Sage Journals: Discover world-class research

Abstract

Purpose:

Dissatisfaction and an impaired quality of life after total knee arthroplasty (TKA) is often associated with postsurgical anterior knee pain (AKP). The underlying pathological mechanisms are not yet fully understood. Therefore, a multifactorial approach encompassing clinical and radiological parameters seemed reasonable and promising to investigate postsurgical AKP.

Methods:

In this cross-sectional study, 25 patients without and 25 patients with postsurgical AKP after unilateral TKA were randomly recruited from a larger cohort of patients. Multiple clinical and radiological parameters—including real-time shear wave elastography (SWE) to measure the patellar and quadriceps tendon elasticity—were acquired and subsequently associated with AKP. For statistical analysis, SPSS (IBM, version 25) was used.

Results:

In total 50 participants (58.0% men, mean age 63.42 years, mean body mass index 29.75 kg/m²), having different prosthetic designs implanted, were included. Independently of key covariates, the strength of the quadriceps muscle (p = 0.021), a thinner inlay (p = 0.041), and a lower position of the patella (p = 0.041) were associated with AKP. Although no correlation with AKP was found (p = 0.346, resp. p = 0.154), we observed significantly decreased Young’s modulus of the patellar and quadriceps tendons for the involved knee compared to the uninvolved knee (p < 0.001).

Conclusion:

In conclusion, quadriceps muscle strength, inlay thickness, and the patella position might be of particular relevance in avoiding postsurgical AKP. Future studies with larger sample sizes are needed to clarify the impact of quadriceps muscle strength and the postoperative patella position as well as the role of SWE as a personalized modifiable prediction marker.

Keywords

anterior knee pain logistic regression analysis multifactorial analysis shear wave elastography total knee arthroplasty

Introduction

High rates of satisfaction varying between 85.2% and 92.0% after total knee arthroplasty (TKA) have been reported.^1

–4 However, anterior knee pain (AKP) has been described by up to 30% of the patients after TKA,⁵ and it has been strongly associated with patients’ dissatisfaction and impaired quality of life.^6

–9 Difficulties in standing after sitting on a chair, in walking up and downstairs, and in riding a bicycle are reported by patients suffering from AKP due to retropatellar or peripatellar pain, which limits their everyday lives.^10,11 Up to 50% of the revision surgeries after primary TKA are due to patellofemoral complications,¹² of which 1.9–7.8% of them are related to painful patellofemoral sensations.^13
–15 Despite numerous studies, the underlying pathology of AKP after TKA is not completely understood, and there are no clear correlations between clinical and radiological findings.¹⁶ Due to the pronounced development of medical engineering methods, advanced diagnostic opportunities are applied in the research as well as the clinical setting. For example, the use of ultrasound (US) with more developed features—such as power Doppler (PD) and real-time shear wave elastography (SWE)—offers a good practical applicability without exposure to radiation. The SWE assesses quantitatively differences in tissue elasticity (stiffness) by using US that measures the shear wave velocity (m/s), which is generated by the US pulse.¹⁷ Good to very good intra- and interobserver agreement for the SWE of the patellar and quadriceps tendon has been reported.^18
–20 Studies focusing on SWE observed that pathological clinical symptoms are often associated with “softer” tendons.^21
–23 So far, there are no studies that have used advanced US tools in the diagnosis and treatment of patients with AKP following TKA. Therefore, a multifactorial study approach seems promising to better understand the underlying pathological mechanisms of AKP. The aim of the present study was to investigate the associations of patients’ history, clinical as well as radiological parameters with the occurrence of postsurgical AKP.

Materials and methods

Participants

Patients with a minimum of 8 weeks postprimary or non-primary unilateral TKA with a minimum age of 18 years and the absence of neurological symptoms were included in the study. Patients with an inflammatory systemic disease (e.g. rheumatoid arthritis), impaired cognitive status, and a compromised cardiorespiratory fitness that could influence the clinical examination were excluded from this study. During their routine appointments at our inpatient clinic, patients with and without postsurgical AKP after TKA, meeting the criteria of inclusion, were randomly selected from a larger collective of TKA patients and invited to participate. We included an equal number of participants with and without AKP after TKA to achieve an adequate possibility for clinical and radiological comparability between these two groups. All patients gave their written and oral informed consent after receiving spoken and written information about the study. The study was approved by the local ethics committee (study number EK 072/17) and was conducted according to the principles found in the Declaration of Helsinki. Table 1 presents the demographic data of our study population.

Table 1.

Demographic, clinical, and radiological characteristics of participants.

	n	Mean (SD)	Range
Demographics
Age (years)	50	63.42 (8.82)	38–82
Male sex (%)	29	58.00%
Body weight (kg)	50	90.82 (20.63)	56–150
Body length (m)	50	1.74 (0.11)	1.50–1.98
BMI	50	29.76 (5.44)	17.67–44.98
Time after surgery (months)	50	25.30 (27.05)	2–132
Primary TKA (%)	25	50.00%
Septic/infectious revision surgery (%)	10	20.00%
Prosthesis design
PCL sparing TKA	30	60.00%
PCL substituting TKA	4	8.00%
Constrained TKA	15	30.00%
Unknown design	1	2.00%
Femoral and tibial angles for the TKA components
Angle α (femoral frontal)	48	94.88° (2.88)	87–101°
Angle ß (tibial frontal)	48	88.79° (1.94)	85–92°
Angle γ (femoral sagittal)	48	3.94° (3.38)	0–15°
Angle δ (tibial sagittal)	48	87.13° (3.15)	79–93°
CDI	48	0.77 (0.23)	0.15–1.28
Thickness inlay (mm)	41	9.54 (2.81)	4–20
Clinical examination
Strength quadriceps muscle (Janda)	50	3.40 (1.370)	0–5
Sit-to-stand transfer from a chair (%)
Moving up freely	21	42.00%
Moving up easily, supported by their hands	3	6.00%
Difficulties with moving up, supported by their hands	17	34.00%
Unable to move up independently from a chair	9	18.00%
Deficit leg extension (in °)	50	0.94 (2.411)	0–10°
Flexion involved knee (in °)	50	108.70 (24.677)	42–145°
Coronal alignment of the leg (%)
Valgus	2	26.00%
Neutral	35	70.00%
Varus	13	2.00%

PCL: posterior cruciate ligament; SD: standard deviation; TKA: total knee arthroplasty; CDI: Caton–Deschamps index; BMI: body mass index.

Medical history

The definition of AKP encompasses pain experienced in the area of the anterior knee and is a syndrome rather than a disease,²⁴ although an up-to-date and consistent definition is still lacking.²⁵ AKP in this study was defined as pain at the front and center of the knee. Information about a previous history of periprosthetic or septic infections and/or an implant exchanges was derived by a clinical evaluation. The thickness of the tibial inlay, more specifically the polyethylene liner construct, was retrieved from the individual surgical reports or, if available, from the prosthesis pass.

Knee Society clinical rating system

The “knee score” (Knee Society Score (KSS))²⁶ and “the function score” (Knee Society Function Score (KSFS)) of the Knee Society clinical rating system, which is a validated scoring system²⁷ to assess the function of the artificial knee joint,²⁸ were determined.

Clinical examination

The function of the quadriceps muscle was tested according to the research of muscle function testing of Janda²⁹ and graded from 0 = plegie to 5 = normal muscle function. Additionally, the strength of the quadriceps muscle was measured by investigating the sitting to standing transfer from a chair. The latter variable was subdivided into (1) moving up freely, (2) moving up easily supported by their own hands, (3) moving up with difficulties supported by their hands, and (4) being unable to move up independently from a chair. The coronal alignment was determined during the clinical examination, which focused on the intercondylar and the intermalleolar distance and, thereby, specified the leg axis into neutral, varus, or valgus alignment. The range of motion was assessed using a goniometer applying the neutral zero method. The cruciate and collateral ligaments were tested as stable or instable, and instability was categorized into three grades. The collateral ligaments were tested at 0° and 30° flexion of the knee by applying valgus and varus stress.

Sonographic evaluation

Real-time SWE, B-mode US, and PD were used to examine the patellar and quadriceps tendon of the involved and uninvolved knee joint. Three modes of US imaging of the patellar and quadriceps tendon (SuperLinear™ SL 15-4, Supersonic Imagine, Aix-en-Provence, France) comparing the TKA with the contralateral knee were conducted: (1) US examination with B-mode, (2) US examination with PD to determine the modified Öhberg score,³⁰ and (3) real-time SWE. The conventional B-mode was applied to qualitatively examine the morphology, the thickness, the echogenicity, and the structure (e.g. the presence of calcifications or fascicle desintegrity) of the patellar and quadriceps tendon in the longitudinal section and; then, the tendons were subsequently categorized as pathological or nonpathological.^31,32 The PD was applied to assess the neovascularization of the patellar and quadriceps tendon in longitudinal sections. We also determined the modified Öhberg score using PD where the amount of neovascularization was quantified.^33
–35 To quantify the neovascularization of the tendon, we applied a modification of the original 5-grade scale (0–4) of the Öhberg score.³⁰ Thus, a modified 4-grade scale (0–3) was used (0 = no flow, 1 = paratendinous flow, 2 = 1–2 intratendinous vessels, and 3 = multiple intratendinous vessels).^36,37 SWE provides the opportunity to measure the elasticity of tendons in real time (E in kPa).³⁸ It was applied at the proximal, medial, and distal third of the patellar and quadriceps tendon in the strictly longitudinal as well as the transverse direction of the extended knee joint in a relaxed state of the tendons, as previously described.^18,39 Mean Young’s modulus was evaluated by placing a region of interest in the most rigid area of the tendon with a standardized diameter of 2 mm. Subsequently, the mean values of the SWE for the patellar and quadriceps tendon and the ratio (involved over uninvolved tendon) were calculated. In addition to the ratio of the SWE signal, the percentage difference between the uninvolved and involved patellar and quadriceps tendon was calculated (Table 2). All US examinations were performed by two radiologists with at least 5 years of experience in US imaging.

Table 2.

US characteristics of the participants.

	n	Mean (SD)	Range
SWE
Patella involved knee	50	45.69 kPa (14.78)	21.53–76.27
Patella involved knee	50	3.85 m/s (0.63)	2.68–5.04
Uninvolved knee	50	60.08 kPa (19.13)	29.42–114.37
Uninvolved knee	50	4.42 m/s (0.70)	3.13–6.17
Quadriceps involved knee	50	36.43 kPa (16.51)	6.68–82.75
Quadriceps involved knee	50	3.40 m/s (0.79)	1.49–5.25
Uninvolved knee	50	45.69 kPa (16.20)	19.50–91.55
Uninvolved knee	50	4.14 m/s (0.67)	2.55–5.52
Ratio involved over uninvolved knee
Patellar tendon	50	0.79 (0.25)	0.36–1.61
Without AKP	25	0.76 (0.27)	0.36–1.61
With AKP	25	0.83 (0.23)	0.45–1.31
Quadriceps tendon	50	0.72 (0.34)	0.14–1.79
Without AKP	25	0.65 (0.24)	0.14–1.21
With AKP	25	0.78 (0.37)	0.37–1.79
Difference uninvolved and involved knee (%)
Patellar tendon	50	−20.62 (24.93)	−64.31–60.64
Without AKP	25	−23.98 (27.11)	−64.31–60.64
With AKP	25	−17.26 (22.59)	−54.99–30.79
Quadriceps tendon	50	−28.20 (31.56)	−86.21–78.59
Without AKP	25	−34.59 (24.20)	−86.21–21.13
With AKP	25	−21.79 (36.92)	−63.48–78.59
B-mode (pathological findings in %)
Patellar tendon with TKA	47	94.00%
Without TKA	14	28.00%
Quadriceps tendon with TKA	46	92.00%
Without TKA	2	4.00%
Öhberg score
Patellar tendon involved knee	50	1.20 (1.23)	0–4
Patellar tendon uninvolved knee	50	0.16 (0.47)	0–2
Quadriceps tendon involved knee	50	1.25 (1.07)	0–4
Quadriceps tendon uninvolved knee	50	0.14 (0.35)	0–1

AKP: anterior knee pain; kg: kilogram; kPa: kilopascal; m: meter; mm: millimeter; m/s: meter per second; PCL: posterior cruciate ligament; SD: standard deviation; TKA: total knee arthroscopy; SWE: shear wave elastography; US: ultrasound.

Caton–Deschamps index

The Caton–Deschamps index (CDI) was applied to describe the position of the patella, and it is calculated by measuring the ratio of the length of the patella tendon and the length of the patella joint area during a 30° knee flexion.^40,41 The physiological range of the ratio is between 0.6 and 1.3; hereby, a CDI >1.3 is defined as a patella alta, and a CDI <0.6 is defined as a patella baja (PB).

Radiographic evaluation of femoral and tibial components

The alignment of the TKA was calculated according to the femoral and tibial axes. According to Ewald,⁴² the axial alignment of the femoral and tibial components are assessed by measuring the femoral (α) and tibial angle (ß) using the coronal view as well as the lateral view to measure the angle γ of the femoral and the angle δ of the tibial component (Figure 1). Thus, the tibial slope of the TKA was retrieved by subtracting the angle δ from 90°.

Figure 1.

Schematic illustration of the radiographic assessment of the femoral and tibial component alignment according to Ewald.⁴² Adapted from Ewald.⁴²

Statistical analyses

For statistical analyses, SPSS (IBM, version 25) was used. The measured values of the SWE were saved in a picture archiving and communication system and were manually transferred to SPSS. For independent nominal data, the χ ² test was used and for expected frequency <5, the exact Fisher’s test was used. The Mann–Whitney U test was applied for ordinal independent data. The two-sided Student’s t-test was used to compare independent samples, whereas the Welch t-test was automatically applied in case of variance heterogeneity. A binary logistic regression was conducted for the key variables of sex, body mass index (BMI), age, primary versus non-primary TKA, the thickness of the tibial inlay, the ratio of the patellar tendon (involved over uninvolved limb), the strength of the quadriceps muscle, the CDI, and the amount of months after surgery as independent variables and AKP as the dependent variable. All calculations were conducted with n = 50 unilateral TKA, including the characteristics of the participants and functional scores in relation to AKP.

Results

The 50 participants (29 men, mean age 63.42 years) had a mean BMI of 29.75 kg/m² and mean time after surgery of 25.30 months (±27.05). No correlation was observed with either age or sex for AKP (p = 0.559, resp. p = 0.774). Also, AKP was not correlated with the number of months after surgery (p = 0.429) while the mean postoperative months for participants without AKP were 28.36 (±33.43) and for participants with AKP 22.24 (±19.00). The mean BMI for participants without AKP was 30.42 and 29.08 kg/m² for participants with AKP (p = 0.390). Among the participants, 50.0% had a non-primary TKA, which was significantly correlated with AKP (p = 0.048). A posterior cruciate ligament (PCL)-sparing TKA was implanted in 30 (60.0%) participants, a constrained TKA was implanted in 15 (30.0%) participants, and a PCL-substituting TKA was implanted in 4 (8.0%) participants. The specific designs of the prostheses were not significantly associated with AKP (p = 0.241). Neither the mean extension deficit of 0.94° nor the mean range of flexion of 108.70° was associated with postsurgical AKP (p = 0.324, resp. p = 0.109). No correlation was observed between the coronal alignment (p = 0.349), the degree of coronal instability (p = 0.128–0.695), and the degree of the sagittal instability (p = 0.270–0.469) (not shown). The mean radiographic angles measured on X-rays were not significantly associated with AKP (p = 0.232–0.859). Ten participants had a history of infection or septic loosening/revision of the TKA, which was also not significantly associated with AKP (p = 0.480). A significant correlation with AKP (p = 0.041) was observed for the mean inlay thickness, which compared participants without AKP (10.45 mm) and with AKP (8.67 mm) (Table 3). A significantly stronger quadriceps muscle (p = 0.012) was observed in participants without AKP (mean 3.88) compared to the group with AKP (mean 2.92). This observation was supported by the fact that patients without AKP had a significantly easier time rising from a chair (p = 0.038). The KSFS and KSS were significantly lower (p < 0.001) in participants with AKP. A mean CDI of 0.77 was observed and associated with AKP (p = 0.001). Of the investigated TKAs, 18.8% (n = 9) had a PB (CDI < 0.6), which was associated with AKP (p = 0.001). In addition, lower KSS and KSFS were significantly associated with AKP (p = 0.040, resp. p = 0.001) (Table 3).

Table 3.

Clinical and radiological results.

	No AKP (n)	AKP (n)	Mean (SD)	p Value	95% CI
Time after surgery (months)	25		28.36 (33.43)	0.429	−9.322–21.562
Time after surgery (months)		25	22.24 (19.00)	0.429	−9.322–21.562
Thickness inlay (mm)	20		10.45 (3.33)	0.041	0.081–3.486
Thickness inlay (mm)		21	8.67 (1.90)	0.041	0.081–3.486
Quadriceps strength (according to Janda)	25		3.88 (1.30)	0.038	0.224–1.696
Quadriceps strength (according to Janda)		25	2.92 (1.29)	0.038	0.224–1.696
KSS score	25		71.44 (19.08)	<0.001	11.951–33.969
KSS score		25	48.48 (19.64)	<0.001	11.951–33.969
KSFS score	25		81.80 (14.57)	<0.001	11.313–33.087
KSFS score		25	59.60 (22.82)	<0.001	11.313–33.087
CDI	23		0.88 (0.15)	0.001	0.082–0.323
		25	0.67 (0.25)^a
	No PB (n)	PB (n)	Mean (SD)	p Value	95% CI
KSS score	39		61.87 (21.42)	0.040	0.830–32.469
KSS score		9	45.22 (20.42)	0.040	0.830–32.469
KSFS score	39		74.74 (19.80)	0.001	11.177–40.533
KSFS score		9	48.89 (19.33)	0.001	11.177–40.533

CI: confidence interval; KSFS: Knee Society Function Score; KSS: Knee Society Score; mm: millimeter; SD: standard deviation; CDI: Caton–Deschamps index; PB: patella baja.

^a Mean CDI < 0.6 in n = 9.

US examinations

When comparing the involved knee with the uninvolved knee a significant difference (p < 0.001) of the Young’s modulus (the tissue elasticity) was observed for the patellar (45.69 kPa, resp. 36.43 kPa) as well as quadriceps tendon (60.08 kPa, resp. 45.69 kPa). Neither the ratio (involved over uninvolved limb) nor the percentage difference (%) of the patellar nor the quadriceps tendon showed a significant association with the occurrence of AKP (p = 0.154–0.346) (Table 4). A larger difference in participants without AKP than in participants with AKP was observed for the percentage difference of the patellar and quadriceps tendon (p = 0.346, resp. p = 0.154). Additionally, a higher ratio of the patellar and quadriceps tendon being associated with AKP (Figure 2) was observed (p = 0.346, resp. p = 0.154). In other words, this larger difference and higher ratio indicate that softer patellar and quadriceps tendons are less associated with the development of postsurgical AKP. Pathological signs using the B-mode were observed in 94.0% of the patellar and in 92.0% of the quadriceps tendon of the involved limb. The uninvolved limb had a pathological B-mode in 28.0% of the patellar and in 4.0% of the quadriceps tendon. No association was observed between experiencing AKP and pathological signs in the B-mode for the patellar (p = 0.235) and the quadriceps (p = 1.0) tendon after TKA. The mean Öhberg score of the uninvolved limb was 0.16 for the patellar and 0.14 for the quadriceps tendon. No statistically significant correlation was found for the Öhberg score of the patellar and quadriceps tendon with AKP (p = 0.832, resp. p = 0.801). Figures 3 to 5 show sonographic B-mode/PD and real-time SWE images.

Table 4.

Results of the US examinations.

	No AKP (n)	AKP (n)	Mean (SD)	p Value	95% CI
SWE: Ratio involved over uninvolved knee
Patellar tendon	25		0.76 (0.27)	0.346	−0.21–0.07
Patellar tendon		25	0.83 (0.23)	0.346	−0.21–0.07
Quadriceps tendon	25		0.65 (0.24)	0.154	−0.31–0.50
Quadriceps tendon		25	0.78 (0.37)	0.154	−0.31–0.50
SWE: Difference uninvolved and involved limb (%)
Patellar tendon	25		−23.98 (27.11)	0.346	−20.91–7.47
Patellar tendon		25	−17.26 (22.59)	0.346	−20.91–7.47
Quadriceps tendon	25		−34.59 (24.20)	0.154	−30.55–4.95
Quadriceps tendon		25	−21.79 (36.92)	0.154	−30.55–4.95
B-mode
Pathological findings patellar tendon	25			0.235
Pathological findings patellar tendon		22		0.235
Pathological findings quadriceps tendon	23			1.0
Pathological findings quadriceps tendon		23		1.0
Öhberg score
Patellar tendon	25		1.36 (1.41)	0.832	−0.832–0.672
Patellar tendon		25	1.44 (1.23)	0.832	−0.832–0.672
Quadriceps tendon	25		1.32 (1.15)	0.801	−0.556–0.716
Quadriceps tendon		25	1.24 (1.09)	0.801	−0.556–0.716

AKP: anterior knee pain; CI: confidence interval; SD: standard deviation; SWE: shear wave elastography; US: ultrasound.

Figure 2.

To assess the postsurgical individual difference of the elasticity of the patellar and quadriceps tendon, examined by SWE, after a TKA, the ratio was calculated by dividing the involved by the uninvolved tendon. Subsequently, this ratio was compared to participants with and without AKP. The nonsignificant trend (p = 0.154–0.346) of an association between a higher ratio and experiencing AKP is observed. The error bars are representing one standard deviation of uncertainty. AKP: anterior knee pain; SWE: shear wave electrography; TKA: total knee arthroplasty.

Figure 3.

Postoperative sonographic images of a patellar and a quadriceps tendon. The sonographic B-mode/PD images show a hypoechoic patellar tendon with intratendinous neovascularization (modified Öhberg score 2) and calcification (a) as well as intratendinous neovascularization (modified Öhberg score 3) of a quadriceps tendon (b). PD: power Doppler.

Figure 4.

Postoperative sonographic image of a quadriceps tendon from a patient with postsurgical AKP. SWE demonstrates a semi-quantitative color-coded map displayed over a regular B-mode image. Real-time quantitative measurement within the circular ROI reveals a mean tendon stiffness of 166.55 kPa. AKP: anterior knee pain; SWE: shear wave electrography; ROI: region of interest.

Figure 5.

Postoperative sonographic images of a patellar tendon. SWE semi-quantitative measurement reveals a turquoise color signal and a reduced mean tendon stiffness of 37.55 kPa accompanied by substantial pathological changes of the patellar tendon. SWE: shear wave electrography.

Multifactorial analyses of AKP

A logistic regression with AKP as the dependent and sex, BMI, age, primary versus non-primary TKA, the thickness of the inlay, the ratio of the patellar tendon (SWE of the involved over uninvolved limb), the strength of the quadriceps muscle, the CDI, and the time after surgery (months) as independent variables was executed (Table 5). This logistic regression was as a model (χ ²(9) = 24.256, p = 0.004, n = 39) statistically significant. The results of the multivariate analysis show that the statistically significant predictors for AKP subsequent to a TKA were the thickness of the inlay (odds ratio (OR) = 0.674), the strength of the quadriceps muscle (OR = 0.304), and the CDI (OR ≤ 0.001). In other words, a thicker inlay implanted, a stronger quadriceps muscle, and a larger CDI are correlated with experiencing less postsurgical AKP. Sex, BMI, age, primary or non-primary TKA, the ratio of the SWE patellar tendon, and the number of months after surgery were not statistically significant predictors of AKP after a TKA (p = 0.207–0.989). The Nagelkerke R ² is 0.619; therefore, the latter logistic regression model shows a good explanatory power for the variance of the dependent variable of AKP.

Table 5.

Multivariate logistic regression for predictors of the dependent variable of AKP (n = 39).^a

Variable	B	SE	OR (95 CI)	p Value
Sex (female)	0.712	1.074	2.039 (0.248–16.747)	0.507
BMI (kg/m²)	−0.074	0.110	0.929 (0.749–1.153)	0.504
Age (years)	0.008	0.073	1.008 (0.873–1.164)	0.912
Primary versus non primary TKA	1.189	1.046	3.285 (0.423–25.534)	0.256
Thickness inlay	−0.395	0.194	0.674 (0.461–0.985)	0.041^b
Ratio SWE patellar tendon (involved over uninvolved knee)	0.043	3.232	1.044 (0.002–588.838)	0.989
Strength quadriceps muscle	−1.190	0.517	0.304 (0.111–0.838)	0.021^b
CDI	−7.649	3.751	<0.001 (0.000–0.742)	0.041^b
Months after surgery	0.035	0.028	1.036 (0.981–1.094)	0.207

BMI: body mass index; B: partial regression coefficient; CI: confidence interval; OR: odds ratio; SE: standard error; TKA: total knee arthroplasty; SWE: shear wave elastography; CDI: Caton–Deschamps index.

^a The model consists of sex, BMI, age, primary versus non-primary TKA, the thickness of the inlay, the ratio of the patellar tendon (SWE of the involved over uninvolved limb), the strength of the quadriceps muscle, the CDI, and the months after surgery as independent variables and the occurrence of AKP as the dependent variable. The reference for the regression coefficient of sex is female.

^b p < 0.05: statistical significance.

Discussion

The present study investigated multiple variables regarding the prevalence of postsurgical AKP. The magnitude of variables demonstrated no clear correlation with postsurgical AKP, whereas some show statistically significant associations. In line with the current literature, the present study found that the patients’ age, sex, and BMI were not found to be predictive of AKP.^43
–45 Knee-specific characteristics—including the prosthetic design, the history of an infection of the knee prosthesis, the mean radiographic angles of the implanted TKAs, the US parameters of the SWE as well as the B-mode, and Öhberg score—were also not related to the occurrence of postsurgical AKP. Our finding that the prosthetic design is not associated with the development of AKP is supported by the meta-analysis of Li et al.⁴⁶ who reported for PCL-sparing and PCL-substituting TKAs similar clinical outcome regarding postoperative knee pain. Although there is no general consensus about the optimal alignment of the femoral and tibial components, there are several recommendations for optimal TKA component placement to avoid postsurgical knee pain and impaired knee function in daily living activities.⁴⁷ The mean angles of the present investigation correspond well with the recommendations for optimal TKA component placement suggested by Gromov et al.⁴⁷ As in the present study done, applying US to investigate tendons has proven to be more sensitive compared to magnetic resonance imaging,⁴⁸ and, therefore, it is reasonable to use to examine soft tissues.⁴⁹ The examination of PD US showed neovascularization in 63.2% of the patellar tendon and 69.4% of the quadriceps tendon for the involved side compared to 12.0% and 14.0% for the uninvolved side. Thus, the PD US of the patellar and quadriceps tendon were significantly more often defined as pathological for the involved compared to the uninvolved side. These US related findings, however, were neither in the B-mode nor in the feature of PD associated with AKP. Future studies with larger study populations are necessary to determine the relevance of pathological findings using US in relation to clinically apparent symptoms. For instance, the therapy of chronic painful Achilles tendinosis³⁰ and patellar tendinopathy³⁶—namely sclerosing the neo-vessels with injections or topical application⁵⁰ of polidocanol—appeared to be effective and could be a treatment option in patients with AKP attributed to patellar or quadriceps tendinosis. During TKA implantation an irritation of the patellar and quadriceps tendons occurs, and, therefore, the subsequently observed neovascularization might be part of the physiological tissue repair, in particular short- to midterm postoperatively.⁵¹ Comparable to the B-mode and PD, the applied SWE revealed that the patellar and quadriceps tendons of the uninvolved side were significantly “harder” (i.e. less elastic). To the best of our knowledge, this is the first study that uses SWE to investigate the presence of AKP in patients with TKA. We calculated a ratio that includes the patellar and quadriceps tendon of the involved limb to the patellar and quadriceps tendon of the uninvolved limb. The rationale behind this approach was to correct for age-related physiological changes of the elasticity of tendons, which has been previously applied for investigating patellar tendinopathies.⁵² Notably, our observations point in the same direction as previous studies investigating patellar tendons, which reported higher patellar tendon elastic modulus in patients with patellar tendinopathies.^39,52 It needs to be considered that decreased elastic modulus and shear wave velocity values are reported in healthy elderly people,¹⁸ and while showing a wide physiological range, the interindividual differences might rather be related to their physical condition in general.⁵³ Literature concerning normative age-related values and values adapted for certain physical training conditions (e.g. athletes vs. nonathletes) is missing. Therefore, to establish normative age-related values of different tendons of interest, further studies are needed that use multiple measurement time points investigating healthy individuals as well as subjects with sustained surgery. Dirrichs et al.²¹ established SWE as a diagnostic tool for the early detection and monitoring of clinically relevant tendinopathies. However, SWE in the diagnosis and treatment of tendinopathies following TKA still needs to be evaluated and future studies will need to elucidate the role of (preoperative) SWE, and in particular the aforementioned ratio (involved over uninvolved limb), as a personalized predictive marker for postsurgical AKP and its impact as a potentially modifiable factor.

In line with the literature,^54,55 a prior history of revision surgery was associated with AKP and impaired functional outcome. Each surgical intervention adds up to a cumulative tissue injury and scar formation, which is hypothesized to be associated with concurrent increased activation as well as the intensity of the nociceptors being responsible for the experienced pronounced pain.^54,55 Furthermore, we observed that the strength of the quadriceps muscle, the ability to stand from sitting on a chair, the thickness of the tibial inlay, and a radiologically measured PB are associated with experiencing postsurgical AKP. Logistic regression revealed that the inlay thickness, the strength of the quadriceps muscle, and the CDI are independently from the sex, BMI, age, revision knee surgery, and the number of months after surgery associated with AKP. Several investigations pointed out the importance of the function of the quadriceps muscle as a predictor of functional abilities in patients undergoing TKAs,^56
–58 especially for the elderly.^59,60 The observed association of AKP with reduced strength of the quadriceps muscle underlines the necessity of strength training and appropriate physical therapy. To accomplish a more profound approach to address the often-observed weakness of the quadriceps muscle, Werner⁶¹, for instance, suggested a multiphasic approach, including an early correction of imbalance between the vastus medialis and vastus lateralis for physically active individuals with AKP. A significant association of the thickness of the tibial inlay with AKP was observed when comparing the samples and, in addition, as part of the logistic regression analysis as an independent variable, we found an independent association of the thickness of the tibial inlay with AKP, whereby participants having a thinner inlay implanted were more prone to experience postsurgical AKP. Multiple studies observed a correlation between different types and thicknesses of inlays and AKP.^62
–64 It is conceivable that thinner inlays are predisposed for instability due to a slackening of the ligaments, which, in turn, may be related to AKP. Therefore, thicker inlays may reduce the slacking; however, unpredictable results for polyethylene exchange only for prosthetic knee instability are reported with regard to pain and instability.⁶⁵ Furthermore, different studies reported an incidence of PB after TKA between 34% and 65%.^66
–68 By applying a modified Caton–Deschamps index (mCD), Prudhon et al.⁶⁹ observed a difference between pre and postoperative mCD of 0.19 and a lower patella in 81.7% of the cases. In line with our observations, Pfitzner et al.⁷⁰ reported a lower CDI (0.48; ±0.15) after TKA. Our results indicated that a CDI <0.6 (“patella baja”) measured by the CDI was significantly associated with lower KSS and KSFS scores. Potential complications of a PB are patellar impingement on the tibial polyethylene insert,⁷¹ a painful limitation of knee flexion as a result of tightening of the collateral ligaments,^67,68 and diminished femoral rollback⁷²; therefore, it should be avoided.

We note several limitations with the present investigation. First, there is no singular definition of AKP, and to differentiate the patellofemoral pain syndrome (PFPS) is difficult. Similar to postsurgical AKP after TKA, an attempt to allocate postsurgical PFPS after TKA to an anatomical correlate was unsuccessful, and therefore, PFPS and AKP are used synonymously.⁷³ Furthermore, due to the rather small sample size and heterogeneous study population, a generalization of the results might be limited. We investigated a magnitude of clinical and radiographic factors, including the novel strategy of applying SWE in patients after TKA. Thereby, we tested the feasibility of the applied study protocol. Also, we aimed to identify especially relevant factors responsible for postsurgical AKP which can be applied in future studies; therefore, this work is best interpreted in the sense of a pilot study. In addition, it is noteworthy that we included patients at a rather early postsurgical stage. However, the number of months after surgery was nor correlated with AKP neither was the postsurgical time a significant independent predictor for the occurrence of AKP. Furthermore, a relevant portion of the participants underwent at least one revision surgery (50%) and only about two-third of the TKAs were PCL sparing. Future larger studies are necessary to draw conclusions from subgroup analyses consisting of primary and non-primary TKAs. However, in line with prior studies,^74,75 we did not find an association of AKP after TKA with the prosthetic design. Furthermore, it is not conclusively possible to clarify if the observed limited strength of the quadriceps muscle in participants with AKP is responsible for the postsurgical AKP or if the postsurgical AKP results in less loading of the leg, which impairs quadriceps muscle strength. To extend prior studies,⁵⁸ future studies may want to also focus on quadriceps muscle strength of the uninvolved limb considering the fact that patients tend to shift weight away from the operated limb.⁷⁶ Therefore, a causal relationship between postsurgical AKP and the ability to stand up from a chair cannot be finally drawn, and a careful interpretation of the obtained results is necessary. Due to radiation hygiene and conflicting literature regarding the impact of coronal alignment, we abstained from making whole leg radiographs, which, in turn, also accounts for the lack of availability of n = 9 radiological plain radiographs. Since preoperative misalignment of the leg axis (varus or valgus knee) is shown to be predictive for postoperative AKP,⁷⁷ future studies aiming to identify risk factors for postsurgical AKP may want to implicate whole leg radiographs to calculate the coronal TKA alignment. Furthermore, recently published large registry data, investigating the controversial issue of patellar resurfacing, indicate that unresurfaced knees are at higher long-term risk for revision surgery after primary TKA⁷⁸; therefore, future studies may also want to include patella resurfacing as an independent variable in multifactorial analyses.

Conclusion

Postsurgical AKP is prevalent and clinically relevant. With respect to an expected increase of TKAs by 143% by 2050 in the United States,⁷⁹ there is an absolute necessity for further investigations. This multifactorial analysis including the novel strategy of applying SWE as a diagnostic tool revealed that the identification of single variables that are likely to predict the occurrence of AKP is limited. Based on the results of this study, avoiding a PB postoperatively and preoperative conditioning of the quadriceps muscle may help to prevent the development of postsurgical AKP.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Markus Laubach

Matthias Gatz

References

Robertsson

Dunbar

Pehrsson

, et al. Patient satisfaction after knee arthroplasty: a report on 27,372 knees operated on between 1981 and 1995 in Sweden. Acta Orthop Scand 2000; 71: 262–267.

Heck

Robinson

Partridge

, et al. Patient outcomes after knee replacement. Clin Orthop Relat Res 1998: 93–110.

Hawker

Wright

Coyte

, et al. Health-related quality of life after knee replacement. J Bone Joi Surg Am Vol 1998; 80: 163–173.

Anderson

Wixson

RLD

, et al. Functional outcome and patient satisfaction in total knee patients over the age of 75. J Arthroplasty 1996; 11: 831–840.

Michalik

Rath

Springorum

H-R

, et al. Vorderer Knieschmerz nach Knie-TEP-Implantation. Der Orthopade 2016; 45: 386–398.

Bonnin

Basiglini

Archbold

HAP

. What are the factors of residual pain after uncomplicated TKA? Knee Surg Sports Traumatol Arthrosc 2011; 19: 1411–1417.

Scott

Howie

MacDonald

, et al. Predicting dissatisfaction following total knee replacement: a prospective study of 1217 patients. J Bone Joint Surg Br 2010; 92: 1253–1258.

Petersen

Rembitzki

Bruggemann

, et al. Anterior knee pain after total knee arthroplasty: a narrative review. Int Orthop 2014; 38: 319–328.

Zmistowski

Restrepo

Kahl

, et al. Incidence and reasons for nonrevision reoperation after total knee arthroplasty. Clin Orthop Relat Res 2011; 469: 138–145.

10.

Biedert

Sanchis-Alfonso

. Sources of anterior knee pain. Clinics Sports Med 2002; 21: 335–347.

11.

Breugem

Haverkamp

. Anterior knee pain after a total knee arthroplasty: what can cause this pain? World J Orthop 2014; 5: 163–170.

12.

Rosenstein

Postak

Greenwald

. Fixation strength comparison of onlay and inset patellar implants. The Knee 2007; 14: 194–197.

13.

Chen

, et al. Systematic review of patellar resurfacing in total knee arthroplasty. Int Orthop 2011; 35: 305–316.

14.

Breeman

Campbell

Dakin

, et al. Patellar resurfacing in total knee replacement: five-year clinical and economic results of a large randomized controlled trial. JBJS 2011; 93: 1473–1481.

15.

Swan

Stoney

Lim

, et al. The need for patellar resurfacing in total knee arthroplasty: a literature review. ANZ J Surg 2010; 80: 223–233.

16.

Petersen

Rembitzki

Brüggemann

G-P

, et al. Anterior knee pain after total knee arthroplasty: a narrative review. Int Orthop 2013; 38(2): 319–328.

17.

Siu

W-L

Chan

C-H

Lam

C-H

, et al. Sonographic evaluation of the effect of long-term exercise on Achilles tendon stiffness using shear wave elastography. J Sci Med Sport 2016; 19: 883–887.

18.

Hsiao

Chen

Lin

, et al. Reduced patellar tendon elasticity with aging: in vivo assessment by shear wave elastography. Ultrasound Med Biol 2015; 41: 2899–2905.

19.

Tas

Onur

Yilmaz

, et al. Shear wave elastography is a reliable and repeatable method for measuring the elastic modulus of the rectus femoris muscle and patellar tendon. J Ultras Med Offi J Am Inst Ultras Med 2017; 36: 565–570.

20.

Davis

Baumer

Bey

, et al. Clinical utilization of shear wave elastography in the musculoskeletal system. Ultrasonography (Seoul, Korea) 2019; 38: 2–12.

21.

Dirrichs

Quack

Gatz

, et al. Shear wave elastography (SWE) for the evaluation of patients with tendinopathies. Acad Radiol 2016; 23: 1204–1213.

22.

Chen

Cui

, et al. Shear wave elastographic characterization of normal and torn achilles tendons: a pilot study. J Ultras Med Offi J Am Inst Ultras Med 2013; 32: 449–455.

23.

Ooi

Richards

Maffulli

, et al. A soft patellar tendon on ultrasound elastography is associated with pain and functional deficit in volleyball players. J Sci Med Sport 2016; 19: 373–378.

24.

Cutbill

Ladly

Bray

, et al. Anterior knee pain: a review. Clin J Sport Med Offic J Canad Acad Sport Med 1997; 7: 40–45.

25.

Collado

Fredericson

. Patellofemoral pain syndrome. Clin Sports Med 2010; 29: 379–398.

26.

Asif

Choon

. Midterm results of cemented press fit condylar sigma total knee arthroplasty system. J Ortho Surgery (Hong Kong) 2005; 13: 280–284.

27.

Lingard

Katz

Wright

, et al. Validity and responsiveness of the Knee Society clinical rating system in comparison with the SF-36 and WOMAC. J Bone Jt Surg Am 2001; 83-a: 1856–1864.

28.

Insall

Dorr

Scott

, et al. Rationale of the Knee Society clinical rating system. Clin Orthop Relat Res 1989: 13–14.

29.

Janda

. Muscle function testing. London: Butterworths, 1983.

30.

Ohberg

Alfredson

. Ultrasound guided sclerosis of neovessels in painful chronic achilles tendinosis: pilot study of a new treatment. Br J Sports Med 2002; 36: 173–175.

31.

Pingel

Starborg

, et al. 3-D ultrastructure and collagen composition of healthy and overloaded human tendon: evidence of tenocyte and matrix buckling. J Anat 2014; 224: 548–555.

32.

Weinreb

Sheth

Apostolakos

, et al. Tendon structure, disease, and imaging. Muscl Ligament Tend J 2014; 4: 66–73.

33.

Sunding

Fahlström

Werner

, et al. Evaluation of Achilles and patellar tendinopathy with greyscale ultrasound and colour Doppler: using a four-grade scale. Knee Surg Sports Traumatol Arthrosc 2016; 24: 1988–1996.

34.

Notarnicola

Maccagnano

Di Leo

, et al. Overload and neovascularization of Achilles tendons in young artistic and rhythmic gymnasts compared with controls: an observational study. Musculoskel Surg 2014; 98: 115–120.

35.

Visnes

Tegnander

Bahr

. Ultrasound characteristics of the patellar and quadriceps tendons among young elite athletes. Scand J Med Sci Sports 2015; 25: 205–215.

36.

Hoksrud

Ohberg

Alfredson

, et al. Ultrasound-guided sclerosis of neovessels in painful chronic patellar tendinopathy: a randomized controlled trial. Am J Sports Med 2006; 34: 1738–1746.

37.

Gisslen

Gyulai

Nordstrom

, et al. Normal clinical and ultrasound findings indicate a low risk to sustain jumper’s knee patellar tendinopathy: a longitudinal study on Swedish elite junior volleyball players. Br J Sports Med 2007; 41: 253–258.

38.

Bercoff

Tanter

Fink

. Supersonic shear imaging: a new technique for soft tissue elasticity mapping. IEEE T Ultrason Ferr Freq Cont 2004; 51: 396–409.

39.

Coombes

Tucker

. Achilles and patellar tendinopathy display opposite changes in elastic properties: a shear wave elastography study. Scand J Med Sci Sports 2018; 28: 1201–1208.

40.

Caton

Deschamps

Chambat

, et al. [Patella infera. Apropos of 128 cases]. Rev De Chir Orthop et Reparatrice de l’Appar Mot 1982; 68: 317–325.

41.

Caton

. [Method of measuring the height of the patella]. Acta Orthop Belg 1989; 55: 385–386.

42.

Ewald

. The Knee Society total knee arthroplasty roentgenographic evaluation and scoring system. Clin Orthop Relat Res 1989: 9–12.

43.

van Jonbergen

H-PW

Reuver

Mutsaerts

, et al. Determinants of anterior knee pain following total knee replacement: a systematic review. Knee Surg Sports Traumatol Arthrosc 2014; 22: 478–499.

44.

Kim

Reitman

Schai

, et al. Selective patellar nonresurfacing in total knee arthroplasty. 10 year results. Clin Orthop Relat Res 1999: 81–88.

45.

Erak

Rajgopal

Macdonald

, et al. Ten-year results of an inset biconvex patella prosthesis in primary knee arthroplasty. Clin Orthop Relat Res 2009; 467: 1781–1792.

46.

Tan

Deng

, et al. Posterior cruciate-retaining versus posterior stabilized total knee arthroplasty: a meta-analysis of randomized controlled trials. Knee Surg Sports Traumatol Arthrosc 2014; 22: 556–564.

47.

Gromov

Korchi

Thomsen

, et al.

What is the optimal alignment of the tibial and femoral components in knee arthroplasty?

Acta Orthopaedica 2014; 85: 480–487.

48.

Erickson

. High-resolution imaging of the musculoskeletal system. Radiology 1997; 205: 593–618.

49.

Friedman

Finlay

Popovich

, et al. Sonographic findings in patients with anterior knee pain. J Clin Ultrasound 2003; 31: 85–97.

50.

Gatz

Schrading

Dirrichs

, et al. Topical polidocanol application in combination with static stretching in tendinopathies: a prospective pilot study. Muscl Ligam End J 2017; 7: 88–97.

51.

Fenwick

Hazleman

Riley

. The vasculature and its role in the damaged and healing tendon. Arthrit Res 2002; 4: 252–260.

52.

Zhang

Lee

, et al. Changes in morphological and elastic properties of patellar tendon in athletes with unilateral patellar tendinopathy and their relationships with pain and functional disability. PLoS One 2014; 9: e108337.

53.

Dirrichs

Schrading

Gatz

, et al. Shear wave elastography (SWE) of asymptomatic achilles tendons: a comparison between semiprofessional athletes and the nonathletic general population. Acad Radiol 2019; 26(10): 1345–1351.

54.

Petersen

Simonsen

Laursen

, et al. Chronic postoperative pain after primary and revision total knee arthroplasty. Clin J Pain 2015; 31: 1–6.

55.

Skou

Graven-Nielsen

Rasmussen

, et al. Widespread sensitization in patients with chronic pain after revision total knee arthroplasty. Pain 2013; 154: 1588–1594.

56.

Mizner

Petterson

Snyder-Mackler

. Quadriceps strength and the time course of functional recovery after total knee arthroplasty. J Orthop Spor Phys Ther 2005; 35: 424–436.

57.

Yoshida

Mizner

Ramsey

, et al. Examining outcomes from total knee arthroplasty and the relationship between quadriceps strength and knee function over time. Clin Biom (Bristol, Avon) 2008; 23: 320–328.

58.

Mizner

Snyder-Mackler

. Altered loading during walking and sit-to-stand is affected by quadriceps weakness after total knee arthroplasty. J Orthop Res Offic Publ Orthop Res Soc 2005; 23: 1083–1090.

59.

Connelly

Vandervoort

. Effects of detraining on knee extensor strength and functional mobility in a group of elderly women. J Orthop Spor Phys Ther 1997; 26: 340–346.

60.

Moxley Scarborough

Krebs

Harris

. Quadriceps muscle strength and dynamic stability in elderly persons. Gait Post 1999; 10: 10–20.

61.

Werner

. Anterior knee pain: an update of physical therapy. Knee Surg Sports Traumatol Arthrosc 2014; 22: 2286–2294.

62.

Keeney

Clohisy

Curry

, et al. Revision total knee arthroplasty for restricted motion. Clin Orthop Rel Res 2005; 440: 135–140.

63.

Breugem

Sierevelt

Schafroth

, et al. Less anterior knee pain with a mobile-bearing prosthesis compared with a fixed-bearing prosthesis. Clin Orthop Rel Res 2008; 466: 1959–1965.

64.

Breugem

SJM

van Ooij

Haverkamp

, et al. No difference in anterior knee pain between a fixed and a mobile posterior stabilized total knee arthroplasty after 7.9 years. Knee Surg Sports Traumatol Arthrosc 2014; 22: 509–516.

65.

Fehring

Baird

Park

, et al. When polyethylene exchange is appropriate for prosthetic knee instability. J Am Acad Orthop Surg Global Res Rev 2019; 3: e031.

66.

Koshino

Ejima

Okamoto

, et al. Gradual low riding of the patella during postoperative course after total knee arthroplasty in osteoarthritis and rheumatoid arthritis. J Arthroplasty 1990; 5: 323–327.

67.

Weale

Murray

Newman

, et al. The length of the patellar tendon after unicompartmental and total knee replacement. J Bone Joint Surg Brit 1999; 81: 790–795.

68.

Floren

Davis

Peterson

, et al. A mini-midvastus capsular approach with patellar displacement decreases the prevalence of patella baja. J Arthroplasty 2007; 22: 51–57.

69.

Prudhon

Caton

Aslanian

, et al.

How is patella height modified after total knee arthroplasty?

Int Orthopaedics 2018; 42: 311–316.

70.

Pfitzner

Perka

Matziolis

. [Patella height after total knee replacement: influence of the radiological setting]. Der Orthopade 2009; 38: 616–621.

71.

Chonko

V Lombardi

R Berend

. Patella baja and total knee arthroplasty (TKA): etiology, diagnosis, and management. Surg Technal Int 2004; 12: 231–238.

72.

Cameron

Jung

. Patella baja complicating total knee arthroplasty. A report of two cases. J Arthroplasty 1988; 3: 177–180.

73.

Thomeé

Augustsson

Karlsson

. Patellofemoral pain syndrome: a review of current issues. Sports Med 1999; 28: 245–262.

74.

Pavlou

Meyer

Leonidou

, et al. Patellar resurfacing in total knee arthroplasty: Does design matter? A meta-analysis of 7075 cases. J Bone Joint Surg Am 2011; 93: 1301–1309.

75.

Metsna

Vorobjov

Märtson

. Prevalence of anterior knee pain among patients following total knee arthroplasty with nonreplaced patella: a retrospective study of 1778 knees. Medicina 2014; 50: 82–86.

76.

Lai

Hong

. Rising from chair after total knee arthroplasty. Clin Biomech (Bristol, Avon) 1998; 13: 176–181.

77.

Zou

Y-G

Chen

Z-W

Feng

Z-Q

, et al. Factors related to anterior knee pain after total knee arthroplasty. Nan Fang Yi Ke Da Xue Bao 2011; 31(8): 1428–1430.

78.

Coory

Tan

Whitehouse

, et al. The outcome of total knee arthroplasty with and without patellar resurfacing up to 17 years: a report from the australian orthopaedic association national joint replacement registry. J Arthrop 2020; 35: 132–138.

79.

Inacio

MCS

Paxton

Graves

, et al. Projected increase in total knee arthroplasty in the United States—an alternative projection model. Osteoarthr Cartilage 2017; 25: 1797–1803.

Anterior knee pain after total knee arthroplasty: A multifactorial analysis

Abstract

Purpose:

Methods:

Results:

Conclusion:

Keywords

Introduction

Materials and methods

Participants

Medical history

Knee Society clinical rating system

Clinical examination

Sonographic evaluation

Caton–Deschamps index

Radiographic evaluation of femoral and tibial components

Statistical analyses

Results

US examinations

Multifactorial analyses of AKP

Discussion

Conclusion

Footnotes

Declaration of conflicting interests

Funding

ORCID iD

References