AANA Journal Course
Update for nurse anesthetists
Epidural analgesia using loss of resistance with air
versus saline: Does it make a difference? Should we
reevaluate our practice?
CDR David Norman, CRNA, MSN, NC, USN
Chesapeake, Virginia
The choice of using air or saline in epidural syringes during
the loss-of-resistance technique, for identifying the epidural
space, has been based largely on personal preference of the
anesthesia provider. A survey of practice in the United King-
dom, thought to be similar to practice in the United States,
revealed that the majority of anesthesia providers use air.
Case reports have appeared in the literature suggesting
that air may be harmful to patients or, at the very least,
impede the onset and quality of epidural analgesia. Two stud-
ies have evaluated air vs saline to determine whether one
may lead to more rapid or better quality epidural analgesia in
Objectives
At the completion of this course, the reader should be
able to:
1. Describe the 2 most common injected sub-
stances used for the loss-of-resistance technique,
including their possible side effects.
2. Identify the possible problems associated with
the use of air as the choice of injected substance.
3. Identify the possible problems associated with
the use of saline as the choice of injected sub-
stance.
4. State why smaller volumes of the injected sub-
stance are preferable.
5. Compare and contrast and choose the injected
substance based on informed scientific rationale.
Introduction
Epidural analgesia is a very popular method of pain
control for laboring parturients. The most common
method for determining entry into the epidural space
is the loss-of-resistance (LOR) technique. During the
* AANA Journal Course No. 23: The American Association of Nurse Anesthetists is accredited as a provider of continuing education in nursing by
the American Nurses Credentialing Center Commission on Accreditation. The AANA Journal course will consist of 6 successive articles, each with
objectives for the reader and sources for additional reading. At the conclusion of the 6-part series, a final examination will be printed in the AANA
Journal. Successful completion will yield the participant 6 CE credits (6 contact hours), code number: 25468, expiration date: July 31, 2004.
5
*6 CE Credits
laboring parturients. Results are mixed. However, they seem
to indicate that the use of saline for the loss-of-resistance
may result in more rapid and satisfactory quality of pain relief
in laboring parturients.
Current anesthesia literature suggests using saline with an
air bubble in the loss-of-resistance syringe. Many anesthesia
training programs continue to teach the use of air, saline, and
saline with an air bubble. Further studies may help to deter-
mine whether there is a scientific or safety basis for using air
vs saline.
Key words: Air, analgesia, epidural, loss of resistance, saline.
LOR technique, 2 to 5 mL of air or saline is drawn into
the syringe, and continuous or intermittent pressure
is applied to the plunger as the epidural needle is
advanced toward the epidural space.1 On entry into
the epidural space, the syringe contents are injected
due to a loss in resistance. Air and saline are the most
widely used substances in the LOR technique
syringe.1 Both are used commonly; the decision
derives mostly from personal preference.1 The pur-
pose of this literature review is to examine our prac-
tice and determine whether both air and saline are
equally safe and efficacious or whether anesthesia
providers may need to consider changing the sub-
stance used in their practice.
The incidence of backache (approximately 30%)
and postdural puncture headaches (up to 50% in par-
turients experiencing an inadvertent dural puncture)
during an epidural insertion, are both considered
common.2,3 Such complications as permanent nerve
injury, systemic toxic effects of drugs, and spinal
hematomas, fortunately, are rare.2,3 However, addi-
AANA Journal/December 2003/Vol. 71, No. 6 449
tional complications cited in the following case
reports are attributed by the authors to the use of air
injected during the LOR technique.
In an early case report, the author attributed a uni-
lateral epidural block in a 64-year-old man undergoing
left-sided inguinal herniorrhaphy to 2 mL of air
injected during the LOR technique.4 In a second case,
using air in the LOR technique during the epidural
insertion in a laboring parturient resulted in inade-
quate analgesia and was cited by the author as a possi-
ble explanation for the patient having unblocked seg-
ments. The presence of peridural air was confirmed by
radiography during the postpartum period.5,6
Another case report describes a 58-year-old man
with acute pancreatitis who was undergoing an
epidural block for pain control.7 The LOR technique
was performed with air. Six attempts at 3 interverte-
bral spaces were required before the successful identi-
fication of the epidural space and placement of the
catheter. Subcutaneous emphysema along the verte-
bral column was confirmed by clinical assessment. A
computed tomography (CT) scan obtained 4 days
after the insertion confirmed the presence of peridural
air bubbles. The authors stated that although there
were no associated symptoms with the subcutaneous
emphysema, it resolved slowly.
One of two patient positions generally is selected
for the insertion of an epidural. The decision to use
the lateral, or sitting, position usually is based on
provider preference. When the sitting position is
selected, it must be noted that air injected into the
subarachnoid space will rise, resulting in a sudden
and often severe headache. Before the invention of CT
scans, pneumoencephalograms were used for radi-
ographic examination of the spine and ventricles of
the brain. During this procedure, air was injected
intentionally into the subarachnoid space at the lum-
bar region, and radiographs were taken during the
ascent of the air into the cerebral ventricles. This cre-
ated a pneumocephalus, which resulted in a severe
headache until the patient was positioned in a head-
down position so the air would flow back through the
subarachnoid space to the lumbar region, where it was
aspirated.
A case report of a 17-year-old parturient receiving
an epidural for labor, using the LOR technique with
air, describes inadvertent dural puncture with 4 mL of
air injected.8 The patient immediately reported a
severe occipital headache. The epidural procedure was
repeated with an LOR using saline at the same inter-
space and was successful. It took 7 days for this
patient’s headache to resolve.
Pneumocephalus was described in a 25-year-old
parturient, “after many attempts” during an LOR tech-
nique using air.9 The epidural was placed for a
450 AANA Journal/December 2003/Vol. 71, No. 6
cesarean section after multiple attempts, and approxi-
mately 20 mL of air were injected during these
attempts. After injection of 16 mL of 0.5% bupiva-
caine with 1:200,000 epinephrine, the patient’s respi-
rations ceased, blood pressure dropped, and the
patient was intubated. The patient remained stu-
porous postoperatively. A CT scan revealed approxi-
mately 25 mL of air in the parietofrontal cerebral cor-
tex. This appeared to resolve within 24 hours, and the
patient was discharged from the intensive care unit
within 72 hours.
Another case report describes a 30-year-old par-
turient at term desiring epidural analgesia.10 The LOR
technique with air was used during the insertion. The
insertion process, with the patient in the sitting posi-
tion, was described as uneventful with the “space
entered easily.” Upon encountering the LOR, 5 mL of
air was injected. The patient described an immediate
sense of severe back pain at the needle site that moved
up the spine, resulting in a severe neck ache and,
finally, a severe bifrontal-temporal headache. The
patient became very distressed (“became uncoopera-
tive and vociferously complaining that she was expe-
riencing the worst headache of her life”), and an
immediate CT scan was obtained to rule out cerebral
bleeding. The CT scan revealed approximately 5 mL
of air in the intracranial subarachnoid space and basal
cisterns. Although no blood or cerebrospinal fluid was
encountered during the LOR technique, the symp-
toms described are classic responses to pneumoen-
cephalography air–related headaches. The patient was
returned to the labor suite, and her analgesic needs
were managed with intramuscular meperidine until
her labor was deemed “failure to progress.” She then
received a general anesthetic for a cesarean section.
Her surgical and anesthetic courses were uneventful.
Her headache resolved fully within 24 hours.
In a case describing the injection of an excessive
amount of air during the LOR technique, the author
attributed the resulting transient paraplegia to the
air.11 In this case, a 52-year-old woman with chronic
back pain was undergoing epidural analgesia. The
insertion required 4 attempts, during which an esti-
mated 40 mL of air was injected. After the injection of
20 mL of 0.25% bupivacaine with 80 mg of methyl-
prednisolone plus 20 mL of normal saline, the needle
was withdrawn. Eight hours after the injection, the
patient complained of persistent left leg numbness
and weakness. A CT scan revealed a large amount of
extradural air. Her symptoms resolved within 24
hours, and a CT scan 72 hours after the injection
revealed that a substantial quantity of the extradural
air had been reabsorbed. The authors thought that her
symptoms were a result of nerve root compression
due to the air. However, it should be noted that 20 mL
of the bupivacaine-methylprednisolone mixture also
was injected into the same space. Since fluid is less
compressible than air, might it also have contributed
to her symptoms?
Nerve root compression also was described when
an epidural pump was used for the management of
cancer pain.12 This case is different from the preced-
ing cases because the air was injected inadvertently
into the narcotic reservoir by the patient during self-
preparation. After a few days, the patient noted
increasing back pain and bilateral hip flexor weak-
ness. A CT scan revealed the presence of peridural air.
The air was aspirated, and the patient experienced
immediate relief.
Neurological damage was described in a case in
which a patient received an epidural anesthetic for
extracorporeal shock wave lithotripsy.13 An LOR tech-
nique with only 2 mL of air was used. Postoperatively,
the patient experienced severe pain in the lower back
and legs for 2 days that delayed her ambulation. She
also experienced tenderness along the lumbar spine
and bilateral limited leg mobility due to pain. A CT
scan revealed air in the right lateral recesses of S1 and
T12. Her pain and limited mobility improved during
the next 6 weeks. At 6 weeks she was able to return to
work but still experienced some backache and limited
motion. The authors felt that the gas gradient between
the air bubbles and the tissue may impede or disturb
the shock waves. They further recommend that air not
be used during the LOR technique during epidural
insertion for extracorporeal shock wave lithotripsy.
An article reviewing the risks of air for the LOR
technique describes one additional concern,14 the
occurrence of a venous air embolism. In addition to
citing all of the aforementioned events (back pain,
headaches, paresthesias, and incomplete blocks)
described in the case reports, the authors state that
because the epidural anatomy is a rich plexus of veins,
the risk of injecting air intravascularly is present.
They cited 2 children and 1 man in whom air was
injected and rapid cardiac deterioration followed. The
authors stated that a patent foramen ovale may lead to
a potentially lethal outcome if a venous air embolism
is produced from the injection of air during the LOR
technique.
Philip15 conducted a study in which she not only
used air during the LOR technique, but also then
injected 10 mL of air after confirmation of entry to the
epidural space. The purpose of this study was to
determine whether the injection of 10 mL of air before
insertion of the epidural catheter might reduce the
incidence of paresthesias encountered during the
insertion of the catheter. She used a sample of 132
parturients, 63 who received 10 mL of air and 69 who
did not. In contrast with the events described in the
preceding case reports, she found that the incidence of
paresthesias reduction was significant (from 49% to
29%) and that there were no complications.
One study found that the use of saline for the LOR
technique resulted in the decrease of anesthetic
effect.16 This dilutional effect on the local anesthetic
injected also resulted in increasing spread of the drug.
Before the mid-to-late 1990s, many anesthesia text-
books described the use of “air or saline” during the
LOR technique. Most did not cite a preference for one
or the other. However, since the late 1990s many of the
leading authors describe the use of saline alone or
saline with an air bubble during the LOR technique.1-3
This transition likely is a result of the many case
reports cited in the literature.
Only a few studies have compared the use of air
and saline during the LOR technique. One study with
a sample of 142 parturients in labor, 71 in each of 2
groups, sought to examine whether there was a differ-
ence between air and saline in the frequency of unac-
ceptable analgesia.1 The investigators found a lower
frequency of inadequate analgesia in the saline group
than in the air group.
The aim of another study examining air vs saline for
the LOR technique was to determine whether there
was any difference in the onset, spread, or character of
the epidural analgesia in laboring parturients.17 The
sample consisted of 50 parturients, 25 in the air group
and 25 in the saline group. The air group experienced
an increased incidence of unblocked segments, but
there was no difference in the spread of analgesia.
A third study examined air vs saline for the inci-
dence of paresthesias during epidural catheter inser-
tion.18 The sample consisted of 67 parturients, 32 in
the air group and 35 in the saline group. The investi-
gators used 10 mL of air or saline during the LOR
technique. No significant difference was found
between the groups in the incidence of paresthesias.
The authors stated that they found no significant dif-
ference in analgesia between groups and concluded
that the decision to use air or saline during the LOR
technique remains a personal choice.
Summary
It is a conservative estimate that during the past 12 to
15 years, hundreds of thousands, if not more than a
million epidurals have been placed. With these num-
bers, the reported cases of complications attributed to
the injected substance are rare on a percentage basis.
In the majority of the case reports cited, the authors
attribute these side effects to air; the symptoms are
transient and rarely life-threatening complications.
The most current anesthesia literature seems to advo-
cate the use of saline rather than air.
Despite reluctance to change practice based on case
AANA Journal/December 2003/Vol. 71, No. 6 451
Table. Textbook chronology

Year Author Quote

19
1987 Shnider et al Loss-of-resistance technique with saline- (or air) filled syringe most commonly
used.
20
1988 Cousins and Bromage We prefer the technique of loss of resistance, using an air-filled syringe.
21
1990 Brown The preferred method of carrying out the loss-of-resistance technique includes
attaching a 3- to 5-mL glass syringe filled with 2 mL saline and a small (0.25
mL) air bubble.
22
1992 Covino and Lambert The most common method to identify the epidural space is the loss of
resistance technique…. Usually, a syringe containing saline or air is attached….
23
1993 Shnider et al Loss-of-resistance technique with saline- (or air) filled syringe most commonly
used….
24
1994 Brown A 3 to 5 mL low resistance glass or plastic syringe is filled with a small (0.25
mL) air bubble.
25
1996 Tetzlaff Whether to inject saline or air with the loss of resistance technique depends on
the preference of the practitioner.
26
1997 Bernards A glass syringe filled with 2-3 mL saline and a small air bubble…
27
1998 Molnar and Pian-Smith A glass or plastic loss-of-resistance syringe containing approximately 3 cc of
air or 3 cc of saline…
28
1998 Cousins and Veering We prefer the technique of loss of resistance, using an air-filled syringe.
29
1999 Brown I prefer that the syringe contain both saline and a small (0.25 mL) compressible
bubble of air.
30
1999 Brown A 3-5 mL glass syringe is filled with 2 mL saline solution and a small (0.25 mL)
air bubble is added.
31
2000 Brown The preferred method of carrying out the loss-of-resistance technique includes
attaching a 3- to 5-mL glass syringe filled with 2 mL saline and a small (0.25
mL) air bubble.
32
2001 Ellis The syringe is filled with 3 to 4 mL of air or 3 mL saline and 1 mL of air to
facilitate compression of the solution.
33
2001 Bernards A glass syringe or a specially designed low resistance plastic syringe is filled
with 2-3 mL saline and a small (0.1-0.3 mL) air bubble.
34
2002 Kleinman A glass syringe filled with approximately 2 mL of fluid or air is attached to the
hub of the needle.
35
2002 Rosen et al Loss-of-resistance technique with saline- (or air-) filled syringe most commonly
used.

reports, the case reports seem to share 3 common
themes. The first is that the side effects reported seem
to be related to the volume of air or saline injected. A
second is that air remains in the peridural area for not
just hours or a day or two as once thought; it often
remains for many days.7,11,13 This was confirmed by
radiography. In light of this fact, relatively few studies
have been conducted to effectively determine whether
there really is an increased risk with air or whether
there exists a clinically significant scientific reason to
advocate the use of one substance over the other. Even
though there are just a few studies reported in the lit-
erature on this subject, it seems that multiple authors
advocate the use of saline rather than air, based pri-
marily on case reports. Finally, the common theory
behind the possibility of air as the cause of inadequate
or slow analgesia is that air actually impedes the
absorption of the local anesthesia molecules.1
A literature review on this topic would be incom-
plete without a review of what the leading anesthesia
textbooks have taught during the past 10 to 15 years.
A chronology with quotes from their description of
the LOR technique is given in the Table.19-35
Clearly, further studies may help determine
whether the use of air or saline leads to a more rapid
onset and better quality of analgesia and a better
safety profile. Before making a premature conclusion
that air or saline is the cause of these complications, it

452 AANA Journal/December 2003/Vol. 71, No. 6

might also be helpful to examine the insertion skills of
the provider and the ease or difficulty of the insertion
procedure.36 These 2 variables and others need to be
studied and excluded before we can be certain that the
injected substance is the cause of the aforementioned
complications.
REFERENCES
1. Beilin Y, Arnold I, Telfeyan C, Bernstein H, Hossain M. Quality of
analgesia when air versus saline is used for identification of the
epidural space in the parturient. Reg Anesth Pain Med. 2000;25:
596-599.
2. Muir H. Epidural misadventures: a review of the risk and compli-
cations associated with epidural anesthesia. 2002:1-5. Available at:
http://www.oyston.com/anaes/local/muir/html. Accessed January
2, 2002.
3. Mehl-Medrona L, Mehl-Madrona M. Medical risks of epidural anes-
thesia during childbirth. 2002:1-21. Available at: http://www.
healing-arts.org/mehl-madrona/mmepidural.htm. Accessed Janu-
ary 2, 2002.
4. Shanks C. Four cases of unilateral epidural analgesia. Br J Anaesth.
1968;40:999-1002.
5. Dalens B, Bazin J, Haberer J. Epidural bubbles as a cause of incom-
plete analgesia during epidural anesthesia. Anesth Analg. 1987;66:
679-683.
6. Boezaart AP, Levendig BJ. Epidural air-filled bubbles and
unblocked segments [letter]. Can J Anaesth. 1989;36:603-604.
7. Veil E, Coussaye J, Bruelle P, Saissi G, Bassoul B, Eledjam J.
Epidural anesthesia: a pitfall due to the technique of the loss of
resistance to air. Reg Anesth. 1991;16:117-119.
8. Kuczkowski K, Benumof JL. Images in anesthesia: headache
caused by pneumocephalus following inadvertent dural puncture
during epidural space identification: is it time to abandon the loss
of resistance to air technique [letter]? Can J Anaesth. 2003;50:159-
160.
9. Katz Y, Markovits R, Rosenberg B. Pneumocephalos after inadver-
tent intrathecal air injection during epidural block. Anesthesiology.
1990;73:1277-1279.
10. Kreitzer J, Reed A, Dauro A, Brodman M, Bronster D. Ascending
back pain and headache during attempted epidural placement. J
Clin Anesth. 1991;3:414-417.
11. Nay P, Milaszkiewicz R, Jothilingam S. Extradural air as a cause of
paraplegia following lumbar analgesia. Anaesthesia. 1993;48:402-
404.
12. Kennedy T, Ullman D, Harte F, Saberski L Greenhouse B. Lumbar
root compression secondary to epidural air. Anesth Analg. 1988;67:
1184-1186.
13. Deam R, Scott D. Neurological damage resulting from extracorpo-
real shock wave lithotripsy when air is used to locate the epidural
space. Anaesth Intensive Care. 1993;21:455-457.
14. Saberski LR, Kondamuri S, Osinubi OY. Identification of the
epidural space: is loss of resistance to air a safe technique? a review
of the complications related to the use of air. Reg Anesth. 1997;22:
3-15.
15. Philip B. Effect of epidural air injection on catheter complications.
Reg Anesth. 1985;10:21-23.
16. Iwama H. Injection volume of saline with loss of resistance
method may affect the spread of epidural anesthesia [letter]. Anes-
thesiology. 1997;86:507-508.
17. Valentine S, Jarvis A, Shutt L. Comparative study of the effects of
air or saline to identify the extradural space. Br J Anaesth. 1991;66:
224-227.
18. Sarna M, Smith I, James J. Paraesthesia with lumbar epidural
catheters. Anaesthesia. 1990;45:1077-1079.
19. Shidner SM, Levinson G, Ralston DH. Regional anesthesia for
labor and delivery. In: Shnider SM, Levinson G, eds. Anesthesia for
Obstetrics. 2nd ed. Baltimore, Md: Williams & Wilkins; 1987:114.
20. Cousins M, Bromage PR. Epidural neural blockade. In: Cousins M,
Bridenbaugh P, eds. Neural Blockade. 2nd ed. Philadelphia, Pa: JB
Lippincott Co; 1988:324.
21. Brown DL, Wedel, DJ. Introduction to regional anesthesia. In:
Miller RD, ed. Anesthesia. 3rd ed. New York, NY: Churchill-Liv-
ingstone; 1990:1398.
22. Covino BG, Lambert DH. Epidural and spinal anesthesia. In:
Barash PG, Cullen BF, Stoelting RK, eds. Clinical Anesthesia. 2nd
ed. Philadelphia, Pa: JB Lippincott Co; 1992:817.
23. Shnider SM, Levinson G, Ralston DH. Regional anesthesia for
labor and delivery. In: Shnider SM, Levinson G, eds. Anesthesia for
Obstetrics. 3rd ed. Philadelphia, Pa: Williams & Wilkins; 1993:
143.
24. Brown DL. Spinal epidural and caudal blocks. In: Kirby R, Graven-
stein N, eds. Clinical Anesthesia Practice. Philadelphia, Pa: WB
Saunders Co; 1994:521.
25. Tetzlaff JE. Spinal epidural and caudal blocks. In: Morgan E,
Mikhail M, eds. Clinical Anesthesiology. 2nd ed. Stamford, Conn:
Appleton & Lange; 1996:233.
26. Bernards CM. Epidural and spinal anesthesia. In: Barash PG,
Cullen BF, Stoelting RK, eds. Clinical Anesthesia. 3rd ed. Philadel-
phia, Pa: Lippincott-Raven Publishers; 1997:651.
27. Molnar R, Pian-Smith MCM. Spinal, epidural, and caudal anesthe-
sia. In: Hurford WE, Bailin MT, Davison JK, Haspel KL, Rosow C,
eds. Clinical Anesthesia Procedures of the Massachusetts General
Hospital. 5th ed. Philadelphia, Pa: Lippincott Williams & Wilkins;
1998:256.
28. Cousins MJ, Veering BT. Epidural neural blockade. In: Cousins M,
Bridenbaugh P, eds. Neural Blockade. 3rd ed. Philadelphia, Pa: Lip-
pincott-Raven Publishers; 1998:297.
29. Brown DL. Spinal, epidural, and caudal anesthesia: Anatomy,
physiology, and technique. In: Chestnut DH, ed. Obstetrical Anes-
thesia. 2nd ed. St Louis, Mo: Mosby; 1999:197.
30. Brown DL, ed. Atlas of Regional Anesthesia. 2nd ed. Philadelphia,
Pa: WB Saunders Co; 1999:335.
31. Brown DL. Spinal, epidural, and caudal anesthesia. In: Miller RD,
ed. Anesthesia. 5th ed. Philadelphia, Pa: Churchill Livingstone;
2000:1510.
32. Ellis WE. Regional anesthesia. In: Nagelhout JJ, Zaglaniczny KL,
eds. Nurse Anesthesia. 2nd ed. Philadelphia, Pa: WB Saunders Co;
2001:1005.
33. Bernards CM. Epidural and spinal anesthesia. In: Barash PG,
Cullen BF, Stoelting RK, eds. Clinical Anesthesia. 4th ed. Philadel-
phia, Pa; Lippincott-Raven Publishers; 2001:695.
34. Kleinman W. Spinal epidural and caudal blocks. In: Morgan E,
Mikhail M, Murray M. Clinical Anesthesiology. 3rd ed. New York,
NY: McGraw-Hill Companies Inc; 2002:270.
35. Rosen MA, Hughes SC, Levinson G. Regional anesthesia for labor
and delivery. In: Hughes S, Levinson G, Rosen M. Anesthesia for
Obstetrics. 4th ed. Philadelphia, Pa: Lippincott Williams &
Wilkins; 2002:127.
36. Russell R. Loss of resistance to saline is better than air for obstet-
rical epidurals. Int J Obstet Anesth. 2001;10:302-306.
AUTHOR
CDR David Norman, CRNA, MSN, NC, USN, is on active duty in the
Navy, assigned to the Naval School of Health Sciences at Portsmouth
Naval Medical Center, Portsmouth, Va. He recently served as the clini-
cal coordinator for the Navy Nurse Corps Anesthesia Program,
Portsmouth Va. Norman is a doctoral student at the Frances Payne
Bolton School of Nursing, Case Western Reserve University, Cleveland,
Ohio.
DISCLAIMER
The material contained in this article and the views of this author in no
way represent the views of the US Naval Medical Center Portsmouth,
the US Navy, or the US Government.
AANA Journal/December 2003/Vol. 71, No. 6 453