Spine surgery in the thoracolumbar region is one of the most common operations performed in the United States to address back and leg pain.1 Such patients frequently present with chronic preoperative pain, but they also experience surgery-related new onset acute pain in the early postoperative period. A variety of regional anesthesia techniques have been shown to be effective in managing acute and chronic pain during the perioperative period.2,3 As general anesthesia becomes increasingly safer, our specialty is tasked with making improvements in the other areas of perioperative care that contribute to morbidity and mortality. In addition, expanded use of regional anesthesia techniques is advocated by Enhanced Recovery After Surgery protocols aimed at minimizing opioid analgesics whenever possible.4-6

Many patients undergoing lower abdominal surgical procedures now benefit from the transversus abdominis plane (TAP) block, a field block targeting the abdominal intermuscular plane that carries sensory nerves from the ventral ramus of the thoracolumbar nerves.7,8 Several studies have shown this block to be safe and effective.9-12 Furthermore, the control of postoperative pain using non-opioid medication has repeatedly been shown to benefit patients in terms of morbidity and mortality.13 Currently, spine surgeons frequently perform a preoperative field infiltration for lumbar surgery with a total dose of anesthetic similar to that of the TAP block14; however, the efficacy of this approach has not been well studied. Nevertheless, the safety of such a procedure with the typically used dose of local anesthetic is supported by the extensive regional anesthesia practice using similar blocks.

In this report, we describe a novel regional anesthesia block that targets the dorsal rami of the thoracolumbar nerves (analogous to the ventral rami for the TAP block) as they pass through the paraspinal musculature (Fig. 1). Accordingly, this newly described block is termed the thoracolumbar interfascial plane (TLIP) block. The purpose of this pilot study is to perform an ultrasound-guided TLIP block showing a reproducible area of sensory blockade in a group of volunteers.

Fig. 1
figure 1

Illustration shows the dorsal and ventral divisions of each spinal nerve. The paraspinal muscles, including multifidus and longissimus, are displayed along with the approximate needle trajectory that was used. Labelled structures include: I = iliocostalis muscle; L = longissimus muscle; M = multifidus muscle; TP = transverse process

Full size image

Methods

Institutional Review Board approval was obtained (IRB # Pro00033609, June 2014), and ten healthy volunteers provided written informed consent to participate in this study. After placement of a peripheral intravenous catheter, each participant was placed in the prone position and standard monitors were applied. No sedation was administered. A SonoSite S-Nerve high-frequency linear (HFL) 50X transducer (SonoSite, Bothell, WA, USA) was placed in transverse orientation in a midline position at approximately the level of the third lumbar vertebra (L3). The corresponding spinous process and interspinal muscles were identified, and the probe was then moved laterally to identify the multifidus muscle (MF) and the longissimus muscles (LG). Initial midline placement of the probe with subsequent lateral displacement avoided misinterpretation of the interspinous/MF interface as the MF/LG interface. These landmarks are seen in Fig. 2. The block was performed at the most caudal level that allowed reliable identification of the MF/LG interface. In our pilot study, this approach resulted in no more than one variation in vertebral level.

Fig. 2
figure 2

Ultrasound image of the paraspinal muscles of the thoracolumbar spine. The fascial plane is easily visible between the muscle bellies (arrow). The structures include: SP = spinous process; F = facet joint; TP = transverse process; M = multifidus muscle; L = longissimus muscle

Full size image

Discrimination of the MF and LG muscles can be difficult as these separate structures can often appear as a single larger muscle. Visualization can be enhanced with lumbar extension and slight rotation of the participant. This maneuver results in “sliding” the LG over the lateral border of the MF better delineating the intended location to inject the local anesthetic. After identifying the muscles, chlorhexidine was applied to the participant’s skin in preparation for performing the subsequent block in a sterile manner.

Lidocaine 1% was used to anesthetize the skin and subcutaneous tissues. A 10-cm 21G Stimuplex® needle (Braun Medical Inc, Bethlehem, PA, USA) was inserted bevel up in a lateral-to-medial orientation at an approximate angle of 30° to the skin. The needle was advanced under real-time in-plane ultrasound guidance through the belly of the LG towards the MF. The needle tip was directed towards the LG-MF interface deep to the midpoint (Fig. 3a). After an attempted aspiration with a 3-mL syringe was negative for blood, a 20-mL syringe was used to inject a small volume of local anesthetic to confirm needle tip placement between the MF and LG (i.e., hydrodissection). The location within the desired plane was facilitated by advancing the needle into the MF and subsequently injecting the local anesthetic as the needle was withdrawn. Ropivacaine 0.2% (without epinephrine) was incrementally injected with intermittently repeated negative aspiration. Anterior spread of local anesthetic was viewed as favourable (Fig. 3b). The block was administered bilaterally to each volunteer by injecting ropivacaine 40 mL in total (i.e., 20 mL injected into each side). The site of injection was subsequently marked with a felt-tip pen. The volunteers were monitored for signs or symptoms of local anesthetic systemic toxicity or other complications, with no report of any adverse reactions.

Fig. 3
figure 3

a Ultrasound image with needle placement. This image was obtained after injection of local anesthetic (approximately 2 mL) to confirm placement. Needle orientation is lateral to medial. SP = spinous process; F = facet joint; LA = local anesthetic; N = needle (arrow immediately below needle showing trajectory); TP = transverse process. b Ultrasound image after injection of local anesthetic (20 mL) and after needle removal. This image is at the level of injection, showing hydrodissection. SP = spinous process; F = facet joint; TP = transverse process; M = multifidus muscle; LA = local anesthetic; L = longissimus muscle

Full size image

The area of the loss of posterior thoracolumbar sensation to pinprick was assessed at five and 20 min after TLIP administration. The assessment was initiated at the mid-axillary line at the level of injection. Pinprick was assessed in the posteromedial direction until loss of pain was reported, and the assessment continued to the midline to evaluate for non-continuous anesthesia. The inferior and then superior borders (starting from the level of injection) were then assessed, before assessing the contralateral side of the back. Areas with loss of discrimination were designated with a marker and a photograph of each participant was taken with a Canon PowerShot SX400 camera (Canon USA, Melville, NY, USA) after bilateral assessment with centimetre rulers placed perpendicularly (X and Y axes) for subsequent measurement and mapping. The location of the injection was marked to provide reference for the distribution of anesthesia. All injections were performed by two of the authors (E.B. and J.T.).

Statistical analysis

Descriptive statistics of the volunteer population (age, height, body mass index [BMI]) as well as creation of the contour plots were conducted in SAS® v.9.3 (SAS Institute, Cary NC, USA). The colour palette was customized by the authors for visual discrimination.

Results

All ten participants underwent complete injections without complications, adverse events, or perturbations in their vital signs. Nine of the participants were male; the mean (SD) age of the participants was 35.4 (3.3) yr ranging from 27-42 yr, and the mean (SD) height was 178 (9.7) cm ranging from 157-198 cm, which was relevant as the extent of the block was reported in centimetres rather than in dermatomes. The mean (SD) BMI in the volunteers was 25.3 (3.4) kg·m−2.

All volunteers described areas of anesthesia demarcated by areas of retained sensation. Figs. 4a and 4b are contour plots showing the number of patients who reported loss of sensation to pinprick at five and 20 min, respectively, at each X-Y coordinate, with a rendering of superimposed vertebrae for visual reference. Five minutes after injection, the participants reported anesthesia to pinprick in a mean (SD) area covering 137.4 (71.0) cm2 of their lower back. After 20 min, this mean (SD) area increased to 217.0 (84.7) cm2. The mean (SD) cephalic spread from the injection site was 6.5 (1.8) cm; caudal spread averaged 3.9 (1.2) cm. All volunteers reported anesthesia to pinprick covering the midline at the level of injection.

Fig. 4
figure 4

a Contour plot showing the number of participants endorsing anesthesia after five minutes. The scale (0-10) on the right shows the number of participants who could discriminate to pinprick. The image of the vertebral column is schematic and meant to serve only as visual reference as there may be considerable variance due to participant height. The stars designate the average lateral location of needle insertion; the arrows approximate the orientation of the needle. b A similar contour plot showing the number of participants reporting anesthesia after 20 min

Full size image

Discussion

This report describes the novel TLIP block that could potentially be an approach for providing anesthesia to the lower back. As seen in Figs. 4a and 4b, the TLIP injection achieved a reproducible area of anesthesia in the volunteer population. While the area anesthetized is not a simple paraspinal rectangle, at both five and 20 min following injection of the local anesthetic, all participants could not discriminate pain and touch at midline at the level of the injection(s). At 20 min, participants reported an average (SD) cephalic spread of anesthesia of 6.5 (1.8) cm, specifically at midline. Midline coverage is emphasized because the impetus for this investigation is a possible clinical application at a single level or minimally invasive lumbar spine surgery. Fig. 2 shows a predictable area of anesthesia following the TLIP injection.

Technical progress continues to be made in minimizing the tissue damage caused during surgery, which allows patients to be candidates for rehabilitation and to be discharged more quickly after surgery.15,16 Further optimization of care may be possible by using the novel TLIP injection to provide focused regional anesthesia to limit postoperative complications associated with pain and opioid analgesia. In future, the authors expect to study the use of TLIP as the regional anesthesia component of an enhanced surgical recovery protocol for selected patients having single-level lumbar spine surgery.13,17

The participants routinely reported cephalolateral spread of anesthesia after 20 min, as seen in Fig. 4b. This pattern may have occurred due to the orientation of the MF/LG groove. In our view, the extension of analgesia outside the level of injection further differentiates this injection from simple field infiltration.

This study was not designed to examine the duration of anesthesia, but it seems reasonable to assume that a shorter or longer duration could be achieved by choice of local anesthetic and the adjuncts injected. Systemic absorption is likely but was not measured in this study. The potential impact of systemic toxicity as well as duration of block will require further study. Furthermore, clinical studies may also be warranted to compare single injection vs infusion in terms of duration of perioperative pain control.

This is a pilot study and therefore has considerable limitations before advocating widespread use of this technique. First, all participants who volunteered for this study were younger than 50 yr, whereas back surgery is most common in patients at least 60 yr old.18 The ability to discriminate between muscle bodies may prove more difficult with aging due to relative atrophy or anatomic changes of the spine itself. Furthermore, no participant had prior surgery in the lumbar spine, so we cannot comment on the efficacy of this technique in patients who have had prior back surgery. Additionally, the ability to discriminate cutaneous pinprick does not necessarily predict loss of sensation to the underlying musculature; therefore, a clinical evaluation with patients undergoing surgery is warranted. Additional validation of this technique with cadaveric injection studies may also be informative.

The thoracolumbar interfascial plane injection consistently provided anesthesia to the midline in all participants in this proof-of-concept pilot study. Further clinical research is needed to optimize volume, concentration, and type of local anesthetic administered. Indeed, we anticipate subsequent prospective research to quantify the effect of TLIP vs surgical “field” infiltration for lumbar surgery in terms of duration of analgesia and postoperative opioid consumption. We speculate that the TLIP block may be able to offer reliable pain control for patients having single-level or minimally invasive back surgery and decrease perioperative morbidity.