Diagnostic efficacy of clinical tests for lumbar spinal instability.

BACKGROUND: Although various clinical tests are utilized to assess lumbar spine instability (LSI), few have documented diagnostic efficacy. We assessed the diagnostic efficacy of four clinical and one radiographic test for LSI in patients with degenerative lumbar disease.
METHODS: A cohort of 52 patients with pain attributed to lumbar spine stenosis and degenerative spondylolisthesis were prospectively evaluated utilizing dynamic X-rays, the passive lumbar extension (PLE) test, instability catch sign, painful catch sign, and the apprehension sign. The results of these preoperative tests were compared with spinal surgeons' intraoperative documentation of spinal instability considered in this study as the "gold" standard.
RESULTS: Intraoperatively, 33 patients demonstrated instability (63.5%) whereas 28 had motion documented on preoperative dynamic radiography. The sensitivity, specificity, positive, and negative predictive value and accuracy of dynamic radiography were 84.8%, 100%, 100%, 79.1%, and 90.4%, respectively. The diagnostic efficiency of PLE was higher than other additional studies - sensitivity 78.8%, specificity 94.7%, positive predictive value 96.3%, negative predictive value 72%, and accuracy rate 84.6%.
CONCLUSION: Dynamic radiography was more reliable than any of the clinical tests in diagnosing LSI. Among the latter, PLE had the highest diagnostic value for establishing LSI.

INTRODUCTION

Lumbar spinal instability (LSI) is a common cause of nonspecific low back pain (LBP).[1] At present, flexion-extension X-rays are the standard method for measuring anteroposterior translation[24] but have several shortcomings.[2] There are several clinical tests for diagnosing LSI,[1235] however, none have been previously proven to be effective measures for LSI.

We investigated the diagnostic efficacy of dynamic X-rays and four clinical tests to assess LSI: (a) passive lumbar extension (PLE) test, (b) instability catch sign, (c) painful catch sign, and (d) apprehension sign utilized to establish the diagnosis of LSI. Furthermore, patients were followed 6 months postoperatively with dynamic X-rays to determine if they became unstable.

MATERIALS AND METHODS

Clinical data

The 52 patients included in this prospective study met the following inclusion criteria: (1) back/radicular leg pain, (2) age between 50 and 70 years, (3) diagnostic studies showing degenerative lumbar spine disease, (4) patients who underwent laminectomy for lumbar spine stenosis and/or degenerative spondylolisthesis [Table 1].

Table 1

Comparing the age, gender and duration of symptoms between stable and unstable patients

Imaging assessment

Preoperatively, all patients had dynamic radiographs and MRI [Table 2]. The need for fusion after laminectomy/decompression was based on intraoperative documentation of active translation of >4 mm or rotation of >10 degrees at the level of listhesis. All images were reviewed by three independent specialists: one radiologist, an orthopedic, and a neurological spine surgeon [Table 3]. Prior to surgery, the two spine surgeons used the JOA score and four clinical examinations to confirm LSI: the passive lumbar extension (PLE) test, the instability catch sign, the painful catch sign, and the apprehension sign [Table 4].

Table 2

Comparing the radiographic and intraoperative findings

Table 3

Showing our material briefly

Table 4

Comparing the outcomes of clinical tests and intraoperative findings

Intervention

Patients underwent decompression by the two attending surgeons who could see the images but were not informed of the outcome of the clinical tests. They could independently judge whether patients had LSI and required fusion or not. In addition, patients underwent dynamic X-rays 6 months postoperatively to see if they developed LSI.

Lumbar canal stenosis was present in 35 patients, whereas stenosis/degenerative spondylolisthesis was found in 17. The average age of patients was 56.7 years.

Statistical analysis

The independent samples t-test (quantitative) and Chi-square test (qualitative) were utilized to compare data between patients with/without LSI. The qualitative data were compared using Pierson's or Spearman's correlation test using SPSS software (version 15.0). P > 0.05 while the sensitivity, specificity, accuracy, positive and negative predictive values of dynamic radiography and clinical tests were calculated.

RESULTS

Although preoperative dynamic X-rays showed LSI in 28 patients, 33 patients were unstable intraoperatively (63.5%). The sensitivity, specificity, PPV, NPV, and accuracy of dynamic radiography in establishing the diagnosis of LSI were uniformly high [Tables 3 and 4]. Even though the PLE test had the highest efficacy in diagnosing LSI, there was no significant correlation with neurologic symptoms (P = 0.65) [Table 4].

DISCUSSION

Intraoperative documentation of LSI best correlated with preoperative dynamic X-ray evidence of instability. Similar to Kasai et al.,[6] our findings confirmed that the PLE test had the highest diagnostic efficacy among all the clinical measures with a sensitivity of 78.8% and specificity of 94.7% [Table 5].

Table 5

Diagnostic efficacy of clinical tests

Limitations

Although two expert spine surgeons evaluated the presence of intraoperative LSI, there may still be nonreproducible intraoperative surgeon-bias constituting a flaw in the study design. Another shortcoming of this study is the relatively low number of cases included.

CONCLUSION

Preoperative dynamic X-rays best predicated the chance of intraoperative documentation of LSI as performed by two spine surgeons. Among the clinical tests for assessing preoperative LSI, the PLE had the highest predictive value.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.