Welcome to Teaching Corner's Radiology Today.
This months topic centers on
Patient Selection for MRI Imaging


Radiology Corner
Basic Principles of MRI Imaging
J. Todd Knudsen, D.C.


Return to Chiropractic OnLine Today


In my last article I listed some basic principles of MRI including terminology. In this issue I would like to address the issues of patient selection for radiography and imaging.

The doctor of chiropractic is faced with decisions about radiography and imaging on a daily basis. With the advent of managed care and an accompanying heightened demand for cost-effective management of common musculoskeletal conditions, real thought must be put into whether the patient needs imaging and if so, what modality should be used.

PATIENT SELECTION

A couple of methods of patient selection are reviewed below.

Guidelines

Documents such as the Guidelines for Chiropractic Quality Assurance and Practice Parameters (Mercy Guidelines) and others achieve a relatively high level of sophistication and introduce a level of scientific validity to the process of spinal imaging decision analysis. These guidelines (relating to imaging in chiropractic practice) were based on the best available literature and an evaluation of the strength of that literature. Because no single document could hope to encompass all diagnostic and imaging possibilities, the necessarily broad scope of this document results in very general guidelines in imaging. I would refer the reader to the Mercy Guidelines for good solid guidelines for the use of radiography in clinical practice.

Another more simple set of guidelines, specifically for plain film radiography, is shown in the table below.

*Adapted from Deyo RA, Diehl AK: Lumbar spine films in primary care: current use and effects of selective ordering criteria, J Gen Intern Med 1:20, 1986: and Howard BA, Rowe LJ: Spinal x-rays. In Haldeman S, editor: Principles and practice of chiropractic, ed 2, Norwalk, Conn, 1992, Appleton and Lange, pp 361-364.
When using guidelines such as these, it is important to remember, that the radiographic examination should never be used as a "shotgun approach" to replace a thorough clinical history and examination, and that routine use (without clinical indication) of diagnostic imaging leads to increasing cost and unnecessary radiation exposure to the patient while failing to increase the probability of detecting significant pathology.

Decision Analysis

A superior way of selecting patients for a specific examination is through Bayesian analysis. This method of decision making appears quite complex at first but in reality is simple if only a few basic concepts are understood.

Before one can select patients for imaging, a few important considerations should be kept in mind. First among these considerations is the fact that the test in question should be indicated based on findings from a thorough history and physical examination of the patient.

Doctors should ask themselves the following questions:

1) Is this examination going to affect my diagnostic certainty about the differential diagnosis I am considering and, if so, how much?
2) Will the information expected to be provided by the examination change my diagnostic thinking enough so that it will significantly affect my choice of treatment?
Other important considerations for the selection of patients for a diagnostic test include:
1) The inherent risk of the examination to the patient.
2) The likelihood that the examination will be of benefit in establishing or refuting a diagnosis.
3) The potential benefit to the patient.
4) The risk of liability if the examination is requested or not requested.
Keeping these points in mind, one cannot dismiss the importance of clinical intuition in the selection process. The doctor’s judgment coupled with these more statistically oriented considerations is the basis of Bayesian analysis.

Bayes' Theorem: In 1763, Bayes described the relationship between the sensitivity and specificity of a test and how those two variables altered with pretest probability. Bayes also constructed a mathematical expression to explain this relationship. Terms that must be defined in order to understand and utilize Bayesian analysis are as follows:

Sensitivity: the ability of a test to detect a certain condition or disease.
Specificity: the ability of a test to exclude those patients who do not have a disease or condition.
Diagnostic Accuracy: the sensitivity and specificity of a certain test for a certain condition as compared to a known "gold standard" (i.e., postmortem, surgical exploration, or biopsy)
Pretest (Prior) Probability: this is the doctor’s impression (including intuition, clinical experience, and history and examination findings) of how likely it is that a patient has the disease or condition in question.
Exclusion Threshold: the point at which the doctor is convinced that the disease or condition in question is not present.
Action Threshold: the point at which the doctor is convinced that the condition or disease is present and institutes treatment.
Posttest (Posterior) Probability: the probability that a condition or disease is present after performing a diagnostic test.
Bayes' theorem has been summarized as follows: "It essentially means that if a test gives either a positive or a negative result, then the chance of the patient having or not having the disease in question depends not only on the sensitivity and specificity of the test, but also on the pretest probability of the presence of the disease."

Once a doctor has arrived at a pretest probability for a certain condition, he or she must then consider if the prior probability is below the exclusion threshold or above the action threshold. If below the exclusion threshold, this means that the doctor is confident that the disease or condition in question is not present, and no further testing is needed at that time for that condition. If above the action threshold, this means that the doctor is certain enough that a condition or disease is present to initiate a trial of treatment and that no additional testing is needed at that time for that condition. It is when the prior probability of a condition or disease being present in a patient lies between the exclusion and action thresholds that a diagnostic test with a reasonable degree of sensitivity and specificity for detection of that condition should be performed to move the prior probability over one of the thresholds.

In the event that the doctor still does not reach one or the other threshold, another test may be needed.

Dr. Knudsen is a board certified chiropractic radiologist (D.A.C.B.R.) and is Director of Clinical Diagnostic Imaging at National College of Chiropractic in Lombard, Illinois. For more information about Dr Knudsen’s radiology consulting services, MRI imaging, or quality control issues, please call Dr. Knudsen at (800)469-9729.

Return to Top


Chiropractic OnLine Today © 1996.
No portion of this page may be reproduced without written permission from
Internet AdCom Services.