin on gadolinium enhancement and revealed that dilution of gadolinium in iodinated contrast material decreased signal intensity on T1- and T2-weighted and gradient-recalled echo images. The mixing of anesthetic with gadolinium could also affect the signal intensity of MRI, depending on the dose or type of anesthetics used. However, this issue is not relevant to the present study, and there are not yet any published reports addressing it in detail. Further studies are required. This study has several limitations. First, the pain scores at baseline and immediately after injection were obtained by a physician who knew the patients group assignments, which could have resulted in observer bias. However, the pain scores at the other three phases were obtained by another physician who was blinded to the patients assignments, and the patients also were blinded to their allocation. Second, the pain level was acquired only from patients who were eager to cooperate, which could result in a selection bias. Third, this study included patients who underwent either CT arthrography or MR arthrography of the shoulder. Nevertheless, CT arthrography showed a pain course similar to that of MR arthrography, and it did not affect the efficacy of the intraarticular local anesthetic. Fourth, the cutoff for the subgroup analysis according to age was 40 years. In a study by Saupe and colleagues [8], the group of patients who were younger than 30 years of age had more severe arthrography-related pain than did patients in other age groups. Therefore, a subgroup analysis using a cutoff age of 30 years could be more useful to evaluate the efficacy of intraarticular local anesthesia. However, in this study, this was not performed because of the small number of patients younger than 30 years of age. In conclusion, this study revealed that an intraarticular injection with 1.5 ml of 2% mepivacaine, which was routinely used in many earlier studies, did not reduce arthrography-related pain. Both the study and the control groups showed similar quadratic pain courses, with a peak in the mean pain score at 2 hours after injection. No patient- or procedure-related condition age, sex, history of shoulder arthroscopic surgery, baseline pain score, rotator cuff or labral tear, leakage of contrast agent, volume of injected contrast agent, level of difficulty of arthrography, physician experience, and imaging modality affected the efficacy of intraarticular mepivacaine. Further investigations are required in order to determine the type and minimal effective dose of local anesthetic to control arthrography-related pain. References 1. Flannigan B, Kursunoglu-Brahme S, Snyder S, Karzel R, Del Pizzo W, Resnick D. MR arthrography of the shoulder: comparison with conventional MR imaging. AJR 1990; 155: Peterson JJ, Bancroft LW. History of arthrography. Radiol Clin North Am 2009; 47: Waldt S, Metz S, Burkart A, et al. Variants of the superior labrum and labro-bicipital complex: a comparative study of shoulder specimens using MR arthrography, multi-slice CT arthrography and anatomical dissection. Eur Radiol 2006; 16: Fogerty S, King DG, Groves C, Scally A, Chandramohan M. Interobserver variation in reporting 866 AJR:200, April 2013

8 CT arthrograms of the shoulder. Eur J Radiol 2006; 186: Baker JF, Byrne DP, Walsh PM, Mulhall KJ. Hu- 2011; 80: Studler U, Kalberer F, Leunig M, et al. MR ar- man chondrocyte viability after treatment with 5. Choi JY, Kim SH, Yoo HJ, et al. Superior labral thrography of the hip: differentiation between an local anesthetic and/or magnesium: results from anterior-to-posterior lesions: comparison of exter- anterior sublabral recess as a normal variant and a an in vitro study. Arthroscopy 2011; 27: nal rotation and active supination CT arthrogra- labral tear. Radiology 2008; 249: Park J, Sutradhar BC, Hong G, Choi SH, Kim G. phy with neutral CT arthrography. Radiology 2012; 263: Hall FM, Goldberg RP, Wyshak G, Kilcoyne RF. Shoulder arthrography: comparison of morbidity after use of various contrast media. Radiology 1985; 154: Binkert CA, Zanetti M, Hodler J. Patient s assessment of discomfort during MR arthrography of the shoulder. Radiology 2001; 221: Saupe N, Zanetti M, Pfirrmann CW, Wels T, Schwenke C, Hodler J. Pain and other side effects after MR arthrography: prospective evaluation in 1085 patients. Radiology 2009; 250: Giaconi JC, Link TM, Vail TP, et al. Morbidity of direct MR arthrography. AJR 2011; 196: Fox MG, Petrey WB, Alford B, Huynh BH, Patrie JT, Anderson MW. Shoulder MR arthrography: intraarticular anesthetic reduces periprocedural pain and major motion artifacts but does not decrease imaging time. Radiology 2012; 262: Huskisson EC. Measurement of pain. Lancet 1974; 2: Knuesel PR, Pfirrmann CWA, Noetzli HP, et al. MR arthrography of the hip: diagnostic performance of a dedicated water-excitation 3D doubleecho steady-state sequence to detect cartilage lesions. AJR 2004; 183: Duc SR, Mengiardi B, Pfirrmann CW, Jost B, Hodler J, Zanetti M. Diagnostic performance of MR arthrography after rotator cuff repair. AJR 15. Zubler V, Mamisch-Saupe N, Pfirrmann CW, Jost B, Zanetti M. Detection and quantification of glenohumeral joint effusion: reliability of ultrasound. Eur Radiol 2011; 21: Steurer-Dober I, Rufibach K, Hodler J, et al. Do patients with structural abnormalities of the shoulder experience pain after MR arthrography of the shoulder? Radiology 2010; 256: Casati A, Fanelli G, Aldegheri G, et al. Interscalene brachial plexus anaesthesia with 0.5%, 0.75% or 1% ropivacaine: a double-blind comparison with 2% mepivacaine. Br J Anaesth 1999; 83: Petty DH, Jazrawi LM, Estrada LS, Andrews JR. Glenohumeral chondrolysis after shoulder arthroscopy: case reports and review of the literature. Am J Sports Med 2004; 32: Chu CR, Izzo NJ, Coyle CH, Papas NE, Logar A. The in vitro effects of bupivacaine on articular chondrocytes. J Bone Joint Surg Br 2008; 90: Anz A, Smith MJ, Stoker A, et al. The effect of bupivacaine and morphine in a coculture model of diarthrodial joints. Arthroscopy 2009; 25: Piper SL, Kim HT. Comparison of ropivacaine and bupivacaine toxicity in human articular chondrocytes. J Bone Joint Surg Am 2008; 90: Baker JF, Walsh PM, Byrne DP, Mulhall KJ. In vitro assessment of human chondrocyte viability after treatment with local anaesthetic, magnesium sulphate or normal saline. Knee Surg Sports Traumatol Arthrosc 2011; 19: Comparison of the cytotoxic effects of bupivacaine, lidocaine, and mepivacaine in equine articular chondrocytes. Vet Anaesth Analg 2011; 38: Piper SL, Kramer JD, Kim HT, Feeley BT. Effects of local anesthetics on articular cartilage. Am J Sports Med 2011; 39: Chu CR, Coyle CH, Chu CT, et al. In vivo effects of single intra-articular injection of 0.5% bupivacaine on articular cartilage. J Bone Joint Surg Am 2010; 92: Moore PA, Hersh EV. Local anesthetics: pharmacology and toxicity. Dent Clin North Am 2010; 54: Newberg AH, Munn CS, Robbins AH. Complications of arthrography. Radiology 1985; 155: Peterson C, Hodler J. Adverse events from diagnostic and therapeutic joint injections: a literature review. Skeletal Radiol 2011; 40: Papacharalampous X, Patsouris E, Mundinger A, et al. The effect of contrast media on the synovial membrane. Eur J Radiol 2005; 55: Blanchard TK, Bearcroft PW, Dixon AK, et al. Magnetic resonance imaging or arthrography of the shoulder: which do patients prefer? Br J Radiol 1997; 70: Montgomery DD, Morrison WB, Schweitzer ME, Weishaupt D, Dougherty L. Effects of iodinated contrast and field strength on gadolinium enhancement: implications for direct MR arthrography. J Magn Reson Imaging 2002; 15: AJR:200, April