兽研1. 引言
髌骨脱位是犬最常被诊断的骨科疾病之一[1,2]，而在猫中，其发生率被认为较低[3]。当存在时，临床症状可能包括疼痛、跛行、患肢负重减少、跳跃能力下降以及膝关节伸展受限[3,4,5]。髌骨脱位可分为四个不同的等级，1级代表最轻微的形式，其中髌骨可被手动脱位，一旦停止操作即返回滑车沟，而4级代表该病症最严重的形式，其中髌骨保持脱位状态且无法被手动复位至滑车沟[2,3]。
在家猫中，内侧髌骨脱位可能源于创伤性或发育性原因[6]。与发育性髌骨脱位相关的主要异常包括滑车沟变平、胫骨粗隆内移、股骨近端前倾角减小以及其他肢体角度畸形[6,7]。然而，在猫中，先天性异常主要与滑车沟变平和股骨内侧髁发育不良相关[7,8]。因此，股骨滑车形态的评估在手术规划中起着重要作用。股骨滑车成形术是猫髌骨脱位手术治疗中主要采用的手术之一，旨在加深滑车沟并改善髌骨稳定性[9,10,11,12]。该手术的目标是将50%的髌骨厚度容纳于滑车沟内[13]；然而，关节暴露可能导致关节软骨损伤，从而可能损害关节功能[14]。
由于是否进行滑车成形术取决于对滑车形态的准确评估，股骨滑车的评价在髌骨脱位的手术规划中起着基础性作用。为此，已采用了不同的影像学方法，每种方法各有其特定的优势和局限性。在犬中，计算机断层扫描能够对骨性结构进行详细评估，并允许对股骨滑车进行客观测量，尽管其在关节软骨评估方面存在局限性[15,16]。此外，它是一种成本较高、需要全身麻醉且可及性有限的影像学检查。相比之下，超声检查可以表征关节软骨，提供对股骨滑车形态更全面的评估。然而，超声检查是一种操作者依赖性检查，在进行评估时应考虑到这一点[15,17]。此外，近期在犬中的研究表明，从切线位X线投影获得的滑车沟深度与股骨滑车的形态学评估之间存在正相关[18,19]。
尽管取得了这些进展，比较不同影像学方法测量猫滑车沟深度的研究仍然有限[7,20]，尤其是那些以离体评估作为参考标准的研究。因此，本研究旨在使用切线位X线摄影、90°关节屈曲下的计算机断层扫描以及45°和90°关节屈曲角度下的超声检查测定健康家猫的滑车沟深度，并将这些测量结果与45°和90°关节屈曲下离体评估所获得的结果进行比较。目的是提供可能支持猫髌骨脱位术前评估和手术规划的信息。本研究的假设是，使用不同影像学方法获得的测量结果将与通过离体评估（被视为金标准）获得的结果具有可比性。
2. 材料与方法
2.1. 选择与描述
本研究经南里奥格兰德联邦大学动物使用伦理委员会批准（方案编号44620）。共纳入来自15只成年猫的27条骨盆肢，年龄2岁或以上，平均体重3.34 ± 0.23 kg。这些动物因与研究无关的原因被安乐死。在纳入研究之前，所有肢体均通过骨科检查进行评估。存在骨科异常的肢体，如髌骨脱位、骨折、影像学半月板病变（半月板矿化）或韧带损伤，被排除。在最终阶段，来自三只不同猫的三条肢体因单侧半月板矿化被排除。所有临床评估均由一名兽医骨科专科医生（M.P.F.）完成。
使用常规正交投影进行X线评估，以仅选择无骨关节异常、骨折、开放性生长板或显著角度偏差证据的膝关节。选择后，尸体保持冷冻状态，并在实验程序前于室温下解冻24–36小时。
2.2. X线评估
X线检查使用数字X线摄影（DR）系统（Siemens®，德国慕尼黑）进行，并使用QXLink图像查看软件（版本3.3）进行评估。
对所有肢体均获取了切线位颅尾侧（天顶位）X线投影。膝关节保持最大屈曲位，X线探测器垂直放置于股骨颅侧，以显示股骨滑车沟。使用置于膝关节水平的2.54 cm圆形金属X线标记物获取X线图像。采用水平X线束获取天顶位投影图像（图1）。X线测量使用vPOP Pro软件（Veterinary Preoperative Orthopaedics Planning，版本2.9.2，VETSOS Education LTD.，Shrewsbury，英国）进行。对图像放大率进行校正以获得真实滑车沟深度（TGD）。TGD的测定方法为：绘制一条与股骨髁相切的水平线，再绘制一条垂直于第一条线的第二条线，该线位于股骨滑车中点并延伸至滑车沟底部。该垂线的长度记录为TGD（图1）。所有图像均由同一评估者（R.K.C.）进行评价。
2.3. 超声评估
超声检查使用HS30超声系统（Samsung®，首尔，韩国），配备10–16 MHz多频线阵探头。成像前对膝关节进行剃毛，并在皮肤上涂抹超声凝胶以减少伪影形成。增益、深度和聚焦设置根据每个膝关节单独调整。
超声评估在膝关节处于两个不同屈曲角度下进行。首先，膝关节屈曲45°，探头置于髌骨近端以测量滑车沟深度（TGD）。随后，关节屈曲90°，将探头置于髌骨远端获取第二次TGD测量值。
为确保肢体位置正确，使用放置在肢体外侧的200 mm数字角度计测定关节角度。以膝关节作为中心参考点，股骨大转子和跗骨分别作为近端和远端参考点。在所有超声评估过程中，探头均横向置于滑车沟并保持与股骨滑车垂直（90°）（图2）。评估过程中未对施加于探头的压力进行量化。滑车沟深度（TGD）的测定方法为：从滑车沟软骨表面的最深点至连接内、外侧滑车嵴最高点的参考线的垂直距离（图2）。图像由同一评估者（G.P.S.）获取并测量。每个膝关节的滑车沟测量一次，随后计算测量值的平均值。
2.4. 计算机断层扫描评估
计算机断层扫描图像使用8层Revolution ACT CT扫描仪（GE Healthcare®，芝加哥，伊利诺伊州，美国）获取。动物俯卧于泡沫槽中，骨盆肢置于尾侧，膝关节屈曲90°（图3）。以1.25 mm层厚获取容积图像，允许进行多平面和三维重建。图像采集后，使用RadiAnt DICOM Viewer软件（版本2023.1）分析DICOM文件。使用多平面重建工具测量滑车沟深度，该工具可获取轴面图像并识别滑车深度。滑车沟深度在髌骨区域远端测量。图像使用计算机断层扫描骨窗设置进行重建。所有图像均由同一评估者（R.K.C.）使用上述相同设置获取并评价。
滑车沟深度的测定方法为：绘制一条与股骨髁相切的水平线，再绘制一条垂直于第一条线的第二条线，该线从滑车沟中央部分延伸至股骨滑车底部。垂线的长度用作滑车沟深度的测量值（图4）。
2.5. 关节切开术与卡尺测量
完成影像学评估后，通过关节切开术暴露滑车沟以进行直接原位测量。使用数字卡尺（Davidson®，柏林，德国；显示分辨率：0.01 mm）进行测量。关节位置和屈曲角度按照超声检查所建立的方案维持。通过位于膝关节外侧的纵向皮肤切口进入关节，随后将髌骨向内侧脱位以显露关节结构。测量过程中，卡尺保持与股骨滑车垂直，再现超声评估时所用的方向（图5）。
2.6. 数据记录与统计分析
所有图像在评估前均经过随机化处理，且所有评估者均无法获取其他评估者的测量结果。所有数据录入Microsoft Excel，随后导出至SPSS 20.0版本进行统计分析。分类变量以频率和百分比描述。定量变量的正态性采用Shapiro-Wilk检验评估。符合正态分布的定量变量以均值±标准差表示。测量结果在左右侧之间以及影像学检查与宏观分析之间采用配对Student t检验进行比较。不同影像学方法之间的比较采用重复测量方差分析（ANOVA），随后进行Bonferroni事后检验。所有比较均以p < 0.05为显著性水平。
3. 结果
最初，共有15只动物纳入研究。然而，仅27条骨盆肢纳入分析。3条肢体因半月板矿化被排除，分别在3只不同猫中单侧发现。混血猫占多数，占研究群体的85.72%（n = 12），其次为1只异国短毛猫（7.14%）和1只暹罗猫（7.14%）。雌性更为常见，占纳入研究动物的64.3%。平均体重为3.61 ± 0.80 kg。
通过宏观评估获得的股骨滑车沟测量值被视为滑车沟深度评估的金标准。因此，所有影像学技术均与解剖的股骨滑车沟（DFTG）进行比较。在影像学方法的比较中，放射学评估与90°离体测量之间（p = 0.157）以及90°超声评估与90°离体测量之间（p = 0.832）观察到相似性。不同影像学方法获得的滑车沟深度与解剖股骨滑车沟的比较数据见表1。左右肢之间平均股骨滑车沟深度的比较未显示统计学显著差异（表2）。在影像学方法之间的比较中，放射学与90°超声评估之间（p = 0.667）以及放射学评估与计算机断层扫描之间（p = 0.999）观察到相似性。所比较的影像学方法见表3。
4. 讨论
本研究采用三种不同影像学技术对无关节异常的猫进行滑车沟深度评估，并将其结果与被视为金标准的离体评估进行比较。因此，研究结果为该物种滑车沟深度的评估提供了新视角。分析结果后，研究假设被部分否定，因为仅在膝关节屈曲90°位时通过放射学和超声获得的测量值与90°离体测量值之间观察到与离体评估的相似性。
滑车沟深度的评估是猫髌骨脱位手术规划中的重要步骤[3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21]。这种重要性归因于患有髌骨脱位的猫与健康猫相比，其滑车沟往往相对较浅[7]。尽管已有数项研究描述了犬滑车沟的评估方法[22,23,24,25]，但猫科动物的研究仍然有限[7,8,9,10,11,12,13,14,15,16,17,18,19,20]，缺乏适用于该物种临床实践和手术规划的客观参数。
滑车成形术延长手术时间并增加患者发病率，因为其代表一项额外干预[26]。此外， prolonged手术操作可能增加手术部位感染率[27]。在犬中，与滑车成形术相关的并发症包括楔形滑车成形术后的骨软骨块移位以及块状滑车成形术后的内侧或外侧滑车嵴骨折[28,29]。通过影像学检查获得的定量值可能有助于更客观地评估滑车沟形态，减少视觉评估中固有的主观性。这种方法在滑车沟形态模棱两可、决定是否进行滑车成形术可能具有挑战性的临界临床病例中可能尤为重要。因此，能够在术前客观评估滑车沟深度的影像学方法可为手术规划提供额外信息，并有助于识别真正需要滑车成形术的患者，避免不必要的手术，降低发病率和手术并发症风险。
滑车沟深度的离体评估已在犬中作为与影像学方法（如超声和放射学）比较的金标准[22,25]。作者认为离体测量是此类比较的可靠参考方法，因为评估者可直接观察解剖结构和精确测量点。因此，使用离体测量作为参考方法对于验证不同影像学技术的准确性以及确定哪种方法与真实解剖结构一致性最高至关重要。
超声检查是一种操作者依赖性成像方式，检查者的经验可能影响图像采集[16]、测量表现，进而影响临床实践中结果的可重复性。然而，超声检查是一种无电离辐射的非侵入性成像方法，在大多数情况下无需镇静即可进行，从而提高了患者安全性[30,31,32,33]。除这一优势外，该技术还可评估关节软骨[33,34,35]，这一特征可能促成了在膝关节屈曲90°时超声测量与90°体位离体评估之间观察到的相似性。相比之下，在膝关节屈曲45°时获得的测量值之间未观察到相似性。对这一发现的一种可能解释是，在该角度下，髌骨相对于探头的位置可能因髌骨的部分重叠而妨碍对滑车沟的充分界定。此外，为尽量减少髌骨在滑车沟上方产生的后方声影伪影所需的探头角度微小变化可能影响测量，从而导致与45°离体评估相比观察到的差异。
膝关节天顶位X线投照用于评估犬滑车沟深度[25,35,36]；然而，作者未发现评估该投照在猫中适用性的研究。在本研究中，通过天顶位X线摄影获得的测量值显示与膝关节屈曲90°体位进行的离体评估相似。这一结果提示，X线摄影可能成为猫滑车沟深度术前评估的有用替代方法，尤其是在超声检查不可用时。无统计学显著差异（p = 0.157）可能与天顶位投照中X线束角度与90°屈曲离体评估时采用的关节体位之间的相似性有关，有利于滑车沟评估区域定位更为接近。相比之下，在膝关节屈曲45°体位进行的离体评估中未观察到这种相似性，提示关节屈曲程度可能影响滑车沟深度的测量。
在将计算机断层扫描测量值与离体评估获得的测量值进行比较时，观察到均值之间存在差异，计算机断层扫描在膝关节屈曲45°和90°时均获得较高的值。对这一发现的一种可能解释是使用了通道数较少、层厚较大的CT扫描仪，这些因素降低了空间分辨率和图像清晰度[37]，可能影响用于测量滑车沟深度的解剖标志的准确识别。此外，计算机断层扫描获得的测量平面与离体评估时使用的测量平面可能并非完全一致，使得方法之间的直接比较更具挑战性。因此，在手术规划中应谨慎解读计算机断层扫描测量值，因为微小差异可能影响关于是否需要滑车成形术的决策。作者认为，与本研究使用的设备相比，使用层厚更薄、各向同性体素重建的多探测器计算机断层扫描系统可能提高形态学测量的准确性。
计算机断层扫描获得的测量值与离体评估之间观察到的差异可能不仅与设备的技术参数有关，还与测量所采用的方法学有关。尽管多平面重建允许在轴位平面获取图像并识别滑车沟最深点，但重建平面方向或所选近远端层面的差异可能影响了获得的测量值。使用相互垂直平面重建和三维重建可能有助于更准确地识别滑车深度最大的区域并使测量标准化。因此，应谨慎解读CT结果，因为滑车沟深度测量的微小差异可能对手术规划和关于是否需要滑车成形术的决策产生影响。
本研究存在一些局限性，应予考虑。首先，样本量相对较小，这可能限制结果的普适性以及在猫滑车沟深度评估中确立明确参考值的能力。此外，未进行观察者内和观察者间重复性分析，因而无法评估采用不同影像技术所获得测量的可重复性和可靠性。再者，计算机断层扫描采集仅使用了单一层厚。计算机断层扫描采集参数的变化，尤其是层厚，可能影响空间分辨率和图像质量，进而可能影响滑车沟深度测量的准确性。未来需要纳入自然发生髌骨脱位猫的研究，以协助解读本研究获得的数据，并评估滑车沟构型及变平对猫髌骨脱位发生及严重程度的影响。
5. 结论
本研究结果表明，关节屈曲角度对猫滑车沟的测量具有重要影响。在45°屈曲时，超声、X线摄影和计算机断层扫描所获得的测量值与参考测量值存在显著差异；而在90°屈曲时，超声和X线摄影测量值与参考测量值的一致性更高。因此，在评估猫滑车沟时，应考虑标准化关节体位和屈曲角度。有必要开展更大样本量并评估观察者内和观察者间可靠性的进一步研究，以确定这些影像方法的可重复性和临床适用性。
查看英文原文▾
1. Introduction
Patellar luxation is one of the most frequently diagnosed orthopedic disorders in dogs [1,2], whereas in cats, its occurrence is considered less common [3]. When present, clinical signs may include pain, lameness, reduced weight-bearing on the affected limb, decreased jumping ability, and limited extension of the stifle joint [3,4,5]. Patellar luxation can be classified into four distinct grades, with grade 1 representing the mildest form, in which the patella can be manually luxated and returns to the trochlear groove once manipulation ceases, whereas grade 4 represents the most severe form of the condition, in which the patella remains luxated and cannot be manually repositioned into the trochlear groove [2,3].
In domestic cats, medial patellar luxation may be traumatic or developmental in origin [6]. The main abnormalities associated with developmental patellar luxation include trochlear groove flattening, medialization of the tibial tuberosity, reduced proximal femoral anteversion, and other angular limb deformities [6,7]. However, in cats, congenital abnormalities are primarily associated with trochlear groove flattening and hypoplasia of the medial femoral condyle [7,8]. Therefore, assessment of femoral trochlear morphology plays an important role in surgical planning. Femoral trochleoplasty is one of the main procedures used in the surgical treatment of patellar luxation in cats, aiming to deepen the trochlear groove and improve patellar stability [9,10,11,12]. The procedure aims to accommodate 50% of the patellar thickness within the trochlear groove [13]; however, joint exposure may result in damage to the articular cartilage, potentially compromising joint function [14].
Because the decision to perform trochleoplasty depends on an accurate assessment of trochlear morphology, evaluation of the femoral trochlea plays a fundamental role in the surgical planning of patellar luxation. For this purpose, different imaging modalities have been employed, each with specific advantages and limitations. In dogs, computed tomography enables detailed assessment of osseous structures and allows objective measurements of the femoral trochlea, although it has limitations in the evaluation of articular cartilage [15,16]. Additionally, it is an imaging examination associated with higher costs, the need for general anesthesia, and limited availability. In contrast, ultrasonography permits characterization of the articular cartilage, providing a more comprehensive assessment of femoral trochlear morphology. However, ultrasonography is an operator-dependent examination, which should be considered when performing the evaluations [15,17]. Furthermore, recent studies in dogs have demonstrated a positive correlation between trochlear groove depth obtained from tangential radiographic projections and the morphological evaluation of the femoral trochlea [18,19].
Despite these advances, studies comparing different imaging modalities for measuring trochlear groove depth in cats remain limited [7,20], particularly those using ex vivo assessment as the reference standard. Therefore, the aim of the present study was to determine trochlear groove depth in healthy domestic cats using skyline-view radiography, computed tomography at 90° of joint flexion, and ultrasonography at joint flexion angles of 45° and 90°, and to compare these measurements with those obtained from ex vivo evaluation at 45° and 90° of joint flexion. The objective was to provide information that may support preoperative assessment and surgical planning for patellar luxation in felines. The hypothesis of this study was that measurements obtained using the different imaging modalities would be comparable to those obtained through ex vivo evaluation, which was considered the gold standard.
2. Materials and Methods
2.1. Selection and Description
The study was conducted following approval by the Animal Use Ethics Committee of the Federal University of Rio Grande do Sul (protocol No. 44620). A total of 27 pelvic limbs from 15 adult cats, aged 2 years or older and with a mean body weight of 3.34 ± 0.23 kg, were included. The animals were euthanized for reasons unrelated to the study. Prior to inclusion in the study, all limbs were evaluated through an orthopedic examination. Limbs presenting orthopedic abnormalities, such as patellar luxation, fractures, radio-graphic meniscal lesions (meniscal mineralization) or ligament injuries, were excluded. At the final stage, three limbs from three different cats were excluded due to unilateral meniscal mineralization. All clinical evaluations were performed by a veterinarian specialized in veterinary orthopedics (M.P.F.).
Radiographic assessment was performed using conventional orthogonal projections to select only the stifle joints without evidence of osteoarticular abnormalities, fractures, open growth plates, or significant angular deviations. Following selection, the cadavers were kept frozen and subsequently thawed at room temperature for 24–36 h before the experimental procedures.
2.2. Radiographic Evaluation
Radiographic examinations were performed using a digital radiography (DR) system (Siemens®, München, German) and evaluated using QXLink image viewing software (Version 3.3).
A tangential craniocaudal (skyline) radiographic projection was obtained for all limbs. The stifle joint was maintained in maximal flexion, and the radiographic detector was positioned vertically cranial to the femur to allow visualization of the femoral trochlear groove. Radiographic images were obtained using a 2.54 cm circular metallic radiographic marker positioned at the level of the stifle joint. A horizontal X-ray beam was used to obtain the skyline projection images (Figure 1). Radiographic measurements were performed using vPOP Pro software (Veterinary Preoperative Orthopaedics Planning, version 2.9.2, VETSOS Education LTD., Shrewsbury, UK). Image magnification was corrected to obtain the true trochlear groove depth (TGD). TGD was determined by drawing a horizontal line tangential to the femoral condyles and a second line perpendicular to the first, positioned at the midpoint of the femoral trochlea and extending to the base of the trochlear groove. The length of this perpendicular line was recorded as the TGD (Figure 1). All images were evaluated by a single evaluator (R.K.C.).
2.3. Ultrasonographic Evaluation
Ultrasonographic examination was performed using an HS30 ultrasound system (Samsung®, Seoul, Republic of Korea) equipped with a 10–16 MHz multifrequency linear transducer. The stifles were clipped prior to imaging, and ultrasound gel was applied to the skin to minimize artifact formation. Gain, depth, and focal settings were individually adjusted for each stifle.
Ultrasonographic evaluation was performed with the stifle joint positioned at two different degrees of flexion. Initially, the stifle was flexed at 45°, and the transducer was positioned proximal to the patella to measure trochlear groove depth (TGD). Subsequently, the joint was flexed at 90°, and a second TGD measurement was obtained with the transducer positioned distal to the patella.
To ensure proper limb positioning, joint angulation was determined using a 200 mm digital goniometer placed on the lateral aspect of the limb. The stifle joint was used as the central reference point, while the greater trochanter of the femur and the tarsus were considered the proximal and distal reference points, respectively. During all ultrasonographic evaluations, the transducer was oriented transversely to the trochlear groove and maintained perpendicular (90°) to the femoral trochlea (Figure 2). The pressure applied to the transducer was not quantified during the evaluations. Trochlear groove depth (TGD) was determined as the perpendicular distance from the deepest point of the cartilaginous surface of the trochlear groove to a reference line connecting the highest points of the medial and lateral trochlear ridges (Figure 2). The images were acquired and measured by a single evaluator (G.P.S.). The trochlear groove was measured once in each stifle, and the mean of the measurements was subsequently calculated.
2.4. Computed Tomographic Evaluation
Computed tomography images were acquired using an 8-slice Revolution ACT CT scanner (GE Healthcare®, Chicago, IL, USA). The animals were positioned in sternal recumbency on a foam trough, with the pelvic limbs positioned caudally and the stifle joint flexed at 90° (Figure 3). Volumetric images were acquired with a slice thickness of 1.25 mm, allowing multiplanar and three-dimensional reconstructions. Following image acquisition, DICOM files were analyzed using RadiAnt DICOM Viewer software (version 2023.1). Trochlear groove depth was measured using the multiplanar reconstruction tool, which allowed the acquisition of axial plane images and the identification of trochlear depth. The trochlear groove depth was measured distal to the patellar region. The images were reconstructed using the bone window settings for computed tomography. All images were acquired and evaluated by a single evaluator (R.K.C.) using the same settings described above.
Trochlear groove depth was determined by drawing a horizontal line tangential to the femoral condyles and a second line perpendicular to the first, extending from the central portion of the groove to the base of the femoral trochlea. The length of the perpendicular line was used as the measurement of trochlear groove depth (Figure 4).
2.5. Arthrotomy and Measurement with a Caliper
After completion of the imaging assessments, the trochlear groove was exposed through arthrotomy to allow direct in situ measurement. A digital caliper (Davidson®, Berlin, Germany; display resolution: 0.01 mm) was used for this purpose. Joint positioning and flexion angles were maintained according to the protocol established for the ultrasonographic examination. The joint was accessed through a longitudinal skin incision located lateral to the stifle, followed by medial luxation of the patella to allow visualization of the articular structures. During measurement, the caliper was maintained perpendicular to the femoral trochlea, reproducing the orientation used during the ultrasonographic assessment (Figure 5).
2.6. Data Recording and Statistical Analysis
All images were randomized prior to evaluation, and none of the evaluators had access to the measurements performed by the others. All data were entered into Microsoft Excel and subsequently exported to SPSS version 20.0 for statistical analysis. Categorical variables were described using frequencies and percentages. The normality of quantitative variables was assessed using the Shapiro–Wilk test. Quantitative variables with a normal distribution were expressed as mean ± standard deviation. Measurements were compared between sides and between imaging examinations and macroscopic analysis using the paired Student’s t-test. Comparisons among imaging modalities were performed using repeated-measures analysis of variance (ANOVA), followed by Bonferroni post hoc testing. A significance level of p < 0.05 was considered for all comparisons.
3. Results
Initially, 15 animals were enrolled in the study. However, only 27 pelvic limbs were included in the analysis. Three limbs were excluded due to meniscal mineralization, identified unilaterally in three different cats. Mixed-breed cats predominated, accounting for 85.72% (n = 12) of the study population, followed by one Exotic Shorthair cat (7.14%) and one Siamese cat (7.14%). Females were more prevalent, representing 64.3% of the animals included in the study. The mean body weight was 3.61 ± 0.80 kg.
Measurements obtained from the macroscopic evaluation of the femoral trochlear groove were considered the gold standard for trochlear groove depth assessment. Therefore, all imaging techniques were compared with the dissected femoral trochlear groove (DFTG). During the comparison of the imaging methods, similarity was observed between the radiographic evaluation and the ex vivo measurement at 90° (p = 0.157), as well as between the ultrasonographic evaluation at 90° and the ex vivo measurement at 90° (p = 0.832). Comparative data regarding trochlear groove depth obtained from the different imaging modalities and the dissected femoral trochlear groove are presented in Table 1. Comparison of the mean femoral trochlear groove depths between the right and left limbs revealed no statistically significant differences (Table 2). During the comparison between the imaging modalities, similarity was observed between the radiographic and ultrasonographic evaluations at 90° (p = 0.667) and between the radiographic evaluation and computed tomography (p = 0.999). The imaging modalities compared are presented in Table 3.
4. Discussion
This study enabled the evaluation of trochlear groove depth in cats without joint abnormalities using three different imaging techniques, comparing their results with ex vivo evaluation, considered the gold standard. Therefore, the findings provide new perspectives for the assessment of trochlear groove depth in this species. After analysis of the results, the study hypothesis was partially rejected, since similarity with the ex vivo evaluation was observed only for measurements obtained by radiography and ultrasonography with the stifle positioned at 90° of flexion, compared with ex vivo measurements performed at 90°.
Assessment of trochlear groove depth represents an important step in the surgical planning of patellar luxation in cats [3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21]. This relevance is attributed to the fact that cats with patellar luxation tend to present a relatively shallower trochlear groove compared with healthy cats [7]. Although several studies have described methods for evaluating the trochlear groove in dogs [22,23,24,25], research in felines remains limited [7,8,9,10,11,12,13,14,15,16,17,18,19,20], with a lack of objective parameters applicable to clinical practice and surgical planning in this species.
Trochleoplasty prolongs surgical time and increases patient morbidity because it represents an additional intervention [26]. Furthermore, prolonged surgical procedures may increase the rate of surgical site infections [27]. In dogs, complications associated with trochleoplasty include osteochondral block migration following wedge trochleoplasty and medial or lateral trochlear ridge fractures following block trochleoplasty [28,29]. Quantitative values obtained through imaging examinations may contribute to a more objective assessment of trochlear groove morphology, reducing the subjectivity inherent in visual evaluation. This approach may be particularly relevant in borderline clinical cases in which trochlear groove morphology is equivocal and the decision to perform trochleoplasty may be challenging. Therefore, imaging methods capable of objectively assessing trochlear groove depth during the preoperative period may provide additional information for surgical planning and assist in identifying patients who truly require trochleoplasty, avoiding unnecessary procedures and reducing morbidity and the risk of surgical complications.
Ex vivo assessment of trochlear groove depth has been used in dogs as the gold standard for comparison with imaging methods, such as ultrasonography and radiography [22,25]. The authors consider ex vivo measurement a reliable reference method for this type of comparison, as the evaluator directly visualizes the anatomy and the exact measurement point. Therefore, the use of ex vivo measurements as a reference method is essential for validating the accuracy of different imaging techniques and determining which method shows the highest agreement with the true anatomical structure.
Ultrasonography is an operator-dependent imaging modality in which the examiner’s experience may influence image acquisition [16], measurement performance, and, consequently, the reproducibility of results in clinical practice. However, ultrasonography is a non-invasive imaging method free of ionizing radiation that, in most cases, can be performed without the need for sedation, providing greater patient safety [30,31,32,33]. In addition to this advantage, the technique allows evaluation of articular cartilage [33,34,35], a characteristic that may have contributed to the similarity observed between ultrasonographic measurements at 90° and ex vivo evaluation with the stifle positioned at 90°. In contrast, no similarity was observed between measurements obtained with the stifle positioned at 45°. A possible explanation for this finding is that, at this angle, the position of the patella in relation to the transducer may hinder adequate delimitation of the trochlear groove due to partial superimposition of the patella. Furthermore, small variations in transducer angulation, required to minimize the posterior acoustic shadowing artifact produced by the patella over the trochlear groove, may influence the measurements, contributing to the discrepancy observed in relation to ex vivo evaluation at 45°.
The skyline radiographic projection of the stifle is used for the assessment of trochlear groove depth in dogs [25,35,36]; however, the authors are unaware of studies evaluating the applicability of this projection in cats. In the present study, measurements obtained by skyline radiography showed similarity with ex vivo evaluation performed with the stifle positioned at 90° of flexion. This result suggests that radiography may represent a useful alternative for the preoperative assessment of trochlear groove depth in felines, particularly when ultrasonography is unavailable. The absence of a statistically significant difference (p = 0.157) was likely related to the similarity between the angulation of the radiographic beam in the skyline projection and the joint positioning adopted during ex vivo evaluation at 90° of flexion, favoring a more similar localization of the trochlear groove assessment area. In contrast, this similarity was not observed in relation to ex vivo evaluation performed with the stifle positioned at 45° of flexion, suggesting that the degree of joint flexion may influence measurements of trochlear groove depth.
When comparing computed tomography measurements with those obtained from ex vivo evaluation, differences between the means were observed, with higher values obtained by computed tomography at both 45° and 90° of stifle flexion. A possible explanation for this finding is the use of a CT scanner with a lower number of channels and greater slice thickness, factors that reduce spatial resolution and image definition [37], potentially compromising the accurate identification of the anatomical landmarks used for measuring trochlear groove depth. In addition, the correspondence between the measurement plane obtained by computed tomography and that used during ex vivo evaluation may not have been exactly the same, making direct comparison between methods more challenging. Therefore, computed tomography measurements should be interpreted with caution during surgical planning, as small differences may influence the decision regarding the need for trochleoplasty. The authors believe that the use of multidetector computed tomography systems with thinner slice acquisition and isotropic voxel reconstruction may improve the accuracy of morphometric measurements compared with the equipment used in the present study.
The discrepancy observed between the measurements obtained by computed tomography and the ex vivo evaluations may be related not only to the technical parameters of the equipment but also to the methodology used for measurement. Although multiplanar reconstruction allowed the acquisition of images in the axial plane and identification of the deepest point of the trochlear groove, differences in the orientation of the reconstructed plane or in the selected proximodistal level may have influenced the measurements obtained. The use of reconstructions in mutually perpendicular planes and three-dimensional reconstructions could help to more accurately identify the region of greatest trochlear depth and standardize the measurements. Therefore, the CT results should be interpreted with caution, as small differences in trochlear groove depth measurements may have implications for surgical planning and the decision regarding the need for trochleoplasty.
This study has some limitations that should be considered. First, the sample size was relatively small, which may limit the generalizability of the results and the establishment of definitive reference values for trochlear groove depth assessment in cats. Additionally, intra- and interobserver repeatability analyses were not performed, preventing the evaluation of the reproducibility and reliability of the measurements obtained using the different imaging techniques. Furthermore, only a single slice thickness was used for computed tomography acquisition. Variations in computed tomography acquisition parameters, particularly slice thickness, may influence spatial resolution and image quality, potentially affecting the accuracy of trochlear groove depth measurements. Future studies involving cats with naturally occurring patellar luxation are needed to assist in the interpretation of the data obtained in this study, as well as to evaluate the impact of trochlear groove conformation and flattening on the occurrence and severity of patellar luxation in cats.
5. Conclusions
The results of this study indicate that the joint flexion angle has an important influence on the measurement of the trochlear groove in cats. At 45° of flexion, measurements obtained by ultrasonography, radiography, and computed tomography differed significantly from the reference measurements, whereas at 90° of flexion, ultrasonographic and radiographic measurements showed greater agreement with the reference measurements. Therefore, standardization of joint positioning and flexion angle should be considered when evaluating the trochlear groove in cats. Further studies with larger sample sizes and assessment of intra- and interobserver reliability are warranted to determine the reproducibility and clinical applicability of these imaging methods.