In a cramped, dimly lit room next to the Oklahoma State Penitentiary’s death chamber, three volunteer executioners push syringes full of lethal drugs into the veins of an inmate they cannot see.

Sometimes the executioners use flashlights to illuminate what they are doing in the drug room. While the three can hear what is said in the death chamber, they use a makeshift system to communicate with the warden, doctor or others in the room.

The executioners stick colored pencils through holes in the wall where two IV lines feed into the inmate’s body.

“If you saw red, there might be possible problems,” a deputy warden explained in a deposition.

The 2007 description of the chaotic scene in the state’s death chamber was documented as part of a legal challenge to Oklahoma’s lethal injection process.

“I use the flashlight to make sure that all the drugs are in the right order,” the deputy warden explained during his deposition.

The state’s failure to require executioners to monitor inmates during the process or require them to receive any training is among serious flaws in Oklahoma’s lethal injection protocol, a Tulsa World investigation has found. When compared to policies in other active death penalty states, Oklahoma’s protocol falls short in key areas that could lead to more botched executions.

The World reviewed execution protocols or policies in Oklahoma and 19 other states that have carried out the death penalty since 2008 and compared each state’s requirements based on 10 factors that can affect the outcome of executions. While the death penalty is legal in 32 states and the federal government, 20 states have carried out an execution since 2008, records show.

The state’s vague requirements for training, qualifications of participating medical staff and procedures to check an inmate’s consciousness may have all played a role in the April 29 botched execution of Clayton Lockett. Preliminary results from an autopsy sought by defense attorneys indicate Lockett’s IV was not properly placed and cast doubt on the state’s claims that his vein “collapsed.”

Among the World’s findings:

No regular training is required of Oklahoma’s execution team. In about half of the states that make detailed protocols public, team members must train for the procedure on a regular basis.

No backup drugs are required by Oklahoma in case the initial doses of lethal drugs don’t work. In Lockett’s execution, the Department of Corrections had no backup drugs on hand when things went wrong, leaving him to die on the gurney after the execution had been halted. Nine states surveyed by the World had protocols requiring a second complete set of drugs for use in such cases.

No specific procedures are required in Oklahoma to ensure an inmate is unconscious. A physician in the death chamber pronounced Lockett unconscious three minutes before he began writhing, mumbling and rising from the gurney. In five states, officials had specific directions on how to determine consciousness.

No method exists for DOC’s director to modify the state’s execution protocol. Director Robert Patton complained about his lack of authority over the process in a letter to Gov. Mary Fallin following Lockett’s execution. However, the current policy was approved while Patton was in charge of the agency. No state protocol reviewed by the World left a prison warden exclusively in charge of deciding how an inmate would be put to death.

No preference is expressed for a more reliable one-drug protocol. At least eight of the states surveyed by the World have switched to using only one drug in lethal injections, which many experts agree is less error prone. Last week, both Missouri and Georgia executed inmates without incident using the single-drug method. Oklahoma has a one-drug option among its five drug combinations but chose not to use it.

The Oklahoma Attorney General’s office defended the current protocol during a series of last-minute court challenges by Lockett’s attorneys. The office cited a 2006 Court of Criminal Appeals ruling that stated: “Risk of accident cannot and need not be eliminated from the execution process in order to survive constitutional review.”

The state does make its protocol publicly available, unlike states including Alabama and Virginia. Several states also had protocols with even fewer safeguards than Oklahoma. In Mississippi, an official state executioner serves at the pleasure of the governor and is paid $500 per execution.

New drug problematic

Oklahoma wrote its new protocol after telling the state Supreme Court it had run out of drugs, delaying the executions of Lockett and Charles Warner. The protocol gives the Oklahoma State Penitentiary warden “sole discretion as to which lethal agent will be used for the scheduled execution.”

For Lockett’s execution, the state chose a new drug combination using the sedative midazolam. The drug has been used by three states — Ohio, Florida and Oklahoma – in nine executions. Two of those were clearly botched, while a third took longer than expected, according to witness and news accounts.

Because the second drug in three-drug protocols paralyzes inmates, witnesses often would not know whether the first drug worked — possibly leading to other botched executions that went unnoticed. Without a proper dose of anesthetic, the third drug, potassium chloride, would cause extreme pain, something attorneys for the state of Oklahoma acknowledged in a 2010 lawsuit.

“Defendants admit that it would be painful to receive a concentrated dose of potassium chloride without first receiving an anesthetic,” states the filing by Assistant Attorney General Stephen Krise.

Krise is now general counsel at the Department of Public Safety, where Commissioner Michael Thompson is leading the state’s investigation into Lockett’s execution.

Dr. Jay Chapman is a retired forensic pathologist and former chief medical examiner who developed Oklahoma’s original lethal injection protocol. It was later copied by other states and nations seeking a more humane way to execute people.

Chapman, who supports the death penalty, dismissed claims that the three-drug protocol and others like it are “untested.”

“How did they want them tested? Did they want someone to go out and take a bunch of prisoners and see what happens?”

While Oklahoma has consulted an expert in the past — Massachusetts anesthesiologist Mark Dershwitz — the state’s latest protocol involved no expert input. Instead, Pruitt’s office took testimony Dershwitz gave in a Florida court case and used it to justify Oklahoma’s use and amount of midazolam, records show.

However, Florida uses 500 milligrams of the sedative, five times the amount called for in Oklahoma’s protocol. Also, comparing state protocols this way is something Dershwitz is professionally barred from doing.

“I am not allowed to compare one state’s protocol to another. … What I have testified in the past, and this is a matter of public record, is that when I have used midazolam for inducing general anesthesia, my dose is around 40 or 50 milligrams.”

But experts in that case and elsewhere have said midazolam is not typically used by itself to anesthetize people before surgeries.

Frank Romanelli, associate dean and professor at the University of Kentucky’s College of Pharmacy, said the drug is used to produce “conscious sedation” for procedures such as wisdom tooth removal and colonoscopies.

“Sometimes it might be used in anesthesia as sort of an induction ... before you get the full blown anesthesia.”

IV errors

Dershwitz, who has testified on behalf of states in lethal injection challenges, said botched executions occur due to errors by medical personnel and not the drugs themselves.

“When an execution has gone awry, it has never been due to the drug or the choice of drug, the dose of the drug. It has always been due to the failure to deliver the drug into a vein,” he said.

Placing an IV into a person’s femoral vein requires a level of skill that phlebotomists and similar medical assistants lack, he said.

Surgeons, cardiologists and other specialists would be trained in starting such IVs, but “it would not be the sort of training I would expect a primary care physician to have,” Dershwitz said.

Exactly what type of medical professional placed Lockett’s IV remains unanswered. The agency’s official timeline reported to the governor states a “phlebotomist” assisted in locating veins for the IV. Phlebotomists aren’t licensed to start IVs in Oklahoma.

The state’s protocol requires an EMT-paramedic “or other licensed person” to start IVs.

When the World questioned DOC about the discrepancy, spokesman Jerry Massie said the reported timeline was wrong and that an EMT was actually present.

As far back as 2006, DOC records refer to using phlebotomists to insert IVs, including a federal lawsuit by two inmates.

Records show improperly placed IVs may have played a role in several other botched executions in the state.

In 2001, witnesses said inmate Loyd LaFevers repeatedly convulsed and his chest rose off the gurney during the six minutes it took him to die.

In an autopsy, an investigator with the Oklahoma State Medical Examiner’s Office states that LaFevers’ IV “infiltrated after the first drug was administered.”

“A prison unit manager had sent two letters to the clemency board stating that this inmate needed to die,” the autopsy report states.

“The family of (LaFevers) is accusing prison officials of diluting the execution drugs or gave it (sic) in the wrong order to make this inmate suffer.”

An expert testifying for plaintiffs in a 2006 lawsuit stated LaFevers had a level of anesthetic in his blood that was not sufficient to render him unconscious.

In 1992, it took inmate Robyn Parks 11 minutes to die, during which he remarked “I’m still awake.”

Witnesses said his body began bucking under the straps as he spewed the air out of his lungs. A reporter from the World who witnessed the execution described it as “overwhelming, stunning, disturbing.”

Consciousness checks

At least five state protocols surveyed by the World contained detailed instructions on how medical professionals should check consciousness of an inmate. Oklahoma’s protocol contains no requirements, leaving it up to the doctor.

Eight minutes after Lockett’s execution began, a physician in the death chamber looked closely at his face and placed a hand on his chest, shaking him gently. Lockett was pronounced unconscious. Three minutes passed with little movement from Lockett until his body began reacting fiercely.

Oklahoma’s protocol states the physician should monitor the offender “through whatever means the physician believes are appropriate to ensure that the condemned is sufficiently unconscious” before the paralyzing drug is given.

During Lockett’s execution, Patton ordered a halt to the process after being told Lockett was still alive but there were no additional drugs on hand.

It’s unclear why the prison did not use the lethal drugs it had available for the execution of Charles Warner, who was set to die two hours after Lockett.

There is no language in Oklahoma’s protocol about what the execution team should do if something goes wrong.

In the event of a stay issued after the execution has begun, the policy is to lower the shades “and medical personnel will take action immediately to render emergency measures.” Officials have refused to say whether they took such measures after Lockett’s execution was halted and his heart was still beating.

Ziva Branstetter 918-581-8306

ziva.branstetter@tulsaworld.com

Cary Aspinwall 918-581-8477

cary.aspinwall@tulsaworld.com