兽研[内容请查看原文]
引言
犬骶髂关节（CSIJ）构成腰盆区域的核心组成部分，介导中轴骨骼与骨盆肢之间的机械相互作用（Breit and Künzel 2001b; Dyce et al. 2010; Frewein et al. 2004; Salomon 2020）。其生物力学功能与关节形态和表面构型内在相关。因此，区分生理性变异与病理性改变具有根本重要性（Standring et al. 2021）。
骨盆和腰骶关节形态的品种特异性差异已有描述。与罗威纳等品种所描述的更倾斜的关节方向相比，德国牧羊犬中观察到的更矢状方向的骶髂关节与增加的颅尾向平移相关（Breit and Künzel 2001a）。这些形态特征可能促进早发性CSIJ改变的发生（Gembardt 1974; Gregory et al. 1986）。
发育因素进一步导致CSIJ的形态变异性。在大型犬中，已有文献记载骶翼内骨化中心（OC）的不完全融合，持续至成年期表现为骶骨滑膜表面区（SSA）的离散缺损或凹陷。有趣的是，此类凹陷可能增强局部关节交锁并减少颅尾向剪切力，从而在机械力学上处于劣势的关节中可能发挥稳定作用（Breit and Künzel 2001b）。然而，这些特征在诊断影像学中有时会被误判为病理性病变（Götz et al. 1993; Prassopoulos et al. 1999）。
除种内变异外，跨物种的比较解剖学研究为骶髂关节（SIJ）的独特形态变异提供了额外证据。在此背景下，SSA对应于耳状面，反映其耳状或新月形形态（Evans and Lahunta 2013）。尽管该术语已标准化，SSA的详细形态表现出显著的种间和种内变异性，在猫、马和人类中具有不同的形状类型（Dalin and Jeffcott 1986; Jesse et al. 2017; Pallandre et al. 2020）。这些变异与应力分布差异和疼痛患病率相关（Jesse et al. 2017）。
在人类中，基于计算机断层扫描（CT）成像已识别出多种SIJ变异，包括未融合骨化中心（NFOC）和副骶髂关节（ASIJ）。后者代表位于主关节背侧的额外关节连接，与退行性改变密切相关（Demir et al. 2007; El Rafei et al. 2018），也与下背痛相关（Hadley 1950; Kang et al. 2017; Prassopoulos et al. 1999; Rosa Neto et al. 2009）。
总之，SIJ形态高度可变，并与跨物种的关节力学、稳定性和临床疾病发展密切相关。因此，全面了解CSIJ的形态谱对于准确解读影像学表现和阐明关节病理的潜在易感性至关重要。
据此，本研究旨在：
描述SSA形状变异，以及
识别和量化发育性骨化异常（包括NFOC和ASIJ）
[内容请查看原文]
材料与方法
本研究使用与先前对CSIJ进行组织学和形态计量学分析相同的29例犬标本队列（表S1）。应用相同的解剖方案。该方案经莱比锡大学兽医学院伦理委员会评估并批准（EK 7/2021，2021年5月3日；Gensicke et al. 2026）。
从每只犬获取成对CSIJ，并细分为四个关节部分：左侧髂骨（LI）、左侧骶骨（LS）、右侧髂骨（RI）和右侧骶骨（RS）。
所有CSIJ均进行宏观形态学评估，并使用iPhone 14 Pro（Apple Inc., Cupertino, CA, USA；4800万像素传感器；最大图像分辨率：8064 × 6048像素）进行摄影记录。对每个关节部分（n = 116）评估SSA形状类型、骶骨SSA的NFOC存在情况以及ASIJ。
[内容请查看原文]
形态学分析
记录每个关节以及每个个体的每种形态特征的出现频率。通过比较左右CSIJ评估侧别相关分布模式，并在呈现双侧发现的标本中评估NFOC和ASIJ的对称性。
评估关节连接特征——包括表面质地、形状、曲率和髂骶一致性。形状类型的分类主要参照Pallandre et al.（2020）提出的分类方法。对于ASIJ病例，记录其相对于相应髂骨和骶骨关节部分的解剖定位。
[内容请查看原文]
常见表面变异的形态计量学特征
SSA、NFOC和ASIJ的形态计量学分析使用先前在Gensicke et al.（2026）中描述的相同二维方法进行。
[内容请查看原文]
ASIJ软骨的组织学
对25号犬的全部四个关节部分（LI、LS、RI、RS）进行了尾背侧ASIJ软骨的组织学评估（表S1），因为这是唯一可用的新鲜标本。首先，整个关节面在4%缓冲福尔马林中固定（Carl Roth GmbH & Co. KG，Karlsruhe，Germany）。随后，周围骨组织在Chelaplex（乙二胺四乙酸（EDTA）和三羟甲基氨基甲烷）（Dr. K. Hollborn & Söhne GmbH & Co. KG，Leipzig，Germany）作用下经超声波浴脱钙并逐步软化，直至浅层软骨可轻易分离。样品包埋于石蜡块中，每个样品切取约1 μm厚的单张矢状切片。四张切片采用苏木精-伊红（H&E）染色（Carl Roth GmbH & Co. KG，Karlsruhe，Germany）。苏木精将嗜碱性结构，特别是细胞核和核仁内的核酸（DNA和RNA），染成蓝色至紫色，而伊红将嗜酸性结构如细胞质蛋白和细胞外基质成分染成各种深浅的粉红色（Mescher 2018）。显微镜检查使用配备Axiocam 208彩色相机（Carl Zeiss Microscopy GmbH，Oberkochen，Germany）和Zen lite软件（版本3.13；Carl Zeiss Microscopy GmbH，Oberkochen，Germany）的光学显微镜（Axioscope 5，Carl Zeiss Microscopy GmbH，Oberkochen，Germany）进行。
根据已建立的组织形态学标准，包括细胞组成和细胞外基质特征，对组织类型进行鉴定和区分。软骨组织通过陷窝内软骨细胞的存在和软骨基质的特征性组织方式加以鉴定。纤维软骨的特征是富含胶原的基质，软骨细胞常呈行排列。透明软骨的特征是均匀的基质，内含圆形软骨细胞，偶见同源群形成。相比之下，弹性软骨基质内含有密集的弹性纤维网，软骨细胞同样位于陷窝内，与透明软骨相比表现出更灵活、更有弹性的结构（Liebich 2012）。
所有图像均按照我们先前研究中描述的方案进行数字化处理（Gensicke et al. 2026）。
[内容请查看原文]
ASIJ的统计分析
额外关节面面积（AASA）的测量数据汇总于Microsoft Excel 2016（Microsoft Corporation，Redmond，WA，USA），并使用jamovi（版本2.6.44.0，The jamovi project，Sydney，Australia）进行分析。正态性采用Shapiro–Wilk检验评估（p < 0.05）。由于AASA数据不服从正态分布，全程采用非参数方法，包括用于配对比较的Wilcoxon符号秩检验和用于相关性分析的Spearman秩相关系数（p < 0.05）。
AASA大小按犬、按关节部分和按解剖位置进行描述性汇总。在推断性分析中，AASA值按关节部分求和，以考虑ASIJ数量变异和双侧不完全出现的情况。
对称性采用配对比较评估，并在同一关节内的髂部和骶部之间进行相关性分析（LI–LS，n = 21；RI–RS，n = 22）。在表现为双侧ASIJ的标本中，还进行了左侧与右侧之间的额外比较（LI–RI；LS–RS；各n = 18）。
为评估ASIJ对CSIJ的相对贡献，将求和的AASA表示为总面积（TSA）的比例。
评估了求和AASA与SSA、滑膜外面积（ESA）、TSA和体型之间的关联。为进行体型标准化，计算每只犬的总AASA并与臀冠长（CRL）进行相关性分析。
选择该方法是为了考虑ASIJ存在的变异性和双侧不完全出现的情况。
性别相关差异（n = 13只雄性，n = 12只雌性）采用Mann–Whitney U检验评估（p < 0.05）。比较了绝对和比例AASA值（n = 24），并额外分析了CRL标准化后的总AASA。
[内容请查看原文]
结果
[内容请查看原文]
SSA形状类型
SSA形状类型根据其出现频率和其他物种中先前描述的形态进行分类（Dalin and Jeffcott 1986；Jesse et al. 2017；Pallandre et al. 2020）。
表1和图1总结了在116个关节部分中鉴定出的形态学类型。
TABLE 1. Synovial surface area shape types ( n = 116 portions).
| Shape type | |||||||
|---|---|---|---|---|---|---|---|
| Auricular | Bifoliate | Phrygian Cap | W | Crescent | Spatula | Trifoliate | |
| Left Iliac | 5 | 7 | 4 | 5 | 4 | 4 | 0 |
| Left Sacral | 8 | 6 | 10 | 1 | 2 | 1 | 1 |
| Right Iliac | 7 | 7 | 4 | 7 | 2 | 1 | 1 |
| Right Sacral | 8 | 7 | 8 | 3 | 2 | 0 | 1 |
| Total | 28 | 27 | 26 | 16 | 10 | 6 | 3 |
[图：左侧犬髂骨骶盆面内侧观示意图，修改自Schaller（2018）。不同形状]
耳状（24.14%）、双叶状（23.28%）和弗里吉亚帽状（22.41%）最为常见，而铲形（5.17%）和三叶状构型（2.59%）罕见。
在29个CSIJ中，对侧SSA形状有17个呈对称，最常见于双叶状和三叶状形态。相比之下，新月形和铲形构型的对称率最低，而耳状、弗里吉亚帽状和W形则呈均衡分布。
六只犬在全部四个关节部分均表现出相同的形态。四例观察到左右侧CSIJ形状不同的双侧对称性，而14只犬表现为单侧对称，五只犬表现为完全不对称。
区域分析仅显示轻微的 compartment 特异性倾向，无一致模式。在所有比较中，SSA大小遵循一致的等级：W形形态的平均表面积最大（313.12 ± 96.19 mm2），其次为双叶形（272.52 ± 221.45 mm2）和耳形（248.50 ± 89.02 mm2）。弗里吉亚帽样（198.36 ± 72.28 mm2）和新月形构型（195.53 ± 91.32 mm2）显示中间值，而铲形形态始终显示最小的SSA（105.25 ± 41.62 mm2）。这些模式在两侧、关节部分和 compartment 间一致，无证据表明与品种、体型或性别相关的聚集。
[内容请查看原文]
骶骨SSA的NFOC
识别NFOC的标准包括骶骨SSA在其尾侧三分之一处、骶翼与骶骨外侧部过渡区域出现部分或完全中断或折叠。这伴有颅侧和尾侧部分之间表面倾斜度的明显变化以及凹形凹陷的形成（图2）。
[图：左侧犬骶骨外侧观。(a) 滑膜表面未融合骨化中心（NFOC）之间的裂隙]
在58个CSIJ中识别出16个NFOC（27.59%），或在29个标本中识别出10个（34.48%），左右侧分布均等（每侧八个CSIJ）。在所有受累关节中，SSA典型的从颅外侧到尾内侧的走向被破坏。在尾侧三分之一，走向一致转变为从颅内侧到尾外侧排列。在每一例中，髂骨表面表现出相应的适应性倾斜变化，朝向颅内侧到尾外侧方向。
在左侧骶翼，这种构型在八个有关节间隙形成的关节中观察到四个，而在右侧骶翼，八个关节中观察到五个。在其余关节中，颅侧和尾侧区域之间一致存在几乎垂直的分界。
尾侧SSA部分的平均测量值在46.44 ± 18.12 mm2至68.89 ± 49.73 mm2之间。当标准化为TSA时，这些区域在骶翼上相对大于髂骨表面（19.99% vs. 16.70%）。
未检测到与品种或性别的相关性。然而，NFOC仅在冠臀长（CRL）超过740 mm的犬中检测到，除了一只法国斗牛犬（CRL：530 mm）例外。
[内容请查看原文]
副骶髂关节
[内容请查看原文]
形态学特征
在58个检查的CSIJ中，共识别出62个邻近SSA的ASIJ。ASIJ存在于29个左侧CSIJ中的21个和29个右侧CSIJ中的22个，表明总体对称分布（左侧CSIJ中30个ASIJ，右侧32个）。
在检查的关节中，58个CSIJ中有29个显示单个ASIJ，九个关节有两个ASIJ，五个关节有三个ASIJ。单个、双重和三个ASIJ的分布在髂骨侧和骶骨侧之间无差异。
在29只犬中观察到18只双侧发生，而仅四个CSIJ在两侧完全缺乏ASIJ。
共识别出六个解剖定位，均位于SSA背侧、髂骨和骶骨结节区域内。最常见的位置是CSIJ的尾背侧缘（n = 32），此处髂翼的骶盆面在颅背侧髂棘的尾内侧与骶骨结节的颅背侧方面相关节。紧邻该区域的颅侧，观察到颅背侧关节（n = 8）。
较少见的是，ASIJ位于ESA边缘的腹侧，靠近第一骶背孔（腹侧 n = 3；背侧 n = 5）。此外，ASIJ被观察为SSA在其颅腹侧（n = 4）和尾背侧肢（n = 10）的延伸（图3）。
[图：(a) 左侧犬髂骨骶盆面内侧观，图示副骶髂关节的不同定位]
ASIJ的髂骨和骶骨组件一致，形成平坦或凹凸关节。27.42%的所有ASIJ中存在局灶性骨性形成。AASA主要呈圆形至椭圆形，偶尔逐渐变细，近一半（48.39%）显示偏离主要关节的走向。
未观察到ASIJ的存在与性别、体型或品种之间的明显关联。
[内容请查看原文]
ASIJ软骨的组织学
25号犬全部四个尾背侧ASIJ的浅层由透明软骨组成（图4）。组织学上，软骨表现出均匀的细胞外基质，呈轻度嗜酸性染色。软骨细胞位于陷窝内，显示嗜碱性细胞核，染成深蓝色至紫色，而细胞质呈 faintly 嗜酸性。软骨细胞单个或排列成小簇，称为同源群。在任何样本中均未检测到指示纤维软骨或弹性软骨的组织学特征。
[图：左侧犬髂骨尾背侧副骶髂关节（ASIJ）透明软骨衬里的组织学表现]
[内容请查看原文]
ASIJ的形态计量学特征
所有AASA的平均表面积为42.83 mm2。最小的AASA（4.20 mm2）见于一只雄性德国牧羊犬杂交犬（9号）左侧髂骨的颅背侧，而最大的AASA（260.80 mm2）见于一只雌性伯恩山犬（30号）右侧髂骨的尾背侧。
在存在一个或多个ASIJ的犬中，每个个体的AASA汇总平均值为61.81 mm2。
就解剖定位而言，最大的AASA位于尾背侧（52.48 mm2），其次为背侧（47.22 mm2）、SSA背侧（39.89 mm2）和SSA腹侧（33.10 mm2）。最小的AASA位于腹侧（17.48 mm2）和颅背侧（14.46 mm2；图3）。
在左侧和右侧CSIJ内，髂骨与骶骨AASA之间均观察到强正相关（ρ = 0.9，p < 0.001）。在存在双侧AASA的标本中，LI与RI之间以及LS与RS之间呈中度相关（ρ ≤ 0.577，p ≤ 0.004）。
LI与LS之间未检测到显著差异（W = 171，p = 0.055），而RI显著大于RS（W = 210，p = 0.007）。在双侧病例中，右侧髂骨AASA显著较大（W = 31，p = 0.016），而骶骨部分无差异（W = 55，p = 0.196）。
总体而言，AASA对TSA的平均贡献值为7.25%，各部位比例分别为5.78%（LI）、6.37%（LS）、8.36%（RI）和8.48%（RS）。比例贡献的两两比较产生的相关模式与对称性分析中观察到的结果一致。
AASA在所有关节部位均与TSA呈显著正相关（ρ = 0.436–0.755，p = < 0.001–0.044；图5a），与ESA的相关性更强（ρ = 0.470–0.831，p = < 0.001–0.028）。与SSA未观察到显著相关性，但LI呈弱趋势（ρ = 0.403，p = 0.071）。每只犬的总AASA与CRL呈中度相关（ρ = 0.523，p = 0.009；图5b）。
[图：(a) 附加关节面面积（AASA）与总表面积（TSA）的关系（* ρ = 0.755，p < 0.001；** ρ = 0.6]
未检测到AASA形态测量学的性别特异性差异。
[内容请查看原文]
讨论
[内容请查看原文]
SSA形状类型
本研究表明，犬的SSA形态表现出相当大的变异性，超出了通常假设的耳状或新月形构型；在检查的116个表面中，仅有38个符合经典模式。所应用的形状分类基于先前在其他物种中描述的变异型（Dalin and Jeffcott 1986；Jesse et al. 2017；Pallandre et al. 2020）。
犬与猫科动物之间观察到最大的形态相似性，因为这两个物种均表现出双叶形、匙形和弗里吉亚帽样构型。相比之下，W形和三叶形构型仅在犬中被发现，据我们所知，迄今尚未在其他物种中报道。
在人类中，耳状SSA更常见于无骶髂（SI）疼痛的个体，而新月形表面更常与SI疼痛相关。这一观察提示SSA形态与SI疼痛发生之间可能存在关系。不同的关节形态被认为会导致不同的负荷分布和力传递模式。与更椭圆形的构型相比，新月形表面提供的负荷传递面积更小。因此，SSA减小可能将更大比例的机械力转移至相邻的稳定结构，如韧带装置。这增加了机械应力并促进退行性改变（Jesse et al. 2017）。
因此，SSA较小的犬——尤其是与弗里吉亚帽形、新月形或匙形相关的犬——可能更容易发生退行性改变，进而发生SI疼痛。此外，缺乏一致或对侧表面对称性的SSA（如在新月形和匙形构型中常见）可能导致有效接触面积进一步减小。这种减小可能导致单位表面积相对负荷增加，从而增加CSIJ上的机械应力，可能引起负荷分布改变和随后的生物力学失衡，这些概念在人类膝关节的相关讨论中也有类似阐述（Gannon et al. 2021；Imhauser et al. 2013）。综上所述，犬与猫、马和人类相似，SSA形态表现出 substantial 变异性（Dalin and Jeffcott 1986；Jesse et al. 2017；Pallandre et al. 2020）。如果谨慎应用源自人类医学的概念，以较小表面积和双侧或对侧表面对称性降低为特征的关节构型可能特别容易发生关节改变。
[内容请查看原文]
骶骨SSA的NFOC
以裂隙或折叠为特征的凹陷，伴有SSA后三分之一倾斜度的改变，对应四个与S1和S2横突相关的OC，以及两个位于其腹侧面的额外孤立OC。在生理条件下，这些OC通常在约25周龄时完全融合，据报道较小品种融合更早。该融合过程的中断即为NFOC，被描述为未成熟SSA对机械过载的反应性应答。通过增加相对关节面之间的局部交锁，NFOC被认为有助于关节稳定。由于小型品种可能缺乏孤立OC，该现象似乎主要发生在大体型犬。犬NFOC最早可在24周龄时检出，偶尔可持续至成年期（Breit and Künzel 2001b）。
人类中的类似发现主要报道于青少年和年轻成人，但患病率明显较低（Götz et al. 1993; Green et al. 2008; Prassopoulos et al. 1999）。在此背景下，NFOC通常被认为是非病理性的（Götz et al. 1993）。这一解释得到一份病例报告的支持，该报告排除了其作为SI疼痛原因的可能性，因为其 presumed 终生存在使得晚年出现疼痛不太可能（Green et al. 2008）。
然而，完全融合可能导致SSA的骨性突起，这可能促成退行性改变。在影像学研究中，NFOC初期可能模仿骨折（Prassopoulos et al. 1999），凸显了与骨折及其他退行性关节改变可靠鉴别的重要性。轴位CT似乎是实现此目的最合适的模态（Götz et al. 1993），其中NFOC通常表现为前上关节间隙内的三角形骨体（Prassopoulos et al. 1999）。
总之，犬骶骨SSA的NFOC似乎比人类更常见。尽管它们不被视为SI疼痛的直接原因，但其准确识别对于避免诊断误判至关重要。此外，它们甚至可能在功能上有助于CSIJ的稳定。
[内容请查看原文]
副骶髂关节
髂骨和骶骨结节共同构成ESA的骨性基础，二者不相称并被纤维软骨覆盖（Evans and Lahunta 2013）。因此，CSIJ被归类为复合关节，包括尾腹侧区域的动关节成分和颅背侧区域的联合成分（Gensicke et al. 2026）。
[内容请查看原文]
ASIJ的形态学特征
ESA内相称的AASA被识别的频率高于预期。这些表面呈平坦或凹凸构型，并被透明软骨覆盖。
类似的关节变异在人类中较少见，但已在解剖剥离和影像学研究中报道，患病率范围为6.8%至50%（CT 1.7%–25.8%，MRI 5%–11%）。这些结构被称为ASIJ，定义为位于原发动关节背侧的AASA（Cihan et al. 2023; Tok Umay and Korkmaz 2020; Trentadue et al. 2023; Wittram and Whitehouse 1995）。
ASIJ在人类中可双侧发生（4.4%–53%）（Ehara et al. 1988; El Rafei et al. 2018; Prassopoulos et al. 1999; Valojerdy and Hogg 1990），本犬类研究 likewise 观察到该模式，其中62%的病例显示双侧受累，主要为对称分布。值得注意的是，约四分之一的犬在单个CSIJ内存在多个ASIJ，这一特征在人类中尚未见报道（Valojerdy and Hogg 1990）。
人类中ASIJ最常报道的位置在很大程度上对应于犬罕见的腹侧ASIJ（4.76%）。更靠腹侧定位的人类ASIJ类似于犬中识别的背腹侧变异（11.29%）（Demir et al. 2007; Ehara et al. 1988; Prassopoulos et al. 1999; Trentadue et al. 2023）。
相比之下，犬中识别出四个额外的ASIJ定位，最常见于尾背侧关节缘（51.61%）。其他变异包括极背侧ASIJ（9.68%）以及腹侧（6.45%）和背侧（16.13%）超出常规SSA的延伸。
在犬中，跨越CSIJ的力传递受水平朝向躯干、后肢经骨盆向脊柱的推进力（Prieur 1980）以及运动过程中的额外旋转力（Breit and Künzel 2001b）影响。相比之下，人类直立姿势中以轴向压缩为主。四足动物中这种更复杂、更动态的负载环境可能促进ASIJ比主要垂直负载的人类SIJ更频繁、更广泛分布的适应性形成。
[内容请查看原文]
ASIJ的形态计量学特征
犬ASIJ尺寸显示出相当大的变异性，范围为14.46至52.48 mm²，相当于TSA的5.78%至8.48%。在人类中，报道的AASA落在可比范围内，平均值约为47 ± 36 mm²（Valojerdy and Hogg 1990），尽管也有描述相当大的表面积，范围为78.00至314.00 mm²（Hadley 1950）。
观察到的ASIJ尺寸变异性似乎主要归因于尺寸依赖性的异速生长缩放，这通过与体型（CRL）和关节尺寸（ESA和TSA）的相关性得以体现。此外，响应个体机械需求的负荷依赖性生物力学适应可能进一步促成了广泛的尺寸谱系。
与人类的研究结果（Valojerdy and Hogg 1990）相反，犬的AASA与SSA之间未能证明存在相关性。这一差异可能反映了与先前在步态变异背景下讨论的相同的物种特异性生物力学条件。
髂骨和骶骨AASA显著相关，在表现出双侧ASIJ的犬中，右侧和左侧AASA也显著相关。LI AASA倾向于大于LS AASA。在右侧，RI显著大于RS AASA。这种不对称性与先前报道的CSIJ形态学模式一致（Gensicke et al. 2026）。
骶骨ASIJ尺寸在两侧之间无显著差异。RI AASA大于LI的原因无法根据现有数据得出确定性结论，可能反映了个体负荷模式或解剖变异性。
ASIJ形态似乎独立于主关节的形态。这一观察支持了ASIJ独立于CSIJ发育的假说，并可能在SI复合体内发挥独特的生物力学作用。
[内容请查看原文]
ASIJ的分类与病因学
本研究中透明软骨和明显关节间隙的鉴定支持将ASIJ归类为动关节，细化了先前的人类研究结果（Ehara et al. 1988）。因此，ASIJ代表了原本以纤维软骨为主的ESA内组织学上独特的结构（Gensicke et al. 2026）。周围ESA以致密纤维结缔组织为特征，含有数量不等的纤维软骨，而ASIJ则表现出滑膜关节的组织学特征，包括由相对的透明软骨表面衬覆的关节腔。这些发现表明，ASIJ不应仅被解释为ESA的局灶性纤维软骨特化，而应被视为一个独立的动关节。
尽管我们的研究群体主要由成年个体组成，但在一个青少年标本（11月龄雌性伯恩山犬）中也鉴定出了动关节性ASIJ，提示犬ASIJ可能是先天性的。
人类中也有人提出先天性病因（Ehara et al. 1988; Trotter 1937）。
然而，大多数人类研究假定ASIJ为获得性起源。其患病率和数量随年龄（Prassopoulos et al. 1999; Trentadue et al. 2023）、体重和产次（Prassopoulos et al. 1999）增加而增加，提示机械应力可能促进其发展。这一假说进一步得到报道15岁以下个体中不存在ASIJ的研究的支持（Rixey et al. 2021）。
本研究中观察到的骨性结构可能与Demir et al.（2007）、Ehara et al.（1988）和Prassopoulos et al.（1999）（报道患病率为14.71%）所描述的突起相对应，可能代表退行性重塑模式，从而支持犬中的获得性起源。
[内容请查看原文]
ASIJ的临床意义
在人类中，ASIJ是最常见的SIJ变异型，并一直被证实与退行性改变（Demir et al. 2007; El Rafei et al. 2018）以及下背痛（Hadley 1950; Kang et al. 2017; Prassopoulos et al. 1999; Rosa Neto et al. 2009; Slobodin et al. 2017; Toussirot et al. 2018）相关。硬化、骨赘形成和关节强直在ASIJ中比在主关节中更常见，据报道在大多数受累个体中具有症状。这些改变至少部分归因于ASIJ相对于SIJ的不一致方向（Hadley 1950; Prassopoulos et al. 1999），导致负荷传递改变和机械应力增加（Hadley 1950）。
本研究表明，ASIJ在犬中的患病率远高于人类报道，并可能在单个CSIJ内以多个结构形式出现。由于ASIJ具有动关节的组织学特征，预计其允许受控运动的同时传递力量。因此，它们的存在可能通过提供额外的局灶性接触区域来改变SI区域的机械负荷分布，特别是在表现出大型或多发ASIJ或额外形态学危险因素（如SSA减小或关节一致性改变）的犬中。
这些发现突显了犬ASIJ潜在的临床相关性，并强调了在犬SI区域的解剖学和影像学评估中识别此类变异的重要性。需要整合影像学、生物力学建模和临床相关性的进一步研究来阐明其在犬疼痛、退行性过程和功能障碍中的作用。
[内容请查看原文]
结论
本研究表明，CSIJ表现出显著的超出于主要关节结构之外的形态学变异性。研究确定了七种SSA形状类型、NFOC和ASIJ为常见变异而非罕见异常。ASIJ在犬中的发生率明显高于人类已报道的情况，且常表现为双侧性或单一关节内多结构。ASIJ衬覆透明软骨的存在支持将其归类为真正的动关节。虽然未获得临床相关性资料，但这些发现提示这些变异可能作为独立关节结构发挥作用，从而影响关节生物力学和载荷分布。因此，认识这种解剖多样性对于准确解读影像学发现至关重要，并为CSIJ未来的生物力学和临床研究提供了形态学基础。
作者贡献
Janek Gensicke：撰写——原始草稿、可视化、验证、项目管理、方法学、研究调查、形式分析、统计分析、数据管理、概念构思。Mirjam Kalusa：撰写——审阅与编辑、验证、监督。Thomas Grochow：撰写——审阅与编辑、验证、监督、可视化。Rebecca L. Tartsch：撰写——审阅与编辑、验证。Alexandra S. Klick：撰写——审阅与编辑、验证。Christoph K. W. Mülling：撰写——审阅与编辑、验证、资源。Simone A. Fietz：撰写——审阅与编辑、验证、资源。Nicole Röhrmann：撰写——审阅与编辑、验证、监督、资源、项目管理、概念构思。
资助
本研究由Dr. Heidi und Karl‐Heinz Kübler Stiftung资助。
利益冲突
作者声明无利益冲突。
支持信息
[内容请查看原文]
致谢
作者衷心感谢Martina Fersterra、Cornelia Weber、Henning Siewert、Elfi Quente和Gregor Pfeifer的技术协助。本研究是德国莱比锡大学解剖学、组织学与胚胎学研究所开展的“犬猫骶髂关节”研究项目的一部分。AI工具（ChatGPT‐5.5，OpenAI，美国加利福尼亚州旧金山）仅用于协助将手稿中部分段落从德语初步翻译为英语。科学内容、研究设计、数据分析、结果解释及最终措辞均由作者独立完成并核实。所有AI辅助翻译随后均由英语母语者仔细审阅、校正并进行语言润色。开放获取资金由Projekt DEAL提供并组织。
数据可用性声明
支持本研究结果的数据可向通讯作者提出合理请求后获取。
参考文献
[内容请查看原文]
关联数据
补充材料
[内容请查看原文]
数据可用性声明
支持本研究结果的数据可向通讯作者提出合理请求后获取。
查看英文原文▾
[内容请查看原文]
Introduction
The canine sacroiliac joint (CSIJ) constitutes a central component of the lumbopelvic region, mediating mechanical interactions between the axial skeleton and the pelvic limbs (Breit and Künzel 2001b; Dyce et al. 2010; Frewein et al. 2004; Salomon 2020). Its biomechanical function is intrinsically linked to joint morphology and surface configuration. Consequently, distinguishing physiological variants from pathological alterations is of fundamental importance (Standring et al. 2021).
Breed‐specific differences in pelvic and lumbosacral joint morphology have been described. More sagittal oriented sacroiliac joints, as observed in German Shepherd Dogs, are associated with increased craniocaudal translation compared with the more oblique joint orientation described in breeds such as Rottweilers (Breit and Künzel 2001a). These morphological characteristics may contribute to the development of early‐onset CSIJ alterations (Gembardt 1974; Gregory et al. 1986).
Developmental aspects further contribute to the morphological variability of the CSIJ. In large‐breed dogs, incomplete fusion of ossification centers (OC) within the sacral wing has been documented, persisting into adulthood as discrete defects or concavities of the sacral synovial surface area (SSA). Interestingly, such concavities may enhance local joint interlocking and reduce craniocaudal shear forces, thereby potentially exerting a stabilizing effect in otherwise mechanically disadvantaged joints (Breit and Künzel 2001b). However, these features can sometimes be misinterpreted as pathological lesions in diagnostic imaging (Götz et al. 1993; Prassopoulos et al. 1999).
Beyond intraspecific variation, comparative anatomical studies across species provide additional evidence for distinct morphological variants of the sacroiliac joint (SIJ). In this context, SSA corresponds to the auricular surface, reflecting its auricular‐ or crescent‐shaped morphology (Evans and Lahunta 2013). Although this terminology is standardized, the detailed morphology of SSA exhibits pronounced inter‐ and intra‐species variability, with distinct shape types in cats, horses, and humans (Dalin and Jeffcott 1986; Jesse et al. 2017; Pallandre et al. 2020). These variations have been associated with differences in stress distribution and pain prevalence (Jesse et al. 2017).
In humans, several SIJ variants have been identified based on computed tomography (CT) imaging, including non‐fused ossification centers (NFOC) and accessory sacroiliac joints (ASIJ). The latter represent additional articulations located dorsal to the main joint and have been strongly associated with degenerative changes (Demir et al. 2007; El Rafei et al. 2018), as well as with low back pain (Hadley 1950; Kang et al. 2017; Prassopoulos et al. 1999; Rosa Neto et al. 2009).
In summary, SIJ morphology is highly variable and closely linked to joint mechanics, stability, and the development of clinical disease across species. Therefore, a comprehensive understanding of the morphological spectrum of the CSIJ is essential for the accurate interpretation of imaging findings and for elucidating potential predispositions to joint pathology.
Accordingly, this study aims to:
characterize SSA shape variations, and
identify and quantify developmental ossification anomalies (including NFOC and ASIJ)
[内容请查看原文]
Materials and Methods
This study was conducted using the same cohort of 29 canine specimens that were previously subjected to histological and morphometric analyses of the CSIJ (Table S1). The same dissection protocol was applied. It was evaluated and approved by the Ethical Committee of the Faculty of Veterinary Medicine, Leipzig University (EK 7/2021, 3 May 2021; Gensicke et al. 2026).
From each dog, paired CSIJ were obtained and subdivided into four joint portions: left iliac (LI), left sacral (LS), right iliac (RI), and right sacral (RS).
All CSIJ underwent macroscopic morphological assessment and were documented photographically using an iPhone 14 Pro (Apple Inc., Cupertino, CA, USA; 48‐megapixel sensor; maximum image resolution: 8064 × 6048 pixels). Each joint portion (n = 116) was evaluated for SSA shape types, the presence of NFOC of the sacral SSA, and ASIJ.
[内容请查看原文]
Morphological Analysis
The frequency of each morphological feature was recorded for each joint as well as for each individual. Side‐related distribution patterns were assessed by comparing left and right CSIJ, and the symmetry of NFOC and ASIJ was evaluated in specimens presenting bilateral findings.
Articulation characteristics—including surface texture, shape, curvature, and iliac–sacral congruency—were assessed. The classification of shape types was primarily guided by the classification proposed by Pallandre et al. (2020). In cases of ASIJ, anatomical localization was documented relative to the respective iliac and sacral joint portions.
[内容请查看原文]
Morphometric Features of Common Surface Variants
Morphometric analysis of SSA, NFOC, and ASIJ was performed using the same two‐dimensional method described previously in Gensicke et al. (2026).
[内容请查看原文]
Histology of ASIJ Cartilage
A histological evaluation of the caudodorsal ASIJ cartilage was performed on all four joint portions (LI, LS, RI, RS) of dog no. 25 (Table S1), as this was the only fresh specimen available. Initially, the entire articular surface was fixed in 4% buffered formalin (Carl Roth GmbH & Co. KG, Karlsruhe, Germany). The surrounding osseous tissue was subsequently decalcified and progressively softened using an ultrasonic bath under the influence of Chelaplex (ethylenediaminetetraacetic acid (EDTA) and tris(hydroxymethyl)aminomethane) (Dr. K. Hollborn & Söhne GmbH & Co. KG, Leipzig, Germany) until the superficial cartilage layer could be readily separated. The samples were embedded in paraffin blocks, and a single sagittal section approximately 1 μm thick was obtained from each sample. The four sections were stained with haematoxylin and eosin (H&E) (Carl Roth GmbH & Co. KG, Karlsruhe, Germany). Haematoxylin stains basophilic structures, particularly nucleic acids within the cell nuclei and the nucleolus (DNA and RNA), in shades of blue to purple, whereas eosin stains eosinophilic structures such as cytoplasmic proteins and extracellular matrix components in various shades of pink (Mescher 2018). Microscopic examination was performed using a light microscope (Axioscope 5, Carl Zeiss Microscopy GmbH, Oberkochen, Germany) equipped with an Axiocam 208 colour camera (Carl Zeiss Microscopy GmbH, Oberkochen, Germany) and Zen lite software (Version 3.13; Carl Zeiss Microscopy GmbH, Oberkochen, Germany).
Tissue types were identified and differentiated based on established histomorphological criteria, including cellular composition and extracellular matrix characteristics. Cartilaginous tissues were identified by the presence of chondrocytes within lacunae and the characteristic organization of the cartilage matrix. Fibrocartilage was characterized by a collagen‐rich matrix with chondrocytes frequently arranged in rows. Hyaline cartilage was characterized by a homogeneous matrix containing rounded chondrocytes, occasionally forming isogenous groups. In contrast, elastic cartilage contained a dense network of elastic fibres within the matrix, with chondrocytes likewise in lacunae and displayed a more flexible, resilient structure compared to hyaline cartilage (Liebich 2012).
All images were digitized according to the protocol described in our previous study (Gensicke et al. 2026).
[内容请查看原文]
Statistical Analysis of ASIJ
Measurements of additional articular surface areas (AASA) were compiled in Microsoft Excel 2016 (Microsoft Corporation, Redmond, WA, USA) and analysed using jamovi (version 2.6.44.0, The jamovi project, Sydney, Australia). Normality was assessed using the Shapiro–Wilk test (p < 0.05). As AASA data were not normally distributed, nonparametric methods were applied throughout, including the Wilcoxon signed‐rank test for paired comparisons and Spearman's rank correlation coefficient for correlation analyses (p < 0.05).
AASA sizes were summarized descriptively per dog, per joint portion, and per anatomical location. For inferential analyses, AASA values were summed per joint portion to account for variability in ASIJ number and incomplete bilateral occurrence.
Symmetry was evaluated using paired comparisons, and correlations were performed between iliac and sacral portions within the same joint (LI–LS, n = 21; RI–RS, n = 22). In specimens exhibiting bilateral ASIJ, additional comparisons were performed between left and right sides (LI–RI; LS–RS; n = 18 each).
To assess the relative contribution of ASIJ to CSIJ, summed AASA were expressed as proportions of the total surface area (TSA).
Associations between summed AASA and SSA, extrasynovial surface area (ESA), TSA, and body size were assessed. For body size normalization, total AASA per dog was calculated and correlated with crown–rump length (CRL).
This approach was chosen to account for variability in ASIJ presence and incomplete bilateral occurrence.
Sex‐related differences (n = 13 males, n = 12 females) were evaluated using the Mann–Whitney U test (p < 0.05). Both absolute and proportional AASA values were compared (n = 24), and CRL‐normalized total AASA were additionally analysed.
[内容请查看原文]
Results
[内容请查看原文]
SSA Shape Types
SSA shape types were categorized based on the frequency of their occurrence and the morphologies previously described in other species (Dalin and Jeffcott 1986; Jesse et al. 2017; Pallandre et al. 2020).
Table 1 and Figure 1 summarize the morphological types identified across the 116 joint portions.
TABLE 1. Synovial surface area shape types ( n = 116 portions).
| Shape type | |||||||
|---|---|---|---|---|---|---|---|
| Auricular | Bifoliate | Phrygian Cap | W | Crescent | Spatula | Trifoliate | |
| Left Iliac | 5 | 7 | 4 | 5 | 4 | 4 | 0 |
| Left Sacral | 8 | 6 | 10 | 1 | 2 | 1 | 1 |
| Right Iliac | 7 | 7 | 4 | 7 | 2 | 1 | 1 |
| Right Sacral | 8 | 7 | 8 | 3 | 2 | 0 | 1 |
| Total | 28 | 27 | 26 | 16 | 10 | 6 | 3 |
[图：Schematic medial view of the sacropelvic surface of a left canine ilium, modified from Schaller ( 2018 ). Different shap]
Auricular (24.14%), bifoliate (23.28%), and phrygian cap–like shapes (22.41%) were the most prevalent, whereas spatula‐shaped (5.17%) and trifoliate configurations (2.59%) were rare.
Opposing SSA shapes were symmetrical in 17 of 29 CSIJ, most frequently in bifoliate and trifoliate forms. In contrast, crescent and spatula‐shaped configurations showed the lowest symmetry rates, while auricular, phrygian cap–like, and W‐shaped forms displayed a balanced distribution.
Six dogs displayed identical shapes across all four joint portions. Bilateral symmetry with differing shapes between the left and right CSIJ was observed in four cases, whereas unilateral symmetry was present in 14 dogs and complete asymmetry in five dogs.
Regional analysis revealed only minor compartment‐specific tendencies, without consistent patterns. Across all comparisons, SSA size followed a consistent hierarchy: W‐shaped forms exhibited the largest mean surface area (313.12 ± 96.19 mm2), followed by bifoliate (272.52 ± 221.45 mm2) and auricular shapes (248.50 ± 89.02 mm2). Phrygian cap‐like (198.36 ± 72.28 mm2) and crescent configurations (195.53 ± 91.32 mm2) showed intermediate values, whereas spatula‐shaped forms consistently displayed the smallest SSA (105.25 ± 41.62 mm2). These patterns were consistent across sides, joint portions, and compartments, with no evidence of clustering related to breed, body size, or sex.
[内容请查看原文]
NFOC of the Sacral SSA
The criteria for identifying NFOC included a partial or complete interruption or folding of the sacral SSA within its caudal third at the transition between the sacral wing and the lateral part of the sacrum. This was accompanied by a distinct change in surface inclination between the cranial and caudal portions and the formation of a concave depression (Figure 2).
[图：Lateral view of a left canine sacrum. (a) Cleft between non‐fused ossification centers (NFOC) at the synovial surface ar]
NFOC were identified in 16 out of 58 CSIJ (27.59%) or in 10 out of 29 specimens (34.48%), with an equal distribution on the left and right sides (eight CSIJ on each side). In all affected joints, the typical craniolateral‐to‐caudomedial orientation of the SSA was disrupted. In the caudal third, the orientation consistently shifted to a craniomedial‐to‐caudolateral alignment. In each case, the iliac surface exhibited a corresponding adaptive change in inclination towards a craniomedial‐to‐caudolateral orientation.
On the left sacral wing, this configuration was observed in four out of eight joints with joint space formation, whereas on the right sacral wing it occurred in five out of eight joints. In the remaining joints, an almost vertical demarcation between the cranial and caudal regions was consistently present.
The caudal SSA portions measured between 46.44 ± 18.12 mm2 and 68.89 ± 49.73 mm2 on average. When normalized to TSA, these areas were relatively larger on the sacral wing compared with the iliac surface (19.99% vs. 16.70%).
No correlations with breed or sex were detected. However, NFOC were detected exclusively in dogs with a crown–rump length (CRL) exceeding 740 mm, with the exception of a single French Bulldog (CRL: 530 mm).
[内容请查看原文]
Accessory Sacroiliac Joint
[内容请查看原文]
Morphological Features
In 43 out of 58 examined CSIJ, a total of 62 ASIJ were identified adjacent to the SSA. ASIJ were present in 21 out of 29 left and 22 out of 29 right CSIJ, indicating an overall symmetrical distribution (30 ASIJ in the left CSIJ and 32 in the right).
Among the examined joints, 29 out of 58 CSIJ exhibited a single ASIJ, nine joints had two ASIJ, and five joints had three ASIJ. The distribution of single, double, and triple ASIJ did not differ between iliac and sacral sides.
Bilateral occurrence was observed in 18 out of 29 dogs, whereas only four CSIJ lacked ASIJ entirely on both sides.
A total of six anatomical localizations were identified, all situated dorsal to the SSA within the region of the iliac and sacral tuberosities. The most common location was at the caudodorsal margin of the CSIJ (n = 32), where the sacropelvic surface of the iliac wing articulates caudomedial to the cranial dorsal iliac spine with the craniodorsal aspect of the sacral tuberosity. Immediately cranial to this region, craniodorsal articulations were observed (n = 8).
Less frequently, ASIJ were located ventrally at the margin of the ESA, near the first dorsal sacral foramen (ventral n = 3; dorsal n = 5). In addition, ASIJ were observed as extensions of the SSA at both its cranioventral (n = 4) and caudodorsal limbs (n = 10) (Figure 3).
[图：(a) Medial view of the sacropelvic surface of a left canine ilium illustrating different localizations of accessory sacr]
The iliac and sacral components of the ASIJ were congruent, forming either flat or concave–convex articulations. Focal osseous formations were present in 27.42% of all ASIJ. The AASA were predominantly round to oval in shape, occasionally tapering, and nearly half (48.39%) showed an orientation deviating from that of the primary joint.
No apparent associations were observed between the presence of ASIJ and sex, body size, or breed.
[内容请查看原文]
Histology of ASIJ Cartilage
The superficial layer of all four caudodorsal ASIJ of dog no. 25 consisted of hyaline cartilage (Figure 4). Histologically, the cartilage exhibited a homogeneous extracellular matrix with mild eosinophilic staining. Chondrocytes were located within lacunae and displayed basophilic nuclei, staining dark blue to purple, while the cytoplasm was faintly eosinophilic. The chondrocytes were observed either individually or arranged in small clusters, referred to as isogenous groups. No histological features indicative of fibrocartilage or elastic cartilage were detected in any sample.
[图：Histological appearance of the hyaline cartilage lining a left iliac caudodorsal canine accessory sacroiliac joint (ASIJ]
[内容请查看原文]
Morphometric Features of ASIJ
The mean surface area of all AASA was 42.83 mm2. The smallest AASA (4.20 mm2) was observed at the craniodorsal aspect of LI in a male German Shepherd Dog crossbreed (No. 9), whereas the largest (260.80 mm2) occurred caudodorsally in RI of a female Bernese Mountain Dog (No. 30).
In dogs exhibiting one or more ASIJ, the summarized mean value of AASA per individual was 61.81 mm2.
With respect to anatomical localization, the largest AASA were found caudodorsally (52.48 mm2), followed by dorsal (47.22 mm2), dorsal to the SSA (39.89 mm2), and ventral to the SSA (33.10 mm2). The smallest AASA were found in ventral (17.48 mm2) and craniodorsal locations (14.46 mm2; Figure 3).
Strong positive correlations were observed between iliac and sacral AASA within both left and right CSIJ (ρ = 0.9, p < 0.001). In specimens with bilateral AASA, moderate correlations were identified between LI and RI as well as between LS and RS (ρ ≤ 0.577, p ≤ 0.004).
No significant difference was detected between LI and LS (W = 171, p = 0.055), whereas RI was significantly larger than RS (W = 210, p = 0.007). In bilateral cases, iliac AASA were significantly larger on the right side (W = 31, p = 0.016), while sacral portions did not differ (W = 55, p = 0.196).
Overall, AASA contributed a mean value of 7.25% to TSA, with proportions of 5.78% (LI), 6.37% (LS), 8.36% (RI), and 8.48% (RS). Pairwise comparisons of proportional contributions yielded correlation patterns consistent with those observed in symmetry analysis.
AASA showed significant positive correlations with TSA across all joint portions (ρ = 0.436–0.755, p = < 0.001–0.044; Figure 5a) and even stronger correlations with ESA (ρ = 0.470–0.831, p = < 0.001–0.028). No significant correlations were observed with SSA, although a weak trend was noted for LI (ρ = 0.403, p = 0.071). The total AASA per dog correlated moderately with CRL (ρ = 0.523, p = 0.009; Figure 5b).
[图：(a) Additional articular surface area (AASA) to total surface area (TSA)‐relation (* ρ = 0.755, p < 0.001; ** ρ = 0.6]
No sex‐specific differences in AASA morphometry were detected.
[内容请查看原文]
Discussion
[内容请查看原文]
SSA Shape Types
This study demonstrates that SSA morphology in dogs exhibits considerable variability beyond the commonly assumed auricular or crescent‐shaped configurations; only 38 of the 116 examined surfaces conformed to the classical patterns. The shape classifications applied were based on variants previously described in other species (Dalin and Jeffcott 1986; Jesse et al. 2017; Pallandre et al. 2020).
The greatest morphological similarity was observed between dogs and felids, as both species display bifoliate, spatula‐shaped, and phrygian cap–like configurations. In contrast, W‐shaped and trifoliate configurations were identified exclusively in dogs and, to the best of our knowledge, have not been reported in other species to date.
In humans, auricular SSA are more frequently observed in individuals without sacroiliac (SI) pain, whereas crescent‐shaped surfaces are more commonly associated with SI pain. This observation suggests a potential relation between SSA morphology and the development of SI pain. Different joint morphologies are thought to lead to distinct patterns of load distribution and force transmission. Compared with more oval configurations, crescent‐shaped surfaces provide a smaller area for load transfer. Consequently, a reduced SSA may shift a greater proportion of mechanical forces to adjacent stabilizing structures, such as the ligamentous apparatus. This increases mechanical stress and promotes degenerative changes (Jesse et al. 2017).
Accordingly, dogs exhibiting smaller SSA—particularly those associated with phrygian cap, crescent, or spatula shapes—may be more susceptible to degenerative alterations and, consequently, to the development of SI pain. Moreover, SSA lacking congruent or opposing surface symmetry, as frequently observed in crescent and spatula configurations, may experience a further reduction in effective contact area. This reduction could lead to an increased relative load per unit surface area and thus elevated mechanical stress on the CSIJ, potentially resulting in altered load distribution and subsequent biomechanical imbalance, concepts that have similarly been discussed in the context of the human knee (Gannon et al. 2021; Imhauser et al. 2013). Taken together, dogs, similar to cats, horses, and humans, exhibit substantial variability in SSA morphology (Dalin and Jeffcott 1986; Jesse et al. 2017; Pallandre et al. 2020). If concepts derived from human medicine are cautiously applied, joint configurations characterized by smaller surface areas and reduced bilateral or opposing surface symmetry may be particularly prone to joint alterations.
[内容请查看原文]
NFOC of the Sacral SSA
Concave depressions characterized by clefting or folding, accompanied by a change in inclination of the caudal third of the SSA, correspond to four OC associated with the transverse processes of S1 and S2, as well as two additional isolated OC located at their ventral aspects. Under physiological conditions, these OC typically fuse completely by approximately 25 weeks of age, with earlier fusion reported in smaller breeds. Interruptions of this fusion process represent NFOC, which have been described as a reactive response to mechanical overload of the immature SSA. By increasing local interlocking between opposing joint surfaces, NFOC have been suggested to contribute to joint stabilization. As isolated OC may be absent in small breeds, this phenomenon appears to occur predominantly in large dogs. Canine NFOC can be detected as early as 24 weeks of age and may occasionally persist into adulthood (Breit and Künzel 2001b).
Similar findings in humans have primarily been reported in adolescents and young adults, albeit with a markedly lower prevalence (Götz et al. 1993; Green et al. 2008; Prassopoulos et al. 1999). In this context, NFOC are generally considered to be non‐pathological (Götz et al. 1993). This interpretation is supported by a case report that excluded them as a cause of SI pain, as their presumed lifelong presence makes an onset of pain later in life unlikely (Green et al. 2008).
Nevertheless, complete fusion may result in osseous prominences of the SSA, which could contribute to degenerative changes. In imaging studies, NFOC may initially mimic fractures (Prassopoulos et al. 1999), highlighting the importance of reliable differentiation from fractures and other degenerative joint alterations. Axial CT appears to be the most suitable modality for this purpose (Götz et al. 1993), where NFOC typically present as triangular osseous bodies within the anterosuperior joint space (Prassopoulos et al. 1999).
In summary, NFOC of the canine sacral SSA appear to be more prevalent than in humans. Although they are not considered a direct cause of SI pain, their accurate identification is crucial to avoid diagnostic misinterpretation. Moreover, they may even contribute functionally to the stabilization of the CSIJ.
[内容请查看原文]
Accessory Sacroiliac Joint
The iliac and sacral tuberosities, which together form the osseous basis of the ESA, are incongruent and covered by fibrocartilage (Evans and Lahunta 2013). Consequently, the CSIJ was classified as a composite joint, comprising a diarthrodial component in the caudoventral region and a symphyseal component in the craniodorsal region (Gensicke et al. 2026).
[内容请查看原文]
Morphological Features of ASIJ
Congruent AASA within the ESA were identified more frequently than expected. These surfaces were either flat or concave–convex in configuration and were covered by hyaline cartilage.
Comparable joint variants are less frequent in humans but have been reported in anatomical dissection and imaging studies, with prevalences ranging from 6.8% to 50% (CT 1.7%–25.8%, MRI 5%–11%). These structures are referred to as ASIJ and are defined as AASA located dorsal to the primary diarthrodial joint (Cihan et al. 2023; Tok Umay and Korkmaz 2020; Trentadue et al. 2023; Wittram and Whitehouse 1995).
ASIJ may occur bilaterally (4.4%–53%) in humans (Ehara et al. 1988; El Rafei et al. 2018; Prassopoulos et al. 1999; Valojerdy and Hogg 1990), a pattern likewise observed in the present canine study, in which 62% of cases showed bilateral involvement, predominantly with symmetrical distribution. Notably, approximately one quarter of dogs had multiple ASIJ within a single CSIJ, a feature that has not been reported in humans (Valojerdy and Hogg 1990).
The most commonly reported location of ASIJ in humans corresponds largely to the rare canine ventral ASIJ (4.76%). More ventrally positioned human ASIJ resemble the dorsoventral variants identified in dogs (11.29%) (Demir et al. 2007; Ehara et al. 1988; Prassopoulos et al. 1999; Trentadue et al. 2023).
In contrast, four additional ASIJ localizations were identified in dogs, most frequently at the caudodorsal joint margin (51.61%). Further variants included extremely dorsal ASIJ (9.68%) as well as extensions exceeding the regular SSA ventrally (6.45%) and dorsally (16.13%).
In dogs, force transmission across the CSIJ is influenced by the horizontally oriented trunk, propulsion from the hindlimbs via the pelvis to the spine (Prieur 1980), and additional rotational forces during locomotion (Breit and Künzel 2001b). In contrast, axial compression predominates in the upright posture of humans. This more complex and dynamic loading environment in quadrupeds may promote a more frequent and more distributed adaptive formation of ASIJ compared to the predominantly vertically loaded human SIJ.
[内容请查看原文]
Morphometric Features of ASIJ
Canine ASIJ dimensions showed considerable variability, ranging from 14.46 to 52.48 mm2, corresponding to 5.78 to 8.48% of the TSA. In humans, reported AASA fall within a comparable range, with mean values of approximately 47 ± 36 mm2 (Valojerdy and Hogg 1990), although considerably larger surface areas, ranging from 78.00 to 314.00 mm2, have also been described (Hadley 1950).
The observed variability in ASIJ sizes appears to be primarily attributable to size‐dependent allometric scaling, as indicated by correlations with body size (CRL) and joint dimensions (ESA and TSA). In addition, load‐dependent biomechanical adaptation in response to individual mechanical demand may contribute further to the broad size spectrum.
In contrast to findings in humans (Valojerdy and Hogg 1990), no correlation could be demonstrated between AASA and SSA in dogs. This discrepancy may reflect the same species‐specific biomechanical conditions as previously discussed in the context of gait variation.
Iliac and sacral AASA were significantly correlated, as were right and left AASA in dogs exhibiting bilateral ASIJ. LI AASA tended to be larger than the LS AASA. On the right side, RI were significantly larger than the RS AASA. This asymmetry is consistent with previously reported morphometric patterns of the CSIJ (Gensicke et al. 2026).
Sacral ASIJ dimensions did not differ significantly between sides. The reason for the larger RI AASA compared with LI cannot be conclusively determined based on the present data and may reflect individual loading patterns or anatomical variability.
ASIJ morphology appeared to be independent of the morphology of the main joint. This observation supports the hypothesis that the ASIJ develops independently of the CSIJ and may fulfil a distinct biomechanical role within the SI complex.
[内容请查看原文]
Classification and Aetiology of the ASIJ
The identification of hyaline cartilage and a distinct joint space in the present study supports the classification of the ASIJ as a diarthrosis, specifying previous human findings (Ehara et al. 1988). Accordingly, the ASIJ represents histologically distinct structures within the otherwise predominantly fibrocartilaginous ESA (Gensicke et al. 2026). Whereas the surrounding ESA is characterized by dense fibrous connective tissue with variable amounts of fibrocartilage, the ASIJ exhibits the histological features of a synovial articulation, including a joint cavity lined by opposing hyaline cartilaginous surfaces. These findings indicate that the ASIJ should not be interpreted merely as a focal fibrocartilaginous specialization of the ESA, but rather as a separate diarthrodial articulation.
Although our study population consisted predominantly of adult individuals, diarthrodial ASIJ were also identified in one adolescent specimen (11‐month‐old female Bernese Mountain Dog), suggesting that canine ASIJ can be congenital.
A congenital aetiology has also been suggested for humans (Ehara et al. 1988; Trotter 1937).
However, most human studies assume an acquired origin of ASIJ. Their prevalence and number increase with age (Prassopoulos et al. 1999; Trentadue et al. 2023), body weight, and parity (Prassopoulos et al. 1999), suggesting that mechanical stress may promote their development. This hypothesis is further supported by studies reporting the absence of ASIJ in individuals younger than 15 years (Rixey et al. 2021).
The osseous formations observed in the present study may correspond to the projections described by Demir et al. (2007), Ehara et al. (1988), and Prassopoulos et al. (1999) (reported a prevalence of 14.71%) and could represent degenerative remodelling patterns, thereby supporting an acquired origin in dogs.
[内容请查看原文]
Clinical Implications of the ASIJ
In humans, ASIJ are the most frequent SIJ variant and have been consistently linked to degenerative change (Demir et al. 2007; El Rafei et al. 2018) as well as low back pain (Hadley 1950; Kang et al. 2017; Prassopoulos et al. 1999; Rosa Neto et al. 2009; Slobodin et al. 2017; Toussirot et al. 2018). Sclerosis, osteophyte formation, and ankylosis occur more frequently in ASIJ than in the main joint and are reported to be symptomatic in most affected individuals. These alterations have been attributed, at least in part, to the discordant orientation of the ASIJ relative to the SIJ (Hadley 1950; Prassopoulos et al. 1999), resulting in altered load transmission and increased mechanical stress (Hadley 1950).
The present study demonstrates that ASIJ are considerably more prevalent in dogs than reported in humans and may occur as multiple structures within a single CSIJ. Because the ASIJ possess the histological characteristics of a diarthrosis, they are expected to permit controlled motion while simultaneously transmitting forces. Their presence may therefore modify the distribution of mechanical loads across the SI region by providing additional focal contact areas, particularly in dogs exhibiting large or multiple ASIJ or additional morphological risk factors such as reduced SSA or altered joint congruency.
These findings highlight the potential clinical relevance of canine ASIJ and emphasize the importance of recognizing such variants in anatomical and radiological evaluations of the canine SI region. Further studies integrating imaging, biomechanical modelling, and clinical correlation are required to clarify their role in pain, degenerative processes, and functional impairment in dogs.
[内容请查看原文]
Conclusion
This study demonstrates that the CSIJ exhibits substantial morphological variability that extends beyond the primary articulation. Seven SSA shape types, NFOC, and ASIJ were identified as common variants rather than rare anomalies. The ASIJ were markedly more prevalent in dogs than has been reported in humans and were often observed bilaterally or as multiple structures within a single joint. The presence of hyaline cartilage lining the ASIJ supports their classification as true diarthrodial articulations. While clinical correlation was not available, these findings suggest that these variants may function as independent joint structures and therefore influence joint biomechanics and load distribution. Therefore, recognition of this anatomical diversity is therefore essential for the accurate interpretation of imaging findings and provides a morphological foundation for future biomechanical and clinical investigations of the CSIJ.
Author Contributions
Janek Gensicke: writing – original draft, visualization, validation, project administration, methodology, investigation, formal analysis, statistical analysis, data curation, conceptualization. Mirjam Kalusa: writing – review and editing, validation, supervision. Thomas Grochow: writing – review and editing, validation, supervision, visualization. Rebecca L. Tartsch: writing – review and editing, validation. Alexandra S. Klick: writing – review and editing, validation. Christoph K. W. Mülling: writing – review and editing, validation, resources. Simone A. Fietz: writing – review and editing, validation, resources. Nicole Röhrmann: writing – review and editing, validation, supervision, resources, project administration, conceptualization.
Funding
This work was supported by Dr. Heidi und Karl‐Heinz Kübler Stiftung.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
[内容请查看原文]
Acknowledgements
The authors gratefully acknowledge the technical assistance of Martina Fersterra, Cornelia Weber, Henning Siewert, Elfi Quente, and Gregor Pfeifer. This work forms part of the research project “Canine and Feline Sacroiliac Joint” conducted at the Institute of Anatomy, Histology and Embryology, Leipzig University, Leipzig, Germany. AI tools (ChatGPT‐5.5, OpenAI, San Francisco, CA, USA) were used only to assist with the initial translation of selected passages of the manuscript from German into English. The scientific content, study design, data analysis, interpretation of the results, and final wording were developed and verified exclusively by the authors. All AI‐assisted translations were subsequently carefully reviewed, corrected, and linguistically refined by a native English speaker. Open Access funding enabled and organized by Projekt DEAL.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
[内容请查看原文]
Associated Data
Supplementary Materials
[内容请查看原文]
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.