The glenohumeral joint has a multitude of potential pain-generating pathologies. These include a variety of arthritides, adhesive capsulitis, labral tears, biceps tendinitis, rotator cuff tears, and impingement. Glenohumeral joint injections are routinely performed for both therapeutic and diagnostic purposes in patients with shoulder pain. Injections have been shown to improve both patient outcomes and satisfaction.1
The two most commonly described techniques for glenohumeral joint injection are the anterior and posterior approaches. The accuracies of these approaches have varied widely in the literature.1–7 More recently, however, a third approach to the shoulder has been explored: the superomedial, or Neviaser, approach (Figure 1).8,9
Front view (A) and top-down view (B) clinical photographs of a right shoulder with superimposed radiographs demonstrating the Neviaser (superomedial) portal indicated by the “X.” The portal is located in the “soft spot” approximately 1 cm posterior to the acromioclavicular joint, aimed slightly lateral and anterior.
Tobola et al8 used a 2-inch, 21-gauge needle and compared the posterior, anterior, and Neviaser approaches for glenohumeral joint injection using post-injection fluoroscopic confirmation. They found no significant difference in successful intra-articular placement of the injection when comparing the 3 approaches. Later, Chernchujit and Zonthichai9 assessed the Neviaser approach to the shoulder by comparing it with the anterior approach. They used a 3.5-inch, 23-gauge needle for the injections and confirmed accurate placement arthroscopically. They found the Neviaser approach to be superior in accurately entering the glenohumeral joint as compared with the anterior approach.
These recent studies beg the following question: Is it the approach, the needle length, or both that leads to the varying degree of success of intra-articular injection reported in the literature? Tong et al10 arthroscopically compared the depth from skin to joint capsule of both the anterior and the posterior approach to the shoulders. The anterior approach led to joint penetration 98.7% of the time, and the posterior approach produced a 19% success rate using a standard 1.5-inch (40-mm) needle. Chernchujit and Zonthichai9 commented on needle depth in their arthroscopic study, reporting that the average depth of skin to capsule was 42±7 mm and 32±7 mm for the Neviaser and anterior approaches, respectively.
The purpose of this study was to establish the needle length required to penetrate the joint capsule via the Neviaser approach based on radiography and magnetic resonance imaging (MRI) measurements. Additionally, the authors sought to identify any correlation between required needle length and a patient's body mass index (BMI). They hypothesized that (1) there would be a positive correlation between a patient's BMI and the required needle length for the Neviaser approach and (2) a needle length of greater than 2 inches may be required to reliably access the joint space.
Materials and Methods
The authors retrospectively reviewed consecutive patients evaluated by a single upper extremity surgeon (B.R.G.) at their institution. Inclusion criteria included a true anteroposterior (AP) radiograph of the shoulder, an MRI within 1 year of radiography of the same shoulder, age younger than 50 years, no previous surgery on the evaluated shoulder, and no history of acromioclavicular pathology. There were 101 patients included in the final analysis.
Using a digital imaging system (Intel-liSpace PACS, Philips Medical Systems, Best, The Netherlands), the intended path of the needle as taken via the Neviaser (superomedial) portal was measured on both the radiography and MRI studies. The authors then established a proxy for an intra-articular space that could be observed on both radiographs and MRI, which they termed the superior triangle, defined as a triangular space made when a line is drawn from the superior aspect of the glenoid to the most superior aspect of the humeral head. Tangential lines were then drawn along the humeral head and along the glenoid (Figure 2). The coronal MRI section that captured the apex of the humeral head determined the image from which MRI measurements were taken. A measurement was then taken from the level of the skin to the top of the superior triangle and recorded for the radiography and MRI studies for each individual patient (Figures 3–4).
The superior triangle as seen on an anteroposterior radiograph of the shoulder.
Measurement of the Neviaser portal on a T2 coronal magnetic resonance image from the skin to the superior triangle.
Measurement of the Neviaser portal from the skin to the superior triangle on anteroposterior radiograph of the shoulder.
Univariate analysis was performed for patient demographics, BMI, and radiographic findings. Fisher's exact test was performed for categorical variables when appropriate, and the Student's t test was performed for continuous variables. Spearman's rho was used to determine correlations. Results were reported as means±SDs, odds ratios, 95% confidence intervals, and P values. A post hoc power analysis was performed to minimize beta error. Statistical significance was denoted by P≤.05. Analysis was conducted using SPSS software version 22.0 (IBM Corp, Armonk, New York).
A total of 101 consecutive shoulders meeting the inclusion criteria were selected for this study. There were 58 (57.4%) male and 43 (42.6%) female patients, with an average BMI of 31.2 kg/m2.
The average skin-to-capsule measurement as determined on MRI was 3.9 cm (range, 2.9–5.98 cm). The average skin-to-capsule measurement as determined on radiograph was 4.27 cm (range, 3.01–5.56 cm) (Figure 5).
A perfect correlation of radiography (XR) to magnetic resonance imaging (MRI) is demonstrated by the red line in this graph. As this graph depicts, radiography had a tendency to measure greater skin-to-capsule measurements compared with MRI. Inset within the graph gives the means and SDs of skin-to-capsule measurements for both radiography and MRI.
Correlation coefficients for BMI and measurements were found for both MRI and radiography. The MRI cohort showed a moderate positive correlation between BMI and measured distance from skin to capsule with a correlation coefficient of 0.53 (P<.0001). Radiographs yielded a weak positive correlation between BMI and measured distance from skin to capsule with a correlation coefficient of 0.36 (P=.0002) (Figure 6).
Plot demonstration of magnetic resonance imaging (MRI) and radiography (XR) values for skin-to-capsule measurements and body mass index (BMI). Correlation coefficients are illustrated by the red and blue lines.
The Neviaser approach to intra-articular injection is relatively easy to identify with palpable landmarks of the shoulder girdle, and, in the authors' experience, it can reliably produce clinical improvement for patients indicated for intra-articular glenohumeral injection. It is also the authors' experience that glenohumeral injections performed through the superomedial approach are typically less painful for patients during the procedure than other approaches. Despite the potential clinical advantages of using the superomedial approach for injections of the glenohumeral joint, this approach is underreported in the literature. Two major predictors of a successful intra-articular injection from this approach are the depth of the capsule from the skin and the length of the needle used for injection. To the authors' knowledge, this is the first study to report the distance from skin to capsule as measured from MRI and radiographs as a primary measurement of needle length requirement for glenohumeral joint access via the Nevaiser approach.
Although there are multiple gaps in the literature, available evidence suggests that glenohumeral injection accuracy is poor in the clinic setting.7,8 For this reason, some authors have recommended that patients with shoulder pathology indicated for intra-articular injection be referred for injection using image guidance (fluoroscopy, computed tomography, or ultrasound) during their glenohumeral joint injection. Outsourcing intra-articular glenohumeral joint injections is suboptimal for several reasons. From the patients' standpoint, they must now sacrifice time to attend an additional appointment and incur the costs of this consultation. If a reliable technique for intra-articular glenohumeral injection was known to the general practicing orthopedist, glenohumeral injections could be reliably performed in the office without the need for additional imaging or outsourcing. This study makes gains toward this goal by reporting the average radiograph and MRI skin-to-capsule depth, which can be used as a measure for average needle length required for reliable access to the glenohumeral joint via the Neviaser approach.
On the basis of MRI and radiograph measurements, the authors found that the average skin-to-capsule measurements were 3.9 cm and 4.27 cm, respectively. When combining all data points, the average skin-to-capsule depth measured was 4.09 cm.
Given that the most commonly used needle lengths for intra-articular glenohumeral injections are 1.5 in and 2 in, this study suggests that 66.5% of intraarticular injections will not penetrate the joint capsule using a 1.5-inch needle via the Neviaser portal, whereas only 8.5% of injections will not reach the capsule with a 2-inch needle.
When taking BMI into account, MRI demonstrates a moderately positive correlation between BMI and needle length requirement, and radiography shows a weak positive correlation. Differences in these imaging modalities are likely related to patient positioning and the effect of gravity on the shoulder girdle during the respective imaging studies. True AP glenohumeral radiographs are more prone to the effects of gravity because they are performed in the sitting or standing position, whereas MRIs are performed in the supine position, minimizing the role of gravity. Body mass index, as a ratio of a patient's height and weight, gives a good impression of overall patient size. However, BMI does not provide insight into where patients carry their weight. In the authors' experience, it is rare that excess weight is carried in the superior shoulder girdle. This is consistent with their findings of a stronger correlation between radiography and BMI than between MRI and BMI.
Limitations of this study were that it involved a retrospective chart review and there was no blinding to BMI and corresponding needle depth measurements. Additionally, BMI alone was analyzed as a correlate for increasing skin-to-capsule depths. There may be other measurements to examine, such as acromioclavicular morphology or skin-to-clavicle depth. Finally, there were no data to verify the clinical validity of these measurements. Further research is currently under way to help shed light on the clinical relevance of the findings of this study.
The results of this study suggest that a minimum of a 2-inch needle be used when performing intra-articular injections in the clinic setting via the Neviaser approach to reliably gain access to the glenohumeral joint. Furthermore, the authors' initial hypothesis was that BMI would play a more integral role in depth of skin to capsule. These data suggest that based on MRI measurements, as the patient's BMI increases, the clinician should use a longer needle to reliably gain access to the glenohumeral joint. This does not seem to be the case, however, in the sitting or standing position as measured from radiographs.
Further investigation is currently under way that will include gadolinium-enhanced intra-articular injections via the Neviaser portal with postinjection radiographs to assess the accuracy and reliability of this approach. Additionally, clinical outcome data will be collected prospectively to examine the relationship between successful intra-articular injections and patient outcomes.
- Patel DN, Nayyar S, Hasan S, Khatib O, Sidash S, Jazrawi LM. Comparison of ultrasound-guided versus blind glenohumeral injections: a cadaveric study. J Shoulder Elbow Surg. 2012;21(12):1664–1668. doi:10.1016/j.jse.2011.11.026 [CrossRef] PMID:22445159
- Raeissadat SA, Rayegani SM, Langroudi TF, Khoiniha M. Comparing the accuracy and efficacy of ultrasound-guided versus blind injections of steroid in the glenohumeral joint in patients with shoulder adhesive capsulitis. Clin Rheumatol. 2017;36(4):933–940. doi:10.1007/s10067-016-3393-8 [CrossRef] PMID:27566475
- Aly AR, Rajasekaran S, Ashworth N. Ultrasound-guided shoulder girdle injections are more accurate and more effective than landmark-guided injections: a systematic review and meta-analysis. Br J Sports Med. 2015;49(16):1042–1049. doi:10.1136/bjsports-2014-093573 [CrossRef] PMID:25403682
- McFarland E, Bernard J, Dein E, Johnson A. Diagnostic injections about the shoulder. J Am Acad Orthop Surg. 2017;25(12):799–807. doi:10.5435/JAAOS-D-16-00076 [CrossRef] PMID:29176503
- Esenyel CZ, Ozturk K, Demirhan M, et al. Accuracy of anterior glenohumeral injections: a cadaver study. Arch Orthop Trauma Surg. 2010;130(3):297–300. doi:10.1007/s00402-008-0811-7 [CrossRef] PMID:19139912
- Axe JM, Axe MJ. 97% accuracy of intraarticular glenohumeral injection with a modified (Delaware) posterior bone touch technique. Del Med J. 2013;85(10):303–306. PMID:24475646
- Sethi PM, El Attrache N. Accuracy of intra-articular injection of the glenohumeral joint: a cadaveric study. Orthopedics. 2006;29(2):149–152. doi:10.3928/01477447-20060201-01 [CrossRef] PMID:16485459
- Tobola A, Cook C, Cassas KJ, et al. Accuracy of glenohumeral joint injections: comparing approach and experience of provider. J Shoulder Elbow Surg. 2011;20(7):1147–1154. doi:10.1016/j.jse.2010.12.021 [CrossRef] PMID:21493103
- Chernchujit B, Zonthichai N. Comparison of accuracy of anterior and superomedial approaches to shoulder injection: an experimental study. SICOT J. 2016;2(13):13. doi:10.1051/sicotj/2015044 [CrossRef] PMID:27163102
- Tong A, Harding R, Graham G. Glenohumeral joint penetration with a 21-gauge standard needle. J Shoulder Elbow Surg. 2012;21(12):e1–e3. doi:10.1016/j.jse.2011.11.034 [CrossRef] PMID:22459266