Relief of Post-Operative Pain
Post-thoracotomy Pain: Assessment and Relief
Notes for a lecture to the Annual Respiratory Symposium
3/14/92
F.W. Grannis jr.M.D.
Thoracotomy with resection of lung tissue is frequently required for
ablation of bronchogenic carcinoma and less often for various benign lung
diseases. The patient with normal lung function tolerates removal of one
lung well in almost all cases. Unfortunately almost all lung cancer patients
have significant pulmonary functional deficits caused by the same cigarette
smoking that caused the cancer. These patients can have major problems
after resection and it is often a difficult problem for the pulmonologist
and thoracic surgeon to decide whether the patient can survive the necessary
surgery. Post- resectional pulmonary function studies at 3-6 months post-op
can be accurately predicted by the formula
FEV1=18-x\18 x FEV1 (pre-op)
where x=the number of segments in the lung tissue removed.
For example if the pre-op FEV1 is 3.0 and the left lung containing 8
bronchopulmonary segments is to be removed then the calculation is
Patients with a predicted FEV1 of > 1.0 can generally function reasonably
well. and patients can survive with as little as 0.7 l.. The problem with
the calculations are that they involve PFTs a number of months after the
operation. Little is known about PFTs in the early post-op period, but
what data does exist indicates that FEV1 and other measurements are strikingly
lower than the predicted values.
The primary tasks of the physician and the nurse are to preserve life
and to alleviate suffering. Management of post-operative pain is important
for both reasons. Post-thoracotomy pain is rated as severe by most patients.
Further, it is well established that post- thoracotomy pain is a major
factor in the morbidity and mortality of thoracic surgical proceedures.
It is readily observable that patients complaining of post-op pain restrict
deep breathing and cough in an effort to prevent the sharp discomfort experienced
during these activities. Multiple studies have demonstrated a decrease
in vital capacity, functional residual capacity (FRC), closing capacity
(CC) ,FEV1, peak expiratory flow (PEF), retention of secretions, atalectasis,
increase in aAO2 gradient, decrease in O2 saturation and CO2 retention.
Breathing at a FRC below CC results in closure of small airways with
resultant >aAO2 gradient and atelectasis . Lowered PEF reflects inability
to cough effectively. Failure to clear secretions causes atelectasis and
nosocomial pneumonia. These problems can result in respiratory failure
in otherwise normal patients but are of increasing importance in the elderly,
the chronically ill and the emphysematous patient. Respiratory failure
is the primary morbidity after thoracotomy and results in high mortality
and morbidity and astronomical medical costs . Many different modalities
have been utilized in an attempt to control post-op pain.
1. PRN narcotics
given either IM or IV is the traditional method of pain relief after
operation. It has been demonstrated in numerous studies to be ineffective.
Variability in patient perception and toleration of pain are enormous.Problems
with assessment of patient pain are ubiquitous. Minimum effective concentration
of narcotics also varies tremendously between individuals as also does
the amount of administered narcotic required to produce a minimum effective
concentration. Inadequate doses of medication are typically given at infrequent
intervals , leaving the patient uncomfortable and subject to the cascade
of problems listed above. The reasons for this are practical. First, the
patient has to have drug levels decrease to the point where he is uncomfortable.
Next, a nurse, often overworked or caring for a more urgent medical problem
elsewhere, is delayed for a significant period of time. Then there are
further delays while narcotics are obtained from a locked cabinet and administered.
In the case of IM injection there is a further delay in uptake. Added to
these logistic problems there is a poor understanding of narcotic analgesics
by many nurses and even more physicians; an unreasoning fear of producing
addiction and a reluctance to give adequate doses . Finally there are a
number of inherent side effects to the use of narcotics ; nausea, vomiting,
constipation, pruritus that make their use unpleasant for the patient.
Other side effects, CNS depression and respiratory depression, if excessive,
can magnify the pulmonary complications of the thoracotomy itself. Clearly
a different approach is needed.
2. PCA (patient controlled analgesia)
employs a microcomputer mechanism to allow pre-specified amounts of
narcotic to be available to the patient immediately upon demand. This technique
has been shown in numerous good studies to provide a more constant blood
level of narcotic and better pain control with smaller total dosage of
medication. It also frees up nurses for more critical tasks. In my experience
PCA works reasonably well but there are often problems with patient compliance.
The idea behind PCA is quite simple. Each time the patient feels pain he
is to push a button. Despite this simplicity, it is frustrating that many
patients will try to "outsmart" the system. "I didn't want
to push the button because I might become an addict." On other occasions
nurses will intervene and push the button for the patient. Obviously the
confused, demented or obtunded patient is a problem. For these reasons
PCA often does not result in optimal pain relief.
3. Continuous drip narcotics
can be given IV, but an important safety feature of PRN or PCA narcotic
use is lost, i.e. if the patient is obtunded by excessive levels of narcotics
then he will not ask for or receive another dose. PCA systems allow the
use of continuous drip narcotics with supplementary doses, but constant
drip levels need to be kept low to enhance safety. If continuous drip narcotics
are used, a coded system that does not allow patient or family tampering
is imperative. Continuous drip narcotics are ideal in the pain control
of the terminal , suffering patient where a death secondary to respiratory
depression in not to be feared. A technique of nurse-controlled intravenous
narcotics has been reported to give good results in post-operative pain.
In a different medico-legal climate this would be an interesting technique
to try here.
4.Paravertebral (intercostal nerve) block
Percutaneous intercostal nerve blocks with local anesthetics have been
studied many times in the past and found effective in controlling post-thoracotomy
pain and improving pulmonary function. The major disadvantage to the technique
is that multiple intercostal nerves above and below the incision (or injury)
must be blocked. This means multiple unpleasant injections for the patient
and multiple opportunities for complications such as pneumothorax, hematoma
and inadvertent subdural injection of local anesthetic to occur. Since
duration of local anesthesia is only a few hours with lidocaine and no
more than 8 to 12 hours with bupivicaine and 1:200,000 epinephrine repeated
injections are necessary. "Rube Goldberg" devices with multiple
indwelling intercostal catheters have been used. The techniques are seldom
used because they are labor intensive, uncomfortable for the patient and
often ineffective.
5 Epidural anesthesia
catheters are placed by the anesthesiologist either before or after
thoracotomy. The technique is difficult and proper placement in the epidural
space can be problematic. Improper technique can result in failure of analgesia,
epidural hematoma, spinal anesthesia. Contamination can cause meningitis
or epidural abcess. The catheter can be infused either intermittently or
by constant infusion , with narcotics, local anesthetics or a combination
of both into the epidural space. Local anesthetics have the disadvantage
that they block sympathetic outflow from the cord and can cause problems
with hypotension. Narcotics locally penetrate the dura and bind with endorphin
receptors in the spinal cord, blocking transmission of pain impulses to
higher centers. This technique can give excellent, even complete, relief
of pain , but it sometimes is unsatisfactory. Some patients are afraid
of spinal injections and refuse the technique. In others with spinal surgery,
coagulation defects or sepsis, the technique is contraindicated. Intermittent
doses are often delayed because the anesthesiologist doesn't want to come
in the middle of the night. The continuous technique works better, but
in some local hospitals, anesthesiologists and nursing administrators are
unwilling to use the technique because of potential medicolegal risks.
Supplemental systemic narcotics must be used judiciously because of the
risks of respiratory depression. This is a special problem when catheters
become displaced and epidural anesthesia fails. The patient then gets little
or no pain relief. Pruritus can become a major problem. Bladder dysfunction
is common. Systemic uptake of narcotics is present and significant respiratory
depression can occur. Despite these potential problems, continuous epidural
narcotic infusion is probably the best technique commonly used today.
6. Cryoanesthesia
is performed by dissecting the intercostal nerves and freezing them
with a liquid nitrogen probe at -60 C at the end of the operation. This
effectively destroys the nerves but leaves a frameword for regrowth of
axons. It has been reported to provide good pain relief but can cause troublesome
neuritis during the late post-operative phase.
7. TENs
uses electrical impulses to "distract" neurons from their
usual recognition and transmission of pain impulses. Pain perception is
carried by the small C fibers of peripheral nerves. High frequency impulses
delivered by commercially available TENS transmittors stimulate activation
of the larger A fibers. The gate theory postulates that this causes inhibition
of neurons in the substantia gelatinosa of the spinal cord which inhibit
the transmission of pain recognition carried by the smaller C fibers to
higher centers in the brain. Although studies have shown a therapeutic
effect in postoperative patients, in my experience it is seldom of significant
value.
8. NSAIDs
non-steroidal anti-inflammatory drugs were little used in the past because
they were not available in a parenteral form. Now ketorolac (Toredol) represents
a powerful addition to the analgesic formulary. NSAIDs inhibit production
of prostaglandin synthesis by inhibiting the enzyme cyclo-oxygenase that
converts arachidonic acid into the prostaglandin cycle. NSAIDs also inhibit
thromboxane A2 synthesis and therefore platelet aggregation and can cause
bleeding complications. NSAIDs cause a significant increase in ulceration
of the stomach and duodenum, especially in older patients. Ketorolac is
a powerful analgesic when an initial dose of 60 mg. is followed by 30 mg.
q 6 hours in adults.
9 Continuous paravertebral block
is a new technique based on intercostal nerve block. At the completion
of thoracotomy the posterior parietal pleura is stripped up off the chest
wall over a five rib area centered over the intercostal incision. The pocket
created extends into the paravertebral area where the intercostal nerves
leave the spinal foramina. A plastic catheter is inserted through an intercostal
space by needle puncture and positioned in the pocket. A continuous drip
of 1% lidocaine at a dose of 1cc./ 10 Continuous Paravertebral Block with
Continuous Lidocaine after Thoracotomy: A Descriptive Pilot Study. by Elizabeth
Sullivan, RN, MN, APN, Frederic W. Grannis Jr., MD, FCCP, Betty Ferrell,
RN, PhD and Mordecai Dunst, MD. From the Departments of Thoracic Oncology,
Pulmonary Medicine and Nursing Research, City of Hope National Medical
Center, Duarte CA Please address requests for reprints to F.W. Grannis
Jr. MD, C/O Department of General and Oncologic Surgery, City of Hope National
Medical Center, 1500 Duarte Road, Duarte CA Abstract Continuous paravertebral
block (PVB) with bupivicaine has been reported to be an effective analgesic
technique in patients after thoracotomy1. We report a retrospective study
of PVB using a continuous drip of 1% lidocaine at a dose of 1 mg./kg/hour.
A posterior parietal pleural pocket was created and cannulated with a sixteen
gauge polyethylene catheter. Lidocaine was perfused over a three day period
following surgery. Patients also had access to morphine sulfate (MS) via
patient controlled analgesia (PCA). Eighteen consecutive thoracotomies
(in seventeen patients) performed over a six month period were reviewed.
Serum lidocaine exceeded the toxic level of 5mg/ml in only one patient,
who experienced no clinical toxicity. Pain was evaluated by verbal analog
scores (zero = no pain and ten = worst pain), which averaged 3.02, 3.14
and 2.8 in the three days following surgery, mean total daily MS doses
were 24.3, 37.75 and 34.32 mg. (range 0-94 mg.). Sedation was scored on
a 1 to 5 scale. Mean scores were 2.78, 2.56 and 2.6. No patient died or
had a major respiratory complication. Continuous PVB with lidocaine appears
to be a promising adjuvant technique in the management of post-thoracotomy
pain. Effectiveness needs to be confirmed in a prospective randomized study.
Introduction: Pain following surgery is a major cause of morbidity and
mortality and improved pain management results in improvement in post-surgical
morbidity and mortality. 2 Transection of skin, muscle and pleura, retraction
of muscles, ligaments, and intercostal nerves, pleural irritation by chest
tubes and fractured ribs all contribute to severe pain following thoracotomy.
If pain is not adequately controlled, a series of events, including shallow
respiration, chest wall splinting and avoidance of cough and deep breathing,
result in atalectasis, ventilation- perfusion mismatch, hypoxemia and pneumonitis.
The end result can be prolonged ICU and hospital stay, or even respiratory
failure and death. Traditional treatment with parenteral PRN opioids provides
inadequate pain relief and superimposes morbidity secondary to the sedative
and respiratory depressant side effects of these drugs. Because of these
problems, in recent years, many surgeons and anesthesiologists have been
studying new techniques of pain control in the post- thoracotomy patient.
Local anesthetic intercostal nerve blocks have been tried by many groups
in the past, and have had some demonstrated benefit, but the problems of
multiple daily injections of multiple intercostal nerves, and the impracticality
of indwelling catheter systems to deliver local anesthetics intermittently
to multiple nerves, have resulted in abandonment of these techniques by
most groups. Sabanathan et al, in 1988 , described a simple technique of
continuous regional anesthesia via a percutaneous catheter through which
continuous bupivicaine could be infused into a retropleural pocket, created
by the surgeon at the time of surgery, by dissecting the posterior parietal
pleura off the chest wall, over a five rib area, centered on the thoracotomy
incision. They reported good analgesia. We report a pilot study, using
a retrospective audit, of the effectiveness of this technique. We have,
however, substituted a continuous infusion of 1% lidocaine at a dose of
0.1 cc/kg./hour (1mg./kg./hr.) for the bupivicaine used by previous investigators.
Study Design: A retrospective, descriptive study was designed to evaluate
the effectiveness of post thoracotomy, continuous infusion lidocaine, paravertebral
block (PVB) in controlling post operative pain. All consecutive patients
who underwent thoracotomy during a six month period from October 1, 1992
to April 30, 1993 were evaluated. In some thoracotomies performed during
this period, the pocket could not be made because of dense adhesions, pleural
plaques or surgical resection of the pleura. In seventeen patients (eighteen
thoractomies) the pocket was dissected. A 16 gauge, 12 inch long polyurethane,
single lumen central venous catheter with side holes * was introduced percutaneously
through a low posterior interspace and sutured in place. Earlier experience
had shown that intermittent dosing and smaller diameter catheters, are
unsatisfactory because of frequent catheter occlusion. Most patients had
10 cc of 1% lidocaine instilled in the pocket intraoperatively, and all
had a continuous infusion at a rate of 0.1 cc (1 mg.)/kg./hr). for three
or four days after surgery. Intravenous PRN morphine sulfate (MS) was given
until the patient was sufficiently awake to begin patient controlled analgesia
(PCA) with MS. Results: Demographics and Surgical Variables: * Arrow International
Inc. Hill and George Avenues, Reading PA 19610; model number AK-04400)
There were 18 thoracotomies on 12 males and 5 females (one male patient
had bilateral thoracotomy) with a mean age of 58 years (range 22-82 years).
67% of patients were operated for treatment of bronchogenic carcinoma,
22% for other neoplasms and 11% for benign disease. Operations included
pneumonectomy 11%, lobectomy 33%, and wedge or segmental resection 55%.
(Table 1). Rib fracture, usually "green stick" was noted in 28%
of patients secondary to retraction. Ribs were reapproximated by figure
of eight doubled #1 Vicryl sutures placed through drill holes in the lower
rib in 83%. This technique is employed to avoid pain consequent to compression
of intercostal nerves. Rib resection was performed in one patient (6%)
and pleural adhesions were dissected in 41%. Mean forced vital capacity
was 3.48 liters (range 1.6-5.5 l; predicted mean 4.07 liters); mean FEV1
was 2.66 liters (range 0.92-4.58 ;predicted mean 3.21 l). (Table 2) Pain
Outcomes: Post op day (POD) #1 was defined as starting at the completion
of surgery and ended at 7 AM the following morning. Each subsequent POD
consisted of three nursing shifts of eight hours each, from 7AM to 7AM.
Serum lidocaine levels were obtained on POD#2 and 3 in 14 and 13 patients
respectively. Mean levels were 2.39 mg on POD# 2 (range 1.2 to 5.9) and
3.7 mg on POD#3 (range 1.8 to 5.0). (Table 3).. One patient had a serum
level over 5mg/ml. at 5.9mg/ml.. No patient experienced clinical toxicity
secondary to lidocaine . (Table 3). The total dose of MS given over each
eight hour shift and twenty four hour day were recorded. Mean total doses
of M.S. were 24.3 mg (range 0-88 mg.), 37.75 mg. (range 9-94 mg.) and 34.32
mg. (range 0-108 mg.) on POD #1-3. (Table 4). Adequacy of analgesia was
documented by multiple assessments by nurses during each shift by verbal
analog, zero to ten pain rating scale . Since the study was retrospective,
differential scores at rest and during cough and activity were not obtained.
Mean pain rating scores were 3.02 (range 0.5-8.0) , 3.14 (range 0.4- 4.8)and
2.8 (range 0.2-4.3) on POD#1-3 respectively. The combination of continuous
PVB and PCA MS provided adequate pain relief in most patients. In one patient
a continuous morphine basal rate was required for pain control for one
day for treatment of unstable angina pectoris. Four patients were given
additional PRN MS doses and one patient received a single dose of ketorolac.
Nurses noted patient state of sedation multiple times each shift on a one
to five sedation scale . Mean sedation scores ( 1=alert; 5=comatose) were
2.78 (range 1-4), 2.56 (range 1.3-3.6) and 2.6 (range 1-3.6) on POD#1-3.
No patient became markedly obtunded or comatose. Complications: One catheter
occluded on post-operative day 3. There were no other complications referable
to the catheter. Fourteen patients were extubated in the recovery room;
one patient was extubated in the ICU later on the first PO day and two
patients were extubated the following morning. Arterial blood gas samples
were obtained daily. Results in patients following extubation are tabulated
in Table 5. Episodes of desaturation below 88% were recorded in two, one
and three patients on POD#1-3 respectively. Atalectasis was noted on exam
or roentgenogram in three, two and two patients on POD#1-3. No patient
required bronchoscopy or reintubation. There were no cases of pneumonia,
respiratory failure or death. One patient, who had undergone previous coronary
bypass grafting and multiple coronary angioplasties, developed unstable
angina pectoris on POD#3 and required angioplasty. Discharge from ICU was
after a mean of 3.29 days (range 0-7). No patient was readmitted to the
ICU. Time of hospital discharge was after a mean of 8 days (range 5-12).
(Table 6). Discussion: Many investigators have attempted to take advantage
of the theoretical advantages of regional anesthesia in the management
of post-thoracotomy pain. Percutaneous intercostal nerve blocks have been
demonstrated to be effective in reducing pain, reducing dosage of opioids
and improving pulmonary function tests in well designed, randomized, double
blind, prospective studies of intercostal nerve block given during surgery
, and post-operatively Although the technique has proven efficacy, logistic
considerations and the patient discomfort and risks associated with repeated
blocks, result in infrequent utilization of this technique in post-operative
patients. A variety of systems of single and multiple indwelling intercostal
catheters have been described to allow serial or continuous regional analgesia.
These techniques have not been adopted by thoracic surgeons for general
usage. The difficulty in explaining how a single intercostal injection
or catheter can provide wide field analgesia has been explained by studies
demonstrating passage of the injected local anesthetic agent retrograde
along the subcostal space into the paravertebral space. While the paravertebral
space can be cannulated percutaneously, there is a high failure rate with
this technique. In an effort to avoid some of the methodological difficulties
inherent in intercostal nerve block, investigators have tried placement
of indwelling intrapleural catheters with intermittent or continuous instillation
of local anesthetic agents for post operative pain control. It is assumed
that the anesthetic will diffuse across the pleura to bathe the intercostal
nerves, thus providing an equivalent to intercostal nerve blocks, without
the need for multiple injections. Three single arm studies have found this
technique to be valuable after thoracotomy in children. and three double-blind,
randomized studies have shown improved pain control, reduced narcotic dose
and improvement of various respiratory parameters. Four other studies have
found the technique to be ineffective, and in two further studies, intrapleural
analgesia was found to be inferior to intercostal block.6, High measured
levels of anesthetic in chest tube effluent help to explain the variable
results . It is important to position the patient supine or operated side
down in order to keep the agent in contact with the posterior parietal
pleura. 13 In 1988, Sabanathan described a technique that combines the
advantages of intercostal block with those of intrapleural analgesia. At
the completion of thoracotomy, the pleura is stripped off the posterior
chest wall over a five rib area, centered on the incision. A percutaneous
plastic catheter is placed in the pocket thus created, and local anesthetic
is infused during closure and for a number of days after surgery into the
paravertebral space.1 Three subsequent randomized double blind studies,
two by the same group and one from other investigators have shown the technique
to be effective, while a fourth study found no advantage over saline infusion.
All of these studies have utilized bupivicaine, presumably because it's
long duration of action carried an advantage when intermittent injections
were used. Toxic plasma levels of bupivicaine are generally considered
to lie between 2 and 4 mg/ml. Numerous studies have used 1:200,000 epinephrine
in conjunction with the bupivicaine. Other groups have avoided it because
of perceived problems with hypertension and tachycardia . There is no improvement
in analgesia with the use of 0.25% vs. 0.5% bupivicaine. Initial doses
of 100 mg. (20 cc of 0.5%) bupivicaine and total doses of 400 mg/24 hours
have generally not resulted in serum levels above 2 mg/ml, but higher levels
can occur especially in children. 13 In one remarkable case 20 cc of 0.375%
bupivicaine was given intrapleural q6 hours for 130 days in a cancer patient.
Plasma levels remained between 1-2mg/ml. at all times. We were concerned
about the incidence of high blood levels of bupivicaine in the series cited
above and the relative dearth of information regarding the pharmacokinetics
and toxicity of bupivicaine. Since regional anesthetic was to be delivered
continuously, there seemed to be little advantage to using a long acting
drug. Accordingly, we elected to utilize lidocaine. Lidocaine, at the dosage
used in this study (1 mg/kg/hour), is routinely given safely, intravenously
in critically ill intensive care unit patients. 300-400 mg/hour of lidocaine
have been given intrapleural with minor toxicity and serum levels at or
below 4.7mg/ml. 28 Our patients experienced no detectable adverse response
to the drug, and measured levels of serum lidocaine remained below the
toxic range (>5 mg/ml.) in all but one patient. At our institution,
verbal pain rating scores are routinely obtained by nurses in post-operative
patients. Optimally, pain scores would have been obtained at rest and with
cough and movement, but since this was a retrospective study, these values
could not be obtained. Nevertheless, the mean pain scores demonstrated
quite good pain control. Nurses and physicians were subjectively impressed
by the apparent level of analgesia provided. Although the PVB provides
analgesia, it is usually not sufficient in itself to provide adequate pain
relief, and adjuvant medications are required. Mean total MS doses/day
were also low. This was true in spite of the fact that the patient who
had bilateral thoracotomies, had by far the highest MS requirements. Low
opioid doses resulted in low sedation scores and an absence of serious
morbidity and mortality. It is not known how this technique compares with
epidural methods. One disadvantage is the technical difficulty of creating
the paravertebral pocket in some patients with pleural disease. On the
positive side, an anesthesiologist with special skills is not required;
it is inexpensive; catheter related malfunction is rare; pruritus was not
seen. In summary, we have confirmed Sabanathan's favorable experience with
the technique of continuous PVB following thoracotomy. We found continuous
PVB with 1% lidocaine to be a simple, safe and effective technique in the
care of our patients. Relative effectiveness of PVB needs to be tested
in prospective, randomized studies comparinkg./hour is started during closure
of the chest and continued until the chest tubes are discontinued. The
originators of the technique note that it provides excellent analgesia
and caused no complications in their experience. I have used the technique
in five patients and was impressed by the analgesia produced , without
the obtundation and respiratory depression associated with narcotic techniques.
I propose to carry out a larger trial of continuous paravertebral block
on the thoracic surgical service at City of Hope. We will place the catheter
in surgery and infuse as above. Although chances of complications appear
to be small, it will be important during the early stages of the study
to carefully monitor the patient for the development of loss of sensation
and motor power in the legs during the infusion. The infusion system will
have to be a closed one so that no inappropriate medications are given.
The catheter will have to be clearly differentiated from an epidural catheter.
Narcotic infusion in the paravertebral area would cause no problems, but
infusion of local anesthetic into an epidural catheter could prove lethal.
The technique does nothing to control the central mechanism of pain perception
and would not be effective for pleural pain outside of the incision area,
e.g., irritation of the diaphragm by the chest tube rubbing on it. It is
necessary therefore to provide a supplementary analgesic. We will use ketorolac
given I.M. while the paravertebral block is in place. If these techniques
provide insufficient pain relief, narcotics provide a fallback position.
The expertise of the Nursing Department at City of Hope in the understanding
and management of pain should provide an opportunity to quantitate in scientific
fashion the value of this technique that would not be available in most
area hospitals. I value your participation in this study and welcome your
feedback. Please do not be shy about suggesting changes and improvements
in the protocol.
Frederic W.
Grannis Jr. M.D
If you have trouble contacting me with the address above,
I may also be reached at 76516,2333@compuserve.com and at fgrannis@cris.com
Back to the Home Page
|