The value of lumbar dorsal root ganglion blocks in predicting the response to decompressive surgery in patients with diagnostic doubt.

BACKGROUND CONTEXT: Pain as a consequence of nerve root compression may not be easy to diagnose. Degenerative changes causing nerve root compression on magnetic resonance imaging (MRI) are common but not necessarily symptomatic while the distribution of pain attributable to a particular nerve root is variable. Selective dorsal root ganglion blocks (DRGBs) have been used in these situations to aid the diagnostic process, although their use remains controversial.

PURPOSE: We sought to investigate the positive predictive value of DRGBs in predicting response to decompressive surgery on a particular nerve root in a patient cohort with diagnostic uncertainty after clinical examination and MRI.

STUDY DESIGN/SETTING: This was a retrospective review of prospectively collected data on 100 consecutive patients.

METHODS: One hundred consecutive patients who underwent diagnostic DRGB under the senior author were identified retrospectively. Clinical records were reviewed for the reason for diagnostic uncertainty, level assessed, whether the DRGB reproduced pain typical for the patient's symptoms, whether there was anatomically appropriate sensory and motor disturbance, whether good pain relief was achieved, and whether they had good response to surgery.

RESULTS: Of 100 patients recruited, four were removed from analysis owing to inadequate surgical decompression proven on postoperative MRI. Of the remaining 96 patients, 74 achieved immediate relief in their symptoms after DRGB. Fifty-one patients underwent surgical decompression after a successful root block; 41 patients achieved a good result from this surgery, and 10 did not. Nine patients who had no relief in their symptoms from DRGB still underwent surgery to decompress the same nerve root; six patients had relief of their symptoms from surgery, two did not respond, and one was lost to follow-up. The most common reason for diagnostic uncertainty was multilevel disease (74%) followed by patients with atypical pain (23%). The most common level assessed was the L5 nerve root. The positive predictive value was found to be 80.4%, the negative predictive value was 22.2%, with a sensitivity of 85.4% and a specificity of 16.7%.

CONCLUSIONS: In patients with diagnostic doubt, a positive DRGB is a good predictor of a positive outcome after surgery to decompress that nerve root. However, the negative predictive value is poor. This result could almost certainly be improved if there was a better definition of what constitutes a positive, and more importantly a negative, DRGB result. In the meantime, DRGBs are a useful adjunct in predicting the outcome of decompressive surgery in people with pain as a consequence of potential lumbosacral nerve root compression.