Secondary Logo

Journal Logo

A Contemporary Medicolegal Analysis of Outpatient Interventional Pain Procedures: 2009–2016

Abrecht, Christopher R. MD*; Saba, Ramsey MD; Greenberg, Penny RN, MS; Rathmell, James P. MD; Urman, Richard D. MD, MBA

doi: 10.1213/ANE.0000000000004096
Chronic Pain Medicine: Original Clinical Research Report
Free

BACKGROUND: Closed malpractice claim studies allow a review of rare but often severe complications, yielding useful insight into improving patient safety and decreasing practitioner liability.

METHODS: This retrospective observational study of pain medicine malpractice claims utilizes the Controlled Risk Insurance Company Comparative Benchmarking System database, which contains nearly 400,000 malpractice claims drawn from >400 academic and community medical centers. The Controlled Risk Insurance Company Comparative Benchmarking System database was queried for January 1, 2009 through December 31, 2016, for cases with pain medicine as the primary service. Cases involving outpatient interventional pain management were identified. Controlled Risk Insurance Company–coded data fields and the narrative summaries were reviewed by the study authors.

RESULTS: A total of 126 closed claims were identified. Forty-one claims resulted in payments to the plaintiffs, with a median payment of $175,000 (range, $2600–$2,950,000). Lumbar interlaminar epidural steroid injections were the most common procedures associated with claims (n = 34), followed by cervical interlaminar epidural steroid injections (n = 31) and trigger point injections (n = 13). The most common alleged injuring events were an improper performance of a procedure (n = 38); alleged nonsterile technique (n = 17); unintentional dural puncture (n = 13); needle misdirected to the spinal cord (n = 11); and needle misdirected to the lung (n = 10). The most common alleged outcomes were worsening pain (n = 26); spinal cord infarct (n = 16); epidural hematoma (n = 9); soft-tissue infection (n = 9); postdural puncture headache (n = 9); and pneumothorax (n = 9). According to the Controlled Risk Insurance Company proprietary contributing factor system, perceived deficits in technical skill were present in 83% of claims.

CONCLUSIONS: Epidural steroid injections are among the most commonly performed interventional pain procedures and, while a familiar procedure to pain management practitioners, may result in significant neurological injury. Trigger point injections, while generally considered safe, may result in pneumothorax or injury to other deep structures. Ultimately, the efforts to minimize practitioner liability and patient harm, like the claims themselves, will be multifactorial. Best outcomes will likely come from continued robust training in procedural skills, attention paid to published best practice recommendations, documentation that includes an inclusive consent discussion, and thoughtful patient selection. Limitations for this study are that closed claim data do not cover all complications that occur and skew toward more severe complications. In addition, the data from Controlled Risk Insurance Company Comparative Benchmarking System cannot be independently verified.

From the *Department of Anesthesia and Perioperative Care, Division of Pain Medicine, University of California San Francisco, San Francisco, California

Department of Anesthesiology, Perioperative, and Pain Medicine, Brigham and Women’s Hospital, Boston, Massachusetts

Controlled Risk Insurance Company Strategies, Boston, Massachusetts.

Published ahead of print 22 January 2019.

Accepted for publication January 22, 2019.

Funding: None.

The authors declare no conflicts of interest.

Reprints will not be available from the authors.

Address correspondence to Christopher R. Abrecht, MD, Department of Anesthesia and Perioperative Care, Division of Pain Medicine, University of California San Francisco School of Medicine, Pain Management Center, 2255 Post St, San Francisco, CA 94143. Address e-mail to christopher.abrecht@ucsf.edu.

See Editorial, p

In the context of the opioid epidemic, interventional pain procedures may be of particular value in that they represent effective pain management options beyond systemic analgesic therapy.1 Previous studies have documented some of the potential complications related to commonly performed pain procedures, such as facet and epidural steroid injections.2–5 Previous studies have also suggested that by 65 years of age, physicians in lower-risk areas of practice will face a 75% probability of having experienced a malpractice claim; by 65 years of age, physicians in higher-risk areas of practice will face a 99% probability of having experienced a malpractice claim.6 The multidisciplinary specialty of pain medicine was not a practice group in this analysis; anesthesiology was identified as an intermediate-risk group. The academic review of malpractice claims has been performed since the 1980s by the American Society of Anesthesiologists (ASA) Closed Claims Project.7

The aim of this study was to use a distinct methodology to review closed malpractice claims related to interventional pain procedures. The objective was to include cases involving pain practitioners from specialties besides anesthesiology and provide the most contemporary analysis. This and other closed claim analyses allow a review of rare but often severe complications, yielding useful insight for improving patient safety and decreasing practitioner liability.

Back to Top | Article Outline

METHODS

Institutional review board approval was obtained for this study. Written consent was waived, given the deidentified, retrospective nature of the study.

Controlled Risk Insurance Company was established as a professional liability insurer for hospitals in the Harvard medical community, and now includes the Controlled Risk Insurance Company Strategies Comparative Benchmarking System. This database contains nearly 400,000 malpractice claims drawn from >400 academic and community medical centers, and is estimated by Controlled Risk Insurance Company to contain 30% of all malpractice claims in the United States. This Comparative Benchmarking System database exists to study and mitigate patient and provider risk using a standardized approach.

The Comparative Benchmarking System database was queried for the interval from January 1, 2009 to December 31, 2016, limiting the search for cases in which a pain medicine practitioner was the primary treating service. Claims closed during this period were included in the study. The cases resulting from this query were individually examined by the study investigators to identify those pertaining to outpatient interventional pain procedures, excluding pain implants.

Each claim file consisted of a detailed narrative of the case, which included relevant medical and surgical histories, a summary of the patient’s clinical course, and expert witness testimony. Each claim also contained discrete coded fields, entered by Controlled Risk Insurance Company taxonomy specialists according to the Controlled Risk Insurance Company proprietary taxonomy system. The accurate compilation of these claims is overseen by a Controlled Risk Insurance Company governance committee of physicians, lawyers, and data analysts. Coded fields included patient age, clinical location of event, pain diagnosis, case type, alleged injury and outcome, total financial legal cost, total settlement cost, and the National Association of Insurance Commissioners outcome severity score.8 A score of 6–9 is defined as “high severity,” representing death or permanent major injury; a score of 3–5 is “medium severity,” representing a temporary major or permanent minor injury; and a score of 0–2 is “low severity,” representing a temporary minor injury or no injury. In other claim analyses performed by anesthesia professional societies, the authors were often able to review the original medical records and legal documents; in this analysis, the authors reviewed the aforementioned clinical and legal records compiled into a file by Controlled Risk Insurance Company.

Another coded field input by the Controlled Risk Insurance Company was the contributing factors thought to have caused the error or event. Technical skill factors included poor technique and inexperience with procedure. Clinical judgment factors included suboptimal patient selection, failure to diagnose a sign or symptom, or too narrow a diagnostic focus. Communication factors included inadequate informed consent, poor communication with patient, and poor communication with other providers. Behavioral factors included patient factors such as noncompliance with treatment regimen or provider factors such as sexual misconduct. Documentation factors included insufficient documentation of clinical findings, clinical rationale, or undocumented phone advice to patient. These factors were applied by Controlled Risk Insurance Company taxonomy specialists based on the claim file, and were then made available to the investigators. A given claim file could have >1 contributing factor.

Back to Top | Article Outline

RESULTS

A demographic and financial overview of the claims is in Table 1.

Table 1.

Table 1.

A total of 126 closed claims were identified, meeting our search criteria for outpatient interventional pain management procedures. The average patient age was 51.9 years, with a range of 13–79 years, and 38.2% of patients were male. The National Association of Insurance Commissioners outcome severity was high (6–9) for 23.8% of claims, medium (3–5) for 71.4% of claims, and low (0–2) for 32.5% of claims. A total of 41 claims (32.5%) resulted in payments to the plaintiffs, with a median payment of $175,000 (range, $2600–$2,950,000) and 95% CI of $144,277–$451,240. When accounting for the total legal costs of all 126 claims (ie, legal costs plus, when applicable, the payment to plaintiffs), the median amount was $17,202, with a 95% CI of $80,625–$201,155. The median cost for cases in which no payment was made was $3411.

The most common pain syndrome associated with these claims was spondylosis (n = 93, 73.8% of claims), followed by myofascial pain syndrome (n = 14, 11.1% of claims) and failed back surgery syndrome (n = 8, 6.34% of claims). The remaining claims included pelvic pain (n = 2), compression fracture (n = 2), complex regional pain syndrome (n = 1), migraine (n = 1), knee osteoarthritis (n = 1), carpal tunnel syndrome (n = 1), painful scar (n = 1), and Eagle syndrome (glossopharyngeal neuralgia, n = 1).

The pain procedures involved in the claims are depicted in Table 2. Interlaminar–lumbar epidural steroid injections represented the largest share of the cases, with 34 (27%) claims involving interlaminar–lumbar epidural steroid injection and 31 (26%) claims involving interlaminar–cervical epidural steroid injection. The second most common claims were trigger point injections, representing 13 (10.3%) claims, and transforaminal–lumbar epidural steroid injection, representing 9 (7.1%) claims. The rest of the procedures consisted of medial branch block, medial branch radiofrequency ablation, sympathetic blocks, scar injection, and others, as detailed in Table 2.

Table 2.

Table 2.

The alleged injuring events are depicted in Table 3. The majority of claims were recorded as occurring because of “improper performance of procedure, not otherwise specified” (n = 38). The majority of these claims were related to procedures that occurred without incident, but after which the patient reported worsening or no improvement in pain (n = 20). Other claims in this category included the development a postprocedural epidural hematoma in patients not on anticoagulants undergoing uneventful cervical epidural steroid injection (n = 5). Another case in this category involved the injection of 10 mL of 98% alcohol into a painful mastectomy scar—the result of which was necrotic breast tissue requiring multiple surgical interventions and worsening pain.

Table 3.

Table 3.

The category “nonsterile technique” (n = 17, 13.5%) was the allegation for most cases involving infections, even if there was no clear evidence of nonsterile technique. Cases in this category included the development of an epidural abscess after an interlaminar–lumbar epidural steroid injection in a patient with poorly controlled diabetes; in another instance, an epidural abscess developed after a cervical epidural steroid injection in a patient with morbid obesity and chronic obstructive pulmonary disease on oral steroids. In 1 case, a joint infection occurred after major joint viscosupplementation injection. In 2 claims, patients developed hepatitis C virus after the nurse providing sedation was later discovered to have used the same needle to draw medications for multiple patients.

The alleged medical outcomes are depicted in Table 4. The most common alleged adverse outcome in this series was worsening pain or no improvement in pain after the procedure (n = 26, 20.6% of claims). Spinal cord infarct was the next most common (n = 16, 12.7%), followed by epidural hematoma (n = 9, 7.2%), soft-tissue infection (n = 9, 7.2%), postdural puncture headache (n = 9, 7.2%), and pneumothorax (n = 9, 7.2%). Death occurred in only a minority of claims (n = 4, 3.2%). In 1 claim, a patient underwent an uneventful lumbar radiofrequency ablation, and then 2 days later, presented in methicillin-resistant Staphyloccoccus aureus septic shock with meningitis, ultimately determined to be due to IV drug use. In another case, a few days after an uneventful interlaminar–lumbar epidural steroid injection, the patient developed weakness determined to be to a new thoracic disk herniation, with the patient dying from complications related to subsequent thoracic decompression and with an allegation that the interlaminar–lumbar epidural steroid injection caused the herniation. In a third case, a patient underwent a transforaminal–epidural steroid injection for chronic spine pain with recent imaging showing disk herniations; a few months later, she was diagnosed with metastatic cancer and alleged failure to diagnose. These cases did not settle. In a fourth case, a cervical epidural steroid injection resulted in an inadvertent intrathecal injection, resulting in a seizure, hospitalization, and subsequent death from pulmonary embolism.

Table 4.

Table 4.

The category “other outcome” represented outcomes that did not fit into any other category, such as the need to have surgical procedures (eg, nephrectomy) after the pain practitioners ordering imaging tests did not follow up on abnormal findings that were ultimately determined to be malignancies. Other outcomes in this category included idiopathic tinnitus after an uneventful cervical epidural steroid injection, globe puncture after supraorbital nerve block, and automatic implantable cardioverter-defibrillator activation during cervical radiofrequency ablation.

Table 5.

Table 5.

The Controlled Risk Insurance Company contributing factors are in Table 5. In this proprietary analysis performed by the Controlled Risk Insurance Company, issues with technical skills were the most common factor contributing to injury. Technical skill factors included possible technical problems (eg, when a known complication occurred after a procedure, such as a dural puncture from a lumbar epidural steroid injection). Also, this category included incorrect body site, inappropriate medication dose or route, or a retained foreign body such as a broken needle tip. Problems with clinical judgment were the next most common factor contributing to injury. This category included but was not limited to failure to diagnose a complication or to have too narrow a differential diagnosis. Contributing factors to communication-related issues included concerns with communication between providers and the patient, such as lack of informed consent, and poor communication among providers. Patient behavior factors included patient noncompliance. Documentation factors included insufficient informed consent and other aspects of the medical record. Clinical system factors included failure to identify care coordinating provider or failure to relay findings to the patient (as with no or wrong report being sent to patient).

Back to Top | Article Outline

DISCUSSION

Specific Procedures: Epidural Steroid Injections

Epidural steroid injections in this series were the procedures most commonly associated with malpractice claims: caudal, lumbosacral, thoracic, and cervical epidural steroid injections represented 80 cases, or 65% of all claims. This predominance of epidural steroid injections among malpractice claims may largely reflect the frequency with which epidural steroid injections are performed by pain practitioners. Studies of Medicare beneficiaries have shown, for instance, an annual increase of 96% for epidural procedures from 2000 to 2014.9 Our present analysis showed that the majority of malpractice claims were for epidural steroid injections performed via the interlaminar route, with interlaminar–lumbar epidural steroid injections representing 34 of our claims and interlaminar–cervical epidural steroid injections representing 31 of our claims. Previous studies of Medicare beneficiaries have reported that interlaminar procedures are more commonly performed in the lumbar spine; in 2011, for instance, 914,324 interlaminar lumbar epidural steroid injections and 200,134 interlaminar cervical epidural steroid injections were performed.10 There is a similar rate of representation of interlaminar lumbar epidural steroid injections and interlaminar cervical epidural steroid injections among our malpractice claim studies despite this 5-fold greater frequency with which lumbar injections are performed. This finding supports the notion that cervical injections pose higher risk in the insured population we studied. Previous closed claims studies have made similar inferences. In one analysis looking at the 3 most common procedure claims in the ASA Closed Claims Database, cervical injections represented 44% of claims and lumbar injections represented 29% of claims. In contrast, the National Anesthesia Clinical Outcomes Registry found that cervical injections represented 23% of cases, whereas lumbar injections represented 61% of cases performed.11 While malpractice claims might be filed at a slightly higher rate for interlaminar–cervical epidural steroid injections compared to interlaminar–lumbar epidural steroid injections, our study does not show a significant difference in the severity of these claims. The National Association of Insurance Commissioners severity score for cervical epidural steroid injections in the present study was an average of 4.7 and a median of 4, compared to the interlaminar–lumbar epidural steroid injection average of 4.6 and median of 4.5. These scores correspond to temporary major or permanent minor injuries and could reflect the heterogeneity of injurious events cited in the claims. Cervical epidural steroid injections in the present study involved severe outcomes, such as 6 cases of epidural hematomas (versus 2 cases for interlaminar–lumbar epidural steroid injections), 4 cases of spinal cord infarcts (versus 1 case for interlaminar–lumbar epidural steroid injections), and 2 cases of epidural abscesses (versus 1 case for interlaminar–lumbar epidural steroid injections). However, cervical epidural steroid injections did also involve 6 cases of worsening or no improvement in pain (versus 7 cases for interlaminar–lumbar epidural steroid injections) and 3 cases of postdural puncture headache (versus 4 cases for interlaminar–lumbar epidural steroid injections). Cervical epidural steroid injections did also represent 5 cases of “other outcomes,” which generally did not result in sustained injury, such as the development of temporary idiopathic tinnitus or a temporary high spinal block from inadvertent intrathecal injection. Previous claim studies have concluded that permanent, disabling spinal cord injuries were dramatically more common among claims associated with procedures carried at the level of the cervical spine.12 However, based on the small size of this Controlled Risk Insurance Company study, no robust conclusions can be made in cervical epidural steroid injections compared to interlaminar–lumbar epidural steroid injections.

Transforaminal–epidural steroid injections, in particular when performed at cervical spinal levels, are associated with rare but serious neurological injury. Previous studies have suggested that brain and spinal cord infarction after such injections may result from the intra-arterial injection and embolism of particulate steroids.13–15 Medicare data suggest that transforaminal–lumbar epidural steroid injections are performed more frequently than transforaminal–cervical epidural steroid injections; in 2011, totals for transforaminal–lumbar epidural steroid injections and transforaminal–cervical epidural steroid injections were 1,109,157 and 65,598, respectively.10 Our study contained one case of a transforaminal–cervical epidural steroid injection, with the outcome being a postdural puncture headache, and 9 cases of transforaminal–lumbar epidural steroid injections. In the latter cases, there were 6 instances of spinal cord infarction, with 3 of those cases involving use of particulate steroid and the others involving deep sedation and incompletely documented image guidance. No strong conclusions can be drawn from this small sample, although it does highlight that serious neurological injury can occur with the use of particulate steroids in transforaminal injections, even when limited to lumbar spinal levels. In our contemporary claims study, the dearth of transforaminal–cervical epidural steroid injection cases involving particulate steroid suggests that clinical practice may well have changed. Perhaps, these injections are performed less frequently than transforaminal–lumbar epidural steroid injections. Another plausible explanation is that the safety of these injections has improved. Expert and society guidelines are increasingly direct about using nonparticulate steroids in transforaminal–cervical epidural steroid injections and at least the initial performance of a transforaminal–lumbar epidural steroid injection. Other safety measures have been suggested, including disciplined use of image guidance during needle placement, the use of imaging in multiple planes to accurately demonstrate the position of final needle placement, and avoidance of deep sedation allowing patients to remain responsive while the procedure is being performed.16–19

Back to Top | Article Outline

Specific Procedures: Trigger Point Injections

Trigger point injections are common, office-based procedures performed for the management of myofascial pain. These procedures are generally considered safe but have rarely been associated with complications such as pneumothorax, soft-tissue infection, and even high spinal due to unintentional intrathecal injection of local anesthetic.20–22 Of the 13 cases in this study related to trigger point injections, the injury was pneumothorax in 7 instances, with the severity ranging from an incidentally discovered pneumothorax managed conservatively to a hemodynamically significant pneumothorax requiring a thoracotomy. Two cases involved scar formation thought to be related to the intramuscular injection of particulate steroid. Two cases involved infections, with one instance of an inadvertent puncture of a breast implant, resulting in an abscess and requiring a surgical intervention. In another case, a tendon rupture occurred, again requiring surgery. In a final case, a needle broke off during the procedure, with the attending physician not being notified until after the patient had left the clinic; in this instance, the patient required surgical removal of the needle. These 3 cases involved indirect supervision, with procedures performed by a medical assistant, a nurse practitioner, and a physician in fellowship training. In none of these instances was ultrasound used for needle placement, although this technique has been described in the literature.23

Previous closed claim studies have identified pneumothorax as the injury in almost all claims; the ASA Closed Claims Project study of 17 cases involving trigger point injections identified pneumothorax as the injury in 15 of those cases.24 Our present study highlights a more heterogenous array of potential complications. When planning to perform trigger point injections around the thorax, it would be reasonable to specifically discuss as part of the consenting process that pneumothorax is a potential complication. In addition, for higher-risk patients at extremes of weight or with nearby implants, it would not be unreasonable for practitioners to consider the use of ultrasound when performing trigger point injections. Practitioners should continue to use their clinical judgment regarding when direct clinical supervision of the performance of trigger point injections is needed.

Back to Top | Article Outline

Outcomes Overview

The majority (71.4%) of cases in this study involved temporary major or permanent minor injury, with only 4 cases (3% of all cases) involving a patient death. Previous closed claims studies include the ASA Closed Claims study on cervical spine procedures for pain management, which, while different in methodology and content from the present study, found that death represented 1 of 64 (3%) claims.12 Death has been shown to be more common in claims involving medication management. Our previous study using the Controlled Risk Insurance Company methodology to investigate medication management in pain medicine found 18 of 37 cases (48.6%) involved a patient death;25 a previous ASA Closed Claims study on medication management in pain medicine found that 29 of 51 cases (56.8%) involved a patient death.26

More common in procedure claims is a permanent injury, as might result from the injuries caused by complications such as epidural hematoma, neuraxial infection, or central nervous system infarction. Among the claims involving a high-severity injury outcome were 6 cases of alleged anticoagulation mismanagement. Two cases involved a pain physician instructing patients to hold their anticoagulation without conferring with their prescribing physicians, with cardiovascular accidents occurring in both cases. Three cases involved not appropriately holding anticoagulation before a procedure due to either not knowing the patient was on anticoagulation, or not holding the medication for a sufficient length of time, with epidural hematoma resulting. One case involved holding anticoagulation before a neuraxial procedure, with the result being worsening ischemia related to peripheral arterial disease. In none of these cases was it mentioned that the pain physicians had referenced anticoagulation guidelines, discussed management with primary providers, or documented risks of altering the anticoagulation. To mitigate provider risk and patient injury, it would be reasonable when considering altering anticoagulation in patients to reference consensus guidelines,27 coordinate with primary care providers, and document the risks of holding or continuing anticoagulation.

The preponderance of claims in this study associated with “medium-severity” injuries stemmed from claims of worsening pain after the procedure (n = 26, 20.6% of claims). In 19 of these cases, the procedures occurred without incident; in 2 of these cases, the patient experienced a temporary paresthesia; in 4 of these cases, the wrong procedure was performed or was performed on the wrong side; and in 2 of these cases, the patient had a fall in the recovery room. Only 5 of these 26 cases resulted in a payment to the patient (19.2% of all claims). In none of these claims was it mentioned in the case narrative that the consent documented “worsening or no improvement in pain” as a possible consequence of the procedure. The majority of these cases involved the claim that the patient did not know the procedure could actually result in worsening pain afterward. Including in consent forms and the provider–patient discussion the possibility of worsening pain or no improvement in pain after the procedure is a reasonable practice and may have deflected some of the claims in this study.

Other medium-severity and low-severity injuries involved the alleged systemic side effects of steroids. Two cases cited iatrogenic Addison disease, 2 cases cited deep venous thrombosis, 1 case cited acute coronary syndrome related to hypertension, 1 case cited end-stage renal disease related to hypertension, and 1 case cited soft-tissue necrosis related to worsening peripheral vascular disease. The primary claim in these cases was that the steroid administered during the pain procedures resulted in the aforementioned complications. In none of these cases was a payment made to the claimant. To promote patient safety and engagement in health care decisions and minimize the risk of such litigation, it is reasonable for pain practitioners to discuss and document the potential risks of repeated steroid injections. Furthermore, pain practitioners would likely benefit from knowing and, when necessary, citing professional society and other relevant practice guidelines on reasonable dose and frequency of steroid used.28

Back to Top | Article Outline

Controlled Risk Insurance Company Contributing Factors

Contributing factors related to the patient injury, as reported by the Controlled Risk Insurance Company, suggest that many of these claims are multifactorial. These Controlled Risk Insurance Company–coded contributing factors have been published in previous studies: technical skill was found to be the most common factor in cases involving implants for pain management; patient behavior (ie, noncompliance) was found to be the most common factor in cases involving medication management.25,29 Given that the focus of the current study is on procedures, it is not surprising that technical skill, as defined by the Controlled Risk Insurance Company Comparative Benchmarking System, is thought to be the most common factor. Inadequate recording of fluoroscopic images, particularly failing to save all of the key images, was cited on multiple occasions and would fit into the technical skill category. Clinical judgment would include proper patient selection and expedient diagnosis and management of complications. Communication was the third most common contributing factor, followed by insufficient documentation.

Back to Top | Article Outline

Limitations

Closed claim limitations have been discussed previously.30 These studies do not include all complications that occur and skew toward more severe complications, and these may be more likely to result in a financial payment. Just as these studies do not provide information on a numerator (the number of complications), they do not provide information on a denominator (the number of procedures performed), and therefore, we cannot comment on incidence. Closed claim data are retrospective and without a control, and as such, cannot determine a cause and effect relationship.

There are also limitations specific to the use of this database, created and maintained by the Controlled Risk Insurance Company. The data are entered and coded in a proprietary manner by risk management–trained taxonomy nurses of the Controlled Risk Insurance Company. The claim files were also initially reviewed by trained nurses and analysts at the Controlled Risk Insurance Company, rather than pain physicians, and as such, the study cannot have independent reliability assessments by other non–Controlled Risk Insurance Company entities. The coded fields and the narrative summaries reviewed by the study authors (C.R.A., R.S., and R.D.U.) at times did not contain sufficient information to make any strong conclusions. This study also lacks independent reliability assessments by board-certified pain management practitioners who were practicing at the time. Another limitation is that the primary medical specialty of the physician involved in the claim was not known.

Back to Top | Article Outline

CONCLUSIONS

Our study is a contemporary analysis of closed claims in interventional pain management. Epidural steroid injections are among the most commonly performed interventional pain procedures. While a familiar procedure to pain management practitioners, this common procedure can result in significant neurological injury. Transforaminal–epidural steroid injection, especially when particulate steroid is used, can result in neural injury. Trigger point injections, while generally considered safe, can lead to pneumothorax or injury to other deep structures.

That being said, it should again be emphasized that closed claim studies such as the present work skew toward more severe complications. Furthermore, while some of the injuries prompting claims are severe and occur out of proportion to the rate at which the procedures are performed, the actual incidence of these complications is likely overall low. In other words, as discussed previously, cervical spine procedural claims such as cervical epidural steroid injections have been shown to involve more severe injuries and occur out of proportion to the rate at which they are performed compared with lumbar spine procedural claims such as from lumbar epidural steroid injections. Viewed from another perspective, the present database, which represents approximately 30% of claims in the United States, contained only 31 interlaminal–cervical epidural steroid injection closed claims from a 5-year period, with only 10 of those claims classified as having a high-severity injury. In contrast, as previously cited, >200,000 interlaminal–cervical epidural steroid injections were charged to Medicare in 2011 alone. Those incomplete data from which no definitive incidence calculations can be performed highlight the need for a comprehensive, pan-payer database to track the number of procedures performed and complications occurring over time. Those data may also suggest that the overall incidence of severe complications for interlaminal–cervical epidural steroid injection is low. The alternatives in pain management, such as opioid and nonopioid pharmacotherapy or spine surgery, are also not without risk of severe complications. However, closed claim studies do allow physicians to have a more nuanced discussion of potential risks.

Indeed, the causes of injury are usually multifactorial, involving aspects of technical execution, clinical judgment, communication with the patient and other medical providers, and documentation. From the review of the available commentary from prosecutors, defense attorneys, and expert witnesses, clinical documentation appears to be of particular importance. On the signed consent form, specifically listing and discussing the most common and most serious risks for a particular procedure is likely a high-yield intervention (eg, for epidural steroid injection mentioning headache, epidural hematoma or abscess, and spinal cord injury); for trigger point injections in the thorax or in patients at extremes of weight, mentioning pneumothorax; and for all procedures, mentioning no improvement in pain or a worsening of pain. Another advisable practice is to document that the practitioner prescribing the patient’s anticoagulants was involved in the decision to pause (or continue) that patient’s anticoagulants for the procedure at hand. Direct reference to institutional or society guidelines on anticoagulant management will likely be useful in particularly high-risk scenarios. Key fluoroscopy images taken during each procedure should be saved as part of the patient record (eg, anteroposterior and lateral images demonstrating the final needle position at the time of injection).

Efforts to minimize patient harm will require more than a single approach. The best outcomes for patients and the lowest liability for practitioners will likely come from in-depth training in procedural skills, familiarity with and attention to published best practice recommendations, thoughtful patient selection, and honest and reasonably detailed discussions that ensure truly informed consent and careful documentation.

Back to Top | Article Outline

DISCLOSURES

Name: Christopher R. Abrecht, MD.

Contribution: This author helped gather and analyze the data, and generate and revise the manuscript.

Name: Ramsey Saba, MD.

Contribution: This author helped gather and analyze the data.

Name: Penny Greenberg, RN, MS.

Contribution: This author helped analyze the data and revise the manuscript.

Name: James P. Rathmell, MD.

Contribution: This author helped revise the manuscript.

Name: Richard D. Urman, MD, MBA.

Contribution: This author helped revise the manuscript.

This manuscript was handled by: Honorio T. Benzon, MD.

Back to Top | Article Outline

REFERENCES

1. Dowell D, Haegerich TM, Chou R. CDC guideline for prescribing opioids for chronic pain–United States, 2016. JAMA. 2016;315:1624–1645.
2. Manchikanti L, Malla Y, Wargo BW, Cash KA, Pampati V, Fellows B. Complications of fluoroscopically directed facet joint nerve blocks: a prospective evaluation of 7,500 episodes with 43,000 nerve blocks. Pain Physician. 2012;15:E143–E150.
3. Manchikanti L, Hirsch JA. Neurological complications associated with epidural steroid injections. Curr Pain Headache Rep. 2015;19:482.
4. Kozlov N, Benzon HT, Malik K. Epidural steroid injections: update on efficacy, safety, and newer medications for injection. Minerva Anestesiol. 2015;81:901–909.
5. McGrath JM, Schaefer MP, Malkamaki DM. Incidence and characteristics of complications from epidural steroid injections. Pain Med. 2011;12:726–731.
6. Jena AB, Seabury S, Lakdawalla D, Chandra A. Malpractice risk according to physician specialty. N Engl J Med. 2011;365:629–636.
7. Cheney FW, Posner K, Caplan RA, Ward RJ. Standard of care and anesthesia liability. JAMA. 1989;261:1599–1603.
8. Sowka M. Malpractice Claims: Final Compilation. 1980.Brookfield, WI: National Association of Insurance Commissioners.
9. Manchikanti L, Pampati V, Hirsch JA. Utilization of interventional techniques in managing chronic pain in Medicare population from 2000 to 2014: an analysis of patterns of utilization. Pain Physician. 2016;19:E531–E546.
10. Manchikanti L, Pampati V, Falco FJ, Hirsch JA. Assessment of the growth of epidural injections in the medicare population from 2000 to 2011. Pain Physician. 2013;16:E349–E364.
11. Pollak KA, Stephens LS, Posner KL, et al. Trends in pain medicine liability. Anesthesiology. 2015;123:1133–1141.
12. Rathmell JP, Michna E, Fitzgibbon DR, Stephens LS, Posner KL, Domino KB. Injury and liability associated with cervical procedures for chronic pain. Anesthesiology. 2011;114:918–926.
13. Malhotra G, Abbasi A, Rhee M. Complications of transforaminal cervical epidural steroid injections. Spine (Phila Pa 1976). 2009;34:731–739.
14. Benny B, Azari P, Briones D. Complications of cervical transforaminal epidural steroid injections. Am J Phys Med Rehabil. 2010;89:601–607.
15. Baker R, Dreyfuss P, Mercer S, Bogduk N. Cervical transforaminal injection of corticosteroids into a radicular artery: a possible mechanism for spinal cord injury. Pain. 2003;103:211–215.
16. Benzon HT, Huntoon MA, Rathmell JP. Improving the safety of epidural steroid injections. JAMA. 2015;313:1713–1714.
17. Manchikanti L, Falco FJE, Benyamin RM, Gharibo CG, Candido KD, Hirsch JA. Epidural steroid injection safety recommendations by the Multi-Society Pain Workgroup (MPW): more regulations without evidence or clarification. Pain Physician. 2014;17: 575–588.
18. Rathmell JP, Benzon HT, Dreyfuss P, et al. Safeguards to prevent neurologic complications after epidural steroid injections: consensus opinions from a multidisciplinary working group and national organizations. Anesthesiology. 2015;122:974–984.
19. Neal JM, Barrington MJ, Brull R, et al. The second ASRA practice advisory on neurologic complications associated with regional anesthesia and pain medicine: executive summary 2015. Reg Anesth Pain Med. 2015;40:401–430.
20. Shafer N. Pneumothorax following “trigger point” injection. JAMA. 1970;213:1193.
21. Elias M. Cervical epidural abscess following trigger point injection. J Pain Symptom Manage. 1994;9:71–72.
22. Nelson LS, Hoffman RS. Intrathecal injection: unusual complication of trigger-point injection therapy. Ann Emerg Med. 1998;32:506–508.
23. Botwin KP, Sharma K, Saliba R, Patel BC. Ultrasound-guided trigger point injections in the cervicothoracic musculature: a new and unreported technique. Pain Physician. 2008;11:885–889.
24. Fitzgibbon DR, Posner KL, Domino KB, Caplan RA, Lee LA, Cheney FW; American Society of Anesthesiologists. Chronic pain management: American Society of Anesthesiologists closed claims project. Anesthesiology. 2004;100:98–105.
25. Abrecht CR, Brovman EY, Greenberg P, Song E, Rathmell JP, Urman RD. A contemporary medicolegal analysis of outpatient medication management in chronic pain. Anesth Analg. 2017;125:1761–1768.
26. Ritzgibbon Dr, Michna E, Rathmell JP, Stephens LS, Domino KB. Medication management in chronic pain malpractice claims. Anesthesiology. 2010;112:948–956.
27. Narouze S, Benzon HT, Provenzano D, et al. Interventional spine and pain procedures in patients on antiplatelet and anticoagulant medications (second edition). Reg Anesth Pain Med. 2018;43:225–262.
28. Manchikanti L, Falco FJ, Singh V, et al. An update of comprehensive evidence-based guidelines for interventional techniques in chronic spinal pain. Part I: introduction and general considerations. Pain Physician. 2013;16:S1–S48.
29. Abrecht CR, Greenberg P, Song E, Urman RD, Rathmell JP. A contemporary medicolegal analysis of implanted devices for chronic pain management. Anesth Analg. 2017;124:1304–1310.
30. Cheney FW. The American Society of Anesthesiologists closed claims project: what have we learned, how has it affected practice, and how will it affect practice in the future? Anesthesiology. 1999;91:552–556.
Copyright © 2019 International Anesthesia Research Society