The wilful nature of children is responsible for a lot of the facial expressions we see in the paediatric emergency department but few parents are as frustrated as those waiting… and waiting… and waiting for the short-lived, poorly-timed, inevitable fountain or dribble of urine they are expected to catch in a sterile pot or bowl for the exclusion of urinary tract infection as a cause of their child’s symptoms. Many imagine a world where their child simply passes urine on demand. And now, in 2017, seven authors from Australia1 seem to be offering us just that. Well, almost…
First of all please note that this paper is Open Access – so there is no excuse for not spending a bit of time with the original article before reading on below!
What is this paper about?
The hypothesis of this study is that the Quick-Wee method (detailed below) can speed up the collection of urine samples from infants in the Emergency Department compared with control – usual care, meaning washing the genitals to reduce contamination then standing around with a pot. Of course, this isn’t the only option; if you don’t care about sterility, urine bags are an option (but the rates of contamination are probably unacceptably high for infants, in whom NICE recommends that further investigations might be indicated in the presence of a confirmed UTI); catheter-specimens and suprapubic aspirates have lower contamination rates but are considerably invasive. As a result for most children we end up waiting around for that elusive specimen.
The authors are following on from their published trial protocol2 (also Open Access) with the results of their study.
What did they do?
The Quick-Wee method they describe involves using cold water (theirs was refrigerated to 2.8ºC), soaked onto gauze held in plastic forceps and wiped on a circular motion over the suprapubic area for up to five minutes following initial cleaning for 10 seconds with room temperature sterile water (this cleaning was undertaken in both the control and intervention groups).
Patients attending the ED aged 1-12months for whom a urine sample was required were identified by ED clinicians and randomly, sequentially allocated to a treatment arm using opaque envelopes (allocation concealment) determined in order by random permuted blocks. The authors note that blinding was not possible due to the nature of the Quick-Wee intervention but it’s possible that they have missed other blinding opportunities (such as blinding of the data analysts to the groups during preparation of the manuscript3).
Babies under a month old were excluded because of the relative importance of neonatal sepsis; in the institution in question, catheter or SPA specimen is preferred in order to be as accurate as possible. The other exclusions seemed sensible too; anatomical or neurological anomalies precluding normal suprapubic sensation would make the intervention meaningless.
In total 344 patients were analysed: 174 were randomised to the Quick-Wee technique and 170 to standard watch-and-wait. The authors had carried out a sample size calculation and very sensibly added 10% to allow for later exclusions or dropouts so they were actively over-recruiting to the study, and this seemed to pay off since their required sample size was 322 (161 in each group), which they achieved despite 10 exclusions/withdrawals after randomisation.
The primary outcome was passage of urine within five minutes of the clock being started – or not. This binary outcome lends itself well to ????2 testing, where proportions of the outcome are compared between the two groups. Several secondary outcomes were considered, including whether the urine was actually caught, the contamination rates (assessed subsequently) and both parents and clinician satisfaction with the method of obtaining the sample. These are all reasonable outcomes in which we are interested as Emergency Physicians, but it’s important to remember that results in these areas are observational at best as they are not the outcomes around which the study was designed or powered.
What’s particularly positive about this study is that everything to do with delivering the study intervention was conducted within the department; patient selection, the intervention itself, the data collection including the results. This is unusual in published studies and certainly adds to the applicability of findings to an ED population.
What did they find?
There was a statistically significant difference in the primary outcome; in the control group a urine sample was obtained in 20/170 subjects (12% – 95% confidence interval 7%-18%) compared with 54/174 subjects in the intervention (Quick-Wee) group (31% – 95% confidence interval 24%-39%). This gave a difference in proportions of 19% (95% CI 11%-28%) which doesn’t quite add up but is, I assume, related to mathematical rounding. In any case the confidence interval is nowhere near zero so we can be reasonably sure that the true population value would be in favour of Quick-Wee.
When they looked at whether the voided sample was actually caught, that was higher in the Quick-Wee group too (30% vs 9%), presumably reflecting how continued attention to the genital area makes us more likely to catch the urine when it comes.
In addition, the authors found slightly lower (but not statistically significant) contamination rates – 27% with Quick-Wee, 46% with control – but these subgroups were small in size and consequently the results have wide confidence intervals attached. A larger, multi-centre study might be able to evidence lower contamination rates with the Quick-Wee technique, something which would certainly sway us towards using it in the ED – but given the time delay in determining the presence of organisms and identification of pure growth versus contaminant, we would likely need an enormous sample size for that particular study.
Lastly, the assessment of satisfaction with the technique fell in favour of Quick-Wee both for parents and clinicians. The use of a Likert-scale here is a little confusing as 1=very satisfied and 5=very unsatisfied so a higher score actually means lower satisfaction, in addition to being a crude assessment of the acceptability of this technique, but at least both groups seemed to prefer it over standard care. I do wonder how much of this was the effect of the novelty of the technique, though.
What does this mean?
You might remember a similar technique, published by Herreros et al in Archives of Disease in Childhood back in 20134 and covered in this ALIEM post. Their technique involved prehydration, washing of the genitals and then alternately rubbing the lower back and tapping the suprapubic area of neonates held under the armpits by a second party, legs dangling. The subtle difference in this current publication, the authors tell us, is that it was undertaken in the ED instead of NICU and it does not involve suspending the child aloft in anticipation of the sample being obtained. Playing Devil’s advocate, I would point out that in the ADC study the success rate was substantially higher (86%, albeit with a far smaller sample and without a sample size calculation) and that the median time to collection was given – just 57seconds. This data is not reported in the Quick-Wee study and it would certainly be nice to see.
In any case, this method seems to be worth a try as a first line attempt at obtaining a urine sample. The biggest challenge I can foresee is not the provision of cold fluid (every Emergency Department should have a drug fridge for insulin etc.) nor the training of ED staff but the fact that the staff members performing the intervention have to do so for five interrupted minutes. Aside from the fact that I doubt the attention span of the average Emergency Physician is that long, there’s good evidence that we are interrupted A LOT – 30.9 times in 180 minutes in one study5 – and I’m sure our ED nurses are exactly the same. Performing the Quick-Wee technique continuously for five minutes might be almost impossible in a busy ED. But from the results of this study, that probably shouldn’t preclude us from giving it a wee go :).
The SGEM – coming soon!
Before you go please don’t forget to…
- Subscribe to the blog (look top right for the link)
- Subscribe to our PODCAST on iTunes
- Follow us on twitter @stemlyns
- See our best pics and photos on Instagram
- PLEASE Like us on Facebook
- Find out more about the St.Emlyn’s team