One High-Volume Posteromedial Knee Capsular Injection Can Provide Safe Anesthetic Infiltration Across the Entire Posterior Knee Capsule

Abstract

Background

Anesthetic injection through the central or posterolateral knee capsule for analgesia after knee surgery risks popliteal vessel and tibial and peroneal nerve injury. We evaluated the distribution of a high volume of fluid injected through only the posteromedial capsule and compared it to a technique involving injections through the posteromedial and posterolateral capsules.

Methods

Four fresh cadaveric knees were injected with 50 ml of Omnipaque 350 diluted 1:4 with normal saline through the posteromedial knee capsule. In four other specimens, we followed a published protocol by injecting 12.5 ml posteromedially and 12.5 ml posterolaterally. The knees were then ranged 20 times over 20 minutes before obtaining a computed tomography scan to evaluate the spread of injected contrast fluid.

Results

Both techniques demonstrated broad distribution of injected fluid posterior to the capsule. An average of 80% of the capsular width at the level of the injection was within 5 mm of the injected contrast with the single injection, while 63% was with the two-injection technique.

Conclusions

One 50 mL injection through the posteromedial knee capsule can provide broad medial, lateral, superior, and inferior distribution of anesthetic without risking injury to neurovascular structures. This has become our preferred method of anesthetizing the posterior knee during knee surgery.

Introduction

Multimodal anesthesia has contributed substantially to improving perioperative care in total knee arthroplasty (TKA). A common component of multimodal anesthesia has been periarticular injections. Periarticular injections are performed around the knee joint during the TKA surgery and can be used for any knee surgery. Protocols for where the injections are made, what is injected, and how much is injected are highly variable [[1], [2], [3], [4], [5], [6]].

Some pain following total knee arthroplasty can be attributed to small sensory nerves that provide sensation to the posterior knee capsule. These sensory fibers originate from articular branches of the posterior obturator nerve, common peroneal, sciatic, and tibial nerves [7]. To address posterior knee pain following TKA, one may inject local anesthetic into the area posterior to the knee capsule where these small sensory nerves exist. The interspace between the popliteal artery and the posterior capsule of the knee block is one option to deliver anesthetic posterior to the knee capsule [8,9]. This is usually done with ultrasound guidance by an anesthesiologist before or after the surgery as a part of a regional anesthesia protocol. Another way to deliver local anesthetic posterior to the knee capsule is to directly inject it into this area through the posterior capsule during total knee arthroplasty. This is usually done by the surgeon after bony resection and before prosthetic implantation when the posterior capsule is well visualized through the surgical wound. Various techniques have been described detailing where these injections should be done and how much anesthetic should be injected [[1], [2], [3], [4], [5], [6]].

Unfortunately, injecting through the posterior capsule of the knee carries risk. The popliteal artery and vein and the tibial and peroneal nerves can lie just millimeters from the posterior capsule centrally and laterally [[10], [11], [12]]. Given the proximity of these structures, direct nerve and vascular injury, as well as intravascular injection of pain medication, can occur [6,13].

One way to avoid this risk and still deliver local anesthetic to the area posterior to the knee joint is to only inject through the posteromedial capsule. This is a reliably safe area that has no major neurovascular structure at risk, even if the needle is advanced deeper than intended in a posterior direction through the posteromedial capsule. The purpose of this study was to visualize and compare the spread of injected fluid behind the posterior knee capsule from (1) a single high-volume posteromedial capsular injection and (2) a combination of posteromedial and posterolateral injections. We hypothesized that a single high-volume posteromedial injection can distribute anesthetic across the entire posterior capsule from medial to lateral in a distribution that is comparable to multiple posterior injections.

Material and methods

This was an in vitro study using 8 fresh frozen cadaveric knee specimens. Institutional review board approval was thus not required. Each knee was opened anteriorly, and an arthrotomy was performed. The patella was mobilized to expose both condyles of the knee. Each knee was positioned into 90 degrees of flexion. A 2 mm drill was used to drill the posterior condyle of the femur from anterior to posterior with the knee in 90 degrees of flexion. Knees that were to have medial and lateral injections had a drill hole placed through both the medial and lateral posterior femoral condyles. Knees that were to have only a medial posterior capsular injection only had the posteromedial femoral condyle drilled. These holes were drilled to allow an 18-gauge spinal needle to pass. The injections were administered through the posteromedial capsule without disturbing the knee articulation. The knee condyles were left intact to allow for knee flexion and extension following the injections, simulating knee manipulation during TKA and the immediate postoperative period.

Omnipaque 350 contrast medium (GE Healthcare, Chicago, Illinois) was diluted 1:4 with normal saline. In addition, a small amount of 0.2% fast green solution was added to dye the injected fluid for future dissection. An 18-gauge spinal needle was passed through the holes either only on the medial side or on the medial and lateral sides. At each site, we attempted to pass the needle just behind the posterior capsule. In the knees getting a single high-volume posteromedial injection, a single bolus of 50 ml of dilute contrast medium was administered only behind the posteromedial capsule. In the knees getting medial and lateral injections, 12.5 ml of the dilute contrast medium was injected behind both the posteromedial and posterolateral capsules. The volume and location of the medial and lateral injections were based on a previously published protocol for posterior capsule injections [1]. Images of a sample knee with drill holes and spinal needle positioned to inject through the posterior capsule are shown in Figure 1, Figure 2.

Figure 1.

A fresh cadaver specimen with drill holes passing through the posterior condyles allowed for the passage of a spinal needle through the posterior capsule of the knee either medially and laterally or just medially.

Figure 2.

A lateral radiograph of a cadaver specimen with a spinal needle passing through the posterior femoral condyle and through the posterior knee joint capsule is shown.

One minute and 20 minutes after injection, the knees were flexed and extended 10 times. After the final round of flexion and extension, a computed tomography (CT) scan of the knee was performed to evaluate the spread of the injected contrast fluid. For each specimen, the axial slice located at the level of injection was selected. The percent posterior capsule within 5 mm of contact with the injected fluid was then manually measured for each specimen 3 times and averaged to give the value for that specimen. We then calculated the mean for each technique and compared the means using a Student’s t-test.

A 3-D CT reconstruction was also done to qualitatively compare the overall spread of the contrast medium. Two cadavers (one from each group) were also dissected to visually evaluate the spread of the injected fluid in the tissues posterior to the knee.

Results

The percent of the posterior capsule within 5 mm of contact with the injected fluid on the axial slice at the level of the injection was 97%, 66%, 96%, and 61% (average 80%) in the 4 samples that had a single medial high-volume injection, and 58%, 77%, 64%, and 52% (average 63%) in the medial and lateral injection knees. (Fig. 3) The number of specimens was small, and this difference was not statistically significant (P = .17). The 3-D reconstructions showed broad spread of the injected fluid superiorly and inferiorly in both groups (Fig. 4). As we reviewed the axial CT slices for each specimen, it appeared that when the anesthetic was not in proximity to the posterior capsule in the one selected axial slice where we made the measurement, it was in close proximity in other nearby slices either proximally or distally. Images of the posterior dissections of 1 knee injected with green-stained fluid medially and another injected medially and laterally are shown in Figure 5.

Figure 3.

Axial CT scan images of each fresh cadaver specimen taken at the level of the posterior capsule injection are shown. The top row shows the specimens that had the single high-volume injection. The lower row shows the specimens that had 2 injections as described previously by Pinsornsak et al. (1) The contrast fluid is seen infiltrating the tissues posterior to the femoral condyles. The translucent yellow band indicates the region within 5 mm of the posterior capsule in each specimen. This band was used to estimate the percent of the posterior capsule that was within 5 mm of contact with the injected contrast fluid.

Figure 4.

(a) A 3-dimensional reconstruction of a knee that had a single high-volume posteromedial injection of radiocontrast is shown. This demonstrates the spread of the injected fluid proximally and distally. (b) A 3-dimensional reconstruction of a knee that had radiocontrast injections through the posterior capsule medially and laterally is shown. This demonstrates a similar spread of the injected fluid proximally and distally.

Figure 5.

Posterior knee dissections in specimens that had medial-only (left) and medial and lateral (right) injections through the posterior capsule. The distribution of green dye shows that there is extensive infiltration posteriorly even with a single medial-sided high-volume injection.

Discussion

Pericapsular injections are effective in reducing pain following knee surgery, including total knee arthroplasty [14]. The injected medications may include combinations of local anesthetic, nonsteroidal anti-inflammatory medications, opioids, corticosteroids, and epinephrine [14]. There is variability in how injections through the posterior capsule are performed, ranging from 2 (1 medial and 1 lateral) [1] to as many as 13 small injections spanning the width of the posterior capsule, as demonstrated in an instructional video [6]. It has been recognized that injection through the posterior capsule centrally and laterally carries the risk of injury or intravascular injection to the popliteal artery and vein and injury to the tibial and peroneal nerves [6]. A single injection just through the posteromedial capsule of the knee would avoid these risks but has not been previously investigated as a method to deliver anesthetic to the entire posterior knee following surgery.

The purpose of this study was to characterize and compare the spread of a high volume of anesthetic agent that is injected only through the posteromedial knee joint capsule with the spread when the anesthetic agent is injected in multiple locations across the knee (medially and laterally). We used a contrast agent as a surrogate for the anesthetic and evaluated its distribution after injection with CT. With the numbers of specimens available, we did not find a statistically significant difference in the percent of the posterior capsule that was within 5 mm of the injected contrast fluid when the axial CT image taken at the level of the injection was evaluated. More importantly, however, when scrolling through the axial CT slices above and below this point, it was clear that this measurement was highly dependent on the specific slice that was evaluated and that at some point in all specimens there was contact or close proximity of the injected contrast fluid with the posterior capsule spanning all the way from medial to lateral. The 3-D reconstructions and open dissection each confirmed our impression that the injected fluid had a broad distribution posterior to the capsule regardless of technique.

This study has weaknesses. The number of cadaver knees was limited to 8 specimens. This limited the power of this study to detect differences in distribution. The point of the study, however, was to demonstrate the ability of injected fluid to spread to both sides of the knee from a single posteromedial-only injection. The number of specimens used in this study was sufficient to demonstrate this. Second, the CT scans revealed that a small amount of contrast fluid did leak back into the joint in some knees. This could have been due to leakage back through the needle hole or a possible drill-related injury to the posterior capsule. The amount of leakage was small compared to the amount of fluid that remained posterior to the capsule and should not have affected the results, and it is actually likely that such leakage happens during real knee surgery. The drill holes were not plugged. Third, the posteromedial-only injection was with 50 ml of contrast medium, while the multiple injection groups had 2 injections totaling 25 ml. We do not know whether the results would be the same if only 25 ml were injected medially, but we did find that the higher volume (50 ml) of injected fluid could disperse from medially all the way across to the lateral side of the posterior knee capsule. This was the primary clinical question we sought to answer in this study. Our current anesthetic cocktail consists of a mixture of 150 ml of 0.2% ropivacaine, 30 mg of ketorolac, and 0.5 mg of epinephrine. We inject 50 ml of this cocktail once through the posteromedial joint capsule, leaving 100 ml to distribute elsewhere around the knee joint.

Conclusions

It is not necessary to inject multiple locations across the posterior knee capsule to achieve complete medial-to-lateral distribution of injected anesthetic posterior to the knee. A single 50 mL injection to the posteromedial knee capsule provides comparable coverage as multiple posterior knee injections without risking injury to neurovascular structures. This is our preferred technique for distributing analgesic medication posterior to the knee during total knee arthroplasty, and this technique can be used for local infiltration analgesia following any knee surgery.

Acknowledgments

The CT scanner used in this study was purchased using funds from the NIH Shared Instrument Grant S10RR026714-01. With appreciation to the Zeego Lab at Stanford University.

Conflicts of interest

S. Sherman receives royalties from ConMed and DJO; is a paid educational consultant for Arthrex, Kinamed, and LifeNet; is a paid advisory board member for Ostesys, Reparel, Sarcio, Sparta Medical, Vericel, and Vivorte; has stock options in LinkX, Ostesys, Moximed, Sarcio, Reparel, and Vivorte; receives research support from JRF, Smith & Nephew, Octane Biotherapeutics, University of Pittsburg, Miach Orthopaedics Inc., and Organogenesis Inc.; and is a board/committee member of AANA Committee membership, AAOS Committee chair, ACL Study Group Committee membership, AOSSM Committee membership, Arthroscopy Journal Editorial Board, Biologic Alliance Committee Member, Cur Rev MSK Med, Editorial board member, ICRS Committee member, General board member, ISAKOS Committee Deputy Chair, ISMF Course Chair, PF Foundation Online Course Chair, and VJSM Editorial board member. A. Wang receives research support from Siemens Healthineers. All other authors declare no potential conflicts of interest.

For full disclosure statements refer to https://doi.org/10.1016/j.artd.2025.101651.

CRediT authorship contribution statement

Ran Atzmon: Writing – review & editing, Visualization, Resources, Investigation, Formal analysis, Data curation. Landon Polakof: Writing – review & editing, Writing – original draft, Visualization, Methodology, Investigation, Data curation. Adam Wang: Writing – review & editing, Supervision, Resources, Funding acquisition. Seth Sherman: Writing – review & editing, Supervision, Resources, Project administration, Funding acquisition. Nicholas John Giori: Writing – review & editing, Writing – original draft, Validation, Supervision, Resources, Project administration, Methodology, Formal analysis, Conceptualization.