Full text of DOI:10.1007/BF03018966

393
Reports of Investigation
Arrhythmias during
spinal anesthesia for
Cesarean section
Chih-Long Shen MD,
Yung-Yuan Ho MD,
Yu-Chun Hung MD,
Ping-Lung Chen MD*
Purpose: Spinal block has long been considered a safe anesthesia technique for surgery. However, severe brady-
cardia, cardiac arrest, and other arrhythmias during spinal anesthesia have been reported and the incidence of
intraoperative arrhythmias is not well established. In this study the incidence of arrhythmias during spinal anes-
thesia was determined.
Methods: We studied 254 healthy women undergoing Cesarean section under spinal anesthesia prospectively.
Spinal anesthesia with 10 mg bupivacaine mixed with 0.2 mg morphine was performed at the L3-4 interspace.
Intraoperative arrhythmias were recorded and verified later by a cardiologist.
Results: First degree atrioventricular block developed in nine patients (3.5%), second degree atrioventricular
block in nine (3.5%), severe bradycardia (heart rate < 50 beats·min–1) in seventeen (6.7%), multiple VPC in three
(1.2%). The height and weight of patients with severe bradycardia, multiple VPCs, or atrioventricular block were
not different from those of the other patients. However, the age of patients in the potentially dangerous arrhyth-
mias group was greater than that in the other group (P = 0.006).
Conclusion: The incidence of arrhythmias as well as hypotension during spinal anesthesia for Cesarean section
was higher than expected. Although most of these arrhythmias were transient and recovered spontaneously, they
might unexpectedly occur and sometimes need immediate and prompt treatment. It is necessary to remain vigi-
lant during spinal anesthesia for Cesarean section and careful monitoring of these patients is warranted, especial-
ly in older parturients.
Objectif : On a longtemps considéré la rachianesthésie comme une technique sécuritaire pour la chirurgie.
Cependant, des cas de bradycardie sévère, d’arrêt cardiaque et d’autres arythmies ont été signalés pendant la
rachianesthésie et l’incidence d’arythmie peropératoire n’est pas bien définie. Dans la présente étude, on a déter-
miné l’incidence d’arythmie pendant la rachianesthésie.
Méthode : Nous avons fait l’étude prospective de 254 femmes en santé qui ont subi une césarienne sous rachi-
anesthésie. L’injection de 10 mg de bupivacaïne avec 0,2 mg de morphine a été réalisée dans l’espace L3-4.
L’arythmie peropératoire a été notée et vérifiée ensuite par un cardiologue.
Résultats : Un bloc auriculoventriculaire de premier degré s’est développé chez neuf patientes (3,5 %), un bloc
de second degré chez neuf patientes également (3,5 %), une bradycardie sévère (fréquence cardiaque < 50 bat-
tements·min–1) chez dix-sept (6,7 %), de multiples ESV chez trois (1,2 %). La grandeur et le poids des patientes
qui ont présenté une bradycardie sévère, de multiples ESV ou un bloc auriculoventriculaire ne différaient pas de
ceux des autres patientes. Toutefois, les patientes du groupe présentant des arythmies potentiellement dan-
gereuses étaient plus âgées que les autres (P = 0,006).
Conclusion : L’incidence d’arythmie autant que d’hypotension pendant la rachianesthésie pour une césarienne
a été plus élevée que prévu. Même si la plupart de ces arythmies ont été transitoires et sont disparues spon-
tanément, elles peuvent survenir inopinément et exiger parfois un traitement rapide et immédiat. Cette situation
appelle à la vigilance et au monitorage suivi de ces patientes, justifié surtout chez les parturientes plus âgées.
From the Department of Anesthesiology and Section of Cardiology,* Ton Yen General Hospital, Hsin Chu, Taiwan.
Financial support was received from the Research Foundation of Ton Yen General Hospital, Taiwan.
Address correspondence to: Dr. Chih-Long Shen, 69, Second Shen-Chen Rd, Department of Anesthesiology, Ton Yen General Hospital,
Chu Pei, Hsin Chu, 302-42 Taiwan. Phone: 886-3-5527000 Ext 1201; Fax: 886-3-5516585; E-mail: yuchun51@ms22.hinet.net
Accepted for publication January 7, 2000.
CAN J ANESTH 2000 / 47: 5 / pp 393–397

394
CANADIAN JOURNAL OF ANESTHESIA
UBARACHNOID block has long been con-
sidered a safe anesthetic technique for surgery.
In addition, small doses of intrathecal mor-
phine produce effective intraoperative and
mence five seconds before the trigger (Hewlett
Packard, REC M1116B). The ECG records were ver-
ified later by a cardiologist. Outpatient department
follow-up and 24-hr Holter study were suggested to
all parturients with second degree AV block. The con-
dition of the neonatal was evaluated by Apgar scores
at one and five minutes.
To identify the most important individual risk fac-
tors associated with potentially dangerous arrhyth-
mias,8 patients with severe bradycardia, multiple
VPCs, or AV block were compared with the other
patients.
S
postoperative analgesia.1,2 Therefore, combined
intrathecal administration of morphine and bupivacaine
for Cesarean section has been widely used during the
current decade. However, severe bradycardia, cardiac
arrest, and other arrhythmias during spinal anesthesia
have been reported.3–6 The incidence of these arrhyth-
mias is not well established.7 Therefore, we prospective-
ly studied 254 healthy women undergoing Cesarean
section under spinal anesthesia to determine the inci-
dence of intraoperative arrhythmias.
Statistical analysis
Statistical analysis was performed using: 1) Spearman’s
rank correlation coefficient to correlate spread of block-
ade with age, body height and weight.9,10 Dermatomal
levels, age, body weight and height were changed to
rank scales before statistical analysis. 2) Unpaired
Student’s t test to evaluate the risk factors for AV block,
severe bradycardia and multiple VPC. Differences were
considered to be statistically significant when P < 0.05
and data were presented as mean SD.
Methods
After ethics committee approval, informed consent
was obtained from all patients. From December 1,
1998 until May 31, 1999, we prospectively collected
254 cases, including twin pregnancy in five parturi-
ents, who had consented to spinal anesthesia for
Cesarean delivery at our hospital. Parturients who had
arrhythmias including multiple VPCs or atrioventricu-
lar (AV) block, who had systemic disease including
hypertension, valvular heart disease, or thyroid gland
dysfunction, who were taking any medication other
than vitamins, those who had received epidural labour
analgesia before Cesarean section, and those who
needed other anesthesia or analgesia for inadequate
spinal block were excluded.
Results
The patient demographics, including age, body height
and weight, are shown in Table I. The maximum cepha-
lad levels of analgesia to pinprick after spinal block are
shown in the Figure. There was no correlation between
patient age, body height, body weight and the maxi-
No premedication was given. In the operating
room, automatic noninvasive blood pressure at two
minute intervals, and pulse oximeter were applied.
The ECG, lead II, was continuously monitored
(Hewlett Packard, CMS 24 OmniCare). Patients
received nasal oxygen supplementation at 2 L·min– 1
and a rapid infusion of 20 mL·kg– 1 lactated Ringer’s
prior to anesthesia. Then, with the patients in the
right lateral decubitus position, spinal puncture was
performed at the L3-4 interspace and 10 mg hyperbar-
ic bupivacaine 0.5% mixed with 0.2 mg morphine
were administered. The patients were immediately
turned supine and the operating table was positioned
in a left lateral tilt to achieve left uterine displacement.
The levels of the block to pinprick were measured at
five minute intervals. If systolic blood pressure was <
100 mmHg or 80% of baseline, intermittent 8 mg
boluses of ephedrine iv were given. All parturients
with severe bradycardia (heart rate < 50 min– 1)
received 0.5 - 1 mg atropine iv.
TABLE I Characteristics of patients.
Age (yr)
Height (cm)
Weight (kg)
Mean SD
Range
29.3 4.5
18 ~ 43
157.8 5.2
140 ~ 173
68.9 10.4
46 ~ 110
TABLE II Case number and incidence of of arrhythmias.
Arrhythmia
Case number Incidence (%)
Mobitz type II
Atypical Wenckebach periods1 2
Mobitz type I
3
4
2
1.2
1.6
0.8
3.5
6.7
1.2
8.3
17.7
8.3
30.3
First degree AV block
Bradycardia
Multiple PVCs
PVCs
9
17
3
21
45
21
77
APC
Supraventricular rhythm
Sinus arrhythmia
The ECG was closely observed by the anesthesiol-
ogist. When arrhythmias occurred, the ECG was
recorded and the unit allowed the recording to com-
Incidence = Arrhythmia case number ÷ All patients number (254)
Bradycardia: heart rate < 50 min– 1
Multiple PVCs: > 6 ·min– 1

Shen et al.: ARRHYTHMIAS DURING SPINAL ANESTHESIA
395
minutes later, she had recovered completely and
surgery was resumed. After that, the anesthetic course
and hospital stay were uneventful.
The body height, and weight of parturients with
severe bradycardia, multiple VPCs, or AV block were
not different from those of the other parturients.
However, the age of patients in the potentially dan-
gerous arrhythmias group was greater than that in the
other group (31.3 4.4 vs 29.0 4.4 yr, P = 0.006).
Discussion
Severe bradycardia, multiple VPCs, or AV block
adversely affected circulatory function, or was likely to
initiate more dangerous ventricular arrhythmias. On
the contrary, APCs, VPCs, sinus arrhythmia, or
supraventricular rhythm had minimal impact on car-
diovascular function and usually did not require treat-
ment.8 Therefore, patients with severe bradycardia,
multiple VPCs, or AV block were compared with the
other patients to identify the most important individ-
ual risk factors of potentially dangerous arrhythmias.
Our study revealed that increasing age increased the
chance of developing arrhythmias and there was no
correlation between arrhythmias and other factors,
such as body height, or weight. These results were
similar to those described by Carpenter et al.7
However, they studied all patients consenting to spinal
anesthesia for surgery, not only for Cesarean section.
High spinal block was required for Cesarean section,
the sensory levels in 242 patients (95.3% of total
patients) in our study were above T6. Sympathetic
blockade might be two segments higher than the sen-
sory block, namely, the thoracic sympathetic ganglia
were almost blocked for the mojority of our patients.
Therefore, the incidence of arrhythmias in our study
was higher than that in Carpenter et al. study (2.1%).
The distinction between Mobitz type I and type II
block is descriptive. Mobitz type I is diagnosed by the
presence of blocked atrial depolarization preceded by
prolongation of the PR interval. Mobitz type II AV
block manifests with several consecutive impulses that
are conducted with a constant conduction time and
that are followed by a block of conduction.1 1 In our
study, there were four cases in whom the PR intervals
increased or decreased haphazardly. It was not neces-
sarily the longest PR interval that was followed by the
blocked impulse. Denes1 2 called such episodes “atypi-
cal Wenckebach periods”, although some authors refer
to them as Mobitz type II,1 3 “Mobitz type II-
like”,14,15 “pseudo mobitz type II”16,17 or “apparent
Mobitz type II”.1 4However, the site of block is prob-
ably a more important determinant of prognosis and
the need for pacemaker therapy than the type of block.
FIGURE Maximum cephalad extent of analgesia to pin prick.
mum cephalad spread of anesthesia. Hypotension
occurred in 160 cases (63%), and the duration of
hypotension varied from two to 16 min. Except for one
premature newborn who died and in whom congenital
heart disease was suspected, 258 newborns had Apgar
scores $ 7 at one and five minutes.
The incidence of arrhythmias is shown in Table II.
The ECG recordings exhibited episodes of transient
second degree AV block in nine parturients. One case
occurred after spinal anesthesia and during uterine
repair, four after spinal anesthesia, one after skin inci-
sion, and three during delivery. The duration of each
episode was < three minutes. There were two cases of
Mobitz type I which had typical Wenckebach periods.
Three cases presented blocked atrial depolarization with
a preceding constant conduction time were Mobitz
type II. The PR intervals of the other four cases
increased or decreased haphazardly; they were atypical
Wenckebach periods. Five parturients with second
degree AV block, including all patients with Mobitz
type II, agreed to 24-hr Holter study. None showed
any degree of AV block. All of these nine parturients
have no symptomatic arrhythmias after more than three
months follow-up.
There were 17 parturients (6.7%) who manifested
severe bradycardia. Most improved after administra-
tion of atropine. Without warning, one patient sud-
denly developed severe bradycardia (heart rate:
23·min– 1) accompanied with loss of consciousness
when the uterus was repaired. Surgery was stopped, 1
mg atropine iv was administered and the lungs were
ventilated with O2 100% via a face mask. About three

396
Usually, Mobitz type I AV block is due to excess
CANADIAN JOURNAL OF ANESTHESIA
administration of atropine,2 7 some studies have shown
that atropine that only minimally improved conduction
in the AV node, but markedly increased the sinus rate.
This could increase the degree of block as a result of the
increasing atrial rate.19,20 Spear et al. demonstrated dif-
ferences in the latency, duration and threshold of the SA
node and the AV conduction in response to the sympa-
thetic effect.2 2 This mechanism was like “tachycardia-
dependent conduction disturbance”, AV block might
occur after increased vagal tone stopped, due to the rel-
atively prolonged AV nodal refractoriness.2 6 Fourth,
hemodynamic, hormonal, autonomic, and emotional
changes related to pregnancy might be associated with
arrhythmias.28,29
Body height has been considered an important
determinant of the dose of spinal anesthesia although
some evidence suggested no correlation between body
height or weight of parturients and spread of spinal
block.30,31 In our study, all parturients were given the
same dose of spinal anesthetic agent. There was no dif-
ference between age, body height, weight and spread
of the blockade. Thus, it is not necessary to vary the
dose of hyperbaric bupivacaine according to parturi-
ent’s age, body height or weight.
In summary, our study showed that the incidence
of arrhythmias as well as hypotension during spinal
anesthesia for Cesarean section was higher than we
expected. Although most of these arrhythmias were
transient and recovered spontaneously, they might
occur unexpectedly and, sometimes, require immedi-
ate treatment. It is necessary to remain vigilant and
careful monitoring of patients during spinal anesthesia
for Cesarean section is warranted, especially in older
parturients.
vagotonia and the site of the block is within the AV
node. The likelihood of the development of complete
AV block may be very low. Usually, Mobitz type II is
associated with a wide QRS complex and is due to dis-
ease of the His-Purkinje system.1 8It is more likely asso-
ciated with a high mortality rate and requires a
pacemaker to prevent more advanced AV block or car-
diac arrest. However, electrophysiologic studies in
Mobitz type II block with narrow QRS complexes
(absence of bundle-branch block) sometimes localized
the block at the AV node or proximal His-bun-
dle.13,19–21 Some studies demonstrated that vagal activi-
ty could modify AV conduction to produce first degree
to complete heart block including Mobitz type
II.13,20–25 For vagally mediated AV block, the decision
regarding the use of pacemakers is not based on the
type of block or on QRS duration but on the underly-
ing clinical settings and the correlation of symptoms.2 6
For all our patients with second degree AV block, the
prognosis was benign, because they had narrow QRS
complexes and only occurred transiently during spinal
anesthesia without significant ventricular pause.
There were 17 cases (6.7%) of severe bradycardia
during spinal anesthesia in our patients. There are no
studies of the relation between neonatal outcome and
maternal bradycardia during anesthesia. However,
maternal bradycardia might reduce uterine blood flow.
To protect both the fetus and the parturient, all of the
parturients with severe bradycardia received 0.5 - 1
mg atropine iv. One case suddenly developed severe
bradycardia and loss of consciousness without warning
Thus, we emphasize the importance of careful moni-
toring of patients and early management of side
effects. These may be the keys in preventing more
dangerous conditions, such as asystole.
References
There are many mechanisms to explain the high inci-
dence of intraoperative arrhythmias. First, the cephalic
spread of spinal block induced a relative increase in
parasympathetic activity by blockade of cardiac sympa-
thetic stimulation or vasovagal attack through decreased
venous return.6 The differences in latency and duration
of various cardiac tissues in response to this unstable
autonomic tone2 2 might encourage arrhythmias. Second,
nausea, vomiting and surgical manipulation during
Cesarean section might also increase vagal tone. The rel-
ative increase in parasympathetic activity was not associ-
ated only with hypotension, but also with bradycardia
and AV block in our study. Third, there were four cases
of second degree AV block occurring immediately after
intravenous ephedrine for hypotension or atropine for
bradycardia. Though AV nodal conduction was
enhanced and intranodal block tended to decrease after
1 Abboud TK, Dror A, Mosaad P, et al. Mini-dose
intrathecal morphine for the relief of post-Cesarean
section pain: safety, efficacy, and ventilatory responses
to carbon dioxide. Anesth Analg 1988; 67: 137–43.
2 Abouleish E, Rawal N, Fallon K, Hernandez D.
Combined intrathecal morphine and bupivacaine for
Cesarean section. Anesth Analg 1988; 67: 370–4.
3 Matta BF, Magee P. Wenckebach type heart block fol-
lowing spinal anaesthesia for Caesarean section. Can J
Anaesth 1992; 39: 1067–8.
4 Nishikawa T, Anzai Y, Namiki A. Asystole during
spinal anaesthesia after change from Trendelenburg to
horizontal position. Can J Anaesth 1988; 35: 406–8.
5 Mackey DC, Carpenter RL, Thompson GE, Brown DL,
Bodily MN. Bradycardia and asystole during spinal
anesthesia: a report of three cases without morbidity.
Anesthesiology 1989; 70: 866–8.

Shen et al.: ARRHYTHMIAS DURING SPINAL ANESTHESIA
397
duction within the atrioventricular conduction system
6 Shen CL, Hung YC, Chen PJ, Tsao CM, Ho YY. Mobitz
type II AV block during spinal anesthesia.
of the dog. Circ Res 1973; 32: 27–41.
23 Harrington JT Jr, DeSanctis RW. Hiccup-induced atri-
oventricular block. Ann Intern Med 1969; 70: 105–6.
24 Thorne MG. Hiccup and heart block. Br Heart J 1969;
31: 397–9.
25 Baron SB, Huang SK. Cough syncope presenting as
Mobitz type II atrioventricular block - an electrophysi-
ologic correlation. PACE 1987; 10: 65–9.
Anesthesiology 1999; 90: 1477–8.
7 Carpenter RL, Caplan RA, Brown DL, Stephenson C,
Wu R. Incidence and risk factors for side effects of
spinal anesthesia. Anesthesiology 1992; 76: 906–16.
8 Atlee JL. Perioperative cardiac dysrythmias. Diagnosis
and management. Anesthesiology 1997; 86:
1397–424.
26 Zaman L, Moleiro F, Rozanski JJ, Pozen R, Myerburg
RJ, Castellanos A. Multiple electrophysiologic manifes-
tations and clinical implications of vagally mediated AV
block. Am Heart J 1983; 106: 92–9.
27 Akhtar M, Damato AN, Caracta AR, Batsford WP,
Josephson ME, Lau SH. Electrophysiologic effects of
atropine on atrioventricular conduction studied by His
bundle electrogram. Am J Cardiol 1974; 33: 333–43.
28 Upshaw CB Jr. A study of maternal electrocardiograms
recorded during labor and delivery. Am J Obstet
Gynecol 1970; 107: 17–27.
9 Meyer RM. Ordinal data are not interval data (Letter).
Anesth Analg 1990; 70: 569–70.
10 Mantha S. Inappropriate use of linear regression analy-
sis (Letter). Reg Anesth 1992; 17: 179–80.
11 Barold SS, Friedberg HD. Second degree atrioventricu-
lar block. A matter of definition. Am J Cardiol 1974;
33: 311–5.
12 Denes P, Levy L, Pick A, Rosen KM. The incidence of
typical and atypical A-V Wenckebach periodicity. Am
Heart J 1975; 89: 26–31.
13 Rosen KM, Loeb HS, Gunnar RM, Rahimtoola SH.
Mobitz type II block without bundle-branch block.
Circulation 1971; 44: 1111–9.
29 Page RL. Treatment of arrhythmias during pregnancy.
Am Heart J 1995; 130: 871–6.
30 Hartwell BL, Aglio LS, Hauch MA, Datta S. Vertebral
column length and spread of hyperbaric subarachnoid
bupivacaine in the term parturient. Reg Anesth 1991;
16: 17–9.
31 Norris MC. Patient variables and the subarachnoid
spread of hyperbaric bupivacaine in the term parturi-
ent. Anesthesiology 1990; 72: 478–82.
14 Massie B, Scheinman MM, Peters R, Desai J, Hirschfeld
D, O’Young J. Clinical and electrophysiologic findings
in patients with paroxysmal slowing of the sinus rate
and apparent Mobitz type II atrioventricular block.
Circulation 1978; 58: 305–14.
15 Viitasalo MT, Kala R, Eisalo A. Ambulatory electrocar-
diographic recording in endurance athletes. Br Heart J
1982; 47: 213–20.
16 El-Sherif N, Aranda J, Befeler B, Lazzara R. Atypical
Wenckebach periodicity simulating Mobitz II AV
block. Br Heart J 1978; 40: 1376–83.
17 Lange HW, Ameisen O, Mack R, Moses JW, Kligfield P.
Prevalence and clinical correlates of non-Wenckebach,
narrow-complex second-degree atrioventricular block
detected by ambulatory ECG. Am Heart J 1988; 115:
114–20.
18 Narula OS, Samet P. Wenckebach and Mobitz type II
A-V block due to block within the His bundle and
bundle branches. Circulation 1970; 41: 947–65.
19 Zipes DP. Second-degree atrioventricular block.
Circulation 1979; 60: 465–72.
20 Mangiardi LM, Bonamini R, Conte M, et al. Bedside
evaluation of atrioventricular block with narrow QRS
complexes: usefulness of carotid sinus massage and
atropine administration. Am J Cardiol 1982; 49:
1136–45.
21 Lichstein E, Chadda KD. Atrioventricular block pro-
duced by swallowing, with documentation by His bun-
dle recordings. Am J Cardiol 1972; 29: 561–3.
22 Spear JF, Moore EN. Influence of brief vagal and stel-
late nerve stimulation on pacemaker activity and con-