Wrist Arthroscopy Using the 2R Portal: Is It Safer Than the 1,2 Portal?

Abstract

Backgroud:

The purpose of this study was to compare the 1,2 with a novel 2R portal in terms of proximity to critical structures.

Methods:

Wrist arthroscopy was performed on 8 fresh frozen cadavers via the 1,2 and 2R portals. External anatomy was then dissected under loupe magnification. The closest distance between the portals and surrounding anatomical structures was measured in millimeters using digital calipers.

Results:

The 1,2 portal was significantly closer to radial artery and first extensor compartment tendons than the 2R portal. The radial artery was on average 1.32 mm from the 1-2 portal and 14.25 mm from the 2R portal. The 2R portal was significantly closer to the second and third extensor compartment tendons. The closest branch of the superficial branch of the radial nerve (SBRN) was on average 2.04 mm from the 1-2 portal and 7.59 mm from the 2R portal, but this was not statistically significant.

Conclusions:

We advocate using the 2R portal preferentially to the 1,2 portal when treating radial sided wrist pathology to decrease the risk of iatrogenic radial artery and SBRN injury.

Introduction

Since its original description in 1986, ¹ the 1,2 portal has been used for treating radial sided disorders.^2-5 Concern exists due to its location in the anatomical snuffbox and resultant proximity to neurovascular structures.^6-10

The recommended location of the 1,2 portal within the anatomical snuffbox has changed over time. The first descriptions recommend placement at the dorsal aspect of the snuffbox,^1,6 so as to avoid proximity to the radial artery (RA). Steinberg et al ¹¹ demonstrated the potential for iatrogenic nerve injury with Kirschner wire (K-wire) placement into the dorsal aspect of the snuffbox. Current recommended placement for the 1,2 portal is no more than 4.5 mm dorsal to the first extensor compartment or 4.5 mm distal to the radial styloid.^11,12 However, this leaves a very small “safe zone” for portal placement, especially in light of the fact that external palpation of anatomy can be distorted by body habitus, trauma, and soft tissue swelling from fluid extravasation if performing arthroscopy with fluid insufflation. To this point, a more recent study by Shyamalan et al ⁹ suggests that both the superficial branch of the radial nerve (SBRN) and the RA remain very much at risk with 1,2 portal placement. Due to these concerns, we propose a novel 2R portal that is placed just radial to the extensor carpi radialis longus (ECRL) tendon at the level of the wrist joint (Figures 1 and 2). To our knowledge, a “2R” portal has not been described to date.

Figure 1.

Trochar placement in the 1,2 (radial trochar) and 2R (ulnar trochar) portals from (a) a coronal plane view and (b) an oblique view.

Figure 2.

Dissected anatomy with both the 1,2 portal (asterix) and the 2R portal (star) in place. Note proximity of the 1,2 portal to the radial artery and both portal’s proximity to the SBRN.

Note. ECRL = extensor carpi radialis longus; EPL = extensor pollicis longus; EPB = extensor pollicis brevis; SBRN = superficial branch of the radial nerve; RA = radial artery.

The purpose of this study was to revisit the anatomy of the anatomical snuffbox in relation to the 1,2 portal and to describe the placement and anatomical structures at risk with a novel 2R arthroscopy portal. We hypothesize that the 2R portal will be further away from neurovascular structures than the 1,2 portal, with a similar arthroscopic field of view.

Methods

After institutional review board approval, a sample of convenience of 8 fresh frozen mid-humerus to fingertip cadaver specimens (3 women, 1 man [age range: 69-94 years]) were placed under 10 pounds of traction. Specimens were excluded if they had a history of wrist surgery or injury. The 1,2 portal was introduced using previously described surface anatomy landmarks. ¹² Specifically, the 1-2 portal is placed in the most volar and proximal area of the anatomical snuffbox, within 4.5 mm of the radial styloid and no more than 4.5-mm dorsal to the first extensor compartment. The 2R portal was established by first palpating the extensor carpi radialis brevis (ECRB) and longus and extensor pollicis longus (EPL) tendons, and then entering the radiocarpal joint just radial to the tendon intersection (Figure 1). Both portals were placed by first localizing the joint space with a 22-gauge needle, followed by a tenotomy scissors to bluntly spread the soft tissues and create a wider opening in the joint capsule. A cannula and blunt trochar were then inserted. A 1.9-mm arthroscope and cannula (total outer diameter of 2.2 mm) [NanoScope™ (Arthrex, Naples, FL] was then introduced into each portal to confirm intra-articular placement and assess field of view. ¹³

The arm was then removed from traction with the trochars in place. With the forearm in neutral rotation, skin and soft tissue were dissected under 3.5-mm loupe magnification by fellowship-trained hand surgeons, exposing the surrounding nerves, arteries, and tendons (Figure 2). Calibrated digital calipers were used to measure the distance between the 1,2 and 2R portals and the SBRN, RA in the snuffbox, extensor pollicis brevis (EPB), EPL, ECRL, and ECRB to the nearest 100th of a millimeter. Measurements were taken by 2 independent fellowship-trained hand surgeons, and the obtained values averaged. If a structure was abutting but not pierced by the trochar, a value of 0.01 mm was recorded.

Data were summarized using descriptive statistics for continuous variables. Differences in distance to anatomical structures between the 1,2 and 2R portals were assessed using 2-tailed t tests for correlated samples. Pearson correlation coefficient for measurements between the 2 portals and the RA was calculated to ensure interrater reliability.

Results

Interrater reliability was excellent, with a Pearson correlation coefficient of 0.97. The average distance from the 1,2 and 2R portals is summarized in Table 1. The 1,2 portal was significantly closer to RA than the 2R portal (P < .0001). The RA was on average 1.32 mm from the 1,2 portal and 14.25 mm from the 2R portal. The closest branch of the SBRN was on average 2.04 mm from the 1,2 portal and 7.59 mm from the 2R portal, although this was not statistically significant (P = .066). There was at least 1 instance of a nerve branch abutting the trochar with both portals. The 2R portal was significantly closer to the second and third extensor compartment tendons (P < .0005). No tendons or neurovascular structures were pierced during trochar placement. Because the initial purpose of our study was to measure the distance between the portals and neurovascular structures, distances between the portals and every tendon were not measured in the first 2 specimens. The number of specimens used for each measurement is noted in Table 1.

Table 1.

Average Distance From Trochar in Radial Portals to Anatomical Structure.

Structure	No. of cadaver arms	1-2 portal distance, mm (SD; range)	2R portal distance, mm (SD; range)	P value
Superficial branch of the radial nerve*	8	2.04 (2.88; 0.01-8.01)	7.59 (4.64; 0.01-14.09)	.066
Radial artery in snuffbox	8	1.32 (0.84; 0.01-2.59)	14.25 (3.09; 10.08-18.58)	<.0001
Extensor pollicis brevis	6	1.84 (1.84; 0.01-4.6)	14.72 (1.96; 11.38-17.3)	<.0001
Extensor pollicis longus	8	12.30 (4.09; 9.04-21.6)	1.72 (3.32; 0.01-8.73)	<.0001
Extensor carpi radialis longus	7	9.41 (3.85; 5.25-15.2)	0.60 (0.83; 0.01-2.24)	.0005
Extensor carpi radialis brevis	6	14.80 (2.96; 12.15-19.99)	2.12 (3.28; 0.01-7.83)	<.0001

^*Closest branch to each portal.

The arthroscopic field of view was similar between the 2 portals. The 1,2 and 2R portals provided both viewing and working access to the radial styloid, proximal pole of the scaphoid and lunate, articular surface and dorsal rim of the radius, and the radial volar wrist capsule (Figure 3).

Figure 3.

Arthroscopic field of view of radial sided structures between the (a) 1,2 portal and (b) 2R portal. RSC = radioscaphocapitate ligament.

Discussion

While minimally invasive arthroscopic wrist surgery has some benefits over open techniques, it is not without risk. The literature reveals an approximate 4.8% complication rate, ¹⁴ the majority of which comprises iatrogenic nerve injuries or tenosynovitis. Injury to the RA has been reported but is a rare complication of arthroscopy unless K-wires are also placed through the snuffbox.^11,14-16 Many studies have recognized the potential for iatrogenic injury specifically with use of the 1,2 portal, and have sought to better define this anatomy.^6-11 These studies varied in their method of assigning portal location, from using an ideal portal location on a magnetic resonance imaging, ⁵ to cadaver measurements using pins^7,9,11 or trocars. ⁶ Arthroscopy was variably done to confirm intra-articular portal placement. The average distance from the 1,2 portal to the SBRN in the Abrams study was 3 mm (range, 1-5 mm), which aligns with our finding of 2.04 mm. The average distance from the 1,2 portal to the RA was previously estimated to be 3 mm by Abrams et al ⁶ ; because we used a more volar trochar placement, it is unsurprising that our average distance decreased to 1.32 mm. Perhaps more important than the actual millimeter distance is noting that all cadaver studies to date report a range in values of at least 5 mm between the 1,2 portal and either the RA or SBRN. This variability highlights the inherent risk of this portal, and moreover, the importance of good surgical technique when placing the portal, including not plunging with the skin knife and blunt spreading to the level of the joint capsule.

In this study, we investigated whether use of a new 2R portal to access the radiocarpal joint would provide a similar field of view with a greater distance between the portal and radial sided neurovascular structures. Our results demonstrate that the 2R portal is significantly further away from the RA than the 1,2 portal. Although not statistically significant, the 2R portal is on average further way from branches of the radial sensory nerve (RSN). The RSN usually bifurcates around 4.2 cm proximal to Lister tubercle, and the dorsal branch bifurcates again at about the level of Lister tubercle.^7,9,17 To be the most clinically relevant, we evaluated distances from portals to the closest RSN branch. For the 1,2 portal, this was either the most radial or the middle branch of the SBRN. At least 1 of these branches was always close to the portal, on average 2.04 mm, with a range of 0.01 to 8.01 mm. For the 2R portal, this branch was always the most ulnar branch. On average, that most ulnar branch was 7.59 mm away from the portal, but due to branching variability, there was significant heterogeneity in this distance (0.01-14.09 mm). The proximity of nerve branches to both portals again highlights the need for meticulous technique when placing radial sided arthroscopy portals. With respect to tendon proximity, the 1,2 portal was significantly closer to the first extensor compartment tendons, and the 2R portal was significantly closer to the second and third extensor compartment tendons. Certainly, being closer to these tendons is the planned intention to ensure it is further away from the neurovascular structures. As with all portal creation, we advocate for a skin incision followed by blunt dissection with tenotomy scissors to protect the adjacent structures akin to how other portals are fashioned.

Due to the increased distance between the RA and the 2R portal and the equivalent field of view within the radiocarpal joint, we recommend considering its use preferentially to the 1,2 portal for radial sided wrist pathology.

There are several limitations to this study. We tried to mitigate the risks inherent with a cadaver study by using fresh frozen cadavers, mimicking true arthroscopic conditions when establishing portals, using loupe magnification to dissect the specimens, and taking 2 independent measurements for each distance. However, it is possible that in dissecting the specimens, we could have disturbed the normal anatomical relationships between structures and portals. Second, we used a sample of convenience, which demonstrated a statistically significant difference in distances between all structures except for the SBRN. While there was a trend toward the 1,2 portal being closer to SBRN branches than the 2R, it was not statistically significant. However, it is important to note that SBRN branches can exist in very close proximity to both portals and therefore have the potential to be injured with either approach. We reported the data as means and standard deviations as opposed to a median with an interquartile range. This is consistent with the existing literature as reported in this study.

Finally, we did perform arthroscopy on every specimen to confirm intra-articular placement, and to assess field of view between the 2 portals. While we did note equivalence of the field of view and arthroscopic access to the radial sided wrist structures, we did not attempt to quantify this finding.

Conclusion

Notwithstanding these limitations, the 2R portal is a desirable alternative to the 1,2 portal when treating radial sided wrist pathology to decrease the risk of iatrogenic RA injury.