False Scents, False Sense, and False Cents

Vincent Starrett, the famous Sherlockian and Baker Street Irregular, wrote a premier biography of Sherlock Holmes titled The Private Life of Sherlock Holmes. ¹ In a review of this book, the renowned William Bennett Bean, professor of medicine and head of internal medicine at the University of Iowa College of Medicine, wrote that novices in medicine should master Sherlock Holmes to “… not be misled by false scents, false sense, or false cents.”² This was good counsel then and remains good counsel now.

During 5 decades of teaching in medical schools, I have made the reading of certain Sherlock Holmes stories mandatory. The use of the mystery genre in medicine is not new and, of course, the roots go back to Sir Arthur Conan Doyle and his teacher, Dr. Joe Bell.³ As the quintessential example of a person who uses observation and deduction to solve mysteries, Holmes recapitulates what physicians do in clinical practice every time they see a patient. Teachers of medicine often tell students to watch how this process works and then emulate it much in the manner that Holmes instructed Dr. John Watson—his roommate, friend, and biographer: “We were simply there to observe and to draw inferences from our observations.”⁴

Holmes's instruction of Watson proceeded by ascension from simpler to more complex tasks. Watson, however, could never win, for Holmes made him the readers' foil, as illustrated here.

Looking, But Not Seeing

In early 1969, I was called by the surgeons at Ft. Gordon Army Hospital to see a captain who had come back from Vietnam with gunshot wounds of the chest and abdomen, sustained 2 weeks earlier. Treatment had required extensive surgery, multiple medications (including antibiotics), a Foley catheter, chest tubes, and various drains before the patient was air-evacuated back to the United States. Recovering now, he nevertheless had unexplained left-lower-quadrant abdominal pain.

When I asked him about the pain, he told me that it consistently occurred as soon as he ate one bite. I didn't find anything helpful on examining him, but his chest x-ray revealed an increased distance between the top of his gastric gas bubble and the air in the base of the left lung. This finding suggested a subphrenic fluid collection.

I called the surgeons and suggested that they drain the presumed left subphrenic abscess. They did, and the pain resolved. I surmised that the mechanism of his pain was a prandial dilation of his stomach, which pressed on the abscess and gave him referred pain to the left lower quadrant. Chest films had been inspected by radiologists and surgeons who looked but did not see the evidence for a subphrenic abscess.

“Not invisible but unnoticed, Watson. You did not know where to look, and so you missed all that was important.”⁵

Seeing, But Not Observing

One of our heart surgeons paged me to see his next-door neighbor in our emergency department (ED). The neighbor, a middle-aged attorney, had been giving a speech when friends noticed he was hyperventilating. So they called an ambulance.

In the ED, the patient was lying comfortably on a cart while being monitored. The ED resident reviewed the history with me and reported no additional symptoms. The man now felt well. His physical examination, electrocardiogram, chest x-ray, and laboratory studies, including cardiac markers, were all normal, and his doctors wanted to send him home.

Looking at the monitor, I asked, “So why does he have sinus tachycardia at 120 beats per minute?” I was told that his doctors thought he was just anxious. I commented that young kids and teenagers will have anxiety-provoked accelerated heart rates lying at rest in the ED, but not middle-aged men. I advised admitting him to the coronary care unit (CCU) for close observation, serial cardiac markers, and follow-up electrocardiograms.

Two hours later, I received an urgent page from the CCU resident. The patient had suddenly become short of breath and cyanotic. The stat ventilation/perfusion (V/Q) lung scan (the then-standard noninvasive diagnostic procedure) showed no perfusion of one lung and huge perfusion defects in the other—all consistent with massive bilateral pulmonary emboli, which we then confirmed with a pulmonary angiogram.

Three days later, after use of thrombolytic recombinant tissue plasminogen activator (rtPA) and heparin, his repeat V/Q lung scan had become normal. Two days after that, he went home, doing well with warfarin therapy. He had no recurrence. His pulmonary emboli had resulted from deep vein thrombosis consequent to a recent long drive from Chicago to Iowa and back. The ED physicians had seen the sinus tachycardia on the monitor but made no observations about it.

“I see no more than you, but I have trained myself to notice what I see.”⁶

Observing, But Failing to Reason

A woman in her 70s with a Starr-Edwards prosthetic valve was admitted with hemiparesis and subsequently underwent cardiac catheterization. A physical examination revealed the hemiparesis, aortic prosthetic sounds, and a systolic ejection murmur. The click caused by the ball in the cage was low-pitched rather than high-pitched. The remainder of her examination was unremarkable except for a soft continuous bruit over her right femoral artery catheterization site.

The diagnosis was ball variance.⁷ The silastic poppet in the cage of the Starr-Edwards ball-valve prosthesis can shrink, swell, or change its shape or consistency, resulting in a rough thrombogenic surface. Any surface clot can embolize and possibly cause a brain infarction, with which she had presented.

However, there was an additional diagnosis. The patient's continuous bruit at her catheterization site was consistent with an arteriovenous fistula that had not been present on admission but appeared after the catheterization. The diagnosis of an iatrogenic arteriovenous fistula is easily confirmed by transcutaneous ultrasonography but will be missed if you fail to reason from the clinical observation.

“On the contrary, Watson, you can see everything. You fail, however, to reason from what you see.”⁸

Reasoning, But Failing to Reason Backward

At an army hospital in early 1970, I became interim chief of the department of medicine and head of cardiology. In this capacity, I consulted in our busy internal medicine clinic staffed by 4 excellent internists. One day, one of these internists called me to see a 60-year-old woman with chronic abdominal pain. He said, “We've all seen her and done every test in the world to find a cause of her pain and nothing is showing up. She's had a complete gastrointestinal series and proctoscopy and all the usual lab tests several times.”

When I saw her in the clinic, she appeared comfortable. Her chart and x-ray file were almost as big as she was. I pushed the chart aside and said to her, “You know if I look at that chart, I'm going to end up with the same conclusions as all the other doctors. I'd rather simply hear you tell me about what's bothering you.” I sat back and for 15 minutes without interruption listened to her describe her abdominal pain.

When she was finished speaking, I said, “It sounds to me as if you think you have cancer.”

Her answer surprised me. She said, “I know I have cancer.”

I asked, “How do you know this?” She then related this story.

Two years earlier, when her husband was stationed in Jacksonville, Florida, they'd gone out for dinner. In the restaurant she developed severe abdominal pain. They drove to the local military hospital's ED, where she was seen by 2 surgeons who found a mass in her abdomen. Because she had eaten, the surgeons operated under spinal anesthesia so that she would be awake during the surgery. She heard one of the surgeons tell the other that he found cancer.

I told her that this was strange, because I'd been told there was nothing in her chart about her having cancer. I suggested that we request her old medical records, operative report, and pathology report from the Jacksonville hospital.

Two weeks later, these records arrived and confirmed that a painful abdominal mass had led to surgery. Instead of finding cancer, the surgeons found an internal hernia with incarcerated infarcted omentum. I could picture them standing opposite each other at the operating table and expecting to find cancer. One surgeon picks up the dead omentum, holds it out to the other, and says, “There's your cancer.”

I showed her the reports, proving that she did not have cancer. Her abdominal pain then disappeared and did not recur, and she returned to work as a roofer. I saw her next 6 months later. She had broken her arm falling off a roof and confirmed that the abdominal pain had never returned.

In this case, the answer to the mystery required reasoning backward 2 years.

Old charts are not invisible but are often unnoticed. Today, if the information is not in the computer files, it will likely be ignored. Reasoning backward can lead you to search for something vital in the old chart.

“In solving a problem of this sort, the grand thing is to be able to reason backward.”⁹

Conclusion

The bedside application of Sherlock Holmes's advice to Watson can sometimes lead to the diagnosis without any tests at all. Once you have the scent, you can use your sense to save some cents.