Pages

.

Computers Selecting Medications: InteMedica Part Two

We’re exploring medical tests. We all know that medical tests can be wrong. In fact, statistically, if 14 tests are run—often done when a pet has a complete blood count including red blood cells, white blood cells, electrolytes, liver, kidney, and thyroid function—at least one of them will be wrong. That’s why if a test is abnormal, the test should be repeated rather than accepted as accurate.

In yesterday’s blog, we explored another area of medical tests—interpretation. Just as the test data can be wrong, the interpretation can be wrong. Even with years of medical education, professionals can misinterpret tests. The example used was an elevation in blood urea nitrogen (BUN), which the veterinarian interpreted as a sign of kidney failure. Although this makes sense because a sick kidney is one reason BUN becomes elevated, in Bella’s case, as we discussed yesterday, this was a mistaken interpretation. Bella’s high BUN was due to dehydration. It would have been a waste of money, an enormous stress on the dog, and a strain on the family to have acted on the misinterpretation.


Unfortunately, misinterpreting test results is not rare. Here’s another example of a blood test that was misinterpreted—this time, a cat’s blood test. And, again, this example is from a pet that has been treated within the last month.

Baron’s saga began when pathologists diagnosed a connective tissue cancer on
his toe. Connective tissue cancers, which are called sarcomas, can be very aggressive; Baron’s veterinarian recommended amputating the leg to control the cancer. A holistic veterinarian would also have recommend changing to a healthy, whole food diet, adding herbs, and using pheromones to decrease stress.

Last week, Baron’s leg was amputated, and after two days in the hospital Baron was discharged home with pain medication and lots of good advice for his rehabilitation. Before Baron left the hospital, blood was drawn to confirm no bacterial infection had set in during the lengthy surgical procedure. There was no sign of infection—the white blood cells were normal—but Baron’s blood sugar was shooting toward the moon—about 350 mg/dl, which is more than two times normal for cats (70-150 mg/dl). The veterinarian called Baron’s family and said it appeared he had developed diabetes
, which is rather common in chubby cats like Baron, and that he should return to the hospital to begin insulin therapy.

When the family discussed the elevated blood sugar test, I suggested that it was due to stress rather than to diabetes. Instead of returning to the hospital for insulin, I recommended they ask for a blood test that clarifies whether there had been a history of elevated blood sugar or whether it was a sudden, short-term elevation. The test that tells us this is called a glycosylated hemoglobin (G Hb). When a pet has longstanding diabetes, the hemoglobin is changed by the sugar in the blood. The change is so marked that in a cat it’s possible to estimate the blood sugar levels of the past 4-6 weeks. (In a dog, the G Hb estimates blood sugar levels for even longer—about 8-12 weeks.) Fortunately, when Baron returned for a blood test, his sugar levels were normal. Absolutely wonderful news for the cat, and the family. No diabetes.

If misinterpreting tests is easy for medical professionals, and if it happens rather frequently, is it more comforting or less comforting to think a computer will interpret the test? Is it more comforting or less comforting to have the computer designed by the company that sells the supplements the computer recommends?

No comments:

Post a Comment