Currently, various preoperative studies including selective nerve root block (SNRB) and electromyography (EMG) have been used as diagnostic tools for lumbar radicular pain before surgery. Notably, SNRB is commonly used for diagnosis and treatment of spinal back pain or radiculopathy5). Many studies have reported its value in the preoperative assessment of patients with radicular pain3,10,17). SNRB and EMG are used for localizing anatomic sources of pain when clinical or radiological studies are equivocal. However, the application of SNRB as a diagnostic tool has not shown consistent efficacy. In the present study, we analyzed patients who underwent spinal surgery for radicular pain to identify the efficacy of SNRB as a preoperative diagnostic tool.

A total of 154 patients (82 women; 53.5±17.1 years of age) met the criteria. Of these, 52 (33.8%) underwent preoperative SNRB as a diagnostic tool. Table 2 shows the baseline characteristics of the enrolled patients. Demographics, medical history, length of hospital stay, diagnosis, number of affected levels, and surgical time were not significantly different between the groups. Although the results did not satisfy statistical significance, patients in the SNRB group had more excellent (42.3% vs. 36.3%) or good (46.2% vs. 36.3%) grade and less fair (11.5% vs. 27.5%) grade compared with those in the non-SNRB group. A poor grade was not observed in any patient from either group. In terms of favorable outcome (excellent or good grade), the SNRB group showed significantly better results (88.5% vs. 72.5%, p=0.041).

Table 3 shows the results when comparing only patients with two or more levels. In patients with multiple levels, there was a difference in postoperative outcome between the 2 groups. However, due to the small number of patients, this result was not statistically significant.

SNRB, when combined with a careful physical examination and quality radiologic studies, is an important tool in the diagnostic evaluation of patients with predominantly radicular symptoms6). SNRB may be used to define the source of pain and is especially useful when clinical examinations, electrodiagnostic studies, and imaging studies are equivocal11). Diagnostic nerve root blocks may be particularly useful to identify the symptomatic level in patients with multi-level pathology1,4). Similarly, SNRB may also be useful when the location of symptoms appears to conflict with abnormalities identified in radiologic findings. In the present study, although not statistically significant, postoperative outcomes defined according to Odom’s criteria were better in the SNRB group than in the non-SNRB group. Moreover, the proportion of patients with favorable outcome was significantly greater in the SNRB group than in the non-SNRB group.

EMG has limitations in localizing the patient’s radicular pain to a single level15). In addition, radiologic studies may be difficult to interpret after spinal surgery due to scar tissue in the epidural space and other anatomic changes14). Furthermore, radicular symptoms may not correspond to classic dermatomal patterns13). The rationale for a diagnostic SNRB is that, if a particular spinal nerve is responsible for causing or mediating patient symptoms, then anesthetizing the nerve should temporarily relieve symptoms. If the symptom is pain, anesthetization of the responsible nerve should relieve the pain. If the symptom is paresthesia, anesthetizing the responsible nerve should produce numbness in the area where the paresthesia was previously present. If the symptom is numbness, anesthetizing the responsible nerve should produce no change in the numbness or perhaps accentuate it. If a nerve anesthetized is not responsible for the patient’s symptoms, pain will not be relieved, and numbness in the distribution will occur2).

The accuracy and effectiveness of SNRB as a diagnostic test have been reported in several studies. Haueisen et al.8) reported on patients with sciatica of unclear etiology who underwent SNRB; an accurate diagnosis was made in 43 of 46 (93%) patients. Schutz et al.12) retrospectively reported on SNRBs performed in 23 patients. In 15 patients, an operation was performed at the level indicated by SNRB, and the operative findings were in agreement with SNRB in 13 (87%) patients12).

In contrast, there are confounding variables that can undermine the validity of our study. Injection of local anesthetics may spread beyond the intended spinal nerve target to structures such as adjacent dorsal rami, spinal nerves, or the sinuvertebral nerves, thereby causing a false-positive result. Anatomical variation is another potential problem. In a cadaver study, nerve root abnormalities were found in 14% of individuals9). The furcal nerve usually arises from the L4 root level and contributes to both the lumbar and sacral plexuses of nerves7). Neurological symptoms, suggestive of 2 roots being involved, frequently result from furcal nerve compression.

Pain is purely subjective, often with uncertain pathophysiology, and may be influenced by psychological, social, financial, and legal factors, as well as by the efficacy of concurrent therapies such as medications and physical therapy. Additionally, spinal injections may be associated with a significant placebo effect. Reportedly, the placebo effect increased in direct correlation to the invasiveness of a procedure16).

The present study had several limitations. First, our data were retrospectively analyzed. In addition, the number of patients included in our study was rather small, which limits the statistical power to appropriately differentiate the relative importance of diagnostic tests studied. Finally, the pain-blocking effect of SNRB can influence the postoperative prognosis. Moreover, the above-mentioned confounding variables render the procedure difficult to precisely perform.

In the present study, the patients who underwent SNRB preoperatively had better prognosis after surgery than patients who did not. Although the statistical significance was not satisfied, further studies are necessary with large series.