High-quality Symptom Information Dramatically Improves the Reports of Lumbar Spine MRI
Hearing directly from patients yielded near-perfect diagnostic agreement with specialists
When radiologists use a questionnaire to collect symptom information from patients before lumbar spine MRI, they can more accurately and confidently diagnose the cause of back pain and radicular pain.
MRI is highly sensitive for degenerative and structural abnormalities of the spine, according to William Palmer, MD, director of Musculoskeletal Radiology at Massachusetts General Hospital in Boston.
“However, MRI also shows spinal abnormalities in individuals who do not have symptoms,” Dr. Palmer said. “Because the same abnormalities can occur in both symptomatic and asymptomatic individuals, it is difficult for radiologists to distinguish incidental MRI findings from the true causes of pain.”
Patient-Reported Information Provides An Assist
In a prospective single-center study published in Radiology, Dr. Palmer’s team investigated a fundamental approach to bolstering diagnostic accuracy: asking the patients to weigh in on
their own symptoms.
The study included 240 patient participants and captured data between May 2022 and February 2023. The investigators designed and implemented an EHR-hosted symptom questionnaire that was sent to patients two days before lumbar spine MRI.
The questionnaire—which only took patients two to three minutes to answer—asked about the location, chronicity and character of symptoms. Depending on patients’ answers to the question, “Where are your symptoms located?” the questionnaire applied conditional logic to ask additional questions pinpointing locations in the back, buttock and/or leg.
Questionnaire results were then immediately available to interpreting radiologists when they launched the MRI exam in their PACS. No extra clicks were required to retrieve the results. Six musculoskeletal radiologists, with experience ranging from two to 24 years, each interpreted studies from two groups: One that included patients’ answers about their symptoms and one that didn’t. For both groups, the investigators were allowed access to clinical information in the EHR, including provider notes.
The investigators compared the findings of radiologists with those of spine specialists, the latter having collected symptom information during face-to-face interviews.
“When radiologists diagnosed pain generators using patient-reported symptom information from our questionnaire, they agreed almost perfectly with the diagnoses made by spine specialists,” Dr. Palmer explained. “Without questionnaire results, discrepancies were common despite access to the MRI requisition and clinical notes in the EHR.”
“Symptom information was high in quality when it was obtained directly from patients using our questionnaire, but low in quality when obtained second-hand from care providers,” Dr. Palmer added.
Accurate and Focused Clinical History and Exam Can Provide Clarity
Masis Isikbay, MD and Vinil Shah, MD discussed the merits of these findings— and the importance of a relevant clinical history when ordering imaging studies for patients with lower back or radicular pain—in an accompanying commentary. The authors give readers an overview of the major patterns of back pain and they offer guidance on making the most of the clinical information at hand.
“It is generally accepted common practice for the treating physician to correlate patient symptoms and/or physical examination findings with imaging results to guide clinical management,” Drs. Isikbay and Shah wrote. “Despite this, it is often the experience of many interpreting radiologists that little to no clinical history is provided at the time of image interpretation.”
Whereas spine specialists are trained to interpret studies and make their own diagnoses—usually with the benefit of in-person visits that allow them to take a customized approach to questioning and physical examination—other providers depend on radiologists’ reports to guide the management of lumbar spine symptoms.
When radiologists interpret MRIs without clinical context, dictated reports often become laundry lists of findings that can’t be accurately ranked in order of importance, Dr. Palmer pointed out.
Using a well-designed questionnaire that is integrated with PACS in the EHR, patient symptom information is within easy reach at the time of image interpretation, Dr. Palmer’s team demonstrated that radiologists can add more value to their dictated reports by improving accuracy and distinguishing incidental findings from actionable pain generators.
“Precise identification of a specific pain generator remains a challenging task,” Drs. Isikbay and Shah wrote. “This is where an accurate and focused clinical history and examination can provide more clarity as to where the source of the pain may exist anatomically. This information is of great relevance to the interpreting radiologist and to proceduralists who may perform an intervention to address the patient’s pain.”
Patient-Reported Symptom Information Benefits Radiologists
Dr. Palmer’s team has shown that collecting symptom information directly from the patient is not only effective in improving diagnosis, it can be done in a streamlined way that complements the existing radiology workflow, providing useful results even in the absence of a face-to-face consult.
“Most importantly, this study validates a new, symptoms-based approach to the interpretation of lumbar spine MRI,” said Dr. Palmer. “It also shows that an effective questionnaire can be designed by radiologists, implemented in the EHR, and delivered automatically to dictating radiologists at the time of MRI interpretation.”
For More Information
Access the Radiology article, "Impact of Patient-reported Symptom Information on the Interpretation of MRI of the Lumbar Spine."
Read previous RSNA News articles on spine imaging:
- Deep Learning Helps Speed MRI of the Spine
- Is Kyphoplasty a Viable Treatment for Vertebral Fractures?
- RSNA Launches Cervical Spine Fracture AI Challenge