Subscribe to eTOC

Secondary Epidural Abscesses Emerge in Meningitis Outbreak

MICHIGAN AND TENNESSEE currently report the highest number of secondary infections to fungal meningitis cases linked with contaminated epidural steroid injections.

A growing number of patients who developed fungal meningitis from mold-contaminated steroid injections from a New England compounding company have developed secondary epidural abscesses, many of them weeks after being discharged from hospitals, according to state and federal health officials. Several cases of secondary arachnoiditis have also been reported.

The exact number of these cases remains unknown. However, in Michigan, the state with the most meningitis cases in the outbreak, there are now 158 meningitis cases, and 86 reported cases of epidural abscess. Abscesses have also been reported in Tennessee, with the second highest meningitis caseload of 82 cases. State health officials had not tabulated the exact number at press time. Cases have been reported in other states as well.

All patients developed meningitis after receiving injections of contaminated preservative-free methylprednisolone acetate (MPA) made by the New England Compounding Center (NECC), in Framingham, MA, which has since been shut down. An affiliate, Ameridose, has also been closed.

Nationwide, the number of cases of secondary infections has yet to be determined, but the Centers for Disease Control and Prevention (CDC) said it is monitoring cases and has issued weekly interim treatment guidelines, including a Nov. 8 update that acknowledged secondary abscesses associated with the outbreak and recommended treatment options. (See “CDC Interim Treatment Options.”)

On Nov. 6, the agency also launched a volunteer Clinicians Consultation Network, with experts in fungal disease treatment and management, to help physicians with such patients. Consultations can be made by telephone or e-mail by calling CDC (1–800–232–4636) or by e-mail at


Health officials are perplexed by the secondary infections because fungal epidural abscesses have been, until the outbreak, extremely rare. Fewer than 100 cases have been reported in the medical literature, and most of these have been single case studies. Moreover, such cases typically have occurred prior to development of meningitis, according to Carol A. Kauffman, MD, professor of internal medicine at the University of Michigan Medical School and the Veterans Affairs Medical Center in Ann Arbor.

“We are seeing patients who were treated in the hospital for meningitis and sent home, only to come back with secondary epidural abscesses and sometimes arachnoiditis,” said Dr. Kauffman, an expert in invasive fungal infections.

“We thought they were doing well on oral therapy, but then some started coming back complaining of back pain. We have found abscesses in a number of patients who were discharged several weeks before.”

In the first week of November, she told Neurology Today, one hospital in Ann Arbor treated 25 to 30 such patients. All treated meningitis patients have since been contacted, told to be aware of any new symptoms, and are undergoing MRI scanning to look for abscesses, Dr. Kauffman told Neurology Today in a telephone interview.

DR. WILLIAM SCHAFFNER: “The concern is that patients who have been doing well on treatment are later becoming symptomatic, and I think it has yet to be determined whether these are in fact new abscesses — this is something that is still being worked out. Some may be new while others are persistent and a consequence of insufficient (dura) penetration by voriconazole. At the very least it is going to be a problem for us.”

“All of us thought it was over, but now I just do not know. It seems like the more you look, the more you find.”


In Tennessee, fewer patients have presented with secondary epidural abscesses or arachnoiditis, said William Schaffner, MD, professor and chair of preventive medicine at Vanderbilt University Medical Center in Nashville, TN.

“There seem to be more cases in Michigan, but we are not sure how many cases there are [in Tennessee] because the state has yet to release any data,” he told Neurology Today in a telephone interview. “We are all hoping this is not a harbinger of things to come.”

Dr. Schaffner added that it remains unclear if meningitis patients are experiencing new abscesses, or if these are existing infections that were not initially detected and have failed to respond to treatment with voriconazole.

“Abscesses have been part of the clinical spectrum from the beginning,” Dr. Schaffner said. “The concern is that patients who have been doing well on treatment are later becoming symptomatic, and I think it has yet to be determined whether these are in fact new abscesses — this is something that is still being worked out. Some may be new while others are persistent and a consequence of insufficient (dura) penetration by voriconazole. At the very least it is going to be a problem for us.”


Neurology Today asked the experts whether drug resistance or length of therapy could be a factor in the secondary infections. There are very little research data on voriconazole resistance in patients with these specific fungi, and what is available pertains primarily in pulmonary infections with Candida albicans or Aspergillis fumigatus, said Dr. Schaffner.

“It remains disturbing that some patients are becoming newly symptomatic during treatment,” he said.“Drug resistance is always a possibility, but is not usually a problem in fungal infections. However there are no data yet, so it needs to be watched.”

Dr. Kauffman noted that all affected patients who had meningitis and then later developed epidural abscesses had remained on therapy, and none had stopped. “In general antifungal drugs do not act rapidly to kill fungi as [quickly] as antibacterials. We could certainly use better, less toxic drugs but there are none on the horizon,” she said.


The first cases of meningitis occurred in early September and the outbreak reached its peak in late October. At least three lots of MPA and one of cardioplegia solution, all manufactured by NECC have been found to be contaminated, most by the fungus Exserohilum rostratum. In addition, one patient was infected with Aspergillus fumigatus and another with a Cladosporium.

“Such fungal infections have not been recognized as a cause of meningitis before. This is uncharted territory,” said Tom M. Chiller, MD, deputy chief of the CDC's Mycotic Diseases Branch, in a CDC podcast commentary.

In October, the CDC and the Food and Drug Administration (FDA) launched an intensive outreach effort to alert all involved physicians and patients of the contaminated products, and all of the 14,000 patients who received injections have been notified. At press time, there were 480 cases (469 cases of meningitis, stroke due to meningitis, or other CNS-related infection meeting outbreak case definition); and 33 deaths in 19 states, however officials remain cautious about the ultimate extent of the outbreak. The numbers of cases rise each week.

While the incubation period for meningitis from fungal infection is usually 42 days, cases have been known to occur 100 or more days after exposure, Dr. Chiller said. The CDC risk period of 42 days ended Nov. 7, yet more than a dozen new cases were subsequently reported by Nov. 12.

Both the U.S. House of Representatives and Senate scheduled hearings on the outbreak, on Nov. 14 and 15, respectively. NECC cofounder and president Barry Cadden was subpoenaed to appear at the Nov. 14 House hearing.


On Nov. 8, the CDC issued the following guidance for treating patients with severe disease and those who remain symptomatic or develop secondary abscesses despite voriconazole therapy:

  • Evaluation of new or worsening symptoms should include appropriate diagnostic imaging at the site(s) including imaging of the injection site approximately 2–3 weeks after diagnosis of meningitis.
  • Early consultation with a neurosurgeon to discuss new symptoms, imaging results, further diagnostic workup, and patient management (strongly encouraged).
  • Voriconazole at a dose of 6 milligrams per kilogram (mg/kg) every 12 hours.
  • “Strongly consider” liposomal amphotericin B in addition to voriconazole (of 5 to 6 mg/kg IV daily), and as an alternative to voriconazole in patients who cannot tolerate voriconazole.
  • Higher doses of liposomal amphotericin B (7.5 mg/kg IV daily) may be considered when patients do not improve, however all patients should be closely monitored for liver toxicity.
  • Early diagnosis and management may require other approaches than those for fungal meningitis alone for patients who develop new or worsening symptoms, complications, or possible progression of fungal infection, including epidural abscesses or masses, arachnoiditis, mycotic aneurysms, or stroke.
  • Because adequate duration of antifungal treatment is unknown, prolonged therapy should be tailored by clinical response to treatment. Depending on a patient's response, a minimum of 3 months of antifungal treatment should be considered.
  • Treatment decisions, including choice of long-term antifungal treatment regimens, should be made in consultation with fungal infection diseases experts.


While the CDC has recommended MRI imaging of patients with symptoms of secondary abscesses, health officials may be overlooking a blood/spinal fluid test that can be used to diagnose invasive fungal infections, even in asymptomatic patients, according to Karen Roos, MD, John and Nancy Nelson professor of neurological surgery at Indiana University in Indianapolis.

“We are in the midst of an unprecedented, life-threatening infection,” Dr. Roos, an anti-fungal therapy expert, told Neurology Today in a telephone interview. As of Nov. 14, 53 meningitis cases and 4 deaths had been reported in Indiana.

Dr. Roos said that a 2009 study in the journal Clinical and Vaccine Immunology showed that a blood and spinal fluid test called β-D-glucan (BG) can detect common and uncommon fungal infections such as the mold associated with the current outbreak. BG detects a cell wall component of black-brown mold and other fungi. The 2009 paper involved 12 patients, six of whom had brain abscesses.

But experts from the CDC expressed concerns about the test. Benjamin J. Park, MD, chief epidemiologist with the CDC Mycotic Diseases Branch, who heads the agency's meningitis outbreak task force, told Neurology Today that the agency does not believe the BG test is specific enough to be used for accurate diagnosis in the current outbreak.

“β-D-glucan is a component of the cell wall of most fungi. Because of this, it is a very non-specific test [and] infection with almost any fungus will yield a positive result,” he commented.

Tom M. Chiller, MD, deputy chief of the CDC Mycotic Diseases Branch, agreed. He told Neurology Today in a telephone interview that because so many fungi exist in the skin and elsewhere, the test is associated with a high positive rate that makes results inconclusive.

“From the beginning we have been asking experts about any tests that might be useful, but none have been recommended,” he said. “This is an unprecedented outbreak, and no one has much experience with it.”

He said the CDC continues to stress a low threshold for imaging of the injection site in all meningitis patients who show any signs or symptoms several weeks after treatment.

— Kurt Samson


• Drazen J, Curfman G, Baden L, Morrisey S. Compounding errors. N Engl J Med 2012; E-pub 2012 Nov. 7.
    • Kainer M, Dreyzehner J, Jones T for the Tennessee Fungal Meningitis Investigation Team. Fungal infections associated with contaminated methylprednisolone in Tennessee. N Engl J Med 2012; E-pub 2012 Nov. 6.
      • Soledad Cuetara M, Alhambra A, Moragues D, et al. Detection of (1–33) β-D-glucan as an adjunct to diagnosis in a mixed population with uncommon proven invasive fungal diseases or with an unusual clinical presentation. Clin Vaccine Immunol 2009;16:423–426.
        Neurology Today archive on fungal meningitis:
          CDC Updates on fungal meningitis :