When Trentina Perry met Dr. Dinesh Shah for the first time on May 12, 2000, both had much to lose. Perry was pregnant and three days past due with a large, healthy baby. She had already named the boy D'Antre. Shah was director of the maternal-fetal medicine department at MacDonald Women's Hospital in Cleveland, a prolific researcher nearing the end of his distinguished career.
Neither recalls their first meeting very well. Only medical details are known. Shah rubbed cold petroleum jelly on Perry's taut stomach. Then he pressed two white transducers to her skin. When the ultrasound was over, Shah reported to Perry's primary physician, Dr. Charles Cowap, that Perry's baby was doing fine. She had enough amniotic fluid to keep him safe for another week.
Six days later, Perry returned to the hospital with packed suitcases, ready to deliver. She was not having regular contractions, but the baby now was almost two weeks past due. The womb that had nurtured her son for nine months was becoming dangerously cramped.
On his first try with a portable ultrasound machine, Dr. Cowap couldn't find the baby's heartbeat. He wasn't concerned -- babies often turn in the womb, making it difficult to find a pulse. Cowap, a resident, summoned two more experienced doctors. Neither of them could find the baby's heartbeat either. Perry was sent to Dr. Shah's department for a more accurate test. Still, Perry wasn't worried. "He was moving the night before and in the morning, so it didn't cross my mind," she later said in court.
Even with more powerful machines, doctors could not find a pulse. Perry's baby was dead. She spent seven hours in labor to deliver her stillborn child. Afterward, the first person she called was her best friend, Lashonda Berry. "She was hysterical," Berry later said in court. "She was screaming into the phone, and the only thing that I could do was hang up."
By the time Berry arrived at the hospital, Perry's family and friends were already in mourning. "Trentina was caressing the baby as if the baby was alive," Berry said. "And when she gave me the baby, she said, 'Isn't he cute, isn't he adorable?' It was almost frightening . . . She just kept saying that 'my baby is not dead.'"
It took years for Perry to discover how her son died -- and how he might have been saved. That's when she learned how far Shah would go to protect his reputation.
Trentina Perry's parents never married and fought often, usually about money. Her mother became especially angry when the ice-cream truck rolled down the street and Perry's father would buy treats for all the neighborhood kids.
Perry's father died when she was 11, according to her testimony. (Perry declined to be interviewed, and Shah did not respond to requests. All information presented here comes from court documents, unless otherwise noted.) Her relationship with her mother "got hard," she says, and she was kicked out of the house. Perry bounced between the homes of cousins, her grandmother, and her older sister. She attended high schools all over the city, from John Hay to Shaw and back to John Hay, then to West Tech and later East High. She fell behind, and by age 17 she was still a sophomore. She found a job changing diapers at a child-care center. In 1996, she became pregnant herself.
Tests showed that the baby was healthy, but Perry's body was not ready to deliver. Two weeks after her due date, she still had no contractions. She was sent to Dr. Kevin Muise, a gynecologist at University Hospitals, for an ultrasound test. Muise found that Perry's amniotic fluid was dangerously low. It's a life-threatening condition called oligohydramnios, in which the baby becomes increasingly confined. There's also a greater risk that the umbilical cord will wrap around the baby's limbs or head, cutting off the flow of blood.
The risks only grow as the pregnancy continues. The placenta hardens near the end of a full-term pregnancy, making it more difficult for it to supply oxygen and nutrients to the baby. This causes amniotic fluid levels to drop even further. "If I thought it was oligohydramnios, but I wasn't really sure, I would err on the side of caution," Muise later testified.
He didn't even bother finishing the ultrasound. Muise called Perry's doctor and recommended that Perry give birth immediately. She was admitted to the hospital and given drugs to induce labor, but her cervix still refused to open. So Muise performed the cesarean section himself. Damonte Perry was born healthy two days before Christmas.
Perry returned home to her duplex on East 116th Street. Hers is the best-kept yard on the block, with trimmed green grass and flower boxes on the porch. But the white paint on the house has mostly chipped away, exposing the bare, battered wood beneath. Just down the street is the Union Drive Thru liquor mart, a cinder-block eyesore where patrons keep their car stereos cranked loud enough to rattle windows. The music booms from the other end of the block too, where young men wash cars in the parking lot of an abandoned gas station late into the night.
Perry found a job as a home-health aide. She tried returning to high school, but the combination of work and raising a child proved too demanding, so she dropped out and earned a GED. On a loud street, she lived a quiet life.
Like Perry, Dinesh Manilah Shah has lived a nomadic life. He was born in rural India and gained entrance to a Bombay medical school in 1968. There he met his future wife, Neena. In 1973, Shah started a three-year residency in obstetrics and gynecology. "I felt I could make a difference in the lives of women and children," Shah said in court. "This was the primary purpose I took up gynecology as a specialty." (He did not return repeated phone calls seeking comment for this story.)
Shah served his residency at University of Chicago Hospitals. "I recognized that the center of knowledge for this field and many others was the United States," he said. From Chicago he moved to Syracuse and then Tampa to complete his medical training. In 1983, he landed his first full-time job, at Vanderbilt University Medical Center in Nashville, where he became a professor and began his practice.
Shah also started his career as a researcher. He published 15 academic articles in 1989 alone, including one in the American Journal of Obstetrics and Gynecology, in which he proposed changing the way doctors perform ultrasound tests. Traditionally, ultrasounds checked four things: Is the fetus breathing? Is it moving? Does it have good muscle tone? Is there enough fluid to keep it safe? Each of these tests was given equal weight in what's called a biophysical profile, which scores the health of the fetus.
Shah suggested that amniotic fluid be removed from the health score. Doing so would allow him to see more pregnant women in less time, without increasing health risks. To support his claim, he excluded amniotic fluid from the health scores of 180 pregnant women. He compared the results to 172 women tested the traditional way and found no difference between the two groups.
But the rest of the medical community rejected Shah's arguments. Critics published a deluge of studies proving that amniotic fluid is a crucial way to gauge the health of a baby. Fifteen years later, medical schools and textbooks continue to teach that fluid should be checked as part of any complete ultrasound.
The adverse reaction to his proposal didn't harm Shah's career. A year after publishing his controversial article, he accepted a job in San Antonio. In 1997, University Hospitals of Cleveland recruited him to become the system's new director of maternal-fetal medicine. He oversaw a staff of doctors and ultrasound technicians in his hospital ultrasound lab. He also joined Case Western Reserve's School of Medicine as a tenured professor. He had won major research grants from the American Heart Association and the National Institutes of Health, and sat on the NIH committee that decided which researchers deserved federal funding.
Shah's workday included a long commute between the hospital's headquarters in University Circle and his home in a subdivision of enormous limestone houses in Pepper Pike, where he and his wife raised their two children. All the while, Shah was performing ultrasounds on hundreds of pregnant women every year. At trial he said of Trentina Perry, "I don't recall her face as a person."
On May 12, 2003, pregnant women were lined up on the third floor of MacDonald Women's Hospital. Ultrasounds were scheduled in precise shifts of 30 minutes each. Trentina Perry's started like all the rest. Probing her belly, Shah checked for fetal breathing, movement, and muscle tone. All looked normal on the grainy screen of the ultrasound machine.
Then Shah used high-pitched sound waves to measure pockets of amniotic fluid in Perry's womb. When he was done, he added all the different measurements together. Any total below 50 millimeters is widely considered dangerously low. In fact, taped to the side of Shah's ultrasound machine were photocopies of a chart published in the medical textbook Maternal-Fetal Medicine, Fourth Edition. The textbook recommends an even higher cutoff point, saying that any total below 80 millimeters could be dangerous for a baby like Perry's, which was almost two weeks past due.
Shah found pockets totaling only 44 millimeters in size. The machine automatically kicked out a report with the diagnosis: oligohydramnios. But instead of calling Perry's primary doctor and urging immediate delivery, as Dr. Kevin Muise had done in the same situation three years before, Shah conducted a different test. This time, he found and measured the largest single pocket of fluid he could find. Using a computer attached to the ultrasound machine, he drew two lines through the pocket. If he could draw lines at least two millimeters tall and two millimeters across, that would mean Perry's fetus was safe.
But when dealing in such fine measurements, details are crucial. To ensure that the pocket was actually large enough to keep the baby safe, Shah's lines had to be perpendicular to each other.
Shah found a pocket that was more than tall enough. But it was narrow. The only way he could get a measurement of two millimeters across was to squeeze in the line, making an extreme angle of almost 45 degrees. Instead of a cross, the shape he created looked more like a slice of pizza.
Nevertheless, Shah gave Perry a perfect health score. Within minutes, that result was phoned in to Perry's physician, Dr. Cowap. Ten days later, the hospital's interoffice mail system finally delivered to Cowap the original paperwork from Shah's test, with the troubling diagnosis written across the top. By then, of course, it was five days too late.
The case of Trentina Perry vs. Dr. Dinesh Shah was assigned to Judge John Angelotta, who had retired in 1996 after 22 years on the bench. He returned as part of the visiting-judge program, to help the overloaded courts handle backlogged cases.
Angelotta's record was less than distinguished, however. In at least two cases, one in 1995 and another in 2003, federal appeals courts overturned his decisions in strongly worded rebukes, saying that he had disregarded basic rules of evidence and failed to give correct instructions to the jury.
Angelotta was best known for presiding over the case of Michael Levine, who was charged with the 1979 kidnapping and murder of Julius Kravitz, a Cleveland supermarket executive. Angelotta found Levine not guilty by reason of insanity. Levine was hospitalized for 14 years, even though state psychiatrists found as early as 1982 that he was quite sane. When Levine finally was let go in 1993, he told the press that the only reason he escaped a jail sentence was that his lawyer was good friends with Angelotta.
A TV reporter attempted to interview Angelotta about Levine's comments in the courthouse cafeteria. Angelotta pushed the reporter, threw a tray of food at him, and threatened to beat him with the tray. "I absolutely could not control myself," Angelotta later told the Ohio Supreme Court's discipline commission. A three-member panel recommended that the judge be punished, but the charges were later dropped.
Three years to the day after Shah and Perry met in a hospital examination room, they met for the second time in Angelotta's courtroom. Shah's attorney, Stephen Walters, assured the jury of the strength of the doctor's case. "First of all, a picture is worth a thousand words," Walters said. "The evidence will be overwelming in this case that there was a two-by-two pocket" of amniotic fluid. He continued, "And when you determine that, with regard to Dr. Shah, you have nothing else to determine. He will have met the standard of care."
But that would be contradicted on the third day of trial, when Dr. Michael Cardwell took the stand. Cardwell, who had been director of departments like Shah's for 22 years, asserted that Shah conducted the test incorrectly. "The measurement should have been perpendicular," Cardwell said. "That's the standard way to do it."
Since the ultrasound machine had already diagnosed Perry with low amniotic fluid, and since Shah couldn't find a big enough pocket of fluid, Shah's course of action should have been clear. "The baby needs to be urgently delivered," Cardwell said. "In some cases doctors can use their judgment, but this is not a judgment call."
Cardwell said that Shah's failure to follow standard procedures led directly to the baby's death. "In my opinion, had the test been interpreted and had Dr. Shah made the appropriate recommendations, the baby would have been delivered on May 12 . . . in an alive and healthy condition."
The jury took a break for lunch. When they returned, Shah's attorney called Dr. Cardwell back to the stand. That's when things got strange.
"Let me show you a picture and ask you if this is a perpendicular line," Walters said. He lifted a poster-sized photograph and turned it so Cardwell and the jury could see. On it was the hazy image Shah took when he measured the largest pocket of fluid. But something was different. In addition to the lines originally drawn by Shah, which formed a narrow wedge, now there was a new set of lines, which met at a perfect right angle.
"Looks perpendicular, doesn't it?" Walters asked.
Cardwell was beside himself. "Well, the, the problem with this is it's a distorted image, and I see what you've done," he stammered. "It's a retouched photograph. It's like something you'd see in the National Enquirer."
"That's where you're wrong, doctor," Walters responded. "It is the actual ultrasound image that was done on Trentina Perry."
But in the next breath, Walters explained the origins of this new piece of evidence. After the the first day of trial, Shah drove to his lab. With the help of a technician, he transferred the original image of Perry's test from the ultrasound machine to a different digital format. Then he blew it up and used a computer to add two new lines. Unable to find an adequate fluid pocket when he originally did the test, Shah testified that now, three years later, he had discovered one. Though he'd resized the image, with no one else watching and his career on the line, he promised that he had not made the pocket appear bigger than it really was. "Absolutely. In God's name," he said.
Dennis Mulvihill, Perry's lawyer, was more than skeptical. "Your Honor, he's introducing evidence that was never disclosed at any time in this case," he said. "This is trial by ambush!"
Judge Angelotta overruled Mulvihill's objection. Walters continued asking questions about the new ultrasound image, but Cardwell refused to cooperate. "Sir, I gave my deposition in this case in my office in February of last year," Cardwell said. "Why was I not provided this evidence at the time of my deposition?"
So Walters switched tack. He asked Cardwell about Rockford Memorial Hospital in Illinois, the doctor's former employer. "Is it not a fact, doctor, that you were approached by the hospital about issues regarding drug abuse, regarding alcohol abuse, regarding mental incompetence?"
Mulvihill objected, but again he was overruled.
Back in 1985, Rockford Memorial conducted a private investigation of Cardwell. Cardwell sued, claiming that the hospital was trying to damage his reputation because he was the only maternal-fetal specialist in town, and he had decided to move his practice to a competing hospital. Rockford Memorial backpedaled, dropping the investigation and promoting Cardwell to director of its own maternal-fetal department. The move hardly suggests that the hospital truly believed Cardwell was an incompetent dope addict.
So when Walters brought up the allegations, Cardwell exploded. "How can a false accusation against me in 1985 be of any relevance today?" he said. "Dr. Shah cannot really in my opinion defend himself based on the protocol and the images. And now you are attacking me!"
Judge Angelotta finally realized his mistake and intervened: "Listen, this may be my fault, doctor, you know. Maybe I should have listened to Mr. Mulvihill, who didn't want the jury to hear these things."
In the confusion, jurors may not have noticed that Walters never questioned Cardwell about the image created during Shah's initial scan of Perry. And they had no way of knowing that he never called his own expert witness to the stand. Stephen DeVoe, a Columbus gynecologist with nearly 30 years' experience, had been hired to review the evidence and testify on Shah's behalf. But in a sworn statement a month before the trial, DeVoe too said that Shah was wrong: "If the lines aren't perpendicular, then it's not done correctly."
Shah's best defense came from Dr. Richard Beigi. Beigi performed an ultrasound three days after Shah and found Perry's amnitoic fluid levels normal. Though he never recorded his intial measurement, Beigi testified that he remembered getting a measurement of 75 millimeters because he was on the hospital floor when Perry came to deliver her stillborn baby. "It kind of stuck in my mind from that," he said.
Dr. Cardwell doubted the result. The amount of fluid in the womb tends to decrease after the due date, he testified, not increase. This was countered by Dr. Harland Giles, the expert witness for University Hospitals, who said it was not uncommon to see amniotic fluid increase after a woman's due date.
After four and a half days of dueling expert witnesses and drama, the jury decided in favor of Shah.
Perry immediately appealed to federal court. A year later, she got a response.
Most court decisions are written like Byzantine lullabies, packed with legal minutiae and references to obscure cases. This decision of the three-judge panel of Ohio's Eighth District Court of Appeals is different. Judge Sean C. Gallagher, who wrote the decision, borrowed Mulvihill's words "trial by ambush" when referring to the new ultrasound image: "The exhibit in this case was an electronically manufactured image that Dr. Shah did not present until the time of the trial."
Gallagher went on to carve Shah's defense -- and Angelotta's decisions -- into little pieces. Perry's case was severely wounded when Angelotta "abused his discretion" by allowing Shah to introduce his altered images. Also, Angelotta never should have allowed Walters to question Cardwell about old allegations of drug abuse, since "No evidence was admitted or even offered to substantiate these allegations."
Gallagher's decision was a rout. "We agree with Perry," the judge ruled. "Had the court's errors not occurred, the jurors may have reached a different decision." Gallagher ordered a new trial. Shah and University Hospitals immediately appealed to the Ohio Supreme Court, which has not yet decided whether to hear the case.
Since the death of D'Antre Perry, many people involved in the case have moved on. Dr. Charles Cowap, Perry's primary physician during the pregnancy, finished his residency at University Hospitals and is now in private practice in Westlake. Judge John Angelotta left the bench this year after turning 80, the mandatory retirement age for judges in Ohio.
Dr. Dinesh Shah has moved most frequently. He left Cleveland in December 2002 and took a job at University of Louisville Hospital. On Dec. 1, 2004, he was scheduled to begin his latest job. Shah is the new director of maternal-fetal medicine at the University of Wisconsin, Madison, where he will continue teaching, researching, and performing ultrasounds on pregnant women.
Meanwhile, Trentina Perry is left with her job, her pain, and Damonte, her surviving son. "Damonte was excited because he was going to have a little brother," she said at trial. "He wanted to take the baby home with him. I had to explain that we couldn't take him home. I told him that he was up in heaven."
D'Antre Perry lies in an 18-inch coffin at Cleveland Memorial Gardens, buried in an unmarked grave.