This article is accompanied by the following Invited Commentary:
Columb M. When comparing apples and oranges is all bananas! Eur J Anaesthesiol 2013; 30:397.
Lumbar epidural blockade is a commonly performed procedure in anaesthesia. Junior anaesthetists tend to gain their first experience of the technique of epidural insertion in the obstetric population wherein almost one in four labouring women requests an epidural.1 In a survey by Wantman et al.,2 99% of obstetric and non-obstetric lumbar epidurals within the UK were inserted with Dogliotti's technique3 of loss of resistance either intermittently or continuously to 0.9% sodium chloride or air. This technique depends on ‘feel’, which is difficult to teach to novices without their hands on the needle, so for most, the feel of the loss of resistance is appreciated for the first time on a patient.
Epidural anaesthesia is not without its complications, the incidence of which is higher among inexperienced trainees.4 Any method of teaching trainees some idea of the feel of loss of resistance before they are exposed to patients is therefore valuable. Various epidural simulators have been developed over the years.5,6 Unfortunately, these have limited availability and are expensive. In addition, all the simulators may not be ideal for the practice of all the available techniques. In the era of three-dimensional computer simulation and other high technology teaching methods, we investigated the use of an innovative, inexpensive and low-technology epidural simulator. There has been anecdotal evidence of fruit being used for the purpose of epidural simulation.7,8 We present a blinded study carried out to establish the best fruit for simulation of the feel of loss of resistance.
This study was discussed informally with a member of our local ethics committee but because no patients or patient information was involved, formal ethical approval was not required. All participation was voluntary.
A frame (hereafter referred to as the ‘epidural simulator’) measuring 55 × 25 × 14 cm, with a base 35 cm wide, was constructed from chipboard. We stacked four equally sized cardboard boxes, each with a circular hole in the outward face, into the frame. The front of the epidural simulator had a narrow strip of chipboard attached by a hinge to the top of the frame with four holes reflecting the holes in the boxes. Each of the holes was covered by two layers of opaque tape, which acted as a false skin and hid the identity of the fruit. The hinge at the top of the epidural simulator enabled the narrow strip to be released from the clasp at the bottom of the frame for the purposes of changing the tape and ensuring that the fruits remained flush against the tape (Fig. 1).
The back of the epidural simulator consisted of a removable piece of wood attached by clasps. This piece of wood was fitted after all the boxes were closed with the respective covers. The fruits were secured firmly in place within the boxes by foam. The frame had a broad base, which was clamped on to a sturdy surface. This represented a patient sitting up in bed.
This was a single blind study because the fruits had to be rotated and changed at regular intervals by the authors. The fruits were chosen for their general availability, low cost, differing textures and lack of a central hard core. The four fruits were kiwi, banana, honeydew melon and orange. We randomly allocated the fruits a letter A to D, which corresponded with the holes of the epidural simulator. We tried to ensure that we used fruits of the same variety, country of origin and approximately the same degree of ripeness. The fruits were all bought from the same supermarket. They were rotated after every two to three participants and the smaller and juicier fruits (being the kiwi fruit and the orange) were changed every four to five attempts. The banana was rotated and changed every six attempts. The honeydew melon was simply rotated and repositioned to expose untraumatised parts of the fruit.
Each participant completed a form detailing his or her grade (consultant, specialist registrar or senior house officer), years of experience (0 to 4, 5 to 9 or more than 10), needle gauge used (16G or 18G) and the participant's choice of 0.9% sodium chloride or air. The form also contained four 100 mm Visual Analogue Scales (VAS) on which 0 mm represented a completely unrealistic ‘feel’ and 100 mm represented a ‘feel’ which was indistinguishable from a real patient. We recruited anaesthetists of all grades with established epidural experience from two hospitals in Glasgow. Prior to their participation, we informed each participant that it was a lumbar epidural simulator so as to ensure similar angles of needle insertion. We also reiterated that simply the ‘feel’ of loss of resistance was being scored and not the feel of the different layers of skin and ligaments. All participants were free to opt out of participating in this study at any time.
Statistical analysis was carried out using Minitab version 15 (Minitab Inc., State College, PA) with a significance level of 5%. The scores for each fruit were compared using a repeated measures general linear model, which consisted of the grade of assessor (consultant, specialist registrar or senior house officer), years of experience (0 to 4, 5 to 9 or more than 9), gauge size (16 gauge, 18 gauge) and technique (0.9% sodium chloride air). Comparisons between fruits were made using the Bonferroni correction factor.
Fifty anaesthetists (29 consultants, 12 specialist registrars and nine senior house officers) completed the study. The anaesthetists were free to choose their normal epidural technique and their preferred method of loss of resistance. Thirty-four anaesthetists chose a 16 gauge (Sims-Portex, Kent, UK) Tuohy needle. Of these, 27 (79%) used 0.9% sodium chloride in the loss of resistance syringe; the remaining seven used air. Sixteen anaesthetists used an 18 gauge needle. Fourteen (88%) of these used 0.9% sodium chloride and two used air. The data from the study are shown in Table 1. The mean VAS scores and 95% confidence intervals for the fruits are shown in Fig. 2.
There were no significant differences between fruits chosen by grade of anaesthetist (P = 0.505), experience (P = 0.640), needle gauge (P = 0.794) or technique (P = 0.751). There was a significant difference between the realism of the fruits (P < 0.001), with the banana having the highest ratings. The results of the paired comparisons are shown in Table 2. There were no significant differences between the honeydew melon and banana, the kiwi and honeydew melon or the orange and kiwi.
The banana emerged as the most realistic fruit for simulating the feel of loss of resistance, and is a very cheap and easy makeshift simulator for every anaesthetic department. We now encourage our junior trainees to practise on bananas to ascertain the feel of ‘loss of resistance’ prior to their first epidural insertion in a patient. The honeydew melon came a close second with the kiwi and the orange placed third and fourth, respectively. There was consensus among all the grades for the first and second places of the banana and the honeydew melon. Consultants, however, considered that the orange should be placed third, in contrast to the trainees, who gave the kiwi fruit third place.
There are some problems inherent in this study. The most important issue was the inability to ensure the constant state of ripeness and thus consistency of the fruits. To alleviate this problem, we bought all the fruits from the same supermarket and made sure of the consistency of their variety and country of origin. Only one author was designated purchaser of the fruits on all occasions to ensure some uniformity and the fruits used in the study were always freshly purchased. All these steps, however, may have not been adequate to ensure the consistency in texture, which was a crucial factor in the comparison process.
Second, 82% of participants used the loss of resistance to 0.9% sodium chloride as their technique of choice. The infusion of 0.9% sodium chloride into the fruit posed two additional problems: fluid shift within the fruit due to osmotic forces and the waterlogging that changed the ‘feel’ of the fruits, leaving them ‘boggy’. The honeydew melon posed a different problem because the 0.9% sodium chloride that was infused pooled in the central core, thereby altering the feel and risking a simulated dural tap! The banana had the least problems with waterlogging. We circumvented this problem by changing the fruits frequently.
Third, with regard to the blinding of the whole process, some of the fruits had very distinct scents and many participants did manage to identify the fruit, which may have led to some bias in the results.
Fourth, we used the VAS because such a scale is mainly used to measure a characteristic or opinion across a continuum wherein a precise value is not directly measurable. This has been most widely validated in the assessment of pain, but has been used previously for measuring participants’ perception of realism in a simulation setting.9,10
There are only a few reports in the literature relating to the realism of epidural simulator devices, despite the presence of a number of these devices on the market. Anderson et al. tested a force-feedback simulator for realism using a 7-point scale (in which 1 = not realistic and 7 = extremely realistic).6 The average score for the feel of the procedure in their study was 4.8 which, when converted to a percentage, equated to 68.6%. This score is not much higher than the mean VAS score of 62 mm obtained for the banana in our study. It is regrettable that such simulators are released on to the market without any objective measurement of their realism. We suggest that the VAS may form an easily reproducible way of assessing the realism of many simulator devices in future studies.
We have shown that a banana can be a cheap and easily acquired simulator for the purpose of teaching the feel of loss of resistance for epidural insertion. This may improve safety profiles among junior anaesthetists.
Assistance with the study: the authors would like to thank Mr Samuel Frickleton Small for his invaluable contribution in the manufacture of the epidural simulator and all the participating anaesthetists from the Victoria Infirmary and the Southern General Hospital, Glasgow.
Financial support and sponsorship: none declared.
Conflicts of interest: none declared.
1. Burnstein R, Buckland R, Pickett JA. A survey of epidural analgesia for labour in the United Kingdom. Anaesthesia
2. Wantman A, Hancox N, Howell PR. Techniques for identifying the epidural space: a survey of practice amongst anaesthetists in the UK. Anaesthesia
3. Dogliotti AM. A new method of block anesthesia. Segmental peridural spinal anesthesia. Am J Surg
4. Paech MJ, Godkin R, Webster S. Complications of obstetric epidural analgesia and anaesthesia: a prospective analysis of 10995 cases. Int J Obstet Anesth
5. Dang T, Annaswamy TM, Srinivasan MA. Development and evaluation of an epidural injection simulator with force feedback for medical training. Stud Health Technol Inform
6. Anderson BD, Macgill JC, Anderson GB, et al. A novel epidural simulator. Anesthesiology
7. Cloote AH, Parke TJ, Kinsella SM. An analysis of three different loss of resistance techniques using the ‘greengrocer's’ epidural simulator. Int J Obstet Anesth
8. Leighton BL. A greengrocer's model of the epidural space. Anesthesiology
9. Price DD, McGrath PA, Rafii A, Buckingham B. The validation of visual analogue scales as a ratio scale for chronic and experimental pain. Pain
10. Devitt HD, Kurrek MM, Cohen MM, Cleave-Hogg D. The validity of performance assessments using simulation. Anesthesiology