Editors' Note: PRF member Terri Lewis is a rehabilitation counselor and volunteer patient advocate. Lewis submitted the following perspective and commentary on the fungal meningitis outbreak that occurred when pain patients received spinal injections of contaminated steroid preparations. We invite our readers to respond to Lewis' remarks by leaving a comment below.
In the late summer of 2012, the event we now refer to as the fungal meningitis event was broadcast as a recall of medications made by New England Compounding Center (NECC) of Framingham, Massachusetts, US. As the event unfolded, we learned that some 14,000 persons had been impacted by the receipt of contaminated injections of a variety of types—epidurals (transforaminal, intralaminar, and caudal) into the cervical, thoracic, and lumbar spine; joint injections; and trigger point injections. Thanks to the work of an alert infectious disease clinician, April Petty, and the Tennessee Department of Health, this event was called relatively early, but not before many people were negatively impacted and not before serious adverse events had already begun to unfold as unrecognized deaths associated with contaminated medications.
The role of the media in overcoming some barriers to collaboration cannot be understated. The public media was instrumental in facilitating a heightened national conversation and creating awareness. The Tennessean newspaper in Nashville, then ground zero, created a social media platform for readers to track the national conversation and to follow the events. As a rehabilitation and mental health specialist myself, familiar with the potential gravity of these events, I dropped into this platform to help steer affected individuals to supports and clinical services. I did this because I am from the Nashville area, am familiar with the Nashville spine community, and I am familiar with the reasons the affected persons sought interventional pain supports—they had an acute, chronic, or intractable pain problem for which they had been referred to one of 75 interventional pain clinics or doctor's offices in 24 states.
My professional interest in tracking chronic spinal pain stems from dealing with consumers with injuries highly correlated to traumatic surgical interventions, epidural steroid injections, and childbirth injection-related injuries. I am also the parent of a young adult who developed an onset of arachnoiditis at the age of 15 after a traumatic spinal injury and multiple failed lumbar fusions. I provide supportive counseling to a rapidly growing group of injured consumers aged two to more than 80 years of age. Some have longstanding injuries, while others demonstrate fresh injury post-epidural steroid injection, surgery, or fungal exposure.
To understand this outbreak, one has to examine it in totality—the characteristics of the consumer who sought treatment; the quickly growing field of interventional pain treatment; the equally rapid growth of compounding manufacturing; the supply chain that feeds pharmacy operations; and the state and federal regulatory systems that failed to understand the scale of this problem or respond quickly enough. While there is plenty of concern at every level about what could or should or will be done differently, our first concern must be for the continued support of those who have been affected and the prevention of similar events in the future.
Epidural steroid injections
The "National Quality Forum (NQF) Standard 0309-Back Pain: Appropriate use of epidural steroid injection" defines the use of this procedure as appropriate for persons ages 18 through 80 years of age who seek short-term relief from pain and have radiculopathy or neural pain radiating down one or both legs. As part of the examination in preparation for this procedure, the standard conditions of "NQF 0322-Back Examination" must also be met. Transforaminal, intralaminar, and caudal injections are associated with varying reports of short-term effectiveness, but none are associated with long-term effectiveness, nor should they be considered a substitute for other rehabilitative approaches that address restoration of function. The Center for Medicare and Medicaid Services (CMMS) has placed these injections on their list of procedures that require investigation due to increasing numbers of complaints. These procedures are being routinely used to deliver steroids purchased from registered manufacturers for off-label applications or to deliver unapproved, un-field-tested medicants produced under less than sterile conditions and derived from compounding manufacturers.
These injections are administered with corticosteroids of a variety of types. However, with few exceptions, these are not recommended by the manufacturer for any use except topical applications, intra-lesion or dermal injection, intra-joint (into the synovial space) or intramuscular injection. Nevertheless, with few exceptions, none are appropriate for any use except intramuscular injection. Systemic and intravenous uses are specifically contraindicated. The FDA has not approved methylprednisolone acetate (MPA) for intrathecal use due to its known neurotoxicity and association with the onset of arachnoiditis. Indeed, the manufacturers of Depo Medrol-MPA notified the FDA in 2002 of adverse events reported with epidural steroid use (Pharmacia letter to FDA, 3 November 2002) and requested a notification to manufacturers. This action was never undertaken, but an associated label change was approved for use. Triamcinolone contains a specific label warning by the registered manufacturer against epidural administration due to adverse events associated with intrathecal administration. The label for betamethasone also identifies adverse events associated with intrathecal administration (see Drugs.com).
Within the group of consumers whom I have had contact with, the youngest was 16 years of age at the time of injection, and the oldest was 92. There are reports of affected persons up to 100 years old, as seen in Michigan (Malani et al., 2013). Presenting problems included chronic lumbar pain due to spondylolisthesis, stenosis, radiculopathy, and prior back fusions. Other conditions include complex regional dystrophy syndrome (CRPS), multiple sclerosis (MS), lupus, carpal and tarsal tunnel syndrome, plantar fascitis, diabetic neuropathy, degenerative disc disease, fibromyalgia, failed back syndrome, trauma from car accidents, neural pain, and more variants too numerous to mention. Not one of the individuals I have come into contact with presents as an addict or a malingerer. All were compliant with physician directives, and all sought to participate as fully as possible in their daily lives.
Their history of pain intervention and treatment in some of these cases is longstanding and in other cases quite spotty. Individuals who presented for treatment as the result of a worker’s compensation-related injury have a history of epidural injections being administered as few as three times or as frequently as every two weeks in the prior 12 months. Individuals who were referred for treatment due to an acute injury (e.g., recovery from a car accident) were more likely to receive a series of three injections. For many affected individuals, it was their first experience with an epidural steroid injection. One woman in the contact group, however, reports that she has had nearly 500 injection procedures in the last four years without relief, as directed by her worker’s compensation carrier.
Interventional pain services
Injections are among the top 20 procedures reimbursed by the Medicare program at a cost of more than US$3.6 billion annually. As a group, these include epidural steroid injections, trigger point and facet injections, and joint injections.
In their recent viewpoint article published in JAMA, Staal, Nelemans, and de Bie (Staal et al., 2013) are quite right to point out that there is increasing evidence that disputes the efficacy of these procedures. Indeed, there is increasing evidence of accidentality and even deaths associated with the administration of epidural steroid injections (FDA Adverse Event Reporting System; Weinstein et al., 2006; Epstein, 2013; and the LESS clinical trial in progress).Among the risks associated with the administration of these procedures at this time are the clear, unresolved issues associated with the lack of effective oversight over compounding manufacturing, unsafe injection practices, inappropriate patient selection, and lack of adequate physician training. There is clear concern that the differences which devolved into the expression of illness throughout this event result from the many variants of injection administration, the conditions of administration, the range of medications delivered, and patient characteristics themselves. Patients who received fungal inoculation of the cerebral spinal fluid appear to have suffered a dural puncture with material placement directly into the CSF. The development of illness and subsequent lumbar punctures may have served to inoculate the CSF with fungi secondary to the epidural injection. In recent days, the CDC has released a summary of pathology studies that document the migration of fungal materials to the brainstem where they then invade venous tissues (Ritter et al., 2013).
The delivery of injection materials into the industry is supported by an unknown number of compounding manufacturers who operate at unknown levels of quality and safety. As of this date, 47 compounding firms in 26 states have been asked to recall their products for production, safety, and sterility issues. Currently, the responsibility for oversight rests with State Departments of Health through their respective Pharmacy Boards. The culprits in the case of the NECC disaster are the fungi aspergillus, exserohilum, and cladosporium, with exserohilum being the most prevalent. An additional number of bacteria and viruses have been identified.
Management of the outbreak
As this event is iatrogenic in nature—the result of a medical treatment and not an infectious disease process—it falls under the purview of federal-state agreements. Had it been a declared national health emergency, it would have tripped the processes that would have resulted in a single management system from which to operate. This single fact has introduced tremendous variability in the national response. Every affected state has mounted its own system of support with varying degrees of resources, communications, and commitment.
Early in the event, this illness was characterized as a fungal meningitis, and the response system for triage resembled that which would ordinarily be associated with a widespread meningitis of a viral or bacterial nature. Communications were deployed into the field that directed the use of lumbar puncture without regard for prior patient characteristics, the trained capacity of community health organizations, or recognition of potential disease and organism spread by the procedure itself. For those of us who were acquainted with these issues, the concern began to elevate that the response had the potential to significantly increase harm due to secondary adverse effects. We have, in fact, seen that through the advent of spinal leaks, dural punctures, and so forth. The CDC messaged early that their modeling indicated that the event would begin to wane at 42 days. At 60 days we were still seeing new cases develop and more states affected. The modeling was revisited, as direct care physicians became more experienced and involved. We are still seeing cases permute into new expressions of illness that indicate systemic fungal presence (see recent news coverage).
In the same issue of JAMA, Malani and colleagues (Malani et al., 2013) describe the use of magnetic resonance imaging (MRI) to identify spinal and paraspinal infections associated with injections of contaminated MPA. Initially, as consumers were directed in local emergency rooms to navigate the variety of procedures employed for diagnosis, I became quite concerned that the MRI was being underutilized by the CDC protocols. Over time, I was ecstatic to see the MRI become the tool of first response due to its noninvasive characteristics and ability to find infections that had not revealed themselves in total as of yet. However, there were barriers to the use of this tool.
First, it must be noted that many patients reported and expressed symptoms that did not become associated with the outbreak because of the way in which the CDC had characterized the expression of illnesses. Second, because this was a fungal inoculation event with widespread variability in onset, the range of indicators and consumer geographic dispersion did not make these indicators apparent to all. Third, differential diagnosis of this symptomology was made much more difficult due to the range of prior injuries associated with chronic pain syndromes present in the affected population. Fourth (and most importantly), across the states, all diagnostic access was passed to consumers—those with insurance or resources got diagnosed. But those who relied on worker’s compensation or their own cash resources often were denied uniform access to MRI. Malani and colleagues note that a significant number of their studies gave results that were not clear enough for unequivocal interpretation. Despite these limitations, it has become clear over time that early detection through the use of MRI with contrast is important to ascertainment and identification of the overall outcomes. Its use can be correlated with other measures.
While MRI has demonstrably proven its worth as the tool of choice for characterizing the disease footprint, there are some secondary effects that we have yet to know very much about. The antifungals are very hard on the liver and kidneys. Because they are combined with repeated exposure to gadolinium, we have yet to determine what we will see over time in terms of consumer health. As expected, a number of consumers have demonstrated liver and kidney impairment over the many months, and some are notably endangered. The knowledge that this problem had the potential to develop has kept us searching for alternatives that are less invasive, efficient, valid, and reliable.
Where we stand today
Pain management in this population has become a crisis. Nearly 60 cases of arachnoiditis have been newly detected among the 800 cases of infections identified to date, and we believe there are many more who remain undetected (see Centers for Disease Control Health Advisory Information). Given the unknown natural history of many of these cases, one cannot clearly state that these individuals got this disease from the fungal meningitis. From patient reports and medical records, there are indications that the arachnoiditis may well have been present before the contaminated injection was administered, particularly where there was a history of failed back surgeries and/or repeated epidural steroid injections. When requested, I have arranged for an independent outside review of imaging studies and medical records on behalf of some patients who cannot navigate this system on their own. Review of historical MRIs, together with recent imaging, leads to the conclusion that in many cases, patients have been overexposed to percutaneous procedures with accumulating injury—the contaminated injection was simply “icing on the cake.”
Increasingly, interventional pain physicians fail to employ the procedures necessary to determine patient health or structural status prior to administration of percutaneous procedures. This shows up as increased accidentality and deviation from the existent NQF standard 0309—if one injection or 80 mg dose is ineffective, then subjectively, more is employed at greater frequency. This approach is documented in patient records and is associated with at least one outbreak-related death. I am very concerned about the path forward for these families in both this outbreak and the newest outbreak associated with contaminated medications purchased from a compounding pharmacy in Tennessee.
It is my belief and observation, based on working with hundreds of cases, that epidural steroid injections are ineffective as used, and their increasing use supplants other therapeutic approaches. We do not learn of adverse events as negative data in studies simply because there is currently no requirement to report injury as a sentinel event. Further, the use of these procedures offers us none of the translational capture of knowledge needed to understand chronic intractable pain as a disease process all by itself. Importantly, it is becoming harder and harder for patients to obtain support for pain without being subjected to a routine that forces epidural steroid injections first and other measures second. More and more, the messaging about chronic intractable pain is interpreted and controlled by an industry with clear designs on the Medicare dollar. More frequently, we see people with disease processes due to legitimate injuries classed as malingerers and addicts when in fact they are being harmed by the very procedures forced upon them.
That we need pain research is clear. This single event demonstrates the necessity for assuring consumers they will not continue to be exposed to highly risky, arbitrary, inefficacious, and even dangerous procedures. The role of epidural injections is being studied as we speak by Janna Friedly at the University of Washington, Seattle, US, in cooperation with George Washington University, Washington, D.C., US. Her results are pending, but other results are offered to us for consideration. I include these in the references.
We have established an informal work group with Thomas Walsh at Cornell University, Ithaca, New York, US; David Perlin at the New Jersey Medical School, Newark, US; and Sarah Sellers, a leader in pharmacy safety. I invite your participation and collaboration as we explore safe interventions for pain treatment in this group of injured consumers. I can be reached at email@example.com.
Surg Neurol Int. 2013 Mar 22;4(Suppl 2):S74-93.
Malani AN, Vandenberg DM, Singal B, Kasotakis M, Koch S, Moudgal V, Jagarlamudi R, Neelakanta A, Otto MH, Halasyamani L, Kaakaji R, Kauffman CA.
JAMA. 2013 Jun 19;309(23):2465-7.
Ritter HM, Muehlenbachs A, Blau DM, Paddock CD, Shieh WJ, Drew CP, Batten BC, Bartlett JH, Metcalfe MG, Pham CD, Lockhart SR, Patel M, Liu L, Jones TL, Greer PW, Montague JL, White E, Rollin DC, Seales C, Stewart D, Deming MV, Brandt MV, Zaki SR.
Am J Pathol. 2013 Jun 21.
Staal BJ, Nelemans PJ, de Bie RA.
JAMA. 2013 May 16:1-2.
Weinstein JN, Tor D. Tosteson TD, Lurie JD, Tosteson AN, Hanscom B, Skinner JS, Abdu WA, MS, Hilibrand AS, Boden SD, Deyo RA.
JAMA. 2006 Nov 22;296(20):2441-50.
Facebook group: Meningitis Outbreak, sponsored by the Tennessean news organization
Samanta A, Samanta J
BMJ.2004 Jun 26;328(7455):1509-10.
Abbott ZI, Nair KV, Allen RR, Akuthota V.
Spine J. 2012 Jan;12(1):35-43. Epub 2011 Dec 3.
Pham JC, Frick KD, Pronovost PJ.
Am J Med Qual. 2013 Mar 24.
Thompson GR, Kontoyiannis DP, Patterson TF.
JAMA. 2013 Jun 19;309(23):2493-5.
Wilson LE, Blythe D, Sharfstein JM.
JAMA. 2012 Dec 19;308(23):2461-2.