Regional Anesthesia

## Cardiac Arrest Following Central Neuraxial Block

*Sadhana S. Kulkarni and Savani S. Futane*

## **Abstract**

Central neuraxial blocks (CNB) are used worldwide in anesthesia practice. They are safe, however, not devoid of untoward complications. Cardiac arrest (CA) is one of the major devastating complications. The anesthesiologists are concerned about CA as it can occur unexpectedly and suddenly even in a young ASA grade I patient, undergoing elective surgery, at any time during and after administration of CNB in spite of continuous vigilance. A better understanding of the physiology of CNB, availability of monitoring devices, and safer local anesthetic drugs contribute to reduced mortality, yet cases of CA are reported even recently. These case reports provide information relevant to particular incidents and may be inadequate to provide comprehensive information to explain the overall clinically important aspects related to CA following CNB. This chapter would provide a summary and analysis of the current recommendations about etiology, predisposing factors, preventive measures, and various measures tried for the treatment of cardiac arrest, although the exact etiology and predisposing factors are still not known. The comprehensive information would be helpful for anesthesiologists during day-to-day practice and to increase the safety of patients undergoing CNB. Proper patient selection, pre-/co-loading of fluids, the modifying technique of CNB as per patient's need, early use of epinephrine during bradycardia refractory to atropine, continuous monitoring, vigilance during intra- and postoperative period would help in prevention, early detection, and prompt treatment of CA. Challenges faced by anesthesiologists during CNB practice and newer modalities used for the treatment of refractory CA are also discussed. The mystery of sudden unexpected CA is yet to be solved and research in this direction is warranted. Electronic medical record keeping and reporting untoward incidence to the national board will also help to improve patient safety in the future.

**Keywords:** anesthesia, epidural, anesthesia, spinal, anesthetic technique, central neuraxial block, complications, bradycardia, cardiac arrest, hypotension

## **1. Introduction**

Central neural blocks (CNB) are commonly used in the perioperative period and are an integral part of anesthetic practice because of well-known reasons [1]. The low rate of complications is one of the reasons for their popularity particularly in regions with the limited health care resources.

The techniques are considered safe but major adverse events such as neurological complications and cardiac arrest (CA) are reported at times, and the techniques are not without risks [1, 2]. It is evident from reports of studies that cardiac arrest following CNB is not rare [3–6]. CA under CNB is a major concern as it is reported in ASA grade I young patients, undergoing elective surgery, and can occur suddenly without warning signs [4, 7]. CA following spinal anesthesia is reported since 1940, yet the exact etiology is not known [1, 8]. Even though the outcome of patients developing CA has improved in the last two decades, the possibility of tragic events does exist despite adequate and timely resuscitation [9]. These case reports provide information relevant to particular incidents and may be inadequate to provide comprehensive information to explain the overall clinically important aspects related to CA following CNB. This chapter would provide a comprehensive view of etiology, predisposing factors, preventive measures, and treatment of cardiac arrest. The information would help to increase patient safety during spinal and epidural anesthesia. The anesthesiologists can make use of this information for proper selection of patients, preoperative optimization of patients, modifying anesthetic technique as well as monitoring as per patient need, to implement measures to prevent severe bradycardia, hypotension, use of different modalities during refractory cardiac arrest and for postoperative care of patients receiving CNB. The importance of vigilance and monitoring during intraoperative as well as in the postoperative period is reinforced as unexpected CA can occur at any time [2, 10]. The chapter would also make the anesthesiologists aware of where the research stands on this critical issue of CNB and in what direction future research is needed.

This chapter is intended to serve as a pragmatic review for use in daily anesthesia practice of CNB (spinal, epidural, and combined spinal-epidural) in adult patients. The manuscript is structured in a way that may help the anesthesiologists to quickly find the most important information about CA relating to the current information and underlying evidence. We did not carry out a systematic literature review. To present a holistic overview of this clinically important subject, a comprehensive literature search was performed in January–April 2022 in MEDLINE, PubMed, and Google Scholar to retrieve articles pertaining to a cardiac arrest related to CNB. The keywords used in various combinations included "Central neuraxial blocks and cardiac arrest"; spinal anesthesia and cardiac arrest; epidural anesthesia and cardiac arrest; local anesthesia systemic toxicity; hypotension and spinal anesthesia. A systematic review would result in a larger and more detailed manuscript that could be difficult to use as a quick clinical reference, even though it would decrease the probability of excluding relevant publications [11].

**Incidence of cardiac arrest:** The exact incidence of CA is not known [1]. The real incidence of CA related to CNB is heterogeneous and has a wide range from 0.07 to 49 per 10,000 patients [2, 4, 5, 10, 12]. In 2002, the incidence of CA following CNB was 10:10,000 [13]. A better understanding of physiological changes following CNB, availability of safer local anesthetic drugs, and improved monitoring have contributed to the reduction in the incidence of CNB [14]. However, even recent reports confirm that CA under spinal anesthesia is not rare [5, 8, 10].

In 2002, the incidence of CA following spinal anesthesia was more as compared to that following epidural, 2.5/10,000 and 0–0.5/10000, respectively [15, 16]. Incremental doses and slower onset of epidural contribute to a lower incidence of CA as compared with spinal anesthesia [3]. However, Cook et al. observed a higher incidence of permanent neurological damage including death following epidural and combined spinal epidural than spinal, 18.2 and 2.8 per 100,000, respectively [2].

Biboulet et al. reported that the incidence of CA was more following spinal than general anesthesia [17]. However, according to the majority of investigators, the incidence is more during general anesthesia [8, 14, 18]. It may be because the more complicated surgeries and high-risk patients are conducted under general anesthesia.

## **2. Etiology**

The real etiology of CA is still not known, even though CA following spinal was reported in 1940. Etiology of CA following spinal and epidural is multifactorial. Due to inconsistent reporting, risk factors leading to bradycardia and CA under spinal anesthesia remain uncertain and contradictory [19]. Etiology of CA is summarized in **Table 1**.


The most likely etiology of CA during spinal/epidural anesthesia is mainly peripheral vasodilatation and reduction of preload resulting from sympathetic blockade. The level of sympathetic blockade extends two to six dermatomes


above the sensory blockade [3, 22]. Cardio-accelerator fibers (T1-T4) can be blocked when a sensory level is at T4 and their blockade produces negative chronotropic, inotropic, and dromotropic effects. Nevertheless, it is not uncommon to see high-sensory blockade levels without hemodynamic changes, particularly in young patients. Reduction in right atrial pressure is likely in 36% of the patients, when the level block is less than T4 dermatome and in 53% of patients when it is above that [13, 23]. Anesthesiologists generally test the level initially till the desired level is achieved for surgical procedure. Higher levels achieved subsequently (due to patient position, baricity, type of local anesthetic, and other factors) may remain unnoticed.

3.Exacerbation of the parasympathetic nervous system

Sympathetic blockade results in significant bradycardia and even asystole. The final pathway is the absolute or relative increase in activity of the parasympathetic nervous system [23]. CA is more common in young individuals as they have a greater vagal tone. The parasympathetic response following spinal anesthesia, traction on viscera, pain, etc., is further exaggerated in these patients [3]. Cardiac arrest during needle insertion is reported particularly in the anxious patients [6]. CA was preceded by bradycardia in many studies [9, 16].

4.Intrinsic cardiac reflexes

A decrease in preload may initiate reflexes leading to severe bradycardia [24] (**Figure 1**).

a.Reflexes involving the pacemaker stretch: The rate of firing of cells of the pacemaker within the myocardium is proportional to the degree of stretch. Decreased venous return to the right atrium results in the decreased stretch and a slower heart rate.

b.The reflex from low-pressure baroreceptors in the right atrium and vena cava.

c.Reflexes arise from inhibitory mechano-receptors in the left ventricle. Decrease in ventricular volume would normally decrease receptor activity leading to tachycardia. However, a rapid decrease in left ventricular volume may trigger a paradoxical increase in the activity of these receptors, which could be due to forceful ventricular contraction around an almost empty chamber. This reflex slowing should allow time for a more complete filling of the heart [25].

Ecoffey et al. studied the effect of sympathetic blockade with echocardiography in unpremeditated volunteers and observed that two out of eight volunteers developed bradycardia and hypotension along with a reduction in left ventricular diameter, with epidural anesthetic levels of T8 and T9. Changes reverted by head-down positioning and rapid infusion of I.V. fluids. The increased levels of human pancreatic polypeptide, a marker of parasympathetic function, associated with these episodes of bradycardia suggest vagal activation. Bradycardia due to an increase in vasopressin levels without changes in catecholamine levels is observed after the head-up tilt in the presence of sympathetic blockade [26]. Pregnant patients undergoing spinal

*Cardiac Arrest Following Central Neuraxial Block DOI: http://dx.doi.org/10.5772/intechopen.106600*

#### **Figure 1.**

*Receptors in heart responsible for cardiac arrest following CNB.*

anesthesia are at increased risk for hypotension and bradycardia due to aortocaval compression and a higher level of spinal block [27].


addition, anxiety or viscous traction, in such individuals, can produce severe bradycardia or atrioventricular heart blocks [3, 29]. A small postural change includes placing legs in the holder, and turning the patient to the left lateral or prone position and CA was reported even after the surgical procedure was over. It is difficult to explain these situations based only on preload changes. Maybe they are due to reflex phenomena similar to those of autonomic dysfunction or hyperreflexia described in patients with a spinal cord section. One should be vigilant during the change of posture of the patient receiving spinal anesthesia [30].

Paradoxically young patients and athletes are frequently classified as low-risk ones, have increased vagal tone, and appear to be at risk of developing severe bradycardia. The highly competitive athlete, in addition, may have "athletic heart syndrome". Its features include sinus bradycardia, sinus dysrhythmias, first-degree and Mobitz type I blockades, and alterations in repolarization. Occasionally, CA has been described during the spinal anesthesia in athletes [31, 32]. Jordi et al. observed the development of first-degree AV block progressing to asystole in patients undergoing spinal anesthesia with the sensory blockade at the T3 dermatome [23]. Retrospective analysis of postoperative holter monitoring indicated persistent first-degree block for six hours after anesthesia. Development of a firstdegree block can be a warning sign for the development of asystole. However, the difficulty in diagnosing first-degree block using a cardioscope limits the applicability of this finding [4].


Causes of Cardiac arrest following epidural block:

Cardiac arrest can occur during epidural anesthesia [6, 21, 34, 35] due to causes similar to that following spinal anesthesia. In addition, unintentional "total spinal" anesthesia, and local anesthetic systemic toxicity (LAST) are common causes of CA during an epidural block. Absorbed local anesthetic from vascular epidural space can add to bradycardia. Occasional severe toxicity and deaths are reported. While using a mixture of local anesthetics, one should not use maximum doses of two local anesthetics in the belief that their toxicities are independent [36]. Heavy intravenous sedation with drugs such as midazolam can mask early signs of LAST, particularly convulsions. Among all, bupivacaine is considered to be 4–16 times more cardiotoxic than lignocaine. The use of ropivacaine and levobupivacaine may help reduce cardiotoxicity due to stereo-selective binding of sodium and potassium channels resulting in less affinity and strength of inhibitory effect [37]. Jacobson concluded that reduction in preload leading to an increase in vagal activity is responsible for arrest rather than blockade of cardiac accelerator nerves from the study on healthy volunteers receiving

*Cardiac Arrest Following Central Neuraxial Block DOI: http://dx.doi.org/10.5772/intechopen.106600*

epidural [25]. Development of third-degree heart block is reported following thoracic epidural block in a patient having preoperative first-degree heart block [38]. Even though there is a segmental block during epidural, partial sympathetic block can be there in lower segments resulting in preload reduction [39].

A combined spinal-epidural technique (CSEA) may be preferable to a continuous epidural technique as is associated with a lower failure rate, better pain scores, and patient satisfaction. Epidural top-ups of local anesthetic should be given in small incremental doses [40].

Causes of early-onset CA may be vasovagal during needle prick, hypovolemia, compromised cardiovascular status, posture-related changes, and accidental intravascular/intrathecal injection during an epidural block, etc. Late-onset CA may be due to blood loss, myocardial infarction, and delayed spread of local anesthetic after spinal anesthesia, surgical stimulus like traction on mysentry, cementing, posture change, tourniquet release, and use of vasodilators such as nitroglycerine or sodium nitroprusside during total hip replacement, etc.

## **3. Predisposing factors for cardiac arrest**

Although the development of CA during spinal anesthesia is considered as the final step of a spectrum of manifestations that starts with bradycardia, establishing an association among factors related to its development can help identify patients at-risk to develop CA during spinal block [3].

**Risk factors for severe bradycardia**: Pollard has suggested the risk factors as shown in **Table 2** [3].

According to Carpenter, the level of the block had the weakest correlation with the development of bradycardia [20]. The presence of two or more listed factors in **Table 2** may place these patients at high risk for bradycardia and cardiac arrest under spinal anesthesia [3]. Due to inconsistent reporting, the risk factor associated with the occurrence of bradycardia and cardiac arrest under spinal anesthesia remains uncertain and contradictory [19].

Patients with a background of vagal dominance, and with a history of vasovagal syncope, may be predisposed to severe bradycardia and even cardiac arrest following spinal anesthesia [10]. I.V. supplementary drugs such as fentanyl, dexmeditomidine, droperidol, beta-blockers, and ondansetron [41–45] can be predisposing factors due to alpha- or beta-receptor blocking effect.


#### **Table 2.**

*Risk factors for bradycardia during central neuraxial block.*


#### **Table 3.**

*Risk factors for hyotension during central neuraxial block.*

**Risk factors for hypotension:** These include hypovolemia, age > 40 to 50 years, emergency surgery, obesity, chronic alcohol consumption, and chronic hypertension, aortocaval compression after 20 weeks of gestation, and alkalinization or excessive doses of local anesthetic (**Table 3**) [7, 22, 46, 47].

CA observed shortly after CNB is due to excessive doses of local anesthetic in the previously hypovolemic patient. Preoperative fasting, dehydration, diuretics and vasodilator drugs for hypertension are common causes. Incidence of CA is more during orthopedic surgeries like hip surgery. Blood loss during surgery, cementing, or postural changes also contribute to CA [16, 48]. The level of sensory blockade in elderly patients is usually higher than that of young adults with the same dose of local anesthetic. According to Biboulet et al. [14], doses as low as 5 mg of bupivacaine, hyper- or isobaric, can cause a sensorial blockade reaching up to T2-T4 [17]. Overdose of local anesthetic using the subarachnoid route is a known cause of CA in elderly patients. It is recommended that the level of the blockade should be limited to T6 and hemodynamic reserves should be evaluated perioperatively to prevent untoward events [48].

When SA is administered by surgeons and non-anesthetist health care providers, the incidence of CA was more [49]. This is due to lack of monitoring, delay in detection, and treatment of complications by non-qualified health care professionals. Lozts et al. postulated that hyperbaric solutions can have delayed cephalad spread even after minutes post-injection and it can take more than 40–60 minutes to fix finally [50]. Obstetric patients have more sympathetic activity so a lower incidence of CA is expected [2, 3]. Adekola et al. observed more cardiac arrests (7.3/10000) in pregnant mothers as compared with non-obstetric patients, however postmortem reports revealed that the causes were not related to spinal anaesthesia [12].

## **4. Prevention of severe bradycardia, hypotension, and cardiac arrest**

Final pathway for the development of severe bradycardia and CA is parasympathetic over activity. Specific strategies to anticipate and prevent vagal predominance *Cardiac Arrest Following Central Neuraxial Block DOI: http://dx.doi.org/10.5772/intechopen.106600*

form the mainstay in the management of severe bradycardia and CA under spinal anesthesia (**Figure 2**).

## **A.** Prevention:

a.**Appropriate patient selection:** When two or more risk factors are present (**Table 2**) and when significant intraoperative blood loss or use of vasodilators such as sodium nitroprusside or nitroglycerine is anticipated, one should reconsider the choice of spinal anesthesia [3].

## b.**Explaining the procedure to the patient during pre-anesthetic checkup:**

This will help to reduce anxiety and fear, which can trigger severe bradycardia. Anxiolytic like oral hydroxyzine in the apprehensive patients, atropine premedication in the selected patients (vagotonic), application of local anesthetic cream or infiltration before insertion of the needle, administration of CNB in the lateral position, etc., can help to reduce incidence of sudden bradycardia during needle insertion [51].

## c.**Prophylactic atropine premedication:**

Routine premedication with atropine is not recommended and does not reduce the incidence of bradycardia and hypotension [52]. Bradycardia of different grades is observed during CNB (mild<60, moderate <50, severe </ min). Clinically significant bradycardia occurs in 10 to 15 percent of spinal anesthetics [53]. The incidence of bradycardia with epidural anesthesia depends on the level and extent of the block. It may be considered in elderly patients having bradycardia and those with the history suggestive of vagotonic symptoms (0.5 mg immediately after spinal anesthesia) [54]. Prophylactic administration of I.V. atropine after spinal did not prevent a decline in blood pressure in parturients even though heart rate was more at 15 and 20th min [55]. I.V. atropine prevented bradycardia when dexmeditomidine sedation was administered but there was increase in the blood pressure so should be used carefully [56]. Epinephrine should be administered in the presence of refractory bradycardia. Early administration of 0.2–0.3 mg adrenaline or drip 0.15 mcg/kg/minute prevents cardiac arrest and subsequent morbidity [3, 40].

d.**Modification in CNB technique:** Hemodynamic consequences can be reduced a) by administering a low dose of local anesthetic with the additive [57], using unilateral spinal anesthesia for lower limb surgery [58] by titrating the required level by using continuous spinal anesthesia [59]. This can reduce the extent of sympathetic block. Care should be executed for local anesthetic toxicity while using a mixture of local anesthetics and epidural dosages when combined spinal epidural anesthesia is administered.

## e.**Maintaining adequate preload and prevention of hypotension**:

Uncorrected hypovolemia increases the risk of hypotension with the onset of CNB and is an absolute contraindication for spinal anesthesia. Risk factors for hypotension include hypovolemia, age > 40 to 50 years, emergency surgvery, obesity, chronic alcohol consumption, pregnant patient with gestational period of more than 20 weeks, and chronic hypertension (**Table 3**) [40]. In vagotonic patients particularly when blood loss is expected during surgery, preload maintenance is essential [3]. Unfortunately, this is not routinely followed [3, 29]. Preloading with colloids or co-loading with colloids or crystalloids administered within 5–10 minutes is effective [60, 61]. Co-loading should be practiced carefully in patients with preeclampsia.

Change from supine to prone or Trendelenburg/lithotomy to supine posture, tourniquet release, intra-operative blood loss, and use of vasodilators can produce preload changes and need preload correction in anticipation. The position should be resumed if hypotension/bradycardia is observed. 30 centimeter leg elevation increases the venous return and is useful in settings with resource constraints [62]. *Cardiac Arrest Following Central Neuraxial Block DOI: http://dx.doi.org/10.5772/intechopen.106600*

10–15 degrees left lateral tilt is beneficial in parturient to reduce aortocaval compression reducing preload. Unfortunately, adequacy of preload is difficult to assess clinically. Assessment of inferior vena cava diameter and left ventricular volume by using non-invasive techniques such as ultrasonography (USG) and transthoracic echocardiography can be useful but may not prevent CA [63, 64].

Invasive blood pressure monitoring can help to increase safety in critical patients [65]. Bradycardia may be an early manifestation of reduction in preload and atropine or vasopressor may be needed to treat vagal manifestation and only fluid administration may not be sufficient. Administration of atropine 0.4–0.6 mg is recommended to prevent cardiac arrest.

f. **Vasopressors**: Vasopressors with different modes of action are tried for the prevention and treatment of hypotension in elderly and obstetric patients, particularly when pre- or co-loading is risky as in patients of preeclampsia (**Table 4**).

Ephedrine, phenylephrine, or noradrenaline can be used for prophylaxis. Ephedrine produces tachycardia and is to be avoided in patients where tachycardia is undesirable as in patients with aortic stenosis. It produces tachyphylaxis when used in repeated doses, and hence is administered as intermittent boluses and not as an infusion. Phenylephrine (alpha-agonist) has duration of action of up to 20 minutes. Noradrenaline increases cardiac output due to its alpha-agonist action and additional weak beta-agonist effect. About 8 mg ondansetron blocks Bezold Jarisch reflex activated by serotonin and is used to limit hypotension. Further evidence is awaited for routine use of ondansetron [66–70].

g.**Continuous vigilance throughout the procedure:** Blood loss, altered consciousness, and signs of vagal over activity such as nausea, sweating, bradycardia, change of posture, traction on viscera, and vital parameters are essential throughout the procedure as well as in the postoperative period also. Awareness about delayed CA is necessary [32].

## **5. Treatment of bradycardia and hypotension**

**Treatment of bradycardia:** Mild bradycardia (<60/min) should be treated in patients with risk factors (3). It is enough to have intra-operative hypotension with bradycardia (<50/min) to rapidly administer atropine plus a vasoconstrictor (e.g., ephedrine). Treatment of moderate and severe bradycardia with hypotension must be quick, intensive, and multimodal. According to Tarkilla et al., atropine is recommended for bradycardia as glycopyrrolate is ineffective [46]. Alexander has suggested that atropine (0.5 mg) or glycopyrrolate (0.2 to o.4 mg) and ephedrine (5 to 10 mg) I.V. can be used for the treatment of bradycardia with hypotension [71]. It does not seem to be wise to administer just one of these drugs and then wait for the result [3, 32]. Pollard recommended a stepwise approach of administering atropine (0.4–0.6 mg), ephedrine (25–50 mg), and epinephrine (0.2–0.3 mg) for the treatment of moderate bradycardia. If there is no improvement after atropine and vasoconstrictors, intravenous epinephrine must be administered without any delay, as recommended by the SOS ALR group in France [74]. Head low position (careful before 30 minutes after spinal) [75] and fluid administration should

