Correction of Tetratorsional Malalignment of the Lower Extremities Improves Patient-Reported Outcomes

Abstract

Background:

Axial malalignment of the bilateral femurs and tibias, previously known as “miserable” malalignment, now renamed tetratorsional malalignment (TTM), presents with hip and/or knee pain refractory to nonoperative treatment.

Purpose:

We sought to investigate whether bilateral rotational osteotomy of the femur and tibia leads to improvement in a deformity-specific patient-reported outcome measure (PROM).

Methods:

A retrospective review of patients who underwent staged rotational correction of the bilateral femur and tibias was performed. Computed tomography (CT) was used to measure the preoperative rotational profile and plan the surgical correction. Stabilization was predominantly with intramedullary nails. The primary outcome measure was the Limb Deformity-modified Scoliosis Research Society (LDSRS) score. Secondary outcomes included change in mechanical limb alignment and complications of the procedure.

Results:

Sixteen patients (13 female and 3 male) with average age of 23.1 years (range: 15–36 years) underwent 4-segment rotational correction. The averages for femoral and tibial deformity correction were 23.5° (6.2° SD) and 20.9° (5.2° SD), respectively. The total LDSRS score improved from 3.67 (0.3 SD) to 4.39 (0.3 SD) (P = .001). The LDSRS sub-scores for function, pain, and self-image also significantly improved. In patients not undergoing concurrent coronal deformity correction, the limb mechanical axis was not significantly changed. No additional procedures were performed to obtain bone union. Three patients required peroneal nerve decompression following the index procedure, and all neurologic symptoms resolved.

Conclusion:

This retrospective review suggests that correction of TTM of the lower extremities may lead to improvements in function, pain, and self-image. There were minimal complications and no iatrogenic deformity among 16 patients reviewed. The new diagnosis, TTM, is descriptive of this debilitating condition without communicating a negative patient image.

Introduction

Atraumatic axial malalignment of the lower limb is among the least recognized osseous deformities because orthogonal radiographs do not clearly demonstrate the abnormality. Even standing radiographs of the entire limb do not expose rotational malalignment without careful inspection. Observation of gait and a prone examination of the hips, knees, and feet are helpful (Fig. 1) but insufficient for diagnosis and not routinely performed in all clinics [10,14,17,23]. Thus, many patients with vague knee and hip pain and negative magnetic resonance imaging (MRI) experience delays in diagnosis and endure physical therapy to treat a bone deformity that stretching or strengthening will not resolve.

Fig. 1.

Prone exam demonstrating increased hip internal rotation (a) and thigh-foot axis (b). When elevated these are indicative of increased femoral anteversion and external tibial torsion.

A particular subset of these patients, those with excessive anteversion of the bilateral femurs and external rotation of the tibiae, have been diagnosed with “miserable malalignment” since 1979, when the term appeared in a textbook on adolescent knee pain [11]. Miserable colloquially captures the experience of walking in these patients, who struggle to position their hips, knees, and feet to minimize stresses in the joints given the relative internal rotation of the knee axis compared with the hip and ankle. This pattern of deformity leads to “squinting” patellae when a patient is viewed in the coronal plane because the kneecap is pulled into the medially rotated trochlea of the femur (Fig. 2). This creates abnormal forces in the patellofemoral region that contribute to anterior knee pain or patellar dislocation [22]. The excessive external rotation of the tibias is felt to be compensatory for the anteverted femurs that fail to decrease in magnitude as is typical during childhood development [19].

Fig. 2.

Standing photographs demonstrating squinting patella with feet forward (a) and externally rotated feet with the patella neutral (b).

A more scientific description of this condition is warranted. We propose for diagnostic and communication purposes the use of the Greek root tetra- to describe the 4 segments of the lower limb (the Latin root quad already denotes the upper and lower limbs), combined with the torsional malalignment of these segments, to form “tetratorsional malalignment” (TTM). This is a slight modification of a prior and broader description of axial deformity by Leonardi et al [13]—torsional malalignment syndrome—because it better captures the magnitude and character of this distinct pathology.

The surgical correction of the condition entails external rotation of the knee joint versus the femoral neck axis through an osteotomy in the femoral diaphysis or metadiaphysis [18]. However, given the prior tibial compensation during development, this creates a foot progression angle too aberrant for normal ambulation so the tibia must be rotated internally as a result. There are numerous tools at the surgeon’s disposal to perform and stabilize a correction, and many of these have been shown to reliably achieve healing with minimal complications [5,13,15,21,24]. What is needed is more data demonstrating that patients who undergo correction, which can take well over a year if all 4 segments are corrected and the implants subsequently removed, report improved function and quality of life. The aim of this study was to document the change in patient-reported outcomes following correction of the bilateral femur and tibia. We hypothesized that patients undergoing tetratorsional rotational correction would report significantly improved outcomes using a validated lower limb deformity measure.

Methods

Institutional Review Board permission was obtained to perform a retrospective review of patient charts to identify patients who underwent rotational correction of the femur and tibia. Thirty-four patients were identified between 2014 and 2021. Patients were indicated for surgery if they presented with hip or knee pain not responsive to nonoperative measures, gait abnormality, and rotational deformity > 10° from normal anatomic values based on computed tomography (CT) scan. Osseous rotational deformity alone without pain or gait disturbance was not an indication for surgery. Patients were excluded if an osteotomy for rotational correction was not performed in all four lower extremity segments. Eleven patients were excluded for unilateral correction only. Four patients were excluded for only bilateral femur or tibia correction, and 3 patients were excluded for 3-segment correction. Patients with concurrent coronal or sagittal plane correction were included. Stabilization with a nail, plate, or circular hexapod external fixation was included. Sixteen patients were included in the final cohort.

Patients were initially evaluated with a clinical exam, including foot progression angle, prone hip rotation, and thigh foot axis measurements. Standing hip to ankle radiographs were obtained to measure mechanical axis deviation (MAD) and joint orientation angles, including the mechanical lateral distal femoral angle (LDFA) and medial proximal tibial angle (MPTA). A CT scan was obtained to measure the preoperative rotational profile and plan the degree of surgical correction. All measurements were made by the fellowship-trained surgeons performing the procedure. The femoral version was measured using a proximal line connecting a point in the center of the femoral head and a point in the center of the femoral neck midway through the neck on axial CT slices, compared with a line connecting the posterior aspects of the femoral condyles on the slice with the largest anterior to posterior width of the distal femur. Values for femoral anteversion (internal rotation of the knee vs the femoral neck) were positive and femoral retroversion negative. Tibial rotation was determined using the same line along the posterior femoral condyles compared with a line connecting the medial and lateral malleoli on the first slice with the talus visible on axial CT (Fig. 3). Normal anteversion of the femur was considered 10°–20° and normal external rotation of the tibia 25°–35°. The amount of deformity on CT was assessed with the values of hip rotation, thigh foot axis, and foot progression angle to determine the amount of correction in each bone. Magnetic resonance imaging of the hip and knee was not routinely utilized unless concomitant pathology was suspected.

Fig. 3.

Rotational CT measurements to determine femoral anteversion and external tibial torsion.

In this example, the right leg has 41° of anteversion and 61° of external tibial torsion, while the left leg has 42° of anteversion and 56° of external tibial torsion. CT computed tomography.

Surgery was performed between 2014 and 2021 by 2 fellowship-trained limb deformity surgeons at a single hospital using previously published techniques [4]. Osteotomies were performed using the percutaneous (1 cm incision) multiple drill-hole technique completed with osteotomes. Acute rotational corrections were visualized by Steinmann pins or Schanz screws placed divergent proximal and distal to the osteotomy site at the planned correction value measured with a sterile goniometer. Gradual corrections were performed using hexapod computer software. Fifty-eight osteotomies were stabilized with intrameduallary nails, 6 were stabilized with external fixation. Need for neurolysis of the peroneal nerve at the level of the fibular neck and fibular osteotomy was considered in tibial rotation > 15°, but was at the discretion of the treating surgeon. Acute corrections of the tibia had prophylactic compartment releases of the anterior compartment using a fasciotome. Tranexamic acid was used in all cases.

The primary outcome measure was the Limb Deformity-modified Scoliosis Research Society (LDSRS) score, a validated patient-reported outcome measure (PROM) [6,9]. The LDSRS consists of 30 questions regarding 4 sub-categories: function/activity, mental health, pain, and self-image (Supplement 1). The responses are recorded with a maximum of 5 and minimum of 1; higher scores represent better status in all 4 categories. The score was obtained during a preoperative visit and at least 1 year following the final corrective surgery (implant removal not included). Other outcome measures determined by the treating surgeon included the change in hip rotation on exam and change in joint orientation angles and MAD on postoperative radiographs. Complications, including fracture of bone or implants, deep infection, nonunion, neuropraxia, compartment syndrome, or need for further surgery were also recorded.

Statistical Analysis

Statistics were performed using Datatab Online Statistics Calculator. The Kolmogorov-Smirnov test for normaldistributions was used to determine parametric or nonparametric hypothesis testing. Comparisons were calculated using paired 2-tailed t tests with significance set at 0.05. Boxplots of the primary outcome were generated using Datatab.

Results

Sixteen patients (13 female and 3 male) with an average age of 23.1 years (range: 15–36) underwent 4-segment rotational correction (Table 1). The average follow-up time was 23.5 months (range: 14–35). The average femoral deformity was 18.5° (9.17° SD) and the average correction was 23.5° (6.2° SD). The average tibial deformity was 16.1° (6.4° SD) with average correction of 20.9° (5.2° SD). For the main outcome measure, the total LDSRS score improved from 3.67 (0.59 SD) to 4.39 (0.44 SD) (P < .001). The LDSRS sub-scores for Function, Pain and Self-image also significantly improved (Fig. 4, Table 2).

Table 1.

List of tetratorsional rotational corrections along with concomitant deformity correction and adjunct procedures.

Case	Deformity correction	Femur rotational correction (R/L)	Tibia rotational correction (R/L)	Coronal plane correction (deg.)	Osteotomy location (stabilization with nail [N], external fixation [E], plate [P])	Fibular osteotomy	Peroneal nerve decompression
1	Axial Coronal (L tibia)	20/25	25/25	L tibia varus—(5)	Diaphyseal (N)	No	Prophylactic (b/l)
2	Axial	35/35	25/25	–	Diaphyseal (N)	No	–
3	Axial	25/25	25/25	–	Diaphyseal (N)	No	Staged (L only)
4	Axial	30/30	25/20	–	Diaphyseal (N)	No	–
5	Axial Coronal (R tibia) LLD (L tibia, 2 cm)	25/40	10/10	R tibia valgus—(5)	Diaphyseal (femurs, L tibia) Metadiaphyseal (R tibia) (N)	Yes	Prophylactic (b/l)
6	Axial	20/20	17/17	–	Diaphyseal (N)	No	–
7	Axial Coronal (b/l tibia)	22/22	20/22	B/l tibia valgus—(5)	Diaphyseal (femur) Metadiaphyseal (tibia) (N)	No	–
8	Axial	17/25	17/20	–	Diaphyseal (N)	No	Prophylactic (R) Staged (L)
9	Axial Coronal (b/l tibia)	25/25	20/20	B/l tibia varus—(7)	Diaphyseal (femur) Metadiaphyseal (tibia) (N)	Yes	Prophylactic (b/l)
10	Axial	15/15	25/15	–	Diaphyseal (N)	Yes	Staged (L)
11	Axial Coronal (b/l femur)	30/30	25/25	B/l femur valgus—R (9), L (15)	Diaphyseal (femur, tibia) (N) Metaphysis (b/l distal femur) (P)	Yes (L only)	–
12	Axial	20/20	20/35	–	Diaphyseal (femur) Metadiaphyseal (tibia) (E)	Yes	–
13	Axial	30/25	25/20	–	Diaphyseal (femur) (N) Metadiaphyseal (tibia) (E)	Yes	–
14	Axial	22/22	25/25	–	Diaphyseal (N)	Yes	–
15	Axial Coronal (b/l tibia)	15/15	15 /15	B/l tibia varus—(7)	Diaphyseal (femur) Metadiaphyseal (tibia) (N)	Yes	–
16	Axial	20/20	20/20	–	Diaphyseal (N)	No	–

LLD limb length discrepancy.

Fig. 4.

Boxplot of the preoperative and postoperative LDSRS scores.

Box includes minimum whisker, quartile 1, mean (dashed line), median (solid line), quartile 3, maximum whisker and outliers (dots). LDSRS Limb Deformity-modified Scoliosis Research Society.

Table 2.

Limb Deformity-modified Scoliosis Research Society (LDSRS) outcome scores.

Domain	Preop. score (SD)	Postop. score (SD)	P value
Total	3.67 (0.3)	4.39 (0.3)	< .001
Function	3.81 (0.5)	4.48 (0.5)	.008
Pain	3.63 (0.6)	4.45 (0.5)	.037
Self image	3.37 (0.5)	4.27 (0.4)	.002
Mental health	4.12 (0.8)	4.38 (0.6)	.35

SD standard deviation. Bold values are statistically significant (P < .05).

The mean preoperative/postoperative prone left hip external rotation was 14° (10.6 SD)/43° (16.4 SD) and internal rotation was 67° (14.9 SD)/39° (13.6 SD). The mean preoperative/postoperative prone right hip external rotation was 19° (8.6 SD)/44° (11.8 SD) and internal rotation was 67° (16.2 SD)/38° (13.8 SD).

In patients not undergoing concurrent coronal deformity correction (n = 10/16), there were no significant differences among the preoperative and postoperative MPTA, LDFA, and MAD values (Fig. 5). The mean pre/post MPTA was 86.7° (3.67 SD)/87.0° (2.86 SD) for the left tibia (P = .25) and 87.1° (3.07 SD)/86.3° (3.06 SD) for the right tibia (P = .24). The mean pre/post LDFA was 86.8° (2.51 SD)/88.5° (2.17 SD) for the left femur (P = .39) and 86.9° (1.9 SD)/88.9° (2.40 SD) for the right femur. The mean pre/post MAD measurement was 13 mm (11.8 SD)/9.67 mm (4.67 SD) for the left lower extremity (P = .42) and 11.3 mm (7.23 SD)/10.7 mm (10.3 SD) for the right lower extremity (P = .24).

Fig. 5.

Standing radiographs demonstrating mechanical axis lines preoperatively (a) and postoperatively (b) after rotational correction with intramedullary nail fixation of the femur and tibias and subsequent implant removal.

Note the mechanical axis is within normal limits (1–15 mm) preoperatively and postoperatively, indicating no iatrogenic deformity.

There were no additional procedures performed to obtain bone union. Three patients had persistent sensory neuropraxia of the superficial peroneal nerve and required open peroneal nerve decompression at the fibular neck following the index procedure. Neurologic symptoms resolved at 2 weeks, 6 months, and 1 year following decompression. There were no instances of deep infection, nonunion, bone or implant fracture, or compartment syndrome.

Discussion

The primary finding in patients who undergo rotational malalignment correction of the bilateral femurs and tibias is improved PROMs of function, pain and self-image. This supports the hypothesis that aligning the knee axis of motion with the hip and ankle improves ambulation and decreases pain caused by abnormal stresses within the joints. This finding is unsurprising given the goal of surgery is to restore normal osseous rotation based on population means of the femur and tibia. The mental health scores also improved, however, the average score was greater than 4 of 5 at baseline, so there was less range for improvement.

There were limitations of this study. The study cohort was small, and the design was retrospective. CT measurements of deformity were determined by the operating surgeon and not compared to other practitioner measurements for reliability. Follow-up CT scans to confirm the intended correction and to correlate with the primary outcome were not obtained given concerns for unnecessary radiation exposure. The follow-up LDSRS measurement was obtained by phone, email, or in person around the time of implant removal, which may have biased results. Longer follow-up and additional data points to confirm durability of the improved outcome would be valuable. Strengths of the study included follow-up of patients to implant removal in most cases, and a consistent pathology was treated with readily available implants and techniques that should be generalizable to other similar patients. Future directions of study could include a broader group of rotational corrections to determine whether all patients with rotational deformity in any segment have similar improvements in patient-reported outcomes.

The identification of pathologic femur and tibia rotation has been present in the literature for many years [2], yet clear surgical indications for osteotomy and rotational correction are debated. Staheli was a prominent author of torsional abnormalities and noted lateral tibial torsion beyond 30° of deformity and femoral anteversion more than 50° as reasonable indications, but this was based on thigh foot axis and hip rotation measurements prior to routine advanced imaging [18]. Even with advanced imaging, there is no consensus on the optimal method of measuring rotation and thus values for normal can vary by up to 15° in the femur and tibia [8]. So aberrant values on imaging should be correlated with symptoms instead of being a firm indication for surgery in isolation. In comparing these results with other studies, it is more valuable to consider the magnitude of rotational correction, 23.5° in the femur and 20.9° in the tibia, than the relative start and end points, which will vary by measurement method. The similarity of these values is indicative of the pathology of TTM—most patients start with foot progression angles near neutral to slight external rotation, and thus need similar rotations of the femur and tibia to maintain a foot forward gait pattern.

Knee pain or discomfort was a prominent symptom in all patients, likely due to stress on the patellar cartilage. In a cadaveric study, Lee et al [12] demonstrated that patellar facet contact pressures increase significantly with increasing rotation of the femur. Only 2 patients had prior attempts at patellar realignment prior to rotational correction in our group, perhaps due to greater recognition that tubercle medialization will not address the underlying pathology [7]. Takagi et al [22] also demonstrated using 3D modeling that both femoral anteversion and external tibial torsion were more associated with recurrent patellar dislocation than coronal and sagittal alignment, confirming the risk to the patella with tetratorsional deformity. In a level II study of only femur rotations, Stambough et al [20] showed that derotating the femur an average of 29° in 28 patients significantly improved International Knee Documentation Committee-9 and Tegner scores. Stevens et al [21] evaluated 23 knees with anterior knee pain or instability with an average of 2 failed prior surgeries, and performed 31 rotational femur and/or tibial osteotomies (only 2 patients had TTM) and found knee pain improved significantly but disappointingly 43% had ongoing patellar instability. Leonardi et al [13] performed bilateral femur and tibial osteotomies on three patients with an average femur and tibial deformity of 21/26.3° with mean follow-up of 16 years, and none of the patients had persistent knee or hip pain. Finally, in a larger series of 14 patients with 27 limbs with “idiopathic miserable malalignment” who underwent average surgical correction of 29/27° in the femur and tibia respectively, Bruce and Stevens [1] reported no patellar instability or persistent knee pain and full satisfaction of all patients with the outcome, although no PROMs were utilized. Overall, these studies would indicate an improvement in knee pain following rotational correction of the femur and tibia which is further supported with the patient-reported outcomes of this study.

The preferred surgical technique utilized was a transverse diaphyseal osteotomy in the femur and the tibia for purely axial realignment. This can be done percutaneously and stabilized with an intramedullary nail which facilitates partial weight bearing immediately after surgery and has minimal impact on lower extremity alignment. However, numerous techniques can be utilized to rotate the lower extremity from the hip to the ankle [16]. For simultaneous correction of coronal alignment, the osteotomy should be placed near the apex of deformity to minimize translation malalignment, usually in the distal femur or proximal tibia metaphyseal region. Whether the fibula needs a simultaneous osteotomy was based on the ability to perform the tibial correction. Any difficulty rotating the tibia acutely led to fibular osteotomy, however, 7 limbs had 25° tibial rotations without necessitating concurrent osteotomy, indicating many tibial corrections are possible without this additional procedure. Leaving the fibula intact imparts stability, reduces blood loss, and presumably leads to less pain because the osteotomized fibula was not separately stabilized.

The results were accomplished with minimal surgical complications, but meticulous planning and technique remain paramount. The heavy reliance on intramedullary nails means accurate start points are critical to prevent iatrogenic deformity. There was no additional surgery needed to obtain bone union. This is likely due to the minimally invasive percutaneous approach, controlled nature of the osteotomy, which uses a fresh drill bit and sharpened osteotomes to create minimal thermal damage to the osteocytes, and the intact periosteum which provides the necessary stem cells for healing. However, appropriate follow-up to confirm healing remains necessary in all cases. There was a need for 3 additional peroneal nerve releases for persistent sensory neuropraxia. This occurs when the nerve is tethered by the proximal anterior compartment fascia. There was no obvious threshold to advise when a prophylactic nerve release is needed. Multiple patients underwent 25° acute tibial corrections without developing symptoms, while 1 patient requiring release was rotated only 15°. While concurrent valgus could be considered an indication for release, the 1 patient (Table 1, Case 10) who had acute valgus correction of the tibia and 20° correction did not develop symptoms. In the series of 27 limbs by Bruce and Stevens [1], none of the patients treated with simultaneous femoral and tibia osteotomy needed peroneal decompression, while one excluded patient treated in a staged manner needed a peroneal neurolysis. External rotation of the knee after femoral osteotomy may sufficiently relax the nerve to permit re-tensioning with the internal rotational of the tibial osteotomy. However, in the series by Delgado et al [3] of 7 limbs, none of the patients required peroneal neurolysis regardless if the treatment was simultaneous or staged. Combined, these results do not delineate a clear indication for peroneal decompression, thus, it is left to the discretion of the surgeon, but it is worth considering in any rotation over 15°. Fortunately, for those who develop a neuropraxia, neurolysis appears curative in most cases.

In conclusion, this retrospective review suggests that correction of TTM may lead to improvement in patient-reported function, pain, and self-image. We found no permanent complications and no iatrogenic deformity associated with the procedures. The new diagnosis, TTM, is descriptive of this debilitating condition and does not communicate a negative patient image in line with current psychosocial medical norms. While harder to identify than sagittal and coronal plane deformity, TTM of the lower limbs deserves equal attention to optimize patient function.