Published online before print March 14, 2013, doi: 10.3174/ajnr.A3544
AJNR 2013 34: E45

T.M. Ryan^a, E.C. Kavanagh^a and P.J. MacMahon^a
aDepartment of Radiology
Mater Misericordiae University Hospital
Dublin, Ireland

We read with great interest the recent article by Miller et al¹ regarding lateral decubitus positioning for cervical nerve root block by using CT image guidance to minimize effective radiation dose and procedural time.

The technique of cervical foraminal injection is outlined in the “Materials and Methods ” section. The authors state, “A slow 1-mL injection of iohexol diluted in 1-mL 1% lidocaine was used in all cases to identify inadvertent direct vessel puncture.” We ask the authors in how many cases did they identify inadvertent direct vessel puncture with the aid of contrast?

From the >1000 procedures we have performed by using the technique outlined in this article, we have never identified vessel puncture with contrast CT. The reason for this is likely to be 2-fold: First, should the contrast be injected intravascularly, it is likely to be washed away by the time CT is performed. Second, it is possible that the given vessel enters the cord at a different level and is therefore not imaged.

In this sense, we believe that contrast administration gives the radiologist a false sense of security. Real-time imaging such as digital subtraction angiography would be needed to reliably exclude inadvertent direct vessel puncture. However, we believe that such measures are also unnecessary on the basis of current best evidence in the literature (case series,² animal experimentation,³ and in vitro microscopy⁴). Dexamethasone sodium phosphate is likely safe if inadvertently injected intravascularly.

On this basis, we propose that this procedure can be made even safer in 2 ways: By eliminating the administration of contrast, the possibility of an adverse reaction is avoided. Furthermore, the number of imaging series could be reduced to give an even lower effective radiation dose and a shorter procedural time.

References

1. Miller TS, Fruauff K, Farinhas J, et al. Lateral decubitus positioning for cervical nerve root block using CT image guidance minimizes effective radiation dose and procedural time. AJNR Am J Neuroradiol 2013;34:23–28 » Abstract/FREE Full Text
2. Scanlon GC, Moeller-Bertram T, Romanowsky SM, et al. Cervical transforaminal epidural steroid injections: more dangerous than we think? Spine 2007;32:1249–56 » CrossRef » Medline
3. Okubadejo GO, Talcott MR, Schmidt RE, et al. Perils of intravascular methylprednisolone injection into the vertebral artery: an animal study. J Bone Joint Surg Am 2008;90:1932–38 » CrossRef
4. Derby R, Lee SH, Date ES, et al. Size and aggregation of corticosteroids used for epidural injections. Pain Med 2008;9:227–34 » CrossRef » Medline

Reply

Published online before print March 14, 2013, doi: 10.3174/ajnr.A3546
AJNR 2013 34: E46

T. Miller^a and A. Brook^a
aMontefiore Medical Center
Department of Radiology and Neuroradiology
Albert Einstein College of Medicine
Bronx, New York

We appreciate the comments submitted by Timothy Martin Ryan, Eoin C. Kavanagh, and Peter J. MacMahon regarding the use of contrast before CT-guided cervical nerve root block. The group certainly has extensive experience with the procedure. In our cases, we did not show an instance of inadvertent direct vessel contrast uptake with the contrast injection. However, we were able to visualize the extent of foraminal or epidural contrast with each injection.

The intent to decrease procedural time and radiation dose by skipping a step is valid, but eliminating the contrast step would be of little incremental value. The accepted safety profile of nonparticulate steroid formulations is growing, and we are believers. We agree that the contrast injection is unlikely to demonstrate intravascular injections. We have begun to rely on the contrast injection to document the location of the injectant. If we see poor perineural or epidural contrast, the needle can be adjusted to allow better medication deposition. Regardless, this process also allows us to document injectant localization.

Although the concern over contrast reaction is plausible, with 1-mL injections, we have yet to elicit a reaction. We concede that it is possible to generate a reaction even with such small volumes; however, we believe that the benefit outweighs the minimal risk. Contrast can be omitted for patients with known contrast allergy, pretreatment can be used, or gadolinium-based agents (off-label) may be substituted for iodinated contrast.

Lidocaine is much more useful for determining intravascular medication injection. Untoward patient reaction signals an inadvertent vascular injection. In such cases, the procedure can be terminated or the needle can be adjusted and the injection can be repeated. This choice is dependent on the patient’s reaction and recovery.

In summary, we agree with the notion that nonparticulates should be the standard of care and that imaging after contrast in CT is of limited utility in demonstrating intravascular contrast. The contrast documents injectant localization and allows us to adjust the injection to maximize localization of the medication. The potential for contrast reaction is minimal, and we believe that another benefit of contrast injection is to document injectant flow. Therefore, we are not yet ready to abandon contrast injections for small incremental reductions in potential contrast reaction, radiation exposure, or procedure time.