Death by lethal injection, adopted in 37 states as a painless method of execution, may actually inflict unnecessary suffering on the condemned because of a routine failure to use enough anesthesia, according to a study of death row autopsies.
An examination of 49 autopsies found that in 43 cases, the concentration of anesthesia in the bloodstream was less than that required to numb a surgical patient before making an incision. In 21 cases, the concentration wasn't sufficient to prevent a patient from responding to a verbal command.
Because other chemicals are used to induce paralysis, inmates do not appear to suffer during their executions. But the toxicology data indicate that the condemned may be fully aware as they are dying, the study showed. It was released Thursday by the British medical journal Lancet.
"We know you need a certain amount of this drug in your blood to be asleep, and when this drug was measured in people a short time after they were killed, it wasn't enough," said Dr. David Lubarsky, chairman of the department of anesthesiology at the University of Miami/Jackson Memorial Hospital and one of the study's authors.
But Dr. Mark Dershwitz, vice chairman of the anesthesiology department at the University of Massachusetts, said he thought the study misinterpreted some of the data from the autopsies and thus exaggerated the problems with lethal injection.
"Any experienced anesthesiologist will tell you that at these doses people will remain unconscious for far longer than the 10 minutes it typically takes to complete the execution," said Dershwitz, who has submitted affidavits and testified in court that lethal injection is humane.
The three-part regimen of sodium thiopental for anesthesia, pancuronium bromide to induce paralysis and potassium chloride to stop the heart was developed in 1977 by an Oklahoma state legislator who opposed the death penalty and wanted to impose it as painlessly as possible. It has been widely adopted since.
Around the country, 789 people have been executed by lethal injection, the Death Penalty Information Center says.
"We thought lethal injection was the most humane possible way to execute people on death row," said Scheree Lipscomb, director of public affairs for the Department of Corrections in Georgia, which adopted the practice in 2001.
But objections to lethal injection have been rising among death row inmates, who have filed claims in many states arguing that the procedure violates the Constitution's 8th Amendment prohibition against cruel and unusual punishment.
Last year, a judge in New Jersey ordered a halt to executions until the Department of Corrections sorted out medical issues related to the procedure.
A judge in Kentucky will consider similar claims at a hearing Monday.
States typically administer 2 grams of thiopental at the beginning of an execution, enough to keep a 220-pound man anesthetized for at least 10 minutes, according to the Lancet study. That was presumed to be long enough to keep an inmate unconscious while potassium chloride is injected into his blood.
"Administration of potassium will cause nerve fibers to fire, and you really get a profound burning sensation," said Dr. Leonidas Koniaris, a surgical oncologist at the University of Miami Miller School of Medicine who cowrote the study.
But 2 grams of anesthesia may not be enough in all cases.
On average, it takes 8.4 minutes to complete an execution, and many executions last more than 10 minutes, the study found. If a mistake is made by the person inserting the IV line -- typically a medic or other technician rather than a doctor or nurse -- the amount of anesthesia that gets into the bloodstream may be less than 2 grams.
In addition, an inmate facing death is likely to be anxious and fearful and would require more sedatives than a typical surgical patient. A history of drug abuse could also require more thiopental, the researchers said.
"It's a process fraught with potential for error," Lubarsky said.
States could respond by increasing the amount of anesthesia given to inmates; California, for example, requires 5 grams of thiopental. But a higher dose wouldn't necessarily alleviate what the researchers found to be evidence of frequent problems in the way the anesthesia was administered.
The researchers -- two physicians, a scientist and a criminal defense lawyer who represents death row inmates -- studied toxicology data from executions in Arizona, Georgia, North Carolina and South Carolina. Those postmortem measurements of anesthesia levels in blood offer a reasonable estimate of concentrations at the time of execution because anesthesia does not degrade once the body stops metabolizing it, Lubarsky said.
The amount of anesthesia in the blood varied widely from case to case, suggesting the technicians who administered the anesthesia weren't skilled enough to do it in a consistent manner, the researchers said. Doctors and nurses are precluded by medical ethics from participating in executions.