兽研[内容请查看原文]
引言
尾部硬膜外局部麻醉给药是小动物麻醉与镇痛中的成熟组成部分，尤其适用于涉及后肢、会阴区和尾侧腹部的操作(1)。在犬和猫中，腰骶或骶尾入路通常作为单次注射技术使用，除非明确指定置入导管。由于单次注射后通常无法根据观察到的节段范围进行滴定，所选剂量-容量方案便成为核心设计变量。以ml/kg表示容量可将注射容量按体重标准化，但并不能确定拟受影响的脊髓节段、神经根或操作特异性伤害性靶点(1, 2)。
当前兽医证据涉及若干问题。解剖学和影像学研究显示注射物可能分布的位置。功能测试在三个独立维度上有所不同：所施加的刺激、所刺激的组织或伤害性区域，以及用于定义衰减的反应。机械压迫或钳夹、受控电刺激和手术操作是不同的刺激；皮肤、肌肉、腹膜、卵巢和其他内脏是不同的靶点；而有意识的伤害防御行为、运动反应以及自主神经或血流动力学变化是不同的反应指标。麻醉状态、刺激强度和时机进一步影响解读。临床研究通过麻醉和镇痛需求、方案规定的补救用药、恢复和术后疼痛来估计复合围手术期表现。这些终点具有信息价值，但不可互换。
靶点特异性阻滞可能有助于临床成功，但临床成功不能回溯性地确定该阻滞的范围或完整性。
在本分析中，可靠的尾侧腹部覆盖是指在预先定义的、与操作相关的躯体 and 内脏靶点上，可归因于硬膜外局部麻醉药的可重复双侧功能衰减，并在相关手术事件期间持续存在。解剖靶点可达性、刺激特异性功能衰减、生理耐受性和复合围手术期临床表现被视为不同但相关的证据领域。
本分析聚焦于尾部单次硬膜外局部麻醉给药。更靠头侧的椎管内或非椎管内区域技术可能适用于特定操作，但这些方法的比较评估不在本分析范围内。硬膜外阿片类药物、α-2受体激动剂和其他椎管内辅助药物不在其主要范围内，除非它们构成正在解读其复合结局的临床方案的一部分。
[内容请查看原文]
证据领域及其支持的论断
区分四个证据领域。解剖分布是指染料、造影剂或影像学证据表明注射物到达了某一椎体水平、椎间孔、神经根区域或其他预先定义的解剖靶点。刺激特异性功能衰减是指在特定条件下，对特定组织或伤害性区域中定义刺激的预定义反应出现可重复的降低。生理耐受性是指该方案对呼吸、血流动力学和运动的影响，包括需要辅助通气或心血管支持。复合围手术期临床表现是指在整体麻醉计划内的麻醉和镇痛需求、方案规定的补救用药、恢复和术后疼痛。
这四个领域并非按证据由弱到强排序。每个领域都可被严格研究，但每个领域支持不同类型的论断。分布研究可以提供解剖可达性的强证据，而不测试神经功能；而临床试验可以确立增量获益，而不确定产生该获益的节段机制。在功能领域内，某一结果仅支持所测试的刺激方法、刺激组织、反应标准、麻醉状态、刺激强度和时机的组合；不能假定其代表另一种组合。
研究为立意选择，因为它们说明四种证据类型中的一种或多种、在同一设计中测量一种以上类型，或直接为所提出的犬和猫比较提供信息。
只有当物种和模型、入路和给药、注射方案、体位和时机、解剖靶点、刺激和反应标准充分一致时，不同领域的结果才能相互关联。表1总结了每项所选研究测量了什么、显示了什么以及哪些问题仍未解决。
Table 1. Selected studies informing the evaluation of cranial abdominal coverage after epidural local anesthetic administration.
| Study | Population/model and design | Epidural strategy | Endpoint(s) relevant to this analysis | Finding relevant to cranial coverage | Question left unresolved |
|---|---|---|---|---|---|
| Freire et al. ( 5 ) | 20 healthy adult female mixed-breed dogs; prospective randomized four-group experiment ( n = 5/group) | Lumbosacral catheter advanced 5 cm cranially; dogs standing; 0.25% bupivacaine plus methylene blue at 0.2, 0.4, 0.6, or 0.8 ml/kg | Mechanical skin pinch; dorsal skin, tail, and pelvic-limb sites; conscious nocifensive response; motor scores; postmortem dural staining | Dorsal cutaneous block increased through 0.6 ml/kg. Staining reached the cervical region in some dogs at 0.4 ml/kg and all dogs at 0.6 ml/kg; gross staining and cutaneous response did not scale identically. | No procedure-specific cranial abdominal somatic or visceral test. Total bupivacaine dose increased with volume; catheter advancement, standing position, and a cranial endpoint ceiling limit extrapolation |
| Aso et al. ( 6 ) | 6 healthy Beagles; prospective randomized three-period crossover experiment | Co1-Co2 injection; equal volumes of 1% ropivacaine and iotrolan 240 at 0.1, 0.2, or 0.3 ml/kg | Controlled electrical stimulation at predefined nonvisceral body sites; HR/MAP response; CT distribution and leakage; motor function | Increasing volume extended the region in which electrical stimulation failed to elicit a 15% or greater HR/MAP response, whereas CT spread did not differ significantly; contrast laterality and response laterality were discordant. | Electrically evoked autonomic responses at nonvisceral sites are not equivalent to cranial abdominal surgical responses. Contrast admixture and foraminal leakage complicate extrapolation |
| Cima et al. ( 7 ) | 10 analyzed living female dogs and 10 thawed canine cadavers; prospective comparative anatomical study | Fluoroscopy-guided L1-L2 injection of 0.2 ml/kg 0.25% bupivacaine plus iohexol | Fluoroscopic cranial distribution | Distribution was more extensive in living dogs than in thawed cadavers, indicating that cadaveric models may underestimate in vivo distribution | More cranial access rather than a caudal single shot; no procedure-specific functional test or individual distribution-to-response analysis |
| Freitag et al. ( 8 ) | 6 healthy female Beagles; prospective randomized crossover experiment | Lumbosacral catheter; 0.25% bupivacaine plus iopamidol; BW dosing at 0.2 ml/kg (1.81 ± 0.21 ml) vs. OCL-based dosing (3.29 ± 0.74 ml); sternal recumbency | CT rostral distribution; postanesthetic sensory and motor assessments; cardiorespiratory variables | OCL-based dosing delivered 1.82-fold more volume and increased mean CT reach from 19 ± 2 to 22 ± 2 vertebrae. No differences were detected in the postanesthetic sensory, motor, or cardiorespiratory outcomes; farther mean spread did not establish a functional advantage or improved predictive precision | Six homogeneous small dogs; no surgical or cranial abdominal target. The SD of CT reach was 2 vertebrae under both methods, and the study was not designed to test whether OCL reduced between-animal variability or prediction error after calibration to a common target. Catheter use and contrast admixture also limit extrapolation |
| Freitag et al. ( 9 ) | 22 canine cadavers; controlled comparative cadaver study | Lumbosacral catheter; mixture of iopamidol and new methylene blue; iopamidol assessed by early CT and dye by later dissection | Early CT distribution of iopamidol; later anatomical distribution of new methylene blue | New methylene blue observed at later dissection extended farther cranially than iopamidol on earlier CT, indicating that these anatomical endpoints are not interchangeable | Marker, assessment method, and assessment time differed, so the discrepancy cannot be attributed to marker alone. Neither endpoint measured neural function |
| dos-Santos et al. ( 10 ) | 50 feral female cats undergoing elective ovariectomy; prospective randomized clinical comparison ( n = 25/group) | Sacrococcygeal 0.25% bupivacaine 0.3 ml/kg vs. bilateral quadratus lumborum block with 0.25% bupivacaine 0.4 ml/kg; multimodal general anesthesia | Intraoperative rescue use and hemodynamics; extubation; postoperative pain classification and motor block | Only 3/25 epidural cats required intraoperative rescue, and postoperative pain classifications were similar to quadratus lumborum block, indicating favorable composite performance. | No anatomical target mapping or segmental functional test. Favorable composite performance cannot establish bilateral T10-T11 reach or ovarian target blockade; the active |
| Hypotension, extubation delay, and motor block were more frequent after epidural administration | comparator and multimodal protocol also contribute to the observed outcome | ||||
| Zahra et al. ( 11 ) | 36 female cats undergoing elective ovariohysterectomy; prospective randomized three-group clinical trial ( n = 12/group) | Sacrococcygeal or lumbosacral 0.33% levobupivacaine 0.3 ml/kg vs. 0.9% saline at the same volume; multimodal general anesthesia | End-tidal isoflurane concentration; intraoperative fentanyl requirements; cardiorespiratory variables; postoperative pain, rescue morphine, and time to standing | Both levobupivacaine groups had modestly lower isoflurane and fentanyl requirements than saline, but postoperative pain and rescue analgesia did not differ. The same 0.3 ml/kg nominal volume therefore produced limited rather than uniformly strong clinical effects | No injectate mapping or segmental functional test. The findings are compatible with incomplete procedure-relevant coverage but do not establish its anatomical or functional extent |
| Wong et al. ( 12 ) | 7 healthy intact female domestic shorthair cats; prospective within-animal CT distribution study | Sacrococcygeal stimulating needle; iohexol 44 mg/mL in four sequential 0.25 mL/kg boluses with CT after each bolus (cumulative 0.25-1.0 mL/kg) | Left- and right-sided cranial CT distribution; bilateral reach to the T10-T11 intervertebral foramina | Bilateral T10-T11 reach occurred in 3/7 cats at 0.75 ml/kg and in 6/7 by 1.0 ml/kg. At 1.0 ml/kg, 1 cat remained short of the target | Iohexol rather than local anesthetic; sequential rather than single-bolus administration; no functional blockade or clinical analgesic endpoint. The study supports 1.0 ml/kg as an adequacy-oriented candidate and 0.75 ml/kg as a lower comparator, not as validated regimens |
| Dias et al. ( 14 ) | 6 healthy adult mixed-breed dogs; prospective randomized crossover experiment | Lumbosacral 0.25% bupivacaine at 0.2 or 0.4 ml/kg during spontaneous ventilation at 1.3 MAC isoflurane | Invasive hemodynamics; ventilation; blood gases; clinical signs through 90 min | The 0.4 ml/kg regimen produced greater mean arterial pressure depression, more hypotension, and transient hypoventilation | No functional coverage endpoint and no 0.6 ml/kg regimen. Constant isoflurane concentration and experimental conditions limit direct clinical generalization |
| Parreño et al. ( 15 ) | 210 dogs undergoing abdominal surgery; single-center retrospective observational study with nonconcurrent cohorts (OCL n = 110; BW n = 100) | L7-S1 levobupivacaine with morphine or buprenorphine; total volume calculated by OCL or BW | Post-epidural hypotension; administered volume; multivariable and machine-learning risk models | Mean volume was 0.448 vs. 0.241 ml/kg and hypotension occurred in 49.1% vs. 33% of dogs with OCL vs. BW calculation | Retrospective nonconcurrent cohorts, combined drug regimens, and multiple perioperative differences preclude a causal volume-only inference; no coverage endpoint |
[内容请查看原文]
颅侧腹部覆盖的间接证据的局限性
[内容请查看原文]
解剖分布不能确立功能性阻滞
解剖学研究确定了注射物能够分布的范围，并识别了影响该分布的因素。经典的人体硬膜外腔冷冻微切片研究表明，硬膜外解剖结构和溶液分布并不均匀，受脂肪、结缔组织、血管结构、神经根袖、筋膜屏障及压力关系的影响（3, 4）。尽管这些研究并非针对犬或猫的种属特异性证据，但它们为将目标节段可见的硬膜外溶液解释为到达范围的证据提供了解剖学基础，而非对神经暴露或神经阻滞的直接测量。
功能性阻滞还要求在相关神经结构处有足够的局麻药暴露，包括浓度和持续时间两方面。在分布模式的颅侧边缘，药物暴露的量、浓度和持续时间可能低于更接近注射部位的节段。因此，不能假定椎体水平或椎间孔处的造影剂或染料能够证明有效的神经根阻滞。脑膜扩散、脑脊液袖内的稀释以及随后对神经组织的接触也可能影响阻滞，但这些过程与影像学表现之间的定量关系在犬和猫中仍未被完全阐明。
所选研究表明，即使在同一实验中，影像学与所测试的功能反应也可能出现分离（表1）。Freire等发现，部分犬在0.4 ml/kg时出现颈部染色，而平均背侧皮肤阻滞在0.6 ml/kg时进一步增加；染料终点已接近其颅侧上限，且所测试的反应为皮肤反应而非腹部内脏反应（5）。Aso等将CT分布与电诱发的自主神经反应进行配对，发现影像学与反应侧别之间存在不一致；所测试的部位为非内脏部位（6）。
表观分布还取决于模型和测量方法。Cima等观察到活体犬中的扩散范围比解冻尸体更广泛（7）。Freitag等发现基于OCL的给药方案比基于体重的给药方案平均CT扩散更远，但该研究并未确立功能优势或更低的动物间变异性（8）；在一项相关的尸体实验中，后续解剖显示亚甲蓝的颅侧分布比先前CT显示的碘帕醇分布更远（9）。
综上，这些研究表明可见扩散随实验模型和评估方法而变化，且可能与所测试的功能反应不匹配。研究设计的细节及未解决的问题总结于表1。
[内容请查看原文]
临床获益不能确立颅侧阻滞的范围
0.3 ml/kg的临床发现存在异质性（表1）。在dos-Santos等的研究中，骶尾部布比卡因与术中补救用药少及术后疼痛分级相似相关，后者与多模式方案中的腰方肌阻滞相当（10）。在Zahra等的研究中，相同标称容积仅产生适度的术中节省效应，且无术后镇痛获益（11）。因此，0.3 ml/kg并不能界定统一的临床特征，也不能揭示阻滞的解剖范围。
Wong等表明，从这些临床发现中得出的解剖学推断仍然十分有限。在0.75 ml/kg时，7只猫中有3只达到双侧T10-T11，在1.0 ml/kg时7只中有6只达到，仍有1只猫未达到目标（12）。注射物、给药顺序、时机和终点的差异妨碍了直接的剂量-反应比较。然而，2.5至3.3倍的容积差异使得难以假定0.3 ml/kg的良好结局通常反映了 consistent、完整的双侧覆盖。这些研究并未表明0.3 ml/kg在任何单只猫中失败。良好结局仍可能反映部分猫的完整覆盖、部分但有用的阻滞，或其他麻醉和镇痛干预的实质性效应。
[内容请查看原文]
候选尾侧剂量-容积方案
[内容请查看原文]
犬：0.6 ml/kg的前瞻性评估
犬类数据为选择候选方案进行前瞻性测试提供了经验性锚点。在Freire等的实验中，15–20分钟时背侧测试被阻滞的皮节平均数目在0.2、0.4、0.6和0.8 ml/kg方案下分别为5.0、14.2、20.2和21.0；T1计为20，颈部区域计为21（5）。在0.6 ml/kg时，所有犬的染料均到达颈椎区域。这些发现支持对0.6 ml/kg剂量-容积方案进行前瞻性评估，以实现犬的广泛颅侧覆盖。
该实验并未确定最小有效容积、最佳方案、常规可接受的临床剂量或容积特异性因果阈值。由于浓度固定为0.25%，布比卡因总剂量随容积增加而增加。导管向颅侧推进5 cm，且在所有的犬中均发现位于腹侧，犬保持站立位，颅侧反应量表存在上限，功能终点为背侧皮肤伤害感受而非颅侧腹部躯体或内脏输入。因此，0.6 ml/kg方案是一个有证据锚定的候选方案，可供直接测试。
[内容请查看原文]
猫：1.0与0.75 ml/kg的前瞻性比较
Wong 等人将 0.75 ml/kg 确定为首次测试的累计容积，在该容积下出现双侧 T10-T11 扩散，但该终点仅在 7 只猫中的 3 只中实现。至 1.0 ml/kg 时，6 只猫出现双侧扩散，尽管在 1 只猫中仍不完全（12）。因此，对于主要旨在测试靶点特异性充分性的研究，1.0 ml/kg 代表了更具可辩护性的主要候选总容积，而 0.75 ml/kg 则是用于考察靶点扩散与过度颅侧分布之间权衡的有用较低比较值。这两个值均未定义最低有效容积或保证均匀覆盖。在 0.75 ml/kg 时，碘海醇至少在 1 只猫中单侧到达 T4-T5；在 1.0 ml/kg 后，另有 3 只猫达到该水平（12）。序贯碘海醇方案也无法确定在任一容积下单次注射局部麻醉药会产生相同的解剖或功能结果。
随后的一项随机临床试验在卵巢子宫切除术后使用 0.75 ml/kg 的总骶尾容积，其中含无防腐剂吗啡 0.1 mg/kg 和罗哌卡因 0.93 mg/kg，发现与对照组相比早期疼痛评分更低且需要补救性 robenacoxib 更少（13）。由于注射在手术后进行，且将硬膜外阿片类药物与局部麻醉药联合使用，该试验评估的是术后镇痛表现，而非术中局部麻醉药特异性的颅侧覆盖。因此，它不能解决 0.75 ml/kg 对于 Wong 等人定义的靶点是否足够的问题。
[内容请查看原文]
生理耐受性的同步评估
覆盖范围和生理耐受性需要同步评估。Wong 等人观察到的颅侧分布确定了超出卵巢靶点扩散的可能性，但并未确定血流动力学或呼吸效应（12）。在六只异氟烷麻醉的犬中，0.4 ml/kg 的 0.25% 腰骶部布比卡因比 0.2 ml/kg 产生更大的平均动脉压下降、更多低血压和短暂性低通气（14）。在一项回顾性队列中，210 只接受腹部手术并使用腰骶部左布比卡因-阿片类药物方案的犬，基于枕-尾长度的计算产生更大的平均容积，并与基于体重的计算相比与更多硬膜外后低血压相关（15）。该关联不能因果性地分离容积，因为队列非同期且多个围手术期因素不同。两只犬在 0.6 ml/kg 布比卡因后出现严重低通气和心血管抑制，提供了病例级安全性信号而非发生率估计（16）。Freire 等人还观察到许多接受较大方案的动物出现虚脱和呼吸抑制，但未进行系统量化（5）。
以充分性为导向的方案的前瞻性评估应预先指定监测、通气标准、血流动力学补救阈值和停止规则。在相同局部麻醉药方案下，对猫进行 1.0 和 0.75 ml/kg 的比较，如果测量双侧靶点扩散、向颅侧胸区的分布、手术相关功能衰减、通气、血流动力学和运动效应，将最具信息价值。生理可接受性还将取决于患者的呼吸储备、气道控制、麻醉深度、监测以及立即获得通气和心血管支持的条件。
[内容请查看原文]
讨论
[内容请查看原文]
将主张与测量终点相匹配
四个问题可以指导对研究或临床报告的解释。第一，所主张的解剖和伤害性靶点是什么？腹壁腹侧切口、壁层腹膜操作、卵巢牵拉和颅侧内脏牵拉不一定代表同一区域。第二，测量了哪个证据领域：解剖扩散、刺激特异性功能衰减、生理耐受性，还是复合围手术期表现？第三，哪些机制与观察到的结果仍相容？第四，如果正在结合来自不同领域的证据，物种、入路、注射方案、靶点、时机、刺激和反应标准是否充分对齐？
当解剖扩散和功能充分性一起讨论时，两个问题特别有用：注射物是否到达预定义的双侧解剖靶点？局部麻醉药是否对预定义的手术相关刺激产生可重复的衰减并持续足够时间？这两个问题共同将一般性覆盖主张转化为两个可独立测试的终点。
[内容请查看原文]
连接解剖、功能、生理和临床证据
未来工作可以分阶段进行，而不是在一项试验中测量所有四种结局。解剖学研究可以首先定义双侧扩散和变异性。活体动物研究随后可以针对预定义的功能刺激测试相同或密切匹配的方案。在功能阶段，可以预先指定手术相关伤害性事件，同时保留适当的背景麻醉和补救方案。单独的研究可以评估生理效应和临床获益。表1显示现有研究覆盖了各个阶段，但尚无方法学可比的关联系列研究验证颅侧腹部覆盖。阶段之间的任何差异都应明确说明。
[内容请查看原文]
为候选方案选择比较值
候选比较方案应遵循所检验的问题。对于犬，0.6 ml/kg的比较方案应根据研究目的预先设定：是将候选方案与反映当前尾部实践的方案比较、与相邻的低剂量-体积方案比较，还是与剂量匹配的替代方案比较。对于猫，1.0 ml/kg可作为以充分性为导向的主要候选方案，0.75 ml/kg作为较低的比较方案。
[内容请查看原文]
区分剂量计算模型与体积-效应研究
基于体重（BW）和基于枕-尾长度（OCL）的计算方法之间的比较，旨在探讨哪一种患者体型描述指标能更一致地预测动物间实现的颅侧扩散。OCL的潜在优势在于降低动物间变异，而非仅仅获得更大的计算体积或更远的平均扩散。BW和OCL公式可产生不同的体积（2），且在六犬交叉研究中，基于OCL的给药增加了平均CT扩散；然而，两种方法报告的SD均为2个椎体（8）。未来研究应将每种公式校准至相同的解剖学目标，并比较剩余变异。若OCL更为一致，则可据此推导目标特异性系数，并对其功能阻滞和生理耐受性进行测试。
该问题与体积本身是否改变硬膜外扩散或阻滞是不同的。为分离体积效应，研究必须说明哪个因素保持恒定。剂量匹配方案可在相同总局麻药剂量下比较不同体积和浓度；在固定体积或固定浓度下的补充比较有助于区分体积、浓度、总剂量或其联合效应。
[内容请查看原文]
局限性
由于文献为目的性选择而非系统性检索，可能遗漏了相关报道。在所综述的研究中，没有一项从双侧目标定位到功能测试、生理评估和临床结局采用相同或密切匹配的局麻药方案。对于犬和猫的若干腹部手术，手术特异性躯体与内脏目标定位仍不完整。
[内容请查看原文]
结论
对于犬，0.6 ml/kg是拟议的前瞻性评估候选方案；对于猫，拟议的比较为1.0 vs. 0.75 ml/kg。两种动物中，双侧目标到达、手术相关功能衰减、生理耐受性和整体临床表现均应分别评估。这些体积是研究候选方案，而非临床给药推荐。
查看英文原文▾
[内容请查看原文]
Introduction
Caudal epidural local anesthetic administration is an established component of small-animal anesthesia and analgesia, particularly for procedures involving the pelvic limbs, perineal region, and caudal abdomen (1). In dogs and cats, lumbosacral or sacrococcygeal access is often used as a single-shot technique unless catheter placement is specified. Because a single-shot injection cannot usually be titrated after injection on the basis of its observed segmental extent, the selected dose-volume regimen becomes a central design variable. Expressing volume in ml/kg normalizes injectate volume to body weight but does not identify the spinal segments, nerve roots, or procedure-specific nociceptive targets intended to be affected (1, 2).
Current veterinary evidence addresses several questions. Anatomical and imaging studies show where an injectate may distribute. Functional tests differ along three independent dimensions: the stimulus applied, the tissue or nociceptive field stimulated, and the response used to define attenuation. Mechanical pressure or pinch, controlled electrical stimulation, and surgical manipulation are different stimuli; skin, muscle, peritoneum, ovary, and other viscera are different targets; and conscious nocifensive behavior, motor responses, and autonomic or hemodynamic changes are different response measures. Anesthetic state, stimulus intensity, and timing further affect interpretation. Clinical studies estimate composite perioperative performance through anesthetic and analgesic requirements, protocol-defined rescue use, recovery, and postoperative pain. These endpoints are informative but not interchangeable.
Target-specific blockade may contribute to clinical success, but clinical success cannot retrospectively establish the extent or completeness of that blockade.
In this analysis, reliable cranial abdominal coverage refers to reproducible bilateral functional attenuation attributable to the epidural local anesthetic at predefined procedure-relevant somatic and visceral targets, sustained for the relevant surgical events. Anatomical target reach, stimulus-specific functional attenuation, physiological tolerance, and composite perioperative clinical performance are treated as distinct but related evidentiary domains.
The analysis focuses on caudal single-shot epidural local anesthetic administration. More cranial neuraxial or non-neuraxial regional techniques may be appropriate for selected procedures, but comparative evaluation of those approaches is outside the scope of this analysis. Epidural opioids, alpha-2 agonists, and other neuraxial adjuncts are outside its primary scope, except when they form part of a clinical regimen whose composite outcome is being interpreted.
[内容请查看原文]
Evidentiary domains and the claims they support
Four evidentiary domains are distinguished. Anatomical distribution refers to dye, contrast, or imaging evidence that injectate reached a vertebral level, intervertebral foramen, nerve-root region, or other predefined anatomical target. Stimulus-specific functional attenuation refers to a reproducible reduction in a predefined response to a defined stimulus in a specified tissue or nociceptive field under specified conditions. Physiological tolerance refers to the respiratory, hemodynamic, and motor consequences of the regimen, including the need for assisted ventilation or cardiovascular support. Composite perioperative clinical performance refers to anesthetic and analgesic requirements, protocol-defined rescue use, recovery, and postoperative pain within the anesthetic plan as a whole.
The four domains are not ordered from weak to strong evidence. Each can be studied rigorously, but each supports a different type of claim. A distribution study can provide strong evidence of anatomical reach without testing neural function, whereas a clinical trial can establish incremental benefit without identifying the segmental mechanism that produced it. Within the functional domain, a result supports only the tested combination of stimulus method, stimulated tissue, response criterion, anesthetic state, stimulus intensity, and timing; it cannot be assumed to represent another combination.
Studies were selected purposively because they illustrate one or more of the four evidence types, measure more than one type within the same design, or directly inform the proposed canine and feline comparisons.
Results from different domains can be related only when species and model, access and delivery, injectate regimen, position and timing, anatomical target, stimulus, and response criterion are sufficiently aligned. Table 1 summarizes what each selected study measured, what it showed, and what remains unresolved.
Table 1. Selected studies informing the evaluation of cranial abdominal coverage after epidural local anesthetic administration.
| Study | Population/model and design | Epidural strategy | Endpoint(s) relevant to this analysis | Finding relevant to cranial coverage | Question left unresolved |
|---|---|---|---|---|---|
| Freire et al. ( 5 ) | 20 healthy adult female mixed-breed dogs; prospective randomized four-group experiment ( n = 5/group) | Lumbosacral catheter advanced 5 cm cranially; dogs standing; 0.25% bupivacaine plus methylene blue at 0.2, 0.4, 0.6, or 0.8 ml/kg | Mechanical skin pinch; dorsal skin, tail, and pelvic-limb sites; conscious nocifensive response; motor scores; postmortem dural staining | Dorsal cutaneous block increased through 0.6 ml/kg. Staining reached the cervical region in some dogs at 0.4 ml/kg and all dogs at 0.6 ml/kg; gross staining and cutaneous response did not scale identically. | No procedure-specific cranial abdominal somatic or visceral test. Total bupivacaine dose increased with volume; catheter advancement, standing position, and a cranial endpoint ceiling limit extrapolation |
| Aso et al. ( 6 ) | 6 healthy Beagles; prospective randomized three-period crossover experiment | Co1-Co2 injection; equal volumes of 1% ropivacaine and iotrolan 240 at 0.1, 0.2, or 0.3 ml/kg | Controlled electrical stimulation at predefined nonvisceral body sites; HR/MAP response; CT distribution and leakage; motor function | Increasing volume extended the region in which electrical stimulation failed to elicit a 15% or greater HR/MAP response, whereas CT spread did not differ significantly; contrast laterality and response laterality were discordant. | Electrically evoked autonomic responses at nonvisceral sites are not equivalent to cranial abdominal surgical responses. Contrast admixture and foraminal leakage complicate extrapolation |
| Cima et al. ( 7 ) | 10 analyzed living female dogs and 10 thawed canine cadavers; prospective comparative anatomical study | Fluoroscopy-guided L1-L2 injection of 0.2 ml/kg 0.25% bupivacaine plus iohexol | Fluoroscopic cranial distribution | Distribution was more extensive in living dogs than in thawed cadavers, indicating that cadaveric models may underestimate in vivo distribution | More cranial access rather than a caudal single shot; no procedure-specific functional test or individual distribution-to-response analysis |
| Freitag et al. ( 8 ) | 6 healthy female Beagles; prospective randomized crossover experiment | Lumbosacral catheter; 0.25% bupivacaine plus iopamidol; BW dosing at 0.2 ml/kg (1.81 ± 0.21 ml) vs. OCL-based dosing (3.29 ± 0.74 ml); sternal recumbency | CT rostral distribution; postanesthetic sensory and motor assessments; cardiorespiratory variables | OCL-based dosing delivered 1.82-fold more volume and increased mean CT reach from 19 ± 2 to 22 ± 2 vertebrae. No differences were detected in the postanesthetic sensory, motor, or cardiorespiratory outcomes; farther mean spread did not establish a functional advantage or improved predictive precision | Six homogeneous small dogs; no surgical or cranial abdominal target. The SD of CT reach was 2 vertebrae under both methods, and the study was not designed to test whether OCL reduced between-animal variability or prediction error after calibration to a common target. Catheter use and contrast admixture also limit extrapolation |
| Freitag et al. ( 9 ) | 22 canine cadavers; controlled comparative cadaver study | Lumbosacral catheter; mixture of iopamidol and new methylene blue; iopamidol assessed by early CT and dye by later dissection | Early CT distribution of iopamidol; later anatomical distribution of new methylene blue | New methylene blue observed at later dissection extended farther cranially than iopamidol on earlier CT, indicating that these anatomical endpoints are not interchangeable | Marker, assessment method, and assessment time differed, so the discrepancy cannot be attributed to marker alone. Neither endpoint measured neural function |
| dos-Santos et al. ( 10 ) | 50 feral female cats undergoing elective ovariectomy; prospective randomized clinical comparison ( n = 25/group) | Sacrococcygeal 0.25% bupivacaine 0.3 ml/kg vs. bilateral quadratus lumborum block with 0.25% bupivacaine 0.4 ml/kg; multimodal general anesthesia | Intraoperative rescue use and hemodynamics; extubation; postoperative pain classification and motor block | Only 3/25 epidural cats required intraoperative rescue, and postoperative pain classifications were similar to quadratus lumborum block, indicating favorable composite performance. | No anatomical target mapping or segmental functional test. Favorable composite performance cannot establish bilateral T10-T11 reach or ovarian target blockade; the active |
| Hypotension, extubation delay, and motor block were more frequent after epidural administration | comparator and multimodal protocol also contribute to the observed outcome | ||||
| Zahra et al. ( 11 ) | 36 female cats undergoing elective ovariohysterectomy; prospective randomized three-group clinical trial ( n = 12/group) | Sacrococcygeal or lumbosacral 0.33% levobupivacaine 0.3 ml/kg vs. 0.9% saline at the same volume; multimodal general anesthesia | End-tidal isoflurane concentration; intraoperative fentanyl requirements; cardiorespiratory variables; postoperative pain, rescue morphine, and time to standing | Both levobupivacaine groups had modestly lower isoflurane and fentanyl requirements than saline, but postoperative pain and rescue analgesia did not differ. The same 0.3 ml/kg nominal volume therefore produced limited rather than uniformly strong clinical effects | No injectate mapping or segmental functional test. The findings are compatible with incomplete procedure-relevant coverage but do not establish its anatomical or functional extent |
| Wong et al. ( 12 ) | 7 healthy intact female domestic shorthair cats; prospective within-animal CT distribution study | Sacrococcygeal stimulating needle; iohexol 44 mg/mL in four sequential 0.25 mL/kg boluses with CT after each bolus (cumulative 0.25-1.0 mL/kg) | Left- and right-sided cranial CT distribution; bilateral reach to the T10-T11 intervertebral foramina | Bilateral T10-T11 reach occurred in 3/7 cats at 0.75 ml/kg and in 6/7 by 1.0 ml/kg. At 1.0 ml/kg, 1 cat remained short of the target | Iohexol rather than local anesthetic; sequential rather than single-bolus administration; no functional blockade or clinical analgesic endpoint. The study supports 1.0 ml/kg as an adequacy-oriented candidate and 0.75 ml/kg as a lower comparator, not as validated regimens |
| Dias et al. ( 14 ) | 6 healthy adult mixed-breed dogs; prospective randomized crossover experiment | Lumbosacral 0.25% bupivacaine at 0.2 or 0.4 ml/kg during spontaneous ventilation at 1.3 MAC isoflurane | Invasive hemodynamics; ventilation; blood gases; clinical signs through 90 min | The 0.4 ml/kg regimen produced greater mean arterial pressure depression, more hypotension, and transient hypoventilation | No functional coverage endpoint and no 0.6 ml/kg regimen. Constant isoflurane concentration and experimental conditions limit direct clinical generalization |
| Parreño et al. ( 15 ) | 210 dogs undergoing abdominal surgery; single-center retrospective observational study with nonconcurrent cohorts (OCL n = 110; BW n = 100) | L7-S1 levobupivacaine with morphine or buprenorphine; total volume calculated by OCL or BW | Post-epidural hypotension; administered volume; multivariable and machine-learning risk models | Mean volume was 0.448 vs. 0.241 ml/kg and hypotension occurred in 49.1% vs. 33% of dogs with OCL vs. BW calculation | Retrospective nonconcurrent cohorts, combined drug regimens, and multiple perioperative differences preclude a causal volume-only inference; no coverage endpoint |
[内容请查看原文]
Limits of indirect evidence for cranial abdominal coverage
[内容请查看原文]
Anatomical distribution does not establish functional blockade
Anatomical studies establish where an injectate can distribute and identify factors that shape that distribution. Classical cryomicrotome studies of the human epidural space showed that epidural anatomy and solution distribution are nonuniform and are influenced by fat, connective tissue, vascular structures, root sleeves, fascial barriers, and pressure relationships (3, 4). Although these studies are not species-specific evidence for dogs or cats, they provide an anatomical basis for interpreting visible epidural solution at a level of interest as evidence of reach rather than a direct measurement of neural exposure or neural block.
Functional blockade additionally requires sufficient local anesthetic exposure at the relevant neural structures, in both concentration and duration. At the cranial margin of a distribution pattern, the mass, concentration, and duration of drug exposure may be lower than at segments closer to the injection site. Contrast or dye at a vertebral level or intervertebral foramen therefore cannot be assumed to demonstrate effective nerve-root blockade. Meningeal diffusion, dilution within cerebrospinal-fluid sleeves, and subsequent access to neural tissue may also influence block, but the quantitative relationship between these processes and imaging appearance remains incompletely defined in dogs and cats.
Selected studies show that imaging and tested functional responses can diverge even within the same experiment (Table 1). Freire et al. found cervical staining in some dogs at 0.4 ml/kg while the mean dorsal cutaneous block increased further at 0.6 ml/kg; the dye endpoint approached its cranial ceiling, and the tested response was cutaneous rather than abdominal visceral (5). Aso et al. paired CT distribution with electrically evoked autonomic responses and found discordance between imaging and response laterality; the tested sites were nonvisceral (6).
Apparent distribution also depends on model and measurement. Cima et al. observed more extensive spread in living dogs than in thawed cadavers (7). Freitag et al. found farther mean CT spread with OCL-based than BW-based dosing, but the study did not establish a functional advantage or lower between-animal variability (8); in a related cadaver experiment, later dissection showed more cranial methylene-blue distribution than earlier CT showed for iopamidol (9).
Together, these studies show that visible spread varies with the experimental model and assessment method and may not match the tested functional response. Details of study design and unresolved questions are summarized in Table 1.
[内容请查看原文]
Clinical benefit does not establish the extent of cranial blockade
Clinical findings at 0.3 ml/kg are heterogeneous (Table 1). In dos-Santos et al., sacrococcygeal bupivacaine was associated with low intraoperative rescue use and postoperative pain classifications similar to quadratus lumborum block within a multimodal protocol (10). In Zahra et al., the same nominal volume produced only modest intraoperative sparing and no postoperative analgesic benefit (11). Thus, 0.3 ml/kg does not define a uniform clinical profile or reveal the anatomical extent of blockade.
Wong et al. show how limited the anatomical inference from these clinical findings remains. Bilateral T10-T11 reach occurred in 3 of 7 cats at 0.75 ml/kg and in 6 of 7 by 1.0 ml/kg, with 1 cat still short of the target (12). Differences in injectate, administration sequence, timing, and endpoint preclude a direct dose-response comparison. Nevertheless, the 2.5- to 3.3-fold volume discrepancy makes it difficult to assume that favorable outcomes at 0.3 ml/kg usually reflected consistent, complete bilateral coverage. These studies do not show that 0.3 ml/kg failed in any individual cat. Favorable outcomes could still reflect complete coverage in some cats, partial but useful blockade, or substantial effects of the other anesthetic and analgesic interventions.
[内容请查看原文]
Candidate caudal dose-volume regimens
[内容请查看原文]
Dogs: prospective evaluation of 0.6 ml/kg
Canine data provide an empirical anchor for selecting a candidate regimen for prospective testing. In the Freire et al. experiment, mean numbers of dorsally tested dermatomes blocked at 15–20 min were 5.0, 14.2, 20.2, and 21.0 for the 0.2, 0.4, 0.6, and 0.8 ml/kg regimens, respectively; T1 was scored as 20 and the cervical region as 21 (5). Dye reached the cervical vertebral region in all dogs at 0.6 ml/kg. These findings support prospective evaluation of a 0.6 ml/kg dose-volume regimen for extensive cranial coverage in dogs.
The experiment does not identify a minimum effective volume, an optimal regimen, a routinely acceptable clinical dose, or a volume-specific causal threshold. Because concentration was fixed at 0.25%, total bupivacaine dose increased with volume. The catheter was advanced 5 cm cranially and was found ventral in all dogs, the dogs were kept standing, the cranial response scale had a ceiling, and the functional endpoint was dorsal cutaneous nociception rather than cranial abdominal somatic or visceral input. The 0.6 ml/kg regimen is therefore an evidence-anchored candidate for direct testing.
[内容请查看原文]
Cats: prospective comparison of 1.0 and 0.75 ml/kg
Wong et al. identified 0.75 ml/kg as the first tested cumulative volume at which bilateral T10-T11 reach occurred, but this endpoint was achieved in only 3 of 7 cats. By 1.0 ml/kg, bilateral reach had occurred in 6 cats, although it remained incomplete in 1 (12). For a study primarily designed to test target-specific adequacy, 1.0 ml/kg therefore represents the more defensible primary candidate total volume, whereas 0.75 ml/kg is a useful lower comparator for examining the trade-off between target reach and excessive cranial distribution. Neither value defines a minimum effective volume or guarantees uniform coverage. At 0.75 ml/kg, iohexol reached T4-T5 at least unilaterally in 1 cat; after 1.0 ml/kg, it reached that level in 3 additional cats (12). The sequential iohexol protocol also cannot establish that a single injection of local anesthetic at either volume would produce the same anatomical or functional result.
A later randomized clinical trial used a 0.75 ml/kg total sacrococcygeal volume containing preservative-free morphine 0.1 mg/kg and ropivacaine 0.93 mg/kg after ovariohysterectomy and found lower early pain scores and less rescue robenacoxib use than in controls (13). Because the injection was administered after surgery and combined an epidural opioid with a local anesthetic, the trial evaluated postoperative analgesic performance rather than intraoperative local-anesthetic-specific cranial coverage. It therefore does not resolve whether 0.75 ml/kg is adequate for the target defined by Wong et al.
[内容请查看原文]
Concurrent evaluation of physiological tolerance
Coverage and physiological tolerance need concurrent evaluation. The cranial distribution observed by Wong et al. identifies a potential for extension beyond the ovarian target, but it does not establish a hemodynamic or respiratory effect (12). In six isoflurane-anesthetized dogs, 0.4 ml/kg of 0.25% lumbosacral bupivacaine produced greater mean arterial pressure depression, more hypotension, and transient hypoventilation than 0.2 ml/kg (14). In a retrospective cohort of 210 dogs undergoing abdominal surgery with lumbosacral levobupivacaine-opioid regimens, occipito-coccygeal-length-based calculation produced a greater mean volume and was associated with more post-epidural hypotension than body-weight calculation (15). The association does not isolate volume causally because the cohorts were nonconcurrent and multiple perioperative factors differed. Severe hypoventilation and cardiovascular depression after 0.6 ml/kg bupivacaine in two dogs provide a case-level safety signal rather than an incidence estimate (16). Freire et al. also observed prostration and respiratory depression in many animals receiving larger regimens, without systematic quantification (5).
Prospective evaluation of an adequacy-oriented regimen should prespecify monitoring, ventilation criteria, hemodynamic rescue thresholds, and stopping rules. A feline comparison of 1.0 and 0.75 ml/kg would be most informative if it measured bilateral target reach, distribution into the cranial thoracic region, procedure-relevant functional attenuation, ventilation, hemodynamics, and motor effects under the same local-anesthetic regimen. Physiological acceptability will also depend on patient respiratory reserve, airway control, anesthetic depth, monitoring, and immediate access to ventilatory and cardiovascular support.
[内容请查看原文]
Discussion
[内容请查看原文]
Matching claims to measured endpoints
Four questions can guide interpretation of a study or clinical report. First, what anatomical and nociceptive target is being claimed? Ventral abdominal wall incision, parietal peritoneal manipulation, ovarian traction, and cranial visceral traction do not necessarily represent the same field. Second, which evidentiary domain was measured: anatomical reach, stimulus-specific functional attenuation, physiological tolerance, or composite perioperative performance? Third, which mechanisms remain compatible with the observed outcome? Fourth, if evidence from different domains is being combined, are species, access route, injectate regimen, target, timing, stimulus, and response criterion sufficiently aligned?
Two questions are particularly useful when anatomical reach and functional adequacy are discussed together: Did the injectate reach the predefined bilateral anatomical targets? Did the local anesthetic produce reproducible attenuation of the predefined procedure-relevant stimuli for sufficient duration? Together, these questions convert a general coverage claim into two independently testable endpoints.
[内容请查看原文]
Linking anatomical, functional, physiological, and clinical evidence
Future work can proceed in stages rather than measuring all four outcomes in one trial. An anatomical study can first define bilateral reach and variability. A living-animal study can then test the same or a closely matched regimen against a predefined functional stimulus. At the functional stage, procedure-relevant noxious events can be prespecified while appropriate background anesthesia and rescue protocols remain available. Separate studies can assess physiological effects and clinical benefit. Table 1 shows that existing studies cover individual stages, but no linked series of methodologically comparable studies has yet validated cranial abdominal coverage. Any differences between stages should be stated explicitly.
[内容请查看原文]
Selecting comparators for candidate regimens
Candidate comparisons should follow the question being tested. For dogs, the comparator for 0.6 ml/kg should be prespecified according to whether the study is intended to compare the candidate with a regimen reflecting current caudal practice, an adjacent lower dose-volume regimen, or a dose-matched alternative. In cats, 1.0 ml/kg can serve as the primary adequacy-oriented candidate and 0.75 ml/kg as the lower comparator.
[内容请查看原文]
Distinguishing dose-calculation models from volume-effect studies
Comparisons of BW- and OCL-based calculations ask whether one patient-size descriptor predicts achieved cranial distribution more consistently across animals. The potential advantage of OCL would be lower animal-to-animal variation, not simply a larger calculated volume or farther mean spread. BW and OCL formulas can yield different volumes (2), and OCL-based dosing increased mean CT spread in the six-dog crossover study; however, the reported SD was 2 vertebrae under both methods (8). Future studies should calibrate each formula to the same anatomical target and compare the remaining variation. If OCL is more consistent, a target-specific coefficient can then be derived and tested for functional blockade and physiological tolerance.
This question is distinct from whether volume itself changes epidural spread or blockade. To isolate a volume effect, the study must state which factor is held constant. Dose-matched regimens can compare different volumes and concentrations at the same total local-anesthetic dose; complementary comparisons at a fixed volume or fixed concentration can help distinguish effects of volume, concentration, total dose, or their combined effects.
[内容请查看原文]
Limitations
Because the literature was selected purposively rather than systematically, relevant reports may have been missed. Among the studies reviewed, none followed the same or a closely matched local-anesthetic regimen from bilateral target mapping through functional testing, physiological assessment, and clinical outcomes. Procedure-specific somatic and visceral targets also remain incompletely mapped for several abdominal operations in dogs and cats.
[内容请查看原文]
Conclusion
In dogs, 0.6 ml/kg is the proposed candidate for prospective evaluation; in cats, the proposed comparison is 1.0 vs. 0.75 ml/kg. In both species, bilateral target reach, procedure-relevant functional attenuation, physiological tolerance, and overall clinical performance should be assessed separately. These volumes are research candidates, not clinical dosing recommendations.