The state has conducted autopsies on less than half of the inmates executed in Oklahoma since 1990 and, in many cases, does not perform tests that could show whether inmates were awake and paralyzed as painful drugs flowed into their veins, a Tulsa World investigation has found.
Because state records are inconsistent and blood is sometimes drawn long after inmates die, it is difficult to say how many inmates were conscious when they received potassium chloride, the third drug in Oklahoma’s lethal injection process. Medical experts, judges and attorneys for the state agree that potassium chloride is excruciatingly painful if given to a conscious person.
The botched execution of Clayton Lockett on April 29 has sparked a nationwide discussion about the death penalty and new scrutiny in how Oklahoma and other states put people to death.
The World created a database using 109 medical examiner’s reports from Oklahoma inmates executed since 1990, including the levels of anesthetic in their blood following death where available. Experts in anesthesiology and clinical pharmacology reviewed the data to spot issues and problem cases.
Among the World’s findings:
The Oklahoma State Medical Examiner’s office conducted full autopsies on all executed inmates until 2000, when state lawmakers passed a measure banning such autopsies except in narrow circumstances. Only 10 out of 50 inmates executed since the law was passed received an autopsy.
Weights of executed inmates varied widely, from 145 pounds to 317 pounds, yet all received the same doses of drugs under the state’s execution protocol. Those inmates had widely varying levels of anesthetic in their blood.
Full or nearly full syringes of anesthetics were returned with inmates’ bodies in at least four cases, an apparent violation of Department of Corrections protocol.
Handling of evidence varied randomly from case to case, with some blood samples being held 30 days and others for five years.
More than 30 inmates had levels of anesthetic in their blood below what experts say might be necessary to produce sufficient anesthesia. In many cases, state pathologists drew blood hours or days after inmates’ deaths and tested it despite those samples being possibly useless in accurately reflecting drug levels.
Medical experts consulted by the World say Oklahoma should investigate the inmates’ deaths and revise its lethal injection protocol to ensure inmates do not suffer “cruel and unusual” punishment during executions, violating the Eighth Amendment of the U.S. Constitution.
“I don’t have enough data, but there is cause to investigate … because of the potential for underdose,” said Frank Romanelli, professor and associate dean at the University of Kentucky’s College of Pharmacy.
Romanelli has advocated for pharmacy students to learn about lethal injection protocols. He has not testified in legal challenges to the death penalty.
Officials with the Oklahoma DOC and Office of Chief Medical Examiner declined to comment on the autopsy data gathered and analyzed by the World. Attorney General Scott Pruitt’s office did not respond to a request for an interview.
During challenges to Oklahoma’s lethal injection protocol over the past decade, attorneys for the state have insisted their method is constitutional.
Key court rulings have found that while inmates cannot expect a pain-free death, the state cannot cause more pain than necessary during executions.
While ruling in favor of states’ rights to execute by lethal injection in 2008, the U.S. Supreme Court said it was “uncontested” that without a proper dose of anesthetic administered to the inmate “there is a substantial, constitutionally unacceptable risk of suffocation” from the final two drugs used in most state’s methods, including Oklahoma’s.
In the years since that ruling, at least seven death penalty states have switched to single-drug lethal injection protocols. States also have scrambled to find new drugs as manufacturers have banned the use of their drugs in executions.
Oklahoma’s protocol allows DOC to choose one of five drug combinations, including a single dose of an anesthetic called pentobarbital. The state opted instead to use a new three-drug combination featuring a sedative called midazolam in Lockett’s execution.
Romanelli said if inmates do not receive a sufficient dose of the initial anesthetic, they may be conscious yet paralyzed before the third drug, potassium chloride, is administered, “which is extremely painful.”
In at least 32 cases from 1990 through 2014, Oklahoma medical examiner’s reports show levels of anesthetic in inmates’ blood were below the threshold used by doctors to render patients unconscious. Each of those executions used a barbiturate — either sodium thiopental or pentobarbital — as the anesthetic in the three-drug cocktail.
Romanelli said the time lapse between when drugs are administered and when blood is drawn heavily influences those test results and the ability to interpret them. In some of the 32 cases, pathologists took samples within an hour, while in others they waited more than one day, records show.
Dr. Mark Heath, a board-certified anesthesiologist and a professor of anesthesiology at Columbia University, also reviewed the inmate autopsy data at the World’s request. Heath has provided expert testimony in challenges to lethal injection protocols in Oklahoma and 15 other states.
Heath said the Oklahoma autopsy data shows “that all the prisoners did receive thiopental or pentobarbital. It does not show whether it was properly delivered into their circulation or whether it was the first drug or the second drug.”
Heath said cases where the level of drugs are below the anesthetic threshold should be investigated to determine why.
In those cases, “that low level is inconsistent with being anesthetized,” he said.
Additionally, the full dose of lethal drugs is not always given, records show.
In at least four cases, full syringes of drugs — including the initial anesthetic — were sent with inmates’ bodies to the medical examiner. The practice appears to violate DOC’s protocol, which requires all drugs to be injected.
In one such case, inmate Olan Robison took 15 minutes to die, according to the medical examiner’s report. One of two syringes of sodium thiopental was returned nearly full with Robison’s body on the night he was executed, March 13, 1992.
Robison’s autopsy shows multiple needle punctures on his arms and bleeding below the surface, indicating problems with his IV.
Law bans autopsies
Preliminary results of an autopsy sought by attorneys for death row inmates indicate Lockett’s IV was inserted improperly. Lockett spent three minutes writhing, mumbling and rising from the gurney before prison officials closed a blind and discovered the drugs had leaked out of his IV.
Department of Corrections director Robert Patton ordered Lockett’s execution halted because the prison had no backup drugs on hand and medical personnel said they could not locate another suitable vein. Lockett died on the gurney 10 minutes later.
Gov. Mary Fallin called for an independent autopsy by the Dallas County Medical Examiner’s Office, which is pending. An investigation headed by Fallin’s public safety commissioner, Michael Thompson, is also pending.
Lockett was sentenced to die in the 1999 killing of Stephanie Neiman, 19, of Perry. Lockett and two accomplices abducted Neiman, two of her friends and a baby.
He shot Neiman twice and ordered her buried in a shallow grave beside a rural road, despite being told she was still alive.
While the state has ordered a full, independent autopsy for Lockett, autopsies are rare following executions in Oklahoma, the World’s investigation shows.
A state law passed in 2000 amended existing law that required the state medical examiner to investigate deaths of all prisoners.
“No autopsy shall be performed on the body of an executed inmate unless requested by the immediate family of the inmate prior to the execution or unless directed by the Department of Corrections or the Chief Medical Examiner,” states the law, sponsored by then-Sen. Glenn Coffee.
The law states the medical examiner “shall not automatically authorize or perform an autopsy” in such cases.
“The Chief Medical Examiner may authorize or perform an autopsy only when the public interest requires it,” the law states.
When Michael Lee Wilson was executed Jan. 9, his last words were “I feel my whole body burning.”
In that case, no autopsy was conducted. A brief medical examiner’s report fails to note how long Wilson took to die or placement of his IVs.
“No acute injuries noted,” it states.
Several death penalty states require coroners to conduct autopsies or investigations following executions, including South Dakota and Georgia.
Romanelli said the state should conduct autopsies and collect better data on levels of drugs in inmates’ blood to ensure the process is constitutional.
“If you have the data you can show one of two things,” he said. “The prison system can show it does what we say it does and the folks were sufficiently anesthetized or we can show it doesn’t do what we say it does.
“Without the data, I don’t understand how anyone could come to the conclusion that it meets the constitutional test.”
Heath said the state’s use of a paralyzing drug is akin to pulling a blind down during an execution, as occurred in Lockett’s case. The drug serves no purpose other than to prevent witnesses from seeing an inmate thrash around, he said.
“In places where prisoners are being paralyzed … that conceals the critical information about whether or not it was done legally and constitutionally and humanely,” Heath said.
Statements and reactions witnessed during the execution of several of Oklahoma’s condemned inmates indicate they could have been conscious while the second or third drugs were administered.
On Jan. 7, 1999, Kay County killer John Castro said, “I can feel it,” after the drugs began flowing into his veins.
Seven months later, after Norman Newsted’s execution began, he remarked: “I can taste it.”
Both had levels of sodium thiopental in their blood below the level needed for sufficient anesthesia, records show.
The autopsy data shows a handful of cases in which medical personnel made multiple failed attempts to place IVs in inmates’ veins. DOC’s protocol requires an EMT-paramedic “or person with similar qualifications” to place IVs in the inmate’s arms. If the EMT has difficulty starting an IV, the physician takes over the process.
Medical personnel made seven attempts to start an IV on inmate John Bolz in 2006 before placing one in his femoral vein, a procedure also used for Lockett.
Records show the execution crew had to resort to using a single IV line in at least six cases.
Romanelli said starting IVs in such a high-pressure situation makes mistakes more likely.
“You take obese patients and patients with a lot of tattoos, and when you think about the high pressure situation of being in an execution environment. … It requires a higher level of expertise.”
World Researcher Hilary Pittman
contributed to this story.
Ziva Branstetter 918-581-8306
Cary Aspinwall 918-581-8477