What’s in that syringe? Standard drug concentrations in critical care

In 2009, the Intensive Care Society (ICS) published guidelines recommending standard drug concentrations for infusions in critical care. ¹ In this issue of JICS, Titiesari et al. ² publish the results of a survey reviewing the adoption of these concentrations across critical care units in the UK. Although the study has limitations found in many such surveys, ³ the results suggest that there has been an increase in the adoption of the ICS standard concentrations but that they have not yet been adopted in all units. This poses two questions: Why would our society recommend standard drug concentrations? Why would some critical care units not implement these standards?

In writing this editorial, I am using a standard keyboard on a computer that uses a standard operating system in a room lit with light bulbs using standard fittings; it’s clear that our modern world depends on standardisation. It has been recognised since the industrial revolution that manufacturing is more efficient if processes are standardised and there has been a drive to increase standardisation in the process of health care for many years. ⁴

Sometimes one way of doing something is better than another and standardising to adopt the better way will have clear advantages. More importantly, standardisation can increase efficiency and safety just because having one way of doing something is easier than using lots of ways of doing it, particularly where things have to be communicated, shared or connected. This may be true even if the chosen method may not otherwise have been the best way of doing something. It would be difficult to argue that the drug concentrations recommended by the ICS are always the best. For example, 5 mg of noradrenaline would seem more logical than 4 mg in 50 ml. The standards are, however, a pragmatic set of concentrations that were based largely on existing practice ⁵ that had evolved over many years, often for long forgotten reasons. The benefits for adopting the ICS standard concentrations are those of standardisation itself rather than because the concentrations themselves are the best possible option.

Staff work in different units; medical staff will have differing memories of their first nights on call in several units in their training. The core standards for critical care ⁶ allow for agency staffing rates of up to 20% and nurses also move between permanent posts. There would have to be very good reasons for confusing these staff at the start of their posts by asking them to become familiar with different concentrations of drugs both in terms of different response profiles and differences in preparation. Standardisation is useful where interconnections are important, and interfaces of care are key opportunities for errors. ⁷ Standardised concentrations should make transfers from theatre and between units safer; unidentified syringe contents are a recurrent theme in patient safety incidents associated with handovers of care.^7,8 Standardisation is also more helpful where staff are relatively inexperienced and during emergencies, ⁹ both common situations in critical care.

The preparation of infusions on critical care units has many problems. The concentrations of locally prepared drugs are often wrong due to preparation errors ¹⁰ and there are risks of potentially catastrophic cross infections. ¹¹ Removing two nurses from direct care for their patients during preparation will also add risks to patient care. Having pre-filled syringes will remove these problems and give additional advantages in terms of consistent drug labelling and bar codes that can be linked to infusion pumps. The adoption of single drug concentrations is already making the manufacture of pre filled syringes a practical reality.

Why then have the ICS standard concentrations not been universally adopted? There is an appropriate organisational inertia that will slow change where that change may not have an immediate benefit. There are also real and perceived disadvantages to standardisation in a random chaotic world. Critical care is a highly complex system, ¹² so that change in one area of care may have unintended consequences in another area. Standardising patient turns to prevent pressure sores will work ¹³ but may disrupt sleep patterns, reducing alarm volumes at night may encourage sleep but could occasionally cause other problems to be missed. Staff rightly perceive a history of external agencies trying to impose standardisation on these complex systems with little resultant improvement in patient care. The standard drug concentrations make some recognition of the complexity of patient response by having three standard concentrations of noradrenaline. This reflects the way that the standard concentrations have evolved as a bottom up process from the experience of many units working in this complex environment. A complex system will also be a changing system and an appropriate standard one year may become outdated in following years. This is clearly true for standard drug concentrations; the ICS standards describe a standard concentration for Vecuronium but do not, for example, mention Remifentanil. The standards should probably be updated to take into account changes in practice, particularly where the usage of some drugs would start to make the production of prefilled syringes economical.

One advantage of identifying variation in outcomes of health care is that variation suggests opportunities for improvement. A review of patient safety incidents associated with medications showed a 10-fold variation in rates between units with over two-thirds of the incidents being judged to be clearly preventable. ⁸ This strongly suggests that we could do more to improve medication safety in critical care. Increasing standardisation in medication practice could improve safety in a number of ways as summarised in Table 1.

Table 1.

Areas of medication practice that could be improved with standardisation and those where other approaches may be helpful.

Areas where standardisation could improve medication safety:

Drug supply and procurement – Using agreements between units to allow procurement of prefilled syringes Prescribing – Moving to standardised order sets, either by locally designed drug charts or sticky labels to put on charts or in  electronic systems to have defined order sets, standardising the provision of pharmacy support to educate prescribers and audit  the process. Communication – Standardising handovers to ensure that medication reconciliation is accurate 14 at transfer into and out of critical  care and during shift handovers. Ensuring that specific issues are reviewed on ward rounds by using check lists. 15 Preparation and administration- Using defined concentrations and ensuring the second checker process is standardised to remove  ambiguity or abdication of responsibility. Providing emergency drug boxes for patients requiring urgent endotracheal intubation,  which have the drugs and labels appropriately presented. Staffing – Setting standards for pharmacy staffing. Examples where standardisation would be unhelpful: Educational interventions should have multifaceted approaches including feedback to prescribers. 14 This reflects variations in  learning styles. Provision of drug information should be multifaceted, again due to different preferences and different information needs by different  staff at different times.

How can the standards set out in Table 1 be achieved? One of the key advantages of standards is that they can be set to improve performance. The recent ‘standards for critical care units’ set ambitious standards for the provision of pharmacy staffing for critical care units. ⁶ These are clearly not yet achieved, ¹⁶ despite clear evidence that pharmacists are self-funding in terms of restricting inappropriate prescribing. ¹⁷ Inspection and benchmarking of critical care units against our new standards, after only the last two years, are clearly already showing improvements in patient care. Many units are now improving their provision of pharmacy support to meet these standards where they would not previously have done so. Where new pharmacists are appointed this should be audited, focusing both on cost savings and patient safety. Audit of medication safety should also include a review of patient safety incidents that uses a logical structure that can allow comparisons across units and over time. This process should be organised within critical care networks using freely available methods. ¹⁸ In many networks, harm due to problems with medication safety stays hidden in un-reviewed patient safety incidents so that this harm cannot be a drive to improvement.

There are additional steps to improving medication safety that are also required. Establishing multidisciplinary groups across units within critical care networks allows sharing of best practice, educational resources and purchasing power, particularly for prefilled syringes. These groups should also form the bases of locally driven and workable local standards. To work well, the groups should have a clear understanding of quality improvement, including small step changes, repeated audit and improvement cycles and staff engagement.

Other areas of critical care would also benefit from standardisation, particularly where novel or unexpected events are unusual, solutions are well understood, subject to analytical reasoning and where interconnections are important. ⁹ Where standardisation identifies ‘best practice’ the process often uses guidelines and protocols. Producing guidelines is increasingly standardised, in part by the development of the GRADE system. ¹⁹ The processes by which these guidelines and protocols actually influence practice, for example by presentation of information to clinical staff, the use of standard order sets and other forcing functions is less clear. Other areas where standardisation may be helpful in critical care include handovers at transitions of care and the management of patients with acute lung injury.

Even with standards, making the best responses to the complex situations encountered in critical care requires judgement. The basic psychological process by which good and bad judgements are exercised ²⁰ should be taught to and understood by the clinical staff who are required to make complex decisions. The use of standard drug concentrations provides us with another standardised building block to build the correct, unstandardised, judgement responses to random situations. The answer to ‘what’s in the syringe’ should not be ‘I’m not sure’!