BOSTON -- Henry Alfonso was arrested for allegedly dealing the prescription painkiller OxyContin. But he's in prison now because a small adhesive patch on his arm tested positive for traces of cocaine.

Alfonso's case is the first in Massachusetts to challenge the reliability of the sweat patches, which are used by federal courts across the country to test for various drugs, including cocaine, heroin, marijuana and methamphetamines.

Alfonso, 32, insists he has never used cocaine. He claims the six positive test results he has had since April were caused either by cocaine residue left in his apartment by a previous tenant or traces of the drug that may have been on cash his wife brought home from her job as a stripper.

Alfonso was arrested in December for allegedly attempting to buy OxyContin from an undercover agent and mailing OxyContin from Florida. After he was released on bail, probation officials began applying the patch on Alfonso each week to test him for drug use. Three months ago, after one of the patches tested positive for cocaine, Alfonso was sent to jail.

On Thursday, he went to federal court to ask a judge to release him while he awaits his trial in the OxyContin case, arguing the sweat patch results are unreliable.

U.S. District Court Judge William Young did not immediately rule on the request, but continued the hearing until Friday to give prosecutors time to present the findings of a scientist who will analyze Alfonso's patch results.

Richard Carelli, a spokesman for the Administrative Office of the U.S. Courts, said the patches are now used for drug testing in more than 40 federal court districts. He said although there have been some challenges to the reliability of sweat patches and other methods of drug testing, the federal court system still has confidence in the patch technology.

"The overwhelming majority of federal judges who have reviewed challenges to the sweat patch have found it to be a reliable indicator (of drug use)," Carelli said.

But Alfonso's lawyer, Matthew Feinberg, said federal courts in at least two districts have stopped using the patches because of questions about reliability.

"Henry has insisted from Day One that he's never used cocaine," Feinberg said. "He did move into an apartment that had been used by drug addicts."

Frederick P. Smith, a professor of forensic science at the University of Alabama at Birmingham, said a 1999 study showed that even trace amounts of drug residue can contaminate the outside of the patch and cause a positive result. The patches absorb sweat on the skin, which is then tested for the presence of drugs.

"The patch may be too sensitive," said Smith, who co-authored the study by the U.S. Naval Research Laboratory in Washington, D.C. Smith said he has been contacted by Alfonso's defense lawyer and may testify about the results of the study.

PharmChem Inc., which makes the sweat patches, claims it would require large quantities of drugs placed on the patch to go through the membrane and create a false positive result.