Schöttle方法是术中透视识别MPFL股骨足印最佳技术_透视识别MPFL股骨足印（2024）

Schöttle方法是术中透视识别MPFL股骨足印最佳技术_五种不同的透视方法识别MPFL股骨足印的比较（2024）Comparison of five different fluoroscopic methods for identifying the MPFL femoral footprint

Emre T Y, Cetin H, Selcuk H, Kilic K K, Aykanat F, Sarikcioglu L, Kose O. Comparison of five different fluoroscopic methods for identifying the MPFL femoral footprint[J]. Arch Orthop Trauma Surg, 2024,144(4): 1675-1684.

转载文章的原链接1：

https://pubmed.ncbi.nlm.nih.gov/38400901/

转载文章的原链接2：

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10965741/

Abstract

Purpose

The success of medial patellofemoral ligament (MPFL) reconstruction is closely linked to the precise positioning of the femoral tunnel. Intraoperative fluoroscopy is commonly utilized to identify the MPFL footprint. This study aimed to ascertain the most accurate fluoroscopic method among the five previously described methods used to determine the MPFL femoral footprint.

髌股内侧韧带(MPFL)重建的成功与股骨隧道的精确定位密切相关。术中透视通常用于确定MPFL足迹。本研究旨在确定五种先前描述的用于确定MPFL股骨足迹的方法中最准确的透视方法。

Materials and methods

Using 44 well-preserved dry femur bones, the MPFL femoral insertion site was demarcated using anatomical bony landmarks, namely the center of the saddle sulcus between the medial epicondyle, adductor tubercle and gastrocnemius tubercle. Fluoroscopic true lateral knee images were acquired and measurements taken, referencing established methods by Schottle et al., Redfern et al., Wijdicks et al., Barnett et al., and Kaipel et al. The distance between anatomic and fluoroscopic MPFL footprints was then measured on digital fluoroscopic images. The accuracy of the locations was compared using a margin of error of 5 and 7 mm.

使用44块保存完好的干股骨骨，利用解剖性骨标志，即位于内上髁、内收肌结节和腓肠肌结节之间的鞍沟中心，划定MPFL股骨止点。参考Schottle等人、Redfern等人、Wijdicks等人、Barnett等人和Kaipel等人的既定方法，获得真实侧膝透视图像并进行测量。然后在数字透视图像上测量解剖和透视MPFL足迹之间的距离。使用5和7毫米的误差范围比较了位置的精度。

Results

The Schottle method consistently emerged superior, showcasing the smallest mean distance (3.2 ± 1.2 mm) between the anatomic and radiographic MPFL footprints and a high in-point detection rate of 90.9% under 5 mm criteria. While the Redfern method displayed perfect accuracy (100%) within the 7 mm criteria, the Schottle method also performed 97.7% accuracy.

Schottle方法一直表现出优势，显示解剖和放射学MPFL足迹之间的平均距离最小(3.2±1.2 mm)，并且在5mm标准下的点内检出率高达90.9%。Redfern法在7毫米标准内显示出完美的准确度(100%)，Schottle法也有97.7%的准确度。

Conclusions

For intraoperative identification of the MPFL footprint using fluoroscopy, the Schottle method is the most consistent and accurate among the assessed methods. Thus, its accuracy in detecting the MPFL footprint makes it recommended for MPFLR to ensure optimal outcomes.

对于术中使用透视法识别MPFL足迹，Schottle方法在评估的方法中是最一致和准确的。因此，它在检测MPFL足迹方面的准确性使其被推荐用于MPFLR，以确保最佳结果。

Level of evidence

Level IV, cadaveric study.

Keywords: Medial patellofemoral ligament, Patellofemoral instability, Patellar dislocation, Fluoroscopy, Femoral tunnel, Schottle point

Introduction

The medial patellofemoral ligament (MPFL) is often injured in acute lateral patellar dislocations. Kluczynski et al. analyzed 35 research studies involving 2558 patients with acute patellar dislocations and found that nearly 95% had injuries to the MPFL [1]. The MPFL, as the main stabilizer of the patella against lateralizing forces between 0 and 30° of knee flexion, is primarily responsible for preventing lateral patellar dislocations, contributing to over half of the restraining force according to previous biomechanical studies. [2–4]. While first-time dislocations can be managed conservatively, recurrent dislocations usually require surgical intervention [5, 6]. MPFLR is the mainstay of surgical treatment for recurrent patellar dislocation, which corrects the ligament deficiency and stabilizes the patella. The success of MPFLR is closely linked to the precise positioning of the femoral tunnel. Recent biomechanical studies underscore the critical role of correct femoral tunnel placement in maintaining normal patellofemoral kinematics [7, 8]. Furthermore, clinical studies have consistently shown that correct femoral tunnel placement is associated with fewer complications and better outcomes [9–11].

髌股内侧韧带（MPFL）常在急性外侧髌骨脱位中损伤。Kluczynski等人分析了35项涉及2558例急性髌骨脱位患者的研究，发现近95%的患者有MPFL损伤[1]。根据之前的生物力学研究，MPFL作为髌骨对抗0 - 30°膝关节屈曲偏侧力的主要稳定器，主要负责防止外侧髌骨脱位，贡献了超过一半的约束力[2 - 4]。虽然首次脱位可以保守处理，但复发性脱位通常需要手术干预[5,6]。MPFLR是复发性髌骨脱位的主要手术治疗方法，可矫正韧带缺损，稳定髌骨。MPFLR的成功与股骨隧道的精确定位密切相关。最近的生物力学研究强调了正确的股骨隧道放置在维持正常的髌股运动中的关键作用[7,8]。此外，临床研究一致表明，正确的股骨隧道放置与更少的并发症和更好的预后相关[9-11]。

There are two primary intraoperative techniques for determining the MPFL femoral footprint: the open and the fluoroscopic methods. The open method identifies the MPFL femoral footprint by surgically dissecting and palpating specific anatomical landmarks, such as the adductor tubercle, gastrocnemius tubercle, medial epicondyle, and saddle sulcus. While it provides high accuracy without radiation exposure, it requires extensive dissection, resulting in significant scarring. Furthermore, it requires considerable experience [12–15]. The fluoroscopic method, on the other hand, relies on radiographic landmarks on a true lateral knee image to locate the MPFL femoral footprint. It is minimally invasive and can be performed through an aesthetically acceptable small incision [16]. However, this technique is affected by even minor errors in obtaining the true lateral knee radiograph [17].

术中有两种主要的方法来确定MPFL股骨足迹:开放和透视法。开放性方法通过手术解剖和触诊特定的解剖标志，如内收肌结节、腓肠肌结节、内上髁和鞍沟，来识别MPFL股骨足印。虽然它在没有辐射暴露的情况下提供高精度，但它需要广泛的解剖，导致明显的疤痕。此外，它需要相当的经验[12-15]。另一方面，透视法依赖于真实膝外侧图像上的放射学标记来定位MPFL股骨足迹。它是微创的，可以通过美观可接受的小切口进行[16]。然而，在获得真正的膝侧位X线片时，即使是很小的错误也会影响这项技术[17]。

Schöttle et al. first described the fluoroscopic technique in 2007 [16]. Subsequently, it gained widespread acceptance and was implemented in practice. Schöttle et al. dissected eight cadavers to identify the MPFL, positioned a metal indicator at its femoral attachment site, and standardized the projection of this point on a true lateral knee fluoroscopic image. According to Schöttle et al., the MPFL footprint was situated 1.3 mm anterior to the posterior cortical extension, 2.5 mm distal to a perpendicular line intersecting the origin of the posterior medial femoral condyle, and 3 mm proximal to a perpendicular line intersecting the posterior point of the Blumensaat line. Nonetheless, subsequent researchers have proposed that this methodology might lack precision and have advocated for alternative fluoroscopic techniques [18–21] (Fig. 1). To date, no research in the existing literature has evaluated the accuracy of these methods compared to one another. This study aims to ascertain the most accurate method among the five techniques to pinpoint the MPFL footprint on intraoperative fluoroscopy.

16. Schöttle PB, Schmeling A, Rosenstiel N, Weiler A (2007)Radiographic landmarks for femoral tunnel placement in medial patellofemoral ligament reconstruction. Am J Sports Med 35(5):801–804

Fig. 1 Schematic representation of fluoroscopic methods to identify the femoral footprint of the MPFL. Posterior femoral cortical line (line 1, yellow line), posterior femoral condylar line (line 2, orange line), and posterior Blumensaat line (line 3, green line). The black, blue, red, purple, and green dots indicate the distance and location of the fluoroscopy methods from line 1, line 2, and line 3, respectively, on the true lateral knee fluoroscopic image

Materials and methods

Study design and specimens

A total of 44 well-preserved dry femur bones were selected from a collection of 142 dry femur bones housed in the Akdeniz University Clinical Anatomy Laboratory and were included in this study. Bones with unclear anatomical landmarks and broken or damaged bones were excluded from the study. None of the bones included in the study were found to have trochlear dysplasia through macroscopic examination. The sex and age at the time of death of the dry femur bones were unknown. The local ethics committee approved the study protocol (20.07.2023-10/10).

Determination of MPFL footprint on dry bones

A committee consisting of an anatomist and two senior orthopedic surgeons examined all bones and collaboratively identified the femoral MPFL insertion site using bony landmarks. The groove called the saddle sulcus between the adductor tubercle, gastrocnemius tubercle, and medial epicondyle reported in previous anatomic studies was accepted as the MPFL attachment site, and its center was marked with a metal 9 mm thumbtack (Fig. 2) [22, 23].

Fig. 2 a The anatomic landmarks used to identify the MPFL femoral attachment. b Appearance of the metal thumbtack with a head diameter of 9 mm

Fluoroscopic imaging

Fluoroscopic imaging of the specimens was performed using c-arm fluoroscopy (Genoray, Fluoroscopic X-ray System, Model: Zen-500, Gyeonggi-do, Korea). The distance between the X-ray source and the image intensifier was 80 cm. The deepest part of the trochlea aligned at the midpoint of this distance and the thumbtack was centralized to the image intensifier. Standard true lateral knee images were obtained while the X-ray source was on the lateral side. The shots were taken with a range of 45–52 kVp and 0.4–1.8 mA, respectively. The fluoroscopy set-up is shown in Fig. 3. To obtain an image with the exact overlap of the femoral condyles, 5–10 images were acquired for each femur, and the most precise image was selected for the study. On a true lateral femur radiograph, the posterior femoral condyles should overlap and appear as a single condyle. All images with a double contour appearance were excluded.

Fig. 3

a The position of the femur and the fluoroscopic set-up.

b The dry femur specimen was positioned at the midpoint of the distance between the X-ray source and the image intensifier.

c A true lateral knee fluoroscopic image was obtained

Radiographic measurements

Digital images were saved in DICOM format and imported into RadiAnt DICOM Viewer software (Medixant Ltd., Poland, V.2023.1). Three reference lines were drawn on the digital images as described by Schöttle et al.: posterior femoral cortical line (Line 1), posterior femoral condylar line (Line 2), and most posterior of the Blumensaat line (Line 3) (Fig. 1). The fluoroscopic MPFL footprint was then marked using the distances reported by Schöttle et al. [16], Redfern et al. [18], Wijdicks et al. [20], Barnett et al. [19], and Kaipel et al. [21]. The distance between the center of the metal thumbtack, the anatomic MPFL point, and these points was measured. Since the head of the metal pin is known to be 9 mm, the measured distances were corrected accordingly to calibrate the magnification.

In addition, a coordinate system was established using the same radiographic reference lines to understand the direction of the deviation. The coordinates of the fluoroscopic MPFL point were set as the origin (x: 0, y: 0), and the coordinates of the anatomic MPFL pins were determined and plotted on the coordinate grid. This enabled evaluation of both the extent and the spatial direction of the deviation. Two criteria were set to measure the accuracy in locating the MPFL anatomical footprint. Schöttle et al. reported that the anatomic MPFL footprint was within a 5 mm diameter circle on their radiologic method and suggested that this distance did not negatively disrupt the isometry of the ligament in the light of previous biomechanical studies [16]. However, Servien et al. later revised this to a 7 mm diameter, aligning with the standard 7 mm drill used for creating the femoral tunnel, and considered a deviation of up to 7 mm as acceptable [24]. Thus, variations exceeding 5 mm (Criterion 1) and 7 mm (Criterion 2) from the midpoint of the radiographic compared to the anatomical MPFL imprints were deemed to be outside the normal range [16, 24].

Two independent observers who were orthopedic surgeons performed the measurements once. Observers were blinded to their own and the other observer‘s measurements. Interobserver reliability was tested using the interclass correlation coefficient (ICC). ICC values were scored according to the following criteria: an ICC value below 0.5 indicates poor reliability, between 0.5 and 0.75 indicates moderate reliability, between 0.75 and 0.9 indicates good reliability, and above 0.9 indicates excellent reliability [25]. Interobserver reliability was good and excellent for all measurements; thus, the mean of observers’ measurements was used for the final analysis (Table 1).

Table 1 Interobserver reliability of distance measurements between anatomic and radiographic MPFL footprint

Fluoroscopic methods ICC (95% CI) Interpretation

Redfern et al. (mm ± SD) 0.868 (0.771–0.926) Good

Barnett et al. (mm ± SD) 0.893 (0.813–0.940) Good

Kaipel et al. (mm ± SD) 0.823 (0.699–0.900) Good

Wijdicks et al. (mm ± SD) 0.947 (0.905–0.971) Excellent

Schottle et al. (mm ± SD) 0.901 (0.826–0.945) Excellent

SD standard deviation, ICC interclass correlation coefficient, CI confidence interval

Statistical analysis

Statistical evaluations were conducted using SPSS Statistics Base v.23 on Windows. Continuous data were characterized by mean, standard deviation, and range, whereas categorical data were represented in frequencies and percentages. The normality of continuous variables was ascertained using the Kolmogorov–Smirnov test. Parametric tests were implemented for data following a normal distribution, and nonparametric tests were chosen for those not fitting this distribution. The paired sample t test was adopted to contrast continuous variables, and the chi-squared test was designated for comparisons of categorical data. A p value less than 0.05 was considered to indicate statistical significance.

Results

Comparison of distance between anatomic and fluoroscopic MPFL points

Table 2 delineates the comparative analysis of distances between the anatomic and the fluoroscopic MPFL points as determined by five methods. The Schöttle method demonstrated the most proximate mean distance of 3.2 ± 1.2 mm. In contrast, the Wijdicks method manifested the most disparate mean distance, registering at 6.8 ± 1.8 mm. Utilizing a one-way ANOVA, a statistically significant variation in distances was identified across the techniques (p = 0.001). Further post hoc comparisons using the Tukey test revealed several pairs of methods with significant differences. Although the Schöttle method was similar to the Redfern method, it significantly differed from the other three methods. Figure 4 shows the accuracy of the methods on the coordinate system.

Table 2 Comparison of distance between the anatomic MPFL and the fluoroscopic MPFL points according to the methods

Fluoroscopic methods Distance (mm ± SD) p value

1 Redfern et al. 3.6 ± 1.5 0.0011

2 Barnett et al. 4.4 ± 2.3

3 Kaipel et al. 6.1 ± 2.1

4 Widjick et al. 6.8 ± 1.8

5 Schottle et al. 3.2 ± 1.2

Post hoc multiple comparisons

Pairs p value Pairs p value

1 vs. 2 0.3302 2 vs. 4 0.0012

1 vs. 3 0.0012 2 vs. 5 0.0332

1 vs. 4 0.0012 3 vs. 4 0.3952

1 vs. 5 0.8452 3 vs. 5 0.0012

2 vs. 3 0.0012 4 vs. 5 0.0012

1ANOVA, 2Tukey test. p ＞ 0.005 is significant. Bold p values are significant

Fig. 4 The distribution of fluoroscopic MPFL points in accordance to anatomic MPFL point (yellow dot). The coordinates of the anatomic MPFL point is (0, 0)

Anatomic MPFL point detection rates: a comparative analysis based on 5 mm and 7 mm criteria

Table 3 presents the MPFL femoral tunnel placement accuracy within the acceptance limits of 5 and 7 mm. Under the stringent 5 mm criteria, the Schöttle method emerged as the most precise, boasting an impressive in-point detection rate of 90.9%. In stark contrast, the Wijdicks method exhibited the least precision, with an in-point detection rate of 11.4%. A chi-squared test confirmed the presence of a significant statistical difference in the detection rates among the methods (p = 0.001). Further analysis through post hoc multiple comparisons highlighted pronounced differences between several method pairings. Similar to distance measurements, the Schöttle method was similar to the Redfern method but significantly better than the others. In the context of the 7 mm criteria, the Redfern technique demonstrated unparalleled precision, achieving an impeccable in-point detection rate of 100%.

Table 3 Comparison of MPFL femoral tunnel placement accuracy within 5 and 7 mm acceptance limits

Variables 5 mm criteria p value

Out n (%) In n (%)

1. Redfern et al. 8 (18.2%) 36 (81.8%) 0.0011

2. Barnett et al. 18 (40.2%) 26 (59.1%)

3. Kaipel et al. 30 (68.2%) 14 (31.8%)

4. Wijdicks et al. 39 (88.6%) 5 (11.4%)

5. Schottle et al. 4 (9.1%) 40 (90.9%)

Post hoc multiple comparisons

Pairs p value Pairs p value

1 vs. 2 0.052 2 vs. 4 ＜ 0.000

1 vs. 3 ＜ 0.000 2 vs. 5 0.001

1 vs. 4 ＜ 0.000 3 vs. 4 0.036

1 vs. 5 0.352 3 vs. 5 ＜ 0.000

2 vs. 3 0.018 4 vs. 5 ＜ 0.000

7 mm criteria p value

Out n (%) In n (%)

1. Redfern et al. 0 (0%) 44 (100%) 0.0011

2. Barnett et al. 9 (20.5%) 35 (79.5%)

3. Kaipel et al. 14 (31.8%) 30 (68.2%)

4. Wijdicks et al. 21(47.7%) 23 (52.3%)

5. Schottle et al. 1 (2.3%) 43 (97.7%)

Post hoc multiple comparisons

Pairs p value Pairs p value

1 vs. 2 0.004 2 vs. 4 0.012

1 vs. 3 ＜ 0.000 2 vs. 5 0.014

1 vs. 4 ＜ 0.000 3 vs. 4 0.190

1 vs. 5 1.000 3 vs. 5 ＜ 0.000

2 vs. 3 0.331 4 vs. 5 ＜ 0.000

Out indicates the outside of the 5 mm circle, and In indicates the inside of the 5 mm circle

aChi-squared test. 2Bonferroni correction p ＞ 0.005 is significant. Bold p values are significant. Numbers in the pairs column designate the fluoroscopic methods

Conversely, the Wijdicks technique revealed the highest propensity for error, with an out-point detection rate of 47.7%. A subsequent chi-squared test validated the significant variability in detection rates among the techniques (p = 0.0011). Further in-depth post hoc analysis illuminated the Redfern method‘s discernible differences when contrasted against the Barnett, Kaipel, and Wijdicks methods but not the Schöttle method.

Discussion

Successful MPFLR requires careful positioning of the femoral tunnel to restore normal patellofemoral kinematics and minimize postoperative complications. However, a key challenge is the accurate intraoperative identification of the MPFL femoral footprint, a prerequisite for optimal outcomes. This study aimed to determine the most accurate of the five fluoroscopic methods used for intraoperative identification of the MPFL footprint. Our results demonstrated the superiority of the Schöttle method in accurately identifying the anatomic MPFL point. While the Redfern method excelled in the 7 mm criteria with perfect accuracy (100%), the Schöttle method appears to be the most consistent across the board, having the smallest distance in the MPFL point comparison and the highest accuracy rate in the 5 mm criteria. It also performed accurately (97.7%) in the 7 mm criteria. Therefore, based on the combined evaluation of distance and detection rate accuracy, the Schöttle method can be considered the best overall among the tested methods.

成功的MPFLR需要仔细定位股骨隧道以恢复正常的髌股运动并减少术后并发症。然而，一个关键的挑战是术中准确识别MPFL股骨足印，这是获得最佳结果的先决条件。本研究旨在确定五种用于术中识别MPFL足迹的透视方法中最准确的方法。我们的结果证明了Schöttle方法在准确识别解剖性MPFL点方面的优越性。虽然Redfern方法在7毫米标准中具有完美的准确性(100%)，但Schöttle方法似乎是最一致的，在MPFL点比较中距离最小，在5毫米标准中准确率最高。在7 mm标准中，准确率为97.7%。因此，基于距离和检出率准确率的综合评价，Schöttle方法可以被认为是所有测试方法中综合效果最好的方法。

There may be several reasons for the variation in recommendations for fluoroscopic detection of the MPFL footprint in the relevant literature. However, the most significant reason for the diversity is the individual variations in knee anatomy. As with all anatomical structures, the femoral attachment site of the MPFL is subject to variations in location and shape. These findings have been demonstrated in many previous anatomic studies. In their systematic review, Aframian et al. reviewed a total of 67 anatomic studies and reported that 16 different locations of the MPFL femoral attachment site were defined [26]. In another systematic review, Placella et al. reviewed 13 anatomic studies, including 312 knees, and found that the femoral insertion was at the adductor tubercle in 29.6% of cases, the medial epicondyle of the femur in 17.8%, and other sites in the remaining 44% [27]. In addition to location, the shape of the MPFL footprint and adjacent anatomic landmarks are variable. Dandu et al. investigated the topography of landmarks used in MPFLR and found significant variability in their morphology and spatial relationship to the MPFL footprint. The medial epicondyle (ME) showed significantly greater variance in volume compared to the adductor tubercle (AT) and gastrocnemius tubercle (GT), which were more consistent in morphology [28]. These findings suggest that the femoral attachment site of the MPFL is subject to frequent anatomical variation. Defining an anatomical structure with a single constant point is challenging given the diverse range of anatomical variations.

在相关文献中，对MPFL足迹的透视检测建议的变化可能有几个原因。然而，造成这种差异的最重要原因是膝关节解剖结构的个体差异。与所有解剖结构一样，MPFL的股骨附着部位在位置和形状上也会发生变化。这些发现已在许多先前的解剖学研究中得到证实。在他们的系统综述中，Aframian等人回顾了总共67项解剖学研究，并报道了MPFL股骨附着部位的16个不同位置的定义[26]。Placella等人在另一篇系统综述中回顾了13项解剖学研究，包括312个膝关节，发现29.6%的病例股骨止点位于内收肌结节处，17.8%的病例位于股骨内上髁处，其余44%的病例位于其他部位[27]。除了位置，MPFL足迹的形状和邻近的解剖标志是可变的。Dandu等人研究了MPFLR中使用的地标的地形，发现它们的形态和与MPFL足迹的空间关系存在显著差异。内侧上髁(ME)的体积差异明显大于内收肌结节(AT)和腓肠肌结节(GT)，后者在形态上更为一致[28]。这些结果表明，强腓肠肌韧带的股骨附着部位经常发生解剖变异。考虑到解剖变异的不同范围，用单个恒定点定义解剖结构是具有挑战性的。

The second reason could be the methodological differences between the studies, mainly regarding sample preparation and imaging. The MPFL is not a true ligament but a functional layer of the medial retinaculum [29]. The dissection of these structures is complex, and unfortunately, several studies do not provide details of the dissection. Some authors in earlier studies could not even clearly identify the MPFL in their dissections [2, 30]. Again, radiographic imaging is subject to several intrinsic errors. It is difficult to obtain a true lateral knee radiograph. In our study, a minimum of 5–10 images were taken to obtain perfect condylar overlap. Balcerek et al. studied how minor deviations from a true lateral fluoroscopic view impact the accuracy of femoral tunnel placement. They showed that even slight deviations (2.5°–5°) in positioning led to significant shifts in the femoral MPFL insertion point, emphasizing the critical need for precision in achieving a true lateral view during surgery [17]. In addition, magnification and calibration errors can alter the result in an area where distance measurements are minute. For example, the point described by Redfern et al. is only 0.5 mm from the posterior cortical line [18]. Wijdicks et al. criticized Schottle’s methodology for not using proper specimen calibration and not performing reliability between examiners to validate their measurements [20]. Distance measurement is also affected by bone size. A 13-year-old girl and a 17-year-old boy will have different knee sizes and, therefore, different distances to the landmarks. The reliability of the measurements is another source of error [31]. Identifying guide points and drawing guidelines are subjective and thus prone to interobserver variations. Finally, almost all these studies were performed on a limited number of cadavers without apparent abnormality. On the other hand, the accuracy of fluoroscopy-guided tunnel placement, particularly in knees with severe trochlear dysplasia (Types C and D), is markedly decreased compared to mild trochlear dysplasia [32].

第二个原因可能是研究之间的方法差异，主要是关于样品制备和成像。MPFL不是真正的韧带，而是内侧支持带的功能层[29]。这些结构的解剖是复杂的，不幸的是，一些研究没有提供解剖的细节。在早期的研究中，一些作者甚至不能清楚地在他们的解剖中识别MPFL[2,30]。再一次，放射成像受到几个固有误差的影响。很难获得真正的膝侧位X线片。在我们的研究中，至少需要5-10张图像才能获得完美的髁重叠。Balcerek等人研究了真实侧位透视视图的微小偏差如何影响股骨隧道放置的准确性。他们发现，即使定位有轻微偏差(2.5°-5°)，也会导致股骨MPFL插入点发生显著移位，这强调了在手术中精确获得真正的侧位视图的关键需求[17]。此外，在距离测量极小的区域，放大和校准误差会改变结果。例如，Redfern等人描述的点距皮质后线仅0.5 mm[18]。Wijdicks等人批评Schottle的方法没有使用适当的标本校准，也没有在检查员之间执行可靠性来验证他们的测量[20]。距离测量也受骨骼大小的影响。一个13岁的女孩和一个17岁的男孩膝盖大小不同，因此到地标的距离也不同。测量的可靠性是另一个误差来源[31]。确定引导点和绘制指导方针是主观的，因此容易在观察者之间发生变化。最后，几乎所有这些研究都是在数量有限的尸体上进行的，没有明显的异常。另一方面，与轻度滑车发育不良相比，透视引导下隧道放置的准确性明显降低，特别是对于严重滑车发育不良(C型和D型)的膝关节[32]。

Based on these challenges, several authors advocated individualized detection of MPFL footprint [22, 33, 34]. Previous anatomical dissection studies have shown that the groove between the AT, GT, and medial epicondyle, also known as the saddle sulcus, is a constant area for the MPFL footprint, and this area can be visualized or palpated through an open dissection [22, 23, 35–38]. Although Zang et al. [15] supported the sulcus localization technique as a reliable and accurate method for MPFL femoral tunnel positioning, Abreu-E-Silva et al. [39] highlighted significant inaccuracies with the palpation method, even when performed by experienced surgeons. Similarly, Koh and Zimmerman reported a palpation error rate of approximately 20% [40]. A recent systematic study by Heindel et al. comparing open and fluoroscopic techniques showed no significant difference in complications or outcomes [41]. Nonetheless, it is crucial to acknowledge that in particular cases, especially during revision surgeries, detecting anatomical landmarks may prove problematic due to altered anatomy. In such instances, reliance on fluoroscopic guidance becomes vital and potentially the sole viable option.

基于这些挑战，一些作者主张个体化检测MPFL足迹[22,33,34]。先前的解剖解剖研究表明，AT, GT和内侧上髁之间的沟，也称为鞍沟，是MPFL足迹的恒定区域，该区域可以通过开放解剖可视化或触诊[22,23,35 - 38]。尽管Zang等人[15]支持沟定位技术作为MPFL股骨隧道定位的可靠和准确的方法，但abure - e - silva等人[39]强调了触诊方法的显著不准确性，即使是由经验丰富的外科医生进行的。同样，Koh和Zimmerman报告的触诊错误率约为20%[40]。Heindel等人最近进行的一项比较开放和透视技术的系统研究显示，并发症或结果没有显著差异[41]。然而，重要的是要认识到，在特殊情况下，特别是在翻修手术中，由于解剖结构的改变，检测解剖标志可能会证明是有问题的。在这种情况下，依赖透视引导变得至关重要，并且可能是唯一可行的选择。

The present study has several notable strengths and inherent limitations. The primary limitation arises from using dry femoral bone devoid of soft tissues. Thus, the identification of the MPFL footprint was based on bone landmarks rather than the ligament itself. The variability between the bone landmarks and the MPFL footprint introduces a potential bias. However, in many previous anatomical and radiological studies, the saddle sulcus has been identified and widely accepted as the site of MPFL attachment [22, 23, 35–38]. There was no trochlear dysplasia in the femurs included in the study. Considering that almost all patients undergoing MPFLR have varying degrees of trochlear dysplasia, the findings may not be generalizable to patients with patellofemoral instability. However, nearly all studies, including Schöttle et al., were performed on normal knees. Despite a rigorous methodological approach, errors in acquiring fluoroscopic images and measurements cannot be excluded. Multiple fluoroscopic images were obtained to minimize these errors, and the best true lateral knee radiograph was selected. In addition, measurements were made by different observers and used after being found reliable. Length measurements were also calibrated to eliminate magnification effects.

本研究有几个显著的优势和固有的局限性。主要的限制是使用没有软组织的干股骨。因此，MPFL足迹的识别是基于骨标记而不是韧带本身。骨标记和MPFL足迹之间的可变性引入了潜在的偏差。然而，在许多先前的解剖学和放射学研究中，鞍沟已被确定并被广泛接受为MPFL附着部位[22,23,35 - 38]。本研究中未发现股骨滑车发育不良。考虑到几乎所有接受MPFLR的患者都有不同程度的滑车发育不良，该结果可能不适用于髌骨不稳患者。然而，几乎所有的研究，包括Schöttle等，都是在正常的膝盖上进行的。尽管有严格的方法方法，在获得透视图像和测量误差不能排除。获得多个透视图像以尽量减少这些误差，并选择最佳的真实侧膝X线片。此外，由不同的观测者进行测量，并在发现可靠后使用。长度测量也经过校准，以消除放大效应。

Conclusions

In conclusion, the comprehensive analysis of this study highlights the Schöttle method as the most accurate and consistent technique for identifying the MPFL femoral footprint using intraoperative fluoroscopy. With the smallest mean distance between the anatomic and radiographic MPFL footprints and a high in-point detection rate, it emerges as the superior approach among the five methods evaluated. Future research should focus on refining these fluoroscopic techniques to cater to the individual variations in knee anatomy, particularly in patients with severe trochlear dysplasia.

总之，本研究的综合分析强调Schöttle方法是术中透视识别MPFL股骨足迹最准确和一致的技术。由于解剖和放射学MPFL足迹之间的平均距离最小，并且点内检出率高，因此在评估的五种方法中，它成为优越的方法。未来的研究应侧重于改进这些透视技术，以适应膝关节解剖的个体差异，特别是严重滑车发育不良的患者。

Abbreviations

MPFL Medial patellofemoral ligament

MPFLR Medial patellofemoral ligament reconstruction

TT–TG Tibial tubercle–trochlear groove