MONTEREY, CALIF. Harvard Medical School hematologist Dr. Jerome Groopman has said that "people talk about technical errors in medicine, but no one talks about thinking errors."
This sentiment registered with Dr. Bari B. Cunningham, a pediatric dermatologist at the University of California, San Diego, and Rady Children's Hospital, prompting her to air her "missteps and misdiagnoses" with colleagues at the annual meeting of the California Society of Dermatology and Dermatologic Surgery.
"The only real mistake is the one from which we learn nothing," she said, quoting self-help author John Powell.
In looking back at her own errors, Dr. Cunningham saw evidence of the four types of errors cited by Dr. Groopman in his book, "Where Doctors Go Wrong":
▸ "I Recognize the Type." These are attribution errors based on stereotypes, she said, such as the Hodgkin's lymphoma that goes undiagnosed in an anxious, neurotic 50-year-old patient with pruritus.
▸ "I Just Saw a Case Like This." After four cases of viral exanthem, it may be tough to recognize the distinctions that mark a drug eruption. On the other hand, a rare diagnosis that leaves a "deep impression" may heighten consideration of that "zebra" in the next 10 patients with more typical conditions, she pointed out.
▸ "I've Got to Do Something." A physician may have a tendency to panic when faced with a rapidly spreading condition, but "as dermatologists we have time" to consider the differential diagnosis before whipping out the prescription pad, said Dr. Cunningham. Look it up. Consult with a colleague, she advised.
▸ "I Hate (or Love) This Patient." It's fairly obvious that one's irritation with a given patient can lead to an oversight, but the reverse is also true: Physicians may be a reluctant to acknowledge signs or symptoms of a serious disease in patients to whom they have grown close.
Dr. Cunningham recalled that she was reassured with the rapid resolution of left-sided facial Sturge-Weber syndrome in a toddler following pulsed dye laser treatments and a check by an ophthalmologist. A few years later, she felt "incredibly responsible" when the child was diagnosed with advanced glaucoma, because she had not realized or informed the family that even low-risk patients with V1 (fifth cranial nerve, ophthalmic division) Sturge-Weber syndrome require annual ophthalmic examinations.
Referring to another case, she said she was lucky to have considered Kawasaki disease in an Asian American baby boy with a groin rash that she might have missed on a busy afternoon in a private practice. An insistent mother emphasized that the rash began with a fever.
"There was a fellow there, and I was trying to do the right thing and said, 'You always want to have Kawasaki disease in your differential.' Lo and behold, this child was sent for an echocardiogram and had a dilated coronary artery," Dr. Cunningham said. Manifestations of this potentially fatal disease may be subtle and incomplete in infants, but in this case the fever was a critical factor leading to further evaluation according to a published algorithm (Pediatrics 2004;114:170833).
" Kawasaki disease has surpassed rheumatic fever as the most common acquired heart disease in children" under 5 years old in the United States, she said. "Never ignore your gut, or the parent's!"
Reviewing a final case, Dr. Cunningham recalled being consulted by the parents of an 8-month-old; they had been told that the fast-growing lesion on his posterior thigh was a hemangioma. "The father was a physician and they didn't buy it," she commented. Her own doubts were somewhat assuaged when a surgical biopsy performed in the operating room was read by a pathologist as a hemangioma.
Still, the rapid growth, the fact that the lesion did not appear until the child was 4 months of age, and the bloodless surgery made her continue to doubt the diagnosis.
MRI revealed the lesion was a virtually avascular "very well-circumscribed, nonenhancing mass" that eventually proved to be a lipoblastoma containing primitive adipocytes, she said. Although benign, such lesions continue to grow and do not resolve on their own, and may recur.
We have time to consider the differential diagnosis before whipping out the prescription pad. DR. CUNNINGHAM