PHOENIX - The state's high court has given those charged with drunken driving a new tool to try to defeat charges against them.

In a unanimous ruling Monday, the Arizona Supreme Court said defendants can present evidence to jurors that there are variables among individuals, including body temperature, that make it impossible to precisely convert a breath reading to someone's specific blood-alcohol content. That can be crucial in determining whether a person is legally impaired.

Chief Justice Rebecca Berch, writing for the unanimous court, rejected arguments by the Tucson City Attorney's Office that those variables are legally irrelevant. And defense attorney James Nesci said this provides defendants with the ability to rebut breathalyzer readings.

But Deputy City Attorney Baird Greene said the ruling, coupled with an earlier one on the same case by the Court of Appeals, is not a total loss for prosecutors.

"It clearly establishes that we have the right to pursue motions to preclude what we view as clearly speculative evidence," he said. "And that's a tool we totally intend to deploy."

The case involves Joseph Cooperman, charged by Tucson police in 2010 with driving while "impaired to the slightest degree." Based on breath tests, prosecutors added a separate charge of driving with a blood-alcohol content of at least 0.08, a per-se violation of the law.

Prior to trial, prosecutors sought to block any evidence the breath test might not be accurate. But the judge refused, resulting in an appeal.

Nesci pointed out state law presumes a blood-alcohol reading of zero to 0.049 shows the motorist is not under the influence of alcohol. There is no presumption either way for a reading of 0.05 to 0.079, with a jury able to consider "other competent evidence" to determine guilt.

Only at 0.08 and above is someone presumed to be legally intoxicated.

Nesci said defense attorneys always have had the opportunity to rebut the accuracy of the readings. Now that path of attack can include physiological variables.

He cited testimony from a defense expert that hematocrit - essentially the concentration of red blood cells - can affect blood-alcohol concentrations by up to 5 percent, plus or minus. The same expert said even a one-degree variance in the temperature of a person's breath from the setting on the breathalyzer could mean a difference up to 8 percent in its blood-alcohol reading.

There also are arguments that environmental factors and breathing patterns can affect readings.

"My breath might be at .090 but I might actually be at .070," Nesci said, where a jury cannot presume a person was intoxicated. "Or I might actually be at .049 and therefore I'm presumed not under the influence."

The high court, without comment, upheld the appellate court ruling that those variables are relevant to the question of accuracy of breath tests.

Greene, however, said this is not a stay-out-of-jail free card for Cooperman - whose case now goes back to the trial court - or any other defendant. He said they cannot simply argue there are variables and, therefore, the readings are unreliable.

He said the ruling still requires a defense attorney to provide specific evidence his or her client's physiology differs from the presumptions the machine makes. Otherwise, Greene said, such testimony is little more than "rank speculation" that would only confuse the jury.