Erector spinae plane block in dogs undergoing hemilaminectomy: A prospective randomized clinical trial.

OBJECTIVE: To compare the perioperative cumulative opioid consumption and the incidence of cardiovascular complications in dogs undergoing hemilaminectomy in which either an erector spinae plane (ESP) block or systemic opioids were administered. STUDY DESIGN: Prospective randomized clinical trial. ANIMALS: A total of 60 client-owned dogs. METHODS: Dogs were randomized to one of three groups: an ESP block (group ESP), a constant rate infusion of fentanyl (group FNT, positive control) or a single dose of methadone as premedication (group MTD, negative control). Intraoperative nociceptive response was treated with fentanyl [1 µg kg-1, intravenously (IV)] boli. Before closure of the surgical site, morphine (0.1 mg kg-1) was applied to the dura mater. The cumulative dose of opioids was recorded and compared between groups. The incidence of intraoperative bradycardia and/or hypotension and the time to extubation were compared between groups. The short form of the Glasgow Composite Pain Scale (SF-GCPS) was used to score nociception before anaesthetic induction and 1, 2, 6, 12,18 and 24 hours postoperatively. Methadone 0.2 mg kg-1 was administered IV if the SF-GCPS score was ≥ 5. RESULTS: Group MTD required more intraoperative rescue analgesia than groups ESP (p = 0.008) and FNT (p = 0.001). The total cumulative intraoperative dose of fentanyl was higher in groups FNT (p < 0.0001) and MTD (p = 0.002) than in group ESP. The incidence of cardiovascular complications was similar between groups. Extubation time was longer in group MTD (p = 0.03). Postoperatively, the time to first rescue analgesia was longer in group ESP than in group MTD (p = 0.03). The cumulative postoperative opioid consumption and pain scores were similar between groups. CONCLUSIONS AND CLINICAL RELEVANCE: The ESP block resulted in a reduced intraoperative opioid consumption compared with the control positive and negative groups.

Ultrasound-guided and 'blind' sciatic nerve injection techniques: Comparison in rat cadavers.

OBJECTIVE: To compare the success rate and extent of sciatic nerve staining with a bupivacaine-dye solution using two injection techniques: 'blind' or ultrasound-guided approach. STUDY DESIGN: Prospective, experimental, randomized, cadaveric study. ANIMALS: Adult female Wistar rat cadavers [n = 24, mass 352 g (323-374)]. METHODS: Each sciatic nerve was randomly allocated to one of two groups: 'blind' (group B) or ultrasound-guided approach (group US) to injection. Following injection of bupivacaine-dye solution (0.1 mL), gross anatomical dissection was performed to visualize nerve staining, categorizing it as either positive or negative. The length of nerve staining was then measured and visual inspection conducted to identify potential nerve damage. Fisher's exact test was used to compare positive or negative nerve staining, and the Wilcoxon signed rank test used to compare the length of nerve staining between groups. RESULTS: In group B, the bupivacaine-dye solution stained 16/24 sciatic nerves (67% success). In group US, staining was successfully observed in all 24 nerves (100% success, p < 0.004). The length of nerve staining [median (interquartile range)] was 2 (2-3) mm in group B and 5 (4-6) mm in group US (p < 0.001). One sciatic nerve in group B had injectate distributed over 16 mm, suggestive of an intraneural injection. No signs of laceration or nerve damage were visible under 6× magnification in either group. CONCLUSIONS AND CLINICAL RELEVANCE: The ultrasound-guided approach for sciatic nerve injection demonstrated a higher success rate with superior injectate distribution when compared with the 'blind' approach. Ultrasound guidance is recommended over a 'blind' approach for sciatic nerve block in rats when possible.

Ultrasound-guided sciatic nerve block in Wistar rat cadavers: Relevant anatomy and injection distribution.

OBJECTIVE: To investigate the feasibility of an ultrasound-guided sciatic nerve block by describing the sonoanatomy and comparing the distribution of two volumes of bupivacaine dye solution for nerve staining. STUDY DESIGN: Randomized, experimental, assessor-blinded cadaveric study. ANIMALS: A total of 40 adult female Wistar rat cadavers. METHODS: After studying the sonoanatomy of the sciatic nerve and adjacent structures using a high-resolution linear transducer (19-5 MHz), rat cadavers were randomly divided into two groups that were administered either 0.1 mL (group 0.1) or 0.2 mL (group 0.2) of bupivacaine dye solution per nerve, delivered via an in-plane technique. The extent of nerve staining was subsequently evaluated following dissection. Statistical analysis consisted of assessing data distribution using the Shapiro-Wilk test, followed by paired t-tests for continuous data, Mann-Whitney U test and McNemar's test for categorical data. Statistical significance was defined as p < 0.05. RESULTS: The sciatic nerve was identified bilaterally as a double ellipsoid-shaped image, surrounded by a hyperechoic fascia separating the biceps femoris from the adductor muscle. The hypoechoic structure formed by the bupivacaine dye solution around the nerve was effectively visualized using ultrasound imaging. Sciatic nerve staining was successfully achieved in all pelvic limbs, with dye spread of 4.82 ± 1.55 mm and 5.47 ± 2.18 mm in groups 0.1 and 0.2, respectively (p = 0.128). CONCLUSIONS AND CLINICAL RELEVANCE: This study achieved a detailed understanding of the sonoanatomy of the sciatic nerve and its adjacent structures, highlighting the feasibility of the ultrasound-guided technique for injection in Wistar rats. Furthermore, the results show a comparable distribution of dye solution in both groups. Use of the ultrasound-guided sciatic nerve block technique in rats not only exhibits substantial potential for regional anesthesia but also opens avenues for translational studies.

Ultrasound-guided caudal quadratus lumborum block combined with the greater ischiatic notch plane block as motor-protective analgesia for the pelvic limb in dogs.

OBJECTIVE: To develop an ultrasound-guided caudal quadratus lumborum block (C-QLB) technique in canine cadavers and to compare sensory and motor blockade resulting from the combination of ultrasound-guided greater ischiatic notch (GIN) plane and C-QLB approaches (GIN-CQLB group) versus a lumbosacral plexus (LSP group) approach [combination of lateral pre-iliac (LPI) and parasacral (PS) techniques] in dogs. STUDY DESIGN: Descriptive anatomical study and prospective randomized, blinded, experimental crossover trial. ANIMALS: A total of six canine cadavers and six adult Beagle dogs. METHODS: Phase I: following ultrasound-guided C-QLB injections of 0.3 mL kg-1 of dye, using the interfascial plane located lateral to the quadratus lumborum muscle at the level of the sixth lumbar vertebra (L6) as injection point, the spread of injectate and nerve staining was evaluated using gross anatomical dissection. PHASE II: sensory and motor blockade achieved with the GIN-CQLB or LSP blocks in Beagle dogs were evaluated and compared. The assigned technique was performed with 2% lidocaine: 0.2 mL kg-1 for the GIN and PS approaches and 0.3 mL kg-1 for the C-QLB and LPI approaches. RESULTS: Dissection revealed distribution of dye around the lumbar hypaxial musculature, extending into the paravertebral spaces, with staining of 3 (2-4) [median (interquartile range)] spinal nerves, spanning L3 to L6. The median motor blockade in the GIN-CQLB and LSP groups was 7 (7-8) versus 16 (10-16) (p = 0.026), whereas the median sensory blockade was 5 (4-5) versus 3 (3-3) (p = 0.025), respectively. CONCLUSION AND CLINICAL SIGNIFICANCE: The GIN-CQLB approach desensitized the thigh dermatomes effectively. Compared with the LSP approaches, GIN-CQLB exhibits a motor-protective effect by preserving tonic muscle function.

Haemodynamic effects of labetalol in isoflurane-anaesthetized dogs that received dexmedetomidine: A randomized clinical trial.

OBJECTIVE: To test whether labetalol improved cardiovascular function in anaesthetized dogs injected with dexmedetomidine. STUDY DESIGN: Prospective, randomized, blinded, clinical trial. ANIMALS: A group of 20 healthy client-owned dogs undergoing ovariohysterectomy. METHODS: Each dog received dexmedetomidine (5 µg kg-1) and methadone (0.2 mg kg-1) intramuscularly. General anaesthesia was induced with propofol and maintained with isoflurane in oxygen. All dogs were mechanically ventilated, and epidural anaesthesia with lidocaine was performed. Standard anaesthetic monitoring, invasive blood pressure, oesophageal Doppler and near-infrared tissue perfusion/oxygenation were applied. Peak velocity (PV), mean acceleration and stroke distance (SD) from the oesophageal Doppler were recorded. Arterial elastance (Ea) was calculated. Tissue oxygenation (rStO2) was also recorded. Prior to surgery, animals received either 0.1 mg kg-1 of labetalol intravenously (IV) over 60 seconds or the equivalent volume of saline. Data were recorded for 20 minutes. Age, weight and propofol dose were compared with a Wilcoxon rank-sum test. The effects of time, treatment and their interaction with haemodynamic and perfusion variables were analysed with mixed-effect models and Tukey's post hoc tests. RESULTS: Significant effects of the interaction between treatment and time were observed whereby heart rate (HR) was higher in dogs given labetalol (p = 0.01), whereas arterial blood pressure and Ea were lower (p < 0.01). Similarly, PV, SD and rStO2 were higher in the labetalol group, and significant effects were detected for the interaction between treatment and time (p < 0.01). CONCLUSIONS AND CLINICAL RELEVANCE: Labetalol at a dose of 0.1 mg kg-1 IV in dogs under general anaesthesia and administered a pre-anaesthetic medication of dexmedetomidine produced mild vasodilation (reduction of Ea), resulting in an increase in HR and left ventricular outflow. Although labetalol could be an effective option to achieve haemodynamic optimization after dexmedetomidine-induced vasoconstriction, future studies are needed to assess long-term effects.

Ultrasound-guided lateral pericapsular hip desensitization of the articular branches of the cranial gluteal nerve: A canine cadaveric study and feasibility study in dogs.

OBJECTIVE: To develop and assess the feasibility, as a diagnostic block, of an ultrasound-guided lateral pericapsular hip desensitization (L-PHD) technique in dogs. STUDY DESIGN: Prospective, randomized, anatomical and feasibility study. ANIMALS: A total of 11 canine cadavers and eight adult dogs scheduled for acetabular surgical denervation. METHODS: After studying the ultrasound anatomy of the lateral aspect of the gluteal region and determining an acoustic window to perform an ultrasound-guided L-PHD in three canine cadavers, the right and left hemipelves of eight canine cadavers were injected in the interfascial plane located lateral (LL-PHD group) or medial (LM-PHD group) to the deep gluteal muscle, with 0.05 mL kg-1 of dye per hip on each cadaver. The staining of the pericapsular nerves was assessed by anatomical dissection. Then, the LM-PHD was performed using 2% lidocaine as a diagnostic block in dogs scheduled for acetabular surgical denervation. Positive predictive value (PPV) was calculated for those animals who had favorable outcomes after acetabular surgical denervation. RESULTS: The ultrasound-guided LL-PHD and LM-PHD could be performed by inserting the needle lateral and medial to the deep gluteal muscle. Ultrasound-guided LL-PHD stained the cranial gluteal nerve and its muscular branches in all injections and partially stained the lumbosacral trunk in two out of eight cadavers. The LM-PHD selectively stained the articular branches of the cranial gluteal nerve in all but one cadaver. The PPV for LM-PHD successful test prediction was 85.7% (95% confidence interval: 48.6% to 98.6%). CONCLUSIONS: and clinical significance Ultrasound-guided LM-PHD using 0.05 mL kg-1 of dye selectively stained the articular branches of the cranial gluteal nerve in canine cadavers. The LM-PHD technique is feasible and could be used as a diagnostic block before acetabular surgical denervation in dogs.

A non-inferiority study comparing the ultrasound-guided parasacral with a novel greater ischiatic notch plane approach in canine cadavers.

OBJECTIVE: To describe the gross and ultrasound anatomy of the parasacral region and an ultrasound-guided greater ischiatic notch (GIN) plane approach aimed at staining the lumbosacral trunk (LST) in canine cadavers. To evaluate if the ultrasound-guided GIN plane approach is non-inferior to the previously described ultrasound-guided parasacral approach at staining the LST. STUDY DESIGN: Prospective, randomized, non-inferiority experimental anatomic study. ANIMALS: A total of 17 (23.9 ± 5.2 kg) mesocephalic canine cadavers. METHODS: Anatomic and echographic landmarks, and the feasibility of performing a GIN plane technique were evaluated using two canine cadavers. The remaining 15 cadavers had each hemipelvis randomly assigned to be administered either parasacral or GIN plane injection of 0.15 mL kg-1 dye solution. The parasacral region was dissected after injections to assess the staining of LST, cranial gluteal nerve, pararectal fossa and pelvic cavity. The stained LST were removed and processed for histological evaluation of intraneural injections. A one-sided z-test for non-inferiority (non-inferiority margin -14%) was used to statistically evaluate the success of the GIN plane versus the parasacral approach. Data were considered statistically significant when p < 0.05. RESULTS: The GIN plane and parasacral approach stained the LST in 100% and 93.3% of the injections, respectively. The success rate difference between treatments was 6.7% [95% confidence interval, -0.6 to 19.0%; p < 0.001 for non-inferiority]. The GIN plane and parasacral injections stained the LST for 32.7 ± 16.8 mm and 43.1 ± 24.3 mm, respectively (p = 0.18). No evidence of intraneural injection was found. CONCLUSIONS AND CLINICAL RELEVANCE: The ultrasound-guided GIN plane technique resulted in nerve staining that was non-inferior to the parasacral technique and may be considered an alternative to the parasacral approach to block the LST in dogs.

Effect of end-inspiratory pause on airway and physiological dead space in anesthetized horses.

OBJECTIVE: To evaluate the impact of a 30% end-inspiratory pause (EIP) on alveolar tidal volume (VTalv), airway (VDaw) and physiological (VDphys) dead spaces in mechanically ventilated horses using volumetric capnography, and to evaluate the effect of EIP on carbon dioxide (CO2) elimination per breath (Vco2br-1), PaCO2, and the ratio of PaO2-to-fractional inspired oxygen (PaO2:FiO2). STUDY DESIGN: Prospective research study. ANIMALS: A group of eight healthy research horses undergoing laparotomy. METHODS: Anesthetized horses were mechanically ventilated as follows: 6 breaths minute-1, tidal volume (VT) 13 mL kg-1, inspiratory-to-expiratory time ratio 1:2, positive end-expiratory pressure 5 cmH2O and EIP 0%. Vco2br-1 and expired tidal volume (VTE) of 10 consecutive breaths were recorded 30 minutes after induction, after adding 30% EIP and upon EIP removal to construct volumetric capnograms. A stabilization period of 15 minutes was allowed between phases. Data were analyzed using a mixed-effect linear model. Significance was set at p < 0.05. RESULTS: The EIP decreased VDaw from 6.6 (6.1-6.7) to 5.5 (5.3-6.1) mL kg-1 (p < 0.001) and increased VTalv from 7.7 ± 0.7 to 8.6 ± 0.6 mL kg-1 (p = 0.002) without changing the VTE. The VDphys to VTE ratio decreased from 51.0% to 45.5% (p < 0.001) with EIP. The EIP also increased PaO2:FiO2 from 393.3 ± 160.7 to 450.5 ± 182.5 mmHg (52.5 ± 21.4 to 60.0 ± 24.3 kPa; p < 0.001) and Vco2br-1 from 0.49 (0.45-0.50) to 0.59 (0.45-0.61) mL kg-1 (p = 0.008) without reducing PaCO2. CONCLUSIONS AND CLINICAL RELEVANCE: The EIP improved oxygenation and reduced VDaw and VDphys, without reductions in PaCO2. Future studies should evaluate the impact of different EIP in healthy and pathological equine populations under anesthesia.

A single-cohort retrospective analysis of factors associated with morbidity and mortality in 193 anesthetized domestic goats.

OBJECTIVE: To define the morbidity and mortality rates in goats undergoing general anesthesia at a large animal teaching hospital. STUDY DESIGN: Retrospective, single-cohort, observational study. ANIMALS: Records of 193 client-owned goats. METHODS: Data were collected from 218 medical records on 193 goats undergoing general anesthesia between January 2017 and December 2021. Demographic data, anesthetic management, recovery period and perianesthetic complications were recorded. Perianesthetic death was defined as anesthesia-related or anesthesia-contributory death occurring within 72 hours after recovery. Records of goats that were euthanized were reviewed to ascertain the cause of euthanasia. Each explanatory variable was individually investigated by univariable penalized maximum likelihood logistic regression, followed by multivariable analysis. Statistical significance was set at p < 0.05. RESULTS: Perianesthetic mortality was 7.3%, but was 3.4% when considering only goats undergoing elective procedures. Multivariable analysis showed that gastrointestinal surgeries [odds ratio (OR) 19.17, standard error (SE) 12.99, 95% confidence interval (CI) 5.08-72.33; p < 0.01] and requirement for perianesthetic norepinephrine infusion (OR 10.85, SE 8.82, 95% CI 2.21-53.33; p < 0.01) were associated with increased mortality. Maintaining other variables equal, the use of perianesthetic ketamine infusion was associated with decreased mortality (OR 0.09, SE 0.09, 95% CI 0.01-0.73; p = 0.02). Anesthesia-related or anesthesia-contributory complications included hypothermia (52.4%), bradycardia (38.1%), hypotension (35.3%), hypoxemia (14.8%), regurgitation/aspiration (7.3%), azotemia/acute renal failure (4.6%), myopathies/neuropathies (4.1%) and fever of unknown origin (2.7%). CONCLUSIONS AND CLINICAL RELEVANCE: In this population, gastrointestinal surgeries and the requirement for perianesthetic norepinephrine infusion were associated with increased mortality in goats undergoing general anesthesia, while ketamine infusion may have a protective effect.

Lumbosacral plexus block using a combination of ultrasound-guided lateral pre-iliac and parasacral approaches in cats.

OBJECTIVE: To describe an ultrasound-guided lateral pre-iliac (LPI) and parasacral (PS) approach in feline cadavers (phase I) and compare the perioperative analgesic use and complications in cats administered LPI and PS blocks (group PNB) or epidural anesthesia (group EPI) for pelvic limb surgery (phase II). STUDY DESIGN: Experimental uncontrolled, anatomic and retrospective cohort study. ANIMALS: A group of eight feline cadavers and 52 medical records. METHODS: Bilateral LPI and PS approaches with 0.1 mL kg-1 of dye to stain the femoral and obturator nerves and the lumbosacral trunk, respectively, were performed on each cadaver. Nerve staining effect was evaluated upon dissections (phase I). Perioperative analgesics use, and complication rates were retrospectively compared between groups PNB and EPI (phase II). Continuous data were compared using the Mann-Whitney U test and the prevalence of events with Fisher's exact test. Differences were considered significant when p < 0.05. RESULTS: Dissections revealed that the LPI approach stained 94% and 75% of the femoral and obturator nerves, respectively. The PS approach stained 100% of the lumbosacral trunks. Cats enrolled in group PNB (n = 23) were administered lower doses of intraoperative opioids than those in group EPI (n = 25) (p = 0.006). Intraoperative rescue analgesia was required in 60% and 17.4% of cats enrolled in groups EPI and PNB, respectively (p = 0.003). Group PNB required more intraoperative anticholinergics than group EPI (p = 0.02). There were no differences in postoperative pain scores, analgesic use and complication rates. CONCLUSIONS AND CLINICAL RELEVANCE: The ultrasound-guided LPI and PS approach stained the femoral/obturator nerves and the lumbosacral trunk, respectively, in feline cadavers. Furthermore, PNB was associated with lower intraoperative opioid use and similar postoperative pain and analgesic use compared with epidural anesthesia in a cohort of cats undergoing surgery of the pelvic limb.

Sedated cesarean sections are associated with increased kid survival compared to general anesthesia in goats: retrospective cohort of 45 cases (2011-2021).

OBJECTIVE: To determine whether anesthesia type (sedation or general anesthesia) affects kid survival to discharge in caprine cesarean sections (C-sections). ANIMALS: Retrospective cohort of 99 caprine C-sections (2011-2021). PROCEDURES: All surgeries were performed via left flank laparotomy in right lateral recumbency. The number of kids alive at presentation, surgery, and discharge was recorded. Kids that were dead on presentation or euthanized intraoperatively were excluded. Goats were classified as "healthy" (American Society of Anesthesiologists status ≤ 2) or "sick" (≥ 3). RESULTS: Kid survival was significantly higher for C-sections performed under sedation (47/52 [90%]) than for C-sections performed under general anesthesia (16/24 [66%]; P = .004). Relative risk was 1.4 and odds ratio was 4.7. CLINICAL RELEVANCE: Performing C-sections in sedated goats may improve kid survival rates over those under general anesthesia.

Endoscopic application of fibrin glue may be a feasible method of treatment for postintubation tracheal lacerations in cats.

OBJECTIVE: To determine the feasibility of endoscopic application of fibrin glue for the treatment of experimentally induced postintubation tracheal laceration (PITL) in feline cadavers. The secondary objective was to determine the optimal technique for application of the fibrin glue. ANIMALS: 20 feline cadavers (n = 10 fresh and 10 frozen). PROCEDURES: An experimentally induced tracheal rupture was created via overinflation of an endotracheal tube cuff. After endoscopic identification of the tracheal tear, fibrin glue was instilled into the tracheal defect in either a bridging or filling fashion. Following the procedure, the airway of each cat was examined and leak tested. Length of tear, volume of glue applied, procedural time, and glue efficacy were recorded. RESULTS: Experimentally induced tracheal lacerations were full thickness with a mean length of 3.27 ± 0.96 cm. A complete seal was attained in 6 of the 9 fresh cadavers when filling the defect with fibrin glue. In the remaining 3 fresh cadavers, air leakage was restricted to the dorsal mediastinum. Bridging the defect with fibrin glue did not attain a seal in fresh or frozen cadavers. The median volume of glue used to fill defects in fresh cadavers was 0.5 mL (range, 0.4 to 2 mL). Procedural time for the application of fibrin glue was 10.5 ± 4.1 minutes for bridging the defect and 7.8 ± 1.5 minutes for filling the defect. CLINICAL RELEVANCE: Endoscopic application of fibrin glue may be a feasible method of treatment for PITL in cats.

Pericapsular hip desensitization in dogs: a cadaveric study and case series.

OBJECTIVE: To develop and assess the efficacy of an ultrasound (US)-guided pericapsular hip desensitization (PHD) technique in dogs. STUDY DESIGN: Prospective, randomized, anatomical study and a case series. ANIMALS: A total of 30 healthy dogs, eight canine cadavers and seven dogs with hip osteoarthritis. METHODS: After studying the US anatomy of the medial aspect of the coxofemoral joint and determining an acoustic window to perform an US-guided PHD in healthy dogs, the US-guided PHD was performed bilaterally in canine cadavers. A low [(LV) 0.1 mL kg-1] and high [(HV) 0.2 mL kg-1] volume of dye was injected per hip on each cadaver. The staining of the pericapsular nerves was assessed by anatomical dissection, and comparison between LV and HV was assessed using Fisher's exact test. Then, the US-guided PHD was performed using a triamcinolone-bupivacaine solution in dogs with hip osteoarthritis. Dynamic pain response was assessed before and after injection. The canine brief pain inventory (CBPI) questionnaire was used to assess treatment efficacy and duration. RESULTS: The US-guided PHD could be performed by inserting the needle between the iliopsoas muscle and the periosteum of the ilium. The articular branches of the femoral and obturator nerves were stained in all cadavers using both volumes. The main femoral nerve was never stained, but the main obturator nerve was stained in 37.5% and 100% of injections using LV and HV, respectively (p = 0.026). Treated animals showed decreased dynamic pain response after the injection. Compared with baseline, CBPI scores were reduced by ≥ 50% for ≥ 12 weeks in all but one dog. CONCLUSIONS AND CLINICAL SIGNIFICANCE: The US-guided PHD with both 0.1 and 0.2 mL kg-1 volumes stained the articular branches of the femoral and obturator nerves in canine cadavers and was associated with clinical improvement in dogs with hip osteoarthritis.

Erector spinae plane block at the thoracolumbar spine: a canine cadaveric study.

OBJECTIVE: To investigate the injectate spread and nerve staining of ultrasound-guided erector spinae plane (ESP) injections at the thoracolumbar spine in canine cadavers. STUDY DESIGN: Prospective, randomized, descriptive, anatomic study. ANIMALS: A total of 15 canine cadavers. METHODS: The location of the medial and lateral branches of the dorsal branches of the spinal nerves (DBSN) from the tenth thoracic (T10) to the third lumbar vertebra (L3) were identified by dissection of three cadavers. ESP injections of dye (0.5 mL kg-1) were performed in seven cadavers using as landmarks the T12 transverse process (ESPTp) on one side and the lateral aspect of the T12 mammillary process (ESPMp) on the opposite side. Additionally, five cadavers were injected with dye (0.5 mL kg-1) bilaterally on the lateral aspect of the L2 mammillary process (ESPMp_L2). Nerve staining effect was analyzed after gross anatomic dissections. The number of stained nerves was analyzed using the Mann-Whitney U test. RESULTS: Gross anatomic dissections showed that the medial and lateral branches of the DBSN change their path in relation to the epaxial muscles caudal to T11. Approaches ESPTp and ESPMp at T12 stained 2 (0-2) and 3 (2-4) medial (p = 0.01) and 3 (3-4) and 2 (0-2) lateral (p = 0.03) branches, respectively. Injection ESPMp_L2 stained 3 (2-4) medial and 2 (0-3) lateral branches. Injections ESPMp and ESPMp_L2 produced a preferential cranial spread from the injection site. No ventral branches of the spinal nerves were stained with either technique. CONCLUSIONS AND CLINICAL RELEVANCE: These results suggest that the mammillary process should be used as anatomic landmark to perform ultrasound-guided ESP blocks in the thoracolumbar spine caudal to T11 when targeting the medial branches of the DBSN. Injections should be performed one spinal segment caudal to the level intended to desensitize.