The success rate and spread of thoracic paravertebral block (TPVB) are variable and difficult to predict. It is now recommended that an ultrasound guidance technique should replace the traditional landmark technique. The objective was to compare anatomical outcomes of both techniques on cadavers.

A landmark technique (loss of resistance technique [LOR]) and a USG technique (three approaches: sagittal, transversal in-plane, transverse out-of-plane) were performed on 27 thawed non-embalmed cadavers. Each of the four approaches was performed in each body (T3–T5 and T9-T11 × right and left). A coloured solution (13 mL, saline 0.9%) was injected in the targeted thoracic paravertebral space (TPVS). A successful thoracic paravertebral injection (TPVI) was defined by the presence of dye in at least one TPVS during anatomical dissection.

In 104 TPVIs analysed, the overall success rate was 78%. Factors associated with success were: USG versus LOR technique (85% vs. 52%, P < 0.0007), sagittal versus both transversal approaches (93%/81%/83%, P < 0.0007) and right side (86% vs. 66%). The median spread was 2 TPVS (min – max 1–5) with a median cephalad-caudal spread of 5 cm (min - max 1-18). By multivariate analysis, the sagittal approach was an independent factor of success (OR 2.75). Dye spread and pleural entry were influenced by neither the approach nor the site of injection.

Paravertebral spread of TPVI is variable. USG technique has higher anatomical success rates than the LOR technique, the sagittal USG approach being the most successful.