Sometimes unlicensed oral liquid medicines are prepared in dispensaries in order to meet the needs of specific patient groups.
Andrew Lowey and Mark Jackson look at methods of improving the quality and formulation of these liquids, using captopril as an example
Pharmacists are responsible for ensuring that drug use is safe and effective. Wherever practicable, medicines are used in line with their product licences, in order to ensure that risks to patient safety are minimised.
Medicines should not be used in an unlicensed manner unless and until the formulation’s quality and fitness for purpose have been assessed by a suitably competent pharmacist.
However, licensed preparations may be inappropriate for the end user, for example, patients may not be able swallow tablets or capsules. Therefore, unlicensed oral liquid medicines are extemporaneously prepared in dispensaries in order to meet the needs of specific patient groups.
Extemporaneous preparation is one of the highest risk activities carried out in pharmacy today and is often supported by a poor quality evidence base. Unlicensed medicines use carries a greater risk of adverse drug reactions than the use of licensed medicines.1,2
The risks of unlicensed medicines use are combined with inherent risks associated with the pharmaceutical compounding process.3
In addition, extemporaneously prepared medicines are commonly given to some of the most vulnerable patients in hospitals and in the community, such as, neonates, children, elderly patients, stroke victims, and patients with feeding tubes. These people are often not capable of alerting carers or staff to any adverse drug events they are experiencing.
There are a number of examples of serious negative outcomes from extemporaneous preparation errors, including the “peppermint water case” (where the use of concentrated chloroform water led to the death of a child).4
This case highlighted the dangers of potent ingredients and calculation errors, particularly where the strength of one or more ingredient is stated in a historical or non-standard fashion. Similar reports have originated from the US, with the death of a child from a superpotent imipramine liquid, and a five-year-old child who received a 1,000-fold overdose of clonidine.5,6
Related administration errors have also been reported, including five-fold, 10-fold, 100-fold and 1,000-fold mistakes in measuring doses. Some of these errors have been attributed to inconsistent labelling approaches, for example, strengths expressed per millilitre or per 5ml spoonful.
The relative lack of research and development work on extemporaneously prepared products is associated with the potential for formulation failure, resulting in the risk of overdose, underdose or poor uniformity of dose. The lack of high quality data to support historical formulae has been acknowledged by authors across the world.7,8,9
Even where a given formulation has been shown to achieve suitable physical, chemical and microbiological stability, the bioavailability and palatability of the preparation may be unproven. Few extemporaneous preparations have any data to demonstrate a suitable absorption profile or bioequivalence with a licensed preparation.
Other issues include concerns regarding substandard facilities and the absence of equipment and materials needed to provide an extemporaneous dispensing service.10,11
Following the peppermint water case, the Royal Pharmaceutical Society published a document laying down brief standards of good professional practice for the dispensing of extemporaneous preparations.12
These standards were subsequently included in “Medicines, ethics and practice: a guide for pharmacists” from the January 2001 issue.13
The launch of the “Guide to the preparation of non-sterile extemporaneous products in NHS hospitals” in 200314 represented the publication of the first set of detailed guidance for this area of pharmacy.
The risks associated with extemporaneous dispensing have been potentiated by the declining expertise in pharmaceutics and formulation within the profession.8,15,16
Moreover, compounding of oral medicines is often delegated to junior or trainee staff and, commonly, there is no quality assurance system in place to support the practice.15
Although changes have been made to pharmacist undergraduate training since the peppermint water case, Chowdury et al reported that of the 16 UK schools of pharmacy surveyed in 2003, only nine responded to say that they would be confident that a student of average ability would, on graduation, be competent to dispense extemporaneously.16
Neill has suggested that some pharmacists may not encounter requests for extemporaneously dispensed products often enough to maintain and develop their confidence, and may be reticent in asking for advice.17
The lack of compounding skills in newly qualified pharmacists means that personnel responsible for historical formulations may have been replaced by staff who lack the necessary skills to reformulate a preparation, as opinions or ingredients are superseded.
While some skills may be retained in NHS production and quality control departments, pharmacists outside these environments are likely to have less experience or access to relevant information sources.
The NHS Modernisation Agency (now superseded by the NHS Institute for Innovation and Improvement) recently funded a research project designed to investigate the quality and formulation of unlicensed oral medicines in the NHS. The project was directed by the NHS Pharmaceutical Quality Assurance Committee’s Working Group for Extemporaneous Dispensing and was hosted at Leeds Teaching Hospitals NHS Trust.
In December 2005, a questionnaire was sent to all NHS trusts in Yorkshire, the north-east of England and London. In the preceding 12 months, 117 different drugs had been extemporaneously prepared. The project considered the quality of evidence available to support the most commonly used 50 extemporaneous preparations, and collated this information in a formal stability summary for each of the drugs.
These stability summaries will be published in a textbook in the near future, together with updated national standards for extemporaneous dispensing. The standardisation and rationalisation of simple extemporaneous formulations in this way will help to decrease the risks associated with complex calculations, potentially dangerous ingredients and variations in bioavailability.
The need for such a collection of standard formulae has been increased since the demise of official formulations in the British Pharmacopoeia and the British Pharmaceutical Codex.
Captopril oral liquid
One of the most controversial preparations to be dispensed out of the 50 most commonly used extemporaneous preparations was captopril oral liquid. Twelve centres reported its preparation, comprising three different formulations with shelf-lives ranging from seven to 28 days.
The variation in shelf-lives correlates well with the research carried out by Mulla et al, who found serious issues with regard to the lack of consistency and potential for toxicity with captopril liquid.18
They described the variation seen in 13 tertiary paediatric centres and 13 large hospitals in the UK. Four hospitals dispensed captopril tablets for crushing and dissolving in water before administration. The remaining 22 hospitals used nine different liquid formulations, including imported products, “specials” and extemporaneous formulations.
The researchers concluded that the serious inconsistency raises issues about optimal captopril dosing such that its use may influence paediatric cardiac surgical and interventional outcomes.
Captopril oral liquid is an ideal example to be considered as part of the modernisation agenda within pharmacy manufacturing:
- Should it be manipulated at ward level?
- Should it be extemporaneously dispensed?
- Should it be commissioned as a “special”?
- Should it be imported?
- Could an alternative therapeutic agent be used?
These are the questions considered by the project team for each of the 50 most commonly used preparations.
Physical and chemical stability
Captopril (molecular weight 217.3) is a white or almost white crystalline powder. It is freely soluble in water and may have a characteristic sulphide-like odour.19,20
The drug exhibits a reported solubility of 160mg/ml at 25C.21 Therefore, it is likely to be in solution rather than in suspension at most usual concentrations in aqueous oral liquids. While this may help to guarantee a uniform and reproducible dose, most drugs are less stable in solution rather than in suspension.22,23
As a solid, captopril exhibits excellent stability, with no significant decomposition detected for up to six months in bulk sample stored at 5C, 33C and 50C.24 Under ordinary conditions, it is not hygroscopic.21
In solution, however, captopril is much more unstable, undergoing an oxygen-facilitated, first order, free radical oxidation at its thiol group and yielding the major degradation product captopril disulphide.21
The reaction is complex, and it has been suggested that the reaction rate may change from first order to zero order when the captopril concentration decreases below a certain level.25
Oxidation is reported to be delayed by lowering pH, adding chelating or antioxidant agents and using of nitrogen or low oxygen head spaces during production.21
Captopril degradation is known to be inversely proportional to drug concentration26 (ie, as the captopril concentration rises, degradation falls). This is due to stoichiometry of the oxidation, with two moles of captopril lost for every half a mole of oxygen consumed.
The oxidation is also known to be pH-dependent, with captopril exhibiting its highest stability at a pH below 3.5. Above pH 4, the rate constants rises rapidly.21
As for many oxidation reactions, the degradation can be catalysed by metal ions, such as copper and iron.27 As little as 1ppm of copper can catalyse the process. So, some formulators have investigated the use of metal ion chelators in captopril formulations.
Captopril in water The data for captopril solutions are wide ranging and contain conflicting opinions. The stability of captopril in water depends greatly on the type of water used. Research has included investigations of the stability of captopril in tap water.28–31
The usefulness of this work is severely limited by the geographical differences in the composition of tap water around the world. Some of the work is also limited by ad hoc sampling routines, small numbers of samples and a lack of detail.
Further research has been conducted on the stability of captopril in sterile or distilled water. Despite the more reproducible standards of the water vehicles, stability results still show a marked variation. A range of researchers have analysed the stability of captopril at a concentration of 1mg/ml in sterile water for irrigation or distilled water.30,32,33
The shelf-lives recommended varied from three to 14 days when stored in a refrigerator (4C or 5C). However, the 14-day recommendation by Nahata et al33 may be open to some doubt due to large standard deviations.
Other researchers have recommended much longer shelf-lives for similar formulations. Using various pH buffers, Schlatter et al investigated the stability of 1mg/ml solutions made from triturated tablets in sterile water. At pH 3 and pH 5, the solution was found to be stable for at least 28 days at 4C.34
Using pharmaceutical grade powder as the raw material, Escribano-Garcia et al found a 1mg/ml captopril solution to be stable for 30 days at 4C, when stored in polyvinyl chloride (PVC) containers and protected from light.34
Nahata et al also showed a significant beneficial effect for the addition of sodium ascorbate.33
Work in 2007 by Berger-Gryllaki et al suggests that captopril powder 1mg/ml in purified water may achieve a much longer shelf-life if disodium edetate 0.1 per cent (EDTA-Na) is added as a chelating agent and preservative.35
Berger-Gryllaki et al found their formulation to be chemically and microbiologically stable at 5C and room temperature for two years. Although this work contributed valuable data for specials manufacturers, most NHS dispensaries would not stock EDTA-Na.
Overall, the body of evidence for the stability of captopril in water appears to suggest that the quality of the water used is extremely important, particularly the metal ion content.
However, given the number of possible variables in the formulations, the evidence is difficult to interpret accurately. Indeed, many of the papers show inconsistencies and unexpected or conflicting results, particularly with regard to the presence or absence of excipients and concentration effects.
For example, Chan et al found degradation did not follow the hypothesis that degradation is inversely proportional to concentration.36 The authors suggest that the increased amount of tablet excipients at higher strengths may be partially responsible.
Given the plethora of issues, the use of water-based captopril solutions is not recommended. Pharmaceutically, there are many disadvantages to this approach, including problems with insoluble tablet excipients and an unacceptable taste.
Near-to-patient manipulation of captopril tablets (eg, dispersing a tablet in water at the bedside) may be an option for a limited number of patients. However, many patients need part-tablet doses, for which this approach is not acceptable.
Captopril in syrup Various researchers have studied the stability of captopril in syrup and other sugar solutions, on occasion mixed with other agents, such as methylcellulose.33,37,38
Although promising data are found in some of the papers, the formulations are often too complex to be considered for extemporaneous preparation in standard NHS dispensaries. Controversial findings are also seen. For example, Lye et al found that tablet-based formulations were more stable than those based on pharmaceutical grade powder.38
The use of syrup as a vehicle for captopril is not recommended. It has poor suspending capabilities with regard to insoluble tablet excipients, some of which may bind active drug. Therefore, dose uniformity is a serious concern.
Moreover, syrup formulations may be vulnerable to microbial contamination when diluted with other agents. Any dilution steps also carry the risk of introducing unwanted metal ions, thus catalysing drug degradation.
Captopril in a suspending agent The use of a suspending agent has some distinct advantages. If tablets are used as the raw material source — as is common practice in the NHS — an appropriate agent will suspend insoluble excipients as well as any drug bound to these particles.
Pre-prepared suspending agents can also be buffered to improve drug stability, and flavoured to mask the unpleasant taste of captopril. The thickening effect of suspending agents may also be useful for some patients.
Allen and Erickson investigated the stability of captopril 750µg/ml in three different vehicles, using tablets as the raw material:39
- A 1:1 mixture of Ora-Sweet and Ora-Plus
- A 1:1 mixture of Ora-Sweet Sugar Free and Ora-Plus
- Cherry syrup (cherry syrup concentrate diluted 1:4 with simple syrup)
It should be noted that the “Ora” preparations contain purified water. The initial pH measurements of the suspensions were in the range 4.1 to 4.2 and changed by less than 0.5 pH units during the study. Significant degradation occurred in all of the bottles over 60 days at 5C and 25C.
The degradation was most apparent in the cherry syrup formulation, which maintained stability for only two days. The Ora-Plus and Ora-Sweet formulation displayed the best stability, maintaining 93.2 per cent (± 1.1) drug concentration by day 14 when stored at 5C, and 90.3 per cent (± 0.6) by day 10 at 25C.
Given that the bottles were only sampled every few days, and that this would not have simulated true “in-use” conditions and normal exposure to the air, it would seem prudent to recommend a conservative shelf-life (eg, seven days expiry for captopril 750µg/ml in Ora-Plus 1:1 Ora-Sweet, or Ora-Sweet SF, stored at 5C).
The low concentration of the suspension may have contributed to the limited stability, given that captopril degradation is known to be inversely proportional to drug concentration.26
Berger-Gryllaki et al conducted a limited pilot into the stability of captopril 1mg/ml in Ora-Plus and Ora-Sweet (stored in 10ml brown flasks).35
Both tablets and powder were used as sources of raw material. The authors describe the stability of both formulations as limited, with the associated formation of a yellow colour within three weeks.
The tablet-based preparation showed 71.5 per cent and 81.9 per cent of drug remaining after one month’s storage at room temperature and 5C, respectively. The powder-based preparation showed 64.7 per cent and 85.8 per cent remaining, respectively. The sampling schedule and method is not supplied and it is not clear whether the solutions remained unopened after preparation until analysis.
Two unpublished studies were carried out by a regional NHS Quality Control Laboratory in 1992 and 1995 (McBride, personal communication, Royal Pharmaceutical Laboratory Service, Belfast City Hospital, June 2006), using Keltrol as a suspending agent.
Concentrations of 2.5mg and 5mg/ml were investigated at 4C and room temperature. The baseline pH of the 5mg/ml suspensions was 2.962, and this did not change significantly after two weeks’ storage at either temperature.
The authors of the reports recommended shelf-lives of seven days and 28 days for formulations stored at ambient and fridge temperatures, respectively, with reference to the 2.5mg/ml formulation. As expected, the stability of the higher strength formulation was slightly improved, allowing a 14-day shelf-life to be recommended at room temperature and 28 days in the fridge.
However, Keltrol is now a lapsed brand name and this xanthan gum-based suspending agent is now prepared to different specifications and pHs by a range of manufacturers. Typically, the formulation contains ethanol and chloroform. Ethanol is a central nervous system depressant and chloroform has been classified as a class 3 carcinogen (ie, substances that cause concern owing to possible carcinogenic effects but for which available information is not adequate to make satisfactory assessments).40
Given the variation in specification and the potential toxicity of the ingredients, Keltrol should only be used as a suspending agent after consultation with a suitably competent pharmacist. As well as being free from chloroform and ethanol, the Ora-based products have a reproducible formulation, and are supported by a significant evidence base for a wide range of drugs.
However, such products need to be imported from the US and are, therefore, relatively expensive. They also contain saccharin and sorbitol, which are associated with potential problems of their own. For example, saccharin was linked with the potential to cause an increased risk of bladder tumours in rats.
Epidemiological studies in humans have not, however, established a link.41,42 Sorbitol may act as an osmotic laxative.43,44
Other approaches Other researchers have investigated more complex formulation approaches, including buffer systems, antioxidants and co-solvent systems.33,37,45
The data from these studies may be useful for the development of “special” preparations of captopril, but the formulations are deemed too complex for application to extemporaneous practice in a typical NHS dispensary.
The evidence behind the formulation of captopril is conflicting and controversial. This reflects the myriad of factors that have been shown or suggested to have an effect on the stability of the drug, including:
- The effect of excipients in both the starting materials and the vehicle
- The raw material source (ie, pharmaceutical grade powder or crushed tablets)
- The effect of metal ions
- The effect of pH
- The effect of drug concentration
- The degree of exposure to oxygen during preparation or analysis
Given the complexity of the issues surrounding the use of captopril oral liquid, the risks associated with the extemporaneous formulation should be avoided where possible. The importation of a product that carries a licence in a foreign country should be considered — an import is known to be available from Australia.
Imported products that bear a licence in a mutually recognised foreign country are supported by safety and efficacy data as part of their licence application, and are subject to pharmacovigilance strategies in their country of origin.
For further information, pharmacists are referred to the document “Guidance for the purchase and supply of unlicensed medicinal products — notes for prescribers and pharmacists”, produced by the NHS Pharmaceutical Quality Assurance Committee.46
However, imports may not be suitable for use in neonatal patients due to their strength or formulation.
Other options include the use of an alternative ACE inhibitor whose stability in oral liquid formulations is less controversial, such as lisinopril oral liquid or ramipril oral liquid. Oral liquid formulations of lisinopril are available from specials manufacturers with evidence to support their choice of formulation and assigned shelf-life.
Where a formulation is purchased from a Specials manufacturer, the supplier should be able to provide suitable information to justify the assigned shelf-life. The use of an extemporaneous preparation of captopril oral liquid should only be considered where there is a demonstrable clinical need.
A simple formula and limited shelf-life should be allocated in order to minimise risk. Given the stability concerns surrounding the presence of metal ions, the use of plastic or glass equipment is recommended during preparation.
Where extemporaneous preparation is absolutely necessary, it is recommended that a simple formula and short shelf-life be used to minimise the significant risks (Panel 1 below).
Panel 1: Captopril oral liquid: extemporaneous formulation
Using a typical strength of 1mg/ml as an example:
|Captopril tablets 25mg||x 4|
|Ora-Sweet or Ora-Sweet SF||to 100ml|
Shelf-life: seven days refrigerated in amber glass. Shake the bottle before use
The formulation of any unlicensed products should be assessed by a suitably competent pharmacist before use, to ensure their quality and fitness for purpose.
In order to assess whether or not a formulation is fit for purpose, the pharmacist must be aware of the needs and vulnerabilities of the individual patient as well as considering the overall quality of the intended medicine.
ACKNOWLEDGEMENTS Regional Pharmaceutical Laboratory Service, Belfast City Hospital. This project was supported by Department of Health funding for Modernisation of Pharmacy Manufacturing and directed by the NHS Pharmaceutical Quality Assurance Committee, Working Group for Extemporaneous Dispensing.
Andrew Lowey, DPharm, MRPharmS, is quality assurance and research pharmacist and Mark Jackson, MPhil, MRPharmS is quality control and quality assurance manager, both at Leeds Teaching Hospital NHS Trust
1. Gill AM, Leach HJ, Hughes J, Barker C, Nunn AJ, Choonara I. Adverse drug reactions in a paediatric intensive care unit. Acta Paediatric 1995;84:438–41.
2. Turner S, Nunn AJ, Fielding K, Choonara I. Adverse drug reactions to unlicensed and off-label drugs on paediatric wards: prospective study. Acta Paediatric 1999;88:965–8.
3. Marshall IW, Daly MJ. Risk management in hospital pharmacy. Pharmacy Management 1996;12:44–47.
4. Baby dies after peppermint water prescription for colic. The Pharmaceutical Journal 1998;260:768.
5. Kaye T. The quandary of compounding for MCOs: administrative costs, risks, and waste. Managed Care 2003;(Jan):42–46.
6. Kirsch L. Extemporaneous quality. Journal of Pharmaceutical Science and Technology 2005;59:1–2.
7. Stewart PJ, Tucker IG. A survey of current extemporaneously prepared paediatric formulations. Australian Journalof Hospital Pharmacy 1982;12:64–68.
8. Crawford SY, Dombrowski SR. Extemporaneous compounding activities and the associated needs of pharmacists. American Journal of Hospital Pharmacy 1991;48:1205–10.
9. Woods DJ. Extemporaneous formulations — problems and solutions. Paediatric Perinatal Drug Therapy 1997;1:25–29.
10. Fisher CM, Corrigan OI, Henman MC. Quality of pharmaceutical services in community pharmacies. Journal of Social and Administrative Pharmacy 1991;8:175–6.
11. Davis NM. Do you really need glacial acetic acid on your shelf? Hospital Pharmacy 1997;32:611.
12. Standards for good professional practice: Council sets standards for extemporaneous dispensing. The Pharmaceutical Journal 2000;265:233.
13. Royal Pharmaceutical Society. Medicines, ethics and practice: a guide for pharmacists and pharmacy technicians. London: The Society; 2007.
14. Fenton-May V, editor. Guide to the preparation of non-sterile extemporaneous products in NHS hospitals. NHS Quality Assurance Committee; 2003.
15. Pappas A, McPherson R, Stewart K. Extemporaneous dispensing: whatever happened to it? A survey of Australian general practitioners. Journal of Pharmaceutical Research. 2002;32:310–14.
16. Chowdury T, Taylor KMG, Harding G. Teaching of extemporaneous preparation in UK schools of pharmacy. Pharmaceutical Education 2003;3:229–36.
17. Neill H. Do not be afraid to ask for help (letter). The Pharmaceutical Journal 2003;270:266.
18. Mulla H, Tofeig M, Bi’Lock B, Samani N, Pandya HC. Variations in captopril formulations used to treat children with heart failure: a survey in the United Kingdom. Archive of Disease in Childhood 2007;92:409–11.
19. British Pharmacopoeia 2007. London: The Stationery Office.
20. Sweetman SC, editors. Martindale: The Complete Drug Reference, 34th edition. London: The Pharmaceutical Press;2005
21. Kadin H. Captopril. In: Florey K, editor. Analytical profiles of drug substances. London: Academic Press; 1982. vol 11: pp79–137.
22. Rusho WJ. Clinical issues and concerns in the use of extemporaneously compounded medications. Journal of Pharmaceutical Care in Pain Symptom Control 1996;4:5–20.
23. Hanson G. Bespoke pharmacy: tailoring medicines to the needs of patients — the pharmacy production unit’s role. Hospital Pharmacy 2003;10:155–159.
24. Shipkowski ER. In: Kadin H. Captopril. In: Florey K, editor. Analytical profiles of drug substances. London: Academic Press; 1982. vol 11: pp79–137.
25. Chen D, Chen H, Ku H. Degradation of captopril in aqueous medium through buffer-catalysis oxidation. Drug Development in Industrial Pharmacy 1995:21:781–792.
26. Connors KA, Amidon GL, Stella ZJ. Chemical stability of pharmaceuticals: a handbook for pharmacists, 2nd ed. 1986.
27. Wang YJ. In: Kadin H. Captopril. In: Florey K, editor. Analytical profiles of drug substances. London: Academic Press;1982. vol 11: pp79–137.
28. Pereira CM, Tam YK. Stability of captopril in tap water. The Pharmaceutical Journal 1992;49:612–5.
29. Pramar Y, Das Gupta V, Bethea C. Stability of captopril in some aqueous systems. Journal of Clinical Pharmacology and Therapeutics 1992;17:185–9.
30. Anaizi NH, Swenson C. Instability of aqueous captopril solutions. American Journal of Hospital Pharmacy 1993;50:486–8.
31. Escribano-Garcia MJ, Torrado-Duran S, Torrado-Duran JJ. Stability study of an aqueous formulation of captopril at 1mg/ml. Farm Hospital 2005;29:30–36.
32. Andrews CD, Essex A. Captopril suspension (letter). The Pharmaceutical Journal 1986;237:734–5.
33. Nahata MC, Morosco R, Hipple TF. Stability of captopril in three liquid dosage forms. American Journal of Hospital Pharmacy 1994;51:95–96.
34. Schlatter J, Sola A, Saulnier JL. Stabilite d’une solution orale de captopril 1mg/ml. Journal de Pharmacie Clinique 1997;16(2):125–8.
35. Berger-Gryllaki M, Podilsky G, Widmer N, Gloor S, Testa B, Pannatier A. The development of a stable oral solution of captopril for paediatric patients. EJHP Science 2007;13:67–72.
36. Chan DS, Sato AK, Claybaugh JR. Degradation of captopril in solutions compounded from tablets and standard powder. American Journal of Hospital Pharmacy 1994;51:1205–7.
37. Liu J, Chan SY, Ho PC. Effects of sucrose, citric buffer and glucose oxidase on the stability of captopril in liquid formulations. Journal of Clinical Pharmacy and Therapeutics 1999;24:145–50.
38. Lye MYF, Yow KL, Lim LY, Chan SY, Chan E, Ho PC. Effects of ingredients on stability of captopril in extemporaneously prepared oral liquids. American Journal of Health-System Pharmacy 1997;54:2483–7.
39. Allen LV Jr, Erickson MA. Stability of baclofen, captopril, diltiazem hydrochloride, dipyridamole, and flecainide acetate in extemporaneously compounded oral liquids. American Journal of Health-System Pharmacy 1996;53:2179–84.
40. CHIP3 Regulations. Chemicals (Hazard Information and Packaging for Supply) Regulations, 2002.
41. Council on Scientific Affairs. Saccharin: review of safety issues. Journal of American Medical Association 1985;254:2622–2624.
42. Morgan RW, Wong O. A review of epidemiological studies on artificial sweeteners and bladder cancer. Food and Chemical Toxicology 1985;23:529–533.
43. Brown AM, Masson E. “Hidden” sorbitol in proprietary medicines — a cause for concern? The Pharmaceutical Journal 1990;245:211.
44. Greaves RRSH, Brown RL, Farthing MJG. An air stewardess with puzzling diarrhoea. Lancet 1996;348:1488.
45. Mathew M, Das Gupta V. The stability of captopril in aqueous systems. Drug Stability 1996;1:161–5.
46. NHS Pharmaceutical Quality Assurance Committee. Guidance for the purchase and supply of unlicensed medicinal products — notes for prescribers and pharmacists. 3rd edition. 2004.