兽研一只10岁雌性已绝育混种犬因癫痫发作及急性进行性颅内症状就诊。神经系统检查符合多灶性脑病。临床检查发现瘀斑、瘀点及双侧视网膜出血。临床病理学评估发现严重血小板减少[血小板计数9×10³/μL，手工涂片评估未见血小板聚集]，平均血小板体积增大。凝血功能检测（活化部分凝血活酶时间和凝血酶原时间）及纤溶标志物（D-二聚体、定量纤维蛋白原）、传染病筛查以及胸腹部影像学检查均在正常范围内，确立了原发性免疫介导性血小板减少症（ITP）的诊断。脑部磁共振成像（MRI）显示双侧大脑半球表面覆盖有轻微扩张的、双侧新月形轴外液体积聚。基于T1加权和T2加权高信号以及T2*加权梯度回波（GRE）成像上的周边磁敏感伪影，诊断为晚期亚急性硬膜下颅内出血。尽管初始麻醉及苏醒过程平稳，该犬仍出现急性神经功能恶化并发生心肺骤停，怀疑由持续出血或脑疝所致。本报告描述了犬原发性ITP的一种罕见神经系统表现。虽然犬ITP通常预后良好，但可发生危及生命的继发性颅内出血。在确认血小板计数正常之前，应将原发性ITP作为急性进行性神经系统表现的鉴别诊断之一。
引言
兽医临床中的出血性脑病可根据病因或定位进行分类。近期提出的一种兽医分类方案将出血性脑病分为原发性、继发性或创伤性，并设有常用的VITAMIN D方案亚类（1）。颅内出血可根据定位分为实质外、硬膜外、硬膜下、蛛网膜下腔、室管膜下、脑室内或实质内出血。在小动物兽医临床中，实质内和实质外出血压已有充分记录，而硬膜下出血或血肿则较为罕见且明显报道不足，尤其是与人医相比。在犬中，硬膜下出血在创伤性脑损伤病例中鲜有报道，其与脑积水及坏死性脑炎等疾病同时发生的情况也属罕见，仅限于个别病例报告（2–4）。与出血性脑病相关的临床症状差异很大，主要取决于出血的急性或慢性阶段，以及出血的位置、范围、继发性改变和整体基础病因。
免疫介导性血小板减少症（ITP）是犬最常见的止血障碍之一（5）。它可以是原发性或继发性的。自发性出血通常在血小板计数低于30,000/μL时才会发生。原发性ITP的临床表现高度可变，血小板减少的严重程度与就诊时的临床症状之间无特定相关性。症状可包括嗜睡、食欲不振或厌食、瘀点和瘀斑以及发热。较少见的表现包括黑便、血尿、巩膜出血、鼻出血和前房积血，这些可能具有潜在的生命威胁（6）。多项研究报告了犬ITP可变的短期和长期结局，总体出院存活预后尚可。然而，犬ITP相关的发病率各异，死亡率可高达27%（7–9）。
虽然典型的黏膜和皮肤出血表现已有充分记录，但ITP中颅内出血所致的神经系统表现在人医和兽医文献中均不常见乃至极为罕见（10）。本报告描述了硬膜下出血作为犬原发性ITP一种罕见且致命的神经系统表现，突出了兽医神经病学中的诊断挑战和临床管理。
病例描述
一只10岁雌性已绝育混种犬因急性起病的无法站立及癫痫样发作病史就诊。据报告，在过去一个月内，该犬活动减少、不愿跳跃、间歇性胸肢跛行，并在就诊前7天内急性起病的进行性癫痫样活动。这些发作表现为全身僵硬伴意识无反应，持续15–20分钟，每日发生数次。转诊前，该犬先在急诊诊所接受评估，随后在初级保健诊所就诊，进行了全血细胞计数、血清生化 panel 及 IDEXX 4Dx Plus SNAP® 检测。血液学评估显示轻度白细胞增多（白细胞计数（WBC）18.4 × 10³/μL；参考区间（RI）4.0–15.5 × 10³/μL）和重度血小板减少（血小板计数 21 × 10³/μL；RI 170–400 × 10³/μL），血涂片可见少量小血小板簇集且估计血小板计数减少。生化分析显示轻度高球蛋白血症（3.7 g/dL；RI 1.6–3.6 g/dL）和碱性磷酸酶轻度升高（138 IU/L；RI 5–131 IU/L）。电解质水平在正常范围内。该犬 4Dx SNAP 检测犬恶丝虫抗原及伯氏疏螺旋体、无浆体属和埃立克体属抗体均为阴性。初级兽医因怀疑骨关节炎而开具口服非甾体抗炎药（NSAID；美洛昔康），并建议若无改善则转诊。
就诊时，该犬安静、警觉且反应灵敏。体格检查除估计约5%脱水外基本无异常。体格检查期间，该犬再次出现癫痫样发作，表现为全身僵硬、跌倒及双侧瞳孔散大伴瞳孔对光反射消失。发作期间血压轻度升高（收缩压 150 mmHg；舒张压 109 mmHg；平均动脉压（MAP）113 mmHg）。发作后神经系统检查显示精神迟钝至木僵，右侧胸肢和盆肢本体感觉缺失，以及发作性胸肢僵硬和颈部伸展。该犬给予单剂量甘露醇 0.5 g/kg 静脉注射（IV）、马罗匹坦 1 mg/kg IV，并开始以 Plasma-Lyte 60 mL/kg/天 IV 进行过夜稳定，直至次日转至神经科服务。过夜期间，该犬出现一次颈部伸展、流涎和呼吸用力增加发作，未经干预自行缓解；进行了胸腹部床旁超声（POCUS），未显示任何 B 线或胸膜、心包或腹膜积液。
转至神经科服务当日，该犬轻度至中度迟钝，头颈部低垂。由于过夜癫痫样发作，治疗中还加入了左乙拉西坦 32 mg/kg IV 每8小时一次。体格检查发现左侧胸部区域两处局灶性瘀斑，长约1.5 cm。神经系统检查显示轻度可行走性四肢轻瘫伴前庭性共济失调、倾向左侧头转、轻度左侧头倾斜及一次自限性躯干摇晃发作。脑神经检查显示双侧威胁反应缺失、等大瞳孔伴中度至重度双侧瞳孔散大、双侧直接和间接瞳孔对光反射完整，以及位置性眼震从垂直向旋转变化，快相向左或向右不定。该犬左侧胸肢和盆肢放置和跳跃反应正常，右侧胸肢和盆肢爪放置和跳跃反应不一致地轻度延迟。眼底检查显示双侧视网膜出血，右眼更重。其余神经系统检查无异常。神经解剖定位为多灶性脑（左侧前脑和脑干）。鉴别诊断包括炎性或感染性脑膜脑炎、颅内肿瘤或血管性疾病。进行了全血细胞计数和生化 panel。血液学评估显示轻度白细胞增多（白细胞计数（WBC）19.2 × 10³/μL；RI 5.7–14.2 × 10³/μL）、网织红细胞增多（绝对网织红细胞计数 125.2 × 10³/μL；RI 11–92 × 10³/μL）伴正常红细胞压积（46%；RI 41–58%），以及重度大血小板性血小板减少（血小板计数 9 × 10³/μL；RI 186–545 × 10³/μL；平均血小板体积（MPV）26.8 fL；RI 8.4–14.1 fL），未见血小板簇集且血涂片血小板计数显著减少。生化分析显示碱性磷酸酶轻度升高（148 IU/L；RI 7–115 IU/L）。全面凝血功能检查，包括活化部分凝血活酶时间（aPTT）、凝血酶原时间（PT）、定量纤维蛋白原（Clauss法）、抗凝血酶活性（抗-IIa发色底物法）及定量交联纤维蛋白降解产物（D-二聚体），均在正常范围内。其他诊断检查包括胸部X线摄影，未见异常；腹部超声显示脾脏中度弥漫性增大且回声不均，最可能代表良性或反应性改变，如淋巴样增生或髓外造血。同日进行了脑磁共振成像（MRI）以进一步明确神经系统临床表现。
在全身麻醉下使用1.5-T扫描仪（Vantage Orian，Canon Medical Systems USA，Inc.，Tustin，CA）进行MRI检查，犬取背侧卧位并使用膝关节线圈。横断面序列包括T1加权、T2加权、T2*加权梯度回波（GRE）、液体衰减反转恢复（FLAIR）、弥散加权成像（DWI）、T1加权磁化准备快速梯度回波（MPRAGE）以及增强前和增强后T1加权图像（包括脂肪饱和增强后T1加权序列和增强后T1 MPRAGE）。矢状面和背侧面包括T2加权和增强后T1加权序列。表观弥散系数（ADC）图由DWI数据生成。钆双胺（Omniscan，GE Healthcare，Chicago，IL）以0.1 mmol/kg的剂量（总计1.3 mL）静脉给药。
在图像分析过程中，观察到双侧轻度扩张的新月形轴外液性积聚，覆盖于双侧大脑半球整个背外侧凸面，未沿内侧和基底缘延伸（图1）。该液性积聚在T1加权、T2加权和FLAIR图像上呈均匀高信号，在T2* GRE图像上可见周边磁敏感伪影。由于该液性积聚在给予对比剂前已呈T1高信号，无法辨别对比增强。未检测到其他异常对比增强部位。大脑半球受液性积聚轻度压迫，并呈轻度弥漫性肿胀，无中线移位。受压和肿胀的大脑使间脑和中脑向尾侧移位，压迫小脑吻侧缘并使小脑向尾腹侧移位，延伸至枕骨大孔但未穿过枕骨大孔。在小脑幕水平的横断面成像证实，该尾侧移位仅限于中线顶盖，海马旁回和大脑皮质位置正常，无尾侧小脑幕疝。在C2和C3椎体水平，脊髓轻度肿胀，伴有边界不清、局部广泛、无对比增强的T2加权高信号（图2）。这些发现被解释为轻度、双侧、晚期亚急性颅内硬膜下出血，伴轻度继发性脊髓水肿。根据脑的颅内移位和脊髓水肿估计颅内压升高。由于未观察到其他创伤迹象，优先考虑继发于严重血小板减少症的硬膜下出血。由于患者严重血小板减少状态和疑似颅内压升高，未进行脑脊液（CSF）分析。
[图：横断面犬脑磁共振图像显示轻度、双侧、晚期亚急性硬膜下出血]
[图：脑磁共振成像（MRI）。（A）矢状面T2加权犬脑MRI图像显示轻度、弥漫性]
根据临床表现、血液检查和影像学发现，该犬被诊断为推定性原发性免疫介导性血小板减少症伴继发性大脑硬膜下出血。该犬在恢复期间给予单剂量地塞米松磷酸钠0.1 mg/kg静脉注射。麻醉和恢复过程平稳，该犬在重症监护笼内可自行站立和行走，精神状态安静。恢复后三小时，主治兽医复查神经系统评估时，该犬急性且严重迟钝，并表现出中度呼吸用力，但心肺听诊无异常。随后突然出现呼吸过缓、舌发绀和呼吸停止，并迅速进展为心肺骤停。鉴于患者为不予复苏（DNR）状态，未进行心肺复苏（CPR）。主人拒绝尸检。
讨论
本报告描述了一只因原发性ITP继发推测的双侧颅内硬膜下出血而出现多灶性神经体征，并最终导致心肺骤停的犬。此前仅有兩篇个案报告描述了与推测ITP相关的犬出血性脊髓病（10, 11）。在Guevar等人的报告中，该犬在经历六周胸腰段感觉过敏和步态异常病史后，于24小时内出现从可行走到不能行走的急性进行性四肢轻瘫。根据血小板计数3×10³/μL且无其他潜在病因，该犬被诊断为推测性ITP。颈段和胸腰段MRI显示胸腰骶段脊髓内多灶性硬膜内-髓外及髓内出血。尽管免疫调节治疗后血小板减少症得以缓解，该犬的神经状态仍恶化至四肢瘫且后肢无痛觉，随后被安乐死。尸检显示整个脊髓多灶性硬膜内、蛛网膜下腔及髓内出血，伴椎管内不同程度压迫。第二例病例描述了相似的神经表现，一只急性起病四肢瘫且四肢感觉减退的犬被诊断为推测性ITP。颈段脊柱MRI显示髓内出血。开始使用地塞米松磷酸钠和环孢素联合进行免疫调节治疗。该犬的血小板减少症改善并于就诊后第5天缓解，神经状态改善为不能行走的四肢轻瘫。出院后5周随访时，该犬仅表现为轻度全身性共济失调，无轻瘫，血小板计数维持正常。
这两例病例凸显了ITP继发神经表现的可能严重性，并强调了尽管血小板计数恢复正常但临床结局仍存在变异性。犬ITP是一种复杂疾病，其特征是血小板减少的严重程度与出血的解剖位置或临床体征之间缺乏直接相关性（12）。虽然犬和猫即使存在严重血小板减少也常见亚临床体征，但其他患畜可能表现为灾难性、危及生命的出血。预后指标研究表明，黑便或高血尿素氮（BUN）浓度可能提示患病犬预后不良（8）。然而，尚无具体研究评估神经体征（脑病或脊髓病）的存在是否构成更高的复发风险或与不良结局相关。在本病例中，未观察到黑便或BUN浓度升高。相反，该犬在诊断为ITP并推测双侧大脑硬膜下出血后24小时内，急性进展至迟钝、突发呼吸浅慢和心肺骤停，随后猝死。由于未进行尸检，无法确认明确的死后死亡原因。虽然肺血栓栓塞（PTE）是犬猝死的公认鉴别诊断，但PTE通常因通气-灌注不匹配和呼吸中枢刺激而表现为过度通气和呼吸急促。相比之下，该患畜在迟钝后突发呼吸浅慢，强烈提示脑干功能障碍——具体而言，即延髓呼吸中枢和上行网状激活系统的抑制，可能由颅内压快速升高所致。MRI时，轻度弥漫性脑肿胀表现为间脑和中脑向尾侧移位，以及吻侧小脑受压和小脑向枕骨大孔尾侧移位。急性继发事件——如颅内出血的动态扩大、进行性血管源性水肿或颅内压升高——是该患畜急性临床恶化的可能促成因素；然而，在缺乏尸检确认的情况下，明确的死亡原因仍无法确定。
本病例的主要局限性在于缺乏尸检，这妨碍了对大脑硬膜下出血及其与患畜临床恶化和死亡直接关系的明确病理学确认。此外，尽管立即进行死后CSF分析是可行的，且可能有助于排除原发性感染性或炎性脑膜脑炎或其他病因，但颅内出血后的CSF结果通常为非特异性，通常表现为细胞群体多变和继发性红细胞吞噬现象。尽管如此，特征性MRI表现、全面的系统性诊断检查以及患畜的临床病程相结合，提供了生前证据支持将大脑硬膜下出血作为原发性免疫介导性血小板减少症的一种罕见神经表现这一推定诊断。
目前，犬ITP仍属排除性诊断(5)。由于其异质性，标准化诊断方法和建立明确标准历来具有挑战性。近期，美国兽医内科学院(ACVIM)发布了关于ITP诊断的更新共识声明，提供了系统性排除血小板减少症其他潜在病因的诊断流程。在本病例中，临床病理学发现包括重度血小板减少(9×10³/μL)和平均血小板体积增大。根据ACVIM共识指南，尽管ITP患犬的血小板减少程度可能比非免疫介导性病因的犬更严重，但临床上仍存在重叠。然而，血小板计数临界值<20,000/μL强烈支持ITP的诊断。本病例未进行骨髓检查，目前文献表明尚无充分证据确定骨髓评估是否能显著提高犬原发性ITP的诊断准确性。共识声明强烈建议在初始诊断检查中进行常规凝血检测(aPTT、PT)和纤溶标志物(D-二聚体、定量纤维蛋白原)，以排除合并消耗性凝血病。在本病例中，这些参数均在正常范围内。ITP指南中的合并症筛查包括媒介传播感染；该犬心丝虫抗原以及伯氏疏螺旋体、无浆体属和埃立克体属抗体检测均为阴性。转诊前曾给予一剂非甾体抗炎药(美洛昔康)。然而，重度血小板减少和临床症状在该药物给药前即已存在。尽管犬药物诱导性ITP的总体证据有限，但头孢西酮和磺胺类药物作为ITP触发因素具有中高级别证据。然而，本病例的临床时间线最有效地排除了美洛昔康作为主要诱发因素。
总体而言，本病例所进行的诊断符合ACVIM关于原发性ITP的共识声明。对于表现为原发性ITP且血小板计数<50,000/μL的犬(如本病例)，共识指南提示有中等证据表明初始入院时血小板计数不影响疾病结局或治疗反应。相反，血小板计数的严重程度可能主要与就诊时活动性临床出血相关，在本病例中表现为瘀斑、双侧视网膜出血和颅内硬膜下出血。视网膜出血已被充分记录为血小板减少症犬的严重眼部病变，其病变严重程度与血小板计数相关(13)。一般来说，自发性出血通常发生在血小板计数<30,000/μL时；然而，尚无前瞻性兽医研究比较原发性ITP患犬轻度(血小板计数>75,000/μL)与中重度(血小板计数<50,000/μL)血小板减少状态下的长期结局或部位特异性病死率。
在人类文献中，ITP是一种不常见的血液学疾病，颅内出血被认为是重度血小板减少患者中罕见但致命的并发症，估计病死率为25%-55%(14,15)。关于危及生命的颅内出血的最佳管理，尤其是神经体征迅速恶化期间为缓解颅内压升高而进行的紧急手术减压，共识有限。仅有一篇人类病例报告记录了紧急脾切除术联合开颅术后良好的临床结局(15)。在本病例中，尽管初始麻醉和苏醒过程平稳，但神经功能恶化和最终心肺骤停进展迅速。鉴于缺乏更大规模的前瞻性研究或兽医患者颅内出血事件的既定手术指南，本病例是兽医文献中首例证明严重且致命的硬膜下出血作为ITP主要神经学表现的病例。
尽管ITP尚未被记录为犬猫脊髓病或脑病的常见病因或鉴别诊断，但继发性中枢神经系统出血可由多种基础疾病引起。出血性脑病的继发性病因包括血管畸形、肿瘤、与感染性或非感染性脑膜脑炎相关的血管炎、凝血功能障碍、伴出血转化的缺血性梗死、代谢性或营养性疾病或热射病。在结构上，出血可根据位置分为实质外、硬膜外、硬膜下、蛛网膜下、室管膜下、脑室内和实质内出血。虽然硬膜下出血在临床上通常定义为血液积聚在硬脑膜与蛛网膜之间的间隙内，但在神经解剖学上它被认为是硬膜内病变，因为不存在真正的生理性硬膜下间隙。相反，这一病理性扩大的间隙是在机械性创伤或自发性血管外渗撕裂结构薄弱的硬膜边缘细胞层并伴有桥接血管破裂时形成的(1, 16)。在人类文献中，硬膜下血肿最常发生于创伤性脑损伤，尽管在血管畸形和动脉瘤等罕见情况下也可发生；公认的危险因素包括年龄、合并凝血功能障碍、抗凝治疗、脑萎缩和酒精性疾病(17, 18)。相反，在兽医文献中，仅报道了少数犬硬膜下出血病例。已记录的潜在病因包括创伤、脑积水、腰椎脊髓造影和脑室腹腔分流术放置的并发症，以及因蜡样脂褐质沉积症导致脑萎缩继发静脉撕裂(2, 3, 19–22)。典型的神经学体征包括癫痫发作、精神状态改变和并发的神经功能缺损。
在MRI上，大脑硬膜下血肿的特征性表现为单个或多个新月形轴外液体积聚，其信号强度因出血的时程而异。随着血液产物降解并沉积于重力依赖层，液体积聚的组成随时间变化，这是充满液体的间隙所特有的现象，不会发生于实体组织。颅内出血的MRI表现高度复杂，其信号强度主要由血红蛋白降解和氧合随时间变化的T1弛豫和T2*磁化率效应决定。随后血红蛋白的降解，从氧合血红蛋白进展为脱氧血红蛋白、细胞内高铁血红蛋白、细胞外高铁血红蛋白，最终变为铁蛋白和含铁血黄素，随时间改变血肿的磁学特性。这些可预测的生化变化使得颅内出血的不同阶段(超急性期、急性期、亚急性早期、亚急性晚期和慢性期)能够在MRI上被确定(23)。
基于本病例的MRI特征，该犬的大脑硬膜下出血最符合亚急性晚期。这一分类得到T1加权和T2加权高信号、T2* GRE上的磁化率伪影以及特征性扩大的新月形硬膜下间隙环绕双侧大脑半球的支持。关键在于，尽管ITP的诊断是在MRI诊断硬膜下出血之前刚刚确立的，但神经学体征，特别是癫痫发作，约在1周前即已开始。这一临床时间线与亚急性晚期至慢性早期出血阶段密切吻合，该阶段通常持续数天至数月，进一步证实了严重血小板减少症和原发性ITP是该患者神经学体征的病因。
结论
本病例记录了兽医文献中此前未曾报道的ITP罕见临床表现。对于表现为急性进行性颅内多灶性神经功能缺损的犬，在确认血小板计数充足之前，应将ITP列为鉴别诊断。尽管ITP在适当管理下通常预后良好，但颅内出血等严重并发症可能导致显著发病率，并可能与不良结局相关。
查看英文原文▾
A 10-year-old female spayed mixed-breed dog was presented with a history of seizures and acute progressive intracranial signs. The neurological examination was consistent with multifocal brain disease. Clinical examination revealed ecchymoses, petechiae, and bilateral retinal hemorrhage. Clinicopathological evaluation identified severe thrombocytopenia [a platelet count of 9 × 103/μL without clumps on manual smear evaluation] with an increased mean platelet volume. Coagulation testing (activated partial thromboplastin time and prothrombin time) and fibrinolysis markers (D-dimers, quantitative fibrinogen), infectious disease screening, and thoracic and abdominal imaging were within normal limits, establishing a diagnosis of primary immune-mediated thrombocytopenia (ITP). Magnetic resonance imaging (MRI) of the brain showed minimally expanded, bilateral crescent-shaped, extra-axial fluid accumulations overlying both cerebral hemispheres. Based on T1-weighted and T2-weighted hyperintensities and peripheral susceptibility artifact on T2*-weighted gradient-recalled echo (GRE) imaging, a late subacute subdural intracranial hemorrhage was diagnosed. Despite initial uneventful anesthesia and recovery, the dog experienced acute neurological deterioration and suffered cardiopulmonary arrest, suspected due to ongoing hemorrhage or brain herniation. This report describes a rare neurological manifestation of primary ITP in a dog. While canine ITP generally carries a favorable prognosis, life-threatening secondary intracranial hemorrhage can occur. Primary ITP should be considered as a differential diagnosis for an acute, progressive neurological presentation until a normal platelet count is confirmed.
Introduction
Hemorrhagic encephalopathies in veterinary medicine can be classified based on etiology or localization. A recently proposed veterinary classification scheme categorized hemorrhagic encephalopathy as primary, secondary, or traumatic, with subcategories for the well-used VITAMIN D scheme (1). Intracranial hemorrhage can be classified based on localization as extra-parenchymal, epidural, subdural, subarachnoid, subependymal, intraventricular, or intraparenchymal hemorrhage. In small animal veterinary medicine, intra-parenchymal and extra-parenchymal hemorrhages are well documented, whereas subdural hemorrhages or hematomas are rare and significantly underreported, particularly compared to human medicine. In dogs, subdural hemorrhages are rarely reported in traumatic brain injury cases, and their occurrence alongside conditions such as hydrocephalus and necrotizing encephalitis is rare, limited only to isolated case reports (2–4). Clinical signs associated with hemorrhagic encephalopathies vary considerably and depend predominantly on the acute or chronic stage of the hemorrhage, as well as the location, extent, secondary changes, and the overall underlying etiology.
Immune-mediated thrombocytopenia (ITP) is one of the most common hemostatic disorders in dogs (5). It can be either primary or secondary. Spontaneous hemorrhage typically does not occur until a platelet count of less than 30,000/μL. Clinical presentation of primary ITP is highly variable, with no specific correlation between the severity of the thrombocytopenia and the presenting clinical signs. Signs may include lethargy, inappetence or anorexia, petechiae and ecchymosis, and pyrexia. Less frequent manifestations include melena, hematuria, scleral hemorrhage, epistaxis, and hyphema, which can potentially be life-threatening (6). Several studies have reported variable short- and long-term outcomes of ITP in dogs, with an overall fair prognosis for survival to discharge. However, variable morbidity is associated with ITP in dogs, and the mortality rate can be up to 27% (7–9).
While typical mucosal and cutaneous hemorrhagic manifestations are well documented, neurological manifestations due to intracranial hemorrhage in ITP are uncommon to exceptionally rare in both the human and veterinary literature (10). This report describes a subdural hemorrhage as a rare and fatal neurological manifestation of primary ITP in a dog, highlighting the diagnostic challenges and clinical management within veterinary neurology.
Case description
A 10-year-old female spayed mixed-breed dog was presented with an acute onset of an inability to stand and a history of seizure-like episodes. Over the past month, the dog was reported to have decreased activity, reluctance to jump, intermittent thoracic limb lameness and an acute onset of progressive seizure-like activity over the preceding 7 days. These episodes were characterized by whole-body rigidity with non-responsive mentation lasting 15–20 min that occurred a few times daily. Prior to referral, the dog was evaluated at an emergency clinic and subsequently at the primary care clinic, where a complete blood count, serum chemistry panel, and IDEXX 4Dx Plus SNAP® test were performed. Hematological evaluation revealed a mild leukocytosis (white blood cell count (WBC) 18.4 × 103/μL; reference interval (RI) 4.0–15.5 × 103/μL) and a severe thrombocytopenia (platelet count 21 × 103/μL; RI 170–400 × 103/μL) with a few small platelet clumps and an estimated decreased platelet count on blood smear. Biochemical analysis showed a mild hyperglobulinemia (3.7 g/dL; RI 1.6–3.6 g/dL) and a mild increase in alkaline phosphatase (138 IU/L; RI 5–131 IU/L). Electrolyte levels were within normal limits. The dog was negative on the 4Dx SNAP test for Dirofilaria immitis antigen and Borrelia burgdorferi, Anaplasma spp., and Ehrlichia spp. antibodies. The primary veterinarian prescribed an oral non-steroidal anti-inflammatory drug (NSAID; meloxicam) for suspected osteoarthritis and recommended referral if no improvement was seen.
On presentation, the dog was quiet, alert, and responsive. The physical examination was largely unremarkable except for an estimated 5% dehydration status. During the physical examination, the dog had another seizure-like episode, characterized by whole-body stiffness, falling over, and mydriasis with absent pupillary light reflexes in both eyes. The blood pressure during this episode was mildly increased (systolic, 150 mmHg; diastolic, 109 mmHg; mean arterial pressure [MAP], 113 mmHg). The neurological examination after the episodes showed a dull to obtunded mentation, absent proprioception in the right thoracic and pelvic limbs, and episodes of stiff thoracic limbs and neck extension. The dog was given a single dose of mannitol 0.5 g/kg intravenously (IV), maropitant 1 mg/kg IV, and was started on Plasma-Lyte at 60 mL/kg/day IV for overnight stabilization until transfer to the neurology service the next day. Overnight, the dog had an episode of neck extension, ptyalism, and increased respiratory effort that resolved without intervention; thoracic and abdominal point-of-care ultrasound (POCUS) was performed and did not show any B-lines or pleural, pericardial, or peritoneal effusion.
On the day of transfer to the neurology service, the dog was mildly to moderately dull with a low head and neck carriage. Due to the overnight seizure-like episodes, levetiracetam 32 mg/kg IV every 8 h was also added to the treatment. Physical examination revealed two focal ecchymoses on the left thoracic region, approximately 1.5 cm in length. Neurological examination showed mild ambulatory tetraparesis with vestibular ataxia, a tendency for a left head turn, mild left head tilt, and a single self-limiting episode of truncal titubation. On cranial nerve examination, the dog had absent menace response bilaterally, isocoric pupils with moderate to severe bilateral mydriasis, intact direct and indirect pupillary light reflexes bilaterally, and positional nystagmus shifting from vertical to rotary, with varying fast phase to the left or right. The dog had normal placing and hopping in the left thoracic and pelvic limbs, and inconsistent mildly delayed paw placement and hopping response in the right thoracic and pelvic limbs. Fundic examination showed bilateral retinal hemorrhage, worse in the right eye. The rest of the neurological exam was unremarkable. The neuroanatomical localization was multifocal brain (left forebrain and brainstem). Differential diagnoses included inflammatory or infectious meningoencephalitis, intracranial neoplasia, or vascular disease. A complete blood count and biochemistry panel were performed. Hematological evaluation revealed a mild leukocytosis (white blood cell count (WBC) 19.2 × 103/μL; RI 5.7–14.2 × 103/μL), reticulocytosis (absolute reticulocyte count 125.2 × 103/μL; RI 11–92 × 103/μL) with a normal hematocrit (46%; RI 41–58%), and a severe macrothrombocytopenia (platelet count 9 × 103/μL; RI 186–545 × 103/μL; mean platelet volume (MPV) 26.8 fL; RI 8.4–14.1 fL) with no visible platelet clumps and a markedly decreased platelet count on blood smear. Biochemical analysis showed a mild increase in alkaline phosphatase (148 IU/L; RI 7–115 IU/L). Comprehensive coagulation profile, including activated partial thromboplastin time (aPTT), prothrombin time (PT), quantitative fibrinogen (Clauss method), antithrombin activity (anti-IIa chromogenic substrate method), and quantitative cross-linked fibrin degradation products (D-dimer), was within normal limits. Additional diagnostics included thoracic radiography, which was unremarkable; abdominal ultrasound showed a moderately diffusely enlarged and heterogeneous spleen, most likely representing a benign or reactive change such as lymphoid hyperplasia or extramedullary hematopoiesis. Brain magnetic resonance imaging (MRI) was pursued the same day to further characterize the neurological clinical signs.
MRI was performed under general anesthesia using a 1.5-T scanner (Vantage Orian, Canon Medical Systems USA, Inc., Tustin, CA) with the dog in dorsal recumbency and a knee coil. Transverse-plane sequences included T1-weighted, T2-weighted, T2*-weighted gradient-recalled echo (GRE), fluid-attenuated inversion recovery (FLAIR), diffusion-weighted imaging (DWI), T1-weighted magnetization-prepared rapid gradient echo (MPRAGE), and pre- and post-contrast T1-weighted images (including a fat-saturated post-contrast T1-weighted sequence and a post-contrast T1 MPRAGE). Sagittal and dorsal planes included T2-weighted and post-contrast T1-weighted sequences. Apparent diffusion coefficient (ADC) maps were generated from the DWI data. Gadodiamide (Omniscan, GE Healthcare, Chicago, IL) was administered intravenously at a dose of 0.1 mmol/kg (1.3 mL total).
During image analysis, minimally expanded, bilateral crescent-shaped, extra-axial fluid accumulations were observed overlying the entire dorsolateral convexity of both cerebral hemispheres and not extending along the medial and basal borders (Figure 1). The fluid collections were homogeneously hyperintense on T1-weighted, T2-weighted, and FLAIR images with peripheral susceptibility artifact on T2* GRE images. Contrast enhancement could not be discerned because the fluid collections were already T1 hyperintense prior to contrast medium administration. No additional site of abnormal contrast enhancement was detected. The cerebral hemispheres were mildly compressed by the fluid collections and mildly diffusely swollen without midline shift. The compressed and swollen cerebrum displaced the diencephalon and mesencephalon caudally with compression of the rostral margin of the cerebellum and caudoventral cerebellar displacement extending to, but not through, the foramen magnum. Transverse imaging at the level of the tentorium confirmed that this caudal displacement was restricted to the midline tectum, with the parahippocampal gyri and cerebral cortex remaining normally positioned without caudal transtentorial herniation. At the level of the C2 and C3 vertebrae, the spinal cord was mildly swollen with poorly defined, locally extensive, non-contrast-enhancing, T2-weighted hyperintensity (Figure 2). The findings were interpreted as mild, bilateral, late subacute, intracranial subdural hemorrhage with mild secondary spinal cord edema. Intracranial pressure was estimated to be high based on intracranial displacement of the brain and spinal cord edema. As no other signs of trauma were observed, subdural hemorrhage secondary to severe thrombocytopenia was prioritized. Cerebrospinal fluid (CSF) analysis was not performed due to the patient’s severe thrombocytopenic state and suspected high intracranial pressure.
[图：Transverse-plane canine brain magnetic resonance images demonstrating mild, bilateral, late subacute, subdural hemorrhag]
[图：Magnetic resonance imaging (MRI) of the brain. (A) Sagittal-plane T2-weighted canine brain MRI image showing mild, diffu]
Based on the clinical presentation, blood work, and imaging findings, the dog was diagnosed with a presumptive primary immune-mediated thrombocytopenia with secondary cerebral subdural hemorrhage. The dog was given a single dose of dexamethasone sodium phosphate of 0.1 mg/kg IV during recovery. The anesthesia and recovery were uneventful, and the dog was noted to be standing and walking in the intensive care kennel on its own with a quiet mentation. Three hours after recovery, on repeat neurological evaluation by the attending veterinarian, the dog was acutely and severely obtunded and exhibited moderate respiratory effort, though cardiopulmonary auscultation was unremarkable. This was subsequently followed by abrupt bradypnea, cyanosis of the tongue, and respiratory arrest, which rapidly progressed to cardiopulmonary arrest. Given the patient’s do-not-resuscitate (DNR) status, cardiopulmonary resuscitation (CPR) was not performed. The owner declined a postmortem examination.
Discussion
This report describes a dog with multifocal neurological signs due to presumed bilateral intracranial subdural hemorrhages secondary to primary ITP, leading to a cardiorespiratory arrest. Only two individual case reports have previously described hemorrhagic myelopathies associated with presumed ITP in dogs (10, 11). In Guevar et al., the dog presented with an acute progressive ambulatory to non-ambulatory tetraparesis within 24 h, following a six-week history of thoracolumbar hyperesthesia and abnormal gait. The dog was diagnosed with presumed ITP based on a platelet count of 3 × 103/μL and no other underlying cause. Cervical and thoracolumbar MRI revealed multifocal intradural-extramedullary and intramedullary hemorrhage within the thoracolumbosacral spinal cord. Despite the resolution of thrombocytopenia with immunomodulatory therapy, the dog’s neurological status deteriorated to tetraplegia with absent nociception in the pelvic limbs, and the dog was later euthanized. Postmortem examination showed multifocal intradural, subarachnoid, and intramedullary hemorrhage throughout the entire spinal cord with variable compression within the vertebral column. The second case describes a similar neurological presentation, where a dog with acute onset of tetraplegia with decreased sensation in all four limbs was diagnosed with presumed ITP. MRI of the cervical spine showed intramedullary hemorrhage. Immunomodulatory therapy was initiated with a combination of dexamethasone sodium phosphate and cyclosporine. The dog’s thrombocytopenia improved and resolved by day 5 after presentation, and the neurological status improved to non-ambulatory tetraparesis. At follow-up 5 weeks after discharge, the dog exhibited only mild generalized ataxia with no paresis present and maintained a normal platelet count.
These two cases highlight the potential severity of secondary neurological manifestations of ITP and emphasize the variability of clinical outcomes despite achieving normalized platelet counts. ITP in dogs is a complex disease characterized by a lack of direct correlation between the severity of thrombocytopenia and the anatomical location or clinical signs of the hemorrhage (12). While subclinical signs in dogs and cats are common even with severe thrombocytopenia, others may present with catastrophic, life-threatening hemorrhage. Studies of prognostic indicators have indicated that the presence of melena or high blood urea nitrogen (BUN) concentration may suggest a poor prognosis for affected dogs (8). However, no specific studies have evaluated whether the presence of neurological signs, either encephalopathies or myelopathies, poses a greater risk for recurrence or is associated with poor outcomes. In the case described here, neither melena nor increased BUN concentration was noted. Instead, the dog suffered sudden death following acute progression to obtundation, abrupt bradypnea, and cardiopulmonary arrest within 24 h of being diagnosed with ITP and presumed bilateral cerebral subdural hemorrhage. In the absence of a necropsy, a definitive postmortem cause of death could not be confirmed. Although pulmonary thromboembolism (PTE) is a recognized differential for sudden death in dogs, PTE typically presents with hyperventilation and tachypnea due to ventilation-perfusion mismatch and stimulation of the respiratory center. In contrast, the abrupt onset of bradypnea following obtundation in this patient strongly suggests brainstem dysfunction—specifically, suppression of the medullary respiratory center and ascending reticular activating system, potentially due to a rapid increase in intracranial pressure. At the time of MRI, mild diffuse cerebral swelling was evidenced by caudal displacement of the diencephalon and mesencephalon, along with compression of the rostral cerebellum and caudal displacement of the cerebellum toward the foramen magnum. Acute secondary events—such as dynamic expansion of the intracranial hemorrhage, progressive vasogenic edema, or increased intracranial pressure—are plausible contributors to the patient’s acute clinical deterioration; however, in the absence of necropsy confirmation, the definitive cause of death remains undetermined.
A primary limitation of this case is the absence of a postmortem examination, which precludes definitive pathological confirmation of the cerebral subdural hemorrhage and its direct relationship to the patient’s clinical deterioration and death. Furthermore, although immediate postmortem CSF analysis was feasible and could have aided in ruling out primary infectious or inflammatory meningoencephalitis or other etiologies, CSF findings following intracranial hemorrhage are often nonspecific, typically yielding variable cell populations and secondary erythrophagocytosis. Nonetheless, the combination of characteristic MRI features, comprehensive systemic diagnostics, and the patient’s clinical course provides antemortem evidence supporting a presumptive diagnosis of cerebral subdural hemorrhage as a rare neurological manifestation of primary immune-mediated thrombocytopenia.
Presently, ITP in dogs remains a diagnosis of exclusion (5). Due to its heterogeneous nature, standardizing the diagnostic approach and establishing definitive criteria have historically been challenging. Recently, the American College of Veterinary Internal Medicine (ACVIM) published an updated consensus statement on the diagnosis of ITP, providing a diagnostic algorithm for systemic exclusion of other underlying etiologies for thrombocytopenia. In the case described here, clinicopathological findings included severe thrombocytopenia (9 × 103/μL) and increased mean platelet volume. Per the ACVIM consensus guideline, although dogs with ITP may have more severe thrombocytopenia than dogs with non-immune-mediated causes, clinical overlap persists. However, a platelet count cutoff < 20,000/μL strongly supports a diagnosis of ITP. A bone marrow examination was not performed in this case, and the current literature suggests there is insufficient evidence to determine whether bone marrow evaluation significantly improves the diagnostic accuracy of primary ITP in dogs. The consensus statement strongly recommends performing routine coagulation testing (aPTT, PT) and fibrinolysis markers (D-dimers, quantitative fibrinogen) during initial diagnostic workup to rule out concurrent consumption coagulopathies. In our case, these parameters were within normal limits. Comorbidity screening in ITP guidelines includes vector-borne infections; the dog tested negative for Dirofilaria immitis antigen and Borrelia burgdorferi, Anaplasma spp., and Ehrlichia spp. antibodies. Prior to referral, a dose of a non-steroidal anti-inflammatory drug (meloxicam) was administered. However, both the severe thrombocytopenia and clinical signs were present before this medication was given. Although the overall evidence for drug-induced ITP in dogs is limited, there is a high-to-intermediate level of evidence for cefazedone and sulfonamide drugs as triggers for ITP. However, the clinical timeline in this case most effectively excludes meloxicam as the primary inciting trigger.
Overall, the diagnostics performed in this case aligned with the ACVIM consensus statement for primary ITP. For dogs presenting with primary ITP and a platelet count < 50,000/μL, such as in this case, the consensus guideline suggests that there is moderate evidence that the initial admission platelet count does not impact disease outcome or therapeutic response. Rather, the severity of the platelet count may correlate predominantly with active clinical hemorrhage at the time of presentation, manifesting in this case as ecchymoses, bilateral retinal hemorrhage, and the intracranial subdural hemorrhage. Retinal hemorrhages are well documented as a severe ocular lesion in dogs with thrombocytopenia, with the lesion severity being associated with platelet count (13). Generally speaking, spontaneous bleeding typically occurs with a platelet count of < 30,000/μL; however, no prospective veterinary studies have compared long-term outcomes or site-specific fatality rates in dogs with primary ITP with mild (platelet count > 75,000/μL) versus moderate to severe (platelet count < 50,000/μL) thrombocytopenic states.
In the human literature, ITP is an uncommon hematological condition, and intracranial hemorrhage is considered a rare but fatal complication in severe thrombocytopenic patients, with an estimated mortality rate of 25–55% (14, 15). There is limited consensus regarding the optimal management of life-threatening intracranial hemorrhage, particularly for emergent surgical decompression to alleviate increased intracranial pressure during rapid deterioration of neurological signs. Only a single human case report has documented an excellent clinical outcome following a combination of emergency splenectomy and craniotomy (15). In our case, neurological deterioration and eventual cardiopulmonary arrest were rapid, despite an initial uneventful anesthesia and recovery. Given the lack of larger-scale prospective studies or established surgical guidelines for intracranial hemorrhagic events in veterinary patients, this case stands as the first case in the veterinary literature that demonstrates severe and fatal subdural hemorrhage as a primary neurological manifestation of ITP.
Although ITP has not been documented as a common etiology or differential for myelopathies or encephalopathies in dogs and cats, secondary central nervous system hemorrhage can occur due to a variety of underlying conditions. Secondary causes of hemorrhagic encephalopathies include vascular malformations, neoplasia, vasculitis associated with infectious or non-infectious meningoencephalitis, coagulopathies, ischemic infarcts with hemorrhagic transformations, metabolic or nutritional disorders, or heat stroke. Structurally, hemorrhage can be categorized as extra-parenchymal, epidural, subdural, subarachnoid, subependymal, intraventricular, and intraparenchymal based on location. While subdural hemorrhage is often clinically defined as blood accumulation within the space between the dura mater and arachnoid membrane, neuroanatomically it is recognized as an intradural lesion, as no true physiological subdural space exists. Rather, this pathologically expanded space is created when mechanical trauma or spontaneous vascular extravasation cleaves the structurally weak dural border cell layer, along with rupture of bridging vessels (1, 16). In the human literature, subdural hematoma most commonly occurs with traumatic brain injury, though it can occur in rare instances with vascular malformations and aneurysms; recognized risk factors include age, concurrent coagulopathies, anticoagulant therapy, brain atrophy, and alcohol disorders (17, 18). Conversely, in the veterinary literature, only a few subdural hemorrhages in dogs have been reported. Documented underlying etiologies include trauma, hydrocephalus, complications of lumbar myelography and ventriculoperitoneal shunt placement, and secondary venous tearing with brain atrophy due to ceroid lipofuscinosis (2, 3, 19–22). Typical neurological signs include seizures, altered mentation, and concurrent neurological deficits.
On MRI, cerebral subdural hematoma characteristically appears as single or multiple crescent-shaped extra-axial fluid collections with variable signal intensities dictated by the chronicity of hemorrhage. As blood products degrade and settle into gravity-dependent layers, the composition of the fluid accumulation changes over time, a unique phenomenon of fluid-filled spaces that does not occur in solid tissues. The MRI appearance of intracranial hemorrhage is highly complex, with signal intensity determined primarily by the T1 relaxation and T2* susceptibility effects of hemoglobin degradation and oxygenation over time. The subsequent degradation of hemoglobin, progressing from oxyhemoglobin to deoxyhemoglobin, intracellular methemoglobin, extracellular methemoglobin, and eventually ferritin and hemosiderin, alters the magnetic properties of the hematoma over time. These predictable biochemical variations allow the distinct stages of intracranial hemorrhage (hyperacute, acute, early subacute, late subacute, and chronic) to be determined on MRI (23).
Based on the MRI features in the present case, the dog’s cerebral subdural hemorrhage is most characteristic of the late subacute stage. This classification is supported by the T1-weighted and T2-weighted hyperintensity with susceptibility artifact on T2* GRE, and a characteristically expanded crescent-shaped subdural space surrounding both cerebral hemispheres. Crucially, although the diagnosis of ITP was established just prior to the MRI diagnosis of a subdural hemorrhage, the neurological signs, specifically seizure episodes, began approximately 1 week prior. This clinical timeline correlates closely with the late subacute to early chronic hemorrhage stage, which generally spans from days to months and further validates the severe thrombocytopenia and primary ITP as the etiology of this patient’s neurological signs.
Conclusion
This case documents a rare clinical manifestation of ITP that has not previously been reported in the veterinary literature. ITP should be considered as a differential diagnosis for a dog presenting with acute progressive intracranial multifocal neurological deficits until an adequate platelet count is confirmed. Although ITP generally carries a favorable prognosis when managed appropriately, severe complications such as intracranial hemorrhage may result in significant morbidity and may be associated with a poor outcome.