Computed tomography in the diagnosis of low-back pain

Update Item Information
Publication Type thesis
School or College School of Medicine
Department Biomedical Informatics
Author Thomas, Randal Jay
Title Computed tomography in the diagnosis of low-back pain
Date 1984-06
Description In the management of patients with low-back pain three general sources of information are utilized by the attending physician to diagnose the specific pain-producing disorder: (a) the demographic and historical data reported by the patient, (b) the physical examination performed by the physician, and (c) the radiographic evaluation of the patient's spine. The proper use of the data obtained from these sources allows the physician to identify the cause of pain in most cases of low-back pain and, as a consequence, prescribe the most effective therapy for the disorder. In order to assess the efficacy of a specialized radiographic technique, known as computed tomography (CT), three studies were performed involving the collection and analysis of demographic, historical, and radiographic data from 460 patients referred with low-back pain for CT examination. Of particular interest in our studies was the assessment of the role of CT in the management of low-back pain patients. Computed tomography is a relatively new radiographic examination in which the computer, using information obtained from x-ray absorption patterns, can construct cross-sectional images of the spinal vertebrae. One study was designed to assess the ability to predict a normal"" computed tomographic examination on the basis of a patient's demographic and historical data. The chi-square test for significance was used to identify demographic and historical parameters of the patient data which most significantly separated the ""normal"" patients from those found to have ""abnormal"" computed tomographic studies. After identifying the pertinent data variables, Bayes' equation and multivariate discriminant analysis were used to test the ability to discriminate between normal and abnormal patients before a CT scan is performed. Receiver Operating Characteristic (ROC) curves were consequently constructed from the data to provide a measure of the ability to predict those patients who had normal CT examinations (based solely upon the patient-supplied demographic and historical data). A similar study was also performed to test the ability to predict a herniated nucleus pulposus from patient-supplied data. It was found that on the basis of information supplied by patients who had been referred for CT examination that it was difficult to successfully separate normal patients from abnormal and herniated nucleus pulposus patients. _We, therefore, concluded that the referring physicians involved in our study were appropriately selecting patients for lumbosacral CT examination. The final study performed was designed to assess, in patients with low-back pain, the degree of association between three spinal conditions which are detectable by CT examination: (a) the vacuum disk phenomenon-defined as the appearance of gas on CT in the intervertebral disk region of the spine, (b) degenerative joint disease--a condition in which the spinal facet joints develop boney outgrowths and/or hypertrophy which may lead to spinal pain and restriction, and (c) herniated nucleus pulposes--a condition often referred to as a ""slipped disk"" in which the intervertebral disk ruptures resulting in the protrusion of some of the disk material onto the spinal nerves within the spinal canal region, and subsequent pain and/or weakness of the lower back and lower extremities. The unique imaging qualities of computed tomography allowed the analysis of these three physiologic phenomena as they appeared exclusively or in conjunction with each other. Bayes' equation was implemented to calculate the probability of a herniated nucleus pulposus appearing on computed tomographic examination when a vacuum disk or degenerative joint disease was either present or absent at the same lumbar interspace. It was found that the presence of herniated nucleus pulposa was correlated with the occurrence of the vacuum disk phenomenon at the fourth and fifth lumbar vertebral interspaces provided that no degenerative joint disease was found at the same interspace regions. Also, the occurrence of herniated nucleus pulposa at the third lumbar interspace was found to be statistically correlated with the occurrence of degenerative joint disease at the same interspace.""
Type Text
Publisher University of Utah
Subject Tomography; Backache
Subject MESH Back; Pain; Tomography, X-Ray Computed
Dissertation Institution University of Utah
Dissertation Name MS
Language eng
Relation is Version of Digital reproduction of "Computed tomography in the diagnosis of low-back pain". Spencer S. Eccles Health Sciences Library.
Rights Management © Randal Jay Thomas.
Format Medium application/pdf
Format Extent 872,574 bytes
Identifier undthes,3850
Source Original University of Utah Spencer S. Eccles Health Sciences Library (no longer available)
Master File Extent 872,613 bytes
ARK ark:/87278/s6zp47w5
Setname ir_etd
ID 190730
Reference URL https://collections.lib.utah.edu/ark:/87278/s6zp47w5
Back to Search Results