Acta Anaesthesiol Scand 2013; 57: 1321–1327 Printed in Singapore. All rights reserved
© 2013 The Acta Anaesthesiologica Scandinavica Foundation. Published by John Wiley & Sons Ltd ACTA ANAESTHESIOLOGICA SCANDINAVICA
doi: 10.1111/aas.12206
The stability of a sulphite-free epidural analgesic solution containing fentanyl, bupivacaine, and adrenaline J. Brustugun1, S. Troland1 and H. Breivik2
1 Hospital Pharmacy at Oslo University Hospital, Rikshospitalet, Sykehusapotekene HF, Oslo, Norway and 2University of Oslo and Department of Anaesthesia and Department of Pain Management and Research, Oslo University Hospital, Rikshospitalet, Oslo, Norway
Background: Thoracic epidural infusion analgesia is optimised by using a triple component infusion containing a local anaesthetic, an opioid, and adrenaline. Adrenaline in solution is prone to oxidation, and stabilisers, such as antioxidants (e.g. sulphites) or chelators (edetates), are therefore commonly added. Sulphites may, however, have unwanted effects, especially allergic reactions. The aim of this study was to evaluate the stability of an analgesic infusion solution for epidural administration free of sulphites, containing adrenaline, fentanyl, bupivacaine, and disodium edetate. Methods: An epidural infusion solution containing adrenaline 2 μg/ml, fentanyl 2 μg/ml, bupivacaine 1 mg/ml, and disodium edetate 0.18 μg/ml was stored at 2–8°C for 4.5 months. A concentrate 11 times more potent, used for the production of the ready-to-use solution, was stored at 2–8°C for 9 months. Concentrations of the active ingredients were determined, pH was measured throughout the period, the clarity of the solu-
B
upivacaine and fentanyl are commonly used for epidural analgesia with concentrations of bupivacaine from 0.06 to 2.0 mg/ml and of fentanyl from 2 to 10 μg/ml. With higher concentrations, there are more potent analgesic effects,1 but also more adverse effects, e.g. motor and sympathetic blockades from the local anaesthetic, as well as nausea, sedation, and immediate respiratory depression from systemic absorption of fentanyl. Adrenaline enhances epidural analgesia from opioids2 and bupivacaine.3 This is true for thoracic and thoracolumbar epidural analgesia.4,5 Adrenaline has a specific analgesic effect in the spinal cord dorsal horn via the α2-receptors,6 and it enhances the spinal cord analgesic effects of bupivacaine and fentanyl so that low concentrations suffice for analgesia, minimising their dose-related side effects.4,7 Clonidine is a more potent α2-receptor agonist;
tions was investigated, and the weight of the infusion bags recorded. Results: After 4.5 months at 2–8°C, the infusion solution contained adrenaline 97.5%, bupivacaine 100.9%, and fentanyl 102.6%. The pH stayed between 4.76 and 4.79, the solutions remained clear, and the weight was 99.9% of that found initially. The solution was also stable for 7 days at room temperature. The concentrate was stable (> 90%) for 9 months at 2–8°C. Conclusion: The solution containing adrenaline, fentanyl, and bupivacaine, stabilised with disodium edetate, is stable for several months at 2–8°C, and at least for 7 days at room temperature without the addition of sulphites. Accepted for publication 2 September 2013 © 2013 The Acta Anaesthesiologica Scandinavica Foundation. Published by John Wiley & Sons Ltd
however, clonidine, contrary to adrenaline, causes sedation and hypotension when coadministered with fentanyl and bupivacaine.8 Levobupivacaine and especially ropivacaine are less cardiotoxic than bupivacaine, but when an opioid and adrenaline are coadministered epidurally, the necessary dose of bupivacaine is far below the cardiotoxic dose. When adrenaline is added to a local anaesthetic epidural infusion/injection in a low lumbar site, the motor blockade of the cauda equina nerve roots is increased, causing weak legs and urinary retention. Adrenaline is prone to oxidation,9 and antioxidants (e.g. sulphites) are therefore often added to solutions of the substance. We documented stability for at least 6 months of our epidural solution containing bupivacaine, fentanyl, and adrenaline, with metabisulphite as the antioxidant.10
1321 bs_bs_banner
J. Brustugun et al.
Worrisome hypersensitivity and allergic reactions to sulphites have been reported, however.11–14 One of our own patients also had a generalised rash that disappeared on discontinuation of our epidural infusion containing metabisulphite. Therefore, the aim of the present study was to investigate the longterm stability at 2–8°C of such a triple component solution without metabisulphite, as well as the stability of the solution during 7 days at room temperature. The stability of a concentrate for the production of the ready-to-use infusion solution was also studied.
Methods This study was a shelf-life study of an epidural analgesic solution. No patients participated; therefore, no approval was sought from the local ethics committee.
Compounding of the solutions We describe briefly how the triple component epidural analgesia infusion is routinely produced via a concentrate that is diluted with saline to a ready-touse solution. The concentrate. The concentrate is produced by mixing adrenaline tartrate 41.5 μg/g (L-form), bupivacaine hydrochloride 10.9 mg/g, fentanyl citrate 34.6 μg/g, and disodium edetate 2.0 μg/g in nitrogen-purged water for injections in bulk. The pH is adjusted to 3.8–4.0 using sterile 1 M hydrochloric acid, and the solution is dispensed on 50 ml injection vials flushed with carbon dioxide. Following this, the vials are capped and autoclaved. They are then stored at 2–8°C, not exposed to light. The ready to use triple component epidural solution. The ready to use triple component epidural solution is prepared by Fresenius-Kabi (Halden, Norway) in polypropylene-based infusion bags (FreeFlex, Fresenius-Kabi) in volumes 110 ml (enough for obstetric practice), 250 ml (for 1–2 days epidural analgesia), or 550 ml – for more prolonged epidural analgesia. The solution is prepared in a gas isolator by diluting appropriate amounts of concentrate in infusion bags containing saline solution. For the present shelf-life study, the bags were prepared by diluting 10 ml of the concentrated solution in 100 ml 0.9% saline. The final volumes of the bags were thus nominally 110 ml, and the final nominal concentrations of the active ingredients in the readyto-use infusion solution were: adrenaline 2 μg/ml,
1322
bupivacaine 1 mg/ml, and fentanyl 2 μg/ml. Because infusion bags contain a certain amount overfill, and the concentrate routinely is added to the infusion bag without withdrawing any 0.9% saline, the actual concentration of the active ingredients in the ready-to-use infusion will be somewhat lower than the nominal concentration – corresponding to the degree of overfill. The bags used in this study could contain up to approximately 20% overfill.
Stability study The concentrate. Following production, the concentrate was stored at 2–8°C, not exposed to light, for 9 months. Every 3 months, the concentrations of the active ingredients were determined, the pH measured, and the appearance recorded. Appearance was investigated using the naked eye and the naked eye aided by polarised light. Six vials were investigated at each time point. The ready to use infusion solution. The 110-ml infusion bags containing ready-to-use triple component infusion solution were returned from FreseniusKabi to the Hospital Pharmacy at Oslo University Hospital, Rikshospitalet, where they were stored at 2–8°C for 4.5 months, not exposed to light. The concentrations of the three active ingredients, adrenaline, bupivacaine, and fentanyl, were determined at the receipt of the bags and again after 1.5, 3, and 4.5 months. The content of six infusion bags was determined each time. The bags were also inspected visually (using the naked eye and the naked eye aided by polarised light), the pH was measured, and the weight recorded. After storage for 4.5 months at 2–8°C, approximately 10 ml sample solution was withdrawn, and the infusion bags (n = 6) were stored in a constant climate chamber at 25°C and 40% relative humidity (RH) for 7 days, not exposed to light. Concentrations were determined, and physicochemical properties were recorded at the start and at the end of the 7-day period. Primary outcome measures. The solutions would be considered stable as long as the concentrations of the active ingredients did not deviate more than 10% from the initial concentrations, a level of change often allowed for drug solutions prepared in hospital pharmacies when the ‘therapeutic window’ is not very narrow. To be considered stable, pH values should be 4.0–5.0 for the ready-to-use infusion solution whereas values of 3.8–4.0 would be considered acceptable for the concentrate, the solu-
Stability of a sulphite-free solution
tions should be clear and free from particles, and the weight of the bags should not change by more than 5%.
Assay The stability indicating method used to determine fentanyl and bupivacaine in the triple component solution has been described in detail previously.15 In short, the method utilised an eluent consisting of 5 mM potassium dihydrogen phosphate and 1.25 mM tetrabutylammonium hydroxide adjusted with phosphoric acid to pH 4.8/methanol, 60 : 40 (v/v). The flow rate was 1 ml/min, and the detection wavelength was 210 nm. Linearity (r > 0.999) was demonstrated over the concentration range 0.05–10 μg/ml for fentanyl and 0.01–0.2 mg/ml for bupivacaine. Interday precision over 6 days was 0.8% of fentanyl and 0.5% for bupivacaine. The mean intraday precision was 1.1% for fentanyl and 0.8% for bupivacaine. The selectivity of the method was demonstrated as solutions of bupivacaine and fentanyl were subjected to pH 1 and 13 under heat (121°C), as well as hydrogen peroxide (10%, 24 h) without any interferences occurring. For the analysis of adrenaline, a gradient high performance liquid chromatography (HPLC) method was used. The first eluent (eluent A) consisted of 5 mM potassium dihydrogen phosphate adjusted with phosphoric acid to pH 3.6/methanol, 80 : 20 (v/v). The second eluent (eluent B) contained the same two ingredients, but in the ratio 40 : 60 (v/v). Eluent A ran from time 0 to 6.5 min, eluent B from 6.5 min to 25 min, and eluent A was used again from 25 min to 45 min. Adrenaline had a retention time of approximately 2.9 min. The detection wavelength was 210 nm, the flow rate 1.0 ml/min, and the injection volume 50 μl. When necessary, samples were diluted in eluent A before injection. The stability indicating the nature of the method for analysis of adrenaline was validated: Linearity over seven points between 0.1 and 10 μg/ml was demonstrated (r2 ≥ 0.999). The system precision (n = 6) was determined at 0.1 μg/ml, at 2.0 μg/ml, and at 7.5 μg/ml. The highest relative standard deviation was found to be 0.4%, at 0.1 μg/ml. The intra- and interday precision was demonstrated by determining the content of adrenaline in six samples of triple component solution on 6 days within a 4-week period. The interday precision was found to be 0.7% (SD% over the 6 days). The highest SD% found any of the 6 days for the six samples determined (intraday precision) was 0.4%. The selectivity of the method was demonstrated by subjecting
adrenaline (4 μg/ml) to acid and heat (pH 1, 121°C, 20 min), and base and heat (pH 11, 60°C, 30 min). No interference with adrenaline measurements was found when analysing the samples so treated.
Materials and instruments Potassium dihydrogen phosphate was obtained from Merck, KGaA (Darmstadt, Germany). Methanol was obtained from Rathburn Chemicals Ltd (Walkerburn, Scotland). Tetrabuthylammonium hydroxide was obtained from Fluka (Neu-Ulm, Switzerland), and Sigma-Aldrich Co (St Louis, MO, USA). Adrenaline bitartrate was obtained from Sigma-Aldrich, and Apotekproduksjon (Oslo, Norway). Bupivacaine hydrochloride was obtained from RTC (Laramie, WY, USA) and Apotekproduksjon. Sodium metabisulphite, disodium edetate, and fentanyl citrate were obtained from Apotekproduksjon. The system used for HPLC analysis consisted of an LC-20AD pump, a SIL-20AC autoinjector with sample cooler, a SPD-M20A diode array detector, a DGU-20A5 degassing unit, and a CBM-20A communication module, all units delivered by Shimadzu Corp. (Kyoto, Japan). For the chromatographic separation, a Supelcosil LC-DP column (100 mm × 4.6 mm, 5 μm) was used (Supelco, Belefonte PA, USA). For storage at controlled room temperature and humidity (25°C, 40% RH), a Termaks cabinet from Houm AS (Oslo, Norway) was used. pH was measured with a 691 pH meter from Metrohm (Herisau, Switzerland). For visual inspection of the solutions in polarised light, an Allen LV28 liquid inspection system was used (P.W.Allen & Co, Evesham, Worcester, UK).
Results The concentrate The concentration of all the three active ingredients in the triple component concentrate, stored at 2–8°C protected from light, remained well within 10% of the initial concentration throughout the whole test period of 9 months and showed little variation (Table 1). The concentrate was clear and colourless with no visible particles at all times, and the pH was 3.8 ± 0.0 throughout the study.
The ready to use infusion solution The concentration of the three active ingredients in the ready to use infusion solution stored in the dark for 4.5 months at 2–8°C also stayed well within 10%
1323
J. Brustugun et al. Table 1 Shelf life of concentrate for production of ready-to-use epidural analgesic solution stored at 2–8°C. Adrenaline % Bupivacaine % Fentanyl %
Start
3 Months
6 Months
9 Months
21.84 (0.11) μg/ml
21.78 (0.12) 99.8 10.75 (0.09) 100.2 21.39 (0.10) 99.4
21.84 (0.06) 100.0 10.78 (0.05) 100.5 21.44 (0.08) 99.7
21.96 (0.09) 100.6 10.76 (0.07) 100.4 20.96 (0.13) 97.4
10.72 (0.03) mg/ml 21.51 (0.11) μg/ml
Concentrations of adrenaline, bupivacaine, and fentanyl in concentrate vials are given as mean (SD), n = 6, and per cent remaining with respect to Start.
Table 2 Concentrations of adrenaline, bupivacaine, and fentanyl in a 110 ml ready to use epidural analgesic solution. Adrenaline % Bupivacaine % Fentanyl %
Start*
1.5 Months
3 Months
4.5 Months
4.5 Months, 7 days at RT
1.68 (0.03) μg/ml
1.70 (0.06) 101.5 0.83 (0.03) 98.5 1.69 (0.07) 101.1
1.61 (0.06) 96.1 0.86 (0.03) 101.5 1.74 (0.05) 103.9
1.63 (0.03) 97.5 0.85 (0.01) 100.9 1.72 (0.02) 102.6
1.67 (0.03) 99.9 0.86 (0.01) 101.2 1.71 (0.02) 102.5
0.85 (0.02) mg/ml 1.67 (0.03) μg/ml
The infusion solutions were stored in the dark, first at 2–8°C for 4.5 months, and then for 7 days at room temperature (RT; 25°C, 40% relative humidity). Data are given as mean (SD), n = 6, and per cent. Per cent values refer to per cent of Start concentration. *The actual start concentrations were lower than the nominal concentrations (2 μg/ml, 1 mg/ml, and 2 μg/ml) because the saline bags initially contained a certain amount of overfill.
Table 3 Physicochemical variables found in the ready to use epidural analgesic solution containing adrenaline, bupivacaine, and fentanyl.
pH Appearance Wight (g)
Start
1.5 Months
3 Months
4.5 Months
4.5 Months, 7 days at RT
4.76 (0.03) Clear, colourless 134.97 (0.50)*
4.78 (0.02) Clear, colourless 134.69 (0.74)
4.76 (0.03) Clear, colourless 134.93 (0.36)
4.79 (0.02) Clear, colourless 134.87 (0.26)†
4.77 (0.02) Clear, colourless 124.01 (0.20)
The solutions were stored in the dark, first at 2–8°C for 4.5 months, and then for 7 days at room temperature (RT; 25°C, 40% relative humidity). Data are given as mean (SD), n = 6. *n = 80. †The weight after sampling was 124.09 (0.20) g, n = 16.
of that initially determined (Table 2). The solutions were clear and colourless at all times, and the pH varied between 4.76 and 4.79, well within the acceptable interval of 4.0–5.0 (Table 3). The weight of the infusion bags at the end of the test period was 99.9% of the initial value. Upon further storing the 4.5 months old bags at room temperature (25°C, 40% RH) for 7 days, the concentrations of the three active substances still were all well within 10% of that determined at day 0 of the study (Table 2). The pH and appearance of the solutions also complied with the acceptance criteria during storage for 1 week at room temperature (Table 3). The weight of the bags was practically
1324
unchanged, 124.09 (0.20) g at the start and 124.01 (0.20) g at the end of the 7-day period.
Discussion The present study shows that a ready-to-use epidural analgesia solution not containing sulphites but consisting of adrenaline (2 μg/ml), fentanyl (2 μg/ml), bupivacaine (1 mg/ml), and disodium edetate (0.18 μg/ml) in 0.9% saline remains stable for both 4.5 months at 2–8°C and 7 days at room temperature (nominal concentrations given, for actual initial concentrations, see Table 2). In our experience, a shelf life of 4.5 months is sufficient
Stability of a sulphite-free solution
for production, quality control, and distribution. Because sulphites are allergenic and can occasionally cause anaphylactic reactions, documenting that adrenaline is stable without sulphites over a time span relevant for use is clinically of major importance. Sulphite allergy is not uncommon; for this reason, winemakers are obliged to declare whether sulphites are added to their finished products! Previously, a solution with the same concentrations of bupivacaine, fentanyl, and adrenaline, stabilised with sodium metabisulphite (1.8 μg/ml) and disodium edetate (0.18 μg/ml), having a pH of 4.7, was shown to be stable for 180 days at 4°C and 4 days at room temperature.10 This solution was made using raw materials in bulk. Other studies have been published on the stability of similar triple component solutions made from commercially available pharmaceutical products containing sulphite. In one such study, a solution containing the same three active ingredients in the same concentrations was made using an adrenaline solution for injection that contained 0.95 μg/ml sulphite (epinephrine injection BP, Phoenix Pharma, Gloucester, UK). The stability of that solution was documented for 184 days at both 4°C and 22°C.16 In another study, a solution containing levobupivacaine (570 μg/ml), fentanyl (3.7 μg/ml), and adrenaline (1.9 μg/ml) was shown to be stable for 60 days at 6°C and 40 days at 22°C.17 The adrenaline solution for injection used for preparing this solution (Adrenalin 0.1 mg/ml, Leiras, Helsinki, Finland) contained edetate and 1 mg/ml sodium metabisulphite. The actual sulphite species added when compounding adrenaline solutions varies: sulphite, bisulphite, and metabisulphite (pyrosulphite) are all used. Which of the sulphite species that is added during production is of less importance as they are all connected through chemical equilibria;9 the concentration and pH value of the solution will determine which species is finally predominant.
Why edetate is an effective stabiliser of adrenaline The results presented in the present study indicate that adrenaline in sulphite-free triple component solutions containing sodium edetate is stable for a period of at least 4.5 months (in the concentrate at least 9 months) when stored at 2–8°C. Several factors may have contributed to the stability. (1) The solutions were stored at a low temperature, 2–8°C. (2) The water used for the preparation of the concentrate was purged with nitrogen to remove oxygen both before the components were mixed together
and also during the mixing process. (3) The 50 ml vials used to contain the concentrate were flushed with carbon dioxide, further preventing oxygen from interacting with the adrenaline. (4) The solutions contained the chelator disodium edetate, and chelators have previously been demonstrated to be effective stabilisers of adrenaline.18 For the edetate to be effective as a stabiliser, the content of trace metal ions should probably be kept as low as possible in the solutions. Using raw materials, drug substances and water, all of high purity, is therefore probably of particular importance in this instance – not only to ensure patient safety, but also to improve drug stability. The ingredients used in the production of the concentrate in our hospital pharmacy comply with the specifications in the European Pharmacopeia. The water used is Water for injections in bulk, in our facility normally having a conductivity of ≤1.0 μS/cm, indicating a low content of ions.
Is edetate biologically safer than sulphites? Edetic acid and edentate salts are substances widely used in pharmaceutical formulations, typically in concentrations ranging from 0.05 to 1 mg/ml.19 The present study documents that disodium edetate at a low level of 0.18 μg/ml is sufficient to stabilise adrenaline (2 μg/ml) in a ready-to-use infusion solution, provided the ingredients are of high purity and steps are taken to remove oxygen (see above). Edetic acid is generally regarded as essentially nontoxic and nonirritant,19 and the nonirritant nature of the salt disodium edetate has been demonstrated through human patch testing.20 Disodium edetate is also included in the U.S. Food and Drug Administration Inactive Ingredients Guide. A concentration of 0.111 mg/ml (0.0111%) being approved for epidural use – a value well above the concentration used in the epidural solution presented here (0.18 μg/ml). As shown in Table 2, the actual initial concentrations of the active substances in the infusion solution are lower than could be expected after a 1 : 11 dilution of the concentrate. The reason for this is that the bags contain a certain amount overfill and that the concentrate is added to the infusion bag without withdrawing any saline. The 100 ml bags used in this study can contain up to 20% overfill, explaining the initial concentrations seen (Table 2). It should be noted that despite the dilution caused by overfill, the concentration of adrenaline in this study remained above the minimally effective concentration of 1.5 μg/ml determined in earlier studies.7 It
1325
J. Brustugun et al.
should also be noted that larger infusion bags contain a lower relative amount of overfill. For instance, the overfill in the 500 ml bags used for preparation of larger volume ready-to-use infusion solution is limited upwards to 7%. Using a larger infusion bag can thus cause less dilution of both the active substances and the edetate – and less dilution of edetate could possibly improve the stability of the infusion solution. In a recent study, it was stated that the degradation of adrenaline in a triple component analgesic admixture containing levobupivacaine, fentanyl, and adrenaline was mostly dependant on light – light having a destabilising effect.17 However, the adrenaline injection used for the preparation of that solution (Adrenalin 0.1 mg/ml, Leiras) contains sulphite, and this may have influenced the result. It has previously been shown that sulphites can make catecholamines, such as adrenaline, less stable when subjected to light.15,21,22 It has also been shown that a triple component solution containing adrenaline, fentanyl, and bupivacaine, but no sulphite, is more stable in light than solutions containing sulphite when tested according to the specifications for photostability testing outlined in the guideline of the International Conference on Harmonisation (ICH), ICH Q1B.15 The photodestabilising effect of sulphite is caused by adrenochrome, a degradation product of adrenaline, first reacting with bisulphite to form an adrenochrome-bisulphite addition product.23 This substance can, when exposed to light, produce reactive singlet oxygen capable of oxidising adrenaline.21 Because the adrenochrome-bisulphite addition product can no longer be formed in a sulphite-free solution, and because a sulphite-free triple component infusion has previously been shown to be stable for periods covering normal use when exposed to light under the ICH criteria, we no longer recommend that infusion bags containing the triple component infusion are protected from light during use at our hospital. A possible limitation of our study is that the HPLC method used to determine adrenaline did not differentiate between the two optical isomers of adrenalin. As the L-form of adrenaline is the more biologically active one, also at the α2-receptors of the spinal cord dorsal horn, detection of racemisation could be of interest. The racemisation of adrenaline is normally a slow process in pharmaceutical products. Stepensky et al. found a racemisation rate of 5.6% over a period of 2 years,24 and others have extrapolated a shelf life (90% remaining) regarding
1326
racemisation of 120 months at pH 3.5.9 However, future studies on the stability of sulphite-free multicomponent epidural solutions containing adrenaline should ideally include investigation of the racemisation of adrenalin as well as the effects of temperature, chelator concentration, water quality, and the use of oxygen-purging agents (i.e. carbon dioxide or nitrogen) on the stability of the active ingredients. In conclusion, this study documents that it is possible to make a multicomponent infusion solution for epidural analgesia, containing adrenaline but no sulphites, that is sufficiently stable for production, distribution, and use. Such a solution containing adrenaline (2 μg/ml), fentanyl (2 μg/ml), bupivacaine (1 mg/ml), and disodium edetate (0.18 μg/ml) adjusted to pH 4.8 has been shown to be stable for at least 4.5 months at 2–8°C in polypropylene-based infusion bags, and also for 7 days at room temperature. A concentrate for the preparation of ready-to-use infusion solution, 11 times higher in concentration of the ingredients, was stable (> 90%) at 2–8°C for 9 months. The important implication of our present findings is that it is now not necessary to expose patients to sulphites when administering an epidural analgesic infusion solution containing adrenaline (with fentanyl and bupivacaine), avoiding the (slight) risk of precipitating allergic reactions in patients who are already allergic to sulphites.
Acknowledgements The authors thank Håvard Hem for technical assistance. The study was supported by Fresenius Kabi, Halden, Norway, and the Hospital Pharmacy at Oslo University Hospital, Rikshospitalet, Oslo, Norway. Conflict of interest: Inven2, a government institution for funding and commercialisation of research at the Oslo University Hospital, receives royalties of sales of the ready-to-use epidural solution from Fresenius Kabi, Halden, Norway. The authors have previously received research funding from Inven2.
References 1. Ginosar Y, Riley ET, Angst MS. The site of action of epidural fentanyl in humans: the difference between infusion and bolus administration. Anesth Analg 2003; 97: 1428–38. 2. Bromage PR, Camporesi EM, Durant PA, Nielsen CH. Influence of epinephrine as an adjuvant to epidural morphine. Anesthesiology 1983; 58: 257–62. 3. Eisenach JC, Grice SC, Dewan DM. Epinephrine enhances analgesia produced by epidural bupivacaine during labor. Anesth Analg 1987; 66: 447–51. 4. Niemi G, Breivik H. Adrenaline markedly improves thoracic epidural analgesia produced by a low-dose infusion of bupivacaine, fentanyl and adrenaline after major surgery. A
Stability of a sulphite-free solution
5.
6.
7.
8.
9. 10.
11. 12. 13. 14. 15.
randomised, double-blind, cross-over study with and without adrenaline. Acta Anaesthesiol Scand 1998; 42: 897– 909. Förster JG, Lumme HM, Palkama VJ, Rosenberg PH, Pitkänen MT. Epinephrine 4 microg/mL added to a lowdose mixture of ropivacaine and fentanyl for lumbar epidural analgesia after total knee arthroplasty. Anesth Analg 2008; 106: 301–4. Collins JG, Kitahata LM, Matsumoto M, Homma E, Suzukawa M. Spinally administered epinephrine suppresses noxiously evoked activity of WDR neurons in the dorsal horn of the spinal cord. Anesthesiology 1984; 60: 269– 75. Niemi G, Breivik H. The minimally effective concentration of adrenaline in a low-concentration thoracic epidural analgesic infusion of bupivacaine, fentanyl and adrenaline after major surgery. A randomized, double-blind, dosefinding study. Acta Anaesthesiol Scand 2003; 47: 439– 50. Curatolo M, Schnider TW, Petersen-Felix S, Weiss S, Signer C, Scaramozzino P, Zbinden AM. A direct search procedure to optimize combinations of epidural bupivacaine, fentanyl, and clonidine for postoperative analgesia. Anesthesiology 2000; 92: 325–37. Connors KA. Epinephrine. In: Connors KA, Amidon GL, Stella VJ eds. Chemical Stability of Pharmaceuticals, 2nd edn. New York: John Wiley & Sons, 1986: 438–47. Kjønniksen I, Brustugun J, Niemi G, Breivik H, Anderssen E, Klem W. Stability of an epidural analgesic solution containing adrenaline, bupivacaine and fentanyl. Acta Anaesthesiol Scand 2000; 44: 864–7. Dooms-Goossens A, De Alam AG, Degreef H, Kochuyt A. Local anesthetic intolerance due to metabisulfite. Contact Dermatitis 1989; 20: 124–6. Smolinske SC. Review of parenteral sulfite reactions. J Toxicol Clin Toxicol 1992; 30: 597–606. Madan V, Walker SL, Beck MH. Sodium metabisulfite allergy is common but is it relevant? Contact Dermatitis 2007; 57: 173–6. Vally H, Misso NLA, Madan V. Clinical effects of sulphite additives. Clin Exp Allergy 2009; 39: 1643–51. Brustugun J, Tønnesen HH, Klem W, Kjønniksen I. Photodestabilization of epinephrine by sodium metabisulfite. PDA J Pharm Sci Technol 2000; 54: 136–43.
16. Priston MJ, Hughes JM, Santillo M, Christie IW. Stability of an epidural analgesic admixture containing epinephrine, fentanyl and bupivacaine. Anaesthesia 2004; 59: 979–83. 17. Helin-Tanninen M, Lehtonen M, Naaranlahti T, Venäläinen T, Pentikäinen J, Laatikainen A, Kokki H. Stability of an epidural analgesic admixture of levobupivacaine, fentanyl and epinephrine. J Clin Pharm Ther 2013; 38: 104–8. 18. Roscoe CW, Hall NA. Chelating agents as color stabilizers for epinephrine hydrochloride solutions. J Am Pharm Assoc 1956; 45: 464–70. 19. Owen SC. Edetic acid. In: Rowe RC, Shestkey PJ, Weller PJ eds. Handbook of Pharmaceutical Excipients, 4th edn. London: Pharmaceutical Press; Washington, DC: American Pharmaceutical Association, 2003: 225–8. 20. Basketter DA, Chamberlain M, Griffiths HA, Rowson M, Whittle E, York M. The classification of skin irritants by human patch test. Food and Chemical Toxicology 1997; 35: 845–52. 21. Brustugun J, Kristensen S, Tønnesen HH. Photostability of epinephrine – the influence of bisulfite and degradation products. Pharmazie 2004; 59: 457–63. 22. Brustugun J, Kristensen S, Tønnesen HH. Photostability of sympathomimetic agents in commonly used infusion media in the absence and presence of bisulfite. PDA J Pharm Sci Technol 2004; 58: 296–308. 23. Heacock RA, Powell WS. Adrenochrome and related compounds. In: Ellis GP, West GB eds. Progress in Medical Chemistry 9. Amsterdam: North Holland Publishing Company, 1973: 275–339. 24. Stepensky D, Chorny M, Dabour Z, Schumacher I. Longterm stability study of L-adrenaline injections: kinetics of sulfonation and racemization pathways of drug degradation. J Pharm Sci 2004; 93: 969–80.
Address: Jørgen Brustugun Hospital Pharmacy Oslo University Hospital Rikshospitalet Pb 4950 Nydalen 0424 Oslo Norway e-mail: [email protected]
1327
© 2013 The Acta Anaesthesiologica Scandinavica Foundation. Published by John Wiley & Sons Ltd ACTA ANAESTHESIOLOGICA SCANDINAVICA
doi: 10.1111/aas.12206
The stability of a sulphite-free epidural analgesic solution containing fentanyl, bupivacaine, and adrenaline J. Brustugun1, S. Troland1 and H. Breivik2
1 Hospital Pharmacy at Oslo University Hospital, Rikshospitalet, Sykehusapotekene HF, Oslo, Norway and 2University of Oslo and Department of Anaesthesia and Department of Pain Management and Research, Oslo University Hospital, Rikshospitalet, Oslo, Norway
Background: Thoracic epidural infusion analgesia is optimised by using a triple component infusion containing a local anaesthetic, an opioid, and adrenaline. Adrenaline in solution is prone to oxidation, and stabilisers, such as antioxidants (e.g. sulphites) or chelators (edetates), are therefore commonly added. Sulphites may, however, have unwanted effects, especially allergic reactions. The aim of this study was to evaluate the stability of an analgesic infusion solution for epidural administration free of sulphites, containing adrenaline, fentanyl, bupivacaine, and disodium edetate. Methods: An epidural infusion solution containing adrenaline 2 μg/ml, fentanyl 2 μg/ml, bupivacaine 1 mg/ml, and disodium edetate 0.18 μg/ml was stored at 2–8°C for 4.5 months. A concentrate 11 times more potent, used for the production of the ready-to-use solution, was stored at 2–8°C for 9 months. Concentrations of the active ingredients were determined, pH was measured throughout the period, the clarity of the solu-
B
upivacaine and fentanyl are commonly used for epidural analgesia with concentrations of bupivacaine from 0.06 to 2.0 mg/ml and of fentanyl from 2 to 10 μg/ml. With higher concentrations, there are more potent analgesic effects,1 but also more adverse effects, e.g. motor and sympathetic blockades from the local anaesthetic, as well as nausea, sedation, and immediate respiratory depression from systemic absorption of fentanyl. Adrenaline enhances epidural analgesia from opioids2 and bupivacaine.3 This is true for thoracic and thoracolumbar epidural analgesia.4,5 Adrenaline has a specific analgesic effect in the spinal cord dorsal horn via the α2-receptors,6 and it enhances the spinal cord analgesic effects of bupivacaine and fentanyl so that low concentrations suffice for analgesia, minimising their dose-related side effects.4,7 Clonidine is a more potent α2-receptor agonist;
tions was investigated, and the weight of the infusion bags recorded. Results: After 4.5 months at 2–8°C, the infusion solution contained adrenaline 97.5%, bupivacaine 100.9%, and fentanyl 102.6%. The pH stayed between 4.76 and 4.79, the solutions remained clear, and the weight was 99.9% of that found initially. The solution was also stable for 7 days at room temperature. The concentrate was stable (> 90%) for 9 months at 2–8°C. Conclusion: The solution containing adrenaline, fentanyl, and bupivacaine, stabilised with disodium edetate, is stable for several months at 2–8°C, and at least for 7 days at room temperature without the addition of sulphites. Accepted for publication 2 September 2013 © 2013 The Acta Anaesthesiologica Scandinavica Foundation. Published by John Wiley & Sons Ltd
however, clonidine, contrary to adrenaline, causes sedation and hypotension when coadministered with fentanyl and bupivacaine.8 Levobupivacaine and especially ropivacaine are less cardiotoxic than bupivacaine, but when an opioid and adrenaline are coadministered epidurally, the necessary dose of bupivacaine is far below the cardiotoxic dose. When adrenaline is added to a local anaesthetic epidural infusion/injection in a low lumbar site, the motor blockade of the cauda equina nerve roots is increased, causing weak legs and urinary retention. Adrenaline is prone to oxidation,9 and antioxidants (e.g. sulphites) are therefore often added to solutions of the substance. We documented stability for at least 6 months of our epidural solution containing bupivacaine, fentanyl, and adrenaline, with metabisulphite as the antioxidant.10
1321 bs_bs_banner
J. Brustugun et al.
Worrisome hypersensitivity and allergic reactions to sulphites have been reported, however.11–14 One of our own patients also had a generalised rash that disappeared on discontinuation of our epidural infusion containing metabisulphite. Therefore, the aim of the present study was to investigate the longterm stability at 2–8°C of such a triple component solution without metabisulphite, as well as the stability of the solution during 7 days at room temperature. The stability of a concentrate for the production of the ready-to-use infusion solution was also studied.
Methods This study was a shelf-life study of an epidural analgesic solution. No patients participated; therefore, no approval was sought from the local ethics committee.
Compounding of the solutions We describe briefly how the triple component epidural analgesia infusion is routinely produced via a concentrate that is diluted with saline to a ready-touse solution. The concentrate. The concentrate is produced by mixing adrenaline tartrate 41.5 μg/g (L-form), bupivacaine hydrochloride 10.9 mg/g, fentanyl citrate 34.6 μg/g, and disodium edetate 2.0 μg/g in nitrogen-purged water for injections in bulk. The pH is adjusted to 3.8–4.0 using sterile 1 M hydrochloric acid, and the solution is dispensed on 50 ml injection vials flushed with carbon dioxide. Following this, the vials are capped and autoclaved. They are then stored at 2–8°C, not exposed to light. The ready to use triple component epidural solution. The ready to use triple component epidural solution is prepared by Fresenius-Kabi (Halden, Norway) in polypropylene-based infusion bags (FreeFlex, Fresenius-Kabi) in volumes 110 ml (enough for obstetric practice), 250 ml (for 1–2 days epidural analgesia), or 550 ml – for more prolonged epidural analgesia. The solution is prepared in a gas isolator by diluting appropriate amounts of concentrate in infusion bags containing saline solution. For the present shelf-life study, the bags were prepared by diluting 10 ml of the concentrated solution in 100 ml 0.9% saline. The final volumes of the bags were thus nominally 110 ml, and the final nominal concentrations of the active ingredients in the readyto-use infusion solution were: adrenaline 2 μg/ml,
1322
bupivacaine 1 mg/ml, and fentanyl 2 μg/ml. Because infusion bags contain a certain amount overfill, and the concentrate routinely is added to the infusion bag without withdrawing any 0.9% saline, the actual concentration of the active ingredients in the ready-to-use infusion will be somewhat lower than the nominal concentration – corresponding to the degree of overfill. The bags used in this study could contain up to approximately 20% overfill.
Stability study The concentrate. Following production, the concentrate was stored at 2–8°C, not exposed to light, for 9 months. Every 3 months, the concentrations of the active ingredients were determined, the pH measured, and the appearance recorded. Appearance was investigated using the naked eye and the naked eye aided by polarised light. Six vials were investigated at each time point. The ready to use infusion solution. The 110-ml infusion bags containing ready-to-use triple component infusion solution were returned from FreseniusKabi to the Hospital Pharmacy at Oslo University Hospital, Rikshospitalet, where they were stored at 2–8°C for 4.5 months, not exposed to light. The concentrations of the three active ingredients, adrenaline, bupivacaine, and fentanyl, were determined at the receipt of the bags and again after 1.5, 3, and 4.5 months. The content of six infusion bags was determined each time. The bags were also inspected visually (using the naked eye and the naked eye aided by polarised light), the pH was measured, and the weight recorded. After storage for 4.5 months at 2–8°C, approximately 10 ml sample solution was withdrawn, and the infusion bags (n = 6) were stored in a constant climate chamber at 25°C and 40% relative humidity (RH) for 7 days, not exposed to light. Concentrations were determined, and physicochemical properties were recorded at the start and at the end of the 7-day period. Primary outcome measures. The solutions would be considered stable as long as the concentrations of the active ingredients did not deviate more than 10% from the initial concentrations, a level of change often allowed for drug solutions prepared in hospital pharmacies when the ‘therapeutic window’ is not very narrow. To be considered stable, pH values should be 4.0–5.0 for the ready-to-use infusion solution whereas values of 3.8–4.0 would be considered acceptable for the concentrate, the solu-
Stability of a sulphite-free solution
tions should be clear and free from particles, and the weight of the bags should not change by more than 5%.
Assay The stability indicating method used to determine fentanyl and bupivacaine in the triple component solution has been described in detail previously.15 In short, the method utilised an eluent consisting of 5 mM potassium dihydrogen phosphate and 1.25 mM tetrabutylammonium hydroxide adjusted with phosphoric acid to pH 4.8/methanol, 60 : 40 (v/v). The flow rate was 1 ml/min, and the detection wavelength was 210 nm. Linearity (r > 0.999) was demonstrated over the concentration range 0.05–10 μg/ml for fentanyl and 0.01–0.2 mg/ml for bupivacaine. Interday precision over 6 days was 0.8% of fentanyl and 0.5% for bupivacaine. The mean intraday precision was 1.1% for fentanyl and 0.8% for bupivacaine. The selectivity of the method was demonstrated as solutions of bupivacaine and fentanyl were subjected to pH 1 and 13 under heat (121°C), as well as hydrogen peroxide (10%, 24 h) without any interferences occurring. For the analysis of adrenaline, a gradient high performance liquid chromatography (HPLC) method was used. The first eluent (eluent A) consisted of 5 mM potassium dihydrogen phosphate adjusted with phosphoric acid to pH 3.6/methanol, 80 : 20 (v/v). The second eluent (eluent B) contained the same two ingredients, but in the ratio 40 : 60 (v/v). Eluent A ran from time 0 to 6.5 min, eluent B from 6.5 min to 25 min, and eluent A was used again from 25 min to 45 min. Adrenaline had a retention time of approximately 2.9 min. The detection wavelength was 210 nm, the flow rate 1.0 ml/min, and the injection volume 50 μl. When necessary, samples were diluted in eluent A before injection. The stability indicating the nature of the method for analysis of adrenaline was validated: Linearity over seven points between 0.1 and 10 μg/ml was demonstrated (r2 ≥ 0.999). The system precision (n = 6) was determined at 0.1 μg/ml, at 2.0 μg/ml, and at 7.5 μg/ml. The highest relative standard deviation was found to be 0.4%, at 0.1 μg/ml. The intra- and interday precision was demonstrated by determining the content of adrenaline in six samples of triple component solution on 6 days within a 4-week period. The interday precision was found to be 0.7% (SD% over the 6 days). The highest SD% found any of the 6 days for the six samples determined (intraday precision) was 0.4%. The selectivity of the method was demonstrated by subjecting
adrenaline (4 μg/ml) to acid and heat (pH 1, 121°C, 20 min), and base and heat (pH 11, 60°C, 30 min). No interference with adrenaline measurements was found when analysing the samples so treated.
Materials and instruments Potassium dihydrogen phosphate was obtained from Merck, KGaA (Darmstadt, Germany). Methanol was obtained from Rathburn Chemicals Ltd (Walkerburn, Scotland). Tetrabuthylammonium hydroxide was obtained from Fluka (Neu-Ulm, Switzerland), and Sigma-Aldrich Co (St Louis, MO, USA). Adrenaline bitartrate was obtained from Sigma-Aldrich, and Apotekproduksjon (Oslo, Norway). Bupivacaine hydrochloride was obtained from RTC (Laramie, WY, USA) and Apotekproduksjon. Sodium metabisulphite, disodium edetate, and fentanyl citrate were obtained from Apotekproduksjon. The system used for HPLC analysis consisted of an LC-20AD pump, a SIL-20AC autoinjector with sample cooler, a SPD-M20A diode array detector, a DGU-20A5 degassing unit, and a CBM-20A communication module, all units delivered by Shimadzu Corp. (Kyoto, Japan). For the chromatographic separation, a Supelcosil LC-DP column (100 mm × 4.6 mm, 5 μm) was used (Supelco, Belefonte PA, USA). For storage at controlled room temperature and humidity (25°C, 40% RH), a Termaks cabinet from Houm AS (Oslo, Norway) was used. pH was measured with a 691 pH meter from Metrohm (Herisau, Switzerland). For visual inspection of the solutions in polarised light, an Allen LV28 liquid inspection system was used (P.W.Allen & Co, Evesham, Worcester, UK).
Results The concentrate The concentration of all the three active ingredients in the triple component concentrate, stored at 2–8°C protected from light, remained well within 10% of the initial concentration throughout the whole test period of 9 months and showed little variation (Table 1). The concentrate was clear and colourless with no visible particles at all times, and the pH was 3.8 ± 0.0 throughout the study.
The ready to use infusion solution The concentration of the three active ingredients in the ready to use infusion solution stored in the dark for 4.5 months at 2–8°C also stayed well within 10%
1323
J. Brustugun et al. Table 1 Shelf life of concentrate for production of ready-to-use epidural analgesic solution stored at 2–8°C. Adrenaline % Bupivacaine % Fentanyl %
Start
3 Months
6 Months
9 Months
21.84 (0.11) μg/ml
21.78 (0.12) 99.8 10.75 (0.09) 100.2 21.39 (0.10) 99.4
21.84 (0.06) 100.0 10.78 (0.05) 100.5 21.44 (0.08) 99.7
21.96 (0.09) 100.6 10.76 (0.07) 100.4 20.96 (0.13) 97.4
10.72 (0.03) mg/ml 21.51 (0.11) μg/ml
Concentrations of adrenaline, bupivacaine, and fentanyl in concentrate vials are given as mean (SD), n = 6, and per cent remaining with respect to Start.
Table 2 Concentrations of adrenaline, bupivacaine, and fentanyl in a 110 ml ready to use epidural analgesic solution. Adrenaline % Bupivacaine % Fentanyl %
Start*
1.5 Months
3 Months
4.5 Months
4.5 Months, 7 days at RT
1.68 (0.03) μg/ml
1.70 (0.06) 101.5 0.83 (0.03) 98.5 1.69 (0.07) 101.1
1.61 (0.06) 96.1 0.86 (0.03) 101.5 1.74 (0.05) 103.9
1.63 (0.03) 97.5 0.85 (0.01) 100.9 1.72 (0.02) 102.6
1.67 (0.03) 99.9 0.86 (0.01) 101.2 1.71 (0.02) 102.5
0.85 (0.02) mg/ml 1.67 (0.03) μg/ml
The infusion solutions were stored in the dark, first at 2–8°C for 4.5 months, and then for 7 days at room temperature (RT; 25°C, 40% relative humidity). Data are given as mean (SD), n = 6, and per cent. Per cent values refer to per cent of Start concentration. *The actual start concentrations were lower than the nominal concentrations (2 μg/ml, 1 mg/ml, and 2 μg/ml) because the saline bags initially contained a certain amount of overfill.
Table 3 Physicochemical variables found in the ready to use epidural analgesic solution containing adrenaline, bupivacaine, and fentanyl.
pH Appearance Wight (g)
Start
1.5 Months
3 Months
4.5 Months
4.5 Months, 7 days at RT
4.76 (0.03) Clear, colourless 134.97 (0.50)*
4.78 (0.02) Clear, colourless 134.69 (0.74)
4.76 (0.03) Clear, colourless 134.93 (0.36)
4.79 (0.02) Clear, colourless 134.87 (0.26)†
4.77 (0.02) Clear, colourless 124.01 (0.20)
The solutions were stored in the dark, first at 2–8°C for 4.5 months, and then for 7 days at room temperature (RT; 25°C, 40% relative humidity). Data are given as mean (SD), n = 6. *n = 80. †The weight after sampling was 124.09 (0.20) g, n = 16.
of that initially determined (Table 2). The solutions were clear and colourless at all times, and the pH varied between 4.76 and 4.79, well within the acceptable interval of 4.0–5.0 (Table 3). The weight of the infusion bags at the end of the test period was 99.9% of the initial value. Upon further storing the 4.5 months old bags at room temperature (25°C, 40% RH) for 7 days, the concentrations of the three active substances still were all well within 10% of that determined at day 0 of the study (Table 2). The pH and appearance of the solutions also complied with the acceptance criteria during storage for 1 week at room temperature (Table 3). The weight of the bags was practically
1324
unchanged, 124.09 (0.20) g at the start and 124.01 (0.20) g at the end of the 7-day period.
Discussion The present study shows that a ready-to-use epidural analgesia solution not containing sulphites but consisting of adrenaline (2 μg/ml), fentanyl (2 μg/ml), bupivacaine (1 mg/ml), and disodium edetate (0.18 μg/ml) in 0.9% saline remains stable for both 4.5 months at 2–8°C and 7 days at room temperature (nominal concentrations given, for actual initial concentrations, see Table 2). In our experience, a shelf life of 4.5 months is sufficient
Stability of a sulphite-free solution
for production, quality control, and distribution. Because sulphites are allergenic and can occasionally cause anaphylactic reactions, documenting that adrenaline is stable without sulphites over a time span relevant for use is clinically of major importance. Sulphite allergy is not uncommon; for this reason, winemakers are obliged to declare whether sulphites are added to their finished products! Previously, a solution with the same concentrations of bupivacaine, fentanyl, and adrenaline, stabilised with sodium metabisulphite (1.8 μg/ml) and disodium edetate (0.18 μg/ml), having a pH of 4.7, was shown to be stable for 180 days at 4°C and 4 days at room temperature.10 This solution was made using raw materials in bulk. Other studies have been published on the stability of similar triple component solutions made from commercially available pharmaceutical products containing sulphite. In one such study, a solution containing the same three active ingredients in the same concentrations was made using an adrenaline solution for injection that contained 0.95 μg/ml sulphite (epinephrine injection BP, Phoenix Pharma, Gloucester, UK). The stability of that solution was documented for 184 days at both 4°C and 22°C.16 In another study, a solution containing levobupivacaine (570 μg/ml), fentanyl (3.7 μg/ml), and adrenaline (1.9 μg/ml) was shown to be stable for 60 days at 6°C and 40 days at 22°C.17 The adrenaline solution for injection used for preparing this solution (Adrenalin 0.1 mg/ml, Leiras, Helsinki, Finland) contained edetate and 1 mg/ml sodium metabisulphite. The actual sulphite species added when compounding adrenaline solutions varies: sulphite, bisulphite, and metabisulphite (pyrosulphite) are all used. Which of the sulphite species that is added during production is of less importance as they are all connected through chemical equilibria;9 the concentration and pH value of the solution will determine which species is finally predominant.
Why edetate is an effective stabiliser of adrenaline The results presented in the present study indicate that adrenaline in sulphite-free triple component solutions containing sodium edetate is stable for a period of at least 4.5 months (in the concentrate at least 9 months) when stored at 2–8°C. Several factors may have contributed to the stability. (1) The solutions were stored at a low temperature, 2–8°C. (2) The water used for the preparation of the concentrate was purged with nitrogen to remove oxygen both before the components were mixed together
and also during the mixing process. (3) The 50 ml vials used to contain the concentrate were flushed with carbon dioxide, further preventing oxygen from interacting with the adrenaline. (4) The solutions contained the chelator disodium edetate, and chelators have previously been demonstrated to be effective stabilisers of adrenaline.18 For the edetate to be effective as a stabiliser, the content of trace metal ions should probably be kept as low as possible in the solutions. Using raw materials, drug substances and water, all of high purity, is therefore probably of particular importance in this instance – not only to ensure patient safety, but also to improve drug stability. The ingredients used in the production of the concentrate in our hospital pharmacy comply with the specifications in the European Pharmacopeia. The water used is Water for injections in bulk, in our facility normally having a conductivity of ≤1.0 μS/cm, indicating a low content of ions.
Is edetate biologically safer than sulphites? Edetic acid and edentate salts are substances widely used in pharmaceutical formulations, typically in concentrations ranging from 0.05 to 1 mg/ml.19 The present study documents that disodium edetate at a low level of 0.18 μg/ml is sufficient to stabilise adrenaline (2 μg/ml) in a ready-to-use infusion solution, provided the ingredients are of high purity and steps are taken to remove oxygen (see above). Edetic acid is generally regarded as essentially nontoxic and nonirritant,19 and the nonirritant nature of the salt disodium edetate has been demonstrated through human patch testing.20 Disodium edetate is also included in the U.S. Food and Drug Administration Inactive Ingredients Guide. A concentration of 0.111 mg/ml (0.0111%) being approved for epidural use – a value well above the concentration used in the epidural solution presented here (0.18 μg/ml). As shown in Table 2, the actual initial concentrations of the active substances in the infusion solution are lower than could be expected after a 1 : 11 dilution of the concentrate. The reason for this is that the bags contain a certain amount overfill and that the concentrate is added to the infusion bag without withdrawing any saline. The 100 ml bags used in this study can contain up to 20% overfill, explaining the initial concentrations seen (Table 2). It should be noted that despite the dilution caused by overfill, the concentration of adrenaline in this study remained above the minimally effective concentration of 1.5 μg/ml determined in earlier studies.7 It
1325
J. Brustugun et al.
should also be noted that larger infusion bags contain a lower relative amount of overfill. For instance, the overfill in the 500 ml bags used for preparation of larger volume ready-to-use infusion solution is limited upwards to 7%. Using a larger infusion bag can thus cause less dilution of both the active substances and the edetate – and less dilution of edetate could possibly improve the stability of the infusion solution. In a recent study, it was stated that the degradation of adrenaline in a triple component analgesic admixture containing levobupivacaine, fentanyl, and adrenaline was mostly dependant on light – light having a destabilising effect.17 However, the adrenaline injection used for the preparation of that solution (Adrenalin 0.1 mg/ml, Leiras) contains sulphite, and this may have influenced the result. It has previously been shown that sulphites can make catecholamines, such as adrenaline, less stable when subjected to light.15,21,22 It has also been shown that a triple component solution containing adrenaline, fentanyl, and bupivacaine, but no sulphite, is more stable in light than solutions containing sulphite when tested according to the specifications for photostability testing outlined in the guideline of the International Conference on Harmonisation (ICH), ICH Q1B.15 The photodestabilising effect of sulphite is caused by adrenochrome, a degradation product of adrenaline, first reacting with bisulphite to form an adrenochrome-bisulphite addition product.23 This substance can, when exposed to light, produce reactive singlet oxygen capable of oxidising adrenaline.21 Because the adrenochrome-bisulphite addition product can no longer be formed in a sulphite-free solution, and because a sulphite-free triple component infusion has previously been shown to be stable for periods covering normal use when exposed to light under the ICH criteria, we no longer recommend that infusion bags containing the triple component infusion are protected from light during use at our hospital. A possible limitation of our study is that the HPLC method used to determine adrenaline did not differentiate between the two optical isomers of adrenalin. As the L-form of adrenaline is the more biologically active one, also at the α2-receptors of the spinal cord dorsal horn, detection of racemisation could be of interest. The racemisation of adrenaline is normally a slow process in pharmaceutical products. Stepensky et al. found a racemisation rate of 5.6% over a period of 2 years,24 and others have extrapolated a shelf life (90% remaining) regarding
1326
racemisation of 120 months at pH 3.5.9 However, future studies on the stability of sulphite-free multicomponent epidural solutions containing adrenaline should ideally include investigation of the racemisation of adrenalin as well as the effects of temperature, chelator concentration, water quality, and the use of oxygen-purging agents (i.e. carbon dioxide or nitrogen) on the stability of the active ingredients. In conclusion, this study documents that it is possible to make a multicomponent infusion solution for epidural analgesia, containing adrenaline but no sulphites, that is sufficiently stable for production, distribution, and use. Such a solution containing adrenaline (2 μg/ml), fentanyl (2 μg/ml), bupivacaine (1 mg/ml), and disodium edetate (0.18 μg/ml) adjusted to pH 4.8 has been shown to be stable for at least 4.5 months at 2–8°C in polypropylene-based infusion bags, and also for 7 days at room temperature. A concentrate for the preparation of ready-to-use infusion solution, 11 times higher in concentration of the ingredients, was stable (> 90%) at 2–8°C for 9 months. The important implication of our present findings is that it is now not necessary to expose patients to sulphites when administering an epidural analgesic infusion solution containing adrenaline (with fentanyl and bupivacaine), avoiding the (slight) risk of precipitating allergic reactions in patients who are already allergic to sulphites.
Acknowledgements The authors thank Håvard Hem for technical assistance. The study was supported by Fresenius Kabi, Halden, Norway, and the Hospital Pharmacy at Oslo University Hospital, Rikshospitalet, Oslo, Norway. Conflict of interest: Inven2, a government institution for funding and commercialisation of research at the Oslo University Hospital, receives royalties of sales of the ready-to-use epidural solution from Fresenius Kabi, Halden, Norway. The authors have previously received research funding from Inven2.
References 1. Ginosar Y, Riley ET, Angst MS. The site of action of epidural fentanyl in humans: the difference between infusion and bolus administration. Anesth Analg 2003; 97: 1428–38. 2. Bromage PR, Camporesi EM, Durant PA, Nielsen CH. Influence of epinephrine as an adjuvant to epidural morphine. Anesthesiology 1983; 58: 257–62. 3. Eisenach JC, Grice SC, Dewan DM. Epinephrine enhances analgesia produced by epidural bupivacaine during labor. Anesth Analg 1987; 66: 447–51. 4. Niemi G, Breivik H. Adrenaline markedly improves thoracic epidural analgesia produced by a low-dose infusion of bupivacaine, fentanyl and adrenaline after major surgery. A
Stability of a sulphite-free solution
5.
6.
7.
8.
9. 10.
11. 12. 13. 14. 15.
randomised, double-blind, cross-over study with and without adrenaline. Acta Anaesthesiol Scand 1998; 42: 897– 909. Förster JG, Lumme HM, Palkama VJ, Rosenberg PH, Pitkänen MT. Epinephrine 4 microg/mL added to a lowdose mixture of ropivacaine and fentanyl for lumbar epidural analgesia after total knee arthroplasty. Anesth Analg 2008; 106: 301–4. Collins JG, Kitahata LM, Matsumoto M, Homma E, Suzukawa M. Spinally administered epinephrine suppresses noxiously evoked activity of WDR neurons in the dorsal horn of the spinal cord. Anesthesiology 1984; 60: 269– 75. Niemi G, Breivik H. The minimally effective concentration of adrenaline in a low-concentration thoracic epidural analgesic infusion of bupivacaine, fentanyl and adrenaline after major surgery. A randomized, double-blind, dosefinding study. Acta Anaesthesiol Scand 2003; 47: 439– 50. Curatolo M, Schnider TW, Petersen-Felix S, Weiss S, Signer C, Scaramozzino P, Zbinden AM. A direct search procedure to optimize combinations of epidural bupivacaine, fentanyl, and clonidine for postoperative analgesia. Anesthesiology 2000; 92: 325–37. Connors KA. Epinephrine. In: Connors KA, Amidon GL, Stella VJ eds. Chemical Stability of Pharmaceuticals, 2nd edn. New York: John Wiley & Sons, 1986: 438–47. Kjønniksen I, Brustugun J, Niemi G, Breivik H, Anderssen E, Klem W. Stability of an epidural analgesic solution containing adrenaline, bupivacaine and fentanyl. Acta Anaesthesiol Scand 2000; 44: 864–7. Dooms-Goossens A, De Alam AG, Degreef H, Kochuyt A. Local anesthetic intolerance due to metabisulfite. Contact Dermatitis 1989; 20: 124–6. Smolinske SC. Review of parenteral sulfite reactions. J Toxicol Clin Toxicol 1992; 30: 597–606. Madan V, Walker SL, Beck MH. Sodium metabisulfite allergy is common but is it relevant? Contact Dermatitis 2007; 57: 173–6. Vally H, Misso NLA, Madan V. Clinical effects of sulphite additives. Clin Exp Allergy 2009; 39: 1643–51. Brustugun J, Tønnesen HH, Klem W, Kjønniksen I. Photodestabilization of epinephrine by sodium metabisulfite. PDA J Pharm Sci Technol 2000; 54: 136–43.
16. Priston MJ, Hughes JM, Santillo M, Christie IW. Stability of an epidural analgesic admixture containing epinephrine, fentanyl and bupivacaine. Anaesthesia 2004; 59: 979–83. 17. Helin-Tanninen M, Lehtonen M, Naaranlahti T, Venäläinen T, Pentikäinen J, Laatikainen A, Kokki H. Stability of an epidural analgesic admixture of levobupivacaine, fentanyl and epinephrine. J Clin Pharm Ther 2013; 38: 104–8. 18. Roscoe CW, Hall NA. Chelating agents as color stabilizers for epinephrine hydrochloride solutions. J Am Pharm Assoc 1956; 45: 464–70. 19. Owen SC. Edetic acid. In: Rowe RC, Shestkey PJ, Weller PJ eds. Handbook of Pharmaceutical Excipients, 4th edn. London: Pharmaceutical Press; Washington, DC: American Pharmaceutical Association, 2003: 225–8. 20. Basketter DA, Chamberlain M, Griffiths HA, Rowson M, Whittle E, York M. The classification of skin irritants by human patch test. Food and Chemical Toxicology 1997; 35: 845–52. 21. Brustugun J, Kristensen S, Tønnesen HH. Photostability of epinephrine – the influence of bisulfite and degradation products. Pharmazie 2004; 59: 457–63. 22. Brustugun J, Kristensen S, Tønnesen HH. Photostability of sympathomimetic agents in commonly used infusion media in the absence and presence of bisulfite. PDA J Pharm Sci Technol 2004; 58: 296–308. 23. Heacock RA, Powell WS. Adrenochrome and related compounds. In: Ellis GP, West GB eds. Progress in Medical Chemistry 9. Amsterdam: North Holland Publishing Company, 1973: 275–339. 24. Stepensky D, Chorny M, Dabour Z, Schumacher I. Longterm stability study of L-adrenaline injections: kinetics of sulfonation and racemization pathways of drug degradation. J Pharm Sci 2004; 93: 969–80.
Address: Jørgen Brustugun Hospital Pharmacy Oslo University Hospital Rikshospitalet Pb 4950 Nydalen 0424 Oslo Norway e-mail: [email protected]
1327
