Wash & go – but with what? Skin antiseptic solutions for central neuraxial block


Infectious complications associated with central neuraxial block (CNB) are rare, but potentially devastating for the unfortunate patient who suffers them. The Royal College of Anaethetists’ Third National Audit Project (NAP3), covering over 700 000 CNBs in the UK, identified 15 epidural abscesses and three cases of bacterial meningitis, giving an approximate risk of 1 in 47 000 and 1 in 233 000, respectively. This is reassuringly low, but inadequate aseptic technique was identified as a factor in several cases [1].

Surgical aseptic technique became enshrined in medicine after the pioneering work by Joseph Lister using carbolic acid solution at Glasgow Royal Infirmary in the 1860s, and healthcare workers are taught how to practise it correctly at an early stage in their training.

The issue of which antiseptic cleaning solution to use to clean the skin before performing CNB has become contentious following two cases of permanent neurological injury after CNB in obstetric patients, one in the UK and one in Australia, in which chlorhexidine gluconate was alleged to have been responsible. In the UK case, the judge concluded that the spinal local anaesthetic injectate was accidentally contaminated with chlorhexidine skin cleanser, although there was no direct evidence that this had happened [2]. The patient developed chronic adhesive arachnoiditis and paraplegia. In the Australian case, 8 ml of colourless chlorhexidine was mistakenly drawn up into a syringe from a gallipot on the anaesthetist’s tray and injected into the epidural space. This patient was also left paraplegic [3]. Following this case, the Western Australian Department of Health issued an updated operational directive on the labelling of injectable medicines [4] and the Queensland Health Department published a guideline on surgical skin antisepsis that recommended that solutions containing > 0.5% chlor-hexidine should not be used before performing an epidural because of the potential risk of neurotoxicity [5]. However, this guideline failed to quote evidence to support this recommendation.

The choice for skin disinfection lies between chlorhexidine (with and without alcohol) and povidone-iodine (e.g Betadine®, Ayrton Saunders Ltd, Liverpool, UK). Chlorhexidine is a potent, broad-spectrum antisepetic that is effective against most gram-positive and negative bacteria as well as yeasts. Adding 70% alcohol to chlorhexidine accelerates onset of action and efficacy. The choice is between 0.5% chlorhexidine with ethanol (e.g. Hydrex® solution, Ecolab Ltd, Leeds, UK) or 2% chlorhexidine in isopropyl alcohol (e.g. ChloraPrep®, CareFusion UK Ltd, Reigate, UK). The datasheet for ChloraPrep specifically states it should not be used for lumbar puncture or come into direct contact with the meninges [6]. As 0.5% chlorhexidine is as effective as 2%, the lower concentration is recommended for chlorhexidine [7].

Chlorhexidine has faster onset and longer duration of action than povidone-iodine. It also has a lower incidence of skin reactions than povidone-iodine solutions. Chlorhexidine also penetrates and adheres to the stratum corneum skin layer more effectively than povidone-iodine and retains its efficacy in the presence of blood contaminating the surgical field [8]. There are few prospective comparative studies of chlorhexidine and povidone-iodine pre-surgical skin disinfection, but Berry et al.’s randomised controlled trial of 1982, that included 371 patients, showed that 0.5% chlorhexidine in spirit (Hibitane®, Regent Medical Overseas Ltd, Manchester, UK) was significantly more effective at reducing surgical site infection than 10% povidone-iodine in alcohol [9]. A systematic review of nine randomised controlled trials, including 3614 patients, found that chlorhexidine resulted in significantly fewer surgical site infections than povidone-iodine (adjusted risk ratio 0.65 (95% CI 0.51–0.80)) [10]. As a result of these advantages, the American Society of Regional Anesthesia, the American Society of Anesthesiologists and the Royal College of Anaesthetists (in the NAP3 report) have all recommended chlorhexidine in alcohol as the skin disinfectant of choice for CNB [1, 8, 11].

However, the British National Formulary recommends avoiding direct contact with the meninges (and middle ear) [12]. Literature review reveals that there are no data on the risk of neurotoxicity or arachnoiditis with chlorhexidine in humans. There are two historical laboratory studies in animals that show neurotoxic effects of concentrated chlorhexidine when applied directly to neural tissue or the meninges. Weston-Hurst injected chlorhexidine directly into the cerebrospinal fluid of monkeys [13] and Henschen and Olson did the same into the anterior chamber of rats’ eyes [14]. There is also evidence of chlorhexidine-induced middle ear damage in guinea pigs [15].

A large retrospective review of 12 000 CNBs over five years, from an institution in which 2% chlorhexidine in 70% isopropyl alcohol is exclusively used for skin asepsis, reported a very low incidence of neurological complications [16]. A total of 57 cases of neurological injury were found in this large series, and cases were independently assessed by three anaesthesiologists (excluding the first author). Spinal anaesthesia was thought to have been potentially implicated in only five cases (0.04%), all of which completely resolved within 30 days. The clinical picture in these five cases was suggestive of direct mechanical (needle) trauma, rather than chemical arachnoiditis, aseptic meningitis or diffuse lumbar plexopathy, as the mechanism of injury. Accepting the limitations of this retrospective case series, there did not appear to be any significant additional risk of neurotoxicity associated with chlorhexidine skin disinfection [16].

Another almost evidence-free zone is whether more than one application of chlorhexidine is necessary to decolonize the skin before CNB. Malhotra et al.’s elegant study in healthy volunteers demonstrated that a single spray application of 0.5% chlorhexidine with ethanol completely sterilised the skin over the lumbar spine, and the authors concluded that repeated application was unnecessary [17]. However, it is important to wait for it to dry before proceeding, to allow time for it to exert its antiseptic effect and also to reduce the risk of contaminating gloves or equipment.

So, where does that leave us, especially in today’s medicolegal climate [2, 18]? Chlorhexidine is a more effective skin disinfectant than povidone-iodine. There is no published evidence to suggest that using chlorhexidine before CNB increases the risk of neural injury. In fact, there is evidence from the large series of Sviggum et al. that 2% chlorhexidine does not result in a higher than expected rate of neurological complications [16]. However, there is some evidence from animal studies that direct application of chlorhexidine to neural tissue or meninges can result in injury, so great care is necessary to avoid accidental splashing of spinal needles, syringes or catheters with cleaning solution when performing CNB. We all know how easy it is to knock over the gallipot containing chlorhexidine accidentally on the spinal or epidural tray, spilling the contents on to the rest of the equipment. If this happens, it is appropriate to ask for fresh equipment. It may be wise to apply the chlorhexidine separately (perhaps by spray as in Malhotra et al.’s study) and then remove it from the vicinity of the CNB equipment, to reduce the risk of contamination. It is also prudent to avoid using colourless skin disinfectant, to reduce the chance of mistaking it for saline or local anaesthetic solution, as happened in the Australian case.

Finally, a plea for attention to detail with aseptic technique for CNB, echoing the wise words of Tony Wildsmith [19]. ALL components of the aseptic technique are probably important to reduce the risk of infective complications of CNB – handwashing, hat, gown, gloves, mask and careful skin disinfection, although the evidence for some of these is a little weak [8]. It was reassuring that a UK survey of obstetric anaesthetists [20] reported high compliance rates for all components of aseptic practice when epidural catheters were inserted (Table 1). However, these figures probably don’t reflect national practice for spinal anaesthesia. A recent paper by Kinsella et al. on ‘rapid sequence’ spinal anaesthesia for category-1 caesarean section [21] generated correspondence expressing concern that aseptic precautions may be compromised, in particular not waiting for the chlorhexidene/alcohol to dry before proceeding [22].

Table 1.   Routine aseptic precautions when inserting an epidural catheter in obstetric patients. Values are number (proportion) of survey respondents. From Mckenzie and Darragh [20].
Cap143 (87%)
Gown163 (99%)
Mask149 (91%)
Gloves164 (100%)
Assistant wears cap148 (90%)
Sterile drape163 (99%)

The NAP3 audit noted that aseptic technique had been poor in a number of the cases where patients came to harm. Perhaps too many shortcuts have crept in. An outbreak of infective complications after regional anaesthetic proceures in New York in 2008 was linked to poor aseptic technique in one individual and led to the closing down of the clinic by the Commissioner of Health [23]. Sloshing skin disinfectant around is no saviour for sloppy aseptic technique.

Competing interests

No external funding and no competing interests declared.