Hint: Break down the impression into individual components. If you’ve coded for an independent diagnostic testing facility (IDTF) for long enough, you’re used to the fact that some spinal magnetic resonance imaging (MRI) scan reports are capable of causing serious headaches. Some of this stress can be mitigated by breaking the impression into individual components — and referring each back to the indication. Have a look at this challenging clinical example of a spinal MRI that will put your diagnosis coding skill set to the ultimate test. Don’t Discount the Power of the Indication Exam: MRI Lumbar Spine W/O Indication: Back pain after seizure. Recent injury. This dictation report proves challenging right off the bat given indicating diagnoses such as these. Radiology coders are hard-wired to expect the bare minimum out of their indications, but this adds an extra degree of difficulty due to two potentially conflicting diagnoses. Without any further elaboration, you are unable to determine the underlying reason for this patient’s MRI scan. Is it to assess for a fracture following an injury that occurred due to the patient’s seizure? Or is it to assess for any underlying spinal etiologies that may have invoked the seizure? Since you’re not supposed to make any sort of diagnostic inferences (assumptions) based on the indication, you should prepare to code for all potential scenarios. This means that you will report diagnoses that apply to both the injury and the seizure, if applicable. Know When to Compare, Contrast Impression With Indication 1. Multiple abnormalities, first, 30% anterior compression at L1 with posterior cortical retropulsion and bony canal stenosis with no compression of the caudal roots or conus. There is a horizontal fracture line L1 and edema with subacute presentation. There is no diffuse infiltrative process, and the etiology may be traumatic, or insufficiency related, but infiltrative disease is not excluded here such as metastatic disease or myeloma. Combing this first portion of the impression with the diagnostically open-ended indication increases the degree of difficulty in coding this MRI substantially. Had the provider not speculated on the etiology of the L1 fracture and edema, you might feel more confident in your diagnosis code selections. However, you’ve now got to contend with two competing diagnoses — one traumatic and one pathologic. While you could decide to send this report back for an addendum, it likely wouldn’t do you any good. Given the nature of the report, the radiologist is just as uncertain of the nature of the fracture and edema as you are. The first diagnosis you’ll be looking to report is the compression. Unless the findings elaborate further that this is actually a fracture, you should not be coding a diagnosis of compression as a fracture. This point is driven home by the etiological speculations, as well. “Due to the etiological uncertainty stated in the report—physician speculates trauma, insufficiency, or pathological etiologies — it would be inadvisable to code it as a compression fracture. Rather, you should stick what the report does conclusively give you,” says Lindsay Della Vella, COC, CMCS, medical coding auditor at Precision Healthcare Management in Media, Pennsylvania. Additionally, you should not conclude that just because the compression and fracture are documented at the L1 site that you are working with a compression fracture. Unless the findings elaborate further that this is, in fact, a compression fracture, you should code the diagnoses separately. Physician’s note: “As is the case with the compression, the retropulsion and the stenosis can either be the result of trauma, or be pathologically induced,” says Barry Rosenberg, MD, chief of radiology at United Memorial Medical Center in Batavia, New York. So, when you see a diagnosis such as retropulsion, which you might associate with trauma, you should not jump the gun in this instance and code it as a result of trauma. The fracture, on the other hand, gives you a little more room to report as the result of trauma. That’s because the physician documents edema with subacute presentation at the fracture site. The “subacute presentation” wording is important, as it implies its was recently occurring. It’s also perfectly feasible that that vertebrae became deficient due to a pathological condition, such as malignancy, and then the weakened bone fractured during a traumatic incident. However, you don’t have enough information at hand to code it as such, so you should report the fracture as traumatic (primary diagnosis), and the compression as non-traumatic: Report Any Clinically Significant Findings, not Just Clinically Relevant 2. There is an unusual smooth solid-appearing mass on the right psoas surface, which may infiltrate or emanate from the psoas muscle. This does not have the typical appearance of an abscess or hematoma and could be an incidental solid lesion such as a neurofibroma or lymph node. This finding is not definitely benign. This next portion of the impression involves a diagnosis that is most likely pathologic in nature, but may be entirely incidental to the patient’s reason for the MRI. If the radiologist was performing this exam exclusively to evaluate for trauma, you might consider leaving this diagnosis out altogether. However, since this might be significant given the clinical findings in the first part of the impression, you should report this diagnosis. You’ll use the index to report a mass of the psoas muscle as M62.9 (Disorder of muscle, unspecified). Coder’s note: It’s highly unlikely that a condition affecting the lumbar spine could result in the patient’s documented seizure. However, if the etiology of the mass on the right psoas surface is malignant, a separate primary or secondary tumor on or near the brain could be the culprit. Include Abnormal Findings as Clinically Significant, When Applicable 3. The lumbar thecal sac and contents have an abnormal appearance at L1 and T12 with abnormal low signal and thickening of the caudal roots. These are indistinct. This may reflect hemorrhage, which is unlikely, inflammatory process, or an infiltrative process related to the other findings. Further raising suspicion is apparent new-onset seizure activity. This last portion of the impression is certainly not one that you should consider incidental. Much like part two of the impression, these abnormal findings may or may not be related to the patient’s documented seizure — and should therefore be included as relevant diagnoses. Once you’ve made that determination, the final piece to the puzzle is coming up with the correct diagnosis code(s) to report the abnormal appearance, abnormal low signal, and thickening surrounding T12-L1. Due to the incredibly ambiguous nature of the diagnoses, you’ll have to report a diagnosis code that’s equally ambiguous. That’s where the “abnormal radiological findings” diagnosis codes come into play. You’ll report R93.7 (Abnormal findings on diagnostic imaging of other parts of musculoskeletal system), which covers each of the abnormal components the radiologist documents in part four of the impression. All you’ve got to do now is determine how to order each of the four respective diagnoses. With the traumatic fracture listed as the primary diagnosis, you may report the remaining three in any order: