Iatrogenic Fungal Meningitis: Lessons Learned?
Abstract & Commentary
By Joseph E. Safdieh, MD
Assistant Professor of Neurology, Weill Cornell Medical
Dr. Safdieh reports no financial relationships relevant to this field of study.
Synopsis: An outbreak of fungal meningitis and paraspinal infections occurred over the past year due to contaminated methylprednisolone injections.
Sources: Smith RM, et al. Fungal infections associated with contaminated methylprednisolone injections. N Engl J Med 2013;369:1598-1609. Chiller TM, et al. Clinical findings for fungal infections caused by methylprednisolone injections. N Engl J Med 2013;369:1610-1619.
In september 2012, cases of fungal meningitis started emerging in Tennessee. The first cluster of cases, eight in total, all received epidural steroid injections at the same ambulatory surgical center. Upon investigation, it was determined that all vials of methylprednisolone used in these patients had been purchased from a single compounding pharmacy in Massachusetts. In response, the Centers for Disease Control (CDC) and Prevention and FDA requested a list of all clinical facilities that had ordered the same lot numbers of methylprednisolone from this compounding pharmacy. Cases of fungal meningitis in patients who received injections using these lots of steroid then started becoming apparent in other states, sparking a serious concern of widespread national epidemic. Epidemiologic investigation by the CDC determined that 13,534 patients were potentially exposed to the contaminated steroid — 89% through epidural, spinal, or paraspinal injection, and 11% through peripheral-joint or other injection.
The pathogenic organism implicated in the development of fungal meningitis in these patients was found to be Exserohilum rostratum, an environmental mold. A total of 749 patients developed fungal infection in 20 states. The E. rostratum organism was difficult to culture, with only 20% of cases actually being culture positive. Median age of infected patients was 64 years old. Of the 749 infections, 31% had meningitis only, 20% had meningitis and concomitant paraspinal infection, and 4% had peripheral joint infection. Forty patients had strokes, mostly in the posterior circulation and in association with meningitis. A total of 61 patients (8%) died from complications of the fungal infection. More than 90% of affected patients were not known to be immunosuppressed.
The median time from steroid injection to presentation of infection was 47 days, although cases of stroke tended to present sooner than paraspinal infections. Most patients with meningitis presented with headache (88%); fever was less common (31%). Most patients with paraspinal infections presented with back pain (63%), making it difficult to diagnose, as many of the patients had back pain as the initial indication for the injection. Mean CSF cell count among all cases was 83, mean glucose was 53, and mean protein was 84. Overall attack rate was 5.5 cases per 100 exposed persons.
This series of infections raises a number of important issues. First and foremost, clinicians must be careful to order medications from reputable pharmacies. The fact that a compounding pharmacy produced steroid contaminated with fungus that was injected into more than 13,000 patients is a clear indication that we need better regulatory control over compounding pharmacies. That said, for neurologists who are more likely to recommend epidural injections than perform them, the main issue that arises from this case is the question of whether many of these patients actually needed the injections in the first place. It reminds us that all invasive procedures, even "routine" and "minor" ones like epidural steroid injections come with inherent risk and patients should be carefully selected. The American Academy of Neurology practice parameter on lumbar epidural injections made a level B (good evidence) recommendation that epidural injections showed no impact on average impairment of function, need for surgery, or pain relief after 3 months. A level C (weak evidence) recommendation suggested that these injections may result in limited improvement for 2-6 weeks post-procedure. Risks and benefits must always be taken into account. The fact is that most patients with radicular syndromes would do just fine without these injections, and we should use the available practice parameters to best select patients for these spinal injections.