We read with interest the article by Loubert et al.  in which the authors could not demonstrate any benefit of using epidural volume extension (EVE) with 5 ml saline 0.9% as part of a combined spinal-epidural technique for women undergoing elective caesarean delivery.
Firstly, we believe that the 5 ml used in the study was a relatively small volume. It has been demonstrated that 10 ml epidural saline generates a higher level of analgesia with hyperbaric agents . Using hyperbaric contrast medium, Takiguchi et al. demonstrated three segments of myelographic cephalad extension after EVE using 20 ml, with the patient in a 45° reverse Trendelenburg position , although some of this extension was likely to be attributable to the diffusion characteristics of contrast medium . Doganci et al. demonstrated a ceiling effect using EVE with 15 ml after normobaric bupivacaine administration .
Secondly, intrathecal fentanyl (25 μg) used in the study might have confounded the results. Although it is clear that intrathecal opioids augment spinal anaesthesia, currently available data have not demonstrated their dose responsiveness . The addition of fentanyl could have affected the results in a study such as this, with its delicate design.
Thirdly, Mardirosoff et al. demonstrated that the effect of EVE is time sensitive . While Loubert et al. administered EVE immediately after the spinal bupivacaine injection, patients in the EVE group were kept in a sitting position during the injection which resulted in a longer procedural time (average 12.9 s), with a much wider SD. We suspect this short delay could nevertheless have allowed enough time for part of the intrathecal hyperbaric bupivacaine to travel caudally, passing the peak of lumbar lordosis and therefore diminishing the EVE effect.
In summary, we believe that the effect of EVE is dependent on multiple variables, including the patient’s position, the timing of the injection, the volume of epidural saline, the baricity of the intrathecal medication, adjuvant medication and patient perception, which may explain the lack of effect found by Loubert et al.