Until his admission to the hospital, John O'Connor, a fifty year-old railroad dispatcher from Charlestown, was in perfect health. He had never been sick a day in his life.
On the morning of his admission, he awoke early, complaining of vague abdominal pain. He vomited once, bringing up clear material, and had some diarrhea. He went to see his family doctor, who said that he had no fever and his white-cell count was normal. He told Mr. O'Connor that it was probably gastroenteritis, and advised him to rest and take paregoric to settle his stomach.
In the afternoon, Mr. O'Connor began to feel warm. He then had two shaking chills. His wife suggested he call his doctor once again, but when Mr. O'Connor went to the phone, he collapsed. His wife brought him to the emergency ward of the Massachusetts General Hospital at 5 P.M., where he was noted to have a temperature of 108° F. and a white count of 37,000 (normal count: 5000 to 10,000).
The patient was wildly delirious; it required ten people to hold him down as he thrashed about. He spoke only nonsense words and groans, and did not respond to his name. While in the emergency ward he had massive diarrhea consisting of several quarts of watery fluid.
The patient was seen by the medical resident, John Minna, who instituted immediate therapy consisting of aspirin, alcohol rubs, fans, and a refrigerating blanket to bring down his fever, which rapidly fell to 100°. He was in shock, with an initial blood pressure of 70/30 and a central venous pressure of zero. Over the next three hours he received three quarts of plasma and two quarts of salt water intravenously, to replace fluids lost from sweating and diarrhea. He was also severely acidotic, so he was given twelve ampule of intravenous sodium bicarbonate as well as potassium chloride to correct his electrolyte balance.
The patient could not give a history. His wife, upon questioning, denied any history of malaria, distant travel, food exposure, infectious disease, headache, neck stiffness, cough, sputum, sore throat, swollen glands, arthritis, muscle aches, seizures, skin infection, drug ingestion, or past suicide attempts. His past history, according to the wife, was unremarkable. He had never been ill or hospitalized. His mother died at age fifty-five of leukemia; his father, at age fifty-nine, of pneumonia. The patient had no known allergies, and did not smoke or drink.
Physical examination was normal except for a slightly distended abdomen and a questionably enlarged liver, which could be felt below the rib cage. Neurological examination was normal except for the patient's stuporous, unresponsive mental state. The patient was cultured "stem to stern," meaning that samples of blood, urine, stool, sputum, and spinal fluid were sent for bacteriologic analysis. He was also given heavy doses of antibiotics, including a gram of chloramphenicol, a gram of oxacillin, two million units of penicillin; later in the evening, kanamycin and colistin were added to the list.
X rays of the chest and abdomen were normal. Electrocardiogram was normal. Hematocrit was normal. The white count was elevated, with a preponderance of polymorphonuclear leukocytes, the cells which increase in bacterial infections. Examination of the urine showed a few white cells. Platelet count and prothrombin time were normal. Measurements of blood sugar, serum amylase, serum acerone, bilirnhin, blood urea nitrogen were normal. Lumbar puncture was normal.
An intravenous pyelogram (an X ray of the kidneys to check their function while they excrete an opaque dye) showed that the left kidney was normal, but the right kidney responded sluggishly. The excretory tubing on the right side seemed dilated. A diagnosis of partial obstruction of the right kidney system was suggested.
Because the abdomen was distended, six abdominal taps were performed in different areas by the surgical residents, Drs. Robert Corry and Jay Kaufman, in an attempt to obtain fluid from the abdominal cavity. None was obtained.
Dr. Minna's diagnosis was septicemia, or generalized infection of the bloodstream, from an unknown source. As possibilities he listed the urinary tract, the gastrointestinal tract, the gallbladder, or the lining of the heart. He felt there was no good evidence for a central nervous system cause for the fever, and no good history of drug ingestion or thyroid problems to account for the fever.
This was essentially the conclusion of the neurological consultants who saw the patient later in the evening. They felt that Mr. O'Connor had suffered a primary infectious process with sudden outpouring of bacteria into the blood, and consequent fever and prostration. They felt the infection was somewhere in the urinary or gastrointestinal system, or perhaps even in a small area of the lungs. They felt that meningitis, encephalitis, subarachnoid hemorrhage, or other central nervous problems were unlikely.
A formal surgical consult, also later in the evening, reported that in the absence of muscle spasm or guarding of the abdomen, and in the presence of six negative taps, an acute abdominal crisis was unlikely.
Genito-urinary consultants examined the patient that evening and reviewed his kidney X rays. They felt that there was a probable partial obstruction of the right kidney, but they could not determine whether this was a recent or a slowly developing change. They found no evidence of infection of the prostate gland to explain the fever.
Mr. O'Connor was placed on the danger list and transferred to the intensive-care unit of the Bulfinch Building. At the end of his first twelve hours in the hospital, his fever had been reduced, but there was still no explanation for the fever itself.
BEFORE continuing with Mr. O'Connor's hospital course, it is worth pausing a moment to consider the patient's initial symptoms and initial therapy. Mr. O'Connor presented with high fever and shock. Classically, the fever of unknown origin is a pediatric problem, and classically it is a problem for the same reasons it was a problem with Mr. O'Connor -- the patient cannot tell you how he feels or what hurts. However, a high fever in a child is less worrisome than it is in an adult, for children have a much greater tolerance for fever. In adults, prolonged high fever is more likely to result in permanent brain damage and death.
The most common cause of fever for anyone, child or adult, is infection; the most common cause of fever of unknown origin is also infection. There are some unusual causes occasionally seen, such as malignancies, bleeding in the brain, drug ingestion, and outpouring of thyroid hormone, but for the most part, unexplained fevers are produced by unidentified infections.
It is now known that one can harbor an infection in a secluded part of the body, and the body will make very little response to it; however, if the infection spreads into the bloodstream, there may be a "shower " of bacteria, and a subsequent sudden rise in temperature. The shower is usually brief, lasting minutes or hours, and often ends before the temperature rises. This makes diagnosis difficult; if one wants to catch bacteria in the blood, one must draw a sample before the temperature spike, and not during it or after it. It was thought that Mr. O'Connor was suffering from precisely this sort of situation: a sequestered infection producing episodic bursts of bacteria into the blood, with episodic fever. However, his fever was threateningly high. And thus a classic conflict in therapy arose. It is a conflict as old as Hippocrates. "For extreme diseases, extreme remedies," Hippocrates wrote. He also said: "For grave diseases, the most exact therapy is best." But obviously, an exact therapy depends upon a precise diagnosis, and here lies the conflict.
What is a diagnosis? The question is not as simpleminded as it first appears, for the notion of what constitutes an acceptable diagnosis has radically changed through the years. A diagnosis is drawn up on the basis of two kinds of knowledge: the physician's concept of disease processes, and his available therapies. Ideally, a diagnosis contains some sense of etiology -- the cause of the disease -- but for most of medical history etiology was either ignored or wrongly ascribed (as in "fever from excess of black bile").
In a modern sense, precise diagnosis is required because precise therapies are available. Yet the need for precise diagnosis is older; in Hippocratic time, this need was based on a prognostic, not a therapeutic, concern. Physicians were unskilled at curing disease, and therefore served mostly to predict the course of an illness which they could not influence. Robert Platt notes that "until quite recently....it did not matter whether your diagnosis was right or wrong....Prognosis mattered rather more, especially to the doctor's reputation. "
Hippocrates was deeply concerned with the prestige of the physician as related to prognostic acumen; much Hippocratic writing shows this preoccupation with "Sleep following upon delirium is a good sign." "Those who swoon frequently without apparent cause are likely to die suddenly." "Labored sleep in any disease is a bad sign." "Spasm supervening upon a wound is dangerous." "Hardening of the liver in jaundice is bad." "If a convalescent eats heartily, yet does not take on flesh, it is a bad sign. "
These observations are still valid today. But we demand something further from diagnosis, as the range of therapies has increased. If a person swoons, for example, it is important to know whether he has aortic stenosis -- and is likely to die suddenly -- or whether he is hysterical, or diabetic, or has some other reason for fainting. In short, we want more precise diagnoses because we have more precise therapies.
Throughout medical history, physicians of every age have felt that they had precise, specific remedies but few of these are still acceptable. As Berton Roueché notes, only three eighteenth-century drugs are still acceptable today: quinine for malaria, colchicine for gout, and foxglove (digitalis) for heart failure. All the other "specifics, " as well as what Holmes termed the "peremptory drastics," have disappeared.
Even as recently as 1910, L. J. Henderson commented that "if the average patient visited the average physician, he would have a fifty-fifty chance of benefiting from the encounter." Much has happened since then; in fact, nearly every diagnostic test and therapeutic procedure performed on Mr. O'Connor during those first twelve hours has been developed since 1910. For clinically, diagnosis and therapy go hand in hand; increasing sophistication in either one demands sophistication in the other.
The proliferation of tests and techniques in this century is staggering. Consider the following list of tests performed on Mr. O'Connor, and the dates those tests were first described in clinically practical terms:
X ray: chest and abdomen (1905-1915)
White-cell count (about 1895)
Serum acetone (1928)
CSF protein (1931)
CSF sugar (1932)
Blood sugar (1932)
Serum albumin/globulin (1923-1938)
Electrocardiogram (about 1915)
Prothrombin time (1940)
Blood pH (1924-1957)
Blood gases (1957)
Protein-bound iodine (1948)
Alkaline phosphatase (1933)
Uric acid (1933)