Twenty healthy volunteers were seated prone with a cushion placed under the abdomen to minimize lumbar lordosis. A curved array transducer (3,5C40+; Siemens; operated with 4 MHz) was applied, and sonograms were obtained as described. The psoas muscle was traced in the transverse sonograms of L3-4 and L4-5 levels on both sides, and measurements of its cross-sectional areas were computed and compared between examiners.
Twenty-one healthy volunteers were examined at L2-3, L3-4, and L4-5. The volunteers were allocated to three groups (Table 1): normal weight (n = 13; body mass index [BMI] 18.5–24.9 kg/m2), overweight (n = 5; BMI 25.0–29.9 kg/m2), and obese (n = 3; BMI >30.0 kg/m2). Posterior paravertebral sonography was considered feasible when the psoas muscle, the quadratus lumborum muscle, and the erector spinae could be clearly delineated in transverse sonograms at each level. Skin-plexus distance measurements were performed by one of the examiners at L2-3, L3-4, and L4-5. Skin-plexus distances were measured between the junction of the posterior third and the anterior two-thirds of the psoas muscle and the skin surface (estimated position of the lumbar plexus) (10). The measurements were computed in the center of the transverse sonograms parallel to the axis of the US beam to minimize measurement errors (11).
Finally, skin-plexus distance measurements were performed equally by means of computed tomography (CT) in 10 embalmed cadavers to obtain reference values. The cadavers were also seated in the prone position to avoid a decrease of the skin-plexus distances caused by body-weight pressure.
Interobserver reliability in the pilot study was determined by using Kendall’s coefficient of concordance, W, to compare the psoas muscle cross-sectional areas measured by the two examiners. The Kruskal-Wallis test was used to reveal significant skin-plexus distance differences among the L2-3, L3-4, and L4-5 levels, as well as skin-plexus differences among the three BMI groups. The Mann-Whitney U-test was performed in case of significant skin-plexus distance differences between the analyzed levels as well as BMI groups (revealed by the Kruskal-Wallis test). To analyze correlations between the measured skin-plexus distances and BMIs, Spearman’s coefficient of correlation was applied. P values <0.05 were considered statistically significant. Skin-plexus distances are presented as median ± sd.
In all 20 volunteers (10 men, 10 women), median age 24 yr (range, 19–43 yr), height 177 cm (range, 164–192 cm), weight 71 kg (range, 50–100 kg), and BMI 22 kg/m2 (range, 18–27kg/m2), posterior paravertebral sonography was performed successfully. The psoas muscle, the quadratus lumborum muscle, and the erector spinae could be traced reproducibly at all examined levels (n = 80). Kendall’s W was 0.9 (P < 0.001).
Twenty-one volunteers (13 men, 8 women), divided into three groups (Table 1), had 126 lumbar levels (L2-3, L3-4, and L4-5, right and left side) examined. Posterior paravertebral sonography was feasible at 112 of 126 lumbar levels (Table 2). In the Overweight and Obese groups, sonography was impossible in one volunteer at all levels (left and right side) and in another at the level of L4-5 on one side (Table 2). For both examiners, sonography was unfeasible in the same volunteers at equal levels.
Although there was an increase of the skin-plexus distances from L2-3 to L4-5, the Kruskal-Wallis test revealed no significant differences among L2-3, L3-4, and L4-5 (P = 0.42). The skin-plexus distances were 5.5 ± 1.4 cm at L2-3, 5.5 ± 1.4 cm at L3-4, and 5.8 ± 1.3 cm at L4-5. However, skin-plexus distance differences among the Normal Weight, Overweight, and Obese groups (Table 2) were significant (P < 0.001). Further, Spearman’s coefficient of correlation showed a significant positive correlation of 0.9 (P = 0.01) between the skin-plexus distances and the BMIs.
In 10 embalmed cadavers (3 male, 7 female) skin-plexus distance measurements were computed at 56 lumbar levels (L2-3, n = 18; L3-4, n = 20; L4-5, n = 18) by means of CT. Their median age at death was 81 yr (range, 51–88 yr) and height, 164 cm (range, 148–178 cm). Seven of 10 cadavers had normal body habitus, and 3 of 10 were obese. Skin-plexus distances increased from L2-3 to L4-5, similar to the skin-plexus distances obtained by sonography, but they showed higher median values: 7.0 ± 1.6 cm at L2-3, 7.3 ± 1.5 cm at L3-4, and 7.9 ± 1.3 cm at L4-5. The Kruskal-Wallis test showed no significant skin-plexus distance differences among the three levels (P = 0.1).
This investigation is the first dealing with a detailed and reliable description of the sonographic appearance of the lumbar paravertebral region. Posterior paravertebral sonography is reliable and accurate, as shown by the results of the interobserver reliability test.
The main study revealed that in 19 of 21 examined volunteers, posterior paravertebral sonography was feasible for both examiners. In 2 of 21 volunteers, posterior paravertebral sonography failed at all examined levels because of obesity. The quality of the obtained sonograms did not allow a sharp delineation of the psoas muscle and the adjacent structures. There are two reasons that sonography may fail in obese subjects: first, the applied transducer provides too low tissue penetration, and second, thick subcutaneous fat tissue causes heavy reflections because of US dispersion in dense tissues. Further, the spread of US may be disturbed in men at L4-5 by tall iliac crests, which are an anthropometric feature of male pelvises (Fig. 1). For that reason, both examiners were unable to perform sonography at L4-5 in 2 of 13 male volunteers, each on one side. These results indicate that posterior paravertebral sonography can be reliably performed in normal weight and in the majority of overweight and obese individuals. Occasionally, a reliable sonographic examination at L4-5 may be unfeasible in men because of the obstructing iliac crests.
Usually, the lower poles of the kidneys reach the level of L3, but during deep inspiration they may descend to reach the level of L3-4, appearing as hypoechoic, oval-shaped structures in the posterior transverse sonograms of L3-4. Distinguishing between the kidneys and the typical echotexture of the psoas muscle [hyperechoic striations on an echo-poor background (12)] was feasible in all successfully examined volunteers. Aida et al. (9) reported two cases of renal subcapsular hematoma caused by lumbar plexus blockade at L3 and stated that a posterior approach to the lumbar plexus must be performed at L4-5 to avoid renal injury. The use of real-time US guidance for approaches at L2-3 and L3-4 should help to avoid such complications by visualizing the structures at risk.
This study revealed that it is necessary to apply curved-array transducers operating at lower frequencies (4–5 MHz) because they provide appropriate tissue penetration and image size but less spatial resolution. Therefore, it was not possible to distinguish between peripheral nerves (13,14) as parts of the lumbar plexus and tendon fibers (which appear as hyperechoic striations, similar to peripheral nerves) within the psoas muscle (12). For a reliable and accurate delineation of the latter, the application of linear array transducers (>7.5 MHz) is recommended (15). Nevertheless, Koyama et al. (16) reported the depiction of parts of the lumbar plexus within the psoas muscle by using a 3.5-MHz curved array transducer.
Consequently, skin-plexus distance-measurements were made indirectly with the use of a reference point (junction of the anterior two-thirds and the posterior third of the psoas muscle in its anteroposterior diameter) that was estimated to be the approximate position of the lumbar plexus (10). The lumbar plexus is situated within the posterior part of the psoas muscle at all lumbar levels (Fig. 2) (8,10,17–20). The median skin-plexus distances measured with US are smaller than those measured with CT at all examined levels. The most likely explanation for this decrease is the pressure of the transducer against the skin (compression of the subcutaneous tissue and paraspinal muscles) that is mandatory for obtaining optimal sonograms. We estimated this reduction to be approximately 1–2 cm (depending on the dimension of subcutaneous tissue). Although the obtained skin-plexus distances did not exactly represent the real values for that reason, we considered them as guidelines for further applications of posterior paravertebral sonography. The increase of the skin-plexus distances from L2-3 to L4-5 (not significant) revealed by both CT and US measurements can be explained with the topographical position of the psoas muscle (as the muscle courses caudad, it moves anteriorly) (21). Nevertheless, skin-plexus distances showed significant differences among the Normal Weight, Overweight, and Obese groups (see Table 1 for median values); this could be confirmed by the significant positive correlation between the BMIs and skin-plexus distances.
In conclusion, we demonstrated detailed and reliable visualization of the lumbar paravertebral region by means of posterior paravertebral sonography. Nevertheless, we were unable to delineate the lumbar plexus. With the use of US, particularly approaches at L2-3 and L3-4 should be feasible without any complications that occur because of “blind” approaches.
1. Kapral S, Krafft P, Eibenberger K, et al. Ultrasound-guided supraclavicular approach for regional anesthesia of the brachial plexus. Anesth Analg 1994; 78: 507–13.
2. Marhofer P, Schrögendorfer K, Koinig H, et al. Ultrasonographic guidance improves sensory block and onset time of three-in-one blocks. Anesth Analg 1997; 85: 854–7.
3. Marhofer P, Schrögendorfer K, Wallner T, et al. Ultrasonographic guidance reduces the amount of local anesthetic for 3-in-1 blocks. Reg Anesth 1998; 23: 584–8.
4. Kapral S, Krafft P, Gosch M, et al. Ultrasound imaging for stellate ganglion block: direct visualization of puncture site and local anesthetic spread. Reg Anesth 1995; 20: 323–8.
5. Chayen D, Nathan H, Chayen M. The psoas compartment block. Anesthesiology 1976; 45: 95–9.
6. Winnie AP, Ramamurthy S, Durrani Z, Radonjic R. Plexus blocks for lower extremity surgery. Anesth Rev 1974; 1: 11–6.
7. Parkinson SK, Mueller JB, Little WL, Bailey SL. Extent of blockade with various approaches to the lumbar plexus. Anesth Analg 1989; 68: 243–8.
8. Hanna MH, Peat SJ, d’Costa F. Lumbar plexus block: an anatomical study. Anaesthesia 1993; 48: 675–8.
9. Aida S, Takahashi H, Shimoji K. Renal subcapsular hematoma after lumbar plexus block. Anesthesiology 1996; 84: 452–5.
10. Farny J, Drolet P, Girard M. Anatomy of the posterior approach to the lumbar plexus block. Can J Anaesth 1994; 41: 480–5.
11. Burns PN, Waldroup L, Pinkney MN. Limitations of ultrasound imaging measurements. In: Goldberg BB, Kurtz AB, eds. Atlas of ultrasound measurements. Chicago: Year-Book Medical Publishers, 1990: 4–18.
12. King AD, Hine AL, McDonald C, Abrahams P. The ultrasound appearance of the normal psoas. Clin Radiol 1993; 48: 316–8.
13. Fornage BD. Peripheral nerves of the extremities: imaging with US. Radiology 1988; 167: 179–82.
14. Silvestri E, Martinoli C, Derchi LE, et al. Echotexture of peripheral nerves: correlation between US and histologic findings and criteria to differentiate tendons. Radiology 1995; 197: 291–6.
15. Steiner E, Nasel C. Sonography of peripheral nerves: basic principles. Acta Anaesthesiol Scand 1998; 42: 46–8.
16. Koyama H, Murakami K, Suzuki T, Suzaki K. Phenol block for hip flexor muscle spasticity under ultrasonic monitoring. Arch Phys Med Rehabil 1992; 73: 1040–3.
17. Berry M, Bannister LH, Standring SM. Nervous system. In: Williams PL, Bannister LH, Berry M, et al., eds. Gray’s anatomy. 38th ed. New York: Churchill Livingstone, 1995: 1277–82.
18. Rickenbacher J, Landolt AM, Theiler K. Rücken. In: Lang J, Wachsmuth W, eds. Praktische Anatomie. Berlin: Heidelberg, 1982: 69–71.
19. Black SM, Chambers WA, eds. Essential anatomy for anesthesia. New York: Churchill Livingstone, 1997:136–9.
20. Dietemann JL, Sick H, Wolfram-Gabel R, et al. Anatomy and computed tomography of the normal lumbosacral plexus. Neuroradiology 1987; 29: 58–68.
© 2001 International Anesthesia Research Society
21. Reid JG, Livingston LA, Pearsall DJ. The geometry of the psoas muscle as determined by magnetic resonance imaging. Arch Phys Med Rehabil 1994; 75: 703–8.