Chapter 48. Paravertebral Anesthesia

1. A paravertebral nerve block involves conduction block of the spinal nerve within the paravertebral space.

2. Dense sensory, motor, and sympathetic block results.

3. Unilateral or bilateral, thoracic, or lumbar segmental block can be obtained.

4. Common indications include thoracic surgery, breast surgery, and hernia repair

5. Benefits include diminished stress response to surgery, decreased opioid consumption, reduced opioid-related side effects (nausea, vomiting, sedation), and hemodynamic stability as well as preservation of pulmonary mechanics, lower extremity strength, and bladder function.

6. Potential adverse effects are rare and include pleural puncture, pneumothorax, epidural or intrathecal injection, and local anesthetic toxicity.

Paravertebral nerve blockade (PVB) involves injection of local anesthetic close to the spinal nerve roots within the paravertebral space (PVS). The resultant unilateral or bilateral segmental anesthesia and analgesia of thoracic or lumbar dermatomes have multiple applications. In this chapter, paravertebral anatomy and PVB techniques are described. The physiologic effects are discussed in relation to the advantages, disadvantages, and contraindications for PVB. The existing literature outlining the use of PVB for a variety of surgical procedures is summarized.

Thoracic Paravertebral Anatomy

The thoracic PVS is a wedge-shaped space that lies on each side of the vertebral column. Detailed descriptions of the anatomic features of the PVS are available.1-7

Figure 48-1 illustrates the wedge-shaped boundaries of the thoracic PVS. Posteriorly, the space is limited by the superior costotransverse ligament. At each thoracic level, the superior costotransverse ligament extends from the lower border of the transverse process above to the upper border of the rib below (Fig. 48-2). Anterolaterally, the thoracic PVS is limited by the parietal pleura. The medial base of the thoracic PVS is defined by the posterolateral segment of the vertebral body, the intervertebral disk, the intervertebral foramen, and its contents.

The PVS is continuous medially with the epidural space via the intervertebral foramen; in addition, dural sleeves may extend into the PVS.1,5,8 Laterally, the thoracic PVS is continuous with the intercostal space, lateral to the transverse processes. Communication with the contralateral PVS may occur by contact through the prevertebral9,10 or epidural spaces.2 The PVS is continuous superiorly and inferiorly across the heads and necks of adjacent ribs. The precise cranial limit of the PVS has not been fully elucidated; however, cervical spread of injectate has been observed after thoracic PVB.2 It was previously thought that, caudally, the thoracic PVS was limited by the origin of the psoas major muscle11; however, continuity with the PVS below the diaphragm has been supported by studies documenting the lumbar spread of dye after thoracic injection.2,12,13

The contents of the thoracic PVS include the endothoracic fascia, the spinal nerves, the sympathetic chain, the intercostal vessels, lymphatics, and loose fatty tissue (Fig. 48-3). The endothoracic fascia is the deep, fibroelastic fascia of the thoracic cavity.14,15 It is continuous medially with the prevertebral fascia that covers the vertebral bodies and intervertebral disks,15 superiorly with the scalene fascia, and inferiorly with the fascia transversalis of the abdomen.2,14 Karmakar2 described the endothoracic fascia dividing the thoracic PVS into 2 potential fascial compartments—the anterior extrapleural paravertebral compartment and the posterior subendothoracic paravertebral compartment. The anterior compartment is thought to contain loose areolar connective tissue (subserous fascia)7 and the sympathetic trunk.2,16 The posterior compartment is thought to contain the intercostal nerves.2,16

The spinal nerves emerge from the intervertebral foramina and course through the thoracic PVS as a collection of small nerve rootlets devoid of fascial covering.2,4,17,18 Early in the course of the spinal nerve, the posterior primary ramus branches to supply the posterior vertebral muscles, ligaments, facet joints, and the overlying skin (Fig. 48-3). The sympathetic chain traverses anteriorly within the thoracic PVS2 and communicates with the spinal nerves through the preganglionic white rami communicantes and the postganglionic grey rami communicantes (Fig. 48-3). The intercostal arteries (originating from the descending aorta) as well as the hemiazygos and accessory hemiazygos veins19 also pass through the thoracic PVS. Lymphatic drainage is to local nodes and subsequently to tributaries of the thoracic duct, which form a plexiform network around the vertebral bodies.20

Several features of the thoracic spine, serving as landmarks for PVB, are important to note. The spinous processes in the thoracic spine are steeply angulated. As a result they lie in the same transverse plane as the transverse processes of the vertebra below (ie, spinous process of T5 is at the same horizontal level as the transverse process of T6) (Fig. 48-2). A spinal nerve exits through its intervertebral foramen to enter the PVS caudal to the transverse process of the same level (ie, T5 nerve root passes inferior to the T5 transverse process). In adults, the thoracic transverse processes project laterally a mean distance of 3.18 cm from midline (range 2.1-4.2 cm),8 and the mean depth from skin to thoracic PVS is 55 mm.21 Depth is greater in the upper thoracic spine (mean 77 mm at T1) compared with the mid and lower thoracic spine (mean 50 mm at T6)21; in addition, considerable variation exists as a result of body habitus (mean depth to T1 PVS 67.5 mm if body mass index [BMI] <25, 78 mm if BMI 25-30, and 84 mm if BMI >30).21

Lumbar Paravertebral Anatomy

A number of pertinent differences exist between thoracic and lumbar paravertebral anatomy. In the lumbar region, there are no ribs or superior costotransverse ligament to mark the posterior boundary of the PVS. The anterior margin of the lumbar PVS is the fascial lining of the abdomen. The endothoracic fascia above the diaphragm is continuous with the fascia transversalis, which lines the abdominal wall14 (Fig. 48-4). In the abdomen, the fascia transversalis blends medially with the anterior layer of the quadratus lumborum fascia and the psoas fascia22 (Fig. 48-5). The cephalad part of the psoas and quadratus lumborum fascias are thickened to form the medial and lateral arcuate ligaments, respectively. The medial arcuate ligament is attached medially to the body of L2 and laterally to the transverse process of L1. The lateral arcuate ligament passes from the lateral aspect of the L1 transverse process to the inferior border of the twelfth rib. The transversalis fascia is in direct communication with the endothoracic fascia at the medial and lateral arcuate ligaments as well as at the aortic hiatus. It is via these communications that there exists continuity between the thoracic and abdominal PVS.2,14,23

The lumbar nerve roots course through the lumbar PVS and subsequently combine to form the nerves of the lumbar plexus. The subcostal, iliohypogastric, ilioinguinal, and lateral femoral cutaneous nerves course anterolaterally over the quadratus lumborum muscle. The genitofemoral and, more distally, the femoral and obturator nerves pass over the anterolateral surface of the psoas muscle. Accordingly, local anesthetic injected into the PVS has the potential to block a segment of the lumbar plexus.

Several features of lumbar spinal anatomy deserve mention. The spinous process of a lumbar vertebra projects posteriorly with less angulation as compared with the thoracic spine (Fig. 48-6). As a result, a given transverse process lies at the same horizontal level as its corresponding spinous process. In addition, the transverse processes in the lumbar region do not articulate with the ribs and are much smaller and thinner in the anteroposterior plane in contrast to the thoracic spine.

Hugo Selheim, an obstetrician, is credited with the development of paravertebral anesthesia in 1905.2 Arthur Lawen, a surgical resident, subsequently refined Sellheim's technique2 and performed PVB on patients presenting with abdominal pain. This provided acute pain relief that facilitated abdominal muscle relaxation and palpation. Subsequent laparotomy or autopsy allowed identification of the underlying pathology. As a result, detailed information was collected on the segmental innervation of various thoracic and abdominal organs. These pioneers are credited with early development of the familiar segmental dermatome map (Fig. 48-7). Selheim, Lawen, and subsequently Kappis further expanded the use of PVB for surgical anesthesia in an attempt to reduce complications and improve survival in the early years of general and spinal anesthesia.2 The technique fell into disfavor for several decades until a publication by Eason and Wyatt renewed interest.3

The patient can be positioned sitting, prone, or lateral decubitus (side to be blocked uppermost). The neck is flexed, the back is rounded, and the shoulders are rounded forward, similar to positioning for a thoracic epidural. Conscious sedation is useful for anxiolysis and analgesia.

The levels (spinal nerve roots) to be blocked are selected based on the surgical procedure as summarized in Table 48-1 (ie, T1 to T6 for mastectomy and axillary dissection) (Fig 48-7). The spinal nerve root is located after identification of its corresponding transverse process. In the thoracic spine, a given transverse process is located in the same transverse plane as the spinous process of the vertebra above (ie, the T4 transverse process is beside the T3 spinous process) (Fig. 48-2). Spinous processes are generally palpable in the midline with the most prominent spinous process in the neck representing C7, the lower border of the scapula corresponding to T7, and the intercristal line marking L4. The superior aspect of the desired spinous process is identified in the midline, and from this point (in adults) a mark is drawn 2.5 cm lateral (Fig. 48-8). The skin is cleaned with disinfectant, and subcutaneous infiltration of local anesthetic is given at all needle entry sites.

A number of techniques have been described to identify the PVS, as described next.

Anatomic or Loss of Resistance

For thoracic PVB, the superior border of the appropriate spinous process is palpated and its midpoint identified. In adults, 2.5 cm lateral from this midpoint, a 10-cm, 22-gauge Tuohy needle is inserted and advanced in the parasagittal plane to contact the transverse process. Upon contact with the transverse process, the block needle is withdrawn to the subcutaneous tissue and redirected caudad to "walk-off" the transverse process (Fig. 48-9). After this caudad redirection, the needle is slowly advanced until the PVS is identified. This occurs approximately 1 cm past the depth at which the transverse process is contacted. Identification of the PVS can be made as the needle traverses the superior costotransverse ligament either by perception of a tactile change in resistance or utilization of a loss of resistance to injected air or saline. The end point with the latter is more subtle and subjective than with epidural anesthesia.3,23,24 As an alternative approach, the block needle may be safely advanced by a fixed distance (1 cm) after caudal redirection from the transverse process,25 followed by tactile identification of the appropriate resistance with initial local anesthetic injection. The depth from skin to thoracic PVS varies by vertebral level and patient size.21

A cautious approach is warranted if bone is contacted deep to the transverse process, as this may represent the rib. Further advancement of the block needle past the rib can result in pleural puncture and pneumothorax. This risk can be minimized by caudad redirection (as opposed to cephalad redirection) of the needle after initial bony contact. If the rib is unintentionally contacted first, caudal redirection will bring the block needle in contact with the transverse process at a shallower depth (Fig. 48-10). Subsequently, a more accurate estimation of the depth of the PVS is provided, and the risk of pleural puncture is decreased. In contrast, an alternative approach has been described in which the needle is redirected cephalad to the transverse process. If initial contact is made with the rib, cephalad redirection will not bring the needle into contact with the transverse process. Further advancement of the needle deep to the rib increases the possibility of pleural puncture and possible pneumothorax.

When the thoracic PVS has been located, 3 to 5 mL of local anesthetic is typically injected at each level. Injection should occur without resistance. Deliberate aspiration to detect pleural puncture or intravascular or subarachnoid needle placement is important. Local anesthetic doses may need to be adjusted when multiple-level or bilateral PVB is performed.

Nerve block adequacy is confirmed by appropriate thoracic dermatomal sensory block, intercostal muscle motor block, and sympathectomy-related changes such as vasodilation and increased skin temperature.

Lumbar PVB involves a modification of the thoracic technique. In contrast to the thoracic level, needle entry site for each spinal nerve at the lumbar level corresponds to the spinous process of the same level. Correction for spinous process angulation is unnecessary in the lumbar region. In addition, after contact with the lumbar transverse process, the block needle is advanced no more than 0.5 cm anteriorly. This is because the lumbar transverse processes are thinner in the anteroposterior plane as compared with the thoracic spine. Finally, the loss of resistance upon entrance into the lumbar PVS is even less defined as compared with that felt in the thoracic spine. This may be related to absence of the costotransverse ligament in the lumbar region.

Nerve Stimulation

The spinal nerve can be located in the PVS using a nerve stimulator and a 21- or 22-gauge short bevel insulated stimulating needle. The needle is inserted as previously described. Direct paraspinal muscle stimulation is observed initially as the needle passes through these muscles. Further advancement caudad to the transverse process brings the stimulating needle in close proximity to the spinal nerve and leads to intercostal or abdominal muscle contractions depending on the level of the block. Similar to other peripheral nerve block technique, appropriate needle placement is confirmed when muscle contractions persist with a current 0.5 mA or less. Isolated posterior spinal muscle contraction should not be accepted, as this may represent direct muscle activation or stimulation of the posterior ramus of the spinal nerve root after it diverges from the spinal nerve.

Proponents of this technique describe its usefulness in challenging cases (ie, morbid obesity, ankylosing spondylitis) and as a way to minimize the occurrence of pneumothorax, though this has yet to be proven.26

Ultrasound Guided

Ultrasound-guided single-injection and continuous paravertebral blockade may prove beneficial to one skilled in its application. Ultrasound guidance allows for real-time identification of pertinent anatomy, visualization of needle-tip advancement, and local anesthetic spread. Ultrasound imaging can be used to provide an estimate of transverse process location and depth.27,28 Pusch et al28 found a close correlation between the ultrasound estimated and the actual demonstrated needle depth from skin to transverse process, thoracic PVS, and parietal pleura. Ultrasound guidance, however, is not without limitations: Hara et al27 visualized both transverse process and parietal pleura at T4 in all patients, but only the transverse process alone at T1.27

Ultrasound guidance can be performed using a transverse or saggital paramedian ultrasound-guided technique.

Transverse Ultrasound-Guided PVB

With the patient in the sitting, prone, or lateral decubitus position, place a low-frequency linear or curvilinear ultrasound probe in the transverse plane at a position lateral to the spinous process on the rib/transverse process of the desired level. While maintaining the transverse orientation, slide the ultrasound probe slightly caudad to a position between adjacent transverse processes to visualize the PVS. Using an in-plane technique, a needle is directed toward the PVS from the lateral aspect of the probe (Fig. 48-11A and 48-11B). Local anesthetic spread is visualized within the PVS with anterior displacement of the pleura and lateral spread to the intercostal space. Alternatively an out-of-plane technique may be used with this approach. Upon subsequent saggital scan, local anesthetic spread can be visualized at contiguous paravertebral spaces with a widening of the PVS.

Paramedian Saggital in-Plane Ultrasound-Guided PVB

With the patient in the sitting, prone, or lateral decubitus position, place a low-frequency linear or curvilinear ultrasound probe in the saggital plane at a position 2 to 3 cm lateral to the spinous process. Rotate the probe to a slight oblique axis orientation from the saggital plane to achieve visualization of the PVS.29 The needle is directed toward the PVS using an in-plane technique (Fig. 48-11C and 48-11D). Alternatively, the needle may be advanced to make contact with the lower border of the transverse process and then redirected toward the PVS, advancing the needle a predetermined distance of 1 cm beyond the transverse process as with the anatomic or loss of resistance approach.25,29 Upon injection, local anesthetic spread is visualized within the PVS with anterior displacement of the pleura. Local anesthetic spread can be visualized at contiguous paravertebral spaces with a widening of the PVS.

Other Techniques

Injection of Radiographic Contrast Dye

Injection of contrast dye and radiologic examination can be used to confirm proper paravertebral needle location. Dye spread occurs in either a longitudinal distribution or a segmental, cloud-like dispersal7 (Fig. 48-12). Naja et al30 hypothesized that dye dispersion is dependent on the location of the needle with respect to the endothoracic fascia. They conjectured that needle position posterior to the endothoracic fascia would be associated with segmental, cloud-like dye spread and needle position anterior to the endothoracic fascia with more longitudinal dye distribution.30

Pressure Transduction

Pressure measurement has been described to confirm paravertebral needle placement.24 When the needle tip is in the erector spinae muscle, the measured pressure is higher during inspiration (mean 29.6 mm Hg) than expiration (mean 19.4 mm Hg).24 This is thought to occur as a result of greater muscle activity during inspiration or due to muscular compression caused by the expanding chest cage.23 As the needle is advanced into the PVS, there is a sudden lowering of pressures, and expiratory pressure becomes higher (7.6 mm Hg) than inspiratory pressure (3.3 mm Hg).24 Unintentional pleural puncture is identified when subatmospheric pressures are recorded, both during inspiration and expiration.

Continuous PVB

Placing a continuous catheter into the PVS can be used to extend the duration of postoperative analgesia provided by PVB. This is typically reserved for more invasive surgery associated with intense postoperative pain, such as thoracotomy and subcostal incisions. Insertion methods are usually modifications of single-injection techniques. A single-injection 22-gauge, 10-cm Tuohy needle is initially used to estimate the depth of the transverse process and paravertebral space. A larger-bore Tuohy (ie, 18 gauge) capable of accommodating a 20-gauge epidural catheter is then substituted. Extension tubing or a hemostatic valve is used to create a closed circuit with the needle to mitigate entrainment of air if the pleura is punctured. The catheter is threaded 1 to 2 cm into the paravertebral space to minimize the risk of migration. Consequently, a catheter with a single distal orifice is used in order to reduce leakage of local anesthetic solution outside of the paravertebral space.

Final catheter position is variable. A cadaveric study by Luyet et al31 revealed correct paravertebral spread of contrast in only 11 of 20 cases, with 1 catheter found in the pleural space, 6 in the epidural space, and 2 with prevertebral spread of contrast dye. A cadaveric study by Riain et al32 demonstrated 8 of 10 catheters correctly placed within the PVS. Ultrasound-guided continuous paravertebral catheters placed in fresh cadavers were found outside of the paravertebral space in 40% of cases.33 In contrast, Renes et al34 determined 100% PVB correct catheter position within the thoracic PVS in 36 patients using a transverse in-plane ultrasound-guided technique with radiologic confirmation.34

An alternative to the classic approach is an intercostal approach to paravertebral catheterization as described by Burns et al.35 At the midpoint of the anticipated surgical incision, the appropriate rib is identified and marked 8 cm lateral to the midline. An 18-gauge Tuohy needle is advanced to come in contact with the rib. With the bevel directed medially (the tip angled 45 degrees cephalad and 60 degrees medial to the saggital plane) in effort to place the bevel and tip away from the pleura, the needle is "walked-off" the inferior border of the rib. Once under the rib, the needle is advanced, maintaining the aforementioned position, 5 to 6 mm to enter the intercostal neurovascular space. After injection of local anesthetic, a 20-gauge epidural catheter is advanced through the needle and inserted 8 cm past the needle tip, brining it to the PVS.35

A novel insertion method described for thoracic surgery is to place the catheter under direct vision in the surgical field4,23,36-40 (Fig. 48-13). At the completion of the thoracotomy just before chest closure, the surgeon strips away the parietal pleura in the paravertebral gutter at the level of the incision, 2 dermatomes cephalad and 2 dermatomes caudad. A small defect is made in the extrapleural fascia. Then, a Tuohy needle is introduced percutaneously and a catheter is placed into the PVS through the small defect in the extrapleural fascia. The parietal pleura is then repositioned and sutured in place. During thoracoscopic procedures, catheter placement may be done with video assistance by the surgeon41 or in combination with the anesthesiologist.42

Single-Level versus Multiple-Level Technique

Paravertebral anesthesia can be provided using a single- or multiple-level injection technique. A single-level injection technique involves depositing a large volume of local anesthetic (10-20 mL) at 1 level. A multiple-level injection technique involves depositing a smaller volume of local anesthetic (3-5 mL thoracic; 5-7 mL lumbar) at each involved dermatomal level. The theoretic advantage of the former is a reduction in potential needle insertion complications such as pleural puncture or pneumothorax, whereas the latter may provide more extensive and complete anesthesia of the desired dermatomes. Although single-level injection PVB has been shown to reduce postoperative pain and hospital stay,3-45 several authors have recommended using a multiple-level injection technique for surgical anesthesia.23,25,46-50 In a study by Naja et al,50 97% of patients had complete loss of sensation with 4-level injection as compared with 11% with single-level injections. In an ultrasound-guided study of fresh cadavers, contrast dye spread more extensively across intercostal segments with a dual-injection technique as compared with a single-injection technique (6 vs 4.5; p < .03).33 There was 40% frequency of epidural spread of contrast associated with paravertebral injection in both techniques.33

Effect of Technique on Distribution of Injectate

A single-level paravertebral injection of 10 to 15 mL of local anesthetic in adults12,48,51 and 0.5 mL/kg in children7 achieves a mean sensory block of 4 to 5 dermatomes independent of age, height, weight, or sex. Naja et al,50 in a prospective randomized trial investigating radiographic and clinical differences in local anesthetic spread as a function of injection number, concluded that increasing the number of paravertebral injections results in more reliable radiographic and sensory-block distribution compared with a single-injection technique. In this study they demonstrated vertical contrast spread with 4 injections to be a mean (SD) of 6.5 (2.01) dermatomes as compared with 3.0 (1.19) dermatomes achieved with 1 injection. The vertical distribution pattern for both single-injection and multiple-injection techniques is slightly asymmetric, with a more extensive caudad spread of the block,50 as previously identified.12

Few studies have been designed to rigorously investigate the potential effect of dose or volume of injectate on spread within the PVS. One study in adults has shown no association between these variables,48 whereas another study of pediatric patients has found a moderate to strong correlation between injected volume and segmental spread of dye.7 It has been proposed that there may be an age-related loss of connective tissue leading to greater spread of injectate in adults.48

Additional factors do affect the spread of injectate within the PVS. Purcell-Jones et al8 studied single-injection PVB using the loss of resistance technique. They radiographically showed greater craniocaudal distribution of solution (5.5 mL) when the PVB was associated with epidural spread. They found sensory anesthesia in a mean 1.43 dermatomes when injectate was confined to the PVS, 2.27 dermatomes when injectate was primarily in the PVS but also spread to the epidural space, 4.66 dermatomes when injectate was confined to the epidural space, and 6.6 dermatomes when the injectate was primarily in the epidural space but also spread to the PVS. In all instances, the spread of contrast dye was greater than the observed sensory block. Cheema et al12 studied patients who received a single-injection PVB using loss of resistance with 15 mL of local anesthetic and dye. They found that the mean extent of vasodilation (sympathetic block) was 8 dermatomes and exceeded the mean sensory block of 5 dermatomes.

Local Anesthetic Regimen

Bupivacaine and ropivacaine are the most frequently used local anesthetics for PVB, and the duration of analgesia is similar to that obtained with brachial plexus anesthesia, a length of 12 to 24 hours. In a study by Hura et al,52 ropivacaine was associated with a more rapid onset of sensory blockade, with 53% of patients demonstrating dermatome coverage to undergo modified radical mastectomy 5 minutes after block completion as compared with 20% in the bupivacaine group (p < .01). Although ropivacaine demonstrated a longer duration of analgesia and a wider extent of segmental anesthesia (p < .05), degree of postoperative pain and analgesic consumption were similar between groups.52 Navlet et al53 concluded that both bupivacaine 0.25% and ropivacaine 0.3% are equally effective to control post-thoracotomy pain at rest with a continuous paravertebral block, but suggested that higher concentrations of both drugs during the first 24 hours might improve visual analog scale (VAS) scores on coughing and spirometry values, including forced vital capacity and forced expiratory volume in 1 second (FEV1). Indeed, a systemic review and meta-regression by Kotze et al54 demonstrated that the use of higher doses of bupivacaine (890-990 mg/24 h compared with 325-472.5 mg/24 h) was found to predict lower pain scores (50%) at all time points up to 48 hours after the operation (p < .006 at 8 h, p < .001 at 24 h, p < .001 at 48 h) and faster recovery of pulmonary function by 72 h (20% improvement in FEV1 (p < .029). Continuous infusions of local anesthetic predicted lower pain scores compared with intermittent boluses (p < .04 at 8 h, p < .003 at 24 h, p < .001 at 48 h). Epinephrine is frequently added to the local anesthetic solution to indicate intravascular injection, to reduce the peak local anesthetic blood level,55 and to improve analgesia. Epinephrine produces a 25% reduction in the mean peak arterial concentration of ropivacaine and delays the time to peak arterial and venous concentrations.55 Some add opioids or clonidine to local anesthetic solutions46,57; however, this has not been extensively studied with PVB. Standard dosing regimens are outlined in Table 48-2. A single-level injection is typically performed with 0.5 mL/kg of dilute local anesthetic solution in children and 10 to 20 mL in adults. Multiple-level injections are accomplished in adults with 3 to 4 mL of local anesthetic per segment in the thoracic region and 5 to 7 mL per segment in the lumbar area. Continuous PVB is dosed with 0.1 to 0.2 mL/kg/h of dilute local anesthetic.

The advantages of PVB are summarized in Table 48-3. PVB provides blockade of afferent nerves, leading to dense anesthesia and postoperative analgesia. Richardson et al57 performed somatosensory evoked potential testing on patients before and after they received a single injection block. They discovered complete abolition of the evoked potentials at the level of the paravertebral injection in 100% of patients (Fig. 48-14). As a result, PVB may be associated with a reduced development of chronic postsurgical pain after breast surgery. In a study by Iohom et al59 at 10 weeks, 80% of patients (12 of 15) not receiving PVB analgesia developed chronic postsurgical pain as compared with 0% (0 of 14) in those receiving PVB analgesia (p < .009). Preoperative PVB may reduce the incidence of chronic pain 1 year after breast cancer surgery60 and reduce the incidence of post-thoracotomy neuralgia.61

PVB also modifies the stress response to surgical stimulation. Giesecke et al62 studied patients having open cholecystectomy under general anesthesia and discovered that those who received single-injection PVB for postoperative pain control had lower plasma adrenaline and cortisol concentrations (p < .05), as well as a smaller rise in plasma glucose concentration (p < .025) compared with patients who did not receive PVB.

PVB is associated with reduced postoperative opioid consumption and incidence of opioid-related side effects when compared with intravenous opioid analgesia43,46,63-68 and preservation of lower-extremity motor strength, greater hemodynamic stability, and less urinary retention when compared with epidural analgesia.69 This technique has the potential to provide improved analgesia with a reduced risk of pleural puncture and local anesthetic toxicity compared with intercostal and intrapleural analgesia.

A retrospective analysis suggested that paravertebral anesthesia and analgesia for breast cancer surgery reduced the risk of recurrence or metastasis during the initial years of follow-up. The proposed mechanism was maintenance of perioperative immune function by attenuation of the stress response and reduction in volatile anesthetic and postoperative opioid requirements.70 Prospective trials are necessary to investigate the conclusions of this retrospective analysis.

Absolute Contraindications

Similar to other percutaneous techniques, PVB should be avoided in patients with skin infections at the proposed needle entry site and in those with an empyema or a deep infection within the chest.2,71 A paravertebral tumor also contraindicates PVB due to the risk of tumor seeding. In addition, this technique is avoided in the setting of local anesthetic allergy, patient refusal, and severe hemodynamic instability.

Relative Contraindications

A severe chest deformity, such as kyphoscoliosis, increases the risk of pneumothorax or unintentional subarachnoid injection and should be considered a relative contraindication to PVB.2,71 Similarly, patients who have had a prior thoracotomy are at higher risk of pleural puncture and pneumothorax due to paravertebral scarring and obliteration of the thoracic PVS. Nevertheless, operator experience and/or imaging adjuncts can result in successful blocks in these circumstances with few complications. Coagulopathy represents a relative contraindication to PVB. Based on the increased compliance of the PVS compared with the epidural space, the neurologic impairment from a paravertebral hematoma will be less than that from an epidural hematoma.71

Two large series provide incidence rates for adverse events after PVB in both adults and children.72,73 (Table 48-4). Naja and Lonnqvist72 provide data on multilevel PVB performed on 662 adult and pediatric patients using nerve stimulation. Lonnqvist et al73 reported their results from single-level PVB performed with a loss-of-resistance technique in 367 adult and pediatric patients. Additional adverse event data can be gleaned from other studies.25,47,68

The most common reported adverse event is technique failure. A failure occurs when a PVB does not succeed in providing adequate surgical anesthesia or when general anesthesia is required to complete the procedure. The published incidence of block failure is 6.1% to 10.7%25,47,68,72,73 and is relatively similar among different operators and different institutions. Interestingly, the incidence of failure was slightly higher in the series in which a single level PVB was performed with the loss-of-resistance technique (10.7%)73 compared with the study in which multiple PVB was performed using nerve stimulation (6.1%).72

Unintentional vascular puncture occurs in up to 6.8%72 and when recognized before injection of local anesthetic is of little consequence. Blood pressure and heart rate are generally maintained within normal limits after both unilateral and bilateral PVBs.2 Hypotension has been reported in up to 5.0% of patients73; however, it is usually mild and easily managed. Other common side effects of PVB includes localized hematoma or pain at the site of injection. These are usually self-limiting and do not require treatment.

Unintentional pleural puncture can result from deep insertion of the nerve block needle. A pneumothorax may result and is considered a serious adverse event. An early warning sign includes cough during needle insertion. Aspiration of air occurs if the lung has been punctured or if air has been introduced into the pleural cavity by the needle.2 Shortness of breath or pleuritic chest pain may follow the PVB. To establish the diagnosis and estimate the size of a pneumothorax, a chest x-ray is often required. Many pneumothoraces that result from PVB are small and can be managed conservatively.2 Although there are no studies directly comparing the incidence of pleural puncture or pneumothorax using various PVB techniques, the incidences in 2 large series can be contrasted. The incidence of pleural puncture and pneumothorax were 2.1% and 0.3% to 2.1%, respectively, in the series who received single-level PVB with loss of resistance73; in contrast, these adverse effects occurred in 0.8% and 0.5% in the series of patients who received multilevel PVB using nerve stimulation.72 Pulmonary hemorrhage is a rare respiratory complication that has been reported after PVB.74 The risk of both pneumothorax and pulmonary hemorrhage is thought to be increased in patients who have an obliterated PVS from scar tissue as a result of previous thoracotomy.74

Studies on the spread of injectate within the PVS have shown spread of contrast dye into the epidural space in a variable proportion of patients8; however, the magnitude with which epidural block contributes to paravertebral anesthesia is unpredictable.2,72,73 Subarachnoid injection,75 total spinal anesthesia, and postdural puncture headache76 are also rare adverse events. They can result either from needle entry into the subarachnoid space or into dural extensions around the spinal nerve roots. The hemodynamic effects of an unintentional thoracic subarachnoid injection may require efficient resuscitation.

Cephalad spread of local anesthetic to the brachial plexus or the stellate ganglion frequently occurs and may result in a brachial plexus block2 or Horner syndrome,77 respectively. These effects are temporary, and patients should be reassured.

Systemic absorption of local anesthetic with resulting toxicity is rare.44 Local anesthetic plasma levels after paravertebral injections have been extensively studied. After a bolus dose of 0.75 to 1.5 mg/kg of bupivacaine, mean peak serum levels varied from 0.705 μg/mL78 to 1.45 μg/mL79 in adults and from 1.03 μg/mL80 to 1.60 μg/mL81 in children. Karmakar et al82 administered a single-level PVB with 2 mg/kg of ropivacaine and randomized patients to receive a solution with or without epinephrine. They found that epinephrine decreased mean peak serum concentration of ropivacaine from 2.47 μg/mL to 1.85 μg/mL and increased mean time to peak serum concentration from 7.5 minutes to 11.25 minutes.82 In other studies, the median time to peak plasma bupivacaine levels ranged from 5 minutes78 to 25 minutes.79 Continuous infusion of 0.2 to 0.5 mg/kg/h of bupivacaine for up to 4 days results in variable mean peak plasma bupivacaine levels. Several studies of adult and pediatric patients report levels between 1.6 μg/mL and 2.5 μg/mL.80,81,83 However, other investigators have found mean peak plasma bupivacaine concentrations as high as 4.92 μg/mL79 to 5.43 μg/mL,84 with some individuals as high as 7.48 μg/mL.79 Despite these high plasma concentrations, no patients in these studies manifested clinical signs of local anesthetic toxicity. Although total bupivacaine increases steadily during paravertebral infusion, free bupivacaine remains unchanged83 due to a perioperative increase in α1-acid glycoprotein, a serum protein that binds local anesthetic molecules. This may explain the low incidence of clinical local anesthetic toxicity seen even among patients with plasma bupivacaine concentrations that are considered above the toxic threshold.

There are published case reports of serious neurologic complications related to PVB: incomplete transverse myelitis occurred in association with the use of efocaine,85 and a Brown-Séquard syndrome resulted from paravertebral injection of alcohol.86 In more recent publications, neurologic injury is rarely reported.72,73 There is only 1 case report of chronic segmental pain after PVB87; however, some authors23 believe that this complication occurs more frequently, but is often attributed to nerve injury from surgical dissection.

The various indications for PVB are summarized in Table 48-1 and described in detail next. The most common indications include breast surgery, hernia repair, and thoracic surgery. A systematic review of the literature, which included 8 randomized clinical trials evaluating PVB for breast surgery or herniorrhaphy, concluded that PVB for surgical anesthesia is associated with less pain during the immediate postoperative period, as well as less postoperative nausea and vomiting and greater satisfaction as compared with general anesthesia.88 A systematic review of randomized trials evaluating regional anesthetic techniques for post-thoracotomy analgesia concluded that continuous paravertebral block was comparable to thoracic epidural analgesia with local anesthetic, but was associated with a reduced incidence of hypotension and a reduced incidence of pulmonary complications compared with systemic analgesia, whereas thoracic epidural analgesia was not.89

Thoracic Surgery

The use of PVB for postoperative analgesia after thoracoscopy or thoracotomy has been investigated in both adults36,37,41,63-67,90-95 and children.80,96 A commonly used approach involves preoperative PVB to provide intraoperative analgesia in addition to a continuous local anesthetic infusion via a paravertebral catheter for postoperative analgesia. Preoperative PVB is performed using either a single- or multiple-level injection technique with the goal to provide sensory block from T4 to T9. A paravertebral catheter can be placed preoperatively by the anesthesiologist percutaneously, intraoperatively by the thoracic surgeon under video assistance during thoracoscopy41 or in combination with the anesthesiologist,42 or under direct visualization at the time of chest closure.36-40

Hill et al92 investigated multilevel PVB after thorascopic procedures in a prospective, double-blinded, randomized, placebo-controlled study. The benefits of PVB included less intraoperative fentanyl (p < .003), reduced cumulative opioid consumption (p < .03), and lower maximum pain scores at 6 hours (p < .02). No significant difference in cumulative morphine consumption was reported at 12 or 18 hours after block placement, and there was no difference in spirometry, cortisol levels, or cytokine production between groups. A prospective, randomized, blinded, placebo-controlled study by Vogt et al94 demonstrated a significant difference in VAS scores at rest and coughing over a 48-hour time course with single-level injection PVB (p < .05). There was no difference with regard to peak expiratory flow rate at 24 and 48 hours between groups. Kaya et al95 demonstrated that preoperative multiple-level injection PVB for postoperative analgesia after video-assisted thorascopic surgery procedures was associated with lower intraoperative fentanyl use (p < .01), longer time to first analgesic requirement (p < .05), lower VAS at first analgesic requirement (p < .01), lower maximum VAS pain scores during the 48-hour study period (p < .01), improved patient satisfaction (p < .05), decreased time to first mobilization (p < .01), and decreased time to hospital discharge (p < .05).

In a prospective, randomized, placebo-controlled trial, Berrisford et al63 documented the beneficial effects of continuous PVB for thoracotomy patients. They found lower pain scores (p < .01), reduced opioid consumption (8.6 mg vs 119 mg over 5 days), fewer postoperative pulmonary complications (8% vs 52.4%; p < .05) and better preservation of pulmonary function (p < .01) in the group who received bupivacaine compared with saline. Comparable results were obtained in a number of studies with similar design64-67 and a non–placebo-controlled study.91 The moment of insertion of PVB catheter, before or after rib-spreading, did not affect analgesic quality after thoracotmy.97 An additional advantage of PVB involves a potential reduction in the incidence of post-thoracotomy neuralgia.61

Continuous PVB has a number of theoretical advantages over both continuous interpleural and intercostal blocks. PVB more reliably achieves block of the dorsal ramus of the spinal nerve. This is important because much of the pain that results from thoracotomy stems from the paraspinal muscles and other areas innervated by the dorsal ramus. In addition, greater cephalocaudal spread of local anesthetic occurs and results in more extensive block with continuous PVB compared with the other 2 modalities.98 Despite these theoretical advantages, prospective studies comparing continuous PVB versus interpleural99 and intercostal100 infusions of local anesthetic failed to show a difference between groups in pain scores and opioid consumption. Nevertheless, PVB was associated with better preservation of lung function as measured by forced vital capacity and FEV1.99

PVB continuous catheters have been compared with incisional (subcutaneous) catheters and a combination of the 2 for postoperative analgesia after thoracotomy.37 VAS reports were lower at rest, on coughing, and on movement at 12 (p <.01, p < .05, and p < .05, respectively) and 24 hours (p < .05, p < .05, and p < .05, respectively) after surgery when incisional catheters were used to supplement paravertebral analgesia. However, excluding the first 4 hours, there was no significant difference in daily morphine use to improve analgesia between the 3 groups. There was no significant difference in spirometry testing observed between groups.

Continuous PVB has also been compared with intermittent boluses of epidural morphine for post-thoracotomy analgesia. Dauphin et al38 found that these 2 modalities resulted in similar pain scores and supplemental intravenous morphine consumption.

A number of prospective, randomized trials exist comparing continuous PVB (local anesthetic alone) versus epidurals (local anesthetic with or without opioid).36,40,69,90,101,102 In a study by Richardson et al,101 the group who received PVB had lower postoperative pain scores at rest (p = .02) and with coughing (p = .0001) as well as lower cumulative morphine consumption over 48 hours (p ≤ .008). However, other studies36,40,90 failed to replicate these findings. Messina et al90 demonstrated increased morphine consumption with PVB as compared with epidural with local anesthetic and opioid infusion; however, VAS scores were not significantly different between the 2 groups. Opioid-related side effects were not reported. Gulbahar et al36 found no significant difference between the 2 groups with regard to VAS, serum cortisol and glucose levels, necessity for additional analgesia, and hospital stay duration. There was a significant difference in side effects, with no side effects in the PVB group (p < .01).36 A number of investigators have shown a lower incidence of side effects such as nausea,36,101 vomiting,36,101 hypotension,36,69,101 and urinary retention36,40,69,101 with PVB as compared with epidurals.

Studies of the effects of PVB on postoperative lung function have been conflicting. Richardson et al101 documented that the postoperative peak expiratory flow rate as a fraction of preoperative value was 0.73 in the PVB group compared with 0.54 in the epidural group (p < .004), suggesting better preservation of lung function with PVB. In contrast, Messina et al90 reported improved spirometry values at 72 hours with return of forced vital capacity to 83% of the preoperative value in the epidural group, as compared with 31% in the paravertebral group. Other studies36,102 found no significant difference between the 2 analgesic modalities.

In addition to their utility for thoracoscopies and thoracotomies, PVB has also been used successfully for other indications. In a prospective, randomized, placebo-controlled trial of patients having pleurectomy, Mozell et al103 established that continuous PVB with bupivacaine provided lower pain scores, reduced opioid consumption, and better preservation of pulmonary function than placebo. In a retrospective study by Patel et al,104 the addition of PVB to general anesthesia after first rib resection demonstrated decreased postanesthesia care unit (PACU) pain scores. PVB has also been used to provide analgesia for rib fractures. Mohta et al105 demonstrated that continuous PVB provided comparable analgesia and respiratory function to continuous thoracic epidural analgesia (TEA) in patients with unilateral multiple rib fractures with a decreased incidence of hypotension. There was no significant difference in VAS scores at rest and on coughing, respiratory rate, morphine requirement, peak expiratory flow rate, incidence of pulmonary complications, infusion duration, length of intensive care unit stay, and length of hospital stay.105 Additionally, PVB has been used to provide analgesia for rib fractures in patients with head106 and spinal cord107 injuries. In patients with head injury, paravertebral analgesia reduced the need for potentially sedating analgesics and enhanced neurologic assessment.106 Unilateral PVBs were administered instead of an epidural block in patients with lumbar spinal cord injury because the PVB enhanced assessment of lumbosacral spinal cord function.107 Moreover, paravertebral analgesia provides quality analgesia and has beneficial effects on pulmonary function among patients with traumatic chest injury.108 Additional applications have been in the management of chest pain caused by pleural effusion.109

Breast Surgery

PVBs can be used as the sole intraoperative anesthetic or as an analgesic supplement to general anesthesia for breast surgery. The block may be performed as a single-level injection or as injections at multiple levels. The dermatomal levels blocked depend on the surgical procedure (Table 48-1). Block of T2-T6 is required for mastectomy. When mastectomy and axillary dissection is scheduled, extending the block to include the T1 dermatome is essential. Supplementing with a superficial cervical plexus block can also enhance analgesia at the superior aspect of the incision. Breast biopsy requires block of the dermatome involved in addition to 1 level cephalad and 1 level caudad. Long-acting local anesthetics such as bupivacaine can provide postoperative analgesia for up to 23 hours.110

A number of prospective, randomized trials exist comparing general anesthesia versus PVB for breast surgery.43,45,46,59,68,111,112 Pusch et al68 performed a single-injection block with bupivacaine using the loss-of-resistance technique at T4 in patients having breast cancer surgery. They found rapid onset of block, with skin incision occurring within 15 minutes of the block. They also documented a shorter emergence time in the group who had PVB (p < .01). The PVB group had lower pain scores during the 13 hours of the study (p < .05) and received fewer analgesics (p < .01). There was less painful restricted motion (p < .001) and a reduced incidence of postoperative nausea and vomiting (p < .05) in the PVB group. Paravertebral anesthesia with propofol sedation was inadequate in 6% of patients; however, supplemental analgesia with intravenous fentanyl was sufficient to avoid conversion to general anesthesia. Naja et al46 used a nerve stimulator and performed injections at multiple levels with a mixture of lidocaine, bupivacaine, epinephrine, fentanyl, and clonidine. They showed an analgesic benefit for the first 5 days postoperatively and also documented a shorter length of hospital stay among patients who received PV (p < .01). Klein et al43 studied patients having unilateral or bilateral cosmetic and reconstructive breast surgery and also showed similar benefits, including lower verbal pain scores up to 72 hours after surgery. In addition, they were able to reduce intraoperative induction time from 24 minutes with general anesthesia to 4 minutes with PVBs by using a preoperative block area. Terheggen et al45 studied patients having minor breast surgery and performed PVB with bupivacaine via a catheter placed at T3-T4. The PVB reduced intraoperative fentanyl requirements and lowered postoperative pain scores only in the first 90 minutes postoperatively for this group of patients having surgery that may be considered less stimulating. Boughey et al111 compared general anesthesia with general anesthesia with multilevel PVB at T1 to T6 for patients undergoing breast cancer surgery. This was a prospective study in follow-up to a retrospective analysis that demonstrated improved postoperative analgesia up until the morning after surgery and a decreased proportion of patients requiring overnight stay after major breast operations.111 The prospective study demonstrated decreased pain scores and a greater percentage of pain-free patients at 1 hour (1 vs 3, p < .014; 44% vs 17%, p < .006) and 3 hours (0 vs 2, p < .001; 54% vs 17%, p < .005) after surgery; however, at 6 hours the difference was no longer evident. The overall worst pain score was lower with PVB (3 vs 5, p < .02), and there was a greater number of subjects in the PVB group reporting to be pain-free during hospital stay (33% vs 12%, p < .032). At 24 hours, there were more subjects reporting to be pain-free in the general anesthesia–alone group (23% vs 54%, p < .011), which might be representative of inadequate analgesia with block resolution. There was no difference in opioid consumption, nausea/vomiting, and hospital length of stay between groups. This study was limited in that it did not include a standardized surgical procedure. An underpowered subgroup analysis of the more extensive surgeries failed to demonstrate a difference between groups at any time. Moller et al112 in a prospective, randomized, double-blind, placebo-controlled study also demonstrated duration of postoperative analgesia provided with PVB to be less than that described in previous studies. Intraoperatively, PVB was associated with a significant decrease in intraoperative fentanyl (p < .0001) and propofol (p < .001). In the PACU, PVB was associated with decreased opioid consumption p < .001) and improved pain control (patients reporting VAS <3; p < .0001). There was no difference in postoperative nausea and vomiting between groups. No benefits were demonstrated beyond the PACU. Another, randomized, placebo-controlled trial44 studied patients scheduled for breast cancer surgery receiving single-injection PVB at T3 with bupivacaine 0.5% (1.5 mg/mL) or saline before general anesthesia. Pain scores were low in both groups; however, patients receiving PVB with bupivacaine had less postoperative pain in the PACU and up to 12 hours after surgery, as measured by longer time to first analgesic (20 min vs 10 min; p < .019) and lower VAS (PACU, 3.0-1.7 vs 4.8-1.4, p <.025; 12 h, p < .096). Postoperative pain was minimal in both groups of this study, as evident by low PACU pain scores and zero opioid consumption after discharge from PACU in either group. There was no significant difference in the number of patients with postoperative nausea and vomiting (PONV) between groups. Iohom et al59 demonstrated improvement of postoperative VAS pain scores with PVB continuous catheters at rest up to 12 hours (p < .035) and movement at 12 hours and on postoperative days 1, 2, 3, 4, and 5 (p < .04.)

A prospective comparison of continuous wound infiltration with ropivacaine versus single-level injection PVB after modified radical mastectomy113 demonstrated low pain scores in both groups. PVB was associated with reduced pain scores at 4 hours (p < .02) and less PONV. Continuous wound infiltration was associated with reduced pain scores at 16, 20, and 24 hours. There was no difference in opioid consumption between groups at any time point.

PVB may be associated with a reduced development of chronic postsurgical pain after breast surgery. In a study by Iohom et al59 at 10 weeks, 80% of patients (12 of 15) not receiving PVB continuous catheters developed chronic postsurgical pain as compared with 0% (0 of 14) in those receiving PVB catheters (p < .009). Preoperative PVB may reduce the incidence of chronic pain 1 year after breast cancer surgery.60

Hernia Repair

PVB can be used for surgical anesthesia or for analgesia as a supplement to general anesthesia for inguinal, umbilical, or incisional hernia repair. When performed for inguinal hernia repair, block of the T10 through L2 dermatomes is typically required. This can be achieved by a single-level injection in the low thoracic PVS or by injections at multiple levels.

PVB may be used as the sole anesthetic for inguinal hernia repair.114,115 Klein et al114 demonstrated that bupivacaine PVB provided surgical anesthesia within 15 to 30 minutes and prolonged postoperative analgesia with a mean time to first opioid of 22 hours. In a study by Weltz et al,115 the failure rate requiring conversion to general anesthesia was 6.7%; however, these investigators documented low pain scores for 48 hours postoperatively.

Studies have evaluated the efficacy of PVB compared with other anesthetic techniques, including general anesthesia, spinal anesthesia, ilio-inguinal block, and local field block for inguinal hernia repair. As compared with general anesthesia, multilevel PVB from T9 to L1 was associated with shorter time to home readiness, greater phase 1 PACU bypass, less postoperative pain, faster ambulation, and quicker discharge home.116 Single-level injection PVB at L1 demonstrated shorter time to ambulation, increased recovery room bypass, and decreased postoperative urinary retention as compared with spinal anesthesia with 12.5 mg of 0.5% bupivacaine.117 Multilevel PVB from T9 to L1 versus unilateral spinal anesthesia with 8.0 mg of 0.5% bupivacaine provided statistically significant shorter hospital stays with shorter time to home readiness (with and without voiding), shorter discharge time, and prolonged postoperative analgesia with a mean time to first analgesic of 16 hours as compared with 7 hours in the spinal group.118 Naja et al119 compared multilevel PVB from T12 to L2 versus general anesthesia versus spinal anesthesia. Again, patients in the PVB group had a shorter duration of hospital stay (1.2 d vs 2.9 d for general anesthesia vs 2.5 d for subarachnoid block; p < .0001), better postoperative analgesia, and a lower incidence of postoperative nausea and vomiting (0 vs 21% for general anesthesia vs 19% for subarachnoid block; p < .001). In a randomized trial, Wassef et al120 compared lidocaine PVB versus lidocaine/bupivacaine field block. The PVB was associated with less frequent intraoperative supplementation (20% vs 41%; p < .01), a lower rate of conversion to general anesthesia (0 vs 6.7%), and greater patient satisfaction (p < .05).

Alternatively, PVB can supplement another primary anesthetic technique (ie, general or spinal anesthesia) to provide postoperative analgesia. In a prospective, randomized study of subjects undergoing inguinal hernia repair under general anesthesia, Klein et al121 compared postoperative analgesia from ropivacaine PVB with ilioinguinal-iliohypogastric nerve blocks with wound infiltration. They found reduced opioid consumption intraoperatively (p = .02) and in the PACU (p = .002), as well as a lower antiemetic use (p < .001) in the PVB group; however, there was no subsequent difference in pain scores and opioid use. In children undergoing inguinal herniorrhaphy under general anesthesia, multilevel PVB from T12 to L1 as compared with general anesthesia with systemic analgesia (GA/SA) demonstrated improved postoperative analgesia during the first 48 hours, increased same day discharge (80% PVB group vs 52% GA/SA), and greater parental and surgeon satisfaction.122 As compared with ilio-inguinal block, multilevel PVB from T12 to L2 in children undergoing inguinal herniorrhaphy under general anesthesia demonstrated a significant improvement in intraoperative hemodynamic stability, postoperative analgesia with decreased consumption of analgesic drugs during the first 36 hours after surgery, and again greater parental and surgeon satisfaction.123

Ozkan et al124 comparing 2-level (T10, L1) PVB versus 4-level (T10-L1) PVB demonstrated shorter block performance time (5 min [SD 1] vs 16 min [SD 4]; p < .001) with 2-level injection but no difference in the other parameters investigated, including intraoperative propofol and remifentanil use, VAS scores, postoperative analgesics consumed, sensory block duration adverse effects, PONV, and patient satisfaction.

When PVB is used for umbilical hernia repair of moderate size, a bilateral block from T9 to T11 is required. In a nonrandomized study, Naja et al56 compared PVB versus general anesthesia. PVB was performed using a nerve stimulator and a mixture of lidocaine, bupivacaine, epinephrine, fentanyl, and clonidine. Interestingly, 10% of patients in the PVB group required supplemental analgesia due to unanticipated extension of the surgical field. Nevertheless, PVB had a number of benefits, including lower pain scores and reduced opioid requirements up to 48 hours (p < .001). PVB was also associated with a decreased incidence of PONV (3.3% vs 26.7%; p < .05) and shorter length of hospital admission (2.3 d vs 4.1 d; p < .05).

Finally, PVB can be used for incisional hernia repair and are performed according to the dermatomal levels involved. Block from T8 to T12 has also been used for ileostomy closure.125

Renal Surgery

Unilateral PVB is well suited to provide postoperative analgesia after renal surgery. A PVB continuous catheter can be placed preoperatively, or alternatively, a catheter can be placed under direct vision by the surgeon at the time of wound closure.126 In a randomized, prospective, placebo-controlled trial, Awwad and Atiyat126 found reduced pain scores for 3 days (p < .026) and decreased opioid consumption (13.3 mg vs 40.13 mg over 3 d; p < .001) in the continuous PVB catheter with bupivacaine versus saline. Lonnqvist127 and Lonnqvist and Olsson128 have used continuous PVB in children having renal surgery. In a nonrandomized study, they compared the postoperative analgesia obtained from paravertebral versus epidural bupivacaine128 and found decreased opioid consumption in the PVB group. In a case report of 30 patients, PVB as a part of a multimodal analgesia regimen provided significant opioid sparing as compared with previous studies129,130 after hand-assisted laparoscopic nephrectomy.131

Jamieson and Mariano132 reported pain scores of zero for 24 hours and no opioid rescue analgesia in 2 patients undergoing lithotripsy under PVB.

Cholecystectomy

Few studies exist evaluating the efficacy of PVB for postoperative analgesia after open cholecystectomy by subcostal incision. Giesecke et al62 suggest that PVB reduces the stress response to surgery; however, Bigler et al87 found no benefit from continuous PVB when compared with thoracic epidural. More recently, PVB has been used to provide postoperative analgesia for laparoscopic cholecystectomy. In a prospective, randomized study, Naja et al133 randomized patients receiving general anesthesia to have PVB or opioid analgesia. Although there was no difference between groups in time to first oral intake or length of hospital stay, patients who received PVB had lower pain scores (p < .05), reduced opioid consumption over 36 hours (p < .05), and a decreased incidence of PONV (p < .05). Paleczny et al134 in a randomized, prospective study compared general anesthesia alone with PVB performed before the induction of general anesthesia. PVB demonstrated a significantly lower mean pain score during the first 72 hours after surgery (p < .005) and improved patient satisfaction.

Appendectomy

Splinter and Thompson135 demonstrated in children (3-16 y) undergoing appendectomy that PVB T11 to L1 using ropivacaine 0.2% 0.25 mL/kg with 1:200 000 epinephrine was associated with decreased opioid consumption (p < .001) with an increased time to first opioid dose (p < .001). There was no significant difference in vomiting and no other adverse effects observed in the 2 groups.

Orthopedic Surgery

PVB can be used as an adjunct to other peripheral nerve blocks or as the sole regional technique for the treatment of postoperative pain in orthopedic patients. In patients receiving an interscalene brachial plexus block for shoulder surgery, the addition of T1 to T2 PVB provides more complete shoulder analgesia. And, in combination with a lumbar plexus block, T11 to T12 PVB provides more comprehensive analgesia for patients having hip surgery. Alternatively, PVB from T12 to L4 and from L2 to S1 has been used to provide postoperative analgesia after total hip and knee arthroplasty, respectively.136 For hip arthroscopy, Lee et al presented 2 cases in which they performed 2-level (L1 and L2) PVB with 5 mL of ropivacaine 0.5% preoperatively in combination with general anesthesia and injection of 20 mL of 0.25% bupivacaine with epinephrine into the hip joint by surgery at the end of the procedure. This regimen provided postoperative analgesia of approximately 36 hours and greater with no opioid requirement.

Bone Marrow Aspiration

Bilateral PVB from T11 to L2 can be used to provide intraoperative anesthesia and postoperative analgesia for patients having bone marrow aspiration. This technique provides an alternative option when general and/or spinal anesthesias are contraindicated (ie, mediastinal tumor and chemotherapy-related thrombocytopenia).

Cardiac and Vascular Surgery

PVB has been used both for analgesia after cardiac surgery as well as peripheral vascular procedures. In a prospective, observational study, Canto et al137 performed continuous bilateral ropivacaine PVB at T3 to T4 for patients having open-heart surgery with cardiopulmonary bypass. They reported excellent analgesia, a low complication rate, and facilitation of early tracheal extubation. Moderate hypotension and bradycardia were repeatedly observed and presumed due to sympathectomy; however, dopamine support was required in only 6.4%.

Bilateral T4 to T5 continuous PVB has also been successfully used for postoperative analgesia after minimally invasive coronary artery bypass138 and may facilitate early extubation.139 In a prospective, randomized trial, Dhole et al140 compared continuous PVB versus epidural and found no significant differences between the techniques with respect to analgesia.

A prospective randomized study in patients undergoing elective robotic-assisted coronary artery bypass grafting comparing thoracic epidural analgesia with PVB revealed no significant difference with regard to hemodynamics, arterial blood gases, pulmonary functions, and analgesia.141

Thoracic PVB with mild sedation was successfully used for implantable cardioverter-defibrillator and laser lead extraction in a patient who had a high opioid tolerance and desired prolonged analgesia.142

Richardson et al143 described paravertebral analgesia after major abdominal vascular surgery in an observational study. They placed bilateral PVB catheters at T10 in 8 patients and administered an infusion of bupivacaine. The continuous PVB provided excellent postoperative analgesia with preserved hemodynamic stability.

A retrospective review comparing paravertebral blockade with propofol sedation with general anesthesia for elective endovascular abdominal aortic aneurysm repair demonstrated decreased intraoperative hypotension (p < .05) and blood pressure lability (p < .01), as well as postoperative nausea (p < .01).144

Hepatic Surgery

A prospective, randomized, placebo-controlled study compared bilateral PVB injection of 25 mL of 0.25% bupivacaine with epinephrine to saline before induction of general anesthesia in 24 patients undergoing right lobe donor hepatectomy. PVB was associated with a 50% decrease in opioid consumption, a significant increase in time to rescue analgesia, and decreased PONV.

In a case report, Ho et al146 described the use of a right thoracic paravertebral catheter for hepatectomy and cite their preference for this technique over epidural analgesia in patients who have or may develop a coagulopathy.

In an additional case report, Hall and Leach147 illustrated the use of continuous paravertebral analgesia for the pain associated with a traumatic liver fracture managed conservatively.

In a feasibility study performed by Culp et al,148 PVB at T8 to T9 and T9 to T10 under fluoroscopic guidance in combination with intravenous sedation was successfully performed for 10 consecutive patients undergoing biliary drainage procedures. This group also performed multilevel PVB injection with 15 mL of 0.25% bupivacaine at T7 and T9 for analgesia after liver mass radiofrequency ablation, with no additional analgesic requirements. Sensory blockade of the T4 to T12 dermatomes was reported.149

Prostate Surgery

A retrospective study of 100 patients undergoing radical retropubic prostatectomy under general anesthesia compared 50 patients (group 1) receiving local anesthetic wound infiltration with 30 mL of 0.25% bupivacaine, intravenous ketorolac, and opioids for postoperative pain management with 50 patients (group 2) who had received an amended analgesia protocol to include, in addition to the preexisting protocol, the use of preoperative, multiple-injection PVB at T10 to T12 bilaterally and a single preoperative oral dose of a COX-2 inhibitor. The study suggested that the analgesic protocol of group 2 might significantly reduce intraoperative and postoperative opioid requirements and reduce hospital length of stay.

Labor Analgesia for Obstetrics

Bilateral PVB has been used to provide analgesia for laboring obstetrical patients in whom epidural anesthesia is impossible or contraindicated. In a case series, Nair and Henry150 described using bilateral T11 and T12 PVB to obtain labor analgesia for patients with coagulation abnormalities (prolonged bleeding time, syndrome of hemolysis, elevated liver enzymes, and low platelets) and lumbar spinal anomalies (neuroblastoma, spina bifida occulta). Suelto151 underlines the limitation of this technique in providing analgesia for the second stage of labor as well as for cesarean section.

Chronic Pain

Several authors have published descriptions of the use of PVB in the management of chronic pain. Kirvela and Antila152 provided retrospective information on 32 patients having 281 PVBs for chronic chest wall pain. Results were similar for those patients with post-thoracotomy chest pain and those with postmastectomy pain. Among the thoracic surgical patients, immediate pain relief occurred in 99% of cases; however, prolonged pain relief was rare. Fifty-eight percent were pain-free at 1 month, 30% at 2 months, 8% at 4 months, and only 3% at 5 months. Among the breast cancer patients, 88% were pain-free for less than 1 month and only 6% were pain-free for more than 5 months. Antila and Kirvela 153 observed similar results in a population of patients who had chronic thoracic pain related to malignant disease with poor prognosis. PVB may offer a useful treatment for thoracic myofascial pain syndrome refractory to traditional therapeutic approaches.154 Limited data exist on the use of PVB for pain relief from acute herpes zoster155 and post-herpetic neuralgia.156

PVB is a technique that offers dense, long-lasting, segmental anesthesia and analgesia. This block takes advantage of the anatomical structure and contents of the PVS. Local anesthetic is delivered at discrete locations to achieve site-specific block of both the dorsal and ventral rami as well as the sympathetic chain. In contrast to neuraxial techniques, PVB offers the flexibility of either unilateral or bilateral blockade and avoids many of the side effects associated with centrally administered local anesthetics.40,69,101 The density of the achieved block is highlighted in the profound analgesia that results and in the mitigation of the surgical stress response.62 Additionally, preoperative PVB may reduce the incidence of chronic pain after breast cancer surgery60 and reduce the incidence of post-thoracotomy neuralgia.61 Further prospective investigation is necessary to determine whether PVB for breast cancer surgery may reduce the risk of cancer recurrence or metastasis.70 By peripherally targeting individual nerve roots or segmental innervation, the technique has wide-scale applicability for nearly all procedures of the thorax and abdomen as well as the lower extremities. Despite these advantages, the greatest impediments to wider scale adoption of this technique are the low but consistent failure rate of up to 10%73 and the concern of pneumothorax.

It has been most extensively studied after thoracotomy, demonstrating profound block, excellent analgesia, opioid sparing, and reduction in opioid- and epidural-related side effects.63-67 In addition, PVB offers better preservation of pulmonary function and a reduced incidence of postoperative pulmonary complications.63-67 The magnitude of these outcome differences suggests that PVB is underutilized, particularly in procedures such as thoracic surgery where iatrogenic pneumothorax is of limited concern.

PVB has also been validated as an effective sole intraoperative anesthetic or as an analgesic adjunct for breast surgery43,45,46,68 and inguinal hernia repair,114,115,119,120 2 of the most commonly performed outpatient surgical procedures. When used for ambulatory surgery, PVB facilitates recovery and has the potential to hasten discharge and reduce costs. Efficacious results have also been produced for procedures as diverse as cardiac surgery,137,138 cholecystectomy,62,87,133 and labor.150 In short, the ability to provide site-specific segmental anesthesia makes PVB one of the most useful and broadly applicable peripheral nerve block techniques.

• Cheema S, Richardson J, McGurgan P. Factors affecting the spread of bupivacaine in the adult thoracic paravertebral space. Anaesthesia. 2003;58:684-687.
• Eason MJ, Wyatt R. Paravertebral thoracic block-a reappraisal. Anaesthesia. 1979;34:638-642.
• Karmakar MK. Thoracic paravertebral block. Anesthesiology. 2001;95:771-780.
• Karmakar MK, Ho AM, Law BK, Wong ASY, Shafer SL, Gin T. Arterial and venous pharmacokinetics of ropivacaine with and without epinephrine after thoracic paravertebral block. Anesthesiology. 2005;103:704-711.
• Naja Z, Lonnqvist PA. Somatic paravertebral nerve blockade. Incidence of failed block and complications. Anaesthesia. 2001;56:1184-1188.
• Naja MZ, Ziade MF, Lonnqvist PA. Nerve-stimulator guided paravertebral blockade vs. general anaesthesia for breast surgery: a prospective randomized trial. Eur J Anaesthesiol. 2003;20:897-903.
• Richardson J, Sabanathan S, Jones J, Shah RD, Cheema S, Mearns AJ. A prospective, randomized comparison of preoperative and continuous balanced epidural or paravertebral bupivacaine on post-thoracotomy pain, pulmonary function and stress responses. Br J Anaesth. 1999;83:387-392.