It was the first time I had to testify in a case without a dead body. The patient was a 30 year-old Hispanic man who got into a fight with another guy on the street. The emergency room physician told me "it looked like someone tap danced on his face." He had severe facial injuries, including a jaw fracture and bruises that had swollen his eyes shut. He lay unconscious in the hospital for a week, and when he came to, he couldn't remember what had happened—not uncommon with brain-injured patients. Witnesses heard shouting and saw the other guy run away, but didn't witness any beating. When the hospital doctors couldn’t tell the DA how many times the victim was struck, the DA called the Office of the Chief Medical Examiner for a forensic assessment.
So, the radiology reports in the medical records documented the bony injury to the face and sinuses—and they also commented on some bruising to the brain. Buried in the description of the findings (but notably omitted in the radiologic diagnoses) was a mention of "swelling" over the scalp. I asked the DA to subpoena the actual CT scans. What I found shocked me. The "swelling" was actually bleeding, in two distinct areas: one on the right back of the head, and another at the left side of the head. Two more planes of injury! The ER doc’s tap dancing analogy was not far off.
I called the DA and told her my opinion. With five planes of injury, this victim was struck at least four times, plus a fall against the pavement—and if he had been struck multiple times in the same location I probably wouldn't be able to tell that there were additional impacts, since they would have occurred along the same plane. After reviewing the radiology from the hospital as well as the medical records, I could attest with medical certainty that these injuries did not come about after a single punch and fall, as the defendant was claiming. When the case went to court I was the only prosecution witness who could tell the jury what actually happened, using the physical evidence from the injured man’s body. His assailant was convicted.
This experience during my fellowship training taught me several lessons about radiology. One was that hospital radiologists tend to focus on bony injuries, and will often omit soft tissue findings entirely. If I hadn't sought out the CT scans and looked at them myself, I would have missed two additional impacts—compounding evidence, in this case of criminal assault. I also learned that just because I am a pathologist doesn't mean I'm not qualified to interpret radiology. As a forensic pathologist, I interpret X-rays most frequently when I'm called to identify decomposed cadavers or when I need to find bullets and other foreign bodies in trauma cases; I also receive radiology reports and hospital scans with a decedent’s medical records, and I study those as a part of the postmortem investigation. They have proven crucial in identifying the cause of death or in directing my autopsy examination in several important cases.
If the patient went to the hospital prior to dying, I can stack the clinical radiological images to create a three-dimensional reconstruction of the body. In one officer-involved shooting case, I used this method to answer the question of whether the victim was facing toward or away from the police when he was shot in the face. So while many forensic offices (including the agency where I currently work) have purchased full body X-rays to improve the speed and accuracy of their radiologic examinations, I really yearn for a CT or MRI scanner with 3-D imaging capacity. I want to see the soft tissues too—not just the bones.
Some radiologic findings, like air emboli following laparoscopic surgery, might be missed entirely at autopsy if a CT or MRI isn't done prior to the autopsy examination. Colleagues have shared with me cases in which CT scans have alerted them to subtle findings—small pulmonary nodules, vertebral artery injury, healing fractures in abused children—prompting them to do a more extensive or targeted dissection to make the forensic diagnosis. Some have gone so far as to perform post-mortem angiography by obtaining expired contrast material from hospital radiology departments and injecting it into a cadaver in order to assess vascular integrity. These scans make excellent courtroom exhibits and are less likely to be thrown out by a judge who doesn't want to upset the jury with gory photos from the autopsy.
Radiology is an imperfect diagnostic avenue for postmortem examination, however. It’s a shadow play. You aren’t looking at the real thing, but rather at a two-dimensional image of it, the shadow cast by the X-rays onto a capture plate, and separated by 1 centimeter intervals between the slices taken of the body. During an autopsy the pathologist is looking at the real thing—and feeling it, and smelling it, and listening to the sounds it makes when she handles it. In the course of thorough postmortem I have found linear non-displaced skull base fractures that weren't visible on radiology and vascular injury no clinician had detected. I have also seen benign congenital lesions misinterpreted as malignancy or trauma. That's why I'm not worried that radiologists with high-tech "virtual autopsies" are going to replace me and my trusty scalpel. Nothing can match the certainty and validity of a direct, hands-on forensic examination. Autopsy means “see for yourself.” You can’t do that by staring at shadows on a screen.