Dozens of Infants Died Mysteriously at a Top Children’s Hospital—I Wanted to Know Why
Years after the suspicious deaths occurred at the very hospital where I spent time as a baby, I decided to investigate what happened. Had this been a mass murder or a tragic coincidence?
When our expert copy editor, Rob, sent back this piece after he had finished working on it, he wrote: “Boy, that certainly did not disappoint!” And that’s exactly how I feel, too. Leigh spent countless hours researching, reporting, writing and shaping this exhaustive story about a chilling mystery she felt a personal connection to — and it comes through in every sentence. This piece — the last in our series collab with Creative Nonfiction, Heart of the Matter — will have you at the edge of your seat, and then surprise you when you least expect it. Enjoy and please share! —Jesse Sposato, executive editor
Kevin Garnett was furious. He shouted and banged his fist against a wall inside Toronto’s Hospital for Sick Children, demanding to know why his son had died.
The boy, Kevin Pacsai, had been born just 25 days earlier with a head of dark hair like his 21-year-old father. His mother, Laurie Pacsai, also 21, thought her son was perfect. But Kevin’s heart had started to fail in his second week of life, and he had barely survived an episode of shock. His heart was structurally normal, but its conduction system seemed to be misfiring. Earlier that day, March 11, 1981, he had been transferred to the world-renowned hospital, known as SickKids.
Youngsters came from all over the globe to the imposing brick building, where doctors had invented techniques that revolutionized pediatric heart surgery. Cardiologists were venerated; they seemed to work miracles to prolong the lives of children born with faulty hearts. The previous spring, the hospital had unveiled a splashy renovation of its cardiac unit, which now stretched over wards 4A and 4B.
But since the summer, deaths on the cardiac unit had soared — 625 percent higher during the nine-month period between July 1980 and March 1981 than in the periods before and after, according to a later government analysis. Most of the deaths had happened overnight, when one of the eight nursing teams, headed by Phyllis Trayner, was on duty. By the fall of 1980, the five nurses had earned a nickname: “the jinx team.”
The cardiologists at the hospital had formally discussed the surge in deaths in the fall, around the time the nickname had surfaced, as well as a few months later in January. But they attributed the children’s deaths to their cardiac conditions and thought that the increase might be a natural cluster. “You have some in which you say, ‘This child was expected to die.’ They were seriously ill or they wouldn’t be in the cardiac unit,” a hospital administrator told the Toronto Star in a May 1982 article.
Kevin, though, didn’t seem to fit in that category. Doctors had told his mother, Laurie, to go home and get some rest; her son was out of harm’s way. He likely had a condition that would respond to treatment. Overnight, two doctors examined the baby and found his heart rate stable. They told the nurse who was looking after him, 24-year-old Susan Nelles, that they saw little cause for immediate concern.
But by around 4 a.m., Kevin’s condition had changed. His heart was oscillating back and forth between slow and fast beats. He was lethargic, his breathing shallow, his extremities blue. Dr. Colm Costigan, the hospital’s chief pediatric resident, checked the boy’s heart rate; it was slow. The doctor suspected Kevin might be suffering from an overdose of digoxin, a medication used to treat congestive heart failure.
Digoxin, a derivative of the foxglove plant, has been used for centuries to help the heart pump. Babies’ doses are tiny, measured in millionths of a gram; the line between a therapeutic dose and a toxic dose is narrow, making the risk of an overdose high, though not fatal in every case. Digoxin toxicity can manifest as an irregular or slow heart rate, vomiting, cardiac arrest and seizures, among other symptoms — all of which can also indicate unrelated cardiac issues. SickKids tested patients who were getting digoxin; if their blood showed a reading of 3.5 nanograms per milliliter (ng/mL) or more, doctors would put the medication on hold.
Dr. Costigan transferred Kevin to the intensive care unit and put a hold on his digoxin. For an hour, the baby’s heart rate returned to normal. By then the sun was up. At around 8:45 a.m., Kevin stopped breathing. The lower chambers of his heart started to contract rapidly, out of sync. Over and over again, Dr. Costigan and the arrest team tried to resuscitate the infant, but it didn’t work. At around 10 a.m. on March 12, 1981, Kevin was pronounced dead.
After the outburst from the boy’s father, Dr. Rodney Fowler, a senior cardiologist, called a coroner and asked him to investigate Kevin’s death. “Maybe it will give the father some relief,” he said, according to the coroner’s 2021 memoir.
Like Dr. Fowler, Dr. Costigan, who was the head of the cardiac arrest team, was also preoccupied with the death. The effort to revive the boy reminded him of the death of 20-day-old Jordan Hines a few days earlier. As with Kevin Pacsai, no matter what they did, no matter what medications they gave him, Jordan’s heart continued to spasm. The two deaths were similar, so Dr. Costigan was surprised to learn later that Jordan hadn’t been prescribed digoxin.
For Kevin, Dr. Costigan requested a digoxin test on a blood sample taken before the infant died. The results came back five days later: more than 10 ng/mL, two to three times higher than a potentially toxic dose and the highest number Dr. Costigan had ever seen. A postmortem blood sample showed an even higher number. He was convinced that digoxin had killed Kevin.
Justin Cook was an early Christmas gift for his parents, Jackie, 19, and Brad, 20. The couple had been living in Western Canada, but when Jackie got pregnant, they moved to Tara, a small town a few hours northwest of Toronto, where Brad worked on his parents’ 300-acre farm.
From the start, Justin did nothing but eat and only cried when he was hungry. One day in March of 1981, when the boy was about 3 months old, he turned blue, a sign that his heart wasn’t getting enough oxygen to his body. His parents bundled him into the car and drove southeast to SickKids, the baby watching the country slide by out the window.
When they arrived, a petite, 5-foot-tall nurse with a blonde bob greeted them. It was Susan Nelles. Just over a week had passed since Kevin Pacsai’s death. That night, Nelles had been looking after a nearly year-old girl named Allana Miller, who had a complicated series of heart defects and multiple spleens. But Justin’s admission was urgent, and Nelles accompanied him to get an ultrasound of his heart. Tests subsequently revealed that he had a constellation of complex cardiac abnormalities, and the hospital scheduled him for emergency surgery.
The next night, March 21, 1981, Nelles was assigned to look after Justin one-on-one, a practice reserved for the sickest babies. His mother, Jackie, later recalled to a reporter how Nelles had held Justin in her arms, sat beside his bed and rocked him, just as if she were his mother. “We felt he couldn’t be safer,” Jackie said.
Meanwhile, as Nelles had been occupied with Justin, Allana had gone into cardiac arrest and died. SickKids doctors met with hospital administrators, coroners and two police officers who were assisting the coroners with fact-finding. The doctors thought they had the situation under control, but the cops floated another possibility: murder.
Allana’s digoxin test came back at 8 p.m. that same evening. The results were more than 70 ng/mL. Off the charts.
The number echoed the reading of another infant who had died in the cardiac unit in January. Born with multiple complex heart defects, Janice Estrella had struggled for most of her short life. At 4 months, she was still the size of a newborn, and her mother considered her death inevitable, according to a 1981 news report.
Janice’s digoxin readings were high when she was alive, and after she died a pathologist tested her levels, likely the first time SickKids had analyzed a post-mortem sample for digoxin. At the time, the result of 72 ng/mL was so high that it looked like a mistake, as if a decimal were missing or the sample contaminated.
Now it didn’t seem like a mistake. The doctors met again and decided to act. They made digoxin a controlled drug, meaning it would no longer be available on open shelves but instead locked in the narcotics cabinet. Only a head nurse or team leader could dispense it. In the early hours of March 22, Dr. Costigan and a colleague visited the wards to disseminate the new protocols.
It would be too late for Justin Cook. At around 3:45 a.m., the baby turned bluer. He had a seizure. His heart rate slowed and the organ contracted wildly, a scene achingly similar to Kevin’s and Allana’s final hours. Justin was pronounced dead at around 5 a.m.
A couple of hours later, the boy’s father, Brad, came to the ward and implored Nelles that he be allowed to see his son. She took him to the morgue and, with tears in her eyes, told him how sorry she was that Justin had died.
The hospital got back the results of Justin’s digoxin test later that day. The sample, taken before the baby died, showed a reading of 72 ng/mL. Justin had never been prescribed digoxin.
This was now a murder investigation.
Staff Sgt. Jack Press, a balding veteran of Toronto’s homicide squad, and Sgt. Anthony Warr, who had just joined the squad, set up shop at SickKids. They took off their ties, put on white medical coats — the better to blend in — pored over the unit’s medical and staff records, and interviewed SickKids staff.
The cops quickly pieced together what some on the wards already knew: The same nursing team had been on shift when all of the babies known to have high digoxin readings — Janice, Kevin, Allana and Justin — had died. No doctor had been consistently on duty during any of the patients’ deaths, going back to January. And only one nurse had been assigned to care for Justin Cook one-on-one: Susan Nelles. She had also given Kevin Pacsai his last dose of digoxin. And Dr. Fowler told the cops that she’d had a strange look on her face as she was writing up Justin’s death report. No signs of grief.
Prosecutors agreed that the cops had grounds to arrest Nelles for Justin’s alleged murder. The nurse and her colleagues on “the jinx team” had been sent home after he had died, but they were scheduled to return for their shift that night. Other patients were at risk. The cops wanted to act.
Nelles lived on a quiet street not far from the hospital, in a brick duplex where she shared an apartment with her roommates. Working in the medical field was in Nelles’s DNA. Her father was the head of pediatrics at the hospital in Belleville, the small Ontario city where she’d grown up. In the 1950s, he had done his early training at SickKids. Nelles’s older brother, David, was then a resident at SickKids. A year after finishing nursing school, she started working there herself, and in June 1980, she got a position in the cardiac unit — her dream job.
Given the inquest into Kevin Pacsai’s death, prompted by the boy’s father, Nelles assumed the police would talk to her. One of her roommates, a law student named Alison Woodbury, advised her to get a lawyer and connected Nelles with her brother, who phoned with a couple of names.
Nelles wrote down the names and slipped the paper into the pocket of her robe. Not long after, at around noon on March 25, two men were at her door.
“Are you Susan Nelles?” the older one asked.
“Yes,” she replied.
Sergeants Press and Warr asked to come in. Nelles led them down a long hallway, and the three took seats at the kitchen table.
“Here were these two big burly police officers and little me,” Nelles recalled in 2015. “All I can remember at the time was feeling alone and so vulnerable.”
Press explained that Nelles didn’t have to talk — the standard “caution” that police officers give. “Justin Cook died of an overdose of digoxin, a drug he wasn’t supposed to have,” he continued. “We believe you gave him the drug, and we’d like to know why. Do you wish to give any explanation for his being given digoxin?”
Nelles remembered her roommate’s advice. “I think I want to speak to a lawyer,” she responded.
Press asked her if she had a particular name, and she pulled the piece of paper from the pocket of her robe.
He said she could make the phone call from the police station.
“Why the police station?” Nelles asked.
“Because I am arresting you on a charge of murder,” Press replied.
That night, the Toronto news was consumed with the story of the nurse who’d allegedly killed one of her infant patients — at the illustrious SickKids, no less. The Toronto Star chartered a private plane for a reporter and photographer to score the first interview with the Cooks, who were back in Tara mourning their loss.
They had buried Justin on March 23, believing he’d died of natural causes. “When they phoned to tell us he was gone, they said he took a blue spell and didn’t come out of it and that his system had just quit,” Brad Cook told the Toronto Star reporter. Now, on March 25, they heard the news on the radio: Their baby was the victim in a murder investigation.
At the precinct, Nelles was afraid to respond to any question where she wasn’t sure of the answer — but equally afraid to stay silent. In a police interview, she mostly responded, “I don’t wish to answer that at this time.” Her selective responses only confirmed the cops’ suspicion of her guilt.
Nelles was reportedly transferred to a 9-by-7-foot cell in a detention center with a bunk, desk, chair, open toilet and sink. She was held in segregation — her fellow inmates might not take kindly to an alleged baby killer. Nelles’s parents, James and Barbara, raced to Toronto to see their daughter. “We sat in a holding room — appropriately named because we sat and held each other and simply cried,” Nelles recalled in 2015.
The nurse was initially charged with Justin Cook’s death, and murder charges related to Janice Estrella, Kevin Pacsai and Allana Miller came a couple of days later. (That’s when Janice’s parents, Allana’s mom and Kevin’s parents learned that their children might not have died of natural causes.) After five days, Nelles was released on bail of $50,000 Canadian (around $170,000 Canadian in today’s dollars, or about $121,000 U.S.).
Nelles’s defense team was immediately convinced of her innocence. “She was clear eyed, reserved, polite, considerate, and above all, apparently quite sane,” eminent defense attorney Austin Cooper recalled in a 2005 article about his first meeting with his client.
The junior lawyer on the case, Mark Sandler, was tasked with reading the prosecution’s four-volume brief of evidence against Nelles. When he was done, he ran into Cooper’s office. “There is just no credible case against her whatsoever,” Sandler exclaimed, a presumptuous move for someone barely a year into his law career. “I don’t understand how they think they can sustain a case.”
The prosecution’s theory was that a single killer had committed the murders — no evidence of a conspiracy between any of the nurses had surfaced — and that Nelles had had the exclusive opportunity to kill Justin, as well as opportunities to kill the others. After consulting with a Chicago-based digoxin expert, they determined the fatal dose would likely have been administered intravenously within 15 minutes of the onset of the infant’s critical symptoms.
But Nelles had worked the day shift on the night Janice died; she had left the hospital hours before the baby’s death. (The cops initially theorized that problems with Janice’s IV might have caused the digoxin to hit her system more slowly, though they later abandoned that line of thinking.) And Nelles hadn’t actually been the only one to care for Justin. Sometime between 1:45 a.m. and 3:30 a.m., Phyllis Trayner, the team leader, had relieved Nelles while she took a break. Nelles returned to Justin’s bedside no later than 3:30 a.m., and the infant’s critical symptoms started at around 3:45 a.m. While the wards that made up the cardiac unit — wards 4A and 4B — each had their own nursing teams, nurses routinely covered one another’s shifts and helped out across the unit while they were on duty.
Still, the defense had a high bar to clear. First, Nelles would face a preliminary inquiry, a legal proceeding to determine whether there was enough evidence to go to trial. At this stage, prosecutors didn’t need to prove Nelles was guilty, only that there was enough evidence to advance the case.
The preliminary inquiry got underway in January 1982, in what Linda Fuerst, then a law student working with the defense team, described as a stuffy courtroom with high ceilings and wood paneling. Every morning, Nelles and her legal team would elbow through a media scrum, Fuerst told me. Rows of spectators, reporters and police officers filled the room. Nelles’s friends organized a calendar to ensure that at least one of them attended the inquiry with her every day. As nurses, doctors, cops and experts submitted to the lawyers’ questioning, Nelles remained stoic.
“Her ability to keep her chin up and to be optimistic about what the outcome would be in the face of huge press scrutiny, the weight of the law being brought down upon her — it was just incredible,” Fuerst told me.
The prosecution’s investigation didn’t stop after Nelles was arrested. They eventually examined records going back to the summer of 1980, nine months before Justin Cook had died, and found a number of “carbon copy” deaths that were also potential digoxin poisonings. One of these infants was Stephanie Lombardo, who had died just before Christmas at 10 days old, six days after heart surgery. SickKids, a teaching hospital, routinely autopsied patients who died, but Stephanie’s parents had declined. In February 1982, they agreed to have her body exhumed. At least eight more babies were exhumed as part of the investigation.
Tests on Stephanie Lombardo’s tissues showed extremely high digoxin levels — but she’d never been prescribed the drug. More important for Nelles’s defense, the nurse had been on vacation for days before the girl’s death. The prosecution’s case was unravelling.
Throughout the proceedings, Nelles’s lawyers had advanced a theory that it was actually Phyllis Trayner who was responsible for the alleged murders. Trayner had been on duty for all of the most suspicious deaths. In September, a three-week respite in suspicious deaths coincided with her honeymoon. And Trayner had been helping Nelles care for Allana Miller and Justin Cook before they died. Trayner denied killing the babies.
Lead prosecutor Robert McGee later testified that throughout the preliminary inquiry he continued to think that Nelles was the culprit in Justin Cook’s death, as well as the most likely killer of Kevin Pacsai and Allana Miller. And he believed he could get a conviction in Justin’s case.
On May 21, 1982, Nelles was in the courtroom with her parents and brother. She wore a high-necked blouse with puff sleeves and a skirt with flowers along the bottom. During the 10-week proceeding, she had watched 122 witnesses testify. Judge David Vanek was ready to decide whether she would go free or go to trial.
He found that a toxic dose of digoxin, administered deliberately, had indeed killed the four babies she was charged with murdering, and that the evidence pointed to a single killer. But it wasn’t Nelles. No direct evidence implicated her, he found. “All of her actions are perfectly consistent with the due and proper performance of her regular duties as a registered nurse,” he wrote. Moreover, the evidence of the so-called carbon copy deaths “appears to point in an entirely different direction.”
Nelles left the courtroom surrounded by her lawyers, as people on all sides shouted questions. “It was like pandemonium hit when he delivered his judgment, and we were swarmed,” Sandler told me.
“I knew my innocence. My friends and family knew my innocence and now I just know the rest of the world knows my innocence,” Nelles told a reporter soon after.
Nelles, her family and her defense team went out for a celebratory dinner. Her father didn’t look well though. “You could see just the toll that it had taken on the family,” Sandler told me. A few months later, Nelles’s father died of a heart attack.
By early 1983, the police had exhausted their leads, and the investigation was over, according to Death Shift: The Digoxin Murders at “Sick Kids,” a 1984 book written by the late CBC reporter Ted Bissland. No direct evidence of Trayner’s involvement ever surfaced, and she was never charged. Much of the public felt that bedrock institutions — SickKids, the police, prosecutors, the media — had failed Canadians.
Kevin Cox, then a 30-something reporter at Toronto’s Globe and Mail, hungry for information about the biggest story of the day, gave his number out to the parents and told them to call him anytime. “People did call me — late at night, 10, 11 o’clock and in tears, totally upset with the system that wasn’t giving them any answers. ‘Was it a cover up? Was there a murderer still loose in that hospital?’” Cox told me. “We had no answers.”
The public wanted accountability. The Ontario government responded by launching a commission to investigate how the babies died and the circumstances surrounding the charges against Nelles. At its head was Justice Samuel Grange, a respected 64-year-old judge with thick white streaks in his hair and a sprinkle of liver spots across his forehead.
Grange allowed cameras to film the proceedings, and front pages and nightly newscasts across Canada were filled with scenes from the courtroom. Bissland filed a story almost every day — and when he didn’t, people would call to ask why. In April 1984, roughly two years after she had cleared her name, Nelles was back in a courtroom, taking the stand for the first time. A week into her testimony, a Toronto cable company started broadcasting each day’s session live, making Canadian television history. Toronto residents were glued to their TVs. “I hate to say it,” Cox said, “but it was like a daytime soap opera.”
One of those Toronto residents consuming the coverage? My mom. For months, the Globe arrived on my parents’ downtown doorstep, filled with stories of infants with critical heart issues, killed in the very place meant to sustain their lives, their apparent murders unsolved. When my mom, then 30, gave birth to me that May, the courtroom proceedings were far from her mind. In her hospital room, all was bliss. Her first child was a healthy baby girl.
But as I lay in the nursery, something caught a nurse’s eye: My lips had a blue tinge. The familiar lub-dub-lub-dub of my heartbeat was swallowed by a murmur, the last beat a vibrating trill. Something was wrong, a cardiologist told my parents. They needed to send me to SickKids.
Suddenly, all those babies dying turned from newsprint to real life. “We were part of that story,” my mom recalled thinking.
I was taken by ambulance to SickKids, where my parents met me. My mom remembers seeing a sign pointing the way to ward 4A. “My God, this is the floor where those babies died,” she thought. “Is something going to happen to her?”
My room had four cribs and a big window overlooking the street. Had the babies died in this room? In that bed? My mom yearned to ask the nurses, but she worried that saying the wrong thing might disrupt my care. “I thought if I had answers to those [questions], I could protect my baby better,” she told me recently. “But I was totally too afraid to ask because, what if it was in the same room?”
Compared with the other babies on the ward, I looked healthy — pudgy and pink, though I got dusky when I cried — but the staff was watching me for heart failure. To ease the work that my heart had to do, they planned to put me on digoxin. To my mom, the word meant poison.
She didn’t exactly think a murderer was stalking the wards — no suspicious deaths had occurred in the three years since Justin Cook had died. The more my mom interacted with the nurses, and the more she observed their kindness and skill, the harder it was for her to believe that Nelles, someone just like these nurses, had killed her patients. But the deaths had shaken my mom’s faith, underscoring the uncertainty, the potential for deadly mistakes, the fine line between safety and collapse.
I underwent different tests, and the diagnosis was confirmed: tricuspid atresia, ventricular septal defect, atrial septal defect — a clunk of syllables like ice cubes clattering in a glass. An accident of cardiac muscle and its absence had given me a rare kind of heart: one valve skinned over, one chamber never formed, two holes where there should be none. In other words, my tricuspid valve was closed up. One of the pumping chambers, or ventricles, wasn’t developed. And there were two holes, one between the upper two chambers and another between the lower two chambers. Altogether, it meant that my heart wasn’t able to get enough oxygen to my body.
I would have two stopgap operations — one at 9 months and another at 2 years old — to keep me going until I grew big enough for the final procedure, an open-heart surgery known as a Fontan. The surgical technique had been in use for a little more than a decade at that point, and my pediatric cardiologist had no data to show my parents about long-term survival rates. It didn’t matter. It was the only option the doctors had.
By the time the Grange hearings concluded in September 1984, they had generated 44,000 pages of transcripts — so much material that the judge hired two people to manage the evidence on a computer. “There is a great future for the computer” in these kinds of proceedings, Grange wrote in his 300-page report, completed that December.
Ultimately, Grange examined 36 deaths that took place on the SickKids cardiac wards between June 30, 1980 and March 22, 1981, including 33 babies and infants, one child and two teenagers. He concluded that eight — including Janice Estrella, Kevin Pacsai, Allana Miller, Justin Cook, Jordan Hines and Stephanie Lombardo — were murdered with a deliberate overdose of digoxin. Fifteen had died in circumstances where a digoxin overdose was suspected or highly suspected, six died of natural causes, and seven weren’t possible to categorize, he found.
Grange ruled out the possibility that medication errors — digoxin accidentally administered to the wrong patients, in the wrong doses or in place of another drug — were to blame. While SickKids staff did make rare mistakes with drugs, Grange found it highly unlikely that they could have occurred on this scale, with one nursing team on a single unit, most in the middle of the night, and all serious enough to kill. “The theory of multiple, repeated, concentrated, fatal error[s] must be rejected as untenable,” he wrote. He also absolved the doctors, since no evidence existed that any of them had consistently been on shift for the suspicious deaths.
And he found that Nelles hadn’t committed the murders. If anything, additional evidence that came out during the hearings suggested that others had opportunities to kill the babies, he found. “In a perfect world, she would not have been arrested, charged or prosecuted,” he wrote.
Grange seemed to think that the culpability lay with another nurse. And he believed that the government order setting up the commission entitled him to identify anyone who played a role in the babies’ deaths. But the nurses and their professional organization, whose lawyers described the commission as a “witch hunt,” contested the idea that Grange could name a culprit, both during the hearing and all the way up to the eventual Ontario Supreme Court ruling. Siding with the nurses, the high court found that identifying a suspect was outside the scope of Grange’s commission: It would be tantamount to rendering a guilty verdict — outside of a trial, without giving any of the accused parties an opportunity to defend themselves, impeding their civil rights.
The ruling came down partway through the commission, as Trayner was testifying, changing the rules as to what could be asked and what Grange could include in his final report. Ultimately, Grange was prevented from assigning guilt. He never named a name.
What if, however, there wasn’t a name? What if the babies hadn’t been murdered after all?
Before placing his hand on the Bible, David Seccombe touched his watch. Seccombe is dyslexic. The watch is on the left, he remembered, putting his right hand on the book in front of him. In 1983, before the Grange commission wrapped up, Seccombe testified as an expert witness about his latest findings on a digoxin-like substance in infants.
As part of his work as a medical biochemist at a Vancouver hospital, Seccombe conducted lab tests on patients. He had recently run a digoxin test on a tiny, premature infant with an irregular heart rate. The test turned out to be positive, but no record of digoxin appeared on the baby’s chart.
Seccombe waited a week and tested the baby again. By that point, any mistaken dose would have been cleared from the infant’s system. Not only was the second test positive, the digoxin levels were higher than before, he said.
“Well, this is strange,” he thought.
Seccombe sent the sample to half a dozen labs; all of them returned positive results, all at different levels, including some at levels high enough to be toxic, he told me. The labs were picking up a naturally occurring substance that wasn’t digoxin but was similar enough to fool the tests.
In further research, Seccombe also found the substance in other infants, and he determined that its effect was additive, meaning that if a baby had the substance in her system and was given the actual drug, her supposed digoxin level could increase to a level that appeared to be toxic.
“What we were saying is, ‘Hey guys, if you’re hanging your hat on digoxin numbers, you really want to be careful because in this particular age group, in these very sick babies, there is something in their circulation that’s not digoxin — but it’s very similar to it — that will give you a false positive reading,” Seccombe said, referring to his Grange testimony.
Seccombe wasn’t the only one to cast doubt on the toxicological evidence. Skepticism had surfaced early as to whether postmortem digoxin tests could be trusted.
George Cimbura, a toxicologist who worked with the prosecution, spent months developing new methods of finding the drug in tissue samples. Experts agreed that tests on samples exhumed months after death didn’t reliably measure exact digoxin levels.
Postmortem blood samples also faced questions. When a person dies, digoxin leaches out of their tissues and into bodily fluids. In one study, Cimbura analyzed blood samples from 14 children who died at SickKids while on digoxin, separate from the suspicious deaths. One child showed a reading of 169.6 ng/mL.
“We couldn’t say conclusively they were poisoned, even after all our research into it,” Cimbura told a reporter in 1982. “As a layperson I could say they were very suspicious but as a scientist, I could only say it was a strong possibility,” he added.
After Justin Cook died, SickKids started routinely testing postmortem digoxin levels. In 608 autopsies, 97 children had positive digoxin readings, including 12 who had no record of taking the drug.
Another doctor blamed the deaths on MBT (mercaptobenzothiazole), a chemical used in the rubber-making process. Gavin Hamilton, a retired radiologist based in London, Ontario, told me he felt morally compelled to expose what he considers to be the dangers of MBT. In the 1980s, some of his patients had allergic reactions — including two that were life-threatening — after routine dye injections. Hamilton had the syringes tested and learned that MBT was seeping out from the rubber seals and plunger stoppers.
Hamilton’s theory, which he set out in a 2011 book, The Nurses Are Innocent: The Digoxin Poisoning Fallacy, held that MBT contamination from syringes and other medical equipment poisoned the babies, who were already on the brink. Now 94, Hamilton told me that some batches of rubber might have higher concentrations of MBT, likening it to pockets of unmixed flour in bread dough. Still, the deaths were concentrated on a single ward of SickKids — not in other parts of the hospital or other Canadian hospitals — and they stopped abruptly after Susan Nelles was arrested.
To Grange, the fact that digoxin was present in babies who hadn’t been prescribed the medication — in their tissues and, in Justin Cook’s case, in blood samples taken before death — was enough to show that they’d received an unauthorized dose. And even if a digoxin-like substance had elevated the readings, it couldn’t account for the sky-high results, he found.
“I may eventually be proven wrong because the toxicologic evidence upon which I in part based my conclusion may be proved wrong or inadequate,” Grange wrote. “But I shall have done the best I can with the material at hand.”
Sandler, Nelles’s lawyer, told me that it was nonsense that the SickKids readings were bogus, even if false positives were possible. “You would’ve expected if there were false digoxin readings, then babies would’ve been dying during another team’s duty. But that’s not what happened,” he said.
Throughout, the cardiologists at SickKids largely attributed the deaths to the infants’ conditions or other natural causes. “Let’s not forget these were very sick babies, and with their serious heart problems they were quite literally born to die,” Dr. Richard Rowe, the hospital’s head of cardiology until 1986, testified at the Grange commission.
Regardless, the hospital had problems, despite its world-class care, according to a 1982 government report cited in Bissland’s book. Among them: No one in the cardiac unit reported the climbing death rate to hospital administrators or patient-safety bodies. Drug-administration errors were too common for a hospital of its stature. Parents had to scrounge for information about their children. And the hospital lacked security. Anyone could wander the hallways, no visitor identification required.
Separately, there seemed to be a disconnect between the nursing staff — most, if not all, of whom were female — and the revered male doctors. Dr. Fowler, the cardiologist who had told police about Nelles’s strange look as she wrote up Justin Cook’s death report, considered himself a troubleshooter with the nurses, but he didn’t know that they worked in teams, Bissland’s book said. Dr. Robert Freedom, who succeeded Dr. Rowe as cardiology chief, told investigators that Nelles “was constantly suggesting to the residents that certain babies be put on an IV in case of emergency. In most cases the residents didn’t agree and thought it unusual,” according to Bissland’s book. The night Kevin Pacsai died, she was concerned about his condition, looked for help from the doctors on duty, and was frustrated when two of them shrugged her off. Shouldn't they have considered her opinion as a medical professional?
The children’s deaths — and, in all likelihood, the ensuing publicity — sparked changes at SickKids. The institution increased its supervision of the staff. By early 1982, the hospital had started to use unit-dose syringes, where drugs are dosed at a central pharmacy, rather than by nurses measuring from vials of medication. “Our ID cards are checked and if you are without one, you’re questioned,” a hospital administrator told a reporter in May 1982. Today, the hospital is once again known as a world leader in pediatric care. (A spokeswoman for SickKids said it had no one available to comment for this story.)
Trayner never returned to SickKids after her team was sent home following Justin Cook’s death. She spent at least the next three and a half years on paid leave from the hospital, according to Bissland. She reportedly went into occupational health sometime after, and died in 2011.
Cox, the reporter, considered Trayner another victim of the whole affair. If she did know anything more about how the babies died, he concluded she would never tell. “She was going to carry that forever,” he told me. “And if she could have done something to stop it, she was going to carry that forever.”
Nelles returned to the nursing profession, first in the kidney dialysis unit of SickKids — after some lobbying on her part — and later at a hemodialysis unit in Belleville, her hometown. In 1985, she married, and she and her husband had three children. Exonerated twice over, she became an advocate for the nursing profession, delivering talks on legal issues to colleagues in the field. In 1999 and 2015, Queen’s University and Nipissing University, respectively, awarded her honorary degrees. Nipissing noted “her courageousness in defending the profession” in a brief write-up about the accolade. (Nelles couldn’t be reached for comment, and her husband didn’t respond to requests for comment.)
“I truly, honestly don’t feel special,” Nelles said in a video related to the Nipissing degree. “My life presented me with circumstances that I wouldn’t wish on anyone, and I hope that no one ever — a nurse [in] particular, never has to go through that kind of thing again.”
“We’re recording now. I want to start at my birth,” I tell my mom.
We both laugh. We’re sitting in her sunny home office in Cambridge, Ontario, in May 2024. Both of us are nervous. We’ve never talked in depth about my heart issues, but over the past couple of years, I’ve been diving into my own cardiac history, and she agreed to an interview.
Five minutes into our conversation, she stops me. She wants me to know some context — how there was an inquiry in the months before my birth, how SickKids patients with heart defects were dying. Seemingly murdered, she says.
I had never heard this before, in all the time I’d spent at the hospital. Before I was 4 years old, I had the two stopgap surgeries and the Fontan operation at SickKids, and every year I went back for a day of tests and an appointment with my pediatric cardiologist. At 18, I graduated to the adult cardiac clinic across the street. It’s unlikely that any of the nurses who treated me overlapped with Nelles, though at least three of the same cardiologists were still there when I was first admitted. My own pediatric cardiologist rejoined the staff at SickKids in 1983, after training there in the late 1970s, but he told me he didn’t remember anything about the so-called digoxin deaths. He was an unflappable doctor, and his care was steady and kind. Though my heart’s anomalies were rarely far from my mind, I flourished.
Last year, I got a digital cache of over 1,000 pages of my SickKids medical records — every dose of medication, every test, every doctor’s observation for 18 years. Poring over it, I could almost hear the click-click of keys turning, locks opening, my story being revealed.
In the nurses’ notes, with their faded loops of cursive and hieroglyphic abbreviations, I could see their care for me. They documented my likes (Cheerios) and dislikes (porridge). One nurse advised against putting an IV in my right hand because I sucked my right thumb; another noted that I’d be accompanied by a blue Care Bear called Bubba.
Later, I read Grange’s final report, and it was easy to see how those babies were like me. Babies whose color turned dusky when they cried, who needed medicine to fortify their hearts, whose parents stayed watchful by their bedsides. Justin Cook’s tricuspid valve was closed up, just like mine. He died hours before he could get his surgery, the same type of procedure I would receive in the same hospital a few years later.
In the end, the public never got the accountability it wanted following the deaths. The parents have had to find their own ways of resolving the question of whether their babies died of natural causes or at the hands of a killer who was never caught. The case remains unsolved, either one of Toronto’s deadliest mass murders or a tragic coincidence.
For most observers, the main questions involving the SickKids deaths are, how did the infants die — and who, if anyone, was responsible? I’m also preoccupied with another question: Why did I live? It seems that I was the beneficiary of those losses, a baby with an abnormal heart who thrived in part because of the measures put in place after the children’s deaths. But in the end, I don’t have an answer to either question. My only response is to sign the closing papers on an apartment with honeyed hardwood floors, to book a plane ticket to visit a friend in Chicago, to cheer on my nephew as he reads me a chapter book for the first time, to board the amusement park ride that spins me upside down, to feel my muscles grind as I curl a dumbbell, to relish the wrinkles that crinkle my eyes, to say yes as much as I humanly can. Yes, let’s do this. Yes, let’s go there. My only response is to live a full life.
To report this piece, I contacted more than three dozen people, companies and organizations. Some didn’t respond; others declined to comment or said they weren’t able to comment on events from so long ago. I attempted to reach everyone mentioned above in multiple ways, as well as others who were involved; some couldn’t be reached, and many have died. This account is based on interviews, books, Samuel Grange’s final report, court documents, newspaper archives, scientific studies and other materials.
This piece is the fourth in our series, Heart of the Matter, a special collaboration from Narratively and Creative Nonfiction that explores love and matters of the heart. You can learn more about this special series and check out the rest of the stories we’ve published here.
Leigh Kamping-Carder is the head of newsletters at The Wall Street Journal. She’s also the writer of The Heart Dialogues, a Substack for people born with heart conditions (and the people who care about them). Her writing has appeared in the WSJ, The New York Observer, The Walrus, Guernica, The Brooklyn Rail and elsewhere. She lives in New York.
Jesse Sposato is Narratively’s executive editor. She also writes about social issues, feminism, health, friendship and culture for a variety of outlets. She is currently working on a collection of essays about coming of age in the suburbs.
Julie Benbassat is an award-winning illustrator, painter and animator.
A riveting piece! Congratulations, Leigh, for a job well done!