Blog with interesting cases and/or problems related to anesthesia with discussion based on best evidence in the literature.

May 10, 2017

35 year old female with history of alcohol abuse

The other day a 35 year old female presented for surgery for lipoma excision from the posterior neck.  She reported a history of severe alcohol abuse in the past for which she was taking naltrexone 100mg qd.  The patient also reported a severe allergy to all steroids.  When I asked for details, she reported that she had severe manic exacerbations with steroids and that this had happened with a one time dose given prior to surgery in the past.  The patient had also experienced at least two seizures in the past, one occurring after surgery.  She was told that this seizure could have been related to the anesthetic she had.  She also reported another seizure episode in the past related to severe depletion of vitamin B12.  She was taking several medications for anxiety, bipolar, alcoholism and PTSD.  These included:

  • Seroquel
  • Tegretol
  • Prozac
  • Prazosin
  • Klonipin
In addition, the patient was concerned because in the past, she had experienced very severe and  unpleasant emergence anxiety and agitation.

There were several issues that played a role in choosing an anesthetic for this patient. First she requested that she did not receive a steroid which I commonly give to female patients (especially non smokers) as they are at increased risk for PONV.  In addition, She was medicated with naltrexone for alcoholism.  Therefore, there was likely to be challenges with pain control in this patient.  Lastly, the patient had had previous seizures following anesthesia during emergence as reported.  She was on a number of medications that could potentially also activate her liver enzymes causing her to be a rapid metabolizer of anesthetics such as propofol, midazolam and fentanyl.

This patient described a past experience of severe agitation and anxiety upon emergence.  Emergence delirium (ED) is not uncommon in pediatrics. However, it has not been well studied in adults.  One review of the literature  estimated an incidence of 3% in adult patients. Another study found an incidence of 4.7% among adult patients [1].  However, in those at risk for PTSD an incidence of 20% is more likely [3].  Emergence delirium in adults can be a significant problem as they are capable of self injury.  A recent case report details a severe episode of emergence delirium in a patient with PTSD [4], and subsequent case where changes were made to help mitigate this issue.  In this patient with a self described episode of emergence delirium and history of PTSD and bipolar disorder, several considerations would be important.  One consideration is the recommendation to avoid midazolam in particular in this patient group.  Some literature suggests that Midazolam in particular can enhance the memory of events surrounding the trauma [5].  If a benzodiazepine is desired, lorazepam may be preferred.  While it has been traditional teaching that ketamine can excerbate PTSD, and should be avoided during anesthesia in these patients, recent literature suggests that ketamine may not be a problem [6], or may be helpful in the prevention of PTSD [7].  Currently, prazosin, an alpha 1 blocker is used in the treatment for PTSD, but also is effective in the treatment in reducing alcohol intake in alcoholics.  Therefore, it would be important to verify that this patient continued to take this medication on their usual schedule prior to anesthesia. Prazosin, as an alpha 1 blocker, can cause sudden drops in blood pressure, and this is particularly problematic when patients stand up suddenly.  In pediatrics, there is quite a large amount of literature looking at interventions that can prevent or decrease the chances of ED. Propofol decreased the incidence from 38% after sevoflurane to 0% in pediatrics.  In the pediatric literature, other studies have corroborated the benefit of propofol in reducing ED. In adults, propofol also was able to decrease the incidence of ED, from 45% to 10% [10]. Another study in the pediatric literature showed that 2 mck/kg of IV Clonidine after induction significantly reduced the incidence of ED. Like Clonidine, Dexmedetomidine may also be a useful (although expensive) adjuvant to help prevent ED.  Given that the patient had already experienced a significantly negative ED event in the past, had a post anesthetic seizure, and was also allergic to steroids, so that decadron was unavailable for PONV prophylaxis, I opted to utilize a propofol infusion for anesthesia maintenance.  Unfortunately, in a patient taking naltrexone for alcoholism, maintaining anesthesia with routine anesthetic dosages could be fraught with problems.  Therefore, I opted to utilize a BIS monitor as an aid in verifying the depth of anesthesia. The BIS system is a proprietary system using data from a processed EEG signal to produce a dimentionless number that correlates with the depth of anesthesia in a sense. It is presumed that a number between 40 and 60 represents general anesthesia and would thus have a low incidence of recall.  It is not clear if an alcoholic or patient who is taking naltrexone will respond in the same fashion to a typical anesthetic.  In this patient, 200 mg of propofol was given for induction along with 100 mcg of fentanyl and 2 mg of versed.  The BIS went from 87 to 34.  This change in BIS was as expected.  Shortly thereafter, with an infusion of propofol running at 200 mcg/kg/min and the BIS reading 34, an injection of local anesthetic by the surgeon was performed at the site of the lipoma.  The patient immediately started moving requiring an additional bolus of propofol and another 100 mcg of fentanyl.  This would be an atypical response at a BIS of 34.  However, given that propofol does not inhibit spinal reflexes like the potent inhalational agents, this could be considered within normal limits.  Given the patient movement at a BIS of 34, the BIS was maintained in the high 30's for the remainder of the very short case.  A propofol infusion of 200 mcg/kg/min was sufficient.  after 30 min the case was over and the BIS reading was 36.  The propofol infusion was discontinued and in 4 minutes, the BIS had jumped from 36 to 74 and the patient began responding to commands and was extubated.   This rapid emergence could likely be explained by rapid hepatic metabolism of the propofol as well as residual naltrexone binding of mu opioid receptors decreasing the efficacy of fentanyl.  The patient was transported to the PACU with no evidence of emergence delirium and seemingly in good spirits with good pain control.  However, 30 minutes later the PACU nurse called me to report that she had given an additional 200 mcg of fentanyl in PACU for pain control and the patient was reporting 10/10 pain.  This was unusual in that the small lipoma excision had been anesthetized with local anesthetic.  Hydromorphone was ordered and no more calls were forthcoming.

Currently there is an epidemic occurring in America with opioid abuse.  It is becoming ever more likely that we will be required to care for one of these patients. Many of these patients will require therapy such as suboxone to prevent cravings of opioid.  Suboxone is a combination of buprenorphine and naltrexone where  the ratio is 4 parts buprenorphine to 1 part naltrexone.  The naltrexone is included to prevent the crushing and snorting or intravenous injection and is not considered to be active when suboxone is taken as intended. Buprenorphine is a partial agonist at the mu receptor.  It is also a kappa antagonist. These receptors are predominantly located in the spinal cord.  As a partial agonist, buprenorphine has a ceiling affect which occurs at about 32 mg / day.  Most patients who are on buprenorphine (suboxone) for maintenance therapy, are on doses that are higher than for treatment of chronic pain (i.e 16 mg for maintenance vs 2 to 4 mg for chronic pain therapy). There are two issues that anesthesiologists need to understand when patients arrive for surgery on suboxone.  First is the underlying pharmacokinetics.  Suboxone, if provided as a patch takes a full seven days to clear.  Buprenorphine is also prescribed to be taken sublingual where the bioavailability is  30 to 50%. The half life of high dose buprenorphine (as used for maintenance therapy in opioid dependent patients)  is 20 to 70 hours. However, the half-life is highly variable and depends on patient characteristics as well as the dosing regimen. It is clear from case reports that maintaining suboxone in the preoperative period will lead to very poor pain control amid escalating pure mu opioid agonist requirements. For elective surgery, where significant post op pain is anticipated, it is recommended that patients discontinue suboxone therapy for five days.  In urgent/emergent surgical patients presenting on suboxone maintenance therapy, it may be very difficult to override the suboxone from the mu receptor. The reason for this difficulty is a result of the very high affinity for the mu opioid receptor by buprenorphine.  It has been noted that buprenorphine binds the mu receptor avidly, and only slowly diffuses away.  Therefore, the mu receptor remains unavailable to pure mu agonists. Consequently,  respiratory depression is a risk at the very high dosages required to achieve pain control.  It may be wise to titrate a rapid onset opioid (fentanyl) prior to surgery in the operating room to help anticipate what dose the patient may tolerate or require for post op pain control.  Suboxone should not be given during the preoperative period to avoid precipitating additional pain.  For patients where post op pain is anticipated to be minor or mainly controlled via other means (i.e. local anesthetics) suboxone may be continued in the preoperative period.
My patient was on naltrexone.  As a mu antagonist, it was anticipated that override might be difficult.  While, indeed, higher doses of fentanyl than typical were required, the requirements were not extraordinary.  Once again, an understanding of pharmacokinetics is important to determine how difficult it might be to override any naltrexone present at the mu receptor.  A single oral dose of 25 mg of naltrexone has an apparent half-life (t1/2) of 14 hours. Veberey et al. [2], looked at naltrexone effectiveness in blockade of heroin 25 mg. After a 100 mg oral dose of naltrexone, there was 96% blockade at 24 hours, 86.5% blockade at 48 hours, and 46.6% at 72 hours. Therefore, it would seem prudent to recommend that patients avoid naltrexone for 72 hours prior to any surgery where it is anticipated post op pain control will require opioid therapy.  In patients who have chronically taken naltrexone for maintenance therapy, there is up regulation of mu receptors in the brain [8]. Therefore, patients who discontinue naltrexone, may be resistant or more susceptible to opioids depending on how much naltrexone remains in the system compared to the degree of mu opioid receptor upregulation.  While naltrexone is a mu opioid antangonist, it is approved by the FDA for treatment of alcoholism because it reduces the euphoria and positive reinforcing effects of ethanol use [9].
Lastly, this patient had a history of vitamin B12 deficiency.  Vitamin B12 is an integral part of two biochemical reactions: the conversion of L-methylmalonyl coenzyme A into succinylcholine coenzyme A and the formation of methionine by methylation of homocysteine.  These reactions are critical for the synthesis of DNA and to the maintenance of the myelin sheath by methylation of myelin basic protein. Active Vitamin B12 contains cobalt in its reduced form (Co+).  Nitrous oxide produces irreversible oxidation to the Co++ form leading to inactive Vitamin B12. The result is an irreversible inactivation of the enzyme methionine synthase  There are several case reports in the literature of patients developing subacute combined degeneration of the spinal cord following nitrous oxide anesthesia [11,12,13].  In these case reports, the diagnosis of subacute combined degeneration of the spinal cord was made several weeks post op after otherwise routine and uneventful anesthesia. Subacute combined degeneration of the spinal cord is characterized by degeneration of the posterior and lateral columns.   The avoidance of nitrous oxide in patients with anemia of unknown cause or in patients at risk of vitamin B12 depletion is recommended.   A significant risk factor for Vitamin B12 depletion is alcoholism for two reasons. 1) alcohol abuse leads to poor nutrition and therefore, vitamin B12 intake is reduced, and 2) excessive alcohol use leads to atrophic gastritis.  Atrophic gastritis is characterized by inflammation and dysfunction on the cells lining the stomach so that production and secretion of intrinsic factor is compromised.   Intrinsic Factor (IF) is essential for the absorption of vitamin B12.  Atrophic gastritis also decreases the production of hydrochloric acid. The increase in pH leads to overgrowth of different bacterial strains that consume vitamin B12 and reduction in efficiency of food breakdown that is necessary for the release of vitamin B12 from the diet for absorption into the body.

Lastly, her seizure episode following anesthesia should be addressed more fully in context of her history of alcoholism.  While there are case reports of severe vitamin B12 deficiency leading to seizures, this isn't typical.  What  is more common in alcoholics undergoing surgery is alcohol withdrawal syndrome (AWS). AWS can result in an incidence of mortality of around 15% if untreated [14]. In patients who abuse alcohol, AWS more commonly manifests during stress such as surgery [15]. It has been suggested that in operative patients, hypotension, hypoxia, or uncontrolled pain in the PACU may precipitate AWS [16].  The key features of AWS include hyperexcitability of the CNS due to reduced GABA activity and increased glutaminergic action [17]. Increased noradrenergic activity can lead to hypertension, tachycardia, sweating, tremor, and hallucinations.  Lastly, seizures can occur. Seizures typically occur early in the process, i.e. 6 to 8 hours after stopping alcohol intake. This is long before the most severe manifestations of AWS, which may take 3 to 4 days. Attributing this patient's prior post anesthetic seizure to AWS may be reasonable, but can't be proven.

In conclusion, this patient had multiple medical issues that were impactful on making an anesthetic plan.  These included a history of seizures related to anesthesia and vitamin b12 depletion, alcoholism treated with naltrexone and prazosin, as well as high anxiety with a  history of severe emergence delirium requiring treatment.  In this patient, using TIVA, local anesthesia for the surgery site, along with a BIS monitor allowed for an uneventful anesthetic and emergence.

1. Lepouse c et al.  Emergence delirium in adults in the post-anesthesia care unit, BJA, 2006, vol. 96(6), 747-53.
2. Verebey K. The clinical pharmacology of naltrexone: pharmacology and pharmacodynamics. NIDA Res Monogr 1981;28: 147-58.
3. Curtis D, Stevens WC. Recovery from general anesthesia. Int. Anesthsiol. Clin, 1991;29(2):1-11.
4. Lovestrand D, Phipps S, and Lovestrand S.  Posttraumatic stress disorder and Anesthesia Emergence.    AANA Journal, 2013(81).3. 199-203.

5. McGhee LL, Maani CV, Garza TH, DeSocio PA, Gaylord KM, Black IH. The relationship of intravenous midazolam and posttraumatic stress disorder development in burned soldiers. J Trauma Inj Infect Crit Care. 2009;66(4 suppl):S186-S190 
6.  Wilson JT, Pokorny ME. Experiences of military CRNAs with ser- vice personnel who are emerging from general anesthesia. AANA J. 2012;80(4):260-265.
7.  McGhee LL, Maani CV, Garza TH, DeSocio PA, Gaylord KM, Black IH. The correlation between ketamine and posttraumatic stress disorder in burned service members. J Trauma Inj Infect Crit Care. 2008;64(2 suppl):S195-S199.
8. Yoburn BC, Luke MC, Pasternak GW, Inturrisi CE. Upregulation of opioid receptor subtypes correlates with potency changes of morphine and DADLE. Life Sciences 1988;43: 1319-24.
9.  Volpicelli JR, Alterman AL, Hayashida M, O'Brien CP. Naltrexone in the treatment of alcohol dependence. Arch Gen Psych 1992;49: 876-80
15. C. Spies, H. Tønnesen, S. Andreasson, A. Helander, and K. Conigrave, “Perioperative morbidity and mortality in chronic alcoholic patients,” Alcoholism, vol. 25, no. 5, pp. 164S–170S, 2001. 

No comments: