Near my office, there’s a breakfast-and-lunch joint where strangers sit down at shared tables. When Joseph, an attorney in his early seventies, heard that I was writing a book about the back pain industry, he started asking questions.
That afternoon, he was scheduled to have the first of three epidural spinal injections meant to relieve the symptoms of spinal stenosis, a condition in which the spinal cord narrows to the point where it squeezes the nerves. The weakness and cramping in his legs were so bad that he couldn’t walk a long city block without stopping to recover.
The injections were his best hope, he said, making it even harder for me to break the news. A few months earlier, in a 2015 review of the medical literature, the Agency for Healthcare Research and Quality had found no evidence that epidural steroid injections were effective in treating symptoms of spinal stenosis or typical lower-back pain. Even in the presence of a recent disc herniation and ensuing sciatica, the benefits of injections were small and not sustained over time. That news followed on the heels of an FDA statement warning that injection of the active medication in these shots, glucocorticoids — a class of corticosteroids — into the epidural space of the spine could result in rare but serious neurological problems, including loss of vision, stroke, paralysis and death. Based on those and other findings, the Journal of the American Medical Association ( JAMA) advised physicians to refrain from recommending injection therapy to patients with any kind of chronic back pain.
That news did not go over well with the doctors known as “interventional pain physicians,” who make a living performing such procedures. In the United States, more than ten million epidural steroid injections are delivered each year, a number that makes them the bread and butter of interventional pain management practices.
I was not surprised that my lunch partner didn’t have the facts. Primary care physicians who ordered the shots were rarely informed about the lack of evidence and the risks of treatment. Even young, healthy people, explained anesthesiologist James Rathmell, the chair of the Department of Anesthesiology, Perioperative and Pain Medicine at Boston’s Brigham and Women’s Hospital, could go in with manageable low-back-pain symptoms and come out with catastrophic neurological injuries. “The bottom line,” said Rathmell, “is that if you come into my clinic with chronic axial back pain, you’re not going to get epidural steroid injections — because they don’t work.
“People should get the best evidence-based treatment they can,” he added. “As a rule of thumb, if you pay practitioners to do stuff, they will do more stuff. Frankly, what’s happened in interventional pain management is just a microcosm of what’s happened in all of medicine.”
As I explained these things to Joseph, he paled. Why would his doctor advise him to undergo a worthless and risky procedure?
Three decades ago, anesthesiologists had no trouble getting jobs in hospital ORs. On a busy morning, they could run five cases at once, and get paid for them all, while depending on registered nurses to keep an eye on individual patients. Starting in the early 1990s, cost-conscious health management organizations (HMOs) realized that the nurses could manage without supervision, and they stopped paying fees to doctors who at best were marginally present. Many anesthesiologists found themselves underemployed. They knew little about treating musculoskeletal disorders or how to address the feelings of depression, anger, and isolation that often afflict back pain patients, but when they retooled, they set up pain management practices.
Torpedoed into entrepreneurship, the most successful interventional pain physicians offered an ever-expanding menu of injections, including facet and sacroiliac joint blocks, selective nerve blocks, discography, needle electromyography, radio-frequency and thermal facet ablations, botulinum toxin, and trigger point injections. They implanted intrathecal drug delivery pumps and spinal cord stimulators and cemented together vertebral compression fractures. They used medical lasers to heat, shave and slice soft tissue. By the turn of the new millennium, interventional pain management, which a decade before had barely existed, had become one of the most profitable aspects of spine care.
There are, indeed, people who undergo one perfectly targeted epidural steroid injection and hit the golf course the next morning, completely cured. In the more typical scenario, however, the first injection — if in fact, it provides any relief at all — is only briefly effective. Then the numbing medicine and the anti-inflammatory effect of the glucocorticoid wear off, and the pain returns.
Generally, the shots are ordered in a series of three, although no expert I asked could say why, and the American Society of Anesthesiologists’ guidelines do not advise the administration of a specific number. “You always do three, even if the first two do no good at all,” wisecracked neurosurgeon Charles Burton, who publicly questioned the safety and effectiveness of the procedure, long before JAMA and the FDA got on board. Some doctors construed the “rule of three” to mean that, in a single visit, they could give three shots at each affected vertebral level, thereby exposing a patient to a colossal dose of glucocorticoid. In fact, when Colorado researchers mined an insurance company’s database, they found that one doctor had billed a single patient for fifty-one such injections in one year. The same database showed that a New Jersey patient had received thirteen shots in a five-month period and had subsequently developed kidney failure.
The FDA had been issuing cautionary statements about epidural steroid injections since 1981. But in 2014, the agency took a further step compelling pharmaceutical manufacturers that produced the injectable glucocorticoids to clearly state the risks on every vial’s label, advising that “serious neurologic events, some resulting in death, have been reported with epidural injection,” and that the “safety and effectiveness of epidural administration of corticosteroids have not been established.” The FDA stopped short of requiring manufacturers to notify physicians or their purchasing departments that things had changed, and most did not notice. One pain physician, Cleveland Clinic’s Richard Rosenquist, told Bloomberg reporter David Armstrong that because he’d used such drugs for his entire career, unless he was alerted to do so, he was “unlikely to go back and spend time reading the package insert.”
The implications were significant: Properly “consented,” a patient who was about to receive an epidural steroid injection would hear about specific risks, including damage to the dura mater (the sturdy sleeve surrounding the spinal cord), nerve root injury, elevated cholesterol levels, vertebral fractures, the death of muscle and bone tissue, staph infection, epidural abscess, immune system deficits, stroke and death. But in reality, if this information was conveyed at all, it was in boilerplate format, which the patient signed after only a cursory glance.
There are two dominant techniques for administering epidural steroid injections. In the first, known as “interlaminar,” the needle is directed into the epidural space, around the spinal nerves. In the second approach, referred to as “transforaminal,” the needle is inserted at an angle, which places it closer to the targeted nerve but also in the vicinity of vessels and arteries. Incorrectly placed, the needle can sever an artery or deliver medication into the blood vessels, clogging them and preventing adequate blood flow to the brain. The result, in either case, may be stroke or paralysis.
Whether the approach is transforaminal or interlaminar, research shows that a quarter of epidural steroid injections miss their targets. In “blind” injections, performed without fluoroscopic guidance, the needle is incorrectly placed in up to half of epidural steroid procedures.
In roughly six percent of epidural steroid injections (a number that sounds small but is not, because thousands of injections are delivered every day), the needle nicks the dura mater, the sturdy sleeve surrounding the spinal cord, allowing cerebrospinal fluid to leak out. Typically, this is not terribly serious. It results in a severe headache, which goes away after the patient lies flat for a couple of days. Sometimes, another procedure, known as a “blood patch,” is used to stop the leak of cerebrospinal fluid.
But when the needle actually punctures the dura mater, it’s a different story. Then the payload of glucocorticoid and anesthetic may be delivered into a region of fragile nerve tissue called the subarachnoid space. From there, the cerebrospinal fluid, bearing its toxic load, circulates to the brain, where the cortisone solution efficiently strips the insulating (and essential) myelin layer of neurons. One result is “adhesive arachnoiditis,” a condition so grossly debilitating that neurologist Dewey Nelson described it as akin to “having a blowtorch up your rectum. It binds the nerves, like gunky cooked spaghetti, and the result is unrelenting pain that may last for a lifetime.”
Cathryn Jakobson Ramin is an investigative journalist and author of Crooked: Outwitting the Back Pain Industry and Getting on the Road to Recovery. Reprinted with permission of Harper, an imprint of HarperCollins Publishers
Image credit: Shutterstock.com