REGIONAL ANESTHESIA

SECTION EDITOR
DENISE J. WEDEL

The Timing of Intravenous Crystalloid Administration and

Incidence of Cardiovascular Side Effects During Spinal
Anesthesia: The Results from a Randomized Controlled Trial
Jose L. Mojica,

MD*,

Hector J. Melendez,

MD*,

and Leonelo E. Bautista,

PhD

*Department of Surgery, School of Medicine, Universidad Industrial de Santander; and Department of Preventive
Medicine and Biometrics, Uniformed Services University of the Health Sciences

We conducted a randomized clinical trial to evaluate the

efficacy of crystalloids in preventing spinal-induced hypotension (SIH) and cardiovascular side effects (CVSE) in
a group of surgical patients. Participants were assigned to
receive lactated Ringers solution at 12 mL/min (Placebo
group, n 142); lactated Ringers at 20 mL/kg starting
20 min before spinal block (n 130); or lactated Ringers at
20 mL/kg starting at the time of spinal block (n 132).
SIH was defined as a decrease of 30% in baseline systolic
blood pressure, and CVSE as SIH plus nausea, vomiting,
or faintness requiring treatment. The incidence of SIH was
similar in all treatment groups. Compared to placebo,

ypotension is a most common cardiovascular

response to spinal anesthesia. More than 30%
of the patients undergoing spinal anesthesia
develop intraoperative spinal-induced hypotension
(SIH) (1). The administration of large volumes of IV
crystalloids 1520 min before spinal anesthesia to prevent SIH has become an increasingly common practice
(23). However, the efficacy of crystalloid administration before spinal block has been tested mostly in
obstetric patients (4 6), and its benefit has been small
(35,7). Only a small number of studies (8 12) have
evaluated the value of crystalloid administration before spinal block versus no crystalloids in general
surgery patients. Two studies with very small sample
size (8,11) found no significant difference in the incidence of SIH between patients receiving and not receiving crystalloids before spinal block, although
blood pressure was significantly increased in the intervention group. No beneficial effect in preventing
SIH was observed in two trials among patients 60 yr
of age (9,10). However, a large study (12) of patients
admitted for orthopedic or general surgery found that
Accepted for publication September 26, 2001.
Address correspondence and reprint requests to Leonelo E. Bautista, 4301 Jones Bridge Road, Room A1039, Bethesda, MD 20852.
Address e-mail to lbautista@usuhs.mil.

432

Anesth Analg 2002;94:4327

crystalloid administration at the time of spinal block resulted in a significant reduction in the proportion of patients developing CVSE from 9.9% to 2.3%. The corresponding relative proportion was 0.23 (95% confidence
interval, 0.07 0.78; P 0.019), and one additional case of
CVSE was avoided for each 13 patients receiving crystalloids at the time of spinal block instead of placebo. Administration of crystalloids at the time of spinal block seems to
be effective because it provides additional intravascular
fluids during the period of highest risk of CVSE after spinal anesthesia.
(Anesth Analg 2002;94:4327)

crystalloids administered before spinal block significantly reduced the incidence of cardiovascular side
effects (CVSE), but only during the first 15 min of
anesthesia.
Although the time elapsed between administration
of fluids and spinal anesthesia may influence efficacy,
no previous study has evaluated the benefits of crystalloid administration at varying times before spinal
block. We have conducted a clinical trial to evaluate
the efficacy of crystalloids administered 20 min before
or at the time of spinal anesthesia, as compared with
not administering crystalloids (Placebo group), in preventing SIH and CVSE in a group of general and
specialty surgery patients.

Methods
The study was approved by the Research Ethics Committee of the School of Medicine, Universidad Industrial de Santander, Bucaramanga, Colombia. Written
informed consent was obtained from all participants
or their parents. General and specialty surgery patients 11 yr old undergoing spinal anesthesia were
eligible for the study, excluding pregnant women,
those with a pelvic mass 10 cm, and those with
cardiac or pulmonary decompensation. Participants
2002 by the International Anesthesia Research Society
0003-2999/02

ANESTH ANALG
2002;94:4327

were randomly allocated to receive a continuous IV

infusion of lactated Ringers solution at a rate of
12 mL/min (Placebo group), lactated Ringers solution at a rate of 20 mL/kg body weight starting 20 min
before the spinal block, or lactated Ringers solution at
a rate of 20 mL/kg body weight starting at the time of
spinal block. Fluids were administered within 6 to
10 min and thereafter all participants received lactated
Ringers solution at 12 mL/min. Treatments were
allocated according to a computer-generated random
sequence.
Hyperbaric 0.5% bupivacaine without epinephrine
was injected between L2-3, L3-4, or L4-5 with the
patient seated or resting in the lateral position at
0.25 mg/kg body weight with a maximum total dose
of 15 mg. The extension of spinal blockade was assessed by cold temperature discrimination using a wet
cotton ball (13). Baseline arterial blood pressure and
heart rate were measured immediately before the administration of bupivacaine and monitored thereafter
using an automated oscillometer. Measurements at 2,
4, 6, 8, 10, 15, 20, and 25 min were recorded and used
for data analysis.
SIH and CVSE were the main study outcomes. In
agreement with previous studies (10 12), SIH was
defined as a decrease of 30% in baseline systolic
blood pressure, which is a change believed to warrant
treatment (14,15). At the beginning of the procedure,
patients were instructed to report any episode of nausea or faintness occurring during the intervention.
Patients were unaware of what treatment they were
receiving. CVSE were defined as SIH plus clinical
symptoms (nausea, vomiting, or faintness) requiring
treatment. CVSE were treated with a bolus of IV
ephedrine (100 g/kg body weight), with additional
doses every 5 min, if deemed necessary.
Patient clinical characteristics were described using
means and proportions and their corresponding 95%
confidence intervals (95% CI). Fishers exact test and the
F test with a 0.05 significance level were used to test
differences in the distribution of dichotomous and continuous variables, respectively. The risk of SIH and
CVSE were estimated for each treatment group. The
relative risk (RR) and its 95% confidence interval (95%
CI) were used to measure the efficacy of each intervention as compared with placebo. Log binomial multiple
regression (16) was used to estimate the independent
effect of clinical risk factors on the risk of SIH and CVSE.
Log binomial regression allows for the estimation of
adjusted RR instead of adjusted odds ratios, which frequently overestimate the effect of the exposure.

Results
A total of 404 patients were recruited into the study,
142 were assigned to the placebo group, 130 to receive

REGIONAL ANESTHESIA
MOJICA ET AL.
CRYSTALLOID ADMINISTRATION DURING SPINAL ANESTHESIA

433

crystalloids before spinal block, and 132 to receive

crystalloids at the time of spinal block. Approximately
half (53.5%) were men and the mean age was 42 yr
(95% CI, 40 44). Four patients were 15 yr old and one
was 11 yr old. No biologically significant betweengroup differences were observed in the distribution of
clinical characteristics immediately before spinal block
(Table 1). However, as expected, the systolic and mean
arterial blood pressure were significantly increased in
the group receiving crystalloids 20 min before spinal
block, which, contrary to the other treatment groups,
had received a significant volume of fluids by the time
of injection of bupivacaine (Table 1).
Among all study subjects, the risk of SIH in the first
25 min after spinal anesthesia was 14.6% (95% CI,
11.1%18.1%), whereas the risk of CVSE was 7.2%
(95% CI, 4.6%9.7%). Although patients who received
crystalloids before spinal block had an incidence of
SIH 1.52 times more frequent than those in the Placebo
group (19.2% versus 12.7%), this difference was not
statistically significant (P 0.143) (Table 2). Subjects
who did not receive and those who received crystalloids at the time of spinal block had a similar incidence
of SIH (12.1% versus 12.7%) and CVSE (9.2% versus
9.9%, respectively). In contrast, the administration of
crystalloids at the time of spinal block resulted in a
statistically and clinically significant reduction from
9.9% to 2.3% in the risk of CVSE as compared with
administering no additional crystalloids. The corresponding RR was 0.23 (95% CI: 0.07, 0.78; P 0.019)
and one additional case of CVSE was avoided for each
13 patients receiving crystalloid at the time of spinal
block instead of no additional crystalloids [1/(0.099
0.023)]. Similar results were obtained when administration at the time of spinal anesthesia was compared
to administration before spinal anesthesia (RR, 0.26;
95% CI, 0.08 0.91). Adjustment for body weight and
bupivacaine dose in the multivariate analysis did not
result in substantial changes in the effect of crystalloids administered before or at the time of spinal
block.
The incidence of SIH and CVSE decreased with age
up to 40 yr and then increased progressively (Table 3).
They also increased with ASA physical status score,
from 9.8% in ASA I to 26.7% in ASA III for SIH and
from 2.7% to 13.3% for CVSE, respectively (Table 3).
Males were also more likely to develop complications
than females, and a peak block height above T5 was
associated with an increased risk of SIH and CVSE
(Table 3).
Treatment, age, and block height were the only
independent variables associated to the risk of SIH in
the multivariate analysis (Table 4). Treatment effects
adjusted by age and block height (Table 4) were practically identical to those estimated in the crude analysis (Table 3). Compared with the group of 30 39 yr
old, patients 30 yr old had a nonsignificant increase

434

REGIONAL ANESTHESIA MOJICA ET AL.

CRYSTALLOID ADMINISTRATION DURING SPINAL ANESTHESIA

ANESTH ANALG
2002;94:4327

Table 1. Distribution of Clinical Characteristics Just Before Spinal Block by Allocated Treatment
Treatment group

Age (yr)
Sex (% male)
Weight (kg)
Bupivacaine dose (mg)
Block height T5 (%)*
ASA physical status (%)
I
II
III
SBP (mm Hg)
DBP (mm Hg)
Heart rate (bpm)

No crystalloids
(placebo)
(n 142)

Crystalloids
before
spinal block
(n 130)

Crystalloids
at time of
spinal block
(n 132)

P value

41.5 (38.544.6)*
59.9 (51.768.0)
61.4 (59.763.2)
14.0 (13.714.2)
63.4 (54.971.3)

42.4 (39.545.2)
50.0 (41.358.7)
64.7 (62.666.8)
14.3 (14.114.5)
68.5 (59.776.3)

42.1 (39.344.9)
50.0 (41.458.6)
64.1 (62.365.8)
14.4 (14.214.6)
73.5 (65.180.8)

0.92
0.16
0.03
0.05
0.20

61.3 (52.769.3)
32.4 (24.840.8)
6.3 (2.911.7)
132.7 (129.7135.6)
75.1 (73.177.1)
82.5 (79.885.2)

60.0 (51.068.5)
37.7 (29.346.6)
2.3 (0.56.6)
137.8 (134.4141.1)
76.9 (74.978.8)
81.5 (78.884.1)

68.9 (60.376.7)
28.8 (21.237.3)
2.3 (0.56.5)
133.2 (130.1136.3)
73.9 (72.175.8)
81.2 (78.683.9)

0.16

0.05
0.12
0.78

Data given as mean (95% confidence interval).

SBP systolic blood pressure; DBP diastolic blood pressure.
* Block height is the maximum block height after spinal block.

Table 2. Absolute and Relative Risk of Hypotension and Cardiovascular Side Effects by Allocated Treatment

Hypotension
Placebo group
Before spinal block
At time of spinal block
Cardiovascular side effects
Placebo group
Before spinal block
At time of spinal block

Absolute risk
(95% CI)

Relative risk
(95% CI)

P value

12.7% (7.719.3%)
19.2% (12.827.1%)
12.1% (7.118.9%)

1.00
1.52 (0.872.65%)
0.96 (0.511.80%)

0.143
0.889

9.9% (5.516.0%)
9.2% (4.915.6%)
2.3% (0.46.5%)

1.00
0.94 (0.451.95%)
0.23 (0.070.78%)

0.860
0.019

Placebo group was administered lactated Ringers solution at a rate of 12 mL/min; Others were administered lactated Ringers solution at 20 mL/kg body
weight 20 min before or starting at the time of spinal block.
95% CI 95% confidence interval.

in risk, whereas among patients 50 yr the risk was

more than three times higher (Table 4). Also, for each
increment of one segment in peak sensory block
height (range from T2 to T12), the risk of SIH increased 25% (95% CI, 4%51%) (Table 3). ASA physical status was also associated with risk of SIH, but
only when age was not included in the model. Patients
with ASA physical status IIIII were two times (RR,
2.32; 95% CI, 1.44 3.76) and four times (RR, 4.03; 95%
CI, 1.56 10.36) more likely to develop SIH than patients with ASA I, respectively.
Regression modeling for CVSE resulted in the selection of the same risk factors as those for SIH (Table 4).
Although the age-specific RR had wide confidence intervals, a pattern similar to that described for SIH was
observed, and the risk of CVSE was more than four
times higher in patients 50 than in those 3039 yr old
(Table 4). In addition, the risk of CVSE increased 1.47
times (P 0.011) for each increment of one segment in
peak sensory block height. When age was excluded from
the model, the RR associated with an ASA physical

status IIIII 3 were 5.33 (95% CI, 2.34 12.13; P 0.001)

and 6.75 (95% CI, 1.5030.37; P 0.013), respectively.
Finally, all groups showed a similar pattern of
changes in systolic blood pressure with time after
bupivacaine administration (Fig. 1). Immediately preceding spinal anesthesia, patients who received crystalloids before spinal block had a higher systolic blood
pressure (Fig. 1, Table 1) than other patients. However, beyond 10 min after bupivacaine administration
the systolic blood pressure was very similar in patients receiving crystalloids before spinal block and
those from the Placebo group, but higher in those
receiving crystalloids at the time of spinal block. Time
trend changes in diastolic blood pressure were similar
to those described for systolic blood pressure.

Discussion
Rapid infusion of large volumes of crystalloid is commonly used as prophylaxis for SIH. The goal of administering fluids before spinal block is to increase

ANESTH ANALG
2002;94:4327

REGIONAL ANESTHESIA
MOJICA ET AL.
CRYSTALLOID ADMINISTRATION DURING SPINAL ANESTHESIA

435

Table 3. Risk of Hypotension and Cardiovascular Side Effects by Clinically Relevant Risk Factors in All Study Subjects
Hypotension
Risk factor
Age (yrs)
1119
2029
3039
4049
5059
6069
7084
ASA physical status
I
II
III
Sex
Male
Female
Block height
T5
T5

Cardiovascular side effects

No. subjects

Risk (%)

95% CI

Risk (%)

95% CI

28
92
82
73
46
47
36

14.3
9.8
7.3
13.7
19.6
25.5
25.0

4.032.7
4.617.8
2.715.2
6.823.8
9.433.9
13.940.3
12.142.2

7.1
5.4
2.4
5.5
8.7
10.6
19.4

0.923.5
1.812.2
0.38.5
1.513.4
2.420.8
3.523.1
8.236.0

256
133
15

9.8
22.6
26.7

6.414.1
15.830.6
7.855.1

2.7
15.0
13.3

1.15.6
9.422.3
1.740.5

216
188

18.6
11.1

13.324.9
7.316.1

9.6
5.1

5.814.7
2.68.9

128
276

10.9
16.3

6.117.7
12.121.2

3.1
9.1

0.97.8
5.913.1

95% CI 95% confidence interval.

Table 4. Multivariate Adjusted Relative Risk for Hypotension and Cardiovascular Side Effects by Clinical Risk Factors
and Treatment in Patients Undergoing Spinal Anaesthesia
Hypotension
Risk factor
Crystalloid administration
Placebo group
Before spinal block
At time of spinal block
Block height
Age (yr)
1119
2029
3039
4049
5059
6069
7084

Cardiovascular side effects

Relative
risk

95% CI

P value

Relative
risk

95% CI

P value

1.00
1.54
0.90
1.25

0.892.66
0.531.85
1.041.51

0.121
0.974
0.019

1.00
0.93
0.25
1.47

0.461.88
0.080.83
1.091.98

0.831
0.023
0.011

0.535.98
0.533.81

0.355
0.479

0.3214.92
0.4811.58

0.427
0.293

0.745.02
1.167.88
1.499.20
1.419.68

0.180
0.023
0.005
0.008

0.4411.93
0.8021.24
0.9322.32
1.8637.71

0.327
0.090
0.061
0.006

1.77
1.43
1.00
1.93
3.03
3.70
3.67

2.19
2.35
1.00
2.29
4.12
4.57
8.38

Placebo group was administered lactated Ringers solution at a rate of 12 mL/min; others were administered lactated Ringers solution at 20 mL/kg body
weight 20 minutes before or starting at the time of spinal block.
95% CI 95% confidence interval.

venous return and preserve central blood volume and

cardiac output (17,18), both of which decrease in spinal anesthesia as a consequence of sympathetic blockade (2).
Our study showed that crystalloid administration before spinal block in general and specialty surgery patients does not reduce the incidence of SIH when compared with placebo (12 mL/min of lactated Ringers
solution). This finding is consistent with those from previous studies (811) and may be partly explained by the
short intravascular half-life of crystalloids (6,12,19). Administering larger volumes of crystalloids may increase

the efficacy of fluids administered before spinal block,

but such practice is not advisable because of the increased risk of pulmonary edema (20), hemodilution
(21), and urinary retention (20). A more rapid administration does not seem to improve the efficacy of crystalloids administered before spinal block and may induce
unwanted increases in central venous pressure (5). Replacing colloids for crystalloids seems to be a reasonable
approach because colloids remain longer in the vascular
space, but there is no conclusive evidence of a decreased
incidence of SIH in patients administered colloids
(10,22). In addition, colloids are more expensive than

436

REGIONAL ANESTHESIA MOJICA ET AL.

CRYSTALLOID ADMINISTRATION DURING SPINAL ANESTHESIA

Figure 1. Change in systolic blood pressure after spinal anesthesia

by treatment.

crystalloids, and they carry a small but significant risk of

anaphylaxis (23).
Administration of crystalloids before spinal block
was also ineffective in preventing CVSE in the 25 minutes after spinal block. This question has been evaluated in a previous study in which a beneficial effect of
crystalloids was limited to the first 15 minutes after
spinal anesthesia (12). These apparently contrasting
results could be partly explained because in our study
crystalloids were administered within 6 to 10 minutes
starting 20 minutes before spinal block, whereas in
Arndt et al.s study (12), crystalloids were administered within the 15 minutes preceding spinal block.
Crystalloid administration at the time of spinal
block resulted in an incidence of SIH almost identical
to that seen in the Placebo group, but led to a significant reduction in the risk of CVSE as compared with
placebo (RR, 0.23; P 0.019; number needed to treat,
13) or with crystalloids administered before spinal
block (RR, 0.26; P 0.014; number needed to treat, 14).
The frequent efficacy of administering crystalloids at
the time of spinal block could be explained by the
timing of hemodynamic events after spinal anesthesia.
Sympathetic nerve blockade is completed within the
first 510 minutes after administration of bupivacaine
(2,24). This period corresponds to the steepest decrease in systolic blood pressure and is followed by a
period of increased risk of CVSE. As a matter of fact,
62% of the CVSE cases in our Placebo group developed more than 10 minutes after bupivacaine administration. Similarly, in the study of Arndt et al. (12)
Sixty-one percent of the cases occurring in the first
45 minutes of follow-up developed after the first
15 minutes. Therefore, crystalloids administered at the
time of spinal block may result in the availability of
extra fluids in the intravascular space during the period of highest risk of side effects, leading to timely
compensatory cardiovascular changes. In contrast, administering crystalloids before spinal block also increases systolic blood pressure but it is not effective in

ANESTH ANALG
2002;94:4327

preventing SIH or CVSE because by the time of highest risk a considerable volume of crystalloids have
already diffused to the interstitial compartment and
can no longer improve venous return and cardiac
output (6,19). The administration of fluids before spinal block may initiate volume overload reflexes that
are then counterproductive during the onset of spinal
block, whereas administration at the time of spinal
block does not require the body to quickly reverse the
direction of volume homeostasis mechanisms.
The use of SIH as an outcome may partly explain
the negative results observed in trials of crystalloids
administered before spinal block (8 11). Although a
reduction of 25%30% in the baseline systolic blood
pressure warrants treatment (14,15), most patients can
safely tolerate such a decrease (20). Thus, this definition of hypotension may result in too many false positive cases and may dilute the estimate of the efficacy
of crystalloids administered before spinal block. On
the contrary, CVSE is a more specific and clinically
meaningful outcome because it includes SIH plus
alarming symptoms usually leading to treatment. Because false positive cases are less likely to occur with
CVSE, this outcome seems to be more appropriate for
evaluating the efficacy of fluid administration. Because patients were instructed to report nausea and
faintness but were unaware of the assigned treatment,
it is unlikely that our results were affected by biased
reporting of cardiovascular symptoms. Similarly, although the anesthesiologists knew the treatment assigned to the patient, observer bias is not a likely
explanation of our findings because blood pressure
was measured with an automatic device.
Age and peak block height were the only variables
independently associated with the risk of SIH and
CVSE in our study. Finding a decreased incidence of
both outcomes in the 30 39 year old group and increasing risk in younger as well as older subjects was
unexpected. Although an increased risk of SIH in patients over 40 years old has been previously reported
(1,25), the actual form of the age-incidence relationship should be further explored in future studies.
Changes in autonomic function and decreased cardiac
reserve may partly explain age-related changes in the
incidence of SIH and CVSE (1).
We found increases of 25% (P 0.019) in the incidence of SIH and 47% (P 0.011) in the incidence of
CVSE for each increment of one segment in peak
sensory block height, and only one case of SIH developed in patients with block height below T6. This
finding is consistent with previous reports of a greater
decrease in systolic blood pressure and an increased
risk of SIH in patients with sensory block level equal
or above T5 (1,8). Higher levels of sensory block correlate with higher cephalic spread of sympathetic
blockade, greater decrease in venous return and cardiac output, and increased risk of SIH. In contrast,

ANESTH ANALG
2002;94:4327

lower levels of sympathetic blockade may result in

compensatory vasoconstriction in the upper extremities and decreased risk of SIH.
In summary, our study showed that administering
crystalloids at the time of spinal block had a beneficial
effect in preventing CVSE in general and specialty
surgery patients undergoing spinal anesthesia as compared with administering crystalloids before spinal
block or administering no crystalloids. Crystalloid administration at the time of spinal block seems to be
more effective because it provides additional intravascular fluids in the period of increased risk of CVSE.

References
1. Carpenter RL, Caplan RA, Brown DL, et al. Incidence and risk
factors for side effects of spinal anesthesia. Anesthesiology 1992;
76:906 16.
2. Covino BG, Scott DB, Lambert DH. Handbook of spinal anesthesia and analgesia. Philadelphia: WB Saunders, 1994.
3. Murray AM, Morgan M, Whitwam JG. Crystalloid versus colloid for circulatory preload for epidural caesarean section. Anaesthesia. 1989;44:463 6.
4. Rout CC, Rocke DS, Levin J, et al. A reevaluation of the role of
crystalloid administration for the prevention of hypotension
associated with spinal anesthesia for elective cesarean section.
Anesthesiology 1993;79:2629.
5. Rout CC, Akoojee SS, Rocke DA, Gouws E. Rapid administration of crystalloid preload does not decrease the incidence of
hypotension after spinal anesthesia for elective cesarean section.
Br J Anaesth 1992;68:394 7.
6. Rocke DA, Rout CC. Volume preloading, spinal hypotension
and Cesarean section. Br J Anaesth 1995;75:2579.
7. Jackson R, Reid JA, Thorburn J. Volume preloading is not essential to prevent spinal-induced hypotension at caesarean section. Br J Anaesth 1995;75:2625.
8. Venn PJ, Simpson DA, Rubin AP, Edstrom HH. Effect of fluid
preloading on cardiovascular variables after spinal anaesthesia
with glucose-free 0.75% bupivacaine. Br J Anaesth 1989;63:
6827.
9. Coe AJ, Revanas B. Is crystalloid preloading useful in spinal
anaesthesia in the elderly? Anaesthesia 1990;45:2413.

REGIONAL ANESTHESIA
MOJICA ET AL.
CRYSTALLOID ADMINISTRATION DURING SPINAL ANESTHESIA

437

10. Buggy D, Higgins P, Moran C, et al. Prevention of spinal

anesthesia-induced hypotension in the elderly: comparison between preanesthetic administration of crystalloids, colloids, and
no prehydration. Anesth Analg 1997;84:106 10.
11. Casati A, Fanelli G, Berti M, et al. Cardiac performance during
unilateral lumbar spinal block after crystalloid preload. Can J
Anaesth 1997;44:623 8.
12. Arndt JO, Bomer W, Krauth J, Marquardt B. Incidence and time
course of cardiovascular side effects during spinal anesthesia
after prophylactic administration of intravenous fluids or vasoconstrictors. Anesth Analg 1998;87:34754.
13. Brull SJ, Greene NM. Zones of differential sensory block during
extradural anaesthesia. Br J Anaesth 1991;66:6515.
14. Blake DW, Donnan G, Novella J, Hackman C. Cardiovascular
effects of sedative infusions of propofol and midazolam after
spinal anaesthesia. Anaesth Intensive Care 1988;16:292 8.
15. Logan MR, McLure JH, Wildsmith JAW. Plain bupivacaine: an
unpredictable spinal anaesthetic agent. Br J Anaesth 1986;58:
292 6.
16. Wacholder S. Binomial regression in GLIM: estimating risk ratios and risk differences. Am J Epidemiol. 1986;123:174 84.
17. Robson S, Hunter S, Boys R, et al. Changes in cardiac output
during epidural anaesthesia for caesarean section. Anaesthesia
1989;44:4759.
18. Morgan P. The role of vasopressors in the management of
hypotension induced by spinal and epidural anaesthesia. Can J
Anaesth 1994;41:404 13.
19. Twigley AJ, Hillman KM. The end of the crystalloid era? A new
approach to peri-operative fluid administration Anaesthesia
1985;40:860 71.
20. McCrae AF, Wildsmith JAW. Prevention and treatment of hypotension during central neural block. Br J Anaesth 1993;70:
672 80.
21. Hahn RG. Increased haemodilution in hypotension induced by
epidural anaesthesia. Acta Anaesthesiol Scand 1993;37:357 60.
22. Koski E, Tuppurainen T, Mattila M, et al. Hydroxyethyl
starches, dextran and balanced salt solution in correction of
hypotension during epidural anaesthesia. Acta Anaesthesiol
Scand 1984;28:5959.
23. Stoelting RK. Allergic reactions during anesthesia. Anesth
Analg 1983;62:34156.
24. Berde CB, Strichartz GR. Local Anesthetics. In. Miller RD, eds.
Anesthesia, 5th. Edition. Philadelphia: Churchill Livingstone,
2000:503.
25. Dohi S, Naito H, Takahashi T. Age-related changes in blood
pressure and duration of motor block in spinal anesthesia. Anesthesiology 1979;50:319 23.