Flashback: Tainted injections turn A2 into a trauma ward
This story originally appeared in The Ann in January 2013.
Over a few weeks in 2012, U.S. doctors unwittingly injected 14,000 people with toxic fungus. An Ann Arbor hospital became treatment central for patients who came down with rare, deadly fungal meningitis.
Story by Lynn Monson
Photos by Benjamin Weatherston
This is a maddening story, a sad story, a story that never should have happened. It involves people getting sick and dying when they shouldn’t have. Right here in America, in Michigan, in Washtenaw County, in 2012, as one of the richest countries in the world continues to debate how to provide health care for its citizens. It’s the story of how a rare, life-threatening illness spread through 19 states and how an Ann Arbor hospital became the epicenter in its treatment.
• • •
Dr. Lakshmi Halasyamani, chief medical officer at St. Joseph Mercy Hospital in Ann Arbor, remembers listening to a National Public Radio story while she was driving home from work one evening in the first week of October. Federal health officials were discussing a puzzling outbreak of fungal meningitis first identified by doctors in Tennessee. Investigators had linked it to 35 illnesses and five deaths in six states.
“Wow, I’m so glad that doesn’t affect us,” she remembers thinking. “I’m so glad I don’t live in Tennessee.”
Within 48 hours, St. Joe’s would become a major player in what grew into a national story. Around the country, hundreds would be tested, treated and hospitalized over the next several weeks because a Massachusetts company distributed a spinal steroid contaminated by a fungus.
Michigan would have the most cases of fungal meningitis and related spinal infections of any state, almost double the next highest caseload, in Tennessee. By mid-December, St. Joe’s had treated 75 percent of the cases in Michigan and more than 25 percent of the 590 people on the national list. Two Washtenaw County residents – a 61-year-old man and a 78-year-old woman – were among 10 people in Michigan and 37 nationwide who died as of mid-December.
Today, three months after the outbreak first hit, the pace has slowed but doctors and staff at St. Joe’s are still treating patients, with no end in sight. It’s not an exaggeration to say that they are writing the book on best practices for a fungal outbreak; at the same time they are writing one of the most important and interesting chapters in the history of the hospital.
It was late September when doctors at St. Joe’s, like the doctors in Tennessee, encountered several patients with what seemed like bacterial meningitis, a highly contagious and potentially fatal disease. The symptoms and test results, however, didn’t quite match the usual bacterial meningitis profile. The strange illness became a topic of conversation at staff meetings and in the hallways as doctors conferred and researched the possibilities. Halasyamani was given updates as Drs. Varsha Moudgal and Anurag Malani worked with pathologists Paul Valenstein and Samuel Hirsch, among others, to solve the medical puzzle.
Then, suddenly, the pieces began fitting together.
It started with a message from the federal Centers for Disease Control and Prevention that was left after hours on an answering machine at a pain clinic in Brighton on Oct. 2. When staff from the Michigan Pain Specialists arrived for work the next morning, they retrieved a message that was part of a life-and-death nightmare that would envelop thousands of Michigan people – patients, families, doctors, medical staffs and state health officials.
The CDC informed Michigan Pain Specialists that it was one of four medical facilities in Michigan (and dozens nationally) that had received tainted vials of methylprednisolone acetate, or MPA, a commonly used injectable steroid that eases back and joint pain. A fungus had been found in MPA manufactured by the New England Compounding Center. The fear was that the tainted injections could cause fungal meningitis. The clinic checked its records and found that 629 of its patients had received the bad meds at its Brighton location from Aug. 9 through Oct. 2, the day the CDC called.
MPS began contacting patients and advising that they be checked for fungal meningitis, which requires a lumbar puncture, also known as a spinal tap. The disease is not contagious like its more feared relative, bacterial meningitis, but both types cause swelling of the brain and spine and can be fatal. The fungal version has the same symptoms as bacterial – headache, fever, nausea and neck stiffness – but they are milder at first and take longer to show up, sometimes a month or more. That meant some recipients of the bad steroid shots who were still feeling fine might not understand the importance of being tested.
Michigan Pain Specialists is not affiliated with St. Joe’s or its parent Trinity Health, but four of the pain clinic’s five doctors are on the St. Joe’s staff: Edward Washabaugh, Lou Bojrab, John Chatas and Alex Shalhoub. That familiarity with the hospital and its staff, along with the need to consolidate and focus expertise on this unusual and developing outbreak, led MPS to steer its 629 patients to St. Joe’s, though some chose other options.
Only the Brighton office of MPS used the tainted steroids; they weren’t used at its Ypsilanti and Adrian offices. The other three Michigan facilities that received the tainted steroid shots were the Michigan Neurosurgical Institute in Grand Blanc, Neuromuscular and Rehabilitation in Traverse City and the Southeast Michigan Surgical Hospital in Warren. Although the Michigan Department of Community Health has not released a breakdown of the number of illnesses attributed to each location, the Brighton MPS clinic apparently had by far the most patients who were affected.
A few national news stories appeared on Oct. 3, focusing on the supposedly small regional outbreak discovered by doctors at Vanderbilt University in Nashville. The next day the CDC and the Food and Drug Administration went public, declaring that it was a much more widespread public health emergency that included clinics in Michigan and many other states. Between the MPS alert and the CDC announcement, doctors at St. Joe’s confirmed that their earlier patients with the strange meningitis symptoms were part of the fungal outbreak. Oct. 5, the hospital announced it was treating six fungal meningitis patients and the media alerted the public that Michigan had joined the national outbreak.
MPS continued to call its patients and the hospital staff scrambled to learn as much as it could about the outbreak and how to treat fungal meningitis.
If doctors usually open the treatment literature and turn to time-tested recommendations for an unfamiliar illness they’ve encountered, this time they found little advice. “The fungi found in both patients and in recalled vials are common in the environment but were not recognized as a cause of meningitis before this outbreak,” the CDC says on its website. People occasionally contract fungal infections – think athlete’s foot or yeast infections – but usually the fungus has arrived on the skin or by spores entering the lungs. This was much different. As Halasyamani succinctly put it, “We don’t go around injecting pathogens into people.”
But that’s exactly what had happened: A fungus had been directly injected into the spinal area of 14,000 people around the country.
Questions began multiplying like the infections. Definitive answers were elusive. Did all the patients receive the same level of toxic fungus? Don’t know. Will analyzing spinal fluid from a lumbar puncture tell us who has fungal meningitis? Yes, but a patient with an initial clean result may develop the disease later. Which of the two main anti-fungal drugs works best? Depends. Should patients receive both? Probably. Why are some patients reporting meningitis symptoms but many are not? Don’t know. Should we start them on the drugs even if they feel fine? Probably not. How long will the patients have to take the anti-fungal drugs? Don’t know. Are the anti-fungal drugs compatible with other medications patients are already taking for other conditions? Frequently not. What consequences will that have for the patients’ health? We’ll have to wait and see.
Those were only a few of the questions hanging in the air on Oct. 5, as the first MPS patients started showing up at the St. Joe’s emergency department. A moderate stream over the weekend turned into a flood that next Monday, Oct. 8, forcing the hospital to implement its system-wide emergency “incident command structure,” which involves regular conference calls of key players across the hospital.
“I don’t think we had any idea what (the initial outbreak) was going to grow into,” Halasyamani said. “We were mostly unsure about how to treat these folks. And we began to see more and more patients presenting to the emergency department with headache and concerning symptoms. We quickly had to mobilize resources to do over 250 lumbar punctures.”
Jennifer Dunn, director of emergency services, said it was difficult in the first days to get a grip on the scope of what the hospital would be dealing with, both in terms of precise treatments and how many patients would be showing up.
“(Early on) there really weren’t clear indications who should get the (spinal) tap,” Dunn said. “(Patients were told to) come if you’re symptomatic … if you have a headache or back or neck pain. Well, they all do, they always do … they have pain all the time, they’re not sure if it’s different.”
In a few days that confusion was cleared up by a decision: Give every fungal patient a spinal tap. More than 370 were given in the first month, including 66 on a single day.
St. Joe’s ED has lots of rooms – 64 – but it wasn’t enough in the first few days. “When you have 110 (or) 120 patients, the rest are in the waiting room and in the halls,” Dunn said. “All you can think of is, oh my gosh, they are so afraid, they’re upset, they’re fearful. They want an answer and (we’re) struggling because this is all new to us. And it was changing almost by the hour.”
For example, a typical meningitis test requires that about 9 cubic centimeters of fluid be withdrawn during the lumbar puncture, Dunn said. So that’s the amount the staff collected at first; the hospital lab tested the fluid while the patients waited for the results. But then the CDC said it also wanted a sample from each patient, so Dunn had to alert the staff to withdraw 30cc, which takes longer and extends the waiting time in the ED.
Oct. 8 had been “chaos in a not-so-organized way,” Dunn said, but she and her staff figured out how to streamline the process over the next few days. A key part was assigning teams that would focus exclusively on the various tasks required for the fungal patients. Depending on the day and time, normal emergency room staffing is between 50 and 70 doctors, nurses and technicians per shift; during the fungal crunch, it was increased to 90. Other hospital departments offered help as it was needed, Dunn said.
If the lab analysis of a spinal tap indicated meningitis, the patient was hospitalized. If the spinal tap results were negative, the patient was sent home and told to monitor their health and report back if headaches or the other symptoms appeared.
Through mid-December, 53 patients were hospitalized with spinal meningitis. The degrees of their illness varied because every person reacts differently to a dose of fungus. Likewise, each patient varies in their reaction to the two main anti-fungal drugs, voriconazole and amphotericin B, which are given by intravenous solution. Depending on which combination of the highly toxic drugs the patients take, they can develop liver and kidney problems, electrolyte imbalances, irregular heartbeats, flank pain, stroke, hallucinations, sensitivity to light, an extreme shaking condition called rigor and other allergic reactions.
Some patients can handle the full dosage of the drugs; some must have lower dosages, which slows their recovery; some can tolerate only one. Some of the meningitis patients could leave the hospital after a week or maybe two, usually continuing to take voriconazole in capsule form at home. In one of the more extreme cases, a meningitis patient was still hospitalized in December after 65 days.
Determining whether a fungal meningitis patient is recovering or ready to go home is difficult, according to Dr. David Vandenberg, director of the Fungal Outbreak Clinic that St. Joe’s established to track the patients.
“Is the patient subjectively feeling better? Is their fever gone? Are their vital signs OK? Are their presenting complaints – headache, stiff neck, light sensitivity – better? A lot of that is subjective on the patient’s part,” Vandenberg said. “The other technical thing we can do is repeat their lumbar puncture, their spinal tap. The problem is we don’t know how quickly that is going to clear, so we’re actually not sure how to interpret the results. If a week into therapy or (in) two weeks we repeat the spinal tap, and they originally had 2,000 white blood cells in their first sample and now it’s 1,000, that’s probably better. But it’s not clear. If they had 2,000 and now it’s zero, does that mean they’re cured? Almost certainly it doesn’t.”
Early on, the hospital started a daily in-house conference call that sometimes included as many as 50 doctors, staffers and administrators. They shared information on new admissions, treatment possibilities, health alerts and so on. Halasyamani said the calls were crucial because of the constant developments, the lack of previous case histories, the sheer scope of what they had to deal with.
Because St. Joe’s was treating more fungal patients than any hospital in the country, it was gaining new treatment knowledge and becoming a resource for others in the state and nationally.
While St. Joe’s was handling the majority of Michigan patients treated by mid-December (159 out of 214), the other 55 were being cared for at seven other hospitals around the state, including the University of Michigan Medical Center, which treated 15. Carol Kauffman, a U-M doctor and medical school professor who is also on the staff of the Ann Arbor VA hospital, shared her expertise in anti-fungal therapy with St. Joe’s. She also was the lead author of an article about the outbreak published on the website of The New England Journal of Medicine on Oct. 19. It detailed the outbreak origins and discussed symptoms, analysis of spinal fluid, treatment options and anti-fungal drugs.
By early November, meningitis seemed to be under control, but a new problem emerged. MSP patients, many of whom were at home with no other symptoms, began complaining of pain at the site of the steroid injection they had received many weeks or months previously. Some were or had been meningitis patients; some were not.
Using MRI’s, doctors discovered that the body was forming abscesses to wall off the fungus that had been injected. The abscesses usually have a liquid center, like a pimple, and the infection begins eating away at surrounding tissue. Antibiotics won’t work so patients were admitted, started on the anti-fungal drugs and scheduled for surgery. Surgeons drain the abscess and remove the damaged tissue. In some cases, the fungus attaches itself to a bone in the spine or lower area, and must be scraped away.
Since most of the steroid shots were given for back pain near the spine, abscess patients ended up undergoing the very back surgery they were trying to avoid by taking the steroid shots, Vandenberg said.
Abscess patients usually stayed in the hospital for five to seven days. By early December, St. Joe’s had given 400-some MRI’s, with about 20 percent finding abscesses.
The new batch of abscess patients admitted to St. Joe’s topped 100 by mid-December, pushing the total number of fungal-related admissions to 159 since the start of the outbreak. The high point at any one time was 79 on Nov. 16; by December the average number of fungal patients at the hospital hovered around 40 or 50. Fungal patients caused the hospital to reach and stay near its 537-bed capacity for weeks, prompting staffing and facility problems. Extra beds were opened with equipment provided by St. Joe’s sister hospitals. The state health department approved fast-track Certificates of Need for additional operating rooms and a mobile MRI to supplement the hospital’s existing units. Extra nurses came from St. Joe’s sister hospitals in Michigan and Ohio; the Michigan nurse-licensing agency quickly authorized the Ohio nurses to practice in Michigan.
Up on the patient floors, the nurses, patient care technicians and other staffers quickly realized that everything had changed. These new fungal patients would be staying longer than most, their medications were more complicated and time-consuming to administer and their reactions to the medication varied widely. Progress in defeating this little-known illness was anything but steady; patients would show improvement for a while, then have a setback.
Hannah Gillmer, a charge nurse on the floor known as 5 North, said nurses like to understand exactly what’s going on and answer patients’ every question, but that isn’t possible with this little-known illness.
“We’re giving these medications to the patients … and they’re asking, ‘Why are my potassium levels all over the board? Why are my legs swelling up?’” Gillmer said. “We know answers to some of the questions, but not all of them. But just by their nature, too, nurses like to try to find the answers to those things, so it’s been a challenge.”
The nurses worked longer shifts, extra days, weekends, often with visiting nurses who needed help transitioning to St. Joe’s routine. As the hospital reached capacity it tried to group the patients together but often a few would be added to a floor usually devoted to other types of patients. That added to the challenge for nurses because of the fungal patients’ unique needs and treatments.
For example, administering the anti-fungal IV’s requires a longer, more complicated and precisely timed set of actions than most medications. That meant a nurse couldn’t deal with as many patients per shift as normal, which affected staffing.
The up and down nature of the patients’ improvement has been one of the most difficult parts, Gillmer said. “I’ve probably never thought so much about work when I’m not at work. I think most if not every nurse on our unit would say the same thing. You think about these people and when they have a bad day, it kind of resonates with you a little bit more because, wow, they’ve had a really good week. And then what’s going on? … They’re having a really bad day. Or they were just about to get to go home and then this happens. You just feel a little more involved, I think.”
As St. Joseph Mercy enters its fourth month of dealing with the fungal outbreak, doctors and staff know the story will continue for years. More patients will develop abscesses, existing patients will linger with the illness, unpredictable complications will develop, long-term consequences will emerge.
Halasyamani, the chief medical officer, says the doctors and staff have moved from the acute “Oh my gosh, what are we going to do?” stage to the more chronic “This is going to be with us for a while” stage.
Is there a light at the end of the tunnel? That implies there is an end and there really won’t be one, Halasyamani says. She prefers to think of it as the staff gaining enough new knowledge to light the tunnel where they are and move forward each day, putting up more lights as they go.
“Unclear about where the tunnel ends and when it will end,” she said.