Free
Correspondence  |   February 2005
Can Epidurography Help to Predict the Extent of Epidural Blockade?
Author Affiliations & Notes
  • Emmanuel Marret, M.D.
    *
  • * Tenon University Hospital, Paris VI University, Assistance Publique-Hôpitaux de Paris, Paris, France.
Article Information
Correspondence
Correspondence   |   February 2005
Can Epidurography Help to Predict the Extent of Epidural Blockade?
Anesthesiology 2 2005, Vol.102, 478. doi:
Anesthesiology 2 2005, Vol.102, 478. doi:
To the Editor:—
We read the study from Yokoyama et al.  1 that sought to predict the extent of epidural blockade from the distribution of contrast medium injected into the epidural space. This study documents a linear correlation between radiographic and analgesic spread in 90 patients. The authors conclude that epidurography helps to predict the exact dermatomal distribution of analgesic block. Although it is not surprising that the distribution of radiographic spread is well correlated with the spread of analgesic block, this does not mean that it is a reliable technique and that the two methods agree. The comparison could be better performed using the Bland and Altman method allowing determination of the bias and the accuracy of measurement.2 We used individual data reported in Yokoyama et al.  ’s1 figure 4 in the 16 patients who received 5-ml and 10-ml epidural injections to calculate the correlation (fig. 1) and also to assess the bias and limits of agreement (fig. 2). The correlation between the methods was confirmed (r = 0.93, P  < 0.001; fig. 1) despite an inadequate agreement (bias = 1.1, SD = 1.0). If we consider the limits of agreement, it appears that from one patient to another, the difference between radiographic and analgesic spread may range from less than one segment to three segments. In these 16 patients, the mean radiographic spread was 6.7 ± 2.1 segments after the 5-ml injections and 9.4 ± 2.7 segments after the 10-ml injections. Consequently, the radiographic spread may overestimate the extension of epidural blockade, at most, of 44% and 32%, respectively. Moreover, the Bland and Altman plot shows that 1) when a difference exists between the two methods, the radiographic spread overestimates the analgesic spread in all cases except one and 2) the scatter of the difference seems to increase with the increase of analgesic and radiographic extension. The injection of a volume of local anesthetic or contrast medium greater than 5 ml or 10 ml, corresponding to the current clinical practice, may thus result in a greater discrepancy between the clinical and the radiographic evaluation of the block. Epidurography is therefore a useful tool to check the adequate position of the catheter but is less reliable to predict the extent of the block.
Fig. 1. Correlation between radiographic and analgesic spread. From individual data reported in fig. 4  1 . 
Fig. 1. Correlation between radiographic and analgesic spread. From individual data reported in fig. 4  1. 
Fig. 1. Correlation between radiographic and analgesic spread. From individual data reported in fig. 4  1 . 
×
Fig. 2. Agreement between radiographic spread and analgesic spread using Bland and Altman plot. The mean bias was 1.12, with limits of agreement of −0.85 and 3.10. From individual data reported in fig. 4  1 . 
Fig. 2. Agreement between radiographic spread and analgesic spread using Bland and Altman plot. The mean bias was 1.12, with limits of agreement of −0.85 and 3.10. From individual data reported in fig. 4  1. 
Fig. 2. Agreement between radiographic spread and analgesic spread using Bland and Altman plot. The mean bias was 1.12, with limits of agreement of −0.85 and 3.10. From individual data reported in fig. 4  1 . 
×
* Tenon University Hospital, Paris VI University, Assistance Publique-Hôpitaux de Paris, Paris, France.
References
Yokoyama M, Hanazaki M, Fujii H, Mizobuchi S, Nakatsuka H, Takahashi T, Matsumi M, Takeuchi M, Morita K: Correlation between the distribution of contrast medium and the extent of blockade during epidural anesthesia. Anesthesiology 2004; 100:1504–10Yokoyama, M Hanazaki, M Fujii, H Mizobuchi, S Nakatsuka, H Takahashi, T Matsumi, M Takeuchi, M Morita, K
Bland JM, Altman DG: Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1:307–10Bland, JM Altman, DG
Fig. 1. Correlation between radiographic and analgesic spread. From individual data reported in fig. 4  1 . 
Fig. 1. Correlation between radiographic and analgesic spread. From individual data reported in fig. 4  1. 
Fig. 1. Correlation between radiographic and analgesic spread. From individual data reported in fig. 4  1 . 
×
Fig. 2. Agreement between radiographic spread and analgesic spread using Bland and Altman plot. The mean bias was 1.12, with limits of agreement of −0.85 and 3.10. From individual data reported in fig. 4  1 . 
Fig. 2. Agreement between radiographic spread and analgesic spread using Bland and Altman plot. The mean bias was 1.12, with limits of agreement of −0.85 and 3.10. From individual data reported in fig. 4  1. 
Fig. 2. Agreement between radiographic spread and analgesic spread using Bland and Altman plot. The mean bias was 1.12, with limits of agreement of −0.85 and 3.10. From individual data reported in fig. 4  1 . 
×