An 82-year-old man came to my office because he had passed out in church. He had been healthy all his life. A few years before our first encounter, he had had a similar experience but had never made much of it. He had told no one about it and had felt fine when he regained consciousness. He attributed the syncopal episode to hunger or anxiety and never considered "bothering the doctor."

His history of syncope and his loud systolic ejection murmur, however, gave me great cause for concern. If he had aortic stenosis, as I suspected, syncope was a bad prognostic sign.

My first step, of course, was to check his insurance. Would prenotification be required before I began spending? Miraculously, he was not a managed care patient but was covered by Medicare and secondary insurance. His wife mentioned that if the problem had anything to do with his heart, the family preferred a particular hospital in our region known for its expertise in cardiac surgery. Money, she added, was no object.

I called a cardiologist, who performed echocardiography the same day. The aortic valve was narrowed. I prided myself. Yes, I thought, I had saved this man's life with a history and a stethoscope. We would get that valve replaced, and he would be fine.

Two hours and a dozen phone calls later, the patient was in the emergency department at the heart hospital his family had requested. His wife called to thank me for my help and to reassure me that he would be back after the heart problem was treated.

When I called her 2 days later, she said it was a good thing that he had gone to the special heart hospital. Although he had yet to leave the emergency department, several important facts had been determined. The aortic valve was narrowed but, according to the cardiologist, not enough to have caused the syncope. The angiogram did show that one of the distal branches of the heart's circulation system was narrowed.

"They also did a special test to see if his heart beat normally," she said.

The patient received a coronary stent and a pacemaker. Discharge medications included warfarin and aspirin. I saw him for follow-up 3 days after discharge.

He looked haggard but said he was relieved to be out of the hospital. We spoke. I examined him and ordered coagulation studies.

"Anything else? Otherwise I'll see you in about a month," I said.

"Well, I didn't mention it the first time, Doctor, because it seemed silly. I've had this problem with my bowels for years, and I wonder if you wouldn't mind helping me with it?" I asked a few more questions and then performed the rectal examination I had formerly deferred. His stool was melanotic, strongly positive for occult blood.

Two hours later, the patient underwent upper endoscopy. No cause of the bleeding could be found. He was admitted. Colonoscopy later revealed cancer of the colon. The patient became bacteremic after the colonoscopy but recovered from the bacteremia and underwent colonic resection, from which he made a satisfactory recovery.

I told the story to a colleague familiar with the staff at the heart hospital.

"Bet he got an electrophysiologic study," she said.

"How'd you know?"

She smiled. "It's the latest thing. Everyone who passes out gets one. Big moneymaker since the government started regulating pacemakers."

"You're exaggerating. Anyway, good thing he had it because his valve gradient wasn't too bad and there had to have been some reason he was passing out." She just kept smiling. I started thinking about the stent in my patient's distal circumflex coronary artery. Why should such a lesion require any treatment? In reality, stenting was for patients in whom angioplasty had failed. In addition, the patient would have needed angioplasty only if his circumflex artery were dominant.

Could the aortic stenosis, combined with anemia due to colon cancer, have resulted in syncope? Had I backed into the diagnosis of colon cancer when a questionably indicated angioplasty went bad, requiring a stent and necessitating medication that had increased colonic bleeding? Had my best intentions and the work of the heart hospital created such an outcome? Was this outcome good or bad? When I try to sleep these questions awaken me, and if I start thinking about them, I'm up for the rest of the night. I try not to think and just recite my litany:

"It worked out fine. We got to the cancer sooner. He probably needed the pacemaker even if he did not need the stent. They're not just doing more electrophysiologic studies to generate dollars lost when the Feds cut down on pacemakers. As much as it seemed too tempting to permit physicians to profit from the tests and procedures they performed, paying them more to do less was worse, wasn't it? Fee for service is better. Less hassle. I am not going to question the cardiologist's judgment. No guideline could have helped me in this case, or could it have? I am not going to say anything. I hate thinking about what I could have done differently. Then again, nothing terrible happened. Stop thinking. Go to sleep."