Abstract
Purpose:
To systematically review available literature comparing location and safety of two common anteromedial portals with nearby neurovascular structures in cadaveric models and to determine the correct positioning and preparation of the joint prior to elbow arthroscopy.
Methods:
The review was devised in accordance with PRISMA guidelines. Inclusion criteria consisted of original, cadaveric studies performed by experienced surgeons on male or female elbows evaluating anteromedial portal placement with regards to proximity of the arthroscope to neurovascular structures. Exclusion criteria consisted of case reports, clinical series, non-English and non-cadaveric studies. Statistical analysis was done to measure reviewer reliability after scoring of each study.
Results:
2,596 studies were identified during screening and 10 studies met final inclusion as original, cadaveric investigations of anteromedial portal proximity to neurovascular structures. The difference in distance between proximal and distal portals was less than 1mm for the brachial artery and less than 1.5mm for the medial antebrachial cutaneous nerve, whereas the ulnar nerve was 4.17mm further from the distal portal and the median nerve was 5.07mm further from the proximal portal. Joint distension increased the distances of neurovascular structures to portal sites with the exception of the ulnar nerve in distal portals. Elbow flexion to 90 degrees increased distances of all neurovascular structures to portal sites.
Conclusion:
The results show proximal anteromedial portals puts less structures at risk compared to the distal portal. Elbows in 90-degrees flexion with joint distension carry a lower risk of neurovascular injury during portal placement. These findings suggest the proximal anteromedial portal to be the safer technique in anteromedial arthroscopy of the elbow.
Clinical Relevance:
Discrepancies in placement of portals have existed in literature indicating differing safety margins regarding surrounding neurovascular anatomy. The present study aims to link the literature-based evidence together in order to describe the safest anteromedial portal variation.
Introduction
In the field of elbow arthroscopy, physicians face many challenges that centralize around the safety and complexity of accessing the joint capsule and preserving neurovascular structures in target joints. Within the limited space of the elbow lies the ulnar, radial, and median nerves, many cutaneous nerves, and major blood vessels - within range of potential injury from arthroscopic instrumentation. Previous notions existed in the early 20th century that elbow arthroscopy was too dangerous and impractical given risks to surrounding neurovascular anatomy.1,2 However, there have been well documented reviews of diagnostic and procedural techniques associated with elbow arthroscopy since the 1980s, which all detail the importance of maintaining adequate safety in what is considered to be one of the most high-risk arthroscopic procedures routinely performed.3,4
One of the earliest reviews by Andrews and Carson concluded the need for extreme delicacy during arthroscopic inspection of the elbow compared to more accessible knee joints since instrumentation must pass through deeper muscular layers and in close proximity to neurovascular structures.3 Over the past few decades, multiple authors have described the risk and benefit of certain arthroscopic portal locations including the anteromedial, anterolateral, and posterior approaches. However, many discrepancies exist among upper extremity surgeons pertaining to the proper location and safety of these portals. These discrepancies are due to the multitude of techniques and approaches utilized in existing cadaveric studies, which were done with different portal locations, variable levels of joint distension, and varying elbow positioning.3–12
The purposes of this study were to systematically review available literature comparing the location and safety of two common anteromedial portals with nearby neurovascular structures in a cadaveric model and to determine the correct positioning and preparation of the joint prior to elbow arthroscopy. Additionally, we aimed to determine the correct positioning and preparation of the joint prior to elbow arthroscopy. We hypothesize that both proximal and direct anteromedial portals, coupled with joint distension and elbow flexion, provide equal risk to surrounding neurovascular structures of the elbow.
Methods
Design, Eligibility Criteria and Search Strategy
An initial search for literature was performed using the PubMed, EMBASE, Cochrane Database of Systematic Reviews, Cochrane Central Register of Controlled Trials, Mendeley, and Google Scholar databases according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines (Figure 1).13 Last search queries were completed in November 2018. The search terms within each database included combinations of “elbow”, “elbow arthroscopy”, “arthroscopy”, “portal”, “anteromedial”, “cadaver”, and “anteromedial portal”. Inclusion criteria consisted of original, cadaveric studies performed by surgeons experienced in elbow arthroscopy evaluating the placement of the anteromedial portal on male and female cadaveric elbows with regards to proximity of the arthroscope to neurovascular structures and safety profile. Exclusion criteria consisted of case reports, clinical series, non-English and non-cadaveric studies. No restrictions were placed on publication date or status.
Figure 1:
Flow diagram illustrating the literature search process
Study Screening and Data Abstraction
Four independent investigators, including one medical student, one orthopedic resident and two orthopedic surgeons, qualified each manuscript that met final inclusion for reference to the present study. Two senior authors, FS and MO, fellowship trained orthopedic surgeons, were the final decision makers for the inclusion of any manuscripts referenced in our study. After this qualifying period, 10 studies met final inclusion into the present study.3–12 Discrepancies in study inclusion were resolved by reviewer consensus and agreement. Two independent reviewers, TC and ZF, one medical student and one orthopedic resident, performed data abstraction and included the following variables: study characteristics (author, year), number of cadavers used, and portal location relative to neurovascular anatomy, joint distension, and arm positioning. Variables were collected and recorded in a Microsoft Excel (2017) worksheet. Results were obtained for each variable where reported and all were systematically compared to each other (Tables 1, 2 and 3). The primary outcome assessed in each cadaveric dissection was the distance between neurovascular anatomy and the path of the arthroscope (Table 2). Some studies incorporated additional variables such as forearm and elbow positioning when portals were placed. These results were added to the comparison and their values reported as a separate entity (Table 3).
Table 1:
Information for each included study, neurovascular structures and portals investigated, elbow position variations and use of distention. QUACS scores for all included studies determined by two independent raters, with individual scores listed for each of the 13 QUACS criteria.18
| Authors, Year | Number of Elbows Dissected | Neurovascular Structures Investigated | Elbow Position Variations | Distension (Y/N/Both) | Type of Portal Investigated (Distal, Proximal, or Both) | Total QUACS Score: | Maximum Possible QUACS Score: | QUACS (%) | |
|---|---|---|---|---|---|---|---|---|---|
| Unlu et al.,9 2006 | 20 | Ulnar n., Median n., Brachial a., MABCN |
Y | Y | Both | Rater 1 | 11 | 13 | 84.62 |
| Rater 2 | 12 | 13 | 92.31 | ||||||
| Lynch et al.,4 1986 | 5 | Median n., Brachial a., MABCN |
N | Both | Distal | Rater 1 | 9 | 12 | 75.00 |
| Rater 2 | 9 | 12 | 75.00 | ||||||
| Zonno et al.,11 2010 | 8 | Ulnar n. | N | Y | Proximal | Rater 1 | 11 | 12 | 91.67 |
| Rater 2 | 11 | 12 | 91.67 | ||||||
| Marshall et al.,12 1992 | 20 | Median n. | N | N | Distal | Rater 1 | 6 | 12 | 50.00 |
| Rater 2 | 6 | 12 | 50.00 | ||||||
| Stothers et al.,8 1995 | 12 | Ulnar n., Median n., Brachial a., MABCN |
Y | N | Both | Rater 1 | 8 | 12 | 66.67 |
| Rater 2 | 8 | 12 | 66.67 | ||||||
| Lindenfeld et al.,5 1990 | 6 | Ulnar n., Brachial a. |
N | Y | Proximal | Rater 1 | 8 | 12 | 66.67 |
| Rater 2 | 8 | 12 | 66.67 | ||||||
| Adolfsson et al.,7 1994 | 14 | Ulnar n., Median n., Brachial a., MABCN |
N | Y | Both | Rater 1 | 9 | 12 | 75.00 |
| Rater 2 | 8 | 12 | 66.67 | ||||||
| Verhaar et al.,6 1991 | 5 | Ulnar n., Median n., Brachial A. |
N | N | Distal | Rater 1 | 7 | 12 | 58.33 |
| Rater 2 | 7 | 12 | 58.33 | ||||||
| Chaware et al.,10 2016 | 12 | Ulnar n., Median n., MABCN |
Y | Y | Both | Rater 1 | 12 | 13 | 92.31 |
| Rater 2 | 12 | 13 | 92.31 | ||||||
| Andrews et al.,3 1985 | Unknown | Median n. | N | N | Distal | Rater 1 | 4 | 12 | 33.33 |
QUACS, Quality Appraisal for Cadaveric Studies; MABCN, medial antebrachial cutaneous nerve; n., nerve; a., artery
Table 2:
The average measured distances to the ulnar nerve, medial nerve, brachial artery and median antebrachial cutaneous nerve (MABCN) reported at each portal site for each study. The average distances and associated standard deviations from neurovascular structure to portal site between all studies, studies that only included joint distension, and studies that did not include joint distension. Standard deviations (reported in table as ± standard deviation) and ranges of measurements reported in each study were included, if available. Of note, the values reported in the table are with the elbow in 90-degrees of flexion wherever applicable to standardize comparison. w/o = measured without joint distention; w = measured with joint distention. There was no standard deviation or range to report from the Andrews and Carlson3 study.
| Average Distance to Ulnar N. (if reported) (mm) | Average Distance to Median N. (if reported) (mm) | Average Distance to Brachial A. (if reported) (mm) | Average Distance to MABCN (if reported) (mm) | |||||
|---|---|---|---|---|---|---|---|---|
| Study | Distal | Proximal | Distal | Proximal | Distal | Proximal | Distal | Proximal |
| Andrews and Carson (w/o) | 6 | |||||||
| Lynch et al (w and w/o) | 4 (range 3–10mm) (w/o), 14 (w) | 9 (range 8–13mm) (w/o), 17 (w) | 1 (range 1–9mm) (w) | |||||
| Lindenfeld (w) | 23.7 ± 1.63 | 22.3 ± 1.63 | ||||||
| Verhaar et al (w/o) | > 25 | 18 (range 12–25mm) | 26 (range 21–32mm) | |||||
| Adolfsson et al (w) | 24 (range 19–26mm) | 21 (range 19–24mm) | 14 (range 11–18mm) | 19 (range 16–22mm) | 22** | 4 (range 2–11mm) | 6 (range 3–12mm) | |
| Stothers et al (w/o) | 12 (range 7–18mm) | 7 (range 5–13mm) | 12.4 (range 7–20mm) | 15.2 (8–20mm) | 18 (12–25mm) | 1 (range 0–5mm) | 2.3 (range 0–9mm) | |
| Unlu et al* (w) | 25.4 ± 1.7 | 20 ± 2.2 | 16.2 ± 4.4 | 17.1 ± 2.8 | 20.3 ± 3.4 | 21.1 ± 3.3 | 9.2 ± 3.9 | 7.9 ± 2.5 |
| Chaware et al (w) | 16.03 ± 4.86 | 13.16 ± 3.73 | 22.12 ± 6.64 | 19.45 ± 7.42 | 4.99 ± 5.03 | 5.14 ± 5.08 | ||
| Marshall et al (w/o) | 6.5 ± 3.3 | |||||||
| Zonno et al (w) | 20.8 (range 14.4–25.1mm) | |||||||
| Average Distance and Standard Deviation (All Studies) | 22.61 ± 4.42 | 18.44 ± 4.72 | 12.98 ± 5.95 | 18.05 ± 3.67 | 20.1 ± 4.24 | 19.55 ± 2.19 | 4.04 ± 3.39 | 5.34 ± 2.33 |
| Average Distance and Standard Deviation (w/ joint distension) | 21.81 ± 5.05 | 19.73 ± 4.28 | 16.58 ± 3.84 | 19.46 ± 2.15 | 19.77 ± 2.54 | 21.1 ± 0 | 4.8 ± 3.39 | 6.35 ± 1.41 |
| Average Distance (w/o joint distension) | 25 ± 0 | 12 ± 0 | 8.3 ± 5.54 | 12.4 ± 0 | 16.73 ± 8.6 | 18 ± 0 | 1 ± 0 | 2.3 ± 0 |
best results from the study in full flexion and neutral pronation/supination rotation9
reported as 8mm further from the mean median nerve distance (14mm) to the distal portal, which therefore was assumed to equal 22mm7
MABCN, medial antebrachial cutaneous nerve; n., nerve; a., artery; w/o, measured without joint distention; w, measured with joint distention.
Table 3:
Three studies that reported measurements from the distal and proximal anteromedial portal to neurovascular structures under settings of various flexion/extension and pronation/supination positions.8–10
| Distal Portal (mm) | |||||
|---|---|---|---|---|---|
| Unlu et al | |||||
| 90 flex+neutral | extension+neutral | full flexion+neutral | 90 flex+supination | 90 flex+pronation | |
| median n. | 12.9 ± 4.6 | 7.9 ± 4.6 | 16.2 ± 4.4 | 13.9 ± 4.3 | 15.1 ± 4.4 |
| ulnar n. | 22.1 ± 2.0 | 17.7 ± 1.6 | 25.4 ± 1.7 | 22.5 ± 1.9 | 22.6 ± 1.9 |
| MABCN | 8.9 ± 3.8 | 8.8 ± 3.4 | 9.2 ± 3.9 | 9.1 ± 3.6 | 9.2 ± 3.8 |
| brachial a. | 16.6 ± 3.6 | 11.5 ± 3.4 | 20.3 ± 3.4 | 17 ± 3.5 | 18.4 ± 3.9 |
| Chaware et al | |||||
| 90 flex+neutral | extension+neutral | full flexion+neutral | 90 flex+supination | 90 flex+pronation | |
| median n. | 22.12 ± 6.64 | 24.77 ± 6.56 | |||
| ulnar n. | 16.03 ± 4.86 | 14.56 ± 5.49 | |||
| MABCN | 4.99 ± 5.03 | 5.35 ± 4.94 | |||
| Stothers et al | |||||
| 90 flex+neutral | extension+neutral | full flexion+neutral | 90 flex+supination | 90 flex+pronation | |
| median n. | 7 | 2.1 | |||
| brachial a. | 15.2 | 9.6 | |||
| Proximal Portal (mm) | |||||
| Unlu et al | |||||
| 90 flex +neutral | extension+neutral | full flexion+neutral | 90 flex+supination | 90 flex+pronation | |
| median n. | 13.8 ± 2.6 | 8.9 ± 2.8 | 17.1 ± 2.8 | 13.8 ± 2.6 | 14.1± 2.5 |
| ulnar n. | 16.2 ± 2.2 | 12.2 ± 2.1 | 20 ± 2.2 | 16.2 ± 2.2 | 16.2 ± 2.4 |
| MABCN | 7 ± 2.6 | 2.9 ± 1.8 | 7.9 ± 2.5 | 7 ± 2.6 | 6.7 ± 2.3 |
| brachial a. | 17.6 ± 3.3 | 12.3 ± 3.3 | 21.1 ± 3.3 | 17.6 ± 3.3 | 17.9 ± 3.1 |
| Chaware et al | |||||
| 90 flex +neutral | extension+neutral | full flexion+neutral | 90 flex+supination | 90 flex+pronation | |
| median n. | 19.45 ± 7.42 | 20.69 ± 7.16 | |||
| ulnar n. | 13.16 ± 3.73 | 12.84 ± 4.60 | |||
| MABCN | 5.14 ± 5.08 | 5.54 ± 5.68 | |||
| Stothers et al | |||||
| 90 flex +neutral | extension+neutral | full flexion+neutral | 90 flex+supination | 90 flex+pronation | |
| median n. | 12.4 | 7.6 | |||
| brachial a. | 18 | 16.6 | |||
flex, flexion; MABCN, medial antebrachial cutaneous nerve; n., nerve; a., artery; 90, 90 degrees
Quality Assessment and Data Analysis
Quality assessment of all included literature was performed using the Quality Appraisal for Cadaveric Studies (QUACS) scoring method.14 QUACS is used exclusively for studies that performed cadaveric dissections. Each of the 13 criteria, if applicable, is given a score of either a 0 or 1, with a maximum score of 13. Studies that do not contain statistical methodology can only receive a maximum score of 12. Two independent reviewers recorded ratings for each included study and the total QUACS score (%) was determined. Studies with a QUACS score of 80% or greater were considered high-quality, studies with QUACS scores of 51–79% were considered moderate-quality and those with QUACS scores of 50% or below were considered low-quality. Discrepancies between scores were discussed and if needed, the senior author were consulted for direction. The Intraclass Correlation Coefficient (ICC) measure of reliability was calculated based on QUACS scores. ICC ranges from 0 to 1 where a value of 1 indicates excellent interrater reliability.
Results
A total of 2,596 studies were identified after searching literature databases (Figure 1). After duplicate articles were eliminated, a total of 2,231 articles remained for review. 2,123 of these studies were non-cadaveric and excluded. 108 full text articles were then assessed for eligibility based on title and abstract. 98 of these potentially eligible studies were excluded because they failed to address distances and safety profiles between the anteromedial elbow portal and neurovascular anatomy. Overall, 10 articles met inclusion criteria for assessment of the anteromedial portal.3–12 Across all studies that met inclusion criteria, there were a total of 102 cadavers dissected (Table 1). However, one of these studies by Andrews and Carlson originally described their clinical experience with 12 patients followed by a cadaveric study of an undisclosed number3, which would therefore increase the sample size, if reported.4–12 The 10 included studies were published between 1985 and 2016 (Table 1). Study quality was determined from average QUACS scores reported by the two independent reviewers (Table 1). As a result, there were two low-quality studies,3,12 five medium-quality studies,4,5,6,7,8 and three high-quality studies.9,10,11 There was considerable agreement between the reviewers during quality assessment (ICC of 0.98).
There were five separate definitions identified for anteromedial elbow portals (Figure 2), each grouped into two main categories according to their location relative to the medial epicondyle: the distal and proximal portal. Andrews and Carson originally described the first anteromedial portal as a distal approach.3 Poehling et al later described the first proximal anteromedial portal.16 Adolffson et al7, Stothers et al8, Unlu et al9 and Chaware et al10 were the four studies that compared both the distal and proximal anteromedial portals.
Figure 2:
Image illustrating the 5 different anteromedial portal locations identified during literature review. A, two versions identified of the 2cm proximal portal. B, 2cm proximal and 1cm anterior portal. C, 2cm distal and 2cm anterior portal. D, 2cm distal and 1cm anterior portal.
Portal Categories, Definitions, and Locations
Distal AM Portal Placement
Eight articles reported the placement of the distal anteromedial portal. Seven of the eight studies reported using the original definition of the distal portal as described by Andrews and Carson3,4,7,8,9,10,12, 2cm distal and 2cm anterior to the medial epicondyle. Verhaar et al6 reported use of a 2cm distal and 1cm anterior portal.
Proximal AM Portal Placement
Six articles reported using a proximal anteromedial approach. Of these six studies, four compared the distal and proximal anteromedial approaches. The earliest definition of the proximal anteromedial portal appears to be established by Linfield et al5, describing it as 1cm proximal and 1cm anterior to the medial epicondyle. Four studies report the use of this original definition of the proximal portal5,7,9,11 and two studies report the use of a 2cm proximal anteromedial portal, with and without specified anterior positioning.8,10
Average distance to Neurovascular Structures
Proximal vs. Distal Portal Technique
The difference in distance between proximal and distal portals was less than 1mm for the brachial artery and less than 1.5mm for the median antebrachial cutaneous nerve (MABCN, whereas the ulnar nerve was 4.17mm further from the distal portal and the median nerve was 5.07mm further from the proximal portal. Across all studies, the median nerve (18.05 ± 3.67mm) and MABCN (5.34 ± 2.33mm) averaged the furthest distance from the proximal anteromedial portal. The ulnar nerve (22.61 ± 4.42mm) and brachial artery (20.1 ± 4.24mm) averaged the furthest distance from the distal portal. A standard deviation of 0mm occurred when only one study reported an average measurement for a certain portal distance to neurovascular structures (Table 2).
Effect of Joint Distention
High variability existed when measuring the distances from the anteromedial portal to neurovascular structures. This variability may be caused by differences in study design and direction of the surgeon’s portal placement toward the joint capsule. One clear difference between these studies was the utilization or absence of joint distension. Four studies performed measurements without distending the joint capsule,3,4,8,12 Five studies utilized joint distension,5–9,11 and one study compared measurements between distended and non-distended joint capsules (Table 2).4 Verhaar et al6 did not utilize joint distension, however they produced measurements comparable to those in other studies with joint distension. Stothers et al8 did not identify differences in measurements between distended and non-distended joints, explicitly stating that distension had minimal effects compared to non-distension.
In a distended joint, the ulnar nerve (21.81 ± 5.05mm) averaged the furthest distance from distal portals and the median nerve (19.46 ± 2.15mm), brachial artery (21.0 ± 0mm) and MABCN (6.35 ± 1.41mm) averaged the furthest distance from proximal portals. In a non-distended joint, the ulnar nerve (25 ± 0mm) averaged the furthest distance from distal portals and the median nerve (12.4 ± 0mm), brachial artery (18 ± 0mm) and MABCN (2.3 ± 0mm) averaged the furthest distance from proximal portals (Table 2). When comparing measurements between distended and non-distended joints, all distal and proximal portal safety margins were largest in distended joints, except for the distance from the distal portal to the ulnar nerve. However, Verhaar et al6 was the only study to measure the distance between the distal portal and ulnar nerve without joint distension, measuring >25mm from the distal portal site. Further studies are necessary to confirm these measurements for a more thorough understanding of the benefits of joint distension in ulnar nerve position.
Insertion with forearm in Extension vs. Flexion
Three studies, Unlu et al9, Chaware et al10, and Stothers et al8, determined the effects of flexion and extension on distal and proximal portal safety profile (Table 3). Flexion produced greater distances from portal site to neurovascular structures, with the exception of the Chaware study, which found the median nerve and MABCN to lie further from distal and proximal portals in extension. This difference is no greater than 2.7mm for any average measurement.
Miller et al15, a study not captured in any tables due to the uniqueness of their approach, measured distances from the humerus and joint capsule to nerves with and without joint distension. They measured seven pairs of insufflated and non-insufflated cadaveric elbows, one of which was arranged in extension and the other six in flexion. They reported greater distances from bone-to-nerve and capsule-to-nerve in a flexed, insufflated elbow compared to non-insufflated elbows and elbows in extension.
Insertion with forearm in Supination vs. Pronation
Unlu et al9 compared supination and pronation of the forearm. Table 3 indicates the differences in distance between pronation and supination of the forearm with the elbow flexed at 90-degrees. There was a slight increase in safety margins from the distal portal when the forearm was pronated. Supination and pronation had less of an effect on measurements from the proximal portal to neurovascular structures.
Discussion
The results reject our original hypothesis and strengthen the notion that the proximal anteromedial portal puts less neurovascular structures at risk compared to the distal portal. Additionally, distended elbows in a 90-degree flexed position have a reduced risk of neurovascular interference during placement of the arthroscope.
The complexity of elbow arthroscopy is affected by the close proximity of major neurovascular structures near the capsule and operative field. Optimizing the portal site placement can minimize risks of nerve and vascular injury. In this review, we have categorized the anteromedial elbow portal site into two groups relative to their position around the medial epicondyle: the distal and proximal anteromedial portals. We identified the proximal anteromedial portal technique as the surgical approach with the greatest safety margin, specifically in a 90-degree flexed and distended elbow.
The anteromedial portal was first described by Andrews and Carlson3, who aimed to compare the anterolateral and anteromedial portal site. They described the anteromedial portal as 2cm distal and 2cm anterior to the medial epicondyle. Soon after, Poehling et al16 defined the first proximal anteromedial portal as 2cm proximal to the medial epicondyle. They placed their proximal portal directly anterior to the intermuscular septum, which served as a guide to avoid damage to the ulnar nerve. Lindenfield et al5 followed shortly after with another definition of the proximal anteromedial portal, which was placed 1cm proximal and 1cm anterior to the medial epicondyle. From then on, there have been five variations of the anteromedial portal investigated, owing to the discrepancy in expert opinion on adequate portal placement. The senior authors have modified the proximal anterior medial portal, as originally described by Poehling et al16, to a more anterior location. This portal, 2.5cm proximal and almost 3cm anterior to the medial epicondyle, provides a complete view of the entire anterior compartment, including the medial side. The present review, although not specifically addressing this portal, does indeed show a satisfactory safety margin for proximal anteromedial portals and requires further evaluation in a cadaveric study.
Many authors have described caution in their approach of elbow arthroscopy as Andrews and Carson once described in their original review of elbow arthroscopy reporting transient median nerve palsy.17 Lynch et al reported a transient median nerve palsy, radial nerve palsy, and a neuroma of the MABCN using the distal anteromedial portal.4 A more recent study by Hilgersom et al looked at permanent ulnar and radial nerve injury following elbow arthroscopy, speculating that portal placement and use of motorized instrumentation could be avoidable causes to nerve damage.18 Kelly et al concluded in their review of 473 patients that temporary or minor complications from elbow arthroscopy, including ulnar and medial antebrachial cutaneous nerve palsies, could be more common than previously thought, which they reported as occurring 11% of the time. However, Kelly et al noted that there were no permanent complications seen in their review.19 Dumonski et al reported a case of ulnar nerve injury after arthroscopic debridement and drilling of the capitulum using a proximal medial portal (2 cm proximal to the medial epicondyle at the level of the medial intermuscular septum) in a patient with symptomatic osteochondritis dissecans. They concluded that extending this portal further proximally negates any protection of the ulnar nerve afforded by the intermuscular septum.20 Under the assumption that larger distances from the portal to neurovascular structures would decrease risk of injury, it appears imperative to choose a portal technique that would maximize these distances and safety profiles.
Joint distension significantly improved the safety profile of the anteromedial portals (Table 2). Carson agrees with the trend seen here that joint distension should precede puncture of the joint capsule because it pushes soft tissues anteriorly, increasing the clearance between neurovascular structures and the arthroscope.9,12,21 Furthermore, Lindenfield et al5 preferred the proximal anteromedial portal in a distended joint due to the dense, tendinous nature of the flexor/pronator group, which reduces fluid extravasation from the capsule. Lindenfield et al5 compared this reduction in fluid extravasation to the softer, thinner radial capsule when using the anterolateral portal, which has less of an ability to maintain joint distension. Because of its superior fluid control, the proximal anteromedial portal maintains a predictable safety margin for a longer period of time.5
In a distended joint, the ulnar nerve was furthest from the distal portal and the median nerve, brachial artery and MABCN were furthest from the proximal portal (Table 2). Many authors prefer the proximal anteromedial portal because it provides greater visualization of the radial head and it provides better accessibility to the radio-ulnar joint.6,7,21 Stothers et al8 reported that the proximal portal provides complete assessment of the anterior elbow joint. Lindenfield et al5 reported versatility of the proximal portal (1cm proximal, 1cm anterior) over the distal portal used by Lynch et al4 (2cm distal, 2cm anterior) because proximal portal placement can be angled in more ways than the distal portal, providing more options for joint visualization. Lindenfield et al5 further concluded that the distal portal is close to the distal aspect of the elbow capsule and a cannula can only be advanced straight medially in a direction toward the median nerve, putting it at an increased risk of injury. A proximal portal allows the cannula to be directed distally, parallel to the median nerve with less risk of nerve injury.5 Trauma and previous surgery to the elbow is an important part of patient history before plans for portal placement can be done. Potential anatomic changes to neurovascular structures, such as ulnar nerve transposition or previous trauma, can affect nerve mobility and deviate from typical anatomy.5,8,11 Park et al studied the use of proximal anteromedial portals in patients with prior transposition of the ulnar nerve and concluded that it was a safe portal technique based on a degree of certainty with which the nerve can be localized by clinical palpation.22
When comparing the safety profile of each of the studies included in this review to the collective mean, a few studies provided the most accurate measurements from the anteromedial portal to neurovascular structures. Adolfsson et al7 reported the closest measurement to the mean from the distal anteromedial portal to the ulnar nerve, median nerve, MABCN and from the proximal anteromedial portal to the median nerve. Unlu et al9 reported the closest measurement to the mean from the distal anteromedial portal to the brachial artery and from the proximal anteromedial portal to the ulnar nerve and brachial artery. Chaware et al10 reported the closest measurement to the mean from the proximal anteromedial portal to the MABCN. Of all of these studies that reported measurements closest to the mean distances from portal to neurovascular structures, Adolfsson et al7 had the greatest number of measurements closest to the mean distances. One important reason that could have led to the accuracy of the Adolfsson et al7 measurements over the other studies, is that Adolfsson et al7 and Zonno et al11 were the only two studies that incorporated a 2.7mm arthroscope rather than a 4mm arthroscope. Adolfsson et al7 alludes to the fact that a smaller arthroscope could be one of the reasons they reported the specific distances they measured. Additionally, it is important to note that Zonno et al11 reported the second closest measurement to the mean from the proximal anteromedial portal to the ulnar nerve, which was the only measurement they reported in their study. These results indicate that the use of smaller arthroscopes may allow for more accurate measurements when conducting a study of this nature. Although, the majority of arthroscopy is performed with the 4mm arthroscope and the utility of these measurements may be limited. Investigation of pressure on the neurovascular structures needs to be investigated based on arthroscope size to further evaluate these differences.
Future studies should investigate the increased safety profile of the proximal anteromedial portal with every aspect analyzed, including but not limited to cannula size, flexion/extension, joint distension/non-distension, and supination/pronation. Potential new studies include analyses of portal distances to neurovascular structures in post-traumatic elbows compared to healthy elbows and the use of technological advancements and their effect on safety.
Limitations
Our review has several limitations. The medial epicondyle is an indiscrete ridge rather than a focused point and the the included studies did not identify a specific point on the medial epicondyle for reference. This could provide a source of error when measuring the placement of the distal and proximal portals as well as measuring distances from the portal to neurovascular structures.5 Though many anteromedial portal techniques were described in this review, many practitioners have modified their anteromedial approach as previously described. Because of this distinction, there may be other anteromedial portals used throughout the orthopedic surgery community that have not been evaluated for safety in this review. Cadaveric studies are limited by changes in tissue characteristics compared to in vivo and there was a significant distinction in preservation technique between the 4 studies that compared the distal and proximal portals.7–10 Chaware et al10 used formalin-fixed cadavers whereas the other three studies used fresh, frozen cadavers7–9, even noting their use within 36 hours of reception.9 Regarding the cadaver preservation techniques of the 6 other studies in this review, all used fresh, frozen cadavers except for Andrews and Carlson et al3 and Marshall et al12, in which there was no mention of cadaver preservation technique. There were only three studies9,10,11 that included accuracy metrics of their technique. Unlu et al9 reported accuracy to 1mm, Chaware et al10 reported accuracy of their digital Vernier calipers of +/− 0.02mm, and Zonno et al11 reported precision of their Optotrack 3020 measurement system of 0.29mm. Because seven studies did not report measurement accuracy, we may be limited by the potential magnitude of deviation from the reported distances in those studies. Chaware et al10 alludes to the fact that fresh cadavers would have been preferable and asserted that formalin-fixed specimens limit the extent of joint distension and neurovascular mobility due to a reduction in elasticity of embalmed tissue. These limitations were not reported in studies using fresh cadavers.
There were two cadaveric studies that were excluded from this review because they did not meet eligibility and did not contain valuable information needed in order to be a significant contributor to this review. Drescher et al23 was a cadaveric study conducted in German that found the mean distance of the median nerve to the arthroscope of the anteromedial portal was 15.5mm (range 8–27mm) when elbow joint position was optimized in 90-degree flexion and distended. This study did not meet inclusion because the large majority of it was non-English and the small amount that was written in English did not indicate whether the anteromedial portal was made distally or proximally. Claessen et al24 was another cadaveric study that measured the proximity of the anteromedial arthroscope to neurovascular structures. This study was excluded because all of the surgeons were inexperienced in elbow arthroscopy and their training for the procedure came solely from a 2-hour introductory course given before cadaveric dissection took place. The inexperience has potential to confound our measurement analysis from errant portal placement.
Conclusion
The results show proximal anteromedial portals puts less structures at risk compared to the distal portal. Elbows in 90-degrees flexion with joint distension carry a lower risk of neurovascular injury during portal placement. These findings suggest the proximal anteromedial portal to be the safer technique in anteromedial arthroscopy of the elbow.
Supplementary Material
Acknowledgment:
All authors assisted in the proof-reading and preparation of this manuscript. Dr. Myers was supported in part by U54 GM104940 from the National Institute of General Medical Sciences of the National Institutes of Health which funds the Louisiana Clinical and Translational Science Center.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Contributor Information
Tucker Cushing, Department of Orthopaedic Surgery, Tulane University School of Medicine, New Orleans, LA, 70112, USA
Zachary Finley, Department of Orthopaedic Surgery, Tulane University School of Medicine, New Orleans, LA, 70112, USA
Michael J. O’Brien, Department of Orthopaedic Surgery, Tulane University School of Medicine, New Orleans, LA, 70112, USA
Felix H. Savoie, III, Department of Orthopaedic Surgery, Tulane University School of Medicine, New Orleans, LA, 70112, USA
Leann Myers, Dept. Global Biostatistics and Data Science, Tulane School of Public Health & Tropical Medicine, New Orleans, LA, 70112
Gleb Medvedev, Department of Orthopaedic Surgery, Tulane University School of Medicine, New Orleans, LA, 70112, USA
References
- 1.Burman MS. Arthroscopy or the direct visualization of joints: an experimental cadaver study. 1931. Clinical Orthopedics and Related Research 2001;390:5–9. [DOI] [PubMed] [Google Scholar]
- 2.Burman MS. Arthroscopy of the elbow joint: A cadaver study. Journal Bone and Joint Surgery 1932;14:349–350. [Google Scholar]
- 3.Andrews JR, Carson WG. Arthroscopy of the elbow. Arthroscopy 1985;2:97–107. [DOI] [PubMed] [Google Scholar]
- 4.Lynch GJ, Meyers JF, Whipple TL, Caspari RB. Neurovascular anatomy and elbow arthroscopy: inherent risks. Arthroscopy 1986;2:190–197. [DOI] [PubMed] [Google Scholar]
- 5.Lindenfeld TN. Medial approach in elbow arthroscopy. American Journal of Sports Medicine 1990;18:413–417. [DOI] [PubMed] [Google Scholar]
- 6.Verhaar J, Mameren HV, Brandsma A. Risks of neurovascular injury in elbow arthroscopy: starting anteromedially or anterolaterally? Arthroscopy 1991;7:287–290. [DOI] [PubMed] [Google Scholar]
- 7.Adolfsson L Arthroscopy of the elbow joint: a cadaveric study of portal placement. Journal of Shoulder and Elbow Joint Surgery 1994;3:53–61. [DOI] [PubMed] [Google Scholar]
- 8.Stothers K, Day B, Regan W. Arthroscopy of the elbow: anatomy, portal sites, and a description of the proximal lateral portal. Arthroscopy 1995;11:449–457. [DOI] [PubMed] [Google Scholar]
- 9.Unlu M, Kesmezacar H, Akgun I, Ogut T, Uzun I. Anatomic relationship between elbow arthroscopy portals and neurovascular structures in different elbow and forearm positions. Journal of Shoulder and Elbow Surgery 2006;15:457–462. [DOI] [PubMed] [Google Scholar]
- 10.Chaware PN, Santoshi JA, Pakhare AP, Rathinam BA. Risk of nerve injury during arthroscopy portal placement in the elbow joint: A cadaveric study. Indian J Orthop 2016;50(1):74–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Zonno A, Manuel J, Merrell G, Ramos P, Akelman E, DaSilva MF. Arthroscopic Technique for Medial Epicondylitis: Technique and Safety Analysis. Arthroscopy 2010:26(5);610–616. [DOI] [PubMed] [Google Scholar]
- 12.Marshall PD, Fairclough JA, Johnson SR, Evans EJ. Avoiding nerve damage during elbow arthroscopy. Journal of Bone and Joint Surgery 1993;75:129–131. [DOI] [PubMed] [Google Scholar]
- 13.Moher D, Shamseer L, Clarke M, et al. Preferred Reporting Items for Systematic review and MetaAnalysis protocols (PRISMA-P) 2015 statement. Syst Rev 2015;4:1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Wilke J, Krause F, Niederer D, Engeroff T, Vogt L, Banzer W. Appraising the quality of cadaveric studies. Validation of the QUACS scale. J Anat 2015;226:440–446. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Miller CD, Jobe CM, Wright MH. Neuroanatomy in elbow arthroscopy. Journal of Shoulder and Elbow Surgery 1995;4:168–174. [DOI] [PubMed] [Google Scholar]
- 16.Poehling GG, Whipple TL, Sisco L, Goldman B. Elbow arthroscopy: A new technique. Arthroscopy 1989;5:222–224. [DOI] [PubMed] [Google Scholar]
- 17.Andrews JR, Carson WG, JR A, & WG C. Arthroscopy of the elbow. Arthroscopy 1985;1(2): 97–107. [DOI] [PubMed] [Google Scholar]
- 18.Hilgersom NFJ, van Deurzen DFP, Gerritsma CLE, van der Heide HJL, Malessy MJA, Eygendaal D, van den Bekerom MPJ. Nerve injuries do occur in elbow arthroscopy. Knee Surg Sports Traumatol Arthrosc 2018;26:318–324. [DOI] [PubMed] [Google Scholar]
- 19.Kelly EW, Morrey BF, O’Driscoll SW. Complications of elbow arthroscopy. J Bone Joint Surg Am 2001;83-a:25–34. [DOI] [PubMed] [Google Scholar]
- 20.Dumonski ML, Arciero RA, Mazzocca AD. Ulnar nerve palsy after elbow arthroscopy. Arthroscopy 2006;22:577.e1–577.e3. [DOI] [PubMed] [Google Scholar]
- 21.Carson W Arthroscopy of the elbow. Instructional Course Lecture 1988:37;195–201. [PubMed] [Google Scholar]
- 22.Park SE, Bachman DR, O’Driscoll SW. The safety of using proximal anteromedial portals in elbow arthroscopy with prior ulnar nerve transposition. Arthroscopy 2016;32:1003–1009. [DOI] [PubMed] [Google Scholar]
- 23.Drescher H, Schwering L, Jerosch J, Herzig M. The risk of neurovascular damage in elbow joint arthroscopy. Which approach is better: anteromedial or anterolateral? [in German] Z Orthop Ihre Grenzgeb 1994;132:120–125. [DOI] [PubMed] [Google Scholar]
- 24.Claessen F, Kachooei AR, Kolovich GP, Buijze GA, Oh LS, van den Bekerom MPJ, Doornberg JN. Portal placement in elbow arthroscopy by novice surgeons: cadaver study. Knee Surg Sports Traumatol Arthrosc 2017;25:2247–2254. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.