Available online at www.sciencedirect.

com

Surgical Neurology 71 (2009) 60 – 65

www.surgicalneurology-online.com
Spine
A prospective randomized trial comparing the technique of spinal and
general anesthesia for lumbar disk surgery: a study of 100 cases
Seyed Hossein Sadrolsadat, MD a , Ali Reza Mahdavi, MD a , Reza Shariat Moharari, MD a ,
Mohammad Reza Khajavi, MD a , Patricia Khashayar, MD b ,
Atabak Najafi, MD a , Abbas Amirjamshidi, MD, MPH a,⁎
a
Sina Hospital, Medical Sciences/University of Tehran, Tehran, Iran
b
Research and Development Center, Sina Hospital, Medical Sciences/University of Tehran, Tehran, Iran
Received 13 July 2008; accepted 5 August 2008

Abstract Background: General anesthesia and regional anesthesia have both been shown to be suitable
techniques for patients undergoing lower thoracic and lumbar spine surgery; however, GA is the
most frequently used method. The purpose of this study was to conduct an acceptable RCT to
compare the intraoperative parameters and postoperative outcome after SA and GA in patients
undergoing elective lumbar disk surgery.
Methods: Patients undergoing laminectomy for herniated lumbar disk during the years 2005 and
2007 were enrolled. They were randomly selected to undergo GA and SA. The variables recorded
during the operation were the patients' HR, MAP, amount of blood loss, and surgeons' satisfaction
with the operating conditions. The severity of pain, nausea, vomiting, and length of stay in the
hospital were recorded in the postoperative course.
Results: The mean blood loss was less in the group undergoing GA; however, the difference was not
statistically significant. The surgeon's satisfaction was reported to be higher in the GA group. No
major intraoperative complication was reported in either series. During the recovery period,
hypertension was reported to happen more frequently in the patients undergoing GA; and
postoperative nausea and vomiting were more frequent among patients recovering from SA.
Conclusion: Contrary to previous studies, the findings of the present study revealed that SA has no
advantages over GA. Moreover, it was showed that GA can reduce the related risks and
complications in several aspects.
© 2009 Elsevier Inc. All rights reserved.
Keywords: General anesthesia; Lumbar disk surgery; Spinal anesthesia; Visual analog scale

1. Introduction
General anesthesia and regional anesthesia have both
been shown to be suitable techniques for patients under-
Abbreviations: ASA, American Society of Anesthesiologists; CVE,
compensatory volume expansion; GA, general anesthesia; HR, heart rate;
IV, intravenous; MAP, mean arterial pressure; NPO, nothing by mouth (nil
per os); PACU, postanesthesia care unit; RCT, randomized controlled trial;
SA, spinal anesthesia; TIVA, total intravenous anesthesia; TNG, trinitrogly-
cerin; TOF, train-of-4 ratio; VAS, visual analog scale.
⁎ Corresponding author.
E-mail address: abamirjamshidi@yahoo.com (A. Amirjamshidi).
0090-3019/$ – see front matter © 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.surneu.2008.08.003
going lower thoracic and lumbar spine surgery, and the
hypothesis is that “GA is of equal value with SA for
lumbar disc surgery.” Although GA is the more frequently
used method, this preference may be due to the greater
acceptance by patients, the ability to perform longer
operations [1,16], or the anesthesiologist feeling more
comfortable seeing that the airway is secured before placing
the patient in the prone position [14]. On the contrary, some
anesthesiologists choose regional techniques because they
believe it is accompanied with less blood loss [8] and less
hemodynamic instability [6,18]. An additional advantage is
the patient's ability to reposition their extremities and chest
 S.H. Sadrolsadat et al. / Surgical Neurology 71 (2009) 60–65
to avoid nerve injury, brachial plexus palsy, or pressure
necrosis to either the face or the chest wall [4,14]. Spinal
techniques may also reduce the incidence of pulmonary
complications compared with GA; however, episodes of
urinary retention and infection were reported following this
method [13,19].
There have been very few RCTs regarding this subject in
the literature [3,7,11,12]. The purpose of this study was to
conduct an acceptable RCT to compare the intraoperative
parameters and postoperative outcome after SA and GA in
patients undergoing elective lumbar disk surgery.
2. Material and methods
2.1. Ethics and inclusion criteria
After approval of the study protocol by the Institutional
Ethical Board Committee, all the patients awaiting lami-
nectomy for herniated lumbar disk in 1 to 2 levels during the
years 2005 and 2007 were enrolled in the study.
Those with an ASA class higher than class III; patients
with multilevel spinal stenosis, coagulopathy, localized
infection (interpreted as a contraindication or a high risk
for SA), or severe cardiac or liver disease; and cases with any
relative or absolute contraindication for SA [11] were
excluded from the study.
An informed consent was obtained from all the patients
after enough description about the different methods being
used for the patient during surgery. The patients were
allocated into 2 groups based on computer-generated
randomization. They were selected to undergo either GA
or SA. A single surgeon and a group of anesthesiologists
were responsible for performing all the operations. A
trained technician unaware of the objectives of the study
and the method by which the patient was operated was
responsible for gathering the required information.
2.2. Techniques
Patients undergoing GA were given 2 mg of IV
midazolam. Induction was carried out using propofol IV
(2 mg/kg) and fentanyl IV (2 μg/kg). Endotracheal
intubation was facilitated with atracurium IV (0.5 mg/kg).
Once the patients were intubated, they were placed in a prone
position on a standard operating frame. Maintenance
anesthesia consisted of oxygen 100% with propofol (100-
200 μg/[kg min]) and alfentanil (2 mL/400 mg propofol).
Anesthetics were modified to maintain hemodynamic
variables within 10% of the baseline values. Ten milligrams
of atracurium was administered every 20 minutes as muscle
relaxant. At the termination of GA, the anesthetic agents
were discontinued; and the patient breathed 100% oxygen.
The patient was then rolled to a supine position on the
postoperative bed. At that time, atropine and neostigmine
were used to reverse the relaxation. When appropriate, they
were extubated and then transported to the PACU.
61
The SA patients were placed in the seated position on
arrival in the operating room. After the injection of local
anesthetics (1 mL of lidocaine 1%), SA was delivered using
a lumbar puncture at the L3-4 intervertebral disk space via
Quincke 25-gauge spinal needle and injection of 4 mL of
bupivacaine 0.5% in the subarachnoid space.
Once the block was obtained, the patients were rolled
into the supine position and allowed to rest until a stable
maximal spread of the block to T6 through T10 was
achieved. After about 10 minutes, the patient was turned
to a prone position on the operating bed and placed on
the standard laterally located chest roles in the semiflexed
position. Oxygen was provided via face mask. During the
operation, the patient was sedated with propofol infusion
at 25 to 50 μg/(kg min) IV. At completion of the
operation, propofol was discontinued; and the patient was
rolled to supine position. The patients were then
transferred to the PACU. There were no dropouts in
this series of patients.
2.3. Data gathering
Age, sex, height, and weight were the demographic
characteristics entered for every patient. Electrocardio-
graphic monitoring, noninvasive blood pressure, pulse
oximeter, and capnography were recorded during the
operation in both groups.
The patient's HR (per minute) and MAP were recorded at
several intervals:
1. The time at which the patient had entered the operation
room,
2. The time anesthesia was initiated,
3. At the initiation of the operation (when incision was
performed),
4. At the termination of the operation (placement of surgical
dressing),
5. Once anesthesia was terminated (when the patient was
transferred to the recovery room), and
6. When the patient was discharged from the recovery room.
Mean arterial pressure and HR were also assessed every
10 minutes for at least half an hour in the recovery room in
both study groups (sample of the graphs taken for each
patient in Figs. 1 and 2).
The amount of blood loss was assessed by measuring
the bloody gauzes' weight plus the volume calculated from
the difference of suctioned fluid and the volume of
irrigation solutions administered (for the purpose of
washing the operation site). The infused liquid during
anesthesia, accounting for those administered before
induction, was documented.
The severity of pain and nausea was evaluated using
100-mm VAS; meperidine (25 mg IV) and metoclopramide
were prescribed in cases with pain intensity more than
50 mm and nausea more than 50 mm, respectively. Both
drugs were readministered in intervals of 30 minutes, if
62 S.H. Sadrolsadat et al. / Surgical Neurology 71 (2009) 60–65

Table 1
The demographic and baseline data of the 2 groups enrolled in the study
GA SA
Age 45.2 ± 5.6 45.7 ± 5.2
Weight 75.2 ± 8.2 77.8 ± 7.5
Baseline MAP 92.6 ± 4.9 93.3 ± 6.2
Baseline HR 78.2 ± 6.4 78.6 ± 7.0

assessed to be hemodynamically stable and comfortable at

Fig. 1. Sample of a graph showing the changes in MAP of a patient during
the study period. Numbers in the x-axis denote the time intervals at which
MAP was assessed: 1, the time at which the patient had entered the operation
room; 2, the time anesthesia was initiated; 3, after the intubation; 4, at the
initiation of the operation (when incision was performed); 5, at the
termination of the operation (placement of surgical dressing); 6, once
anesthesia was terminated (when the patient was transferred to the recovery
room); and 7, when the patient was discharged from the recovery room.
required. The reinjection of the drugs every 4 hours was
possible when the patient was transferred to the ward.
The number of blood bags injected during the operation,
the preoperative and postoperative values of hemoglobin,
and the satisfaction rate of the surgeon were recorded.
Surgeon's satisfaction was evaluated using a 100-mm
VAS based on the satisfaction from the amount of blood in
the surgical field, patients' muscular relaxation, being
motionless, and the possibility of assessing neurologic
status immediately after the operation. The plausibility and
internal validity of the variants considered for VAS of the
surgeon's satisfaction rate were examined in several
introductory sessions with the attending surgeons and
gained acceptable rates.
Patients were discharged from the recovery room based
on Miller's Anesthesia 2005 guidelines, TOF ratio equal to
0.85 in GA patients, and the regression of at least
4 dermatomes block in SA patients. All the patients were
the time of discharge. The total stay of the patients in PACU
was also recorded.
Hypotension or bradycardia was defined as the reduction
of MAP or HR to less than 80% of the baseline amounts.
Ephedrine IV 10 mg or more was administered in such cases.
On the other hand, hypertension or tachycardia was
considered when MAP or HR had raised to more than
120% of the baseline. These cases were treated using 20 to
30 mg of propofol IV and/or TNG with an initial dose of
10 μg/min.
During the pilot study, the difference between the
hypotension rate in the 2 groups was calculated (15% vs
40%); according to this findings and with α = .05 and a
power of 80%, a sample size of 100 (50 patients in each
group) was required.
The data were analyzed using SPSS (Chicago, IL) version
11.5 software. The χ2 test and the Student t test were used
for comparing the variables between the 2 groups. P values
lower than .05 were considered as significant.
3. Results
One hundred patients were enrolled in this study. The
mean age of the patients was 45.5 ± 5.4 years, and 50% were
male. There was no difference between the demographic data
and the baseline MAP and HR of the 2 groups (Table 1).
The perioperational findings of both groups are outlined
in Table 2. Regarding the surgical and anesthesia durations,
no statistically significant difference was observed between
the 2 groups (P = .932 and .084, respectively). In comparison
with the SA patients, the GA group lost about 26.5 ± 2.7 mL
less blood during the operation, which was marginally
significant (P = .054). Anyhow, no episodes of excessive
hemorrhage or cases requiring blood transfusions occurred in

Table 2
Intraoperative data for SA vs GA
GA (n = 50) SA (n = 50) P
Total anesthesia time (min) 126.4 ± 19.1 120.1 ± 16.9 .084
Operation time (min) 94.1 ± 17.9 94.4 ± 17.3 .932
Blood loss (mL) 438.0 ± 66.6 464.5 ± 69.3 .054
Hypotension 6 (12%) 24 (48%) b.001
Bradycardia 17 (34%) 17 (34%) 1.000
Tachycardia 13 (26%) 3 (6%) .006
Fig. 2. Sample of a graph showing the changes in HR of a patient during the
IV fluid (mL) 2890.0 ± 290.8 2930.0 ± 378.1 .555
study period. Numbers in the x-axis denote the time intervals at which HR
Surgeon's satisfaction 76.7 ± 7.5 64.0 ± 6.5 .038
was assessed (see Fig. 1 for interpretation of the numbers).
 S.H. Sadrolsadat et al. / Surgical Neurology 71 (2009) 60–65 63

Table 3 4. Discussion
Postoperative data for SA vs GA
GA SA P OR (95% CI)
A good anesthetic technique should have both rapid onset
and reversal of effects. It should provide desirable
Hypertension 19 (38%) 3 (6%) b.001 9.6 (2.6-35.2)
PACU time 23.8 ± 7.8 21.7 ± 8.8 .212 –
intraoperative hemodynamic conditions and, if possible,
Nausea and vomiting 3 (6%) 10 (20%) .037 3.9 (1.1-15.1) contribute to a reduced need for blood transfusion. More-
in PACU over, it should permit the earliest possible discharge from the
Nausea and vomiting 9 (18%) 5 (10%) .249 – PACU and minimize the common postoperative problems
after 24 h such as pain, analgesics consumption, nausea, and vomiting
Analgesic requirement 31 (62%) 11 (22%) b.001 5.7 (2.4-13.9)
in PACU
[5,17].
Analgesic requirement 37 (74%) 38 (76%) . 817 - The literature is largely supportive regarding SA used for
after 24 h lumbar laminectomy, although not in an evidence-based
OR indicates odds ratio; CI, confidence interval. manner; however, GA is the method of choice for spinal
procedures according to some authors [7]. This concept
either may be related to the lack of familiarity with the use of
either group. Furthermore, just 12% of GA patients spinal techniques for anesthesia in such procedures or may
compared with 24% of SA ones experienced hypotension reflect the uncertainty concerning the potential risks and
during the operation (P b .001). complications incurred when spinal operations are carried
Bradycardia was similarly seen in both groups, whereas out in a conscious patient under SA [20].
tachycardia was more frequent in those undergoing GA Several studies have compared SA and GA in lumbar
(26% vs 6%, P = .006). It is noteworthy that, in 3 cases of the laminectomy surgery, and many of which have concluded
SA group, coughing impeded the surgeons from proceeding SA as the preferred method [7,3,11,12]. However, the pre-
for a while and that, in 1 case, hiccups made the condition sent study did not support this hypothesis. Table 4 outlines
tense. However, no major intraoperative complication was the findings of the similar studies.
reported in either of the series. On the whole, surgeon's In the present study, lower prevalence of nausea and
satisfaction measured by VAS was higher in GA patients vomiting in the following day of the surgery, shorter
(76.7 ± 7.5 mm) compared with SA patients (64.0 ± 6.5 mm), recovery time, and lower need for analgesics during the
which was statistically significant (P = .038). recovery period were reported in patients undergoing SA.
The postoperative findings are summarized in Table 3. The patients in the GA group were reported to experience
Duration of recovery period was rather similar in the intraoperative hypotension as well as nausea and vomiting
2 groups (GA, 23.8 ± 7.8 minutes vs SA, 21.7 ± 8.8 minutes). (in the recovery room) less frequently than the other group.
During the period, 19 of GA cases (38%) had hypertension The surgeons were also more satisfied with this method.
episodes, whereas just 3 cases of the other group experienced A number of confounding factors could explain this
it (P b .001). Postoperative nausea and vomiting were more discordance. (a) Tolerating the prone position for about an
frequent among patients recovering from SA compared with hour, during disk surgery, is not possible unless the patient
GA (20% vs 6%, P = .037). Conversely, in the day after the is sedated. One of the best sedative agents administered for
operation, nausea and vomiting occurred 8% more in GA this purpose is propofol. It has short-term effects; and
cases than SA ones, although it was not statistically therefore, the sedative effects disappear within a few minutes
significant (P = .249). after the interruption of the drug infusion. The antivomiting
Patients in the SA group required less analgesics effect of this drug even in low doses (10-15 mg) is well
immediately after the operation, and the difference was known [10]; moreover, it has been shown that it can reduce
statistically significant (62% vs 22%, P b .001). However, bleeding by decreasing the systemic vascular resistance by
the need became similar on the day after the operation 15% to 25% [15]. Propofol infusion was administered as
(74% vs 76%). the sedative agent in the group undergoing SA in several

Table 4
Comparing SA vs GA outcomes of the RCTs conducted on patients undergoing laminectomy in the systematic review performed in Medline, Cochrane,
Scopious, and advanced G-scholar using the key words GA, (and) SA, (and) comparison, (and) laminectomy, compared with the present study
Study n N&V Bleeding RT AC Complication
Jellish et al (1996) [7] 122 GA N SA GA N SA SA N GA GA N SA SA = GA
Demeril et al (2003) [3] 60 GA N SA GA N SA SA = GA GA N SA GA = SA
McLain et al (2004) [11] 200 GA N SA GA N SA SA N GA GA N SA GA N SA
McLain et al (2005) [12] 400 GA N SA GA N SA SA N GA GA N SA GA N SA
This series 100 SA N GA GA = SA SA = GA GA N SA GA = SA
N&V indicates nausea and vomiting in the recovery room; bleeding, the amount of intraoperative bleeding; RT, recovery time; AC, analgesic consumption in the
recovery room.
64 S.H. Sadrolsadat et al. / Surgical Neurology 71 (2009) 60–65
studies; however, the patients in the other group did not
receive such an agent. Therefore, administration of the
propofol was one of the factors altering the results of those
studies [7,11,12]. In the present study, propofol was used as
the sole hypnotic and sedative agent in both the GA and SA
groups. (b) Inaccurate calculation of the patients' fluid
requirement was the second confounding factor cited in
those studies. The fluid infused during the operation is
normally assumed by considering the patients' weight,
the duration of anesthesia, the minimum NPO duration, the
amount of the fluid loss, the CVE, the third space, the
insensible water loss, and the bleeding compensation.
According to Jellish et al [7], the amount of infused fluid
was reported to be approximately 800 1100 mL less than
the required amount. They also showed that the average
amount of bleeding and the duration of anesthesia were also
reported to be more in the GA group. In other words, the
required fluid was not completely compensated in the
abovementioned study; thus, the hemodynamic changes
cannot exclusively be attributed to the kind of anesthesia
applied. (c) Regarding the operation time and the amount of
blood loss during operation, both of them look to be
“confounding factors” rather than independent variables
because the surgeons usually turn the patients into the prone
position according to their own will and satisfaction, so no
neurosurgeon is actually blind about the patient allocation
to either GA or SA group! That makes a major bias for the
surgeon. First, the surgeon may take his/her time for a
satisfactory and complete surgery without being worried
about the length of operation being inconvenient for the
patient in the prone position. Second, the longer the
operation, the more will be those minimal changes in
the amount of bleeding during operation.
According to the findings of the present study, each
method has its own preferred characteristics:
1. Considering the hemodynamic stability, the incidence
of bradycardia was equal in both methods.
2. The prevalence of hypotension and, consequently,
the administration of ephedrine were higher in SA;
conversely, hypertension was more frequently
reported after the operation in the patients who had
undergone GA.
3. The incidence of tachycardia was higher in GA,
occurring mainly during the intubation period. For-
tunately, none of the 100 patients under study reported
any complication due to the hemodynamic changes.
4. The average intraoperative bleeding and need for fluid
were not significantly different in the 2 groups.
5. In the recovery room, the number of patients with pain
who needed analgesics was less in the SA group; this
could be due to the anesthetic agent's effect.
6. On the other hand, nausea and vomiting were more
prevalent among the patients recovering from SA
because of the unopposed parasympathetic activity
occurring with this method [2].
7. It should also be kept in mind that the surgeons are not
able to evaluate the patients' neurologic status
immediately after the operation in the SA because of
the motor block state that happens during this type of
anesthesia. That is one of the major drawbacks of SA
for disk surgery. It is noteworthy that very few studies
in this field have evaluated the surgeons' satisfaction
rate not only because there may be no standard scaling
technique for it but because close cooperation between
the anesthesiologist and the attending surgeon is
necessary to develop a reliable VAS.
Along with these concerns, there are some other
important points that reduce the anesthesiologists' prefe-
rence for SA, such as:
1. Although anesthesiologists are interested in SA as a
more reliable method, experience shows the prolonged
operation performed in the prone position under SA
increases the anesthesiologist's stress and anxiety.
2. Heightening of the blocked level leading to apnea is
another possible worrying complication using this
method, compared with GA anesthesia.
3. Managing an apneic patient, providing an airway
access, and placing an endotracheal tube in the prone
position are difficult.
4. In addition, the infusion of the anesthetic agent may be
accompanied with amnesia; any decrease in the
sedation depth (which is common) is accompanied
with involuntary, sudden movements in the patient's
upper body, resulting in possible nerve root injuries
[9,21]. Cough and hiccups are also recurrently
reported during this type of anesthesia [3,16].
According to the abovementioned results, it could be
concluded that GA causes less intraoperative bleeding
accompanied by a higher hemodynamic stability; it also
provides a relatively higher satisfactory condition for the
surgeon and the patient. It should be noted that an increased
risk of bleeding, nausea, vomiting, and hemodynamic
instability is reported after a typical GA sedation using
inhaled anesthetics, although the present study revealed that
the administration of propofol via TIVA in these operations
reduces the incidence of such adverse effects.
5. Conclusion
The findings of the present study revealed that SA has no
acceptable advantage over GA in lumbar disk surgery.
Moreover, it could support the hypothesis that GA
can reduce the related risks and complications in several
aspects and rule out the first hypothesis suggested in
the introduction.
However, further well-designed studies including more
sample cases are still needed to evaluate the accuracy of the
abovementioned findings.
 S.H. Sadrolsadat et al. / Surgical Neurology 71 (2009) 60–65
Acknowledgment
We are indebted to the Research and Development Center
of Sina Hospital for their support.
References
[1] Backofen JE, Schauble JR. Hemodynamic changes with prone
positioning during general anaesthesia. Anesth Analg 1985;64:194.
[2] Brown DL. Spinal, epidural and caudal anesthesia. Miller's anesthesia. 6th
ed. UK: Elsevier Churchill Livingstone; 2005. p. 1654-83. Chapter 43.
[3] Demeril CB, Kalayci M, Ozkocak I, et al. A prospective randomized
study comparing peri-operative outcome variables after epidural of
general anesthesia for lumbar disc surgery. J Neurosurg Anesthesiol
2003;15(3):158-92.
[4] Dripps RD, Vandam LD. Long term follow up of patients who received
10098 spinal anesthetics: failure to discover major neurological
sequelae. JAMA 1954;156:1486-91.
[5] Hassi N, Badaoui R, Cagny Bellet A, et al. Spinal anesthesia for disc
herniation and lumbar laminectomy: apropos of 77 cases. Cah
Anesthesiol 1995;43:21-5.
[6] Jellish WS, Shea JF. Spinal anaesthesia for spinal surgery 2003;17:
332-4.
[7] Jellish WS, Thalji Z, Stevenson K, Shea J. A prospective randomized
study comparing short and intermediate term perioperative outcome
variables after spinal or general anesthesia for lumbar disk and
laminectomy surgery. Anesth Analg 1996;83:559-64.
[8] Kakiuchi M. Reduction of blood loss during spinal surgery by epidural
blockade under normotensive general anesthesia. Spine 1997;22:
889-94.
[9] Kennedy F, Efron AS, Perry G. The grave spinal cord paralysis caused
by spinal anesthesia. Surg Gynecol Obstet 1950;91:385-97.
[10] McCollum JS, Milligan KR, Dundee JW. The antiemetic action of
propofol. Anesthesia 1988;43:239-40.
[11] Mclain R, Bell GR, Kalfas I, et al. Complications associated with
lumbar laminectomy: a comparison of spinal versus general anesthesia.
Spine 2004;29:2542-7.
[12] McLain RF, Kalfas I, Bell GR, et al. Comparison of spinal and general
anesthesia in lumbar laminectomy surgery: a case controlled analysis
of 400 patients. J Neurosurg Spine 2005;2:17-22.
[13] Nazon D, Abergel G, Hatem CM. Critical care in orthopedic and spine
surgery. Crit Care Clin 2003;19:33-53.
[14] Phillips OC, Ebner H, Nelson AT. Neurological complications
following spinal anesthesia with lidocaine: a prospective review of
10440 cases. Anesthesiology 1969;30:284-9.
65
[15] Reves JG, Glass PSA, Lubrasky DA, McEvoy MD. Intravenous non-
opioid anesthetics. Miller's anesthesia. 6th Ed. UK: Elsevier Churchill
Livingstone; 2005. p. 317-78. Chapter 10.
[16] Riding JE. Minor complications of general anaesthesia. BJA 197;47:
91-101.
[17] Riegel B, Alibert F, Becq MC, et al. Lumbar disc herniation with
surgical option: general versus local anesthesia. Agressologie 1994;34:
33-7.
[18] Scott NB, Kehlet H. Regional anesthesia and surgical morbidity. Br J
Surg 1988;75:299-304.
[19] Silver DJ, Dunsmore RH, Dickson CM. Spinal anesthesia for lumbar
disc surgery: review of 576 operations. Anesth Analg 1976;55:
550-4.
[20] Tetzlaff JE, Dilger JA, Kodsy M, et al. Spinal anesthesia for elective
lumbar spine surgery. J Clin Anesth 1998;10:666-9.
[21] Thorsen G. Neurological complications after spinal anesthesia and
results from 2493 follow up cases. Acta Chir Scand Suppl 1947;121:
1-272.
Commentary
How far we have come since the advent of general
anesthesia. The complications of ether anesthesia, in part at
least, led to the development of regional anesthesia
techniques as pioneered in the United States by John
Bonica (http://www.painresearch.utah.edu/crc/CRCpage/
Bonica.html), especially for the pain of childbirth. Never-
theless, improvements in the techniques of general
anesthesia have proven the method as equal to or superior
to regional anesthesia as is demonstrated in the present
study. Surgeons are clearly more comfortable operating
on a paralyzed asleep patient as this study also confirmed.
The authors are to be commended for an appropriate and
useful study.
Ron Pawl, MD
Center for Pain Treatment and Rehabilitation
Lake Forest Hospital
Lake Forest, IL 60045, USA