nesthesiologists have long been at the forefront of the practice of pain medicine, particularly in the application of technical procedures in the management of acute, chronic and cancer pain. Although there are recognized limitations in the analysis, the ASA Closed Claims Project database provides valuable information on the adverse outcomes in chronic pain management from 1970 through December 2000.¹

Chronic pain management claims increased from 2 percent in the 1970s to 10 percent in the 1990s. Out of a total of 284 chronic pain management claims, 276 (96 percent) were related to invasive procedures. These procedures included nerve blocks (using local anesthetic as the primary agent), injections (epidural steroids, trigger points, botulinum toxin, tendon or joint injections), ablative procedures, implantation or removal of devices, and maintenance of devices. Of these procedures, blocks and injections comprised 78 percent of the claims. The most common complications involved nerve injury and pneumothorax [Figure 1]. Death or brain injury was less frequent and was related to epidural steroid injections and device maintenance.

Figure 1 Primary outcome in chronic pain management claims (solid bars) versus surgical/obstetric claims (open bars). *p<0.05 Fitzgibbon et al. Anesthesiology. 2004; 100:98-105.

Half of the 63 nerve injury claims involved spinal cord injury, including 14 epidural steroid injections, six of which resulted in paraplegia and one quadriplegia. Other procedures associated with nerve injury involving the spinal cord included blocks (5), ablative procedures (3), cervical facet (1), implantation or removal of devices (2) and device maintenance (4). There were 18 claims for paraplegia or quadriplegia, and they included epidural abscess (4), chemical injury from injection into the spinal cord (8) and epidural hematoma (4). Other nerve injury claims included the lumbosacral nerve root (21), the sciatic nerve (2) and the brachial plexus (2).

There were 59 claims for pneumothorax, 40 of which were associated with blocks and 18 from injections, predominantly trigger-point injections. More than half (34) of the patients were diagnosed with pneumothorax after they had been discharged, and of these, 15 patients were diagnosed and treated in an emergency room.

Infection accounted for 13 percent of all claims from invasive procedures. Most were associated with epidural steroid injections and some with implantation, removal or maintenance of implanted devices. Meningitis (12) was the most common presentation followed by epidural abscess (7) and osteomyelitis (3). Surgical intervention was required in six of the seven epidural abscesses with one patient sustaining permanent neurological deficits.

Claims for death or brain damage were associated with epidural steroid injections (9) and device maintenance (9). Of the 114 claims related to epidural steroid injections, 61 of the injections used local anesthetics and or opioids. All nine epidural steroid injections resulting in death or brain damage contained local anesthetics with or without opioids [Figure 2]. Events include unintended intrathecal injection (5), allergic reaction (1), cardiovascular collapse and respiratory depression from inadvertent intrathecal injection at the thoracic level (1), and delayed respiratory depression from epidural morphine administered with the steroid (3).

Figure 2 Most common outcomes in epidural injections. Injections with steroids only (solid bars); injections with addition of local anesthetic and or opioid to the steroid (open bars). *p<0.05 between proportion of injection group with that outcome. Fitzgibbon, et al. Anesthesiology. 2004;100:98-105.

Claims for death (4) or brain damage (5) associated with the maintenance of devices included implanted pumps, epidural injections and patient-controlled analgesia. Opioids were prescribed in all nine claims. Nearly all claims involved the administration of the wrong dose of opioid. Other events included pump programming error, drug interactions and intrathecal migration of the epidural catheter.

Epidural steroid injection, commonly performed to treat radicular pain, has been utilized for more than 40 years and is generally perceived to be a safe procedure. In the ASA Closed Claims Project database, however, epidural steroid injections accounted for 40 percent of all chronic pain claims and more than 80 percent of the claims related to injections. The injuries were serious, relating to nerve injury, infection, death or brain damage. In more recent years, catastrophic events related to cervical transforaminal injection of steroids became apparent.² These events included severe spinal cord infarction with severe neurologic sequelae. Case reports included cortical blindness and neurologic injury³ and brainstem herniation from cerebellar infarct.⁴ Paralysis from lumbar transforaminal injection also has been reported.⁵ It is perceived that particulate steroid injected into a radicular artery critical to spinal cord perfusion results in an embolic phenomenon causing spinal cord infarction. Other factors that may have contributed to these devastating events are lack of detection of intra-arterial injection (with or without real-time fluoroscopy), patient sedation and variable anatomy, particularly in the course of the small radicular vessels in relation to the nerve root.

Last fall the Anesthesia Patient Safety Foundation⁶ reported claims collected by an insurance company that had “noted an alarming incidence of major claims relating to cervical epidural steroid blocks. In fact the number of claims for these blocks consistently exceeds the combined total of claims for steroid blocks performed at all other levels….” The company collected and reviewed 13 anesthesiology claims over three years. It was not reported whether the interlaminar or transforaminal approach was used. Injuries reported included arachnoiditis, paralysis, anoxic brain damage and death.

Claims related to chronic pain management have increased over the last several decades.¹ Recent catastrophic events associated with epidural steroid injections — in particular, cervical transforaminal injections — will likely lead to a further increase in claims in the coming years. As the practice of pain medicine expands and evolves with new drugs, new devices and new techniques, there is a need for greater awareness of potential injuries to patients and the institution of guidelines regarding safe practice. Within this context, we should continue to address issues that may potentially 1) improve outcomes related to invasive procedures by minimizing risks for death or brain damage, nerve injury, bleeding, infection and pneumothorax, and 2) decrease the occurrence of claims: issues such as the emphasis on a clear discussion of informed consent and indications for the proposed procedure with alternative treatments presented; adequate training of interventionalists; the consideration of the use of nonparticulate steroid and the use of real-time fluoroscopy with digital subtraction to maximize detection of intra-arterial injection in transforaminal injections; patient sedation; patient monitoring; immediate availability of resuscitation equipment; postprocedure instructions; patient communication; and follow-up.


References:
1. Fitzgibbon DR, Posner KL, Domino KB, et al. Chronic pain management: ASA Closed Claims Project. Anesthesiology. 2004; 100:98-105.
2. Rathmell JP, Aprill C, Bogduk N. Cervical transforaminal injection of steroids. Anesthesiology. 2004; 100:1595-1600.
3. McMillan MR, Crumpton C. Cortical blindness and neurologic injury complicating cervical transforaminal injection for cervical radiculopathy. Anesthesiology. 2003; 99:509-511.
4. Beckman WA, Mendez RJ, Paine GF, Mazzilli MA. Cerebellar herniation after cervical transforaminal epidural injection. Reg Anesth Pain Med. 2006; 31(3):282-285.
5. Huntoon M, Martin D. Paralysis after transforaminal epidural injection and previous spinal surgery. Reg Anesth Pain Med. 2004; 29:494-495.
6. Lofsky A. Complications of cervical epidural blocks attract insurance company attention. APSF Newsletter. 2005; 20(3):45-48.
.

May L. Chin, M.D., is Professor of Anesthesiology, Director, Division of Pain Medicine, and Director, Pain Management Center, George Washington University Medical Center, Washington, D.C.