American Airlines kept an as-yet-unnamed pilot from his plane in Little Rock (AR) on Monday, after a security agent said he smelled alcohol (and mint) on the pilot's breath.

The pilot was reportedly taken away to the county jail, after pasing a field sobriety test; and he was given a BAC test. He was said to have blown .004%, 1/10 the FAA's limit, and 1/5 American's.

Passengers were delayed some three hours, nearly enough time to drive to their St Louis destination, as the pilot was still not allowed to fly the airplane.

How could this happen?

We wondered a lot of things, and we asked a lot of folks. "Is there any common way to give a test result of .004%, without having consumed something alcoholic?" we asked some friends in medicine. One highly-qualified flight surgeon told us, "Yes-biological variation in the test result.  A negative result could well be in the range 0-5.  The presence of  0.4mg% (ie. 0.4mg alcohol per 100ml blood) could easily be within the 'clutter' around zero. For example, the limit here for driving is 80mg%. Any attempt to claim that this pilot was unsafe at 0.4 is clearly ludicrous." There's more to it than that, though: "However," he continued, "the pilot may have been 'hung over,' and there is no way of measuring this, apart from subjective feeling of ill health. In order to interpret this result you need to know whether he was drinking the night before."

What about other things? Could a recently-'installed' set of dentures, that had been sitting overnight, fool the machine? What about the simple act of using mouthwash? We called Procter and Gamble, makers of Scope; but there wasn't anyone there who was able to go on the record. A Pfizer (Listerine) spokesman told us that proper use of their (26.9% alcohol) mouthwash could render a breathalyzer result that would show the suspect was over the legal driving limit, even ten minutes later. How long would it take, from the proper use of the mouthwash, until a level just 1/20 of that would show up? He didn't know; he'd have to have 'someone from legal' get back to us.

We had another question for our experts: Is there likely any reliability to a number that low, on the equipment typically available at major US airports or police stations? The answer we received, unanimously, from industry and the medical field, was, "no."

We got the same, unequivocal answer (from the doctors), when we asked if a BAC of .004% could impair a pilot's abilities, in any measurable way. "No." In fact, we found that other physiological circumstance or condition (missing an hour's sleep, for instance) could commonly produce similar, or worse results, than having a BAC of .004%

But... if the pilot weren't drunk, how did the "agent" smell booze? One respondent (a qualified medical man) told us, "In the course of my duties I have often examined Airmen who have smelt strongly of alcohol but have a BA of zero. They are hungover from the night before. I have advised on occasions that they are unfit for duty because of their hangover." Since we don't know what the pilot was doing the day before, we just can't say...

We also queried American Airlines, DavTech (makers of an infrared-technology "Intolilyzer") and Draeger (who make the Breathalyzer® -- yes, it's a trademark), to add to the story. None could get back to us, in time for our deadline last night.

It's an interesting story, and leaves the headline question largely unanswered. We'll undoubtedly get more comments and opinions on this one. Expect a followup.

FMI: www.amrcorp.com