Anesth Analg 2002;94:417-420
© 2002 International Anesthesia Research Society
PAIN MEDICINE
Epidural Infusion Pressure in Degenerative Spinal Disease Before and After Epidural Steroid Therapy
Stuart A. Dunbar, MB*,
P. Manikantan, MD*, and
J. Philip, MD, PhD
*Tufts University School of Medicine, Baystate Medical Center, Springfield, Massachusetts; and Harvard Medical School, Department of Anesthesiology, Brigham and Women’s Hospital, Boston, Massachusetts
Address correspondence and reprint requests to Stuart A. Dunbar, MB, Department of Anesthesiology, Baystate Medical Center, 759 Chestnut Street, Springfield, MA 01199. Address e-mail to Dunbar.StuartMD{at}bhs.org
 |
Abstract
|
|---|
The analgesic mechanism of epidural steroids in reducing pain associated with degenerative spinal disease (DSD) is poorly understood. We report increased inline epidural infusion pressure in patients with DSD and assess whether this phenomenon is affected by administration of an epidural steroid injection. We collected data during epidural placement for routine surgery or epidural steroid therapy. Using a 17-gauge Tuohy needle, with patients in the right lateral decubitus position, loss of resistance to 2 mL of saline identified the epidural space. Two minutes later the needle was attached to saline-filled tubing connected to a pressure transducer (Baxter PX 260 pressure monitoring kit with Truwave TM disposable pressure transducer). In the first part of the study, 4 successive boluses of 3 mL of local anesthetic were administered at a rate of 6 mL/min to 15 patients (age 47 ± 6 yrs) with radicular back pain and magnetic resonance imaging (MRI) or computed tomography (CT) evidence of DSD, and to 8 control patients with no history of back pain (age 44 ± 5 yr) while inline epidural infusion pressure was measured. In the second part of the study 44 patients with low back pain and MRI or CT evidence of DSD presenting to the pain clinic were infused with 8 mL of 0.125% bupivacaine and 40 mg of methylprednisolone (20 mg/mL) at a rate of 6 mL/min while inline epidural infusion pressure was measure and recorded. This was repeated 3 wk later. Initially, DSD patients had significantly increased infusion pressures over normals, which most likely reflects outflow resistance or obstruction. A significant decrease in inline epidural infusion pressure was observed after epidural steroid treatment. This change in pressure may indicate efficacy from epidural steroid injection.
IMPLICATIONS: During injection into the epidural space we observed increased resistance in patients with degenerative spinal disease. This resistance was significantly less when measured 3 wk after an epidural steroid injection. This change in pressure may indicate efficacy from epidural steroid injection.
|
Introduction
|
|---|
The analgesic mechanism of epidural steroids in degenerative spinal disease (DSD) is poorly understood. We assessed epidural infusion pressure in patients with DSD and compared these infusion pressures to a group of patients without a history of back pain. This Control group came to the operating room (OR) for elective knee surgery. We further assessed epidural infusion pressure in the DSD group during and after a subsequent epidural injection of steroid 3 wk later.
|
Methods
|
|---|
The experiments were conducted in two parts with differing paradigms. Permission was obtained from the institutional research committee to collect epidural infusion pressure data during the course of administration of an epidural injection for anesthesia or during epidural steroid therapy in the pain clinic.
Part 1
In the first set of experiments, two groups of patients were selected according to the following criteria: Patients were chosen between 40 and 55 yr of age. Group 1 was chosen from patients coming to the pain clinic for an epidural steroid injection with a history of low back pain and with magnetic resonance imaging (MRI) or computerized tomography (CT) evidence of disk herniation or spinal stenosis. Significant DSD was defined as that described by the radiologist as spinal stenosis attributable to disk herniation, with or without facet hypertrophy, significant enough to occlude a neural foramen at any lumbar level. These patients were assigned to the DSD group. Group 2 was selected from patients coming to the OR for knee surgery using epidural anesthesia with no history of back pain. This group was assigned as the Control group. The selection of patients between 40 and 55 yr of age was based on preliminary sampling that showed the Control group was typically in this age range. The epidural was placed in the right lateral decubitus position in all cases. An exclusion criterion was a desire by the anesthesiologist to place the epidural in the sitting position. The epidural needle was placed midline at the level of a horizontal line between the ileac crests. This ensured, as best as possible without radiographic guidance, that the same level was entered when the patient returned for the second injection. Two minutes after identification of the epidural space by loss of resistance to 2 mL of saline using a 17-gauge Tuohy needle, the pressure measurement and infusion apparatus was attached to the epidural needle, and baseline pressure was recorded. The infusion apparatus was an IVAC Model 560computer-controlled infusion pump (IVAC-Alaris, San Diego, CA) programmed to deliver a controlled volume at a controlled infusion rate while monitoring pressure. Both groups then received a 3-mL bolus of solution, followed by a delay of 3 min, and thereafter three further boluses of 3 mL at the same infusion rate of 6 mL/min at 2-min intervals. This paradigm was chosen because it mirrored the customary practice when starting an epidural anesthetic at the institution when the studies were performed. Pressure was recorded for 2 min after the last bolus. The solution used in Group 1 was 0.125% bupivacaine, whereas the solution used in Group 2 was 2% lidocaine. In Group 1 the local anesthetic infusion was followed by a 2 mL bolus dose of methylprednisolone after termination of the study. In Group 2 after administration of this 12 mL infusion, patients went on to have further local anesthetic administered for surgery as needed. These solutions were normally administered for either treatment in Group 1, or the institution of an anesthetic level for surgery in Group 2. The same volume of solution was used in both groups as well as the same infusion paradigm. It was customary to give incremental boluses of local anesthetic to achieve an anesthetic level for surgery in the institution at that time. The anesthesiologist was in control of the needle at all times and could stop the infusion at any time.
Part 2
In the second part of the study, all subjects were selected randomly from patients coming to the pain clinic for an epidural steroid injection for the first time who also had a history of low back pain with a radicular pattern of pain referral to one or other foot as well as evidence of DSD on MRI or CT. There were no other inclusion or exclusion criteria. After identification of the epidural space by loss of resistance to 2 mL of saline using a 20-gauge Tuohy needle, the pressure measurement and infusion apparatus was attached to the needle. The infusion apparatus was a Baxter Model AS40A approved for epidural infusion. The infusion was administered through a pressure transducer line (Baxter Model PX260) attached directly to the needle. Two minutes after identification of the epidural space, baseline pressures were recorded. The pump was then set to deliver a 10-mL bolus of a mixture of 8 mL of bupivacaine 0.125% and 2 mL of methylprednisolone (20 mg/mL) at 6 mL/min, during which time inline pressure was recorded continuously from a stopcock side arm attached to the epidural needle. Pressure was recorded for a further 100 s after cessation of the infusion. When patients returned to the pain clinic 3 wk later for a scheduled second injection, the procedure was repeated in exactly the same fashion. Just as for the first part of the study, the procedure involved administration of an epidural injection of steroid according to the practice at the institution at the time, with the only exception being that the injection was administered by an epidural pump approved for this purpose.
Statistics
Data from the two groups in Part 1 of the study, corresponding to baseline, 30, 180, 210, 240, 270, 300, 330, and 360 min, were compared by analysis of variance with significance at P < 0.05. A post hoc Scheffé test (95% confidence interval) was applied in all cases of stated significance. In Part 2 of the study, data were analyzed at 10-s intervals and data from the two infusions were analyzed by a paired Student’s t-test, with significance at P < 0.05.
|
Results
|
|---|
Part 1
There was no significant difference in age between DSD patients (Group 1) (mean age 47 ± 6 yr, n = 15) and controls (Group 2) (44 ± 5 yr, n = 8). DSD patients had significantly increased peak and residual infusion pressures over Group 2 at 240, 270, 300, 330, and 360 s after the onset of the infusion, analysis of variance P < 0.05, Scheffé 95% (Figure 1).
View larger version (20K):
[in this window]
[in a new window]
|
Figure 1. Mean peak and residual pressures (± SE) during the administration of 4 successive 3-mL epidural local anesthetic boluses (6 mL/min) in patients with a history of low back pain with a radicular pattern of referral to the lower extremity, and imaging evidence of either a herniated disk or spinal stenosis (degenerative spinal disease [DSD], n = 15, mean age 47 ± 6 yr, solution 0.125% bupivacaine) and a second group that had no history of low back pain presenting for knee surgery, (Normals, n = 8, mean age 44 ± 5 yr, solution 2% lidocaine). *P < 0.05, DSD versus Normals, analysis of variance, Scheffé 95%. Data shows that DSD was associated with an increase in inline epidural infusion pressure.
|
|
Part 2
Infusion pressures from 30 s through 100 s were significantly less during the second infusion, paired Student’s t-test, P < 0.05, Figure 2. No significant differences in residual infusion pressures were observed after discontinuation of the infusion.
View larger version (17K):
[in this window]
[in a new window]
|
Figure 2. Inline epidural infusion pressures in patients presenting for an epidural steroid injection for the first time with low back pain (n = 44) and a radicular pattern of referral during and 3 wk after administration of an epidural steroid injection of 8 mL of 0.125% bupivacaine and 2 mL of methylprednisolone (20 mg/mL). Data show the mean pressure readings at 10 min intervals during first and second injections. *P < 0.05 first versus second epidural steroid injection, paired Student’s t-tests.
|
|
|
Discussion
|
|---|
Epidural steroids are widely used in the treatment of pain associated with spinal disease yet their mechanism of action is poorly understood. In the first part of this study we observed increased inline infusion pressure during epidural injection in patients with DSD (Fig. 1). To further evaluate this observation we analyzed the change in infusion pressure during a first and second epidural steroid injection in patients with DSD and observed significantly decreased infusion pressure in patients with DSD between first and second infusions after an interval of three weeks (Fig. 2). Although infusion pressure significantly decreased during the second treatment, this effect may be secondary to causes unrelated to the steroid administered. Thus, simply entering the epidural space and/or infusing a bolus of local anesthetic could have led to a decrease in pressure on reentering the space three weeks later. However, we would propose that the likely explanation for this effect is secondary to the effect of the steroid component of the injection. We did not attempt to control for these potential variables because it would have compromised the current practice and care in our clinic that involves the administration of local anesthetic and steroid.
Although several studies of epidural infusion pressure have been performed in normal, obese (1,2), and pregnant patients (3), there are few studies in patients with spinal disease (4). Studies have shown that intraabdominal pressure, subarachnoid pressure, intrathoracic pressure, and epidural venous pressure all have a direct effect on epidural pressure (1). The effect of position has also been examined and epidural pressure has been shown to vary in the parturient in different positions (3,5,6). Other studies have looked at pressure changes in response to epidural infusions of saline, local anesthetic and blood (7–9). Husemeyer and White (10) examined the pressure changes of infusions of local anesthetics into the epidural space in the parturient. They investigated the pressure response to a continuous infusion of local anesthetic infused at a constant rate over a time range of 30–70 seconds at the level of the L2-3 interspace using a compressed carbon dioxide syringe pump. The rate of injection was altered from 0.33 to 0.143 mL/s, (19.8 mL/min to 85.8 mL/min) and it was found that the maximum pressures generated at the site of injection showed no relationship to the injection rate. In addition, there was no significant correlation between the extent of analgesia, the rate of injection, the maximum injection pressure, or the residual pressure two minutes after injection. However, they did observe that although there was a large variation in the pressures obtained there was a constant pattern generated by each pressure curve.
Pressure change in the epidural space is a complex relationship between subarachnoid pressure, intraabdominal pressure and epidural venous plexus pressure. To investigate the effect of an infusion it is necessary to examine the compliance of the system. This has a direct effect on the spread of infusate in the space, which contains a number of elastic components in what can be described as a rigid leaky container. The generation of pressure by an infusion will not follow a normal compliance curve dictated by pressure and volume relationship alone because of continuing leakage of fluid out of the space. To examine the system, it is necessary to consider the resistance to flow out of the system and the speed of flow into the system. This cannot be done easily with an epidural catheter because the narrow gauge of the catheter and resulting high inflow resistance confound the ability to measure pressure and resistance distal to the catheter in the epidural space (11,12). Thus, to measure the dynamic response of the system it is necessary to use a system that offers little resistance to injection as compared with the resistance in the space itself, so that the time constants of the events being measured are much greater than that of the system itself.
The infusion model used here for examining the response of the system to injection of a fixed inflow rate has been used previously (4). In this study an attempt was made to compare the infusion pressures of patients with and without degenerative spinal disease and to examine any effect that epidural steroid therapy might have on such pressures. In the first part of the study, the response to successive infusions of epidural saline showed a gradual increase in peak and trough infusion pressures toward a plateau pressure. At plateau pressure the infusion pressure would be equal to that in the space, however this equilibrium is not always achieved. This methodological problem negates the steady-state infusion pressure as a target in making comparisons between groups. Despite this, we observed and compared significant differences in mean peak and trough infusion pressures between groups. The significant increase in peak infusion pressures seen in patients with DSD can be explained by a reduction in elastance in the epidural spaces of these patients.
Usubiaga et al. (8) reported that elderly patients given an epidural bolus of local anesthetic had unpredictable and often high levels of anesthesia. This study would support the hypothesis that spinal stenosis in the elderly is the likely cause of higher epidural anesthetic levels because, as was observed in this study, higher infusion pressures were evident in patients with DSD when age was not a factor. Not all elderly patients have DSD explaining the wide and unpredictable variation in infusion pressures seen in this age group in Usubiaga et al.’s study. In the second part of the study, when administration of the epidural infusion of local anesthetic and steroid was repeated three weeks later, there was a significant decrease in infusion pressure that, as discussed, could be secondary to either local anesthetic or steroid, or to simply entering the space and infusing a volume of fluid. Further investigation controlling for these effects would appear to be indicated based on these preliminary results but, as discussed, this would involve obtaining consent to perform a randomized blinded prospective study. Spinal surgical decompression procedures are currently used to treat back pain where some degree of spinal stenosis compromising exiting lumbar roots has been identified. Thus it is possible that epidural steroids may produce pain relief by reducing edema and decompressing the spinal canal. Even a small alteration in the degree of stenosis could presumably relieve pain. This may be reflected in a change in infusion pressure as was observed in this study. To further evaluate this hypothesis it would be necessary to correlate any observed effect on epidural infusion pressure with changes in pain levels.
In summary, increased inline infusion pressure was observed in patients with DSD, most likely reflecting outflow obstruction. An increase in epidural infusion pressure and spread of local anesthetic has been reportedly associated with increasing age (8). In this study there was no significant difference in the age profile of selected DSD patients and a control group of patients without back pain, thus implicating DSD as the likely cause of decreased elastance. A significant decrease in infusion pressure was observed after epidural steroid therapy, which could be secondary to either steroid, local anesthetic, or the volume in which these drugs were administered. A prospective study to control for these variables would be necessary to evaluate whether the steroid component of such treatment is responsible for this effect on infusion pressure.
|
References
|
|---|
-
Hodgkinson R, Husain FJ. Obesity, gravity and spread of epidural anesthesia. Anesth Analg 1981; 60: 421–4.[Abstract/Free Full Text]
-
Hodgkinson R, Husain FJ. Obesity and the cephalad spread of analgesia following epidural administration of bupivacaine for cesarean section. Anesth Analg 1980; 59: 89–92.[Abstract/Free Full Text]
-
Messih M. Epidural space pressures during pregnancy. Anesthesia 1989; 44: 750–2.[Web of Science][Medline]
-
Rocco AG, Philip JH, Boas RA, Scott D. Epidural space as a Starling resistor and elevation of inflow resistance in a diseased epidural space. Reg Anesth 1997; 22: 167–77.[Web of Science][Medline]
-
Galbert MW. Extradural pressures in the parturient patient. Anesthesiology 1974; 40: 499–502.[Web of Science][Medline]
-
Bromage PR. Physiology and pharmacology of epidural analgesia. Anesthesiology 1967; 28: 592–622.[Web of Science][Medline]
-
Usubiaga JE, Usubiaga LE, Brea LM. Effects of saline injection on the epidural and subarachnoid pressure and its relation to post spinal anesthesia headache. Anesth Analg 1967; 46: 293–6.[Free Full Text]
-
Usubiaga LE, Wilkinski JA, Usubiaga LE. Epidural pressure and its relationship to spread of local anesthetic solutions in the epidural space. Anesth Analg 1967; 46: 440–6.[Free Full Text]
-
Coombs DW, Hooper D. Subarachnoid pressure with epidural "patch". Reg Anesth 1979; 4: 3–6.
-
Husemeyer RP, White DC. Lumbar extradural injection pressures in pregnant women: an investigation of relationships between rate of infection, injection pressures and extent of analgesia. Br J Anesth 1980; 52: 55–9.[Abstract/Free Full Text]
-
Hughes NJ, Johnson MD, Datta S, Philip JH. Epidural space resistance cannot be measured through a standard lumbar catheter. Proceedings of the Annual Meeting of the Society for Obstetric Anesthesia and Perinatology. Boston: 1991.
-
Hughes NJ, Johnson MD, Datta S, Philip JH. Epidural space resistance monitoring is difficult during continuous epidural infusion. Reg Anesth 1992; 17: 292–4.[Web of Science][Medline]
Accepted for publication October 19, 2001.
|
Top
Abstract
Introduction
