Spinal Cord Bioavailability of Methylprednisolone After

Home , Methylprednisolone, Pancuronium bromide

156

Anesthesiology 2000; 92:156–63 © 2000 American Society of Anesthesiologists, Inc. Lippincott Williams & Wilkins, Inc. Spinal Cord Bioavailability of Methylprednisolone after Intravenous and Intrathecal Administration The Role of P-Glycoprotein Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/92/1/156/398816/0000542-200001000-00027.pdf by guest on 30 September 2021 Kari L. Koszdin, D.V.M.,* Danny D. Shen, Ph.D.,† Christopher M. Bernards, M.D.‡

Background: High-dose intravenously administered methyl- cord bioavailability of methylprednisolone is poor after intra- prednisolone has been shown to improve outcome after spinal venous administration. The studies in knockout mice suggest cord injury. The resultant glucocorticoid-induced immunosup- that this poor bioavailability results from P-glycoprotein–medi- pression, however, results in multiple complications including ated exclusion of methylprednisolone from the spinal cord. ;(sepsis, pneumonia, and wound infection. These complications (Key words: Friend leukemia virus strain B; mdr 1a/1b (؊/؊ could be reduced by techniques that increase the spinal bio- mice.) availability of intravenously administered methylprednisolone while simultaneously decreasing plasma bioavailability. This HIGH-dose methylprednisolone has been shown to be an study aimed to characterize the spinal and plasma bioavailability of methylprednisolone after intravenous and intrathecal effective treatment for acute spinal cord injury. The administration and to identify barriers to the distribution of Third National Acute Spinal Cord Injury Study demon- methylprednisolone from plasma into spinal cord. strated that patients receiving an intensive 24- to 48-h Methods: The spinal and plasma pharmacokinetics of intra- -؊1 ؊1 intravenous methylprednisolone regimen (30-mg/kg bo venous (30-mg/kg bolus dose plus 5.4 mg ⅐ kg ⅐ h ) and Ϫ1 Ϫ1 ؊ ؊ lus dose plus 5.4 mg ⅐ kg ⅐ h ) within8hoftheinjury intrathecal (1-mg/kg bolus dose plus 1 mg ⅐ kg 1 ⅐ h 1) methylprednisolone infusions were compared in pigs. In addition, had improved 6-month recovery compared with place- 1 wild-type mice and P-glycoprotein knockout mice were used to bo-treated patients. This “megadose” steroid therapy determine the role of P-glycoprotein in limiting spinal bioavail- was not without adverse effects, however. Patients in ability of methylprednisolone. the methylprednisolone treatment groups experienced a Results: Despite the greater intravenous dose, concentrations 2.6-fold increase in the incidence of severe pneumonia of methylprednisolone in pig spinal cord were far higher and plasma concentrations much lower after intrathecal adminis- and sepsis, an increased incidence of wound infection, tration. After intraperitoneal administration in the mouse, the an increased number of days spent receiving mechanical concentrations of methylprednisolone in muscle were not dif- ventilation, and an increased number of days spent in the ferent between mice expressing P-glycoprotein (2.39 ؎ 1.79 intensive care unit.1–4 These complications are believed ␮ ؎ ␮ g/g) and those lacking P-glycoprotein (2.83 0.46 g/g). In to result from glucocorticoid-induced immune suppres- contrast, methylprednisolone was undetectable in spinal cords of wild-type mice, whereas concentrations in spinal cords of sion. P-glycoprotein–deﬁcient mice were similar to those in skeletal In addition to these documented complications, there -muscle (2.83 ؎ 0.27 ␮g/g). are also valid theoretical concerns associated with high Conclusions: These pig studies demonstrate that the spinal dose glucocorticoid therapy in the setting of neural injury. These include steroid-induced hyperglycemia and * Senior Fellow, Department of Comparative Medicine. sepsis-related hypotension, both of which may contrib- 5,6 † Professor, Department of Pharmaceutics. ute to secondary neuronal injury. This may explain ‡ Associate Professor, Department of Anesthesiology. why some studies investigating use of high-dose steroids Received from the Departments of Comparative Medicine, Pharma- for spinal cord injury have failed to demonstrate a ben- 7,8 ceutics, and Anesthesiology, University of Washington, Seattle, Wash- eﬁt. ington. Submitted for publication April 15, 1999. Accepted for publi- One approach to improving methylprednisolone ther- cation July 22, 1999. Supported in part by grants no. RO1 NIDA apy in acute spinal cord injury would be to increase the RR07019-06 and RO1 NS 38911-01 from the National Institutes of Health, Bethesda, Maryland. fraction of the steroid dose reaching the spinal cord while minimizing systemic drug exposure. This goal Address reprint requests to Dr. Bernards: Department of Anesthesi- ology, University of Washington, Box 356540, Seattle, Washington could be accomplished by administering methylpred- 98195. Address electronic mail to: [email protected] nisolone directly at the site of injury or by developing

Anesthesiology, V 92, No 1, Jan 2000 157

SPINAL BIOAVAILABILITY OF METHYLPREDNISOLONE

strategies to increase spinal cord drug exposure after 0.8% and 2.0%. The left femoral vein was cannulated for intravenous administration of methylprednisolone. infusion of either methylprednisolone or 0.9% saline (70 This study therefore was designed to characterize the ml/h) containing pancuronium bromide (0.04 mg/ml). spinal cord and plasma bioavailability of methylpred- Pancuronium bromide was used to ensure that the ani- nisolone after intravenous and intrathecal administra- mal would not move during and after placement of the tion. This was accomplished using a pig microdialysis microdialysis probe. When methylprednisolone was ad-

model to characterize the spinal cord, cerebrospinal ministered intravenously, the saline/pancuronium solu- Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/92/1/156/398816/0000542-200001000-00027.pdf by guest on 30 September 2021 fluid (CSF), and plasma pharmacokinetics of methylpred- tion was administered into an ear vein. Rectal tempera- nisolone during intravenous and intrathecal administra- ture was maintained at 38°C using a servo controlled tion. heat lamp and a rectal temperature probe (YSI model An additional goal was to identify potential barriers to 73A; Yellow Springs Instrument Co., Yellow Springs, the spinal cord bioavailability of methylprednisolone af- OH). ter intravenous administration. One potential barrier is Manufacture of Microdialysis Probes. Two types of P-glycoprotein (mdr1 gene product),9 an efflux trans- microdialysis probes were used to sample concentra- porter for which methylprednisolone has been identified tions of methylprednisolone in this study: linear probes, recently as a substrate. This membrane-bound protein which were inserted into the spinal cord, and loop has been shown to limit the cellular accumulation of probes, which were inserted into the subarachnoid many drugs and is expressed in spinal cord capillary space. Both types of probes were prepared from cellu- endothelium.10 Thus, we hypothesized that the poor lose microdialysis fibers (Spectrum Medical Industries, bioavailability of methylprednisolone after intravenous Houston, TX) with a 215-␮m ID, a 235-␮m OD, and a administration results from active exclusion of the drug molecular weight cutoff of 6,000 D. India ink was used from the spinal cord by P-glycoprotein. To address this to paint calibration marks on the dialysis probes to de- question, we used a transgenic mouse model that lacks fine a “dialysis window.” For the linear probes, the dial- two key P-glycoprotein genes (mdr 1a and mdr 1b)to ysis window was 2 mm long; for the loop probes, the determine the role of this transporter in limiting the dialysis window was 20 mm long. Epoxy cement was spinal cord bioavailability of methylprednisolone after spread evenly over the probes on either side of the systemic administration. dialysis windows by running a 1-cm length of polyethylene 10 tubing along the dialysis fiber. The polyethylene 10 tubing had an ID of 280 ␮m; thus, the finished dialysis Methods probes had an OD of 280 ␮m. A 90-␮m diameter wire was inserted into the lumen of All experiments were performed in accordance with a the loop dialysis probe, and the probe was bent at the protocol approved by the Animal Care and Use Commit- center of the dialysis window to form a dialysis loop. The tee at the University of Washington. American Associa- wire ensured that the probe remained patent after being tion for Laboratory Animal Care guidelines were fol- bent. A conical bead of silicone caulk was then placed lowed throughout. along the neck of the probe. For intrathecal administration of methylprednisolone, an intrathecal catheter (Bec- Pig Microdialysis ton Dickinson and Co., Franklin Lakes, NJ) was secured Animals. Mixed-breed pigs (n ϭ 20) of both sexes, to the neck of the lumbar loop probe with silicone caulk. weighing 10–15 kg, were used. Each animal was anes- The probes were allowed to cure for Ն24 h but Ͻ72 h thetized via face mask with halothane and nitrous oxide before insertion. (70%) in oxygen. After intramuscular injection of succi- In Vitro Calibration of Probes. Four loop probes nylcholine (100–200 mg), the pigs were intubated oro- and four linear probes, which had not been implanted in tracheally and ventilated mechanically. Minute ventila- animals, were placed in a solution of methylpred- Ϯ ␮ tion was adjusted to maintain end-tidal CO2 at 40 3 nisolone (5 g/ml) and perfused with mock CSF for 1 h mmHg. at 10 ␮l/min. Samples were collected at 10-min intervals The left femoral artery was cannulated for blood pres- and analyzed for concentration of methylprednisolone. sure monitoring and blood sampling. Mean arterial pres- Probe efficiency was calculated by dividing the mea- sure was maintained between 60 and 100 mmHg by sured dialysate concentration of methylprednisolone by adjusting the inspired halothane concentration between the known concentration in the solution. Mean and SD

Anesthesiology, V 92, No 1, Jan 2000 158

KOSZDIN ET AL.

were calculated for each set of probes. The in vitro nisolone regimen currently used to treat acute spinal probe efﬁciency was 21 Ϯ 8% for the loop and linear cord injury in humans. probes. After administration of drug by either route, samples of Placement of the Dialysis Probe. To access the dialysate were collected continuously between the fol- spinal cord, the vertebral bodies at L4 and T13 were lowing time points: 0, 5, 10, 15, 20, 25, 30, 35, 40, 45, exposed bilaterally, and a dorsal laminectomy was per- 50, 55, 60, 70, 80, 90, 100, 110, 120, 140, 160, 180, 200,

formed at these two sites. Microdialysis probes were 220, and 240 min. As a result, each sample represents Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/92/1/156/398816/0000542-200001000-00027.pdf by guest on 30 September 2021 then inserted into the spinal cord and the CSF at both the average spinal cord or CSF dialysate concentration of vertebral levels. To insert the microdialysis probes into methylprednisolone present during the period for which the spinal cord, a hollow 30-gauge needle was inserted each sample was collected. through the meninges and spinal cord from the center of Arterial blood samples were collected at 0, 5, 20, 40, the left lateral aspect just caudal to the spinal nerve. A 60, 90, 120, 150, 180, 210, and 240 min for determina- wire glued into the dialysis probe lumen was then in- tion of concentration of methylprednisolone in plasma. serted into this needle and the needle removed. The After 4 h, the pig was killed with an intravenous injec- wire was then used to pull the dialysis probe through the tion of saturated KCl. Spinal cord sections (each 1 cm spinal cord so that the dialysis window was in the center long) from the site of insertion of the microdialysis Ϫ of the spinal cord. The probe was then secured with probe were removed and frozen at 20°C until analysis cyanoacrylate glue. The placement of the probe in spinal for concentrations of methylprednisolone. cord tissue was verified at the end of the experiment when the cord was removed for tissue analysis for meth- Mouse Experiments Mdr 1a/1b Ϫ Ϫ ylprednisolone. ( / ) on a friend leukemia virus strain B background and wild-type male mice with friend leuke- To insert microdialysis probes into the CSF, a small mia virus strain B (weight, 25–30 g) were obtained from (1-mm) incision was made in the dorsal aspect of the Taconic Farms (Germantown, NY). To determine the dura and arachnoid mater, and the loop portion of the role of P-glycoprotein in limiting entry of methylpred- probe was inserted 1 cm into the subarachnoid space. nisolone into the spinal cord, the mice were given an The meningeal hole was sealed by the tapered silicone intraperitoneal injection of methylprednisolone (15 mg/ plug affixed to the neck of the probe, and the probe was kg) and were killed 1 h later with an overdose of halo- secured in place with cyanoacrylate glue. When methyl- thane. The entire spinal cord and a leg muscle specimen prednisolone was to be administered intrathecally, a were removed from each mouse and stored at Ϫ20°C loop probe with an attached intrathecal catheter was until analysis for concentrations of methylprednisolone. inserted into the lumbar intrathecal space. Mock CSF (NaCl, 140 mEq; NaHCO3, 25 mEq; KCl, 2.9 Analysis of Methylprednisolone mEq; MgCl2, 0.4 mEq; CaCl2, 2.0 mEq; urea, 3.5 mM; Concentrations of methylprednisolone were measured glucose, 4.0 mM; pH 7.38–7.42; 295 mOsm) was oxygen- using high-performance liquid chromatography analysis ated and adjusted to pH 7.3 by bubbling with 95% O2/5% (Hewlett-Packard 1,050 series model 79,853C, Palo Alto, CO2 and was then pumped through the dialysis probes CA) with ultraviolet detection at 254 nm in a modifica- ␮ at 10 l/min. tion of a previously published method.11 Chromato- Administration of Drug. At time 0, methylpred- graphic separation was achieved on a Supelco 5-␮m nisolone sodium succinate (Solu-Medrol; Pharmacia Up- LC-18-db 15-cm ϫ 4.5-mm column (Supelco Inc., Belle- john Co., Kalamazoo, MI) was injected either intrathe- fonte, PA) using a mobile phase of 65:35 vol/vol mixture cally or intravenously. The intrathecal dose was of 0.01 M KH2PO4:acetonitrile, at a flow rate of 1 ml/min. administered as a 1-mg/kg bolus dose in a volume of 100 For the spinal cord and CSF dialysates, samples were ␮l given over 1 min. The bolus dose was followed by a injected directly onto the column, and the sample peak Ϫ Ϫ continuous infusion of 1 mg ⅐ kg 1 ⅐ h 1, for a total dose heights were compared with a standard curve prepared of 5 mg/kg over the 4-h infusion. The intravenous dose from methylprednisolone standards bracketing expected was administered as a 30-mg/kg bolus dose given over 1 sample concentrations. For the samples of plasma and min, followed by a continuous infusion at the rate of 5.4 tissue, 700 ng fluoxymestrone (internal standard) was mg ⅐ kgϪ1 ⅐ hϪ1, for a total dose of methylprednisolone of added to each sample. Samples were extracted with 4 ml 52 mg/kg over the 4-h infusion; this is the methylpred- methylene chloride and the aqueous layer discarded.

Anesthesiology, V 92, No 1, Jan 2000 159

SPINAL BIOAVAILABILITY OF METHYLPREDNISOLONE

Fig. 1. Concentrations of methylprednisolone (mean ؎ SD) in pig spinal cord Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/92/1/156/398816/0000542-200001000-00027.pdf by guest on 30 September 2021 dialysates after either intravenous (IV) or intrathecal (IT) administration. TC- ؍ thoracic cord—IT delivery; LC-IT ؍ IT -tho ؍ lumbar cord—IT delivery; TC-IV lumbar ؍ racic cord—IV delivery; LC-IV cord—IV delivery. All data were corrected for dialysis probe efﬁciency.

The organic layer was washed once with 4 ml 0.05 M reported as mean Ϯ SD, and the differences between the NaOH and again the aqueous layer discarded. The meth- mdr 1a/1b (Ϫ/Ϫ) and wild-type mice were assessed by ylene chloride extracts were evaporated to dryness and Student unpaired t test. the residue reconstituted in the high-performance liquid chromatography mobile phase. Calibration standards were prepared in the same manner using blank plasma Results spiked with methylprednisolone. The analyte/internal standard peak height ratio was plotted, and sample con- Pig Studies centrations were estimated from the standard curve. The Figure 1 shows the concentration of methylpred- limits of quantification for the assays were 10 ng/ml for nisolone in the dialysates from thoracic and lumbar spi- the plasma, 50 ng/g for tissue samples, and 45 ng/ml for nal cord segments over time. The steady-state concen- the dialysates. trations of methylprednisolone were greater in the thoracic and lumbar cord dialysates after intrathecal Statistical and Pharmacokinetic Analyses compared with intravenous administration (table 1). For the pig pharmacokinetic studies, the individual Concentrations of methylprednisolone in spinal cord animal data were analyzed by calculating the area under tissue samples confirmed this large difference in bioavail- the concentration–versus–time curve (AUC) over4hfor ability between routes (table 2). There were no statisti- methylprednisolone in plasma and in dialysates from cally significant differences between concentrations in spinal cord and CSF for each pig, using the trapezoidal dialysate at the two spinal cord sites with either intrave- rule. In addition, steady-state concentrations of methyl- nous or intrathecal administration (intravenous, P ϭ prednisolone in these three compartments were deter- 0.72; intrathecal, P ϭ 0.55). With intrathecal administra- mined by averaging concentrations of drug in dialysate tion, a slight time lag was observed before methylpred- or plasma over the interval of 100–240 min for each pig. nisolone reached peak concentration in the thoracic Because the data were not distributed normally, medians cord dialysate (fig. 1). At steady-state, however, concen- and 25th–75th percentiles were calculated for the trations of methylprednisolone were the same for both steady-state concentration and AUC data across each spinal cord segments after intrathecal administration. group of pigs. The Mann–Whitney rank-sum test was After intravenous administration, peak concentrations of used to compare medians between intravenous and in- methylprednisolone were reached simultaneously at trathecal administration groups. both spinal cord sites. The data on spinal cord and muscle concentrations of Figure 2 shows concentrations of methylprednisolone methylprednisolone from the mouse experiments are over time in the CSF dialysate after either intravenous or

Anesthesiology, V 92, No 1, Jan 2000 160

KOSZDIN ET AL.

Table 1. Steady-state Methylprednisolone Concentrations in Cerebrospinal Fluid and Spinal Cord Dialysates and Plasma

IV (␮g/ml) IT (␮g/ml)

25th–75th 25th–75th Median Percentiles n Median Percentiles n IT/IV P*

Thoracic CSF 0.058 0.044–0.067 9 164 32.1–235 10 2,828 Ͻ 0.001 Lumbar CSF 0.043 0.028–0.068 8 87 29.4–170 10 2,023 Ͻ 0.001 Thoracic cord 0.024 0.013–0.047 6 1.27 0.307–2.21 9 52.9 0.002 Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/92/1/156/398816/0000542-200001000-00027.pdf by guest on 30 September 2021 Lumbar cord 0.026 0.012–0.04 8 1.09 0.571–2.92 10 41.9 Ͻ 0.001 Plasma 3.96 3.19–6.76 9 0.192 0.145–0.278 11 0.048 Ͻ 0.001

Steady state was determined by taking the mean dialysate or plasma concentration over the 100–240-min period for each animal. IV ϭ intravenous; IT ϭ intrathecal; CSF ϭ cerebrospinal fluid. * Significant probability from the Mann–Whitney rank sum test. intrathecal administration. As with the spinal cord dialy- intravenous administration, although the total dose of sates, concentrations of methylprednisolone in the CSF methylprednisolone was more than 10-fold lower in the dialysates were greater after intrathecal compared with case of intrathecal administration. intravenous administration (table 1). However, the intrathecal/intravenous ratio of steady-state concentrations of Mouse Studies drug in the CSF is much higher than for the spinal cord. Concentrations of methylprednisolone in skeletal mus- In contrast with the spinal cord and CSF data, the cle did not differ between the mdr 1a/1b knockout concentration of methylprednisolone in plasma was mice (2.83 Ϯ 0.46 ␮g/g) and the wild-type mice (2.39 Ϯ lower after intrathecal than after intravenous administra- 1.79 ␮g/g). The spinal cord concentration of methyl- tion (fig. 3 and table 1). prednisolone in mdr 1a/1b knockout mice (2.83 Ϯ 0.27 To better quantify differences between spinal cord and ␮g/g) was nearly identical to that in skeletal muscle, plasma exposure to methylprednisolone, we calculated a whereas the drug was not detectable in the spinal cord relative spinal cord exposure index (REI) for each route tissue of the wild-type mice. Muscle tissue was chosen as of administration of methylprednisolone. The REI was a comparison tissue because previous work has shown defined as the ratio of the AUCs (table 3) for the spinal no difference in the accumulation of drug within muscle ϭ cord and plasma (REI AUCspinal cord/AUCplasma). The between mdr 1a/1b (Ϫ/Ϫ) and wild-type mice for sev- greater the REI, the greater the spinal cord exposure to eral P-glycoprotein substrates.11–13 methylprednisolone relative to plasma exposure. After intrathecal administration, the median thoracic cord REI was 29 (range, 3.38–36.85), and the median lumbar cord Discussion REI was 52.7 (range, 10.4–87.1). In contrast, after intravenous administration, the median thoracic cord REI was Using a pig microdialysis model, we have demon- only 0.015 (range, 0.012–0.023), and the median lumbar strated that steady-state concentrations of methylpred- cord REI was 0.015 (range, 0.001–0.021). Therefore, the nisolone and total spinal cord exposure to methylpred- spinal cord REI after intrathecal administration is a re- nisolone are substantially greater after intrathecal markable 1,933–3,513 times greater than the REI for administration than intravenous administration. We

Table 2. Methylprednisolone Concentration in Spinal Cord Tissue after Intravenous or Intrathecal Administration

IV (␮g/g) IT (␮g/g)

25th–75th 25th–75th Median Percentile Median Percentile IT/IV P*

TC 3.01 2.25–4.3 309 142–657 103.0 0.002 LC 3.03 2.68–3.75 278 140–635 91.7 Ͻ 0.001

IV groups, n ϭ 9; IT group, n ϭ 11. IV ϭ intravenous; IT ϭ intrathecal; TC ϭ thoracic cord; LC ϭ lumbar cord. * Signiﬁcant probability using the Mann–Whitney rank sum test.

Anesthesiology, V 92, No 1, Jan 2000 161

SPINAL BIOAVAILABILITY OF METHYLPREDNISOLONE

Fig. 2. Concentrations of methylprednisolone (mean ؎ SD) in pig cerebrospinal ﬂuid dialysates after either intravenous Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/92/1/156/398816/0000542-200001000-00027.pdf by guest on 30 September 2021 (IV) or intrathecal (IT) administration. ;thoracic intrathecal—IT delivery ؍ TIT-IT ;lumbar intrathecal—IT delivery ؍ LC-IT ;thoracic intrathecal—IV delivery ؍ TC-IV .lumbar intrathecal—IV delivery ؍ LC-IV All data were corrected for dialysis probe efﬁciency.

found this to be true although the total intravenous dose prednisolone after intravenous administration is poor of methylprednisolone was 10 times greater than the compared with intrathecal administration. intrathecal dose. Microdialysis underestimates the actual Although a previous study demonstrated that higher concentration of methylprednisolone in the tissue or concentrations of methylprednisolone can be achieved fluid. The dialysis recovery rate depends on probe dial- in CSF and brain tissue after intrathecal bolus adminis- ysis efficiency and dialysis flow rate. The in vitro recov- tration,12 this is the first study to compare the spinal ery rate for our probes was 21 Ϯ 8%. Presumably, the in cord pharmacokinetics of the drug after a bolus dose vivo recovery rate was similar or lower than the in vitro plus infusion (as is done clinically) via the two routes of rate but was comparable for the probes in the intrathecal administration. and intravenous administration regimens. Therefore, the One obvious explanation for the low spinal bioavail- dialysate concentrations of methylprednisolone can be ability of methylprednisolone after intravenous adminis- compared directly for the two routes of administration. tration is poor penetration of the blood–spinal cord Clearly, the spinal cord tissue bioavailability of methyl- barrier. Because methylprednisolone is a small, hydro-

Fig. 3. Concentrations of methylprednisolone (mean ؎ SD) in pig plasma after either intravenous (IV) or intrathecal (IT) administration.

Anesthesiology, V 92, No 1, Jan 2000 162

KOSZDIN ET AL.

Table 3. Integral Exposure to Methylprednisolone in Cerebrospinal Fluid and Spinal Cord Dialysates and Plasma

IV (␮g ⅐ mlϪ1 ⅐ minϪ1)IT(␮g ⅐ mlϪ1 ⅐ minϪ1)

25th–75th 25th–75th Median Percentile n Median Percentile n IT/IV P*

Thoracic CSF 13.3 10.3–16.9 9 34,450 14,790–41,807 10 2,590 Ͻ 0.001 Lumbar CSF 10.3 7.1–19.1 8 18,670 8,320–43,900 10 1,813 Ͻ 0.001 Thoracic cord 25.6 16–38.2 6 1,480 695–2,210 8 57.8 Ͻ 0.001 Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/92/1/156/398816/0000542-200001000-00027.pdf by guest on 30 September 2021 Lumbar cord 28.4 15.9–35.1 8 2,670 817–3,700 10 94.1 Ͻ 0.001 Plasma 1680 1,250–2,390 9 46.3 29.4–55.3 11 0.028 Ͻ 0.001

Area under the curve values were calculated for individual animals using the trapazoidal rule. IV ϭ intravenous; IT ϭ intrathecal; CSF ϭ cerebrospinal fluid. * Significant probability from the Mann–Whitney rank sum test. phobic molecule, however, it would be expected to rior to intravenous administration. Greater concen- cross endothelial barriers passively with relative ease. trations of methylprednisolone are achieved in the target The finding that it does not raises the possibility that organ, whereas plasma drug exposure is markedly re- methylprednisolone is actively excluded from the spinal duced. However, it must be kept in mind that greater cord. Several lines of evidence suggested that this might concentrations of methylprednisolone in spinal cord be the case. First, other glucocorticoid molecules have may not be necessary or even desirable. Animal studies been shown to be actively excluded from human cells in suggest that methylprednisolone exhibits a biphasic bio- vitro by P-glycoprotein, an ATP-driven efflux trans- chemical and clinical effect in spinal cord injury mod- porter.13 Second, P-glycoprotein present on brain capil- els.20,21 Low doses have no benefit, whereas doses lary endothelial cells has been shown to be an important higher than those in clinical use actually worsen neuro- part of the blood–brain barrier.14–16 Third, P-glycopro- logic outcomes. tein–deficient mice have enhanced penetration of dexa- Thus, the benefit of the intrathecal route of adminis- methasone into brain tissue.17,18 Finally, P-glycoprotein tration may not be the greater concentrations of meth- has been shown to play an important role in restricting ylprednisolone in spinal cord that can be achieved. absorption of methylprednisolone from the rat small Rather, the real benefit is likely to be reduced concen- intestine.9 trations of methylprednisolone in plasma and a conse- Based on these observations, we hypothesized that quent reduction in the incidence and severity of steroid- P-glycoprotein present on spinal capillary endothelial induced complications such as sepsis, wound infection, cells was responsible for the poor penetration of the and pneumonia. drug from blood into spinal cord. Using mdr 1a/1b In addition, this study suggests possible strategies to (Ϫ/Ϫ) mice, which lack functional P-glycoprotein,19 we overcome the poor spinal cord penetration of methyl- were able to confirm this hypothesis. Mice lacking func- prednisolone after intravenous administration. Because tional P-glycoprotein allowed ready penetration of meth- P-glycoprotein clearly plays a role in excluding methyl- ylprednisolone into the spinal cord. In contrast, methylprednisolone from the spinal cord, coadministration of prednisolone was not detectable in the spinal cords of P-glycoprotein inhibitors may increase the penetration wild-type friend leukemia virus strain B mice, which of methylprednisolone into the spinal cord. This ap- express P-glycoprotein normally. The finding that con- proach would permit the use of lower doses of methyl- centrations of methylprednisolone in muscle were not prednisolone, which would decrease systemic exposure different between mdr1a/1b (Ϫ/Ϫ) mice and wild-type to the drug, thereby limiting side effects. mice indicates that methylprednisolone was equally bio- Our results show that spinal cord bioavailability of available to tissues in which P-glycoprotein does not methylprednisolone is poor after intravenous administra- limit drug entry.11–13 Thus, our results are the first to tion and that bioavailability can be markedly improved demonstrate that a drug can be actively excluded from by intrathecal administration. In addition, we have the spinal cord by P-glycoprotein. shown that P-glycoprotein plays a major role in exclud- The results of our studies have several important im- ing methylprednisolone from the spinal cord after sys- plications. First, from a pharmacokinetic standpoint, in- temic administration. Finally, P-glycoprotein inhibition trathecal administration of methylprednisolone is supe- offers a potential mechanism by which the spinal cord

Anesthesiology, V 92, No 1, Jan 2000 163

SPINAL BIOAVAILABILITY OF METHYLPREDNISOLONE

bioavailability of methylprednisolone can be improved 10. Sugawara I, Hamada H, Tsuruo T, Mori S: Specialized localization after intravenous administration. of P-glycoprotein recognized by MRK 16 monoclonal antibody in endothelial cells of the brain and the spinal cord. Jpn J Cancer Res 1990; 81:727–30 The authors thank Karen Powers and Brian Phillips for excellent 11. Prasad VK, Ho B, Haneke C: Simultaneous determination of technical assistance. prednisolone acetate, prednisolone, prednisone, cortisone and hydro- cortisone in swine plasma using solid-phase and liquid-liquid extraction techniques. J Chromatogr 1986; 378:305–16 Downloaded from http://pubs.asahq.org/anesthesiology/article-pdf/92/1/156/398816/0000542-200001000-00027.pdf by guest on 30 September 2021 References 12. Braughler JM, Hall ED: Pharmacokinetics of methylprednisolone 1. Bracken MB, Shepard MJ, Holford TR, Leo-Summers L, Aldrich EF, in cat plasma and spinal cord following a single intravenous dose of the Fazl M, Fehlings M, Herr DL, Hitchon PW, Marshall LF, Nockels RP, sodium succinate ester. Drug Metab Dispos 1982; 10:551–2 Pascale V, Perot PL, Jr, Piepmeier J, Sonntag VK, Wagner F, Wilberger 13. Ueda K, Okamura N, Hirai M, Tanigawara Y, Saeki T, Kioka N, JE, Winn HR, Young W: Administration of methylprednisolone for 24 Komano T, Hori R: Human P-glycoprotein transports cortisol, aldoste- or 48 hours or tirilazad mesylate for 48 hours in the treatment of acute rone, and dexamethasone, but not progesterone. J Biol Chem 1992; spinal cord injury: Results of the Third National Acute Spinal Cord 267(34):24248–52 Injury Randomized Controlled Trial. National Acute Spinal Cord Injury 14. van Kalken CK, Broxterman HJ, Pinedo HM, Feller N, Dekker H, Study. JAMA 1997; 277:1597–604 Lankelma J, Giaccone G: Cortisol is transported by the multidrug 2. Bracken MB, Shepard MJ, Collins WF, Jr, Holford TR, Baskin DS, resistance gene product P-glycoprotein. Br J Cancer 1993; 67:284–9 Eisenberg HM, Flamm E, Leo-Summers L, Maroon JC, Marshall LF, 15. van Asperen J, Mayer U, van Tellingen O, Beijnen JH: The Phanor LP, Piepmeier J, Sonntag VKH, Wagner FC, Wilberger JL, Winn functional role of P-glycoprotein in the blood-brain barrier. J Pharm Sci HR, Young W: Methylprednisolone or naloxone treatment after acute 1997; 86:881–4 spinal cord injury: 1-year follow-up data: Results of the second National 16. Schinkel AH, Wagenaar E, Mol CA, van Deemter L: P-glycopro- Acute Spinal Cord Injury Study [see comments]. J Neurosurg 1992; tein in the blood-brain barrier of mice inﬂuences the brain penetration 76:23–31 and pharmacological activity of many drugs. J Clin Invest 1996; 97: 3. Galandiuk S, Raque G, Appel S, Polk HC Jr: The two-edged sword 2517–24 of large-dose steroids for spinal cord trauma. Ann Surg 1993; 218: 17. Schinkel AH, Wagenaar E, van Deemter L, Mol CA, Borst P: 419–27 Absence of the mdr1a P-Glycoprotein in mice affects tissue distribution 4. Gerndt SJ, Rodriguez JL, Pawlik JW, Taheri PA, Wahl WL, Miche- and pharmacokinetics of dexamethasone, digoxin, and cyclosporin A. als AJ, Papadopoulos SM: Consequences of high-dose steroid therapy J Clin Invest 1995; 96:1698–705 for acute spinal cord injury. J Trauma 1997; 42:279–84 18. Meijer OC, de Lange EC, Breimer DD, de Boer AG, Workel JO, de 5. Drummond JC, Moore SS: The inﬂuence of dextrose administra- Kloet ER: Penetration of dexamethasone into brain glucocorticoid tion on neurologic outcome after temporary spinal cord ischemia in targets is enhanced in mdr1A P-glycoprotein knockout mice. Endocri- the rabbit. ANESTHESIOLOGY 1989; 70:64–70 nology 1998; 139:1789–93 6. Derendorf H, Mollmann H, Krieg M, Tunn S, Mollmann C, Barth J, Rothig HJ: Pharmacodynamics of methylprednisolone phosphate 19. Schinkel AH, Mayer U, Wagenaar E, Mol CA, van Deemter L, Smit after single intravenous administration to healthy volunteers. Pharm JJ, van der Valk MA, Voordouw AC, Spits H, van Tellingen O, Zijlmans Res 1991; 8:263–8 JM, Fibbe WE, Borst P: Normal viability and altered pharmacokinetics 7. George ER, Scholten DJ, Buechler CM, Jordan-Tibbs J, Mattice C, in mice lacking mdr1-type (drug-transporting) P-glycoproteins. Proc Albrecht RM: Failure of methylprednisolone to improve the outcome Natl Acad SciUSA1997; 94:4028–33 of spinal cord injuries. Am Surg 1995; 61:659–664 20. Hall ED, Braughler JM: Effects of intravenous methylpred- ϩ ϩ 8. Gerhart KA, Johnson RL, Menconi J, Hoffman RE, Lammertse DP: nisolone on spinal cord lipid peroxidation and Na ϩ K )-ATPase Utilization and effectiveness of methylprednisolone in a population- activity: Dose-response analysis during 1st hour after contusion injury based sample of spinal cord injured persons. Paraplegia 1995; 33: in the cat. J Neurosurg 1982, 57:247–53. 316–21 21. Hall ED, Wolf DL, Braughler JM: Effects of a single large dose of 9. Saitoh H, Hatakeyama M, Eguchi O, Oda M, Takada M: Involve- methylprednisolone sodium succinate on experimental posttraumatic ment of intestinal P-glycoprotein in the restricted absorption of meth- spinal cord ischemia: Dose-response and time-action analysis. J Neuro- ylprednisolone from rat small intestine. J Pharm Sci 1998; 87:73–5 surg 1984; 61:124–30

Anesthesiology, V 92, No 1, Jan 2000