Friday, December 19, 2008

Hunchbacks and Bony Variants

Photo by TCM Hitchhiker

The Hunchback of Notre Dame is a novel by Victor Hugo, set in 1482 in Paris, in and around the cathedral of Notre Dame de Paris. It has had numerous theatrical and movie interpretations. I'm no literary snob, and have to confess that my only real exposure to this story is via Walt Disney. One of the perils of having several children is unrelenting exposure to Uncle Walt and his minions.

There are hunchbacks in the medical world as well. The os styloideum was first described by the Frenchman Fiolle, in 1932. He termed this entity the "carpe bossu", which can be translated as "hunchback carpal bone". This anatomic variant is located on the dorsal side of the wrist, at the base of the third metacarpal. It is typically a separate ossicle, but can fuse with the adjacent bones, typically the second and third metacarpals, but sometimes with the trapezoid or capitate.

The carpe bossu is usually an incidental finding, but can cause symptoms in some patients, and can be confused with a ganglion cyst. When troublesome, patients will complain of focal dorsal pain, and a tender bone mass. The lesion usually becomes symptomatic in adulthood, but can rarely present in childhood.

In this case, a fourteen year old boy presented with wrist pain, and was discovered to have a scaphoid fracture on MRI. Review of the coronal images along the dorsal aspect of the hand reveals a carpal boss (red arrows) located between the bases of the second and third metacarpals:

The carpal boss is also well seen on this sagittal T1-weighted image:

The patient also had edema-like changes within the os styloideum and the base of the third metacarpal, seen on this coronal T2-weighted image:

Interestingly, the patient did not report any symptoms referable to this region. One wonders whether this area will eventually become symptomatic, given the degree of abnormal signal in this area.

Vic David MD

Friday, December 12, 2008

Lumbar Spine MRI— See More, Miss Less

What do you see in this diagram?

Some see a vase while others see two faces looking at each other. They are both there.

"You see what you look for and recognize what you know". This is a hoary but true dictum in radiology.

When looking at an MRI, radiologists will detect a greater percentage of abnormalities if they are specifically looking for them. Radiologists are in part a product of their training, and carry with them biases learned during that process.

One place where this tends to show up is in lumbar spine MRI scans, which are interpreted by radiologists with significant differences in their training. It's easy to miss subtle abnormalities that are clinically relevant on this exam.

Early in their post-training career, musculoskeletal-trained radiologists tend to miss small disk herniations, while neuroradiology-trained radiologists tend to miss bone findings such as stress reactions and stress fractures.

Here is an example of a subtle right foraminal disk herniation:

Here is an example of stress-related changes in the posterior elements of L4:

Both are subtle findings, but in my experience, musculoskeletal-trained radiologists will be more likely be miss the first example, while neuroradiology-trained radiologists will be more likely to miss the second example.

If you keep this in mind, whatever your training, you will be less likely to miss either type of finding.

Vic David MD

Friday, December 5, 2008

Osteomyelitis and Fly Larvae

Musculoskeletal infections can be a vexing, serious problem. For example, consider knee replacement surgery, a common procedure that is done over 400,000 times a year worldwide. If postoperative infection is avoided, the outcome is typically excellent. A post-operative infection, however, will often spell trouble, and a prolonged recovery.

Over time, we have gotten better at treating infections of bones and joints, particularly with the advent of antibiotics. In the pre-antibiotic era, some novel treatments were tried, including fly larvae:

In World War II, combat induced wounds to the extremities had a 20-25% rate of osteomyelitis. By the Vietnam War, this figure had dropped to about 8%, thanks to better treatment.

In the civilian world, infection is much rarer, but can still occur. Radiologists are often asked to assess for the possibility of bone, joint, or soft tissue infection, and the define its extent.

The question of whether or not osteomyelitis is present comes up most often in the foot. Collins el al. wrote an excellent article on the use of T1-weighted images when looking for pedal osteomyelitis (AJR 185:386-393, 2005). While signal abnormalities on T2-weighted and STIR images tend to be nonspecific, they noted that T1 signal abnormality that is medullary and confluent is highly suspicious for osteomyelitis.

Since then, I have used the principles outlined in their article when analyzing pedal osteomyelitis, and have had good success. I am not aware of any articles that address the MRI analysis of osteomyelitis in the hand and wrist. Given that the hand and foot are ontogenic homologs, I have transferred the principles of Collins et al. to analysis of osteomyelitis of the hand and wrist.

In this case, a 37 year-old man was sent for a finger MRI, due to pain and swelling. He suffered an injury at his work five weeks before the MRI.

A sagittal STIR image shows extensive marrow edema (red arrows) in the middle and distal phalanx, as well as a distal interphalangeal joint effusion (white arrow):

Marrow edema and excess joint fluid are worrisome findings, but tend to be nonspecific. Next, we examine the corresponding T1 weighted image:

The T1-weighted image depicts confluent medullary signal abnormality in the head of the middle phalanx (red arrow), and depicts extensive soft tissue edema. The presence of confluent medullary signal abnormality allows us to predict the presence of osteomyelitis with a high degree of confidence. Osteomyelitis and septic arthritis were confirmed at surgery.

Vic David MD