Closed Claims Case Review Case 2023-11: Unanticipated Outcome after Peripheral Nerve Block: Role for Communication and Consent

A 63-year-old male, a left-handed woodworking craftsman with diabetes and a BMI of 37, presented for an outpatient surgery arthroscopic repair of a left shoulder labral tear sustained after a fall that did not respond to conservative treatment. The patient and his daughter stated that he was “bullied” into a nerve block. There was no written consent for the regional block in the medical record and no documentation of a verbal discussion with consent. The patient was sedated with midazolam (total 4 mg) and fentanyl 100 mcg for the block. A left-sided/“supraclavicular-interscalene” block was performed using ultrasound guidance with a 21-gauge B-echo needle without pain or paresthesia. 50 ml of bupivacaine 0.375% with epinephrine 1:300,000 was used for the block with the doses given incrementally. The patient then received an unremarkable general anesthetic for surgical repair. The OR time was 50 minutes.

The patient recalls burning pain in the shoulder in the recovery room. Nursing notes in the medical record do not document this pain. The patient denied pain on a postoperative phone call the following day. However, the patient stated that when the block wore off, he began to experience severe left arm pain and weakness.

He saw the surgeon four days postoperatively, who said the injury was secondary to the block. No EMG or conduction studies were ordered. The patient saw a neurologist a month later, and an EMG revealed loss of deltoid and biceps function. Six months later, the patient’s left arm weakness was improved but still present. He was unable to work. He continued to have chronic neuropathic pain requiring treatment with gabapentin, duloxetine, and meloxicam.

Expert review suggested that a direct nerve injury by the block needle could not be ruled out. In addition, the amount of bupivacaine could have been excessive and caused compression and ischemia. Importantly, the informed consent process was ambiguous, and there was a lack of informed consent for possible nerve injury, including pain, weakness, and sensory changes.

The case settled before trial.

Peripheral nerves are fortunately relatively resilient due to their fibrous epineurium. The incidence of long-term postsurgical peripheral nerve injury has been reported to be 0.02%-0.05%, or about two to five in 10,000 cases. Interestingly, this incidence has not changed with the use of ultrasound guidance (Anesthesiology 2018;28:11-26; Reg Anesth Pain Med 2015;40:401-30; Curr Opin Anaesthesiol 2018;31:614-21). Recognition and presentation of injuries are often delayed – occurring after hospital discharge – making these cases likely underreported, particularly in the outpatient surgical environment.

The ASA Practice Advisory for the Prevention of Perioperative Peripheral Neuropathies 2018 (Anesthesiology 2018;28:11-26) includes a list of risk factors that make patients more vulnerable to perioperative neuropathies. Male sex, age, high BMI, and diabetes are some of the risk factors in this case. General population-based reported incidences of peripheral neuropathies have been described at over 6% for those over age 60. We do not know if the patient had any preexisting neuropathies or preexisting neurologic disease. These may increase vulnerability to peripheral neuropathies, described as “the double-crush theory.” The 2015 ASRA Practice Advisory on Neurologic Complications Associated with Regional Anesthesia and Pain Medicine also describes these risk factors, along with an algorithm for evaluation and treatment (Figure) (Reg Anesth Pain Med 2015;40:401-30; Reg Anesth Pain Med 2015;40:491-501).

Most postoperative peripheral nerve injuries impact sensory over motor function and are transient, improving over time. Recovery depends in part on the extent of injury – from neuropraxia to axonotmesis to neurotmesis (Reg Anesth Pain Med 2015;40:401-30; Mayo Clin Proc 2020;95:355-69). In neuropraxia, there is impaired conduction and focal demyelination, but the axon remains intact. These cases have an excellent prognosis. In axonotmesis, there is a break in the axon leading to distal nerve Wallerian degeneration. These cases have variable and slower improvement. In neurotmesis, there is complete disruption of epineurium and connective tissue, and surgical reconnection is necessary.

Importantly, during the management of postsurgical neuropathy, the clinical experts surrounding the patient should remain a team. Unfortunately, in this case, the surgeon told the patient that his symptoms were from the block. We do not know if the surgeon and anesthesiologist were in communication with each other. It could be argued that it was actually unclear if the injury was due to trauma, stretch, compression, positioning, or a combination of factors.

Another important member of the team is the neurologist. Neurological consultation and assessment should be actively considered, particularly when there is severe motor involvement, severe pain, progressive worsening of symptoms, or multifocal findings that are difficult to localize. Direct communication with a neurologist can be helpful. Neurologists can be unfamiliar with the OR set-up and patient positioning during surgery and are likely unfamiliar with anesthesia record-keeping. During evaluation and diagnosis, the team should keep in mind that immune-mediated inflammatory neuropathy can also occur postoperatively (Reg Anesth Pain Med 2015;40:401-30). This can be a missed diagnosis and should be considered especially when the clinical findings don’t match a specific expected nerve distribution or timeline. A nerve biopsy will demonstrate microvasculitis, and there are potential immunotherapy treatment options. Therefore, this missed diagnosis may result in missed treatment.

Early referral to a physiatrist is also strongly advocated to speed and improve functional recovery. The physiatrist can be an important point person of contact within the collaborative clinical team. Ideally, the anesthesiologist should continue to communicate with the patient. Active disclosure after adverse events is a best practice that has repeatedly proven to be beneficial for the patient, the clinician, and for mitigating adversarial legal action.

It cannot be overemphasized that informed consent is a process with a conversation – not a single event, signature, checklist, or form. The components of an informed consent include voluntariness (i.e., without coercion), decision-specific capacity, disclosure (of risks, benefits, and alternatives), an understanding of the risks, and documentation. If full disclosure of risks, benefits, and alternatives, including the option of no treatment, is challenged, a negligence claim can be made (Anaesthesia 2021;76:18-26; Anaesthesia 2017;72:93-105; Can J Anaesth 2014;61:832-42).

The Closed Claim Database has reported that 22% of cases had inadequate informed consent (asamonitor.pub/3Zk4EvV). Since that report, this database continues to review cases where poor or lack of informed consent compromises the legal case for the anesthesiologist. A clear cause of injury may not be linked to anesthesia care, and anesthesia care can even be deemed appropriate – yet the flawed informed consent can lead to litigation and closed claims payment or settlement.

Legally, morally, and ethically, a patient must be fully informed of all the risks, benefits, and alternatives of what a “reasonable patient” under similar circumstances would want to know to make an informed decision. In the current case, that would include recognizing that this woodworking craftsman would have specific self-interests to understand the probability risks of long-term neuropathy given his occupation and his underlying risk factors (e.g., diabetes).

The statement that the patient and daughter felt “bullied” might indicate coercion and/or biases during the discussion. Conscious and unconscious biases can include: a) recency effect, where items discussed last are remembered more clearly; b) nocebo effect, which emphasizes negative effects; and c) implying a perceived “best practice” to perform a block as a sole option, when a block may not be absolutely necessary.

In planning what and how much to disclose, it is important to note that patients want to know both the risk of common but less consequential complications AND the rare yet severe complications (Acta Anaesthesiol Scand 2013;57:342-9). Studies both in the United States and United Kingdom have found that the complications clinicians discuss with their patients are quite variable, especially the rare complications. In one study, none of the patients who had interscalene block for shoulder surgery were informed about the risks of local anesthetic systemic toxicity (LAST) (J R Soc Med 2015;108:451-6). In others, severe complications were infrequently discussed. When discussed, incidence rates were highly variable and inconsistent with the literature (Reg Anesth Pain Med 2007;32:7-11; Reg Anesth Pain Med 2008;33:395-403). Furthermore, the opinion that clinicians should limit discussion of complications for fear that it will provoke harmful anxiety in patients has been discredited. A study by Burkle et al. demonstrated that patients (in an almost 4:1 ratio) felt the benefit of knowing the risks outweighed the anxiety generated by discussion of the risks (asamonitor.pub/3PCcD4p).

Some ABCDEF “best practices” for informed consent discussions include:

• Awareness of Biases (conscious or unconscious).
• Consider enhancing communication by utilizing both written and verbal information. When studied, patients do not have great postoperative recall of the risks discussed preoperatively for peripheral nerve blocks. However, utilizing a combination of written, online, and verbal sources along with visual aids could improve information exchange and retention.
• Discuss both common/less consequential risks AND rare/highly consequential risks.
• Ensure the “reasonable patient” standard.
• Frequencies: Use frequencies instead of percentages to quantify risk. Numerical illiteracy makes statistical risk difficult to understand. The Royal College of Anaesthetists (see graphic above) recommends describing numerical risk in the form of frequencies. Some other numerical risks that can be useful are a) “death from accidental drowning” (1:1,000) and b) “hit by lightning” (1:15,000). In addition, as risks for peripheral nerve blocks may be challenging to quantify, it may be useful to characterize risk as recommended in the graphic using a broader scale of “very common” to “very rare.”

Finally, with regard to documentation, The Joint Commission mandates that elements of the informed consent discussion are documented somewhere in the medical record. They do not specifically require a separate anesthesia consent form, although many anesthesia practices have gone to a separate anesthesia, if not regional anesthesia, consent form. If a template is used, it is prudent to make specific custom notations germane to the patient at hand and avoid only boilerplate checkboxes. Be aware that there may also be state-specific legal requirements that are important to review for one’s own practice location.

Beyond the informed consent documentation, there is also documentation of the procedure to consider. There is currently no “standard” for documentation of peripheral nerve block procedures. However, recently, 77 international regional anesthesia experts utilized a modified Delphi process and published a consensus of suggested documentation components for both neuraxial and peripheral nerve blocks (Reg Anesth Pain Med 2022;47:301-8).

Utilization of peripheral nerve blocks will undoubtedly continue to increase in our anesthesia practice. It is important to stay up to date on the incidence of complications, on how to provide optimal informed consent preoperatively, to document both the informed consent and procedure, and to provide timely evaluation, treatment, and support should the rare postoperative neuropathy occur.