Prophylactic Common Peroneal Nerve Release for Total Arthroplasty of the Valgus Knee: Surgical Technique and Early Outcomes

Abstract

Background

Common peroneal nerve (CPN) palsy after primary total knee arthroplasty represents a relatively rare but serious complication. Recently, there has been a growing interest in prophylactic CPN decompression in high-risk patients with significant combined valgus and flexion deformity. This study aimed to examine outcomes at our institution in those undergoing prophylactic CPN decompression at the time of total knee arthroplasty.

Methods

A retrospective evaluation of a single-institution experience with selected patients at high risk for CPN palsy who underwent prophylactic nerve decompression through a separate incision at the time total knee arthroplasty was performed between July 1, 2018 and December 31, 2022. Patient demographics as well as perioperative and intraoperative clinical and radiographic measurements were collected and analyzed.

Results

A total of 14 patients (15 knees) met our inclusion criteria. The mean preoperative femorotibial angle was 18.6° of valgus (range 13°-22°). The mean preoperative flexion contracture was 4.3° (range 0°-25°). The patients with flexion contractures preoperatively had a mean combined valgus/flexion contracture deformity of 28.8° (range 23°-38°) . There was preservation of nerve function in all knees. No knees required subsequent operative intervention within 90 days of surgery.

Conclusions

Early experience with prophylactic CPN release in our high-risk population demonstrates preservation of nerve function in all patients and is reasonable to consider in patients with a large preoperative combined valgus/flexion deformity. Further studies with larger sample sizes would be beneficial in verification of the results with this technique, as well as determining an angular deformity threshold for which CPN release should be considered.

Introduction

Common peroneal nerve (CPN) palsy after primary total knee arthroplasty (TKA) is a serious postoperative complication that impedes postoperative rehabilitation and extends recovery [1]. Complete CPN palsy is characterized by both motor and sensory components. Physical examination findings include weakness in ankle dorsiflexion/eversion, high stepping gait, and dysesthesias in the lateral leg and dorsal foot. In addition, many patients with CPN palsies after TKA experience pain with both knee flexion and knee extension which impedes postoperative rehabilitation [1]. Patients who suffer this devastating complication struggle to achieve functional range of motion and have sensory deficits that interfere with their activities of daily living [1].

Approximately 10% of patients undergoing TKA have a valgus deformity greater than 10° [2]. This subset of patients has higher incidence of CPN palsies after TKA with reported rates ranging from 0.4%-4% [[3], [4], [5], [6]]. Furthermore, the incidence of CPN palsy increases with angular deformity [7]. Both isolated and combined flexion contracture with valgus deformity further increases the risk of CPN palsy [8]. Additional comorbidities associated with increased nerve palsy risk secondary to TKA include obesity, increased age at surgery, and previous history of neuropathy [7]. Other risk factors include perioperative and intraoperative exposures including anesthesia type, tourniquet time/pressure, traction, and compression [5,7,9,10]

There has been a growing interest in prophylactic CPN release to avoid a postoperative nerve palsy in patients with valgus deformity. Three recent studies have shown promising results with CPN release, with all of the knees demonstrating preserved neurologic function postoperatively [[11], [12], [13]]. The goal of this study was to add to the current literature by describing the technique of prophylactic CPN release at the time of primary TKA in the severe valgus knee at our institution, analyzing the results, and augment the knowledge base utilized in determining the eligibility of patients for nerve release procedures.

Material and methods

Ethics approval was granted by a institutional review board. This study was conducted according to The Strengthening the Reporting of Observational Studies in Epidemiology guidelines [14]. Utilizing Current Procedural Terminology codes 27447 and 64078, we found patients who underwent TKA and CPN release under the same anesthetic between July 1, 2018, and June 31, 2022. In all cases but one, TKA was performed by 1 of 2 fellowship-trained arthroplasty surgeons (NKP or JS). In all cases but one, nerve release was performed by a fellowship-trained hand surgeon with experience in treatment of nerve pathology (JI). In one case, the nerve release and arthroplasty were both performed by a sports fellowship-trained surgeon. Electronic medical records were reviewed for each patient, including clinical notes and radiographs.

Baseline and preoperative characteristics collected included age; gender; body mass index; preoperative diagnosis; femoral-tibial angle; knee flexion contracture; motor and sensory function of peroneal nerve; and history of diabetes mellitus, neuropathy, prior knee surgery, or prior lumbar surgery. The femoral-tibial angle was measured on preoperative and postoperative anteroposterior full leg-length radiographs and defined as the angle formed between the intersection of the femoral and tibial anatomical axes. When a full leg-length radiograph was not available, a standing anterior-posterior knee radiograph was used for postoperative measurement. All measurements were performed by the same investigator by using a standardized method described in the study by Sirik [15]. Sensory and motor function was determined clinically. Surgical outcomes of interest included tourniquet time, and type of anesthesia used. Postoperative measures of interest included femoral-tibial angle, development of CPN palsy, subsequent procedures within 90 days of surgery, all-cause complications, and length of follow-up (days).

Surgical technique

Preoperative planning, templating, and patient positioning

All patients underwent standard surgical optimization and risk assessment per institutional protocol. Preoperative standing radiographs including long-leg alignment radiographs were obtained. Digital templating was performed based on the anteroposterior and lateral radiographs to account for the planned angular deformity correction. Sizing of the components was based on lateral radiograph to allow for tibial coverage without overhang and restoration of anterior and posterior femoral condylar offset.

The patient was positioned supine on the operating table and a nonsterile tourniquet was placed high on the thigh. A nonsterile bump was placed under the ipsilateral hemipelvis to facilitate lateral dissection. Standard technique was used for prepping and draping. In all cases, a sterile foot positioner was used, and care was taken not to wrap stockinette or cohesive bandage higher than the positioner to allow for adequate distal extension of the posterolateral incision if necessary.

Surgical approach to the CPN release

Bony anatomic landmarks including the tibial tubercle, medial and lateral joint line, and fibular head were palpated and marked. Marking was done for both incisions at the outset. For the TKA procedure, a standard midline incision was drawn. An oblique incision approximately 6-cm long over the posterolateral aspect of the knee, one third below and two thirds above the fibular head (Fig. 1). The leg was exsanguinated with an elastic bandage and the tourniquet was inflated to 250 mmHg. The skin was incised with a scalpel and subcutaneous tissue was divided using electrocautery. The nerve release was performed utilizing a similar technique described by Krakow et al. [16]; however, our patients were positioned supine rather than lateral decubitus, as we found that this still allows for sufficient visualization of the posterolateral knee. The peroneal nerve was identified at the biceps femoris insertion and followed from proximal to distal sharply releasing the overlying fascia with dissection scissors. The deep and superficial branches were traced into the anterior and lateral compartments, respectively, and freed from any overlying bands of tissue most commonly related to the leading edges of the muscle fascia. A deep retractor was inserted proximally and superficial to a layer of fascia overlying the common nerve trunk to allow adequate visualization, and the CPN followed proximally for 5-6 cm using long dissecting scissors. Adequate decompression was achieved when a finger could be passed alongside the nerve trunk proximally and the major nerve branches distally. Occasionally, fibrotic adhesions were noted along the nerve at the fibular neck. In some patients, the nerve appeared lax after neurolysis was completed (Fig. 2).

Figure 1.

Incision (a) and sequential release (b-e) of the peroneal nerve.

Figure 2.

(a) With the knee in extension and prior to correction of valgus deformity, there is laxity in the visualized CPN. Following TKA with correction of deformity, there is less laxity (b) although the nerve remains tension-free (c).

Surgical approach to TKA

As previously mentioned, a standard midline incision is used for the TKA portion of the procedure, followed by a medial parapatellar arthrotomy. A minimal medial release is performed off the tibia. Lateral soft tissue is released off the proximal tibia to the mid coronal plane. Bony cuts were made utilizing a measured resection technique. An intramedullary guide was used to make a valgus distal femur cut based on preoperative templating. Femoral component rotation was estimated using Whiteside’s line and the trans-epicondylar axis with posterior up referencing. An extramedullary guide was then used to make a proximal tibia resection. In most cases, further lateral soft tissue release was required to allow for appropriate balancing of the knee. Generally, release of the posterolateral capsule was performed first with a 15-blade or 18-gauge needle. If the knee remained unbalanced, pie-crusting of the iliotibial band, lateral collateral ligament, and popliteus was performed using the Ranawat technique based on which structures were felt to be contracted on manual palpation [17]. The CPN exposure allowed direct visualization of the nerve while pie crusting, ensuring that no iatrogenic injury was caused.

After satisfactory balancing of the knee in the coronal and sagittal planes, the nerve is again visualized to ensure it is not overly tensioned. Final components are then cemented in place per routine.

Wound closure and dressing

The tourniquet is released prior to closure of either wound to ensure adequate hemostasis. This is achieved using electrocautery and tranexamic acid. The midline wound irrigation uses normal saline, pulsatile lavage for the midline wound and a low-pressure bulb syringe for the posterolateral wound. A total joint cocktail consisting of bupivacaine or ropivacaine, clonidine and ketorolac was injected in the posterior capsule and adductor canal in 8 of 15 knees. Both wounds are closed in layered fashion and dressed in silver-impregnated adhesive dressings. The extremity was then loosely wrapped with a cotton dressing and lightly compressive bandage.

Postoperative care and evaluation

All patients were admitted to the hospital and underwent routine evaluation by physical therapists. Three patients were placed in hinged knee braces as a night splint for severe preoperative flexion deformity for 2-4 weeks. They were allowed free total range of motion during the day and locked in extension at night to encourage maintenance of full knee extension. All patients were evaluated in clinic at 2 weeks, 6 weeks, and 12 weeks postoperatively. Supine radiographs were routinely obtained in the postoperative recovery unit, and standing radiographs were obtained at the 12-week visit.

Statistical analysis

A paired sample t-test was used to compare 2 the preoperative and postoperative means. A P value < .05 was considered statistically significant. The Statistical Package for the Social Sciences (SSPS), version 28.0.1.1 (SPSS Inc., Chicago, Illinois, USA), was utilized for statistical analysis.

Results

A total of 14 patients (15 knees) met our inclusion criteria. All patients underwent CPN neurolysis at the time of TKA. Twelve knees were replaced due to end-stage primary osteoarthritis of the knee prior to surgery. Three knee arthroplasties were performed due to rheumatoid arthritis. Twelve of the patients (85.7%) were women and 2 patients (14.3%) were men; the mean age was 67.6 years (range of 57-80 years), and the mean BMI was 30.8 kg/m² (range, 20.7-38.1 kg/m²). Twelve (80%) were diagnosed with primary osteoarthritis, while 3 (20%) knees were diagnosed with rheumatoid arthritis. Six of our 14 (42.9%) patients had previously been diagnosed with diabetes mellitus. Two (14.3%) of our patients had a previous diagnosis of diabetic neuropathy. Two patients had previous lumbar surgery. Two of the 15 knees (13.3%) had undergone ipsilateral knee surgery prior to TKA. Twelve knee replacements (80%) were performed under combined spinal epidural anesthesia and 2 (13.3%) under general anesthesia; anesthesia documentation was unavailable for one knee. Demographic information and type of anesthesia used are summarized in Table 1. The mean tourniquet time was 66.1 minutes (range, 0-165).

Table 1.

Demographic information, perioperative measures, and outcome measures of our patient population.

Patient	Gender	Age	Body mass index	Prior surgeries and comorbidities	Anesthesia used	Tourniquet timed	Length of follow-up in weeks	Complications
1	Female	58	30.9	Prior ipsilateral knee surgery	Combined spinal-epidural, adductor canal block	165	89	None reported
2	Female	75	30.4	None reported	Combined spinal-epidural	85	13	None reported
3	Female	73	29.9	None reported	Combined spinal-epidural, sedation	91	94	None reported
		74	26.7	None reported	Combined spinal-Epidural	77	13	None reported
4	Female	80	38.1	Diabetes mellitus	Combined spinal-epidural, local	109	18	None reported
5	Male	61	31.1	Diabetes mellitus, peripheral neuropathy	Combined spinal-epidural, local	41	54	None reported
6	Female	68	35.4	None reported	Combined spinal-epidural, local	127	90	None reported
7	Female	68	31.2	Prior lumbar surgery	General, local	0	26	None reported
8	Female	57	31.5	Diabetes mellitus	Combined spinal-epidural, adductor canal block, local	65	106	Postoperative hematoma
9	Male	60	31.4	Diabetes mellitus, peripheral neuropathy	Combined spinal-epidural, adductor canal Block, Local	130	116	None reported
10	Female	76	29.1	Prior lumbar surgery	Unavailable	Unavailable	71	None reported
11	Female	63	33.7	Diabetes mellitus, prior ipsilateral knee surgery	Combined spinal-epidural, local	10	74	Superficial wound infection of CPN incision
12	Female	70	32.4	None reported	Combined spinal-epidural, local	80	29	None reported
13	Female	68	29.2	Diabetes mellitus	General	53	58	None reported
14	Female	63	20.7	None reported	General, adductor canal block	115	12	None Reported

All patients had stable sensory and motor CPN function before and after surgery. One of these patients had decreased sensation preoperatively in the CPN dermatome which was unchanged after surgery. The other patients reported normal function. The mean preoperative FTA was 18.6° of valgus (range 13°-22°). The mean postoperative FTA was 4.1° of valgus (range 0°-11°, P < .001). The mean preoperative flexion contracture was 4.3° (range 0°-25°). The mean postoperative flexion contracture was 0.9° (range 0°-5°, P = .035). The 5 patients with flexion contractures preoperatively had a mean combined valgus/flexion contracture deformity of 28.8° (range 23°-38°) and postoperatively of 4.4° (range 0°-8°). Radiographic and clinical function variables are summarized in Table 2. There were 2 (13.3%) knees with postoperative complications: the first was an intra-articular hematoma which did not require any additional intervention and resolved without sequelae. The second complication was a wound dehiscence of the peroneal nerve incision with superficial wound infection. Aspiration was performed and periprosthetic joint infection was ruled out. This was treated with oral antibiotics. Complications are summarized in Table 1. The mean follow-up was 57.5 weeks (range 12-116 weeks). No patients necessitated subsequent surgical intervention on the operative knee during the first 90 days after surgery.

Table 2.

Summary of mean and ranges of preoperative and postoperative radiographic angles, flexion contracture, and tourniquet time in our patient population.

Clinical and radiographic variables (n = 15)	Mean (range)
Preoperative femorotibial angle	18.6° of valgus (13-22)
Postoperative femorotibial angle	4.1° of valgus (0-11)
Preoperative flexion contracture	4.3° (0-25)
Postoperative flexion contracture	0.9°(0-5)
Tourniquet time	66.1 min (0-165)

Discussion

CPN palsy is a rare but serious postoperative complication of TKA with a poorly understood clinical course. Previously proposed intraoperative etiologies of CPN palsy secondary to TKA include direct stretching, traction, direct injury during lateral soft-tissue release, and ischemia due to tourniquet or vascular damage. Although rare, CPN palsy can significantly affect rehabilitation and lead to stiffness and suboptimal function of the knee [1]. In our study, none of our patients developed CPN palsy after prophylactic CPN release at the time of TKA. Prophylactic CPN release is one method that has shown promising results in at-risk patients undergoing arthroplasty and can be a helpful tool for surgeons in avoiding postoperative palsy. Further studies are needed to determine the threshold of valgus deformity or combined flexion contracture and valgus deformity which would necessitate a CPN release.

This study is not without limitations. The sample size of the study was small, and the retrospective nature of the study led to variations in coding, charting, and clinical assessment between providers/patients. There was also selection bias in that there were no standardized selection criteria for prophylactic CPN release. Full leg-length radiographs were not available postoperatively for all patients, so standing anterior-posterior films had to be used for most measurements. Finally, our study did not compare to controls or other protective treatments. A study with more participants or a comparison group would be beneficial to lend further support to our results. Despite these limitations, we were able to demonstrate preservation of nerve function in our patient population.

Prophylactic release of the CPN allows for preservation of nerve function in patients with increased femorotibial angulation. An added advantage is the ability to directly protect the nerve at the time of pie crusting of the posterolateral structures, which is classically described for severe valgus deformity [17]. In our patient population, the mean preoperative FTA of 18.6° and mean postoperative FTA of 4.1° were statistically different. All patients in our patient population had no change in CPN function after surgery. As previously mentioned, Xu et al. [13] also found that “prerelease” of the CPN is thought to be protective against nerve injury in patients with severe valgus deformity undergoing primary TKA. This prospective study included 34 patients with a mean preoperative FTA of 31.3 ± 8.0° who underwent CPN release at the time of TKA. All patients had full preoperative CPN function, and none of their patients developed functional deficits following TKA with common peroneal neurolysis. In a recent study, Makhdom et al. [12] included 10 knees in a retrospective study of CPN release in the valgus knee replacement. Their population had a mean preoperative anatomic alignment of 20° of valgus and all patients had full function of the CPN both preoperatively and postoperatively. Puijk et al. [11] recently published a retrospective study which compared patients with a valgus deformity of >15° or combined valgus deformity >10° and flexion contracture >15° who received a CPN release vs patients who did not receive a CPN release. Twenty-six knees received CPN release concurrently with TKA and there was no incidence of CPN palsy postoperatively. The non-CPN release group included 55 knees and 5 cases of postoperative CPN palsy were reported. However, the incidence of CPN palsy between the 2 groups was not found to be significantly different. Our results align with the findings in these 3 studies. However, it is important to note that these previous studies excluded or did not evaluate certain risk factors for postoperative CPN palsy. For example, Makhdom et al. excluded patients with previous peripheral neuropathies and did not discuss combined valgus/flexion contracture while Xu et al. and Puijk et al. did not include any patients with peripheral neuropathies. None of these studies addressed previous ipsilateral knee surgery, spinal surgery, or type of anesthesia beyond using local. In our study, we identified and included patients with risk factors including previous knee surgery, spinal surgery, and history of peripheral neuropathy. We found that this procedure may be protective in preventing CPN palsy in these patients.

Several studies have found that more severe valgus deformity and combined valgus deformity with flexion contracture are strong risk factors in development of CPN palsy after primary TKA [[3], [4], [5],18]. Idusuyi and Morrey calculated the relative risk of CPN palsy in patients with a valgus deformity ≥12° undergoing TKA was 12 times greater than patients with less severe valgus deformity [18]. This association was supported by Christ et al. [10] in their study which included 383,060 total knee arthroplasties and calculated an odds ratio of 4.19 for postoperative CPN palsy in knees with valgus deformity. In our patients, the mean preoperative FTA was 18.6° of valgus and ranged from 13-22°; the mean preoperative flexion contracture was 4.3 and ranged from 0°-25°. Despite significant deformity, all patients had preserved CPN function postoperatively. The mean preoperative FTAs in the study by Puijk et al. was 17° in the CPN release group vs 13° in the non-CPN group. The mean flexion contracture in the CPN release group was found to be 10° compared to 3° in the non-CPN group. Incidence of CPN palsy was 9% in the non-CPN group compared to no reported cases of CPN palsy in the prophylactic release group. In summary, prophylactic CPN release decreased the incidence of CPN palsy, although it was not significantly different [11].

CPN decompression at the time of arthroplasty appears to be a relatively safe procedure with few reported complications elsewhere in the literature [[11], [12], [13]]. In our population, there were 2 post-operative complications and only one that was related to the CPN decompression. One patient developed an intraarticular postoperative hematoma which resolved without any intervention. The other knee developed a superficial surgical site infection at the site of the CPN decompression which was successfully treated with oral antibiotics. Our data support CPN release as a safe procedure. In this series, the majority of nerve releases were performed by one of the senior authors from our institution who is a hand surgeon (JI). If no hand surgeon was available, nerve release could reasonably be performed by an arthroplasty surgeon during TKA at their discretion.

Despite being identified as major risk factors in prior studies, none of our patients with a history of lumbar surgery, knee surgery, diabetes mellitus, or neuropathy developed postoperative nerve palsies. Christ et al. [3] found that previous history of spinal conditions had an odds ratio of 1.98 for developing postoperative CPN palsy when compared to patients without history who were undergoing TKA. Similarly, it is debated whether or not diabetes mellitus increases the likelihood of CPN palsy after TKA. Studies by Indusuyi and Morrey [18], Shetty et al. [10], and Park et al. [19] found no increased risk of CPN palsy after TKA and diagnosis of diabetes mellitus. Carender et al. [4] hypothesized that patients within this subgroup may have clinical or subclinical peripheral neuropathy and the “second hit” caused by minor injury to the nerve during TKA causes overt neuropathy. The authors also hypothesize a similar mechanism of injury in those with previous lumbar surgery. However, since none of our patients within these subgroups developed CPN palsy after TKA, prophylactic neurolysis may serve as a viable option in patients with a variety of risk factors.

Additionally, type of anesthesia has been identified as another risk for CPNP [10,19]. In our population, despite a variety of anesthesia types there was preservation of nerve function in all knees. Studies performed by Beller et al. and Indusuyi and Morrey proposed that use of intraoperative epidural anesthesia could be a risk factor for postoperative palsy via indirect damage to the CPN by unintentional compression of the limb [18,19]. Horlocker et al. found that post-operative epidural analgesia has the potential to mask peroneal nerve palsy and delay intervention, noting a relatively high incidence of CPN palsy and a delayed presentation in this population [20]. There was preservation of CPN function in all of our patients despite an array of anesthetic methods.

Conclusions

Prophylactic CPN decompression could be a valuable aid to the arthroplasty surgeon in minimizing the risk of post-operative nerve palsy in patients with a large preoperative valgus or combined valgus/flexion deformity. Further studies with a larger sample size would be beneficial in verification of the results of the technique as presented here as well as determining a preoperative deformity threshold for which CPN release should be considered.

Conflicts of interest

The authors declare there are no conflicts of interest.

For full disclosure statements refer to https://doi.org/10.1016/j.artd.2023.101205.