Full text of DOI:10.1007/BF03019864

REPORTS OF INVESTIGATION
1171
Addition of dextrose
3.5% to intrathecal sufen-
tanil for labour analgesia
reduces pruritus
Amr E. Abouleish MD MBA,*
Dmitry Portnoy MD,*
Ezzat I. Abouleish MD†
Purpose: To determine whether the addition of a low concentration (3.5%) of dextrose would minimize pruri-
tus while maintaining the quality of analgesia.
Methods: In a double blind study 48 parturients in early labour were randomized to one of two study groups:
dextrose (Dex, n = 24; 10 µg sufentanil in dextrose 3.5%), or normal saline (NS, n = 24; 10 µg sufentanil in
normal saline). Parturients received the study drug as the intrathecal injection of the combined spinal-epidural
(CSE) technique for labour analgesia. Duration and degree of analgesia were measured until epidural analgesia
was initiated or delivery of the baby. The intensity and distribution (above T6, T -L1, and below L ) of pruritus
6
1
were measured at five minute intervals during first 25 min after injection.
Results: Quality and duration of analgesia did not differ between groups, but the overall incidence of pruritus was
less in the Dex group (88% vs 42%, P = 0.001). Within each region, the incidence of pruritus was less in the
Dex group. In patients who had pruritus, for the Dex group, the incidence of pruritus in the upper region (>T6)
was lower than the NS group. There was no difference in the lower regions.
Conclusion: The addition of dextrose 3.5% to intrathecal sufentanil reduced the incidence of pruritus without
affecting the duration or quality of analgesia in parturients in early labour. The distribution of pruritus in the Dex
group was limited to below T6 suggesting that pruritus to intrathecal sufentanil is mediated at the spinal level.
Objectif : Vérifier si l’ajout d’une faible concentration (3,5 %) de dextrose réduit le prurit pendant le maintien
de l’analgésie.
Méthode : Lors d’une étude à double insu, 48 parturientes en début de travail obstétrical ont été réparties en
deux groupes : dextrose (Dex, n = 24; 10 µg de sufentanil dans du dextrose à 3,5 %) et soluté physiologique
(SP, n = 24; 10 µg de sufentanil dans un soluté physiologique). Elles ont reçu le médicament étudié en injection
intrathécale comme composante d’une analgésie rachidienne et péridurale combinée (RPC). La durée et le degré
de l’analgésie ont été mesurés jusqu’à ce que l’analgésie péridurale soit commencée ou jusqu’à la naissance du
bébé. L’intensité et la distribution (au-dessus de T6, T6-L1, et sous L1) du prurit ont été mesurées à cinq minutes
d’intervalle pendant les 25 premières min après l’injection.
Résultats : La qualité et la durée de l’analgésie n’ont pas présenté de différence intergroupe, mais l’incidence
globale de prurit a été plus faible dans le groupe Dex (88 % vs 42 %, P = 0,001). Pour chaque zone, l’incidence
de prurit a été moindre dans le groupe Dex. L’incidence de prurit dans la zone supérieure (> T6) a été plus faible
chez les patientes du groupe Dex que chez celles du groupe SP. Le prurit de la zone inférieure ne présentait pas
de différence intergroupe.
Conclusion : L’addition de dextrose à 3,5 % à du sufentanil intrathécal a réduit l’incidence de prurit sans chan-
ger la durée ou la qualité de l’analgésie chez des parturientes en début de travail. La distribution du prurit chez les
patientes du groupe Dex s’est limitée à la zone sous T6, ce qui laisse croire à l’existence d’une médiation médul-
laire du prurit relié au sufentanil intrathécal.
From the Department of Anesthesiology,* The University of Texas Medical Branch, Galveston, Texas and the Departments of
Anesthesiology and Obstetrics and Gynecology,† The University of Texas Medical School at Houston, Houston, Texas.
Address correspondence to: Amr E. Abouleish MD MBA, Department of Anesthesiology, The University of Texas Medical Branch, 301
University Boulevard, Galveston, Texas 77555-0591, USA. Phone: 409-772-1221; Fax: 409-772-1224; E-mail: aaboulei@utmb.edu
Sources of funding: Department of Anesthesiology, The University of Texas Medical Branch, Galveston, Texas.
Presented at the following meetings: American Society of Anesthesiology, Annual meeting, Orlando, Florida, USA, October, 1998 and
Society of Obstetric Anesthesia and Perinatology, Annual meeting, Vancouver, B.C., Canada, May, 1998.
Accepted for publication August 17, 2000.
CAN J ANESTH 2000 / 47: 12 / pp 1171–1175

1172
CANADIAN JOURNAL OF ANESTHESIA
NTRATHECAL opioids have been used suc-
cessfully in providing excellent analgesia for
labour.1, 2 In the last decade, the popularity of
the combined spinal-epidural (CSE) technique
made by taking 2 ml of the following 10 ml solution:
1 ml 50 µg·ml–1 sufentanil, and 9 ml normal saline.
Each parturient received CSE labour analgesia. A
fluid bolus of 500 ml lactated Ringer’s solution was
administered. With the patient in the sitting position,
the L2-3 or L3-4 interspace was identified. The epidural
space was identified with loss of resistance to air with
a 9-cm, 18-ga Tuohy needle. A 12-cm, 27-ga
Whitacre needle was then passed through the epidur-
al needle into the subarachnoid space (both needles
were manufactured by Becton Dickinson Company,
Franklin Lakes, New Jersey). After seeing free-flowing
cerebrospinal fluid, 2 ml of the study solution were
injected and the spinal needle was removed. A 20-ga
epidural catheter was then inserted 4 cm into the
epidural space. The patient was positioned in a semi-
recumbent position (head of the bed 20°–40°) and in
left uterine displacement.
The parturients continued to labour and epidural
analgesia was only administered if the patient request-
ed additional pain medication. Duration of action for
the study drug was defined as the time from intrathe-
cal injection to request for epidural analgesia and was
only calculated for those patients who requested
epidural analgesia. If a patient delivered without
requesting epidural analgesia, they delivered while the
spinal sufentanil was still working, and duration of
action could not be calculated.
Measurements of the degree of analgesia, and the
intensity and spread of pruritus were taken at the time
of injection, three minutes after injection, and every five
minutes thereafter up to 25 min after injection.
Additional measurements were noted when patients
requested additional analgesic medications (if applica-
ble) and when the baby was delivered. A visual analog
score (VAS) of pain (0 = no pain, 10 = worst pain), and
a pruritus score were recorded. The intensity of pruri-
tus was measured on a 0 to 3 scale, where 0 = no itch-
ing, 1 = itching only when questioned, 2 = complaining
of itching, and 3 = medication requested to relieve itch-
ing. Pruritus was considered clinically important if the
maximum measured score was 2 or 3. Further, the dis-
tribution of pruritus was recorded for specific anatomic
regions, including the chest, neck, or face (above T6),
abdomen or lower back and buttocks (T6-L1), and legs
and perineum (below L1).
I
has led to increased use of intrathecal sufentanil or
fentanyl to provide analgesia for labour. Although
CSE with sufentanil is associated with fast onset and
excellent analgesia, the limited duration of action and
the high incidence of pruritus have limited its wide
acceptance as standard of care.
Drugs, including bupivacaine, have been added to
intrathecal sufentanil to improve the duration of anal-
gesia.3–5 Similarly, adding dextrose to make a hyper-
baric solution has been attempted to control pruritus
by limiting the spread of sufentanil.6,7 A spinal-level
interaction producing pruritus is consistent with the
onset and duration of pruritus.8 Because normal saline
solution is slightly hypobaric, and because parturients
usually stay in a head-up position following induction
of regional analgesia for labour, the intrathecal injec-
tion of sufentanil in saline tends to spread cephalad.9,10
The addition of dextrose 7.5% or 5% to sufentanil
resulted in reduced spread of pruritus in previous
studies,6,7 but it also reduced the duration of analge-
sia. We hypothesize that too much dextrose adversely
affected the spread of sufentanil and that a lower con-
centration of dextrose would minimize pruritus but
maintain the quality of analgesia. Thus, in a prospec-
tive, double-blind study, we examined the effect of a
lower concentration (3.5%) of dextrose added to
intrathecal sufentanil on the duration and quality of
analgesia as well as the incidence, intensity, and distri-
bution of pruritus.
Materials and methods
After receiving Institutional Review Board approval of
the study protocol and informed consent from the
participants, 48 parturients were enrolled in the study.
The study size was estimated by power analysis, and
the study population consisted of healthy parturients
with singleton gestations in early labour (cervical dila-
tion < 6 cm) requesting regional analgesia for labour
and delivery. Patients with preeclampsia, diabetes, or
other serious medical diseases were excluded.
The participants were assigned to one of two
groups in a double-blind, randomized study: dextrose
(Dex) or normal saline (NS). The study solutions were
made by an anesthesiologist who was not the data-col-
lector. For the Dex group, a solution of 10 µg sufen-
tanil in dextrose 3.5% was made by taking 2 ml of the
following 10 ml solution: 1 ml 50 µg·ml–1 sufentanil,
3.5 ml dextrose 10%, and 5.5 ml normal saline. For
the NS group, 10 µg sufentanil in normal saline was
Statistical analysis
A two-sample t test was used to compare continuous
data (age, height, weight, cervical dilation, and dura-
tion of analgesia) with serial data (VAS for pain). The
Fischer Exact test was used to compare the proportion
data for the incidence of nulliparity with pruritus

Abouleish et al.: INTRATHECAL SUFENTANIL
1173
between the two groups. The Fischer Exact test was
also used to compare proportion data for the severity
and distribution of pruritus between those patients
who reported pruritus in each group. P-values < 0.05
were considered statistically significant.
Because the overall incidence of pruritus was differ-
ent, the incidence of clinically important pruritus and
the distribution of pruritus were determined and com-
pared only in the subgroup of patients who experi-
enced pruritus rather than in the entire study
population. The incidence of clinically important pru-
ritus, defined as a score of 2 or 3, among patients who
had pruritus was lower in the Dex group (NS, 19 of
21; Dex, 5 of 10; P = 0.012) (Figure 2).
In addition, there was a difference between the two
groups in the distribution of pruritus (Figure 3). For
the above T6 region (chest, neck, face), the Dex group
(2 of 10) showed less pruritus than the NS group (19
of 21) (P = 0.005). In contrast, there were no differ-
ences in the T6-L1 and below L1 regions.
Results
Forty-eight parturients (24 in each study group) were
enrolled and received combined spinal epidural labour
analgesia. There were no differences in age, height,
weight, percentage nulliparous, or cervical dilation
between the two groups (Table I).
Quality and duration of analgesia did not differ
between the two groups (Table II). In both groups, all
parturients had excellent analgesia with no difference
in minimum VAS for pain between the groups. There
was no difference in the number of patients who deliv-
ered under the study medication alone (NS, n = 8;
Dex, n = 10). The duration of action did not differ
between the groups.
Discussion
Pruritus occurs in 80% to 95% of parturients who
receive spinal sufentanil for labour analgesia.1,2 Limiting
pruritus would help improve patient satisfaction with
intrathecal sufentanil and the CSE technique used for
labour analgesia. Because both analgesia and pruritus
have a similar onset and duration, the interaction pro-
ducing pruritus appears to be spinally mediated.8
When designing our study, we performed a pilot
study with sufentanil in 5% and 7.5% dextrose, as well
as in the lower concentration of dextrose (3.5%).
Similar to previous studies, we found that the duration
of analgesia appeared to be reduced in the more
hyperbaric solutions (i.e., 5% and 7.5% dextrose). In
contrast, we found that analgesia did not appear to be
reduced in the 3.5% solution (unpublished data).
Hence, we performed this prospective, randomized,
double-blinded study.
The incidence of pruritus, defined as any score other
than 0 (no itching), was 88% (21 of 24 parturients) in
the NS group, and 42% (10 of 24) in the Dex group (
P
= 0.001) (Figure 1). In addition, within each anatomic
region, the NS group had a higher incidence of pruritus.
TABLE I Demographics.
Demographics
NS group
(n=24)
Mean
Dex group
(n=24)
Mean
P
Age (yr)
23.5 5.2
82 21.7
163 4.7
58%
25.3 6.3
81 16.1
162 7.8
75%
0.21
0.79
0.35
0.22
0.33
Weight (kg)
Height (cm)
% Nulliparous*
Cervical Dilation
In contrast to previous studies, which studied high-
er concentrations, we found that the duration of anal-
gesia was not reduced in the hyperbaric 3.5% solution
compared with the normal saline solution. Because
4.6 1.2
4.3 1.3
NS = Normal Saline; Dex = Dextrose. Mean SD.
* Percentages and P-value reported from Fischer’s Exact test.
TABLE II Quality and duration of analgesia.
Demographics
NS group
(n=24)
Dex group
(n=24)
P
Mean SD
Mean SD
Minimum VAS
Maximum percent decrease VAS
Time to delivery with the addition of epidural analgesia
Time to delivery without the addition of epidural analgesia
Duration of action (min)
0.1 0.3
0.2 0.6
0.55
0.58
0.58
0.87
0.55
0.07
98.6% 3.8
344 116.0
110 43.9
112 29.5
33.3% (n/a)
97.7% 6.6
315 152.0
113 35.7
131 26.4
41.7% (n/a)
Percent delivered without epidural analgesia*
NS = normal saline; Dex = dextrose; VAS = visual analog scale.
* For percent delivered without epidural analgesia, no Mean or SD was reported, in which case the Fischer Exact test was performed to
obtain a percentage and a P-value.

1174
CANADIAN JOURNAL OF ANESTHESIA
FIGURE 1 Number of patients with pruritus. The Dextrose
(Dex) group (n = 24) had less pruritus than the normal saline
(NS) group (n = 24) both overall and in each anatomical region
(P = 0.001). Patients may have had pruritus in more than one
FIGURE 3 Distribution of pruritus: comparison of percentage
among patients who had any pruritus. When comparing the distri-
bution of pruritus among women in both groups who reported
experiencing pruritus, patients in the Dextrose (Dex) group had
anatomic region. above T = chest, neck, or face; T-L1 =
significantly less incidence of pruritus in the above T region than
6
6
abdomen and lower back;6below L1 = buttocks, legs, and per-
the patients in the normal saline (NS) group (P = 0.05). above T
6
ineum.
= chest, neck, or face; T- L = abdomen and lower back; below L1
6
1
= buttocks, legs, and perineum.
of pruritus is patient dissatisfaction, the degree of pru-
ritus causing this dissatisfaction was identified in
patients by complaints or requests for medication (a
score of 2 or 3). In addition, because we wanted to
examine the overall incidence of pruritus as well as the
incidence of clinically significant pruritus, we also
included a score for pruritus where the patient did not
complain (a score of 1). If no pruritus was present, a
score of 0 was recorded.
The overall incidence of pruritus was markedly dif-
ferent between the two groups (Figure 1) . In addi-
tion, when comparing the whole population, the
differences at each region are significant (Figure 1). A
more accurate comparison is to examine the distribu-
tion of pruritus in patients who had pruritus. When
this was done, we still found a difference, but only in
the upper region (Figure 3). Because there was no dif-
ference in the lower regions, the addition of dextrose
only limited cephalad spread of pruritus. Similarly,
pinprick levels were reduced in the Dex group within
the upper region. The combination of these two
observations provides further evidence that pruritus
interaction occurs at a spinal level.
FIGURE 2 Incidence of clinically significant pruritus: compari-
son of incidence among patients who had any pruritus. When
comparing the severity of pruritus among women in both groups
who reported experiencing pruritus, patients in the Dextrose
(Dex) group had significantly less incidence of severe pruritus than
patients in the normal saline (NS) group (P = 0.012). Patients
with clinically significant pruritus were defined as having a maxi-
mum pruritus score of 2 or 3.
the 3.5% solution was less hyperbaric than the 5% or
7.5% solution, sufentanil distribution was not expect-
ed to be as limited. This may account for the preser-
vation of the analgesia duration.
On the other hand, similar to previous studies, the
incidence and severity of pruritus was decreased. For
severity, we chose not to use a VAS, but rather a clin-
ically oriented, 0 to 3 scale. Since the major drawback
The severity of pruritus was also decreased. Similar
to the study of distribution, the comparison was made
only in the subgroup of patients who experienced pru-
ritus. When a patient in the NS group had pruritus, the
patient complained or requested medication almost all
the time. Conversely, only half of the patients who
experienced pruritus in the Dex group complained or
requested medications.

Abouleish et al.: INTRATHECAL SUFENTANIL
1175
9 Richardson MG, Wissler RN. Densities of dextrose-free
intrathecal local anesthetics, opioids, and combinations
measured at 37°C. Anesth Analg 1997; 84: 95–9.
10 Parlow JL, Money P, Chan PSL, Raymond J, Milne B.
Addition of opioids alters the density and spread of
intrathecal local anesthetics? An in vitro study. Can J
Anesth 1999; 46: 66–70.
Finally, since the use of ephedrine was not con-
trolled or measured in this study, we cannot comment
on whether the limited spread of intrathecal sufentanil
reduced the incidence of hypotension. In addition,
although no fetal heart rate changes were noted in the
study, we did not specifically examine this issue. Both
of these clinical questions should be examined in
future studies.
In this population, creating a slightly hyperbaric
solution for intrathecal sufentanil limited the inci-
dence, severity, and spread of pruritus without affect-
ing the analgesic properties of the sufentanil. By
limiting pruritus, patient satisfaction may improve,
and practitioner acceptance of the CSE technique for
labour analgesia may increase. Further, the reduction
in the distribution of pruritus and the similar reduc-
tion in the pinprick level of the hyperbaric group add
further evidence that pruritus, from intrathecal opi-
oids, is spinally mediated.
Acknowledgments
The authors wish to thank Christy Perry and Jordan
Kicklighter in the editorial office of the Department of
Anesthesiology at UTMB for their editing and prepa-
ration of this manuscript.
References
1 Cohen SE, Cherry CM, Holbrook RH Jr, El-Sayed YY,
Gibson RN, Jaffe RA. Intrathecal sufentanil for labor
analgesia—sensory changes, side effects, and fetal heart
rate changes. Anesth Analg 1993; 77: 1155–60.
2 Camann WR, Minzter BH, Denney RA, Datta S.
Intrathecal sufentanil for labor analgesia. Effects of
added epinephrine. Anesthesiology 1993; 78: 870–4.
3 Grieco WM, Norris MC, Leighton BL, et al. Intrathecal
sufentanil labor analgesia: the effects of adding mor-
phine or epinephrine. Anesth Analg 1993; 77:
1149–54.
4 Abouleish A, Abouleish E, Camann W. Combined
spinal-epidural analgesia in advanced labour. Can J
Anaesth 1994; 41: 575–8.
5 Campbell DC, Camann WR, Datta S. The addition of
bupivacaine to intrathecal sufentanil for labor analgesia.
Anesth Analg 1995; 81: 305–9.
6 Ferouz F, Norris MC, Arkoosh VA, Leighton BL, Boxer
LM, Corba RJ. Baricity, needle direction, and intrathe-
cal sufentanil labor analgesia. Anesthesiology 1997; 86:
592–8.
7 Gage JC, D’Angelo R, Miller R, Eisenach JC. Does
dextrose affect analgesia or the side effects of intrathe-
cal sufentanil? Anesth Analg 1997; 85: 826–30.
8 Ballantyne JC, Loach AB, Carr DB. Itching after
epidural and spinal opiates. Pain 1988; 33: 149–60.