How Accurate Is An MRI Report?

Can cutting-edge technology deliver faulty results?

By Kevin RR Williams

HEALTH During Lymphoma & Leukemia Awareness Month wWe have revealed when a lump may not be a cause for concern and why one might require investigation. Advanced equipment and methods for arriving at a diagnosis have been presented.

Imagine enduring 30 minutes of ear-deafening clicks during an MRI procedure. Anxiously, you await the pathology results. It's finally in: The palpable mass has not been detected. Good news or concerning contradiction? Could what was earlier felt, or perhaps seen, no longer exist or is this assessment a false negative?

True or False?

A false negative is when a test reveals favorable news that contradicts empirical evidence. Conversely, a false positive occurs when an incorrect unfavorable condition is reported. One might assume that results from a million-dollar piece of equipment would represent an authoritative conclusion. Quality of the imaging coils put around the body part being scanned and the computer programs used to control the imaging and to analyze the images are important. But perfectly tuned equipment is only as reliable as the person who views the images and prepares the report.

You should be able to wear a tampon during an MRI with no problem. To be on the safe side though even though it is embarrassing you should tell the technician or nurse you are wearing one as it can cause images on the scan if they are looking at that area of your body. –Scarlett Chu

I know of a case where a radiologist assumed a uterine mass was a tampon and therefore reported no anomaly. By the time the true diagnosis was made, the patient had stage-4 cancer requiring aggressive chemotherapy. This week, her outlook is grim as the cancer has metastasized to her bones. How common could such errors be? Would you say perhaps less than 3 percent? The answer may shock you.

Statistics From Clinical Studies

Error existed in 47% of breast MRI studies.

Radiologist Emmanuelle Bouic Pagès, MD, and colleagues at CHU Lapeyronie, Montpellier, France, reported in the magazine Radiology that potential observer error existed in 47% of breast MRI studies performed from January 2005 to December 2010. Medscape clarifies that the many critics who came to the defense of the equipment may not have paid attention to the report content. "Machines don't make diagnoses; people do," Herbert Y. Kressel, MD, editor of Radiology, explained to Medscape. "This is a study about errors made by people using the results of MRI exams." Misinterpretation produced most of the false-positive results, according to the study. [1] With false-positive readings, a patient may undergo an unnecessary mastectomy or amputation.

It is hoped that the error rate is much lower wherever we might have tests performed. To determine if better resolution (higher Tesla) improves evaluation, a study by Grossman JW, et al. compared accuracy rates of 3-T and 1.5-T MRI in diagnosing medial and lateral meniscal tears in 200 patients. Fifteen of the 26 missed meniscal tears were not seen in retrospect even with knowledge of the tear type and location. The study concluded comparable accuracy of 3-T and 1.5-T MRI[2]

A negative MRI should not influence the care of a lesion.

A study by Akiko Shimauchi, et al. of 220 sequentially diagnosed breast cancer lesions found seven (3.2%) false negative results — considerably fewer than other published studies. Although the overall sensitivity of cancer detection was high (96.8%), it should be emphasized that a negative MRI should not influence the management of a lesion that appears to be of concern on physical examination or seen by other imaging methods. [3]

Living up to his name, an orthopedic surgeon who blogs under the nom de plume Angry Othopod vents: "Did you know that a growing number of doctors don’t even read the tests themselves? MRI is unnecessarily overused. In a study of 221 patients who had MRIs, the results showed that only 5.9% actually needed to have an MRI done.… If you suspect your doctor is just being quick or using MRI to reach that 'aha' moment, then you're in a bad scenario. When I order an MRI, I am 90% certain about what the results are going to show. Doctors need to have a clear-cut idea on what they can expect to see from the results. Next time you're told to get an MRI, and your doctor has little clue to your diagnosis, you may want to get a second opinion. Also, be sure to ask the physician if they read the MRI themselves." [4]

Blog: CT and PET Scan vs MRI

Unless it is of considerable volume, a mass detected on an x-ray, MRI, or CT scan is often a dot requiring a judgment call. Determining whether a dot is an anomaly depends upon the skill, experience and alertness of the technician. Add these variables to the need for properly tuned equipment and you can see there is a margin for error. Therefore, if a radiologist reports back with positive results, get a second opinion. When results are negative, a second opinion is still advised.

If a doctor reports that an MRI reveals no tumor or more alarmingly, no cancer when there other symptoms or tests contradicting the claim, be wary. Malignancy is determined by biopsy not radiology. Don't let your life be cut short by a douche bag who cannot tell the difference between a tampon and a tumor.

Tags: differential, erroneous diagnosis, errors, medical mistakes, misdiagnosis

  1. Critics Question Study's 47% Observer Error Rate for Breast MRI. ^
  2. Comparison of the accuracy rates of 3-T and 1.5-T MRI of the knee in the diagnosis of meniscal tear. Grossman JW, De Smet AA, Shinki K. AJR Am J Roentgenol. 2009 Aug; 193(2):509-14. ^
  3. Breast Cancers Not Detected at MRI: Review of False-Negative Lesions. ^
  4. MRI overuse is widespread, and dangerous to patients. ^
  5. Image of Philips MRI from the Wikimedia Commons via