Anesthesiology 2003; 98:511–5 © 2003 American Society of Anesthesiologists, Inc. Lippincott Williams & Wilkins, Inc.

Spinal Anesthesia
Functional Balance Is Impaired after Clinical Recovery
Charles O. Imarengiaye, F.W.A.C.S.,* Dajun Song, M.D., Ph.D.,† Atul J. Prabhu, F.R.C.A.,* Frances Chung, F.R.C.P.C.‡

Background: The ability of patients to walk without assis- predict full recovery. Moreover, the underlying assump-
tance after spinal anesthesia is a determining factor in the time tion is that resumption of motor function signifies am-
to discharge following ambulatory surgery. The authors com-
pared clinical markers of gross motor recovery with objective
bulatory readiness, which may not be accurate.
data of functional balance after spinal anesthesia. Balance Master (NeuroCom International Inc., Clacka-
Methods: Twenty-two male patients with American Society of mas, OR), a computerized force platform, has been re-
Anesthesiology physical status I or II who were scheduled for

Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/98/2/511/336160/0000542-200302000-00033.pdf by guest on 07 September 2020

ported to be useful in objective assessment of postoper-
perineal surgery were studied during recovery from spinal an- ative balance function following general anesthesia and
esthesia to compare the predictive accuracy of clinical markers
of ambulatory readiness (e.g., full knee flexion and extension)
sedation.3– 6 We hypothesized that functional balance
with that of an objective method of measurement focused on (i.e., balance when ambulating) parameters measured
functional balance. Lumbar puncture was performed at the using Balance Master could directly identify ambulatory
L2–L3 or L3–L4 interspace using a 25-gauge Whitacre needle, ability of a patient following ambulatory procedures un-
with patients in the sitting position. A 3-ml mixture of 5 mg der spinal anesthesia and would not correlate with clin-
bupivacaine (heavy) and 10 ␮g fentanyl was injected. Block
regression and restoration of motor function were assessed and
ical markers indicating functional motor recovery (e.g.,
recorded. Functional balance was measured using a computer- the deep knee bend, etc.). The common clinical markers
ized force platform method. would not be reliably predictive for a patient’s time to
Results: The majority of patients maintained motor function ambulation or discharge.
and proprioception sensation at the onset of surgical anesthe- Lidocaine used in spinal anesthesia has been reported
sia, as indicated by performance on clinical tests of function:
96% were able to perform the straight leg increase; 82, 77, and
to cause postoperative transient neurologic symptoms in
91%, respectively, were able to perform full knee flexion and 0 – 40% of patients, which is unrelated to anesthetic
extension, perform heel-to-shin maneuvers, and identify joint concentration and baricity.7–11 To avoid risk of such
position in the supine position. Postoperatively, clinical return complications, we chose low-dose, saline-diluted bupiv-
of motor function occurred much earlier than recovery of func- acaine combined with fentanyl for spinal anesthesia in
tional balance. At 60 min after onset of spinal anesthesia, 22
patients (100%) had recovered sensory and gross motor func-
our study.12
tion, but only 36% could stand, and 8% could walk without
assistance (P < 0.01). At 150 –180 min after onset, 96 –100% of
patients achieved the levels of functional balance that permitted Materials and Methods
adequate ambulation.
Conclusions: The results suggest that the recovery time to
Following institutional review board approval (Univer-
unassisted ambulation is longer than has been assumed, and
that the standard clinical markers of gross motor function are sity of Toronto, Toronto, Ontario, Canada), 22 male
poor predictors of functional balance following ambulatory outpatients aged 18 – 65 yr with American Society of
surgery. Anesthesiologists physical status I or II who were sched-
uled for elective ambulatory perineal surgery were asked
RECENTLY, it was suggested that spinal anesthesia with to participate in this prospective study. Informed con-
lidocaine allows patients to walk from the operating sent was obtained. Patients with a history of allergy to
room following ambulatory surgery.1,2 However, all mea- the study medications, previous or current psychiatric
sures of ambulatory readiness in these cases were clini- illness, medical conditions affecting balance and coordi-
cal, i.e., the Romberg test and patient performance of nation, neurologic or vestibular disease, or morbid obe-
various maneuvers indicative of adequate motor func- sity were excluded from study.
tion, including the straight leg increase, deep knee bend,
and heel-to-shin touch. Evaluation of these indicators is Anesthetic Technique
subjective, potentially resulting in variable definitions of In the operating room, intravenous access was estab-
return of motor function, which may make it difficult to lished in one of the forearms. Following initial vital signs,
the patients were placed in the sitting position for ad-
* Clinical Fellow, † Research Fellow, ‡ Professor of Anesthesia. ministration of spinal anesthesia. The L2–L3 or L3–L4
Received from the Department of Anesthesia, Toronto Western Hospital, intervertebral space was then infiltrated with 2% lido-
University Health Network, University of Toronto, Toronto, Ontario, Canada. caine, and the subarachnoid space was identified via the
Submitted for publication June 19, 2002. Accepted for publication October 9,
2002. Support was provided solely from departmental sources. midline using a 25-gauge Whitacre needle. On reflux of
Address reprint requests to Dr. Chung: Department of Anesthesia, Toronto clear cerebrospinal fluid, a 3-ml mixture of 5 mg heavy
Western Hospital, 399 Bathurst Street, Toronto, Ontario, Canada M5T 2S8.
Address electronic mail to: frances.chung@uhn.on.ca. Individual article reprints
bupivacaine (7.5%), 10 ␮g fentanyl, and 0.9% N saline
may be purchased through the Journal Web site, www.anesthesiology.org. was injected slowly into the subarachnoid space. The

Anesthesiology, V 98, No 2, Feb 2003 511

512 IMARENGIAYE ET AL.

patient was then placed in the supine position to achieve 3. Tandem walk: Patients were asked to walk, heel to
sensory blockade to T10 –T12 within 10 min. Analgesia toe, from one end of the force plate to the other, to
was supplemented intraoperatively (on patient request) determine individual step width, speed, and end point
by intravenous administration of 25–50 ␮g fentanyl. center of gravity sway (degrees per second).
All balance tests were performed 30– 60 min before
anesthesia (with instruction and proper practice), 60 min
Clinical Assessment of Sensory and Motor Function
after spinal injection (first postoperative assessment), and
Motor function of the lower extremities of each pa-
then at 30-min intervals until patients were discharged
tient was assessed using straight leg raises (measured
home. Prior to each balance test, motor function and
approximately every 15° from 0° to 90° in the supine
ability to ambulate were assessed clinically. Patients did
position), deep knee bends (full knee flexion and exten-
not proceed to postoperative balance evaluation until
sion in the supine position), heel-to-shin maneuvers

Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/98/2/511/336160/0000542-200302000-00033.pdf by guest on 07 September 2020

they were able to perform a 90° leg increase, a deep
(touching the shins with the opposing heels in the su-
knee bend, and a heel-to-shin maneuver and had
pine position), and modified Bromage scores (1 ⫽ un-
achieved modified Bromage scores of 4 or greater. Am-
able to move the feet or knee; 2 ⫽ able to move only the
bulatory readiness and candidacy for discharge were
feet; 3 ⫽ just able to move the knee; 4 ⫽ full flexion of
defined by the patients’ ability to walk steadily without
the knee with weakness; 5 ⫽ full flexion of the knee
assistance.
without weakness). These tests were performed 30 – 60
min before anesthesia (baseline), 5 min after spinal in-
jection (before skin preparation), 60 min after spinal Temporal Measurements
injection (time of first postoperative assessment), and The time to onset of spinal anesthesia, the duration of
every 30 min following the first postoperative assess- anesthesia and surgery, and the times to return of motor
ment, until the patient was discharged home. The level and sensory function, recovery to ambulatory readiness,
of sensory blockade was determined at the same inter- postanesthetic care unit discharge, and discharge home
vals (except baseline) by testing responses to pin-prick were recorded.
stimulation using a 23-gauge needle bilaterally in the
midclavicular lines. Data Analysis
Parametric data obtained from balance and motor func-
tion testing before and after spinal anesthesia were com-
Evaluation of Functional Balance Using Balance pared using the paired t test. Nonparametric data from
Master these tests were analyzed using the chi-square test. A P
The Balance Master system (model 6.1) is a computer- value of less than 0.05 was considered statistically
ized force platform in which the patient’s feet are placed significant.
on two foot plates, each resting on a transducer that
transmits movement-generated signals to a computer.
The computer calculates and tracks the force and move- Results
ment of the patient’s center of gravity and displays the
value on a monitor. The data can also be stored or All 22 male outpatients completed the study, and all of
printed out by the computer. them achieved satisfactory surgical anesthesia to T11–
The functional balance tests chosen for this study were T12. None required supplemental intraoperative analge-
as follows: sia. Table 1 provides patient demographics, types of
1. Sit-to-stand test: Patients were asked to rise quickly surgical procedures, and recovery times.
from a seated to a standing position, during which data Five minutes after spinal injection, the level of sensory
on weight transfer (time of center of gravity moving blockade determined by response to pin-prick stimula-
from sitting to standing position, in seconds), rising tion averaged T11 (table 2). Most patients maintained
index (percentage of body weight exerted to rise [the motor function and proprioception sensation: 96% were
higher the better]), and end sway (center of gravity able to achieve the straight leg increase (averaging a 69°
movement immediately after standing, in degrees per angle lift), 82% were able to perform deep knee bends,
second) were obtained. 77% were able to perform heel-to-shin maneuvers, and
2. Step-up/-over test: Patients were asked to step 91% were able to identify joint positions.
quickly onto a 8-in curb using one foot and swing over At 60 min after spinal injection (first postoperative
and step down with the other foot on the force platform assessment), the level of sensory blockade averaged T12
to allow determination of rising index (percentage of (T3–L3). All patients were able to achieve the straight leg
body weight exerted to rise to the curb), impact index increase (averaging a 79° lift), and the percentage of
(percentage of body weight to step down to the force those able to perform the deep knee bend, perform the
plate), and movement time (from start to end of the heel-to-shin touch, and identify joint position was the
movement, in seconds). same as that at the time of 5 min after spinal injection

Anesthesiology, V 98, No 2, Feb 2003

FUNCTIONAL BALANCE AFTER SPINAL ANESTHESIA 513

Table 1. Patient Demographics, Surgical Procedures, and Table 2. Change of Sensory and Motor Functions
Perioperative Time Variables
Degree/
Demographics Sensory/Motor Function Number Percentage Score

Age, yr 45 ⫾ 12 Leg Raising

Height, cm 171 ⫾ 6 Preanesthesia 22 100 90 ⫾ 0
Weight, kg 76 ⫾ 12 5 min after spinal injection 21 96 69 ⫾ 32*
ASA physical status (I/II), n 17/5 60 min after spinal injection 22 100 79 ⫾ 18*
Surgical procedures, n 90 min after spinal injection 22 100 88 ⫾ 6
Hydrocelectomy 7 Deep knee bend
Fissurectomy 5 Preanesthesia 22 100 NA
Varicocelectomy 4 5 min after spinal injection 18 82 NA
Hemorrhoidectomy 4 60 min after spinal injection 18 82 NA
90 min after spinal injection 22 100 NA

Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/98/2/511/336160/0000542-200302000-00033.pdf by guest on 07 September 2020

Circumcision 1
Orchidopexy 1 Heel–shin touch
Perioperative time variables, min Preanesthesia 22 100 NA
Duration of surgery 23 ⫾ 16 5 min after spinal injection 17 77 NA
Injection to PACU discharge 126 ⫾ 46 60 min after spinal injection 17 77 NA
Injection to voiding 175 ⫾ 47 90 min after spinal injection 22 100 NA
Injection to home discharge 193 ⫾ 46 Joint position
Preanesthesia 22 100 NA
Values are expressed as number or mean ⫾ SD. 5 min after spinal injection 20 91 NA
60 min after spinal injection 21 96 NA
90 min after spinal injection 22 100 NA
Bromage score
(table 2). In contrast, functional balance tests demon- Preanesthesia NA NA 5 (0)
strated that only 36% of patients (8 of 22; P ⬍ 0.01) 5 min after spinal injection NA NA 4 (3–5)
could actually stand, and even fewer, 18% (4 of 22; 60 min after spinal injection NA NA 4 (3–5)
90 min after spinal injection NA NA 5 (4–5)
P ⬍ 0.01), could complete the Balance Master assess-
ment (table 3). Those who completed the Balance Values are expressed as number, percentage, mean ⫾ SD, or median (range).
Master assessment had significantly (P ⬍ 0.05) lower * P ⬍ 0.05 compared to the preanesthesia value.
rising indexes compared to their preanesthesia baselines NA⫽ not applicable.
(table 3).
At 90 min after spinal injection, the level of sensory At 150 min after spinal injection, 96% of patients (21 of
blockade averaged L2 (T3–S2). All patients could achieve 22) had fully recovered the ability to walk steadily with-
the straight leg increase (averaging an 88° lift), and all out assistance, and their balance scores achieved the
could perform deep knee bends, perform heel-to-shin preanesthesia values (within ⫾20% ranges). Only 1 pa-
maneuvers, and identify joint positions. The percentage tient obtained full ambulatory recovery at 180 min. The
able to stand increased to 73% (16 of 22; P ⬎ 0.05), and median level of blockade at the time patients achieved
55% (12 of 22; P ⬍ 0.01) fulfilled functional balance tests preanesthesia balance scores was L2, with a range from
(table 3). T10 to S2.

Table 3. Change of Functional Balance Parameters

Functional Balance Parameter Preanesthesia 60 min After Spinal Injection 90 min After Spinal Injection 120 min After Spinal Injection

Sit to stand, n 22 8* 16* 22

Weight transfer, s 0.63 ⫾ 0.25 1.16 ⫾ 0.69* 0.64 ⫾ 0.39 0.61 ⫾ 0.19
Rising index, % weight 23.9 ⫾ 8.59 16.38 ⫾ 8.77* 17.33 ⫾ 8.11* 19.9 ⫾ 4.16
End sway, degree/s 2.38 ⫾ 0.96 2.70 ⫾ 1.13 2.90 ⫾ 1.57 2.77 ⫾ 1.04
Step up/over, n 22 4* 12* 21
Left leg
Rising index, % weight 31.8 ⫾ 6.7 25.1 ⫾ 5.13* 29.5 ⫾ 6.3 31.38 ⫾ 5.51
Impact index, % weight 33.1 ⫾ 11.3 32.0 ⫾ 12.5 30.9 ⫾ 11.0 32.0 ⫾ 9.85
Movement time, s 1.75 ⫾ 0.36 2.12 ⫾ 0.69 1.96 ⫾ 0.6 1.95 ⫾ 0.50
Right leg
Rising index, % weight 35.4 ⫾ 6.0 28.3 ⫾ 7.20* 32.9 ⫾ 8.10 31.7 ⫾ 8.39
Impact index, % weight 33.6 ⫾ 9.11 27.0 ⫾ 13.8 31.3 ⫾ 11.7 30.4 ⫾ 6.60
Movement time, s 1.67 ⫾ 0.30 2.01 ⫾ 0.82 2.11 ⫾ 0.70 1.95 ⫾ 0.50
Tandem walk, n 22 4* 12* 21
Step width, cm 6.87 ⫾ 2.89 7.49 ⫾ 1.81 7.9 ⫾ 2.69 7.9 ⫾ 1.13
Speed, cm/s 27.8 ⫾ 10.6 27.3 ⫾ 12.1 25.9 ⫾ 6.94 27.0 ⫾ 8.12
End sway, degree/s 3.25 ⫾ 1.48 3.85 ⫾ 2.01 3.51 ⫾ 1.59 3.89 ⫾ 2.24

Values are expressed as number or mean ⫾ SD.

* P ⬍ 0.05 compared to the preanesthesia values.

Anesthesiology, V 98, No 2, Feb 2003

514 IMARENGIAYE ET AL.

Discussion To produce spinal anesthesia, we used a low-dose

(5-mg), dilute solution of bupivacaine combined with 10
An essential component of early patient discharge fol- ␮g fentanyl. Our goal was to produce satisfactory surgi-
lowing outpatient surgery using spinal anesthesia is the cal anesthesia while reducing the likelihood of residual
ability to walk steadily without assistance. To date, out- postoperative motor block, thereby minimizing the time
patient ambulatory readiness has been assumed when to adequate motor function, ambulation, and discharge.
clinical indicators, such as return of motor function and Saline dilution of low-dose spinal bupivacaine appropri-
adequate Romberg test results, are present. Positive per- ately decreased this agent’s characteristically long dura-
formances on tests, such as the straight leg increase and tion of action,12 making it useful in our outpatient time
deep knee bend, have been considered a marker of frame, and use of a small dose enabled some control of
ambulatory capacity and were used to suggest that pa- the dose-dependent magnitude of motor block.13,14 The

Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/98/2/511/336160/0000542-200302000-00033.pdf by guest on 07 September 2020

tients could walk unassisted from the operating room addition of fentanyl improved the quality of the spinal
following spinal anesthesia.1–2 The results of the anesthesia without prolonging recovery.15 We achieved
present, prospective study suggest otherwise. an early return of motor function (60 min after induction
We found a disparity between the time to recovery of of anesthesia), but did not improve recovery time to
motor function and the time to achieve the postural ambulation and discharge beyond that previously re-
control and balance essential for safe ambulation. Walk- ported by other investigators (150 –180 min after
ing balance remained impaired long after (90 –120 min) induction).
clinical criteria for functional recovery from spinal anes- The limitation to the use of the Balance Master as a
thesia were met. Specifically, all patients had clinically clinical tool is that repetitive testing is required to deter-
fully recovered motor function at 90 min after induction mine satisfactory recovery. This process is both time-
of anesthesia, whereas only 55% were able to walk with- consuming and cumbersome. However, the use of such
out assistance and achieved their preanesthesia balance a system as a research tool can provide insight into the
function parameters. problem of functional balance after anesthesia.
The relation among motor function, balance, and pos- In conclusion, we found a disparity between the time
tural stability is complicated. They are determined by the to return of gross motor function and the time to recov-
integration of visual, somatosensory, and vestibular in- ery of functional balance after spinal anesthesia with
puts by the brainstem and cerebellum. Ideally, the ability low-dose bupivacaine. The functional balance remained
to walk should be determined independently of motor impaired long after (90 –120 min) the motor function
function, by testing the several complex components of recovery was judged adequate using clinical indicators.
balance and posture. The Balance Master is such a tool Tests of gross motor function are inadequate as indica-
that could objectively measure certain sensitive balance tors of the ability to ambulate in readiness to discharge.
parameters when patients perform real-life movement. These results suggest that the ability to walk without
The results could provide an immediate and accessible assistance after spinal anesthesia requires a longer recov-
evaluation of patients’ ability to walk after ambulatory ery period than predicted solely by gross motor recov-
surgeries. ery, making its return inadequate as a sole marker of
The literature on objective assessment of postural ambulatory ability and readiness for discharge. Support-
functions after spinal anesthesia is limited. Previous stud- ing this finding is our use of an objective method of
ies with a force platform have consistently showed im- determining postoperative ambulatory ability, i.e., a
pairment of stability and balance after sedation or anes- computerized force platform system that quantifies the
thesia.3– 6 The influences of general anesthesia and spinal patient center of gravity and measures multiple aspects
anesthesia on posture and balance are different. Al- of postural control and balance during repetitive testing.
though there is no residual cortical depression after
spinal anesthesia, maintenance of equilibrium for the
tasks of daily living depends on a well-integrated muscu- References
loskeletal system. In disturbances of equilibrium, there is
1. Vaghadia H, Viskari D, Mitchell GWE, Berrill A: Selective spinal anesthesia
liability to fall. Such circumstances demand prompt cor- for outpatient laparoscopy: I. Characteristics of three hypobaric solutions. Can J
rections of body position, which require adequate mus- Anesth 2001; 48:256 – 60
2. Vaghadia H, Solylo MA, Henderson CL, Mitchell GWE: Selective spinal
cle tone and coordination. It is possible that the ability of anesthesia for outpatient laparoscopy: II. Epinephrine and spinal cord function.
the patients to initiate an appropriate response to dis- Can J Anesth 2001; 48:261– 6
3. Song D, Chung F, Wong J, Yogendran S: The Assessment of postural stability
turbed equilibrium in the immediate postoperative pe- after ambulatory anesthesia: A comparison of desflurane with propofol. Anesth
riod may still be impaired. Thus, ambulation without Analg 2002; 94:60 – 4
4. Korttila K, Ghoneim MM, Jacob L, Lakes RS: Evaluation of instrumental
assistance should still remain a major factor in determin- force platform as a test to measure residual effects of anesthetics. ANESTHESIOLOGY
ing home readiness of the ambulatory surgical patient. 1981; 55:625–30

Anesthesiology, V 98, No 2, Feb 2003

FUNCTIONAL BALANCE AFTER SPINAL ANESTHESIA 515

5. Gupta A, Ledin T, Larsen LE, Odkvist M: Computerized dynamic posturog- does not alter the incidence of transient neurologic symptoms. ANESTHESIOLOGY
raphy: A new method for the evaluation of postural stability following anaesthe- 1999; 90:445–50
sia. Br J Anaesth 1991; 66:667–72 11. Hampl K, Schneider M, Pargger H, Gut J, Drewe J, Drasner K: A similar
6. Hiller A, Pyykko I, Saarnivaara L: Evaluation of postural stability by comput- incidence of transient neurologic symptoms after spinal anaesthesia with 2% and
erised posturography following outpatient paediatric anesthesia: Comparison of 5% lidocaine. Anesth Analg 1996; 83:1051– 4
propofol/alfentanil/N2O anesthesia with thiopental/halothane/N2O anesthesia. 12. Ben-David B, Levin H, Solomon E, Admoni H, Vaida S: Spinal bupivacaine
Acta Anaesthesiol Scand 1993; 37:556 – 61 in ambulatory surgery: The effect of saline dilution. Anesth Analg 1996; 83:
7. Pollock JE, Neal JM, Stephenson CA, Wiley CE: Prospective study of the 716 –20
incidence of transient radicular irritation in patients undergoing spinal anesthe- 13. Liu SS, Ware PD, Allen HW, Neal JM, Pollock JE: Dose response charac-
sia. ANESTHESIOLOGY 1996; 84:1361–7 teristics of spinal bupivacaine in volunteers: Clinical implications for ambulatory
8. Raeder JC: Regional anesthesia in ambulatory surgery. Can J Anesth 2001; anaesthesia. ANESTHESIOLOGY 1996; 85:729 –36
48:R1–5 14. Gentili M, Senlis H, Houssel P, Monnier B, Bonnet F: Single shot spinal
9. Ben-David B, Maryanovsky M, Gurevitch A, Lucyk C, Solosko D, Frakel R, anesthesia with small doses of bupivacaine. Reg Anesth 1997; 22:511– 4
Volpin G, DeMeo PJ: A comparison of minidose lidocaine-fentanyl and conven- 15. Ben-David B, Solomon E, Levin H, Admoni H, Goldik Z: Intrathecal fentanyl
tional-dose lidocaine spinal anesthesia. Anesth Analg 2000; 91:865–70 with small-dose dilute bupivacaine: Better anesthesia without prolonging recov-
10. Pollock JE, Liu SS, Neal JM, Stephenson CA: Dilution of spinal lidocaine ery. Anesth Analg 1997; 85:560 –5

Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/98/2/511/336160/0000542-200302000-00033.pdf by guest on 07 September 2020

Anesthesiology, V 98, No 2, Feb 2003