Nailing the diagnosis

According to Wikipedia, the differential approach to diagnosis was first suggested a century ago by a pioneering German psychiatrist, Emil Kraeplin, but contributions to the evolution of the differential diagnosis (Ddx) have been made by countless outstanding physician-scientist-educators.  The golden age of U.S. medicine nurtured the great diagnosticians, including Internal Medicine’s Tinsley Randolph Harrison, Pediatrics’ Frank Oski and Dermatology’s Samuel Moschella and Walter Shelly.  In those days, the most revered physicians were masters of the Ddx.  As a medical student, I sat in awe at Dr. Ralph Feigin’s “Rounds”, a series of case presentations with comprehensive Ddx. Since then, nailing the diagnosis has been my loftiest goal.

Almost 30 years have passed since I was inspired by Feigin Rounds. I am now very familiar with a long (and ever-expanding) list of common and rare conditions that apply to complex patients in search of a diagnosis, especially because cutaneous signs are readily available clues that require a trained eye, rather than imaging or laboratory testing. So for me, compiling the Ddx list is now much less difficult than communicating the possibility of a life-altering diagnosis, especially to an unsuspecting patient, family and referring physician.

One of my most memorable early challenges was an infant referred for “eczema.”  When I walked in the room, his other features took precedence:  failure-to-thrive, sparse, brittle hair and vertical nystagmus. Although I had never seen a case, trichothiodystrophy was at the top of my Ddx.  However, I was much less sure about the best way to communicate my concerns to the baby’s medically unsophisticated mother.  I started by gently pointing out his unusual features followed by a phrase that has served me well since then:  “Has your doctor ever mentioned this to you?”

A year or so later, a more medically sophisticated mother brought her 8-month-old son in for evaluation of “eczema.”  She was convinced that he had food-allergy triggered atopic dermatitis, and had restricted his diet to elemental formula, so I allowed her opinion to sway my inexperienced diagnostic skills.  Although my documented impression was incorrect, I commented on the atypical distribution of his erythroderma, accentuated at the skin folds and diaper area.  I did not see him again until almost age 3, when his obviously brittle hair moved Netherton’s to the top of the list. A little later, his similarly affected sister was born.

In the past 2 decades, I have diagnosed hundreds of children with uncommon and rare disorders.  In many cases, medical advances have defined standard evaluation, and even successful treatment: confirmatory gene testing and IL1 receptor antagonists for the cryopyrinopathies; MRA/MRV and propranolol for PHACE; gene sequencing panels for infants with epidermolysis bullosa or collodion phenotype, and the miraculous clinical trial evaluating the impact of neonatal protein replacement on the early development of teeth, sweat glands, and hair follicles in children affected by  X-linked hypohidrotic ectodermal dysplasia. But more often, establishing a rare diagnosis prompts more angst than action.

Furthermore, the value of nailing a diagnosis has gradually diminished, especially relative to the revenue generated by procedures or patient volume. So, for many busy clinicians, diagnostically challenging patients are like hot potatoes. But diagnosis remains my passion, and occasionally an annoying obsession.  Sometimes, I can’t help but assess friends, distant relatives and acquaintances, then agonize about whether to divulge my suspicions.  In a moment of irony, my brother (who has a touch of oppositional defiant disorder) diagnosed me with another ODD (obsessive diagnosing disorder).

One day last week, my first patient was a 15 year old with a chief complaint of acne.  The first glimpse prepared me for a longer-than-expected visit.  He was tall and thin, with arachnodactlyly. On closer exam, he had pectus excavatum, a high arched palate, crowded teeth, hyperextensible joints, flat feet and prominent striae (in addition to mild acne). His family history was non-contributory.  The reply to my question about his doctor’s mention of the distinguishing features was that a Genetics consultation years earlier did not recommend any additional evaluation.  This left me with 3 no-win choices: recommend a second opinion, ignore the signs or explain both options in more detail. I chose the last and most time-consuming (at the expense of my patients waiting to be seen).

I face a similar conundrum every time I see an infant with a low-risk midline sacral birthmark or a congenital nevus at with low-likelihood of neurocutaneous melanosis: how to decide on the relative risk-to-benefit of high cost imaging requiring general anesthesia? Sometimes my tendency to minimize expensive evaluation has failed patients. Years ago, I delayed a diagnosis of occult Crohn’s in a young, asymptomatic teenager with pyoderma gangrenosa, because I recommended only a screening stool guiac, and left the responsibility of sampling to her father, an adult gastroenterologist who did not make it a priority. And just last year, my negative history and physical exam did not prompt further evaluation for another patient who presented with erythema nodosum (EN), and developed hematochezia a year later. I learned about the IBD via a smug electronic message from the pediatric gastroenterologist who performed the colonoscopy.  Rather than sending a defensive eReply, I called her to discuss the case. She was unaware of the possibility of “idiopathic” EN, or the wide range of possible associated extracutaneous problems.

Investing the time in direct communication with colleagues, rather than waging chart wars, is usually a very worthwhile win-win-win for me, patients and colleagues.   It took me years to recognize the importance of these informal discussions in fostering valuable collaboration, and expanding my own Ddx.  These days, when I receive a request for a biopsy alone, rather than a diagnostic evaluation, or a limited impression like “not a surgical candidate,” I make every effort to contact the clinician for more valuable information hidden in the sound bite.  In the age of EMR, the value of direct communication should be part of medical school curriculum.   Perhaps the ghosts of golden-age medicine and the current champions of accountable care and cost-effective medicine will make it so.