Stability of Hydrocortisone in Extemporaneously Compounded Suspensions

Abstract

Rationale: The recent withdrawal in Canada of the only previously commercially available hydrocortisone suspension (Cortef®) from the market, due to concerns regarding its potency, has prompted the need for extemporaneous compounding of hydrocortisone, a corticosteroid used for a variety of conditions. While hydrocortisone is available in tablet form for oral use, pediatric and other patients who are unable to swallow tablets require an alternative dosage form.

Objective: To evaluate the stability of 1 mg/mL and 2 mg/mL hydrocortisone suspensions in a vehicle consisting of equal parts of Ora-Sweet and Ora-Plus after storage at 4°C and 25°C for up to 91 days.

Methods: Suspensions of hydrocortisone 1 mg/mL and 2 mg/mL in a 1:1 mixture of Ora-Sweet and Ora-Plus were prepared in 50-mL amber plastic prescription bottles. Three bottles of each suspension were stored at 4°C (refrigerated), and 3 were stored at 25°C (room temperature). Physical characteristics, including pH, colour, odour, viscosity, precipitation, and ease of resuspension, were observed weekly for 91 days. Aliquots were removed from each bottle weekly for 91 days and stored at -85°C until analysis by a validated high-pressure liquid chromatography method. A suspension was considered stable if it maintained at least 90% of its initial concentration.

Results: No change in pH was observed in suspensions of either concentration after storage at 4°C and 25°C for 91 days. Changes in colour and odour were slight, if any. Viscosity was constant. Precipitates were easily resuspended, and there was no caking or clumping of material. Suspensions of both concentrations of hydrocortisone maintained at least 90% of initial concentration at both temperatures throughout the 91-day period.

Conclusions: Hydrocortisone suspensions of both 1 mg/mL and 2 mg/mL in a 1:1 mixture of Ora-Sweet and Ora-Plus were physically and chemically stable for a period of up to 91 days, with or without refrigeration. The expiry date for these products can therefore be set at 91 days.

J Inform Pharmacother 2003;13:100-110.

Introduction

Many medications are not available in a dosage form suitable for the child or adult who is unable to swallow tablets or capsules. Hence, a liquid formulation is needed. However, in the case for hydrocortisone, a corticosteroid commonly prescribed for a variety of conditions, the only previously available suspension (Cortef®) on the Canadian market is no longer manufactured due to concerns of potency. At the authors’ hospital, until the time of this study, hydrocortisone was extemporaneously compounded by crushing hydrocortisone 10 mg tablets and then suspending in a methylcellulose vehicle prepared by their pharmacy; it was given an expiry date of 7 days for refrigerated storage. This short expiry date meant that hospital pharmacy staff had to prepare the suspension frequently and that discharged patients had to return to the hospital frequently for refills of their prescriptions, leading to relatively high expenditures of both money and time for all concerned.

A combination of Ora-Sweet and Ora-Plus (Paddock Laboratories Inc., Minneapolis, Minnesota), a sweetening and suspending agent, respectively, has recently come on the market and is available to both hospital and community pharmacies. A 1:1 mixture of these products used as a drug vehicle helps to ease product preparation and improve palatability. At this time, however, there is no published information on the stability of extemporaneously compounded hydrocortisone suspensions.

This study examined the physical characteristics and chemical stability (defined as maintenance of at least 90% of initial concentration) of extemporaneously prepared oral hydrocortisone suspensions of 1 mg/mL and 2 mg/mL in a 1:1 mixture of Ora-Sweet and Ora-Plus, stored at 4°C and 25°C throughout a 91-day study period.

Methodology

Preparation of Suspension

Hydrocortisone suspensions (1 mg/mL and 2 mg/mL) were prepared in triplicate from commercially-available hydrocortisone 10 mg tablets (Cortef®, Pharmacia & Upjohn, Mississauga, Ontario), and a 1:1 mixture of Ora-Sweet and Ora-Plus (Paddock Laboratories Inc., Minneapolis, MN, lots OH6525 and OE6324, respectively). Ten tablets were triturated at a time in a mortar and the remaining powder was transferred to a beaker, following multiple rinsings with Ora-Sweet and Ora-Plus. Six replicates of each concentration were prepared in separate 50-mL amber plastic prescription bottles; three were stored at 4°C (refrigerated) and three at room temperature (approximately 25°C with minimal variation). All bottles were exposed only to fluorescent light in the laboratory.

The physical appearance of the suspensions was evaluated qualitatively at the time of preparation and at weekly intervals up to 91 days. All suspensions were examined for changes in colour (against white and black backgrounds). As there was no fresh suspension for comparison purposes, subtle differences in colour could not be detected. Also examined were odour, pH, viscosity, formation of precipitates, and ease of resuspension. All samples were allowed to equilibrate to room temperature before each evaluation, including measurement of pH, and the pH meter (model 8000, VWR Canlab, Mississauga, Ontario) was calibrated at the beginning of each testing period. Immediately after the physical observations were completed, each bottle was manually shaken for 10 seconds, pH was determined, and 1 mL of the suspension was removed. These samples were stored at -85°C until batch analysis by a stability-indicating high performance liquid chromatography (HPLC) method.

Results

Regression analysis of the peak area ratio of hydrocortisone to internal standard versus concentration demonstrated linearity over the working range of the concentrations, with coefficients of determination (r2) greater than 0.996 (n=4). The intra-day (n=4) and inter-day (n=4) coefficients of variation for the 3 different concentrations were within acceptable limits: 5.39% and 2.93%, respectively, for the 0.25 mg/mL suspension; 0.10% and 2.05%, respectively, for the 1.00 mg/mL suspension; 0.35% and 1.98%, respectively, for the 2.00 mg/mL suspension.

When hydrocortisone was subjected to degradation, there was no significant peak (of hydrocortisone) observed at 3.4 minutes. An unidentified peak that did not interfere with the quantification of the parent hydrocortisone compound or the methylprednisolone internal standard was noted at 5.01 minutes (Figure 1). Thus, the HPLC method was deemed capable of indicating stability.

Figure 1. Upper panel: Representative chromatogram showing the hydrocortisone peak (A) 2 mg/mL at 3.4 minutes and the methylprednisolone peak (B) at 4.6 minutes. Lower panel: Chromatogram depicting the degradation study.

No significant changes in physical appearance or odour of the suspensions were observed throughout the 91 days. Each cloudy, white suspension had a faint sweet smell, maintained constant viscosity, and was easily resuspended throughout the study period. Furthermore, no significant fluctuations of pH were observed (Figure 2). The mean pH (± standard deviation) was 4.47±0.03 for the 1 mg/mL suspension stored at 4°C, 4.30±0.03 for the 1 mg/mL suspension stored at 25°C, 4.48±0.02 for the 2 mg/mL suspension stored at 4°C, and 4.34±0.02 for the 2 mg/mL suspension stored at 25°C.

Figure 2. pH data of hydrocortisone suspensions 1 mg/mL and 2 mg/mL pH stored at 4°C and 25°C over the 91-day study period.

The retention time for hydrocortisone was 3.4 minutes, whereas the retention time for the internal standard, methylprednisolone, was 4.6 minutes (Figure 1). The HPLC analysis showed that, at both storage temperatures, the 1 mg/mL and the 2 mg/mL suspensions maintained at least 90% of their initial concentrations on every study day (Table 1). Furthermore, more than 91% of the initial hydrocortisone concentration remained on day 91, according to linear regression analysis of the concentration-time data. The lower limit of the 95% confidence interval also indicated that more than 91% of the initial concentration remained on day 91 (Table 1).

Table 1. Mean hydrocortisone concentration ± standard deviation (and mean percentage remaining*) during 91 days of storage at 4°C and 25°C.

1 mg/mL† 2 mg/mL†
Study Day 4°C 25°C 4°C 25°C
0 1.126±0.109 1.134±0.111 2.004±0.211 1.990±0.141
7 1.187±0.070 (105.42) 1.187±0.438 (104.67) 2.108±0.517 (105.19) 2.066±0.339 (103.82)
14 1.078±0.220 (95.74) 1.097±0.201 (96.71) 2.008±0.334 (100.20) 2.019±0.279 (101.46)
21 1.143±0.306 (101.51) 1.167±0.337 (102.91) 1.948±0.149 (97.21) 2.019±0.308 (101.46)
28 1.122±0.085 (99.64) 1.115±0.333 (98.32) 2.360±0.552 (117.76) 1.995±0.293 (100.25)
35 1.210±0.207 (107.46) 1.285±0.195 (113.32) 2.356±0.120 (117.56) 2.144±0.251 (107.74)
42 1.165±0.126 (103.46) 1.153±0.012 (101.68) 1.954±0.112 (97.50) 1.860±0.312 (93.47)
49 1.067±0.041 (94.76) 1.122±0.154 (98.94) 2.051±0.042 (102.35) 1.977±0.216 (99.35)
56 1.248±0.155 (110.83) 1.173±0.018 (103.44) 2.220±0.222 (110.80) 2.136±0.477 (107.32)
63 1.168±0.196 (103.76) 1.103±0.377 (97.24) 1.830±0.416 (91.30) 2.160±0.403 (108.53)
70 1.350±0.170 (119.86) 1.034±0.152 (91.18) 2.029±0.193 (101.26) 2.307±0.126 (115.93)
77 1.033±0.117 (91.71) 1.128±0.277 (99.47) 1.828±0.495 (91.22) 1.819±0.030 (91.42)
84 1.212±0.433 (107.64) 1.227±0.314 (108.17) 2.008±0.329 (100.22) 2.208±0.210 (110.94)
91 1.258±0.176 (111.72) 1.110±0.256 (97.85) 2.131±0.259 (106.34) 2.091±0.350 (105.06)
% remaining on day 91 by linear regression‡ 91.68 102.41 104.93 94.26
Lower limit of 95% CI for % remaining§ 91.61 96.04 96.00 93.19

CI = confidence interval.

* Percentage remaining was calculated in relation to the initial concentration (day zero).

† Nominal concentration.

‡ Calculated from concentration on day 91 as determined by linear regression and concentration observed on day zero, according to the following formula: concentration on day 91/concentration on day zero x 100.

§ Calculated from lower limit of 95% CI of the slope of the curve relating concentration to time, determined by linear regression, according to the following formula: lower limit of 95% CI of concentration on day 91/concentration on day zero x 100.

Discussion

Until the time of this study, hydrocortisone suspension had been prepared at the authors’ institution from hydrocortisone 10 mg tablets and methylcellulose. This product had an empiric expiry date of 7 days when kept refrigerated.

To the authors’ knowledge, there are no published stability studies on hydrocortisone suspension prepared in a 1:1 mixture of Ora-Sweet and Ora-Plus. A MEDLINE search (1966 to 2003) with the terms hydrocortisone, stability, and HPLC yielded no relevant articles. However, information regarding the 7-day stability of hydrocortisone suspension in simple syrup and methylcellulose is available from the Hospital for Sick Children web site.

In the weekly analysis, colour and odour changes were slight, and samples were easily resuspended without caking, clumping, or crystal formation. Although these measures are qualitative, the observations were documented by the same individual throughout the 91 days, thus eliminating inter-personnel bias. Variation in pH was not notable; mean pH ranged from 4.30 to 4.47 in the 1 mg/mL samples and from 4.34 to 4.48 in the 2 mg/mL samples.

A limitation of this study design relates to the freezing of the samples at -85°C until the time of batch analysis. It was assumed that hydrocortisone would not degrade at this low temperature, and that no volume losses would occur because of freeze-drying during storage. In addition, it was assumed that errors due to serial analysis would have been greater than errors occurring with batch analysis.

Conclusions

According to qualitative and HPLC analyses of weekly samples, hydrocortisone suspensions of 1 mg/mL and 2 mg/mL in a 1:1 vehicle of Ora-Sweet:Ora-Plus stored at either 4°C and 25°C remained stable and maintained at least 90% of their original concentration for a period of up to 91 days.

Acknowledgements

The authors thank Donald P. Hamilton, B.Sc. (Pharm.), and Paul M. Koke, B.Sc. (Pharm.) for their assistance in project logistics.  This study was supported by the British Columbia Children’s Hospital Telethon Innovations Fund and the B.C. Research Institute for Children’s & Women’s Health.

Appendix: Preparation of Stock and Standards, Degradation of Hydrocortisone, Preparation of Samples, and Statistical Analysis

Preparation of stock and standards

Stocks of hydrocortisone at 0.25, 0.50, 1.00, 2.00, and 4.00 mg/mL were prepared by diluting triturated hydrocortisone 10 mg tablets in equal parts of Ora-Sweet and Ora-Plus. The internal standard was methylprednisolone (Sigma Aldrich, Oakville, Ontario, lot 101K1318) at a concentration of 1.0 mg/mL in HPLC-grade methanol (Fisher Scientific, Richmond, British Columbia, lot 011625). Standards were prepared by combining 0.1 mL aliquot of each stock and 0.1 mL aliquot of methylprednisolone 1.0 mg/mL and further diluting with HPLC grade methanol to make up 1.0 mL volumes. The final concentrations of hydrocortisone in the samples injected onto the chromatograph were 0.025, 0.050, 0.100, 0.200, and 0.400 mg/mL. This dilution step was necessary to achieve optimal chromatographics. Prior to injection, all standards were passed through a 0.45 mm microfilter (Acrodisc GHP syringe filter, Gelman, Ann Arbor, Michigan, lot 2864) to prevent injection of impurities onto the column.

The HPLC instrumentation (model 2690, Waters Alliance Systems, Waters Ltd., Mississauga, Ontario) consisted of a delivery pump, an automatic injector equipped with a 200 mL injector, a Symmetry 3.9 x 20 mm guard column (Waters Ltd., lot T22611), a Symmetry Shield RP C18 4.6 x 100 mm column (Water Ltd., lot W22281K25), and an ultraviolet detector (model 2487 dual-wavelength absorbance detector, Waters Ltd.,). The mobile phase was developed in the authors’ laboratory and consisted of a 20:60:20 (v/v) mixture of water (Fisher Scientific, lot 023349), methanol (Fisher Scientific, lot 011625), and 10 mmol/L potassium phosphate buffer (Sigma Aldrich, lot 90K0923; pH 3.0). All solvents were HPLC-grade and filtered before use. The flow rate was set at 1.2 mL/min.

A 5-point calibration curve was prepared, with a blank (methanol only) at the beginning of each run, to ensure that there was no carry-over from one run to the next. The range of this calibration curve (0.25 to 4.00 mg/mL before dilution) encompassed the diluted test concentrations of the 1 mg/mL and 2 mg/mL. The calibration curve was generated by least-squares regression of the peak area ratio of hydrocortisone to methylprednisolone and the concentration of each standard. The precision of the assay was evaluated by intra-day and inter-day validation methods. Intra-day variability was determined by running 0.25, 1.00, and 2.00 mg/mL stocks (diluted to standards of 0.025, 0.100, and 0.200 mg/mL) in quadruplicate throughout a single day, whereas inter-day variability was determined by running the same concentrations (as in the testing for intra-day variability) in quadruplicate daily for 4 days. The means, standard deviations, and coefficients of variation were then calculated. Acceptable limits for the coefficients of variation were defined a priori as less than 10%.

Degradation of hydrocortisone

Triturated hydrocortisone 10 mg tablets in methanol were diluted in water to a concentration of 2 mg/mL, and the diluted sample was made basic with 10% of sodium hydroxide 10N solution 30% w/v (Fisher Scientific, log SC9105042) to a pH of 12.98. This preparation was incubated in a water bath at 90°C for 72 hours. The chromatogram obtained for the degraded preparation was compared with a chromatogram obtained from a standard (2 mg/mL) to determine any changes in concentration, retention time, and peak shape.

Preparation of samples

Hydrocortisone study samples were thawed and centrifuged at 3,000 RPM for 5 minutes, and then processed in a manner similar to that for the stocks. For each study sample (both concentrations, both temperatures, in triplicate), a 0.1 mL aliquot was diluted with 0.8 mL HPLC-grade methanol, and a 0.1 mL aliquot of methylprednisolone 1 mg/mL in HPLC-grade methanol was added. The final theoretical concentrations were 0.1 and 0.2 mg/mL, respectively. Each sample was passed through a 0.45 mm microfilter before a 10 mL sample was withdrawn and injected onto the column.

Statistical analysis

The means, standard deviations, and coefficients of variation were calculated for samples analyzed in triplicate and quadruplicate. For each study day, the percentage of initial hydrocortisone concentration remaining was calculated for each sample. The percentage of hydrocortisone remaining on day 91 was calculated from the concentration on day 91 as determined by linear regression and concentration observed on day zero, according to the following formula: concentration on day 91/concentration on day zero x 100%. The 95% confidence interval (CI) of the amount remaining on the last study day was calculated from the lower limit of the 95% CI of the slope of the curve relating concentration to time, determined by linear regression, according to the following formula: lower limit of the 95% CI of the concentration on day 91/concentration on day zero x 100%. Stability was defined as maintenance of at least 90% of the initial hydrocortisone concentration.

Latest Posts