Accuracy of dilution of morphine for intrathecal use.
Morphine is administered intrathecally alone or in combination with other drugs to provide spinal analgesia. Dose-finding studies have recommended 100 [micro]g be used intrathecally to optimise analgesia and minimise side-effects for caesarean section and hip replacement surgery. Dilute solutions of morphine are generally not available, mandating preparation from a 10 mg/ml ampoule. We postulated that diluting morphine would be inaccurate and imprecise, contributing to the variability in patient response often reported. Twenty consultant and trainee anaesthetists were recruited and asked to prepare 100 lug of morphine from 10 mg/ml vials and from a hypothetical prediluted 500, [micro]g/ml solution. The resultant samples were analysed using liquid chromatography. Prepared morphine doses ranged from 25 [micro]g to 289 [micro]g. Dilution of morphine was less accurate (P=0.001) and more imprecise (P=0.001) compared with using a prediluted solution. A single-step dilution technique using 0.1 ml of a solution diluted to 1.0 mg/ml was more accurate than when a double-dilution technique was used (P=0.047). Given that dose-finding studies suggest that analgesia and side-effects vary at the dose range found in this study, we advocate the use of prediluted solutions. If dilution is to be performed a single-step dilution technique should be used.
Key Words: spinal analgesia, morphine, dilution, predilution
Opiate analgesics have long been used for spinal analgesia (1-3). Morphine, because of its low lipid solubility, provides analgesia for up to 40 hours after spinal administration. Side-effects such as pruritis, urinary retention and respiratory depression have a similar duration (4-7).
As spinal anaesthesia has gained popularity in obstetrics and orthopaedic surgery, much work has been done to determine the optimal dose of intrathecal morphine to provide postoperative analgesia. Doses between 25 [micro]g and 200 [micro]g have been advocated for caesarean section and found to provide good postoperative analgesia both alone and in combination with other analgesic agents (8-10). Doses greater than 100 [micro]g do not appear to offer additional analgesia but result in a higher reported rate of side-effects (8,10). Similarly, for hip arthroplasty, 100 [micro]g appears to optimise analgesia and minimise side-effects (11). A higher dose may be required for major knee surgery (12).
Morphine is currently available in Australia in multiple preparations. Ampoules containing 5 mg/ml, 10 mg/ml and 15 mg/ml are produced commercially and are commonly available. Some hospital pharmacies produce dilute solutions for use in spinal anaesthesia, whilst 1 mg/ml and 0.5 mg/ml solutions are available internationally. As the dose range for intrathecal administration of morphine is 50 to 200 [micro]g, when these low concentration solutions are not available the anaesthetist is required to prepare a dilute solution.
Toyoyama et al (13) advocated the addition of one drop of 10 mg/ml solution from a 23 gauge needle to produce 150 [micro]g of morphine. There have been no other descriptions or attempts to standardise the dilution of morphine reported in the literature. This study was designed to examine the accuracy and precision of dilution techniques used in intrathecal injections.
After ethics committee approval and informed consent, 11 consultants and nine trainees (four registrars, five fellows) were recruited from a Melbourne teaching hospital. There were no refusals. Data collection included years of training and specialist status, previous experience with intrathecal injection of morphine, usual method of dilution of intrathecal morphine and typical intrathecal morphine dose administered.
Subjects were asked to prepare morphine 100 [micro]g (8-11) (see Appendix in website version of this issue). They were provided with a single ampoule of morphine 10 mg/ml and could prepare a solution using any combination of needle and syringe they requested. This preparation was injected into a glass vial, simulating the addition to a syringe used for intrathecal injection. This procedure was repeated three times.
Finally, the subjects were asked to prepare morphine 100 [micro]g from a mock solution said to contain morphine 500 [micro]g/ml. All subjects chose to do this with a 1 ml syringe. This latter solution consisted of normal saline and was used as a volume measurement to calculate the amount of morphine which would have been present at such a concentration.
Our 10 mg/ml solution of morphine contained morphine in sulphate form (10 mg of morphine sulphate is equal to 8.6 mg of morphine as a free base). For this reason the amount of morphine sulphate rather than morphine as a free base was calculated. The morphine sulphate concentration delivered into each vial was determined as follows:
1. Each empty vial was labelled and then weighed.
2. After the morphine sulphate doses were delivered into the labelled vials, the vials were then weighed again and the weight of solution in each vial was calculated by subtracting the weight of the empty vial.
3. This weight of morphine sulphate solution in each vial was converted to a volume (1:1) assuming a density of 1.00 for the purpose of this study.
4. All vials which contained approximately 0.1 ml of morphine sulphate solution were diluted a further 1/10 with 0.9% saline. This was done in order that all samples could be analysed at similar concentrations, regardless of what volume had been used by the subject. This dilution was performed as follows: 20 [micro]l of the solution in the vial was dispensed into a clean labelled vial and 180 A1 of saline was added and the solution was well mixed using a vortex mixer. The concentration of diluted morphine sulphate solution was determined by injecting 20 [micro]l of the contents of each vial onto the high performance liquid chromatography (HPLC) column--see below. The concentration of morphine sulphate in each vial was determined by reference to the calibration curve and then multiplied by the dilution factor (x 10) to calculate the concentration of morphine sulphate in the original vials.
5. All vials which contained approximately 1 ml of morphine sulphate solution did not require further dilution before analysis and 20 [micro]l of each of these vials was injected onto the HPLC column. The concentration of morphine sulphate in each vial was determined by reference of the area of each morphine peak to the calibration curve to calculate the concentration of morphine sulphate in the vials.
6. The dose of morphine sulphate (in [micro]g) was calculated by multiplying the concentration of morphine sulphate as determined by HPLC by the volume of solution contained in each vial (see 1 to 3 above).
7. For the mock morphine solution, the dose of morphine sulphate was calculated by multiplying the volume of solution in the vials by the stated concentration of 500 [micro]g/ml.
The morphine concentration in each sample was measured using HPLC (Varian Star HPLC System, California, U.S.A.; model 9010 Solvent Delivery System), with a 9100 autosampler and a 9050 Varichrom variable wavelength detector, and a Dupont Zorbax SB-CN (5 [micro]m, 150 x 4.6 mm) column fitted with a guard column cartridge (12.5 x 4.6 mm) of the same material as the analytical column. The mobile phase consisted of a mixture of 10 mM sodium acetate/acetic acid buffer (pH = 5.5) (A) and acetonitrile (B). The initial composition was 90% A, 10% B, increasing to 55% A, 45% B in 6.5 minutes, the final composition being maintained for a further 3.5 minutes. The total analysis time was 10 minutes. The solvent flow rate was maintained at 0.8 ml/min. Morphine was detected at a wavelength of 283 nm. The morphine peak eluted at 5.7 minutes. A 10 mg/ml ampoule of morphine sulphate solution was used to prepare working standards. This solution was diluted 1/10 and 1/100 with 0.9% saline to prepare working standards of 1 mg/ml and 0.1 mg/ml; 20 [micro]l of each working standard was injected onto the HPLC column and the area of the morphine peak obtained was plotted against the concentration of morphine sulphate. This calibration curve was a straight line which passed through the origin.
We estimated that at least 14 subjects were required in order to have 80% power to detect a 25% difference (SD 20) between groups, using alpha 0.05. The three samples prepared by dilution were averaged and were treated as a single sample for subsequent analysis (n=20). Collected data were first analysed using a one sample Komolgorov-Smirnov test to assess whether the results were consistent with a normal distribution (subsequent P>0.15). Parametric tests were hence used for further analysis. Morphine solutions prepared by one- and two-step dilution techniques were compared. Precision was compared using the Levene's test of equality of variances and accuracy was compared using a two-sided Student's t-test. The association between level of experience of intrathecal morphine techniques and accuracy and precision were measured using rank correlation. The status of the subjects (registrar, fellow, consultant) was compared using Kruskall Wallis ANOVA. All analyses were done using SPSS for Windows v11.0 (SPSS Inc, Chicago, IL, U.S.A.). A P value of less than 0.05 was considered statistically significant.
Very few (n=4) subjects had used intrathecal morphine on 10 or more occasions; only one subject had used intrathecal morphine on more than 40 occasions. Two main techniques were used to prepare morphine when dilution of the solution was required. Eight subjects used a one-step technique in which 1 ml of the 10 mg/ml morphine solutions was diluted with 9 ml of normal saline to produce a solution containing 1 mg/ml. One tenth of a ml of this solution was then used. Nine subjects used a two-step dilution technique in which 1 ml of the 10 mg/ml solution was diluted to 1 mg/ml with 9 ml of normal saline. One ml of this solution was taken and a further 9 ml of normal saline was added. One ml of this 100 [micro]g/ml solution was then used. The morphine samples which were prepared without dilution were found to have less variability than those which required dilution, 95 vs. 666 (P=0.001); the resultant mean (SD) doses prepared were 101 [micro]g (9.8) and 125 [micro]g (25.8); mean difference 24 [micro]g (95% CI: 12-37), P=0.001 (see Figure 1).
A single-step dilution technique using 0.1 ml of a solution diluted to 1.0 mg/ml was more accurate than when a double-dilution technique was used (mean values, 113 vs. 133.6 (P=0.046)). Samples prepared with a single step were also more precise, although this was not statistically significant (variance, 70 vs. 1220 (P=0.088)).
The level of experience with previous intrathecal morphine administrations and status of the subjects (registrar, fellow, consultant) had no significant effect on precision of preparation (P=0.15 and 0.84, respectively).
[FIGURE 1 OMITTED]
The principal aim of the study was to determine whether the dilution of morphine was less accurate and precise when compared with adding morphine which did not require dilution. We found that dilution increased variability (decreased precision) and decreased accuracy. The resultant morphine dose was 25% higher when the samples required dilution. Thus, the dilution process introduces greater opportunity for error.
We believe that there is a clinically significant difference between samples which require dilution in their preparation and those that do not. The latter technique resulted in a mean dose of 101 [micro]g. Assuming a normal distribution, 95% of prepared samples are between 82 [micro]g and 120 [micro]g. The two-step technique resulted in a mean dose of 126 [micro]g and 95% of samples were between 52 [micro]g and 199 [micro]g, thus providing increased opportunity for underdosage and overdosage.
In previous studies, intrathecal morphine 100 [micro]g has been found to be more effective for postoperative analgesia when compared with 50 [micro]g after hip arthroplasty and caesarean section (8-11). Our results suggest that the method of preparation of morphine for intrathecal injection could have a substantial effect on postoperative analgesia if the variation in prepared dose is as large as in our study. Similarly, when larger doses of intrathecal morphine are administered, dose-finding studies have shown a significantly increased rate of pruritus (7,8,10,12) and postoperative nausea and vomiting (7,12). Given this information, accurate administration of intrathecal morphine is necessary to avoid complications and optimise analgesic efficacy.
It is conceivable that there is greater potential for error in usual clinical practice than occurred in this study, as the setting was without stress and the subjects were not fatigued. Also, there is a potential for a Hawthorne effect due to the increased vigilance associated with study scrutiny.
There were two main techniques used to prepare the morphine solutions. A one-step technique was shown to be more accurate (mean values, 113 vs. 133.6 (P=0.046)) than when a two-step technique was used. A one-step dilution technique was also more precise although this was not statistically significant.
We did not find a statistically significant difference in the accuracy of the final preparation in those with greater experience with intrathecal morphine dilution. However this was not the principal aim of the study and hence the study may not have been powered appropriately to detect such a difference if one exists.
The unacceptably large range of morphine doses prepared by our subjects supports the use of prediluted solutions of morphine for intrathecal administration. If this is not available, then a single-step dilution should be used to accurately approximate the desired dose of morphine. As addition of opiate to local anaesthetic solution has been shown to alter the volume, baricity and hence spread of the solution (14,15), this technique is also less likely to modify the final block obtained as a much smaller volume is added.
We would like to thank Odette Youdell and Natalie Wigley for their assistance with the morphine analysis.
Accepted for publication on December 11, 2006.
(1.) Goerig M, Agarwal K, Schulte am Esch J. The versatile August Bier (1861-1949), father of spinal anaesthesia. J Clin Anesth 2000; 12:561-569.
(2.) Wulf H. The centennial of spinal anesthesia. Anesthesiology 1998; 89:500-506.
(3.) Brill S, Gurman G, Fisher A. A history of neuraxial administration of local analgesics and opioids. Eur J Anaesthesiol 2003; 20:682-689.
(4.) Stoelting R. Pharmacology and physiology in anesthetic practice. 3rd Ed. Philadelphia. Lippencott-Raven. 1999.
(5.) Sjostrom S, Tamsen A, Persson M, Hartvig P Pharmacokinetics of intrathecal morphine and meperidine in humans. Anesthesiology 1987; 67:889-895.
(6.) Nordberg G, Hedner T, Mellstrand T, Dahlstrom B. Pharmacokinetic aspects of intrathecal morphine analgesia. Anesthesiology 1984; 60:448-454.
(7.) Rathmell J, Lair T, Nauman B. The role of intrathecal drugs in the treatment of acute pain. Anesth Analg 2005; 101:30-43.
(8.) Palmer C, Emerson S, Volgoropoulos D, Alves D. Doseresponse relationship of morphine for postcesarean analgesia. Anesthesiology 1999; 90:437-444.
(9.) Gerancher J, Floyd H, Eisenach J. Determination of an effective dose of intrathecal morphine for pain relief after cesarean delivery. Anesth Analg 1999; 88:346-351.
(10.) Yang T, Breen T, Archer D, Fick G. Comparison of 0.25mg and 0.1mg intrathecal morphine for analgesia after caesarean section. Can J Anaesth 1999; 46:856-860.
(11.) Slappendel R, Weber E, Dirksen R, Gielen M, van Limbeek J. Optimization of the dose of intrathecal morphine in total hip surgery: A dose-finding study. Anesth Analg 1999; 88:822-826.
(12.) Rathmell J, Pin C, Taylor R, Patrin T, Viani B. Intrathecal morphine for post operative analgesia: A randomized, controlled, dose ranging study after hip and knee arthroplasty. Anesth Analg 2003; 97:1452-1457.
(13.) Toyoyama H, Mizutani K, Toyoda Y. One drop of morphine added to local anesthetics by means of a 23-gauge injection needle can relieve postoperative pain under spinal anesthesia. Anesth Analg 2000; 90:1000.
(14.) Hallworth S, Fernando R, Columb M, Stocks G. The effect of posture and baricity on the spread of intrathecal bupivacaine for elective cesarean delivery. Anesth Analg 1998; 87:336-340.
(15.) Parlow J, Money P, Chan P, Raymond J, Milne B. Addition of opioids alters the density and spread of intrathecal local anaesthetics? An in vitro study. Can J Anaesth 1999; 46:66-77.
A. R. BAKER *, D. M. RUTHERFORD ([dagger]), E S. MYLES ([double dagger])
Department of Anaesthesia and Clinical Biochemistry Unit, Alfred Pathology Service, The Alfred Hospital, Melbourne, Victoria, Australia
* M.B., Ch.B., B.Sc., Registrar, Magill Department of Anaesthesia, Intensive Care and Pain Management, Chelsea and Westminster Hospital, London, United Kingdom.
([dagger]) B.Sc. (Hons), M.Sc., Senior Scientist, Clinical Biochemistry Unit, Alfred Pathology Service.
([double dagger]) M.B., B.S., M.PH., M.D., F.C.A.R.C.S.I., F.A.N.Z.C.A., Director, Department of Anaesthesia and Perioperative Medicine, Alfred Hospital, Professor, Academic Board of Anaesthesia and Perioperative Medicine, Monash University and NHMRC Practitioner Fellow, Centre for Clinical Research Excellence, Canberra.
Address for reprints: Dr A. R. Baker, St Vincent's Hospital, St Vincent's Health, PO Box 2900, Fitzroy, Melbourne, Vic. 3065.
TABLE 1 Method of preparation of morphine 100 [micro]g intended for intrathecal use Number of subjects Method of preparation 8 One-step technique: morphine 10 mg/ml diluted with 9 ml of normal saline, to produce a solution containing 1 mg/ml; 0.1 ml of this solution was then injected Two-step technique: morphine 10 mg/ml diluted 9 with 9 ml of normal saline, to produce a solution containing 1 mg/ml; 1 ml of this solution was then diluted with another 9 ml of normal saline to produce a 10 [micro]g/ml solution; 1 ml of this solution was then injected Other 3
|Printer friendly Cite/link Email Feedback|
|Author:||Baker, A.R.; Rutherford, D.M.; Myles, E.S.|
|Publication:||Anaesthesia and Intensive Care|
|Article Type:||Clinical report|
|Date:||Jun 1, 2007|
|Previous Article:||William Russ Pugh's qualifications.|
|Next Article:||NT pro BNP secretion and clinical endpoints in cardiac surgery intensive care patients.|