The question reflected Kristrinah Ayala’s increasing desperation, and her attempt to make sense of her 18-month ordeal. “I’m a therapist,” she told the Washington rheumatologist who began treating her in the fall of 2011. “Is it possible that this problem is all in my head?”

Ayala’s anguish and the nature of her question took Victoria Shanmugam, an assistant professor at Georgetown University Medical Center, aback.

Shanmugam says she quickly reassured Ayala that the painful, five-inch-long open wound at the back of her knee was not a product of her psyche but the manifestation of a serious physical problem that a team of specialists at Georgetown were struggling to diagnose.

After 10 failed surgeries, no one knew whether Ayala’s wound, which had refused to heal for more than a year, was caused by her lupus — an underlying autoimmune disorder that can impede healing — or an infection. And if it was an infection, why had multiple tests failed to identify a bacterial culprit?

“It felt like a nightmare that I couldn’t wake up from,” Ayala, 57, recalled. “That wound was just sapping my existence.”

It took 10 surgeries and nearly two years of misery to figure out why a woman’s infected knee would not heal. (Owen Freeman/For The Washington Post)

A resident of College Park, Ayala led an active life before her knee began giving her trouble. In addition to practicing psychotherapy, she was an amateur painter, raised cockateils, kept a flock of chickens, walked four miles a day for fitness and on weekends clocked 30-mile bicycle rides with her husband. Her diagnosis of lupus in July 2009 did little to change that. The disease, which causes inflammation and occurs when the body attacks its own tissue, seemed well controlled on medication, including corticosteroids, although her joints sometimes hurt. That was not surprising: In 2007, doctors had told her she had osteoarthritis, which also causes joint pain.

In June 2010, she developed sharp pains — sometimes so strong she was unable to stand — in the front of her left knee. Ayala consulted several doctors, who seemed unable to decide whether the pain was caused by her arthritis or lupus. An orthopedist gave her a series of cortisone injections, which eased the knee pain temporarily. An MRI scan performed in early 2011 revealed a possible cause: a torn meniscus, a common injury affecting the disk that cushions the knee.

Knee surgery in April revealed something different: Her meniscus did not require repair, but the orthopedist who performed the operation diagnosed septic arthritis, essentially an inflamed joint due to infection. Ayala took antibiotics, but the pain shifted, spreading to the back of her knee and down her calf.

A few weeks later, the orthopedic surgeon decided that the pain might be caused by a Baker’s cyst, a benign, fluid-filled growth. Such cysts, which sometimes disappear without treatment, are common in patients with arthritis or autoimmune disorders.

A month later, the pain still intense, Ayala was wheeled back into a suburban Maryland operating room for her second knee operation in less than two months. The surgeon removed the Baker’s cyst but told her he didn’t think it was the reason for her pain, which he suspected was somehow related to lupus.

That surgery, in June 2011, inaugurated a pattern: An operation to clean out the wound or place a skin graft to spur healing was followed, often within days, by the opening of the incision.

Doctors increased her lupus medication and tried to determine if an infection was the reason the surgical site had not healed. One culture did unearth a potential clue — an atypical form of bacteria often found in tap water; doctors decided it was a contaminant, not the culprit, and didn’t pursue it.

A month later, Ayala sought treatment in a Maryland emergency room for a high fever; doctors determined she had methicillin-resistant Staphylococcus aureus, or MRSA, a dangerous form of staph that is often acquired in hospitals. “Don’t come back here,” she said a doctor advised her, suggesting that she seek treatment at a hospital with greater expertise. “We don’t know what to do for you.”

Feeling powerless

In August 2011, Ayala became a patient at MedStar Georgetown University Hospital’s Center for Wound Healing. At that point, she said, she was wrung out from dealing with an in­trac­table leg wound, two parents with advanced dementia and several doctors whom she characterized as having “the sensitivity of a rock.”

Exhausted and in pain, she said, she tearfully confided to her husband, “ ‘I can’t take it anymore. Why don’t they just cut my leg off?’ I couldn’t see clients, I couldn’t paint, I couldn’t go to Minnesota to see my granddaughter who’d just been born.”

At Georgetown, her doctors, who included a plastic surgeon, an orthopedic surgeon and an infectious-disease specialist, were baffled by her case.

The team called in Shanmugam, a principal investigator for the federally funded research study called WE-HEAL, which involves patients with autoimmune disorders and chronic wounds that won’t heal.

Shanmugam said she was struck by the paradoxical nature of Ayala’s case. Often, she said, aggressive treatment of an underlying autoimmune disorder speeds healing. “But her case was the opposite,” Shanmugam noted. “The more aggressive people got with her lupus, the worse she got.”

In reviewing Ayala’s records, the rheumatologist noted the earlier test that was regarded as a lab error, possibly caused by tap water. It piqued Shanmugam’s interest.

Several years earlier, she had seen a patient whose persistent hand infection was caused by an unusual organism called Mycobacterium kansasii, a cousin of tuberculosis, which is found in soil and water.

She wondered whether the same organism had somehow burrowed into the area surrounding Ayala’s knee, where it festered. “I’m sure that case made me more aware of it,” she said. “There really weren’t a lot of other things it could be.”

Ayala said that when Shanmugam told her that an uncommon organism might be to blame, her spirits lifted. “For the first time it seemed like someone was saying, ‘It’s not your fault, and we’re going to investigate to see what it could be.’ ” Shanmugam, Ayala recalled, was “like a bulldog. She just wouldn’t let go.”

But growing the mycobacterium in the lab requires exacting conditions: The culture must be done in a specific way using a special medium at just the right temperature.

After an attempt in October to culture the wound failed, Shanmugam and her colleagues decided to wait to see if the wound ultimately closed on its own, since Ayala seemed to be improving slowly.

But in February 2012, the incision reopened. Doctors decided to operate again to clean out the wound and try to retrieve bacteria for a culture.

Answer from the lab

In March 2012 a surgeon removed what Shanmugam called “cheesy goop” buried deep in the back of Ayala’s knee, which was hustled off to a lab at MedStar Washington Hospital Center for analysis.

Several weeks after this, her 10th operation, the results came back. Shanmugam was right: Ayala’s knee was infected with Mycobacterium kansasii, an opportunistic infection that occurs most often in patients with damaged immune systems. According to a 2012 article, the number of infections, which most often affect the lungs, has increased since the advent of the HIV/AIDS epidemic. But the infection also strikes those with diabetes, heart disease and others, like Ayala, on long-term steroid therapy.

Ayala said that this time she noticed a distinct difference after surgery. “It just kept getting better and better. I could bend my knee and the pain was less.” It took weeks, but this time the incision closed.

How was Ayala infected?

“Honestly, we don’t know,” says Shanmugam, who credits teamwork for the lab result that finally led to a diagnosis.

Ayala wonders whether she was exposed by the steroid shots she received in 2010 to ease her knee pain. Shanmugam is dubious and said that it is more likely that Ayala was exposed to the bacterium outdoors, not from exposure to tap water during the injections.

Her lupus clearly played a role. “If she had a fully intact immune system, she could have walled it off and gotten rid of it,” Shanmugam said.

Ayala was immediately put on a cocktail of potent drugs that should be taken for about a year to eradicate infection. But the medicines made her so sick that she had to stop after a month.

Even so, the extent of her improvement has surprised her doctors. Ayala has resumed riding her bike — something Shanmugam and her colleagues never imagined she’d be able to do — and recently flew to Minnesota for her first visit with her granddaughter, who is now 16 months old.

“I have a life again,” she said. “I feel like I’m on the road to recovery.”

Shanmugam says that while she hopes Ayala’s infection is gone for good, she is not sure. “It’s hard to know, because it can take only one or two organisms” for the infection to resurface. “We’re letting her body guide us.”

Have a medical mystery that’s been solved? E-mail