PERI-OPERATIVE APPLICATIONS OF REGIONAL ANESTHESIA FOR CHEST AND ABDOMINAL SURGERY
Postoperative epidural analgesia has been provided in the cervical,
thoracic, lumbar, and caudal regions. Intrathecal postoperative analgesia
has been provided by single injection of lumbar narcotics or by repeated
injections of narcotics and/or local anesthetics through small bore catheters.
Renewed interest has occurred in attempting to decrease pain medication
requirements postoperatively by a technique termed preemptive analgesia.1
Through the administration of epidural opiates preoperatively, MSO4
was required postoperatively and visual analogue pain scores were significantly
improved. This may be explained by blocking the input obnoxious
peripheral afferents to the spinal cord and subsequently preventing spinal cord
wide-up.
Many different opiate compounds have been administered and include
morphine sulfate (MSO4), meperidine, hydromorphone, methadone,
fentanyl, and sufentanil. While other compounds such as the α2-agonists
are currently undergoing investigation, opiates remain the gold standard for
intraspinal injection. In the following discussion the term intraspinal
will be used generically to refer to both epidural and intrathecal
administration.
Epidurally administered opiates must initially cross the dura mater
before exerting their effect in the substantia gelatinosa (Rexed's laminas
II-III). These agents are also subject to uptake into the rich epidural
plexus of veins. Uptake and distribution into plasma after epidural
administration resembles that seen after intramuscular injection. The portion
of drug that is not taken into the vascular compartment is available to cross
the dura. Hydrophilicity determines how much drug will cross the
dura. Lipophilic agents such as fentanyl and sufentanil will cross the
dura rapidly but also tend to retrace into the epidural space.
Hydrophilic MSO4 crosses the dura poorly but tends to stay
subarachnoid once there.
The clinical effects of this characteristic is manifested in the
intraspinal:systemic dose ratio, and the ability of an opiate to truly provide
analgesia via a spinally mediated mechanism versus redistribution into plasma
with subsequent action supraspinally.
In comparison, MSO4 has been shown in many studies to
provide prolonged (12-16 hrs), excellent analgesia using a reduced dose.2,3
If the dura mater is circumvented by placing the MSO4 directly into
the subarachnoid space a minute dose (0.2-0.5 mg) can be utilized with
similarly impressive duration. The lipophilic agents do not enjoy these
advantages. Doses similar to systemic doses must be administered to
achieve analgesia.4 This does not necessarily mean there is no
spinal cord effect with these agents. It can be theorized that the
quality of analgesia may be enhanced by the intraspinal administration of these
agents, a result of the opiate being present spinally in much higher
concentrations and subsequently binding to the subarachnoid opiate receptor.
The ideal intraspinal opiate would be hydrophilic in nature, have a
high affinity for the opiate receptor, require occupation of a small percentage
of receptors to provide analgesia, demonstrate a prolonged duration of
analgesia, be free of side effects, and have no propensity to redistribute
cephalad through cerebrospinal fluid (CSF) and cause delayed respiratory
depression. An agent to match these characteristics does not presently
exist. Table 1 shows the intraspinal doses of commonly administered
opiates.
TABLE 1. Postoperative Intraspinal Opiate Doses
|
|
Systemic Dose |
Epidural
Dose |
Subarachnoid
Dose |
Epidural
Infusion
Rate |
|
MSO4 |
5 - 10 mg |
2 - 5 mg |
0.2 - 0.5
mg |
0.5
mg/hr |
|
Meperidine |
50 - 100 mg |
30 - 50 mg |
10
- 20 mg |
10 - 15 mg/hr |
|
Methadone |
5 - 10 mg |
3 - 6 mg |
N/A |
.5 mg/hr |
|
Fentanyl |
100 - 150 μg |
100 - 200 μg |
75 - 150
μg |
70 - 100 μg/hr |
|
Sufentanil |
25-50 μg |
25 - 50 μg |
25
- 50 μg |
25 - 50 μg/hr |
MSO4 was the initial intraspinal opiate injected and remains
the prototypical, available opiate. Its biggest advantage is its
hydrophilicity which provides long analgesia, and, if injected intrathecally,
requires minute doses. Large series have all demonstrated the safety of
intraspinal MSO4.2,3 In its preservative-free form,
MSO4 remains the only opiate which has FDA approval for epidural or
intrathecal injection. Epidurally administered MSO4 penetrates
the dura poorly with an onset of action being 30 minutes and a peak effect not
usually seen prior to 60 minutes. Thus, the injection of MSO4
must be anticipated if adequate analgesia is to be achieved at the onset.
With this understanding MSO4 has been administered routinely in the
postoperative setting by intermittent injection and continuous infusion.
While MSO4 has been utilized intraspinally by patient-controlled
technology (PCA), its characteristically delayed onset may make it less than
optimal by this route.
Meperidine has also been injected extensively and possesses the unique
characteristic of having local anesthetic properties. Its local
anesthetic potency is sufficient when administered intrathecally to provide
sensory anesthesia in some patients and surgical analgesia in most. The
significance of its local anesthetic properties when injected epidurally is
less clear. It does not produce sensory anesthesia in normal epidural
doses. However, very low doses of local anesthetics such as bupivacaine
are synergistic with opiates. The dual properties which meperidine
possesses may provide synergism within the single compound.
Fentanyl has achieved widespread popularity over the past 7-10
years. Its initial use stemmed from the hope that fentanyl might not
suffer the same risk of respiratory depression that MSO4 had been
shown to have. This question currently remains unanswered. The bulk
of information presented would support that fentanyl does redistribute
rostrally through the cerebrospinal fluid (CSF) less than the hydrophilic agent
MSO4, and, as a result, may have fewer side effects associated with
it. Nonetheless, fentanyl does alter the CO2 response curve,
shifting it to the right in a manner similar to MSO4. Like MSO4
this property has rarely been a problem in the clinical setting.
The biggest unanswered question regarding fentanyl surrounds whether
the analgesia obtained is a true spinal cord effect or a result of the drug
being redistributed through plasma to supraspinal levels. Multiple
conflicting studies have appeared over the past several years. The dose
required to produce analgesia intraspinally is similar to that necessary by
intravenous administration. Recently, in a comparison of lumbar epidural
vs., intravenous Fentanyl infusions for post thorocotomy pain patients acheive
similar pain relief but required more drug when administered epidurally.
Whether the quality of analgesia obtained in the early postoperative period is
different has been more difficult to ascertain. An explanation may be
that with initial injection there is a spinal cord effect but with time either
habituation or redistribution through plasma occurs to such an extent that the
spinal cord effect is negligible.
If one accepts that fentanyl does not extensively travel cephalad
through CSF then one must place the spinal or epidural catheter at the
dermatomal level appropriate for the surgical procedure. A lumbar
epidural catheter will not provide analgesia by a spinal cord mechanism for
thoracically mediated pain.5,6 This is in contradistinction to
MSO4, which has been shown to travel cephalad in sufficient quantity
to provide analgesia at dermatomally distant sites from the catheter placement.
Whether the effect after intraspinal injection is spinally mediated or
systemic, the rapidity of analgesic onset after fentanyl administration is so
pronounced that there is utility for the agent when patients are found to be
acutely experiencing unacceptable pain. Patients in this setting are
often not well served if they have to wait 30-60 minutes for the effects of an
injection of intraspinal MSO4. 100-200 μg of fentanyl in
these patients can provide quick analgesia while the longer lasting MSO4
begins to set up.
Sufentanil has also been used in the intraspinal canal. The
characteristics of the lipid soluble fentanyl are magnified with
sufentanil. Systemic doses are necessary and may be exceeded when used
intraspinally. Its theoretical advantage results from the strong affinity
that sufentanil has for the spinal μ-receptor. Whereas MSO4
requires 80-90% occupancy of the receptor to provide analgesia, sufentanil has
to occupy less than 30%. This characteristic has relevance in the cancer
population where tolerance and down regulation of the receptor is important,
but is probably unimportant in the acute postoperative setting.
If opiates are to be injected by intermittent technique, the drugs can
be administered either on a pre-timed schedule or when the patient complains of
pain (prn). If medications are to be injected on a pre-timed schedule,
one must anticipate the large interpatient variation in the analgesic t?.
For example, scheduling re-injection of MSO4 every 12 hours would
render a small percentage of patients with inadequate pain relief prior to the
scheduled re-injection while others would not have required the subsequent
injection for 6-12 hours past the scheduled time. Unnecessary pain or
possible drug accumulation and side effects could result. The flexibility
which is necessary to allow for interpatient variability has been most
successfully done when trained nurses are allowed to inject epidural catheters
during routine postoperative care. Unfortunately, most state nursing
boards preclude registered nurses from re-injecting spinal catheters. If
intraspinal opiates are injected only on demand, the qualitatively superior
analgesia seen with intraspinal narcotics often regresses before a qualified
person can respond to reinject. Patients suffer unacceptable pain before
adequate analgesia is re-established. This roller-coaster effect is less
than optimal in postoperative management.
Continuous infusion of intraspinal narcotics prevents the regression of
analgesia seen with intermittent injections by keeping patients in a steady
state condition of continuous analgesia. If inadequate analgesia occurs,
a change in the infusion rate can be ordered by the physician and performed by
the nursing staff. Continuous infusions are particularly well-suited for
the lipophilic agents such as fentanyl and meperidine and/or local
anesthetics. We have also had much experience in using continuous infusion
with epidural morphine, with excellent analgesia. Additional set-up is
required using this technique, particularly volumetric pumps and tubing.
The largest risk inherent to the technique is inadvertent injection into the
tubing by personnel mistaking the epidural line for an IV line. Drugs
safe for the intravascular space could easily be catastrophic in the
intraspinal canal. Our institution uses nitroglycerine tubing, which has
no injection ports and makes this risk less likely to occur. Also, most
volumetric pumps do not prevent patients from changing the rate of infusion
themselves, behavior that would be expected to occur most frequently in
patients with drug-seeking personalities. PCA pumps with locked drug
chambers may provide a better alternative.
With the advances in PCA technology, studies have reported the efficacy
of epidural opiates by this route of administration. Fentanyl appears to
be the logical agent since its rapid onset meshes better with the principles of
PCA dosing and would avoid the need for long lockout intervals. However,
Sj–strom et. al. examined epidural morphine and pethidine by PCA.7
Pain relief was excellent in both groups although large interpatient variation
existed in both groups. The plasma levels drawn for both drugs were
significantly decreased compared to minimum analgesic concentrations after IM
administration. Further work will need to be performed in this area
before specific recommendations can be made for routine clinical use.
Several studies have examined the effect of epidural opiates on outcome
in post-surgical patients. One of the initial studies to measure outcome
criteria was performed by Rawal et. al. in which a group of morbidly obese
patients receiving epidural narcotics post-operatively had shorter hospital
stays, improved analgesia, and decreased morbidity.8
The most frequently cited article concerning outcome measures in
epidural patients has been the work of Yeager and Glass, published in 1987.9
Twenty-eight patients received epidural analgesia and 25 received IV
narcotics. Morbidity, complication intensity, and mortality rates were
all significantly less in the epidural group. Physician and hospital
costs were significantly lower than in the group receiving epidural narcotics.
One explanation for the beneficial effects of epidural anaesthesia in high-risk
surgical patients may relate to the effect of high thoracic epidural anesthesia
on the diameter of normal and diseased epicardial arteries. Blomberg et.
al. found thoracic epidural analgesia anesthesia to increase luminal diameter
in stenotic segments from 1.35 +/- 0.11 to 1.56 +/- 0.13 ml but had no effect
on non-stenotic segments.10 This effect could be quite
significant in patients undergoing surgery who had poor cardiac reserve
secondary to ischemic heart disease.
A recent study (n=173) tried to compare the effects of thoracic
epidural anesthesia in combination with light general anesthesia to
"balanced" general anesthesia in high-risk surgical patients
undergoing abdominal aortic reconstructive surgery.11 No
difference in morbidity factors or mortality between groups was noticed.
Postoperatively, patients in each group received either subcutaneous morphine
sulfate, epidural fentanyl, or epidural bupivacaine. The effect of
postoperative management on outcome was not addressed.
Risks initially identified with epidural narcotic analgesia included
respiratory depression, urinary retention, pruritus, and nausea and
vomiting.
Much has been published concerning the risk of respiratory depression
from the administration of intraspinal opiates, reflecting the serious
morbidity and mortality which can occur if respiratory depression goes
unnoticed. Early respiratory depression results from the redistribution
of narcotics from the epidural space into the vascular bed and subsequently to
the respiratory center of the fourth ventricle. This occurs 20-45 minutes
after an injection, and mimics the effects which can be seen after an opiate is
administered intramuscularly or subcutaneously. Early respiratory
depression occurs much more frequently than delayed, late respiratory
depression, but its consequences can be equally catastrophic if patients
are not observed closely. Undetected early respiratory depression is
uncommon since most clinicians and nursing personnel know to monitor patients
closely for 30-45 minutes after an opiate has been administered, regardless of
the route utilized.
Delayed respiratory depression most commonly occurs 3-6 hours after
administration but has been reported 18 hours after a single intraspinal
injection. Delayed respiratory depression results from passive rostral
flow within the CSF to the fourth ventricle. The risk of delayed
respiratory depression increases many fold if a concomitant dose of systemic
narcotic is delivered. The clinician can easily forget that an
intraspinal opiate injected 5 hours earlier will have peak ventricle levels at
the same time that a systemic injection is made. This combination of
epidural and intravenous administration was shown in Gustafsson's survey to be
the single greatest predictor of respiratory depression.12
Systemic opiates should never be administered while an intraspinal technique is
being used.
Increasing age decreases the dose requirement of epidural
narcotics. If a routine intraspinal opiate dose or infusion is
administered to an elderly, frail individual, the risk of respiratory
depression will be increased. Likewise, if during placement of an
epidural catheter, a dural puncture has occurred, the transfer kinetics of the
agent passing from the epidural space to the subarachnoid space can be
significantly altered. If a dural puncture occurs and a subsequent
epidural catheter is successfully positioned, the initial dosage used should be
extremely conservative, possibly reduced by a factor of 10 and the patient
initially watched for any signs of respiratory depression. Doses can
subsequently be cautiously increased if analgesia is incomplete.
The risks of respiratory depression from intraspinal narcotics have
affected the management of patients receiving this technique, yet their risks
may be no different than after systemic administration. A study by Miller
et. al.13 reported 860 patients who received MSO4
orally or systemically and found the instance of severe respiratory depression
to be 0.9%, a number that compares favorably to the studies by Gustafsson,
Ready, and Rawal.12,14,15
No consensus of opinion exists on what degree of respiratory monitoring
is necessary with the use of intraspinal narcotics. Unfortunately, a
respiratory monitor that can be used on a regular nursing floor and is both
reliable and predictive of a respiratory event has not been developed. In
a 1986 survey of 73 centers by Mott and Eisele, they found that 67% of the
centers kept intraspinally medicated patients in specialized care units.
They found no difference in the incidence of respiratory depression: 0.4%
overall, but 0.6% in centers where patients were managed only in specialized
units.16
We rely on hourly respiratory checks by our floor nurses and provide
continuous inservicing. Standing orders provide guidelines for all
potential emergencies. A patient is monitored in a specialized unit only
if his surgical risk status and postoperative analgesic risk factors deem it
appropriate.
Treatment of respiratory depression is simple and effective:
naloxone will reverse the respiratory depression of all pure mu-opiate
agonists. Routine cases will require naloxone in doses of 2-5
μg/kg/hr to reverse the respiratory depression, but this must be titrated
to effect and much higher doses may be necessary if an inadvertent overdose has
occurred. Naloxone can be initially administered by a bolus injection
but, should always be followed by infusion when treating respiratory depression
from intraspinal narcotics since its t1/2 is much shorter than any
of the narcotics.
Nausea and vomiting is associated with epidural narcotics, the extent
of which may never be appreciated since many surgical patients experience
nausea or vomiting from their surgical procedure. Estimates have ranged
from 17-34% for MSO4, while the lipid-soluble agents have been
reported to have a decreased incidence of nausea and vomiting. The emetic
effect of epidural narcotics has been attributed to their effects on the
chemoreceptors in the brain.
Treatment of intraspinally related nausea and vomiting can involve
several regimens. An anti-emetic such as chlorpromazine can be included
on standard order forms. Naloxone has also been reported effective in the
treatment of nausea and vomiting, although recent studies support the
ineffectiveness of naloxone to reliably reverse nausea and vomiting.
Recently, two studies have demonstrated marked effectiveness with scopolamine
when applied as a patch the night before surgery. Odanterisin may prove
effective in refractory cases of nausea and vomiting.
The risk of urinary retention has often been understated and its
significance unappreciated when compared to the life-threatening complications
associated with respiratory depression. The incidence of urinary
retention ranged from 22-50% in several studies and may be less with the more
lipid-soluble agents. The morbidity associated with urinary retention and
prolonged catheterization is far from insignificant and can preclude
intraspinal analgesia as the optimal technique in otherwise healthy patients
undergoing surgical procedures with only minimal postoperative pain.
The mechanisms of urinary retention have not been clearly elucidated in
patients receiving epidural narcotics. Several studies have implicated a
relaxation of the detrusor muscle and an increase in bladder capacity.
Treatment of urinary retention is often unsuccessful. Reports of
naloxone to 0.8 mg have been described or infusions of 2-5 μg/kg/hr prior
to the administration of epidural narcotics. Our own clinical experience
has often been less than encouraging.
Pruritus represents a commonly reported side effect of epidural
narcotics. This can vary from a benign localized reaction to a diffuse,
relentless problem that demands treatment. Ready's study of 1106 patients
and Stenseth's report on 1085 patients revealed an incidence of 24% and 11%,
respectively.14,17 This wide variation most likely reflects a
difference in extent of questioning by investigators and a somewhat subjective
determination of whether the pruritus was epidurally related or not.
Pruritus resulting from intraspinal narcotics can be managed
successfully in most patients. Diphenhydramine reduces many localized
reactions, whether of epidural origin or not and can be listed on the standard
orders. Naloxone can be administered in refractory cases and is almost
universally effective in doses of 2-4 μg/kg/hr.
Epidural narcotics, particularly MSO4, have been reported to
cause an increased incidence of recurrent herpes simplex infections in the
obstetric population. With the exception of a single case report, this
has not been reported in other postoperative cases. Why this association exists
remains unclear.
The use of intrathecal opiates in postoperative patients has been
somewhat limited until recently because of unacceptable catheter
technology. Single-dose injections have been administered at the time the
subarachnoid block was placed, or, in the case of some back operations,
surgeons would elect to administer a dose during the surgical procedure.
This technique could provide long-lasting analgesia (18-24 hours in some
patients) when MSO4 was employed. The lipid-soluble agents do
not have equally long analgesic profiles when administered intrathecally by
single shot.
For operative procedures, where the duration of postoperative pain
would not be expectedly prolonged and a subarachnoid technique is chosen for
surgical anesthesia, single dose subarachnoid opiates are an excellent choice
and probably underutilized. Risks, potential complications, and
monitoring should be no different than after epidural administration if
equipotent doses are used (Table 1).
The practitioner must be acutely aware of which opiates and
preparations are suitable for intrathecal administration. Spinal cord
toxicity issues vary greatly between epidural and intrathecal
administration. A drug that is safe for epidural injection may be
catastrophic is injected subarachnoid. The epidural space can be quite
forgiving of some preservatives, while the subarachnoid space is characteristically
not. MSO4 in its preservative-free preparation has been
demonstrated safe for intrathecal administration. Fentanyl has been
approved in Canada for subarachnoid use and its extensive chemical use in the
United States would support its safety. While appropriate spinal cord
toxicity studies have not been performed with meperidine, the clinical evidence
supports its use in preservative-free forms. Interestingly, one recent report
in sheep suggests that sufentanil may exhibit some toxicity when given
subarachnoid. This would conflict with earlier cat and dog studies which
showed sufentanil free of any spinal cord toxicity. Until further spinal
cord studies have been performed, one should inject intrathecal sufentanil with
caution.
With newer spinal catheter technology, prolonged analgesia may become
possibly either by repeated injections through the catheter or continuous
infusion. To date, we have not been able to find a pump that will
reliably infuse opiate through the small spinal catheters. Another concern will
be the risk of infection with catheters placed in the subarachnoid space.
Future work in this area is surely forthcoming.
Outcome measures can justify the use of epidural or intrathecal
analgesia for subsets of postoperative pain patients. Alternative
techniques exist for other patients'. For any given technique, different
agents are available, many of which can be combined, when appropriate, in
synergistic fashion. In comparison with systemic routes of administration,
these techniques provide excellent analgesia with small amounts of drugs.
The future will yield new techniques, new technology to aid in the performance
of regional analgesia, and new agents for administration. A final caveat
remains that any reasonable technique has its own inherent risks and the
postoperative pain practitioner must act as true consultant to the surgeon in
recommending a regional technique or systemic route for postoperative
analgesia.
1. Katz J, Kavanagh
BP, et al: Preemptive analgesia: clinical evidence of
neuroplasticity contributing to postoperative pain. Anesthesiology
77:439-446, 1992
2. Stenseth R,
Sellevold O, Breivik H: Epidural morphine for postoperative pain:
Experience with 1085 patients. Acta Anaesthesiol Scand 29:148‑156, 1985
3. Ready LB, Loper
KA, Nessly M, Wild L: Postoperative epidural morphine is safe on surgical
wards. Anesthesiology 1991(in press)
4. Loper KA, Ready
LB, Downey M, Sandler AN, Nessly M, Rapp S, Badner N: Epidural and intravenous
fentanyl infusions are clinically equivalent after knee surgery. Anesth Analg
70:72‑75, 1990
5. Sandler AN,
Stringer D, et al: A randomized, double-blind comparison of lumbar epidural
and intravenous fentanyl infusions for postthoracotomy pain relief.
Anesthesiology 77:626-634, 1992
6. Guinard JP,
Mavrocordatos P, Chiolero R, Carpenter RL: A randomized comparison of
intravenous versus lumbar and thoracic epidural fentanyl for analgesia after
thoracotomy. Anesthesiology 77:1108-1115, 1992
7. Sj–str–m S,
Hartvig D, Tamsen A: Patient‑controlled analgesia with extradural
morphine or pethidine. Br J Anaesth 60:358‑366,
1988
8. Rawal N, Sjostrand
U, Christoffersson E, Dahlstrom B, Arvill A, Rydman H: Comparison of
intramuscular and epidural morphine for postoperative analgesia in the grossly
obese: influence on postoperative ambulation and pulmonary function.
Anesth Analg 63:583‑592, 1984
9. Yeager MP,
Glass DD, Neff RK, Brinck‑Johnsen T: Epidural anesthesia and analgesia in
high‑risk surgical patients. Anesthesiology 66:729‑736, 1987
10. Blomberg S,
Emanuelsson H, Kvist H, Lamm C, Pont=n J, Waagstein F, Ricksten S‑E:
Effects of thoracic epidural anesthesia on coronary arteries and arterioles in
patients with coronary artery disease. Anesthesiology 73:840‑847, 1990
11. Baron J-F,
Bertand M, BarrČ, Godet G, Mundler O, Coriat P, Viars P. Combined
epidural and general anesthesia versus general anesthesia for abdominal aortic
surgery. Anesthesiology 75:6-11-618, 1991
12. Gustafsson LL,
Schildt B, Jacobsen KJ: Adverse effects of extradural and intrathecal
opiates: report of a nationwide survey in Sweden. Br J Anaesth 54:479‑486, 1982
13. Miller RR,
Greenblatt DJ: Drug Effects in Hospitalized Patients. New York, John Wiley and
Sons, 1976, pp 151‑152
14. Ready LB, Loper
KA, Nessly M, Wild L: Postoperative epidural morphine is safe on surgical
wards. Anesthesiology 1991 (in press)
15. Rawal N, ArnČr
S, Gustafsson LL, Allvin R: Present state of extradural and intrathecal opioid
analgesia in Sweden. A nationwide follow‑up survey. Br J Anaesth 59:791‑799, 1987
16. Mott JM, Eisele
JH: A survey of monitoring practices following spinal opiate administration.
Anesth Analg 65:S105, 1986(Abstract)
17. Stenseth R,
Selleveld O, Breivik H: Epidural morphine for postoperative pain:
Experience with 1085 patients. Acta Anaesthesiol Scand 29:148-156, 1985