Peripheral Nerve Blocks: A Color Atlas, 3rd Edition

31. Continuous Thoracic Paravertebral Block

A. Classic Approach

Bruce Ben-David

Patient Position: Continuous thoracic paravertebral block (TPVB) is performed most easily with the patient in the sitting position with feet dangling over the side of the bed. It is also possible to perform the block in the lateral or prone position.

Indications: (See Table 31-1.)

Needle Size and Catheter: The use of a Tuohy needle is preferred because its blunt, rounded tip provides a distinct “pop” on penetrating the costotransverse ligament and may diminish the chance of perforating parietal pleura. For continuous TPVB use an 18-gauge Tuohy needle with graduated markings and 20-gauge polyamide closed-tip, multiport catheter. More flexible epidural catheters are not recommended as they will be difficult to insert.

Volume and Infusion Rate: An initial injection through the needle of 5 mL of ropivacaine 0.5% to help distend the paravertebral space and facilitate catheter insertion. This is followed by the injection of an additional 10 mL through the catheter. After the surgery, the catheter is infused with 0.2% ropivacaine at a rate of 7 to 10 mL/hr.

Anatomy and Landmarks: The thoracic paravertebral space is a triangular space whose boundaries include anteriorly the parietal pleura, posteriorly the superior costotransverse ligament, and medially the vertebral body, intervertebral disc, and intervertebral neural foramen. The apex of the triangle laterally is continuous with the intercostal space. The space is bisected by the very thin endothoracic fascia which creates two “compartments.” The anterior compartment contains the sympathetic chain, and the posterior compartment contains the intercostal nerve, dorsal ramus, intercostal blood vessels, and rami communicantes. Spinal nerves in the paravertebral space are relatively devoid of fascial covering making them uniquely and exceptionally sensitive to local anesthetic blockade. The important anatomic surface landmark for TPVB is the relevant spinous process or processes for the desired dermatomes to be blocked. Note that the steep angulation of thoracic spinous processes brings them opposite the transverse processes of the adjacent more caudad vertebra.

Table 31-1. Indications

General Surgery

Urologic Surgery

  Hernia—ventral, inguinal


  Laparoscopic Surgery



  Ureteral Surgery


  Axillary Dissection

Hepatic Surgery

Gynecologic Surgery


  Hysterectomy with abdominal debulking


  Node dissection
  Exploratory Laparotomy


Thoracic Surgery

Plastic Surgery


Breast Reconstruction


Breast Reduction


  Tissue Flaps

  Open Heart Surgery
Minimally Invasive CABG

Pain Conditions

  Sternal Surgery

  Herpetic Neuralgia
  Rib Fractures/Thoracic Trauma
  Traumatic Liver Capsule Pain
  Intercostal Neuralgia


Approach and Technique: The relevant spinous processes are identified and marked on the skin and a point 2.5 cm lateral to the spinous process is also marked (Fig. 31-1). Disinfectant is applied in standard fashion and local anesthetic (1% lidocaine) injected at each injection point using a 1.5-cm 25-gauge needle. The Tuohy needle is advanced onto the transverse process and the depth from skin to paravertebral space marked by placing the index or third fingers on the needle shaft 1 cm from the skin (Figs. 31-2, 31-3). This will now serve as both a depth gauge and a guard against excessive insertion of the needle. The needle is then walked caudally off the inferior border of the transverse process and inserted to a depth 1 cm deeper than the transverse process—that is, to the depth allowed by prior placement of the index fingers on the needle shaft. Typically at this point one feels a confirmatory “pop” upon penetration of the costotransverse ligament. A drop of fluid is placed in the needle hub and the patient is asked to inspire deeply (Fig. 31-4). Correct placement is denoted by lack of movement of the fluid bubble. A drawing inward of the fluid indicates intrapleural needle placement, in which case the needle should be immediately withdrawn. After correct needle placement is thus confirmed, 5 mL of local anesthetic is injected through the needle. It is helpful to have an assistant inject through an extension tube as this helps avoid significant movement of the needle. Following the injection the extension tube is disconnected and the catheter inserted a depth of 3 to 5 cm beyond the tip of the needle (Fig. 31-5). The catheter is affixed in standard fashion using adhesive strips and transparent dressing (Fig. 31-6).


Figure 31-1. The skin is marked at 2.5 cm lateral to the spinous process.

Figure 31-2. The finger placement on the needle shaft will serve as both a depth gauge and a guard against excessive insertion of the needle.


1.   Depth of the adult paravertebral space from the skin ranges between 3 and 7 cm except for the morbidly obese. Surprisingly, depth correlates only weakly with height, weight, and BMI.

2.   Limit volume of injectate through the needle—the higher pressures when injecting through the needle may force the local anesthetic through the intervertebral foramen into the epidural space.

3.   Any significant resistance to injection through the needle indicates improper needle placement.

4.   When using paravertebral catheters there is no need to perform injections at multiple levels; clinically the block is extended over more dermatomes by simply increasing the local anesthetic infusion rate (Fig. 31-7).

5.   Abdominal and retroperitoneal surgery (e.g., nephrectomy) typically require placement of bilateral paravertebral catheters.

6.   The catheter is positioned with the understanding that the spray is more caudal (3 to 4 dermatomes) than cephalad (1 or 2 dermatomes).

Figure 31-3. The Tuohy needle is advanced onto the transverse process and the depth from skin to paravertebral space marked by placing the index or third fingers on the needle shaft 1 cm from the skin.

Figure 31-4. A drop of fluid is placed in the needle hub and the patient is asked to inspire deeply.



Figure 31-5. Following injection of 5 mL of local anesthetic, the extension tube is disconnected and a catheter inserted to a depth of 3 to 5 cm beyond the tip of the needle.

Figure 31-6. Catheter fixation technique.

9.   Verification of catheter placement with x-ray although interesting, is not necessary.

10.       Inadvertent placement of the catheter interpleurally is acceptable although not desirable and will provide analgesia as an interpleural block instead. In the case of thoracic surgery, an interpleural block will not provide acceptable analgesia, but the surgeon will clearly see the catheter in the chest and alert you to the need for repositioning it.

Figure 31-7. Injection of 10 mL omnipaque contrast via PVB catheter following thoracotomy demonstrates extensive caudo–cephalad paravertebral spread.

11.       Routine postoperative x-ray to rule out pneumothorax may be comforting but is not absolutely necessary. Nevertheless, one should always have an index of suspicion for this uncommon complication.

12.       In the case of trauma and spine injury, in the absence of “spine clearance,” or in the anticoagulated patient it would be preferable to use an alternative technique to continuous thoracic paravertebral blocks such as percutaneous continuous intercostal blockade or a continuous interpleural block.

13.       Avoid medial redirection of the needle as this may predispose to inadvertent neuraxial blockade.

14.       If one encounters a “wall of bone” in walking the needle caudad–cephalad in the sagittal plane then the needle is probably too medial encountering the lamina and superior-inferior articular processes. Redirect slightly laterally.

15.       Use graduated needles as this helps avoid excessive needle penetration. In addition use index fingers as depth gauge/guard.

16.       Use a 22-gauge Tuohy single-shot needle to define location and depth of the transverse process prior to use of the larger catheter insertion needle, especially in morbidly obese patients.

17.       Walk caudally, not cephalad, off the transverse process as the distance between the costotransverse ligament and the pleura is greater here. The rib, which is deep to the transverse process, usually rises more cephalad. Thus walking the needle cephalad will mean that the needle is walking off the rib, not the transverse process.

18.       Tunneling of the catheter is not necessary if proper taping technique is used.

19.       Verify needle position using the hanging droplet technique. Deep inspiration will pull the droplet in if the needle is interpleural. The droplet will bulge outward if the needle is in the paraspinal muscles.

20.       The catheter exits the needle at approximately an angle of 30°. If one encounters difficulty passing the catheter even after expanding the paravertebral space by injecting local anesthetic or saline one may try, with the Tuohy needle opening directed caudad, applying a slight cephalad torque to allow the catheter greater ease of deflecting off the pleura. One may also try rotating the needle to pass the catheter laterally. If this fails, try reinserting the needle at a steeper angle.

21.       Continue to monitor the patient after placement of the block. Neuraxial block onset and its sequelae may be delayed.

22.       On initial injection of local anesthetic into the paravertebral space patients will frequently complain of a momentary “pinch” or “cramp” in their side. This paresthesia is a good indication of proper placement.

23.       As with other continuous peripheral nerve blocks, patients may be discharged home with disposable infusion pumps.

24.       Up to 8% of patients demonstrate a vagal response during performance of the block. Therefore, a prepared syringe of an anticholinergic drug (e.g., glycopyrrolate) and a syringe with a pressor agent (e.g., ephedrine) should be immediately at hand.

Suggested Reading

Davies RG, Myles PS, Graham JM. A comparison of the analgesic efficacy and side-effects of paravertebral vs. epidural blockade for thoracotomy—a systematic review and meta-analysis of randomized trials. Br J Anaesth 2006;96:418–426.

Ganapathy S, Nielsen KC, Steele SM. Outcomes after paravertebral blocks. Int Anesthesiol Clin 2005;43:185–193.

Karmakar MK. Thoracic paravertebral block. Anesthesiology 2001;95:771–780.

Karmakar MK, Critchley LA, Ho AM, et al. Continuous thoracic paravertebral infusion of bupivacaine for pain management in patients with multiple fractured ribs. Chest 2003;123:424–431.


Lönnqvist PA. Pre-emptive analgesia with thoracic paravertebral blockade? Br J Anaesth 2005;95:727–728.

Lönnqvist PA, MacKenzie J, Soni AK, Conacher ID. Paravertebral blockade. Failure rate and complications. Anaesthesia 1995;50:813–815.

Naja Z, Lönnqvist PA. Somatic paravertebral nerve blockade: incidence of failed block and complications. Anaesthesia 2001;56:1184–1188.

Naja MZ, El-Rajab M, Al-Tannir MA, et al. Thoracic paravertebral block: influence of the number of injections. Reg Anesth Pain Med 2006;31:196–201.

Naja Z, Ziade MF, Lönnqvist PA. Bilateral paravertebral somatic nerve block for ventral hernia repair. Eur J Anaesthesiol 2002;19:197–202.

Naja MZ, Ziade MF, Lönnqvist PA. General anaesthesia combined with bilateral paravertebral blockade (T5-6) vs. general anaesthesia for laparoscopic cholecystectomy: a prospective, randomized clinical trial. Eur J Anaesth 2004;20:489–495.

Purcell-Jones G, Pither CE, Justins DM. Paravertebral somatic nerve block: a clinical, radiographic, and computed tomographic study in chronic pain patients. Anesth Analg 1989;68:32–39.

Richardson J, Cheema SP, Hawkins J, Sabanathan S. Thoracic paravertebral space location. A new method using pressure measurement. Anaesthesia 1996;51:137–139.

Richardson J, Lönnqvist PA. Thoracic paravertebral block. Br J Anaesth 1998;81:230–238.

Richardson J, Sabanathan A, Jones J, et al. A prospective, randomized comparison of preoperative and continuous balanced epidural or paravertebral bupivacaine on post-thoracotomy pain, pulmonary function and stress response. Br J Anaesth 1999;83:387–392.

Vogt A, Stieger DS, Theurillat C, Curotolo M. Single-injection thoracic paravertebral block for postoperative pain treatment after thoracoscopic surgery. Br J Anaesth 2005;95:816–821.

B. Intercostal Approach

David Burns

Patient Position: For blockade above T7, the patient is best positioned sitting with the arms dangling between the knees to bring the scapulae apart, thereby allowing better access to the upper thoracic ribs.

Indications: The indications are similar to those for the classic paravertebral block. These include all surgical procedures of the chest and abdomen including thoracic surgery—both open and video assisted, conventional and minimally invasive cardiac surgery, breast surgery, plastic reconstructive or cosmetic surgery, and pectus repair. Urologic surgical indications include renal, adrenal, ureteral, bladder, and prostatic surgery. General surgical indications include both open and laparoscopic surgeries such as bariatric surgery, cholecystectomy, herniorrhaphy, pancreatic and liver surgery, and bowel surgery. Gynecologic surgical indications include hysterectomy, myomectomy, fallopian tube, and ovarian surgery. Nonsurgical indications include liver capsule pain, rib fractures, acute postherpetic neuralgia, and chronic pain management for both benign and malignant neuralgias. The surgical site will of course dictate whether unilateral or bilateral neural blockade is needed. For retroperitoneal and abdominal surgeries, typically bilateral blockade is necessary. This technique may be advantageous in situations where the classic technique would be contraindicated (e.g., coagulopathies, thrombocytopenia, anticoagulation therapies, spinal trauma, or previous spinal surgery).

Needle Size and Catheter: An 18-gauge, 2-inch Tuohy needle and 20-gauge closed-end, multiorifice, stiff plastic catheter. For obese patients a 4-inch Tuohy needle may be needed.

Anatomic Landmarks: The intercostal nerve lies with the artery and vein under the rib in the subcostal groove between the internal and innermost intercostal muscles. The puncture site is at the angle of the rib about 8 cm lateral to the midline. The inferior border of the scapulae is marked bilaterally. The line connecting the two inferior angles usually runs through the vertebral body of T7. The desired rib is then delineated. The lower third of the rib 8 cm from the midline is the entry site.

Approach and Technique: The skin is prepped and draped in a sterile fashion. Thereafter the skin and subcutaneous tissue under the entry site is anesthetized with 1% lidocaine until the rib is contacted and the periosteum is generously anesthetized. The Tuohy needle is then advanced directly inward to contact the lower third of the rib. The needle is then angled laterally at 45° to 60° with the bevel facing the spinous process and is gently walked off the inferior border of the rib while maintaining this orientation. Once under the rib the needle is advanced 5 to 6 mm to lie within the subcostal groove. After negative aspiration of heme or air and a negative fluid drop aspiration test, 5 mL of local anesthetic is injected followed by passage of the catheter toward the midline. The catheter is left at the combined distance of 8 cm plus the depth of the needle such that the tip is left at the ipsilateral paravertebral space. After securing the catheter, bolus another 10 mL of the local anesthetic to establish adequate spread. Postoperative infusions are typically run at 10 to 12 mL/hr (Figs. 31-8, 31-9 and 31-10).


Figure 31-8. Anatomic landmarks.

Figure 31-9. Needle orientation.


1.   For blockade below T7, the patient can be positioned either sitting or laterally.

2.   0.5 to 1 mg of midazolam with 25 to 50 µg of fentanyl may improve patient comfort, but in most patients, sedation is not necessary.

3.   The shorter needle is used to allow more precise control over the needle, as the usual depth to the subcostal groove is 2 to 4 cm.

4.   To facilitate walking under the rib without having to drop the angle of the needle (and by so doing miss the neurovascular bundle tucked up along the rib in the subcostal groove), the skin should be lifted cephalad somewhat before entering the skin thereby allowing it to easily come with the needle as it is walked off instead of impeding the movement.

5.   To confirm extrapleural placement after advancing the needle into the subcostal groove and before threading the catheter, the stylet should be removed and a drop of fluid placed within the needle and the patient is asked to inhale. If the fluid is entrained the needle should be pulled back 1 to 2 mm and the test repeated.

6.   Upon passing the catheter to the paravertebral space there should be a small amount of resistance. If there is no resistance, the catheter may actually be interpleural. See Chapter 30 on interpleural catheters for their management.

Figure 31-10. Threading the catheter.

7.   0.5% ropivacaine is typically used for the initial bolus at placement and 0.2% ropivacaine at 10 mL/hr postoperative infusion. The nursing staff also has orders for 5 to 6 mL boluses available q1 hr for breakthrough pain.

Suggested Readings

Burns DA, Bruce Ben-David B, Chelly JE, Greensmith JE. Continuous percutaneous intercostal block: an alternative approach to continuous paravertebral blockade. Anesth Analg 2007; in press.

Detterbeck F. Efficacy of methods of intercostal nerve blockade for pain relief after thoracotomy. Ann Thorac Surg 2005;80:1550–1559.

Eng J, Sabanathan S. Site of action of continuous extrapleural intercostal nerve block. Ann Thorac Surg 1991;51:387–389.

Mowbray A, Wong KKS, Murray JM. Intercostal catheterization: An alternative approach to the paravertebral space. Anaesthesia 1987;42:958–961.

Murphy DF. Continuous intercostal nerve blockade: An anatomical study to elucidate its mode of action. Br J Anaesth 1984;56:627–629.

Richardson J, Sabanathan S, Jones J, et al. 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.

Savage C, McQuitly C, Wang DF, Zwischenberger JB. Postthoracotomy pain management. Chest Surg Clin N Am 2002;12:251–263.

Tobias MD, Ferrante FM. Complications of paravertebral, intercostal, and interpleural nerve blocks. In Finucaine BT, ed. Complications of regional anesthesia. New York: Churchill Livingstone, 1999;77–93.