Risk management and simulation

A pilot study of cytotoxic contamination in day care units in Germany has shown heavy contamination in some centres, according to Rudolf Schierl (Head of Analysis and Monitoring, Institute for Occupational, Social and Environmental Medicine, University Hospital, Munich; WHO Collaborating Centre for Occupational Medicine). Working practices varied considerably and some centres either did not know, or simply ignored, standard guidelines, he added. A total of 40 centres, comprising private practices and day hospitals, participated in the study. All centres responded to a questionnaire and wipe samples were collected from 28.

The results showed that 19% of treatment areas had a dedicated place for receiving prepared cytotoxic doses from the pharmacy. Only one clinic received the doses with primed administration sets attached; the remainder primed the administration sets on the ward. About two thirds of units flushed the infusion lines with saline after administration. Flushing with 75–100ml of saline is ideal because this dilutes any residual drug to negligible levels, said Dr Schierl.

Many centres had little idea about how the drug administration areas were cleaned or whether dedicated cleaning equipment was used. Using the same mop and bucket for the clinic and general areas can spread cytotoxic contamination. This is a topic that needs to be well understood and properly organised, commented Dr Schierl.

Wipe samples were analysed for 5-FU, platinum-containing drugs, cyclophosphamide, ifosfamide, methotrexate, gemcitabine, docetaxel and paclitaxel. Some busy clinics, where many doses were administered, had low levels of contamination whereas some less busy centres had high levels of contamination. In general, floors were more likely to be contaminated than work surfaces and this reflected more frequent cleaning of work surfaces, said Dr Schierl. Bathroom floors and patients’ toilets were frequently contaminated, he added. In some places gloves were worn for long periods and could, themselves, become a source of contamination.

Dr Schierl recommended frequent, careful cleaning to minimise contamination of floors. Separate cleaning equipment should be kept for patients’ toilets. He also advocated the use of closed systems and primed administration sets (from the pharmacy) to minimise contamination.

Pharmacists could have an important role in educating cancer patients about safety in the home, Paul Sessink (Managing Director, Exposure Control AB, Sweden) told the audience. Patients themselves can be sources of environmental cytotoxic contamination as they excrete cytotoxic drugs for some hours after treatment. This has implications for carers and relatives. Patients and relatives need sensible advice about how to avoid putting children and pregnant women at risk.

Previous studies have shown that patients’ toilets are the most heavily contaminated areas in hospital wards. This is likely to be due to urine splashes and poor cleaning – often cleaners are unaware of the importance of careful cleaning to remove cytotoxic agents and just spread contamination from one area to another, commented Dr Sessink.

A study performed in collaboration with the University of Fukushima, Japan involved three patients at home after receiving doses of cytotoxic drugs (cyclophosphamide or 5-FU). The results showed that in all cases, family members were exposed to the same drug (that is, they also excreted cyclophosphamide in their urine). Wipe samples of surfaces in the home showed that levels of contamination with cytotoxic drugs were comparable to those seen in the areas in wards where cytotoxic drugs are administered. He noted that cyclophosphamide is absorbed from the skin in about ten minutes.

Dr Sessink recommended that patients who have received a dose of cytotoxic drug in hospital should use a separate toilet at home if possible (for the period of drug excretion). In addition, both male and female patients should urinate in the seated position, the toilet lid should be closed for flushing and the toilet should be flushed several times after each use. Furthermore, both detergent and bleach should be used to clean the toilet and thick gloves should be worn. Clear guidelines for patients and families are needed; the guidelines already in use for cytotoxic administration areas in hospitals could be a starting point, said Dr Sessink.

Small-scale preparation of medicines in pharmacies can involve many different active substances and differing exposure levels. Operations can range from breaking tablets for arthritic patients to processing freeze-dried antineoplastic agents in capsules for children, explained Yvonne Bouwmans-Brouwer (Dutch Pharmacists’ Association, The Hague, The Netherlands). A risk matrix has been developed to help to determine priorities in protecting staff from occupational exposure to cytotoxic agents. Risk depends on the hazard presented by a product and the likelihood of exposure. Hazards are determined by human toxicity data and reports of effects of occupational exposure. Few studies are concerned with exposure.

A study conducted in The Netherlands had examined operator exposure to powders with varying levels of ventilation. Exposure was assessed by measuring the level of airborne powder – in this case riboflavin – in the breathing zone. The degree of exposure was dependent on the quantity of material manipulated and the duration of the activity. The results were used to construct an inhalation exposure model that predicts the likely level of exposure depending on the physical form of the product and the type of ventilation in use. For example, processing a large amount of powdered drug with no ventilation could result in high levels of exposure, whereas processing a similar quantity of liquid drug in a safety cabinet would present a low risk of exposure. Once the degree of exposure is known, this can be entered into a risk matrix together with toxicity data to determine the level of protection required for safe handling. The model works for quantities of active substance between 100mg and 100g, explained Dr Bouwmans-Brouwer. For larger or smaller quantities, the Advanced REACH tool, which is available in English, French, German and Dutch, can be used, she added (see Resources).

Disinfectants are effective if applied to clean surfaces, in the correct concentration for the correct length of time, explained Tim Sandle (Head of Microbiology, Bio Products Laboratory Ltd, UK) and it is important to understand the different roles and properties of detergents and disinfectants when managing a clean room.

Detergents are surface active agents that penetrate soil residues – including grease, dirt, protein, skin and oils – and allow them to be removed easily. They make surfaces appear clean and can dilute or rinse away micro-organisms but not kill them. This is important because a ‘clean’ surface is easier to disinfect than a visibly soiled surface. The ideal detergent works with both hard and soft water, does not inactivate the disinfectant or damage surfaces, is effective against a range of soils and is non-foaming. Neutral detergents are ideal and they may need to be sterile, added Dr Sandle.

Disinfection is the inactivation or destruction of micro-organisms, thereby reducing numbers of microbes. It is not the same as sterilisation, he emphasised. Many disinfectants kill only vegetative bacteria and specific, sporicidal products are required to kill bacterial spores. There is a wide range disinfectants – some are bacteriostatic and some are bactericidal. One way of classifying disinfectants is to separate them into oxidising disinfectants (for example, hydrogen peroxide, peracetic acid) and non-oxidising disinfectants (for example, alcohols, aldehydes, quaternary ammonium compounds). Oxidising disinfectants have a non-specific mode of action, wide spectrum of activity and are mostly sporicidal.

A number of factors can affect the efficacy of a disinfectant including the number, type and location of organisms, the concentration of the disinfectant and the contact time. Other important factors are temperature, pH, the presence of soil and the type of water used to dilute the disinfectant.

When selecting disinfectants, a checklist of the desired properties should be used to guide the process, suggested Dr Sandle. All disinfectants used in GMP facilities must be validated to demonstrate their efficacy and it is wise to select two products – ideally with different modes of action – to be used in rotation. This is not supported by scientific studies but is commonly required by EU Medicines Inspectors, he explained.

Media-fill tests, in which microbiological growth media are used in place of drugs, are well-established simulation processes for validating both aseptic procedures and the performance of individual operators. It is rare to find positive results in hospital aseptic processing and this could be partly due to the relatively small numbers of products that are made, Sylvie Crauste-Manciet (Assistant Professor, University Hospital of Bordeaux, France) told the audience. Hospital aseptic units prepare relatively small numbers of injections and it would require at least 3000 preparations to have a 95% chance of detecting a contamination rate of 1 in 1000, she continued.

Although media-fill tests in ISO 5 sterile compounding isolators are usually negative, visual observation shows that some operators adopt inappropriate aseptic techniques. For this reason, a ‘challenged media-fill test’ has been developed (see Resources). This involved artificially contaminating the septa of vials with Streptococcus faecalis (2000 colony-forming units (CFU)/cm2) before the simulated transfers were performed. In a study, ten experienced operators, all of whom were validated for conventional media-fill tests, each made ten preparations involving 29 transfers. In total, four of the operators failed one test and one failed two tests. Visual observation only identified one operator with very poor technique, said Dr Crauste-Manciet. The overall contamination rate was 2.3%. The causes of contamination included inadequate alcohol decontamination of septa and touching of contaminated surfaces. The challenged media-fill test was incorporated into the routine assessment of staff in the aseptic unit as a means of improving education and adherence to good aseptic technique.

Dr Crauste-Manciet concluded that it is important to understand the sensitivity of the test that is being used and to beware of creating a false sense of security.

Disposable syringes were developed for immediate use but increasingly they are being used as final containers for ready-to-use injections. Integrity testing – to demonstrate that the closure system protects the product against ingress of micro-organisms during storage – is essential in this situation. Passive methods for testing the integrity of syringes always produce positive results but more stringent test procedures have now been developed to reflect exceptional situations, Alison Beaney (Regional Quality Assurance Specialist, North East England) explained.

Passive testing involves filling syringes with a nutrient broth and leaving them in a box or the back of a van for the test period. A positive result would be indicated by turbidity of the broth. A more stringent test involves immersion of the broth-filled syringes in a suspension of micro-organisms. Escherichia coli  (E coli) is often used for this purpose because it is a highly motile organism that can penetrate small gaps. The main disadvantage is that it is very smelly. Brevundimonas diminuta is also used because it is the smallest test organism available.

The broth-filled syringes are immersed in the suspension of bacteria and incubated at 32°C for 14 days. In the case of E coli, the syringe is immersed for 15 minutes then rinsed and incubated at 32°C for 14 days. If the syringes pass the test (that is, show no growth), then the growth medium has to be validated. In order to do this, two of the test syringes are inoculated with 10–100 CFU of the test organism and incubated for three days at 32°C. The test is satisfactory of growth is seen in both syringes. Testing in this way has shown that many syringes are robust containers but filling large syringes to their full capacity can lead to leaks at the plunger end because sideways pressure on the plunger can compromise the seal, said Dr Beaney. When used as a storage container, a syringe should not be filled to more than 85% of its capacity, she added.

A dynamic, dye-intrusion test has also been developed to test the performance of the syringe hubs and plungers. This involves filling the syringe and inserting a pin to hold the plunger in the fully extended position. The syringe is then immersed in a bath of dye such as methylene blue and agitated for two hours. Leakage is clearly evident if the fluid inside the syringe changes colour.

Palatability is an issue that goes far beyond the question of taste alone – it can also involve the appearance, feel and even sound (when taken) of a product, according to Catherine Tuleu (Reader and Director of the Centre of Paediatric Pharmacy Research, School of Pharmacy, University College, London). It is important not only for oral products but also for ophthalmic and nasal formulations, which may all be tasted after administration. Poor palatability is a consistent feature in non-adherence with paediatric medicines. “Children do not think that if it tastes bad it must be doing you good. If it tastes bad they just refuse subsequent doses,” explained Dr Tuleu. One child given a bitter-tasting, pink prednisolone syrup subsequently refused all pink or red syrups. Similarly, a child given a yellow methotrexate tablet (for juvenile arthritis) experienced severe nausea and subsequently refused to take any yellow tablets.

Adult human panels, using the ‘swirl and spit’ method, assess the acceptability of medicines at Phase II of product development at present. Ideally, acceptability aspects should be evaluated during the clinical studies with patients from the target age groups, said Dr Tuleu. The taste of active pharmaceutical ingredients (APIs) and formulations need to be assessed at the early stages of product development so that unpleasant tasting products are identified early and taste-masking approaches can be developed. Another problem is that generic products, that are bioequivalent to the originator products, may have very different taste characteristics. For example, in one study eight generic versions of co-amoxiclav syrup had been tested and none tasted as pleasant as the original. This is a factor that is rarely taken into account by purchasing bodies, she added.

____________________________________________________________

• Augmented reality glasses (‘Google glasses’) allowed staff to prepare injections more quickly and more accurately. Preparation instructions were displayed on the glasses leaving the operator free to use both hands for the task. (Ecole Centrale, Lille and University of Lille, France).
• A self-cleaning cytotoxic drug compounding robot (Kiro) has been developed. Validation tests at Onkologikoa Hospital (San Sebastian, Spain) showed that the one-hour self-cleaning process was able to reduce levels of paclitaxel, 5-FU and cyclophosphamide by more than 90%.
• The installation of a temporary, positive-pressure clean room (Steriroom, EuroBioConcept) proved to be an economical and timely solution to the problem of insufficient aseptic preparation capacity. (Lens Hospital, France).
• A formal scheme for post-chemotherapy self-monitoring by cancer patients receiving FOLFOX or FOLFIRI regimens ensured that critical information was not missed and allowed patients to be actively involved in their own treatment (Angoulême Hospital, France).
• Batch production of cyclophosphamide injections using the Pharmashaker (Medical Dispensing Systems) and the Baxa Repeater pump (Baxter) for reconstitution and filling, respectively, resulted in better quality products, more timely delivery of doses and reduced risk of repetitive strain injury for pharmacy technicians (Hospital Saint Louis, Paris, France).
• The introduction of a secure cytotoxic administration set (CareFusion) reduced the risk of exposure of staff to cytotoxic droplets and the risk of under-dosing as a result of not flushing the line after administration. Injections were easier to prepare using the new administration set and the estimated additional cost was €43,000 for 30,000 infusion bags. (Hotel Dieu Hospital, Nantes, France).
• A pilot study to assess the feasibility of reducing occupational exposure to doxorubicin during chemo-embolisation by the use of Kemoline (Doran International) showed that leakage (shown by contamination of drapes) was reduced from 67% to 17% (University Hospital, Lille, France).
• Evaluation of a new formulation of mitomycin-C eyedrops showed that it was more stable than the unlicensed product that is currently used with no loss of cytotoxicity. The findings could reduce manufacturing costs and improve the treatment of several ocular pathologies (Stobhill Hospital, Glasgow, UK).

____________________________________________________________