Hey, Hospitalists, Those Hoofbeats Sometimes May Be a Zebra

Medscape

15-05-2025

Kelly Chellis could have dismissed the random swollen knee on her then-6-year-old, Trys. At that age, children get injured all the time, and Trys could have tripped, twisted it on the playground, or done one of many other things and simply forgotten to mention it.
While Trys's swollen knee at first seemed to fall under the category of typical childhood bumps and bruises, the journey that followed is an important example of the need for hospitalists to stay curious and keep an open mind in patient diagnosis.
'The week of Trys's sixth birthday, they woke up one morning with their right knee swollen to the size of a softball,' Chellis recalled. 'We kept saying, 'Does it hurt?' but Trys really wanted to just go on to school.'
Kelly Chellis
Chellis' instincts, fine-tuned where Trys was concerned, given that the child had survived a congenital heart defect, gave her pause, though. She took Trys to their regular pediatrician, where x-rays were taken. Although the films showed nothing irregular, the pediatrician had them proceed to Nationwide Children's Hospital in Columbus, Ohio.
'We were put on the infectious disease floor because they thought that there was an infection in the kneecap, and they were just (administering) IV [intravenous] antibiotics,' Chellis recalled, saying that all of the tests that the epidemiologists ran came back normal. Protocols on the floor restricted their movements in case Trys had something communicable. As Chellis kept them occupied inside their room, a break in the case emerged from a strangely timed art project.
'Out of the blue, [Trys] wanted to draw a hand turkey. In March, that's kind of odd — to want to draw a turkey,' Chellis said. 'But I'm like, great, I'll trace your hand. Then when Trys laid their hand down on the table, their fingers would not actually lay flat.'
Chellis immediately drew the care team's attention to Trys's hand, which led to the hospital's rheumatology team joining the case. Shortly thereafter, Trys was diagnosed with juvenile idiopathic arthritis (JIA). Chellis was stunned, but they finally had an answer.
Don't Let the Obvious Block the Possible
Occam's Razor, often attributed to 14th-century friar William of Ockham, said that if you have two competing ideas to explain the same phenomenon, you should prefer the simpler one. A frequently used, more colloquial explanation of the principle is 'When you hear hoofbeats, look for horses, not zebras.'
While Occam's razor is helpful in many situations, it has a caveat: Zebras still exist, especially in the medical field. Trying to find the balance between relying on medical training, which is based on pattern recognition and hard data, and the potential for a variety of diseases, like JIA in Trys's case, to manifest on the diagnostic spectrum can be difficult, especially for doctors just entering the field.
Rebecca Carter, MD
'In any diagnosis, I try to consider the two extremes: One, what is most likely, and two, what would be most dangerous to miss,' said Rebecca Carter, MD, associate professor of pediatrics and division head of General Pediatrics at the University of Maryland School of Medicine, Baltimore. 'While much of what we do in medicine, and certainly most of what I do in general pediatrics, comes down to the notion that common things are common, our most significant impact, and the true foundation of our training, comes down to not missing the rare or less common presentations.'
While there are many such diseases that affect all ages, JIA is a textbook example of a diagnostic outlier in many ways: First, as profiled in Trys's case, the symptoms of JIA — a group of disorders in which the immune system mistakenly targets the body's own joints and tissues, triggering widespread inflammation — can overlap with more common conditions like sports injuries or infections.
Many other early symptoms compiled by the Arthritis Foundation , like morning stiffness and mild fever, are nonspecific, and symptoms come and go and can be anywhere from mild to excruciatingly painful and disabling during attacks.
All of these factors make JIA diagnostically elusive, even though it is the most common form of arthritis in young people, affecting about 300,000 children in the United States, according to statistics from the Arthritis Foundation. As a result, the American College of Rheumatology reported that some children wait months, even years, for a diagnosis.
'JIA is a challenge to diagnose, particularly because most kids just compensate for any joint issues they could possibly have by limping or just making up for it in any way that helps them keep going,' said Aixa Toledo-Garcia, RhMSUS, and managing partner at The Center for Rheumatology in Albany, New York. Toledo-Garcia is not affiliated with Trys's case but said the pattern of their diagnosis is very common. 'It's usually the parent that notices the limp or the swollen joint.
Practical Care and Uncommon Diagnoses: Maintaining a Balance
Carter said that longtime pattern recognition experience in a clinical setting is what alerts doctors to medical outliers like JIA.
'As physicians, we have to see hundreds or even thousands of cases of the same common illness in order to detect when something feels off or different about the patient in front of us,' she said. 'All those hours of medical school and residency give us this perspective, to help us to fully trust our gut when there is something uncommon about how a child presents, and to know when to dive a bit deeper to explore alternative diagnoses to make sure we get it right.'
'Yet' is a great addition to carrying out the process of diagnosis, added Dympna Weil, MD, a board-certified obstetrician/gynecologist and master coach at Prescribing Possibility in Upstate New York.
'By this, I mean, we must use those skills we learn and continue to hone with time. We need to let go of any ego we have and embrace the fact that we don't and cannot possibly know everything,' Weil said. 'Sometimes, patients' bodies don't read the textbooks. They may present differently than the books would tell us they 'should.' So admitting we aren't sure exactly what is going on yet can go a long way in keeping our minds open to other diagnostic possibilities.'
If the patient doesn't improve after being given the standard treatment for the diagnosis reached, Weil said clinicians should stay open to the possibility that perhaps the initial diagnosis was incorrect.
'We may need to go back to the history and listen again to the patient's presentation or reassess their physical exam to better inform us. It is critical to confidently know what you know, but equally, to know what you don't know and to not be afraid to ask for a consult or pursue a different path.'
Trys's diagnosis of JIA, while shocking, brought clarity. It shifted the family's focus toward proactive care, which included immunosuppressive medications to reduce inflammation and slow disease progression. Since then, both Chellis and her child have become advocates for JIA and participants in the fight for better healthcare policies both statewide and nationally. Partially as a result of their experience, Trys is now studying for their Medical College Admission Tests, and their medical career will clearly be informed by how their diagnosis and subsequent treatment played out.
Standing Out in the Herd
Medical training emphasizes pattern recognition and probability for a reason: You don't have time to agonize over each and every one of the thousands of possibilities that the rash on the patient in bed 5 might indicate. However, there are times that the most meaningful diagnoses come from the smallest observations, the details that might otherwise seem inconsequential.
More than anything else, the fact that a child's random idea to draw a hand turkey in the middle of March ended up being a critically important key to diagnosis highlights the fact that curiosity is an essential tool in medicine, right alongside the stethoscope. You may even have to fight a few people on this, but sometimes, when you hear hoofbeats, they don't belong to a horse. Sometimes, they really do belong to a zebra.