The diarrhea started the night Tony Lewis came home from the hospital after routine leg surgery. Dazed, he collapsed on the floor of his Natomas apartment and then on the front lawn as his daughter struggled to guide him to her car.

They raced back to Sacramento's Sutter General Hospital, where he was readmitted, X-rayed, tested. The diarrhea got worse. Tests pinpointed a virulent infection wracking his intestines, destroying his bowels.

"I just didn't understand," said his daughter Alison Gilbreath, 39, of Fair Oaks. "But my mother did. She said, 'What have they done to him?' "

Doctors tried to stop his descent, finally wheeling him into surgery to carve out his sickly colon, but his body was failing. Eight days later, on May 8, 2009, Anthony P. Lewis, 75, a retired British industrial chemist, was dead.

Records signed by his doctor list the underlying cause as colitis linked to Clostridium difficile, a powerful infection that most often afflicts patients in hospitals and nursing homes. Now on the rise in U.S. hospitals, it can withstand sophisticated antibiotics and live for months on patient wards.

As is often true in hospital infection cases, however, questions linger about Lewis' death. Sutter officials, for example, say knowing where he got the infection is impossible, and the Sacramento County coroner's death certificate does not even mention it.

His doctors and other staff convened after his death to review the case, concluding that the steps they took to care for him were medically sound, said Janet Frain, who oversees infection prevention at Sutter General.

"It was a very tragic case," Frain said. "It was the worst-case scenario."

Gilbreath has no doubts that her father got C. difficile at Sutter General, and that it killed him.

"He came in with a broken femur. You don't die of a broken femur," she said.

Infections in the shadows

One in 20 hospital patients get infections. In California, roughly 200,000 people get hospital infections annually, and 12,000 of them die, according to state Department of Public Health statistics. That makes such infections one of the state's leading causes of death, ahead of automobile accidents and Alzheimer's disease.

Yet these deaths have remained mostly in the shadows. They often are classified as "deaths from complications," an oblique term used in obituaries and often unquestioned by relatives and friends.

Even the best doctors can be baffled whether an infection was acquired before or after a patient was admitted, and if it was the principal cause of death or no factor at all.

Many health care providers historically have viewed hospital infections - going by obscure names or acronyms such as C.diff, CLABSI, VRE and the more familiar MRSA - as a sometimes inevitable consequence of being hospitalized.

That has begun to change. Research has demonstrated that many hospital infections are preventable, and most hospitals, including Sutter, have launched campaigns to rid their facilities of infections.

The change has not come nearly fast enough for Lewis' widow. Leila Lewis, 74, of Fair Oaks, is a small woman with a passion in her voice that belies her soft British accent.

"Hospitals are in denial," Lewis said. "Because the public does not have knowledge and understanding of C. difficile, it is easy for the hospitals to sweep this serious situation under the carpet, and the relatives of the deceased are none the wiser."

She and her daughter provided a reporter with copies of Tony Lewis' medical records. They signed a waiver of patient privacy rules to allow Sutter representatives to speak to a reporter about his medical history. At Sutter, Frain agreed to discuss the case in three interviews.

What emerged is a story of how a common operation led to infection and death.

The fall into illness

The spring of 2009 was flush with promise for Tony and Leila Lewis, as they traveled to California from their home in St. Annes on northern England's Fylde Coast.

The couple were considering a move to the Sacramento area to be close to their only daughter, Gilbreath, and her family - husband James and 2-year-old Alexander, their first grandson.

Lewis was in good health, his family said, although his longtime myositis, a condition caused by inflamed muscles, meant that he fell easily and supported himself with a walker.

Six weeks into the visit, Lewis was walking across a kitchen floor that his wife had just mopped. He slipped and fell.

Gilbreath and her husband drove him to Sutter General, the first hospital they spotted. He was diagnosed with a broken left femur, or thighbone, and admitted on April 19, 2009, medical records show.

"This is a 75-year-old male who is pleasant, cooperative, alert and attentive, fairly groomed and nourished," the attending doctor wrote. When Lewis was discharged four days after leg surgery, he was back on his feet.

Then the diarrhea hit. Back at Sutter, his white blood cell count had gone up, which can signal infection, so he was started on antibiotics and admitted. His belly puffed up like a balloon, said his widow.

When he tested positive for C. difficile, a consulting doctor described him as "fatigued, very pale-appearing," but still alert. His abdomen was painful. His white cell count climbed. His blood pressure dropped.

"It appears that he has a case of severe Clostridium difficile, colitis with possible impending toxic megacolon despite medical therapy," the doctor wrote, noting that he had suggested surgery, but that Lewis was opposed.

Lewis was moved to intensive care.

Attacking spores

Although C. difficile is well known in England, it has only recently started attracting public attention in the United States. Cases here have tripled in the past 10 years, scientists say, and those cases are growing more serious.

In some U.S. hospitals, it is outstripping the better-known MRSA, or methicillin-resistant Staphylococcus aureus.

More than 300,000 U.S. cases occur annually, leading to 15,000 to 30,000 deaths and adding billions of dollars a year to health care costs, according to the federal Centers for Disease Control and Prevention. Most susceptible are patients recently given antibiotics, as Lewis was before his leg surgery.

C. difficile is an insidious, Jekyll-and-Hyde pathogen. Its microscopic rod-shaped bacteria, resembling clusters of long-grain rice, can live unnoticed in the intestines of many children and a tiny fraction of adults, held in check by healthy or "good" bacteria.

But it produces spores so resilient that they can live outside the human body for months, on hospital beds, tables, bedpans.

Gilbreath said she learned all this by happenstance. When she first learned her father had C. difficile, she held his hand and touched his hair, only hearing later that she should have been wearing gloves.

Although the hospital provided plastic dispensers of an alcohol-based cleanser on the hallways of the ICU unit, a nurse cautioned her that the sanitizer wouldn't fight off C. difficile.

That's because the spores can resist alcohol-based disinfectants, and experts recommend hand washing with soap and water instead. Patient rooms containing spores should be decontaminated with special cleansers such as bleach, they say.

Sutter representatives said that they follow national cleaning standards and use bleach-based products to clean the rooms of C. difficile patients.

Staff members are taught to educate families to protect themselves with gloves and other safeguards, said Frain, administrative director for integrated quality services at Sutter Medical Center, Sacramento, which includes Sutter General and Sutter Memorial.

The two hospitals' combined C. difficile infection incidence, 82 cases over 15 months starting in 2009, is in the average range for Sacramento hospitals, according to state records. That amounts to 5.6 cases per 10,000 patient days. The records do not show how many infected patients died.

Final hours

As Anthony Lewis grew weaker, doctors asked again if he would consider a total colectomy to remove his diseased colon before it ruptured and threatened his life.

The surgery would mean a retooling of his intestines called an ileostomy, a hole in the groin area to allow his feces to pass into a colostomy bag.

Lewis was horrified, his daughter said. He didn't want to live with a bag. After long talks with his family and doctors, though, Lewis relented. Surgeons removed his colon.

He never regained consciousness. Soon, only a ventilator kept him alive.

His family met with Sutter representatives to ask why he was so sick and to discuss his future. Gilbreath remembers that someone suggested her father may have had C. difficile in his system before he came to Sutter, and her temper flared. "I said that was conjecture," she said.

Lewis earlier told his doctors that he wanted "full code," the term for being resuscitated if his heart or lungs failed. His family decided on May 7 to change that to DNR, or "do not resuscitate," but to keep him on the ventilator.

The next afternoon, Lewis' family told his physician, Dr. Mimi Reiss, they had decided to withdraw life support. Reiss declared him dead at 5:10 p.m.

His family rushed to find a Sacramento burial plot and the money to pay for it.

The death was so unexpected that Lewis' friends and family in England could not make it in time. Only four guests attended - his widow, daughter, son-in-law and his Texas-based son.

They stood at the graveside surrounded by rows of empty chairs.

Reporting complications

The official accounting of Tony Lewis' death suggests one reason why hospital infections have been so invisible to the public. Although C. difficile was almost certainly a contributing factor to Lewis' death, government records don't show that.

Reiss signed a draft death certificate stating that the immediate cause of death was respiratory failure. The document identifies the underlying cause: Clostridium difficile colitis. The draft was faxed to the Sacramento County coroner's office, according to Lewis' records.

The official certificate issued by the coroner, however, lists only one cause of death: "Complications of left femur fracture due to ground level mechanical fall."

Assistant coroner Edward E. Smith said that the term "complications" was probably intended to encompass Lewis' C. difficile. His office's pathologists often use terms more generic than hospital physicians, he said.

"Because it does not match what the hospital doctor put on (the) worksheet is not significant to us," Smith said. "If our doctors don't want to put it there, they don't put it on there."

That practice may help mask the number of deaths in California caused by hospital-acquired infections, he acknowledged. "To get that changed would probably be an educational process," he said.

The state Department of Public Health recently began requiring hospitals to report the numbers and types of infections among their patients, a step mandated by a 2008 state law. The numbers of related deaths are not disclosed.

"There's no requirement to report that," said epidemiologist Dr. Jon Rosenberg, chief of the state's health care acquired infection program.

"Just think about the logistics of that. What if (patients) have recurrent Clostridium difficile infections, and they die at home? You can't say the hospital should go back three months later and report it."

But Lewis died in a hospital. Should the state have known about his death?

"That's not part of our mandate," Rosenberg said.

© Copyright The Sacramento Bee. All rights reserved.

Deborah Schoch is a senior writer at the California HealthCare Foundation Center for Health Reporting. The center is a journalistically independent organization devoted to reporting on health care issues in California. It is based at the USC Annenberg School for Communication & Journalism and funded by the nonpartisan California HealthCare Foundation.

• Read more articles by Deborah Schoch