Date: September 26th, 2017
Reference: Kaufman, et al. Faster clean catch urine collection (Quick-Wee method) from infants: randomised controlled trial. BMJ April 17.
Guest Skeptic: Dr. Natalie May trained as an emergency physician with subspecialty paediatric emergency medicine in the UK and worked in Manchester and Oxford before moving to Australia in 2015 to work for Sydney HEMS. She’s been there for the last 18 months working in prehospital and retrieval medicine and then in Emergency Medicine. She is a medical education enthusiast and has been an editor and regular contributor to the St. Emlyn’s blog and podcast since 2012. We appeared on stage at SMACC Chicago doing a parody of Jimmy Fallon’s Tight Pants skit.
Case: You are working the afternoon shift in the paediatric emergency department. It has been the usual busy after-school, after-work time. Your shift is almost over and the department finally seems to be under control. The next patient you see is a five-month old girl who has been brought by her parents with a history of fever and vomiting in the last 24 hours. She has not had any diarrhoea and her parents have noticed she seems particularly miserable when she is producing wet nappies (diapers), which she is doing more frequently than often. You wonder whether it might be a urinary tract infection (UTI) causing her symptoms and you explain to the parents the importance of a clean catch urine sample. They look at you in despair – all three of them are tired and they really want to go home. You know that it can take some time to get a urine sample in this age-group and wonder, is there a quicker way to get the wee?
Background: Urine samples are frequently obtained in the paediatric emergency department to rule in or rule out UTIs in those children with vomiting, fever, abdominal pain or non-specific illness.
In 2016, the American Academy of Pediatrics (AAP) reaffirmed their clinical practice guidelines for the diagnosis and management of UTIs. They state that the diagnosis of a UTI requires a urinalysis and a urine culture.
Obtaining the sample can be particularly tricky and potentially messy in children who are not yet continent. The rationale for definitive diagnosis included ensuring good antibiotic stewardship, missed UTIs can lead to renal scarring and other delayed pathology and a presumptive diagnosis can result in expensive and unnecessary imaging.
The decision to obtain an invasive sample is a clinical one. If the child looks sick and requires antimicrobial therapy then immediate collection of a urine sample through urethral catheterization or supra-pubic aspiration is recommended strongly by the AAP.
However, if the child is looks well, but the clinician determines that the febrile infant is not in the low risk group for a UTI, the AAP recommends two options. One option is to obtain the urine sample using an invasive method.
The second option is a two-step approach to identification of a UTI. This two-step process can decrease the rate of catheterization by over 30% without missing UTIs or increasing length of stay in the pediatric emergency department (Lavelle et al Pediatrics 2016).
The first step is to screen using a non-invasive collection method like a bagged urine specimen. If the sample is negative (no evidence of a UTI) then cultures may be omitted. A second step is needed if the bagged sample result is positive or shows evidence of a UTI.
Non-invasive methods are preferable for all but those at risk of neonatal sepsis or the ill appearing infant. However, catching a fresh sample is time consuming.
Just for completeness, the AAP does make a strong recommendation that if the clinician determines the febrile infant to have a low likelihood of a UTI, then no urine testing is necessary as long as there is clinical follow-up monitoring.
An open-access study published this year in the BMJ suggests a new method called the Quick-Wee could speeded up collection of clean catch urine samples in patients for whom we are considering a diagnosis of UTI.
Clinical Question: Does the Quick-Wee method increase the success rate of getting a clean catch urine collection in infants suspected of having a UTI?
Reference: Kaufman, et al. Faster clean catch urine collection (Quick-Wee method) from infants: randomised controlled trial. BMJ April 17.
- Population: Pre-continent infants age 1-12 months where a urine sample was required and the clinician felt the appropriate method of collection was a clean catch urine.
- Excluded: Neonates (age less than one month) and infants with neurological or anatomical abnormalities affecting sensation or voiding.
Intervention: Quick-Wee method – Genital area was cleaned for 10 seconds with sterile water at room temperature. This was followed by continued rubbing of the suprapubic area in a circular pattern with gauze (soaked in cold saline of kept in fridge at 2.8C) held by forceps. This was done for until clean catch urine was obtained up to five minutes. A parent/care-giver/clinician stood by ready to catch the urine sample.
- Comparison: Usual Care – Genital area was cleaned for ten seconds with sterile water at room temperature. A parent/care-giver/clinician stood by ready to catch the urine sample.
- Primary – Proportion of subjects voiding urine within five minutes (binary outcome)
- Secondary – Voiding with successful catch of urine sample, rates of contamination, and parental and clinician satisfaction with technique on a 5-point Likert scale (1- very satisfied and 5- very unsatisfied).
Authors’ Conclusions: “Quick-Wee is a simple cutaneous stimulation method that significantly increases the five minute voiding and success rate of clean catch urine collection.”
Quality Checklist for Randomized Clinical Trials:
- The study population included or focused on those in the ED. Yes
- The patients were adequately randomized. Yes. A 1:1 ratio using random permuted blocks of different sizes, envelopes taken sequentially.
- The randomization process was concealed. Yes. They used opaque envelopes.
- The patients were analyzed in the groups to which they were randomized. Yes. They performed an intention-to-treat analysis.
- The study patients were recruited consecutively (i.e. no selection bias) – Unsure
- The patients in both groups were similar with respect to prognostic factors. Yes. However, no P values are given but values in Table 1 appear similar.
- All participants (patients, clinicians, outcome assessors) were unaware of group allocation. No. This was not a blinded study.
- All groups were treated equally except for the intervention. Yes. The study intervention was very quick with a binary outcome so there’s little opportunity to add in any potential confounders.
- Follow-up was complete (i.e. at least 80% for both groups). Yes
- All patient-important outcomes were considered. Yes
- The treatment effect was large enough and precise enough to be clinically significant. Yes
Key Results: 354 subjects were recruited of whom 344 participated in the analysis. There were 170 in the control group and 174 in the intervention group. The mean age was 5.4 months with a 50/50 male/female split. The most common clinical indication for urine collection being fever (42%) followed by unsettled baby (38%). Only 17% of infants had urine collected because a UTI was specifically suspected.
Primary Outcome: Voiding within 5 minutes 31% Quick-Wee vs. 12% with usual care
This gives an absolute difference of 19% (95% CI of 11% to 28%) or a NNT of 5
- Secondary Outcomes: They had higher success rate with QuickWee overall, no difference in contamination and greater parent/clinician satisfaction and we will put the details in the blog post.
- Voiding with successful catch of urine sample (30% QuickWee vs. 9% Usual Care, p<0.001)
- Rates of contamination (27% QuickWee vs. 45% Usual Care, p=0.29)
- Median Parent/Care Giver Satisfaction (2 QuickWee vs. 3 Usual Care, p<0.001)
- Median Clinician Satisfaction (2 QuickWee vs. 3 Usual Care, p<0.001)
1) Selection Bias and External Validity: There are some question marks over the integrity of the selection procedure as patients were identified and recruited by emergency depargtment staff, which is great, because it’s very practical, but it looks as though those same staff may have then undertaken the intervention. This creates potential for bias if the recruiting staff are also undertaking the intervention. The published study protocol does not offer any further clarity here so we have to assume there was a lack of independence.
It is also important to point out this was a single center, large, tertiary pediatric hospital in Australia. The population may not reflect the same pediatric population you have presenting to your community or rural emergency department. They also excluded neonates and those over 12 months of age and so the results can not be applied to these age groups.
2) Lack of Blinding: There was no blinding in the study, which the authors tell us was because of the nature of the intervention. That seems fair as the observer will naturally be aware of which method of obtaining the sample was used; however, there were missed opportunities to use blinding in this study. The person starting the timer could have been independent, standing outside the cubicle and measuring five minutes or until the observer produced a sample.
In addition, the researchers assessing the data could have been blinded to which intervention group each data set came from. This is the simplest way of introducing blinding into a study but it is rarely utilised in isolation – and perhaps it should be used more.
3) Statistical Significance and Clinical Significance: The sample size was calculated with 80% power to detect a difference of 15% between groups in the primary outcome and the sample size was achieved. The difference in proportions was 19% with a 95% confidence interval that went from 11% to 28%. This gives which gives an NNT of 5 to successfully obtain one additional sample within five minutes.
With regards to clinical significance, they interviewed 20 experts (Paediatric Emergency Medicine Specialists and Paediatricians) and based on their expert opinion they felt it would need a difference of 15% to be clinically significant to their clinical practice. What about asking the parents/caregivers what they think would be clinically significant? We know that clinicians and patients (parents/care-givers) can have different levels of what they would consider “significant”.
4) Adverse Events and Contamination: There are no adverse events in either group, excluding crying and mild distress, which was thought to be standard for children in whom you are trying to obtain a urine sample – I’m inclined to agree.
There also seemed to be lower rates of contamination in the Quick-Wee method. However, the study was not designed or powered for this secondary outcome. The authors hypothesize this might be because the sample is obtained “faster and more forceful voiding” and that could reduce contamination.
In any case, contamination rates were low in both groups. It should be noted that culture results were only available for 55 patients in total (12/44 samples sent for culture in the Quick-Wee group grew contaminants compared with 5/11 in the control group). There was no robustness around this measure since the sending of the sample for culture was apparently at clinician’s discretion.
If the method really does reduce the proportion of contaminants then that would make it a bit of a game changer. But based on the available evidence from this study we cannot draw that conclusion about the Quick-Wee method.
5) Satisfaction Scores: This was pretty crudely measured too. They used the five-point Likert scale. However, it was a little counterintuitive in the paper because they have used 1 to mean “very satisfied” and 5 to mean “very unsatisfied”. So, a higher score does not suggest higher satisfaction. Median satisfaction scores for parents and clinicians were 2 for Quick-Wee and 3 for the standard clean catch urine, suggesting they were more satisfied with the Quick-Wee method.
Comment on Authors’ Conclusion Compared to SGEM Conclusion: We generally agree with the authors’ conclusion. Children aged 1-12 months do seem more likely to pass urine within five minutes if the Quick-Wee Method is used.
SGEM Bottom Line: If you have five minutes to spare, you can increase the proportion of infants in whom a clean catch urine sample can be obtained using the Quick-Wee method.
Case Resolution: You suggest to the parents that, now that the department is a little less busy, you could try the Quick-Wee method to make their infant pass urine a little bit faster. After a brief explanation of what’s involved they agree. After about 90 seconds of cold saline suprapubic stimulation you successful have a clean catch urine sample.
Clinical Application: This certainly seems like a relatively quick and reasonably effective way of increasing your chances of obtaining a clean catch urine within five minutes. If you have staff to spare to spend five minutes rubbing the suprapubic area of your patients with cold-saline-soaked gauze them great. However, we might not get to five minutes of uninterrupted stimulation in the emergency department before something more urgent comes through the door.
What Do I Tell My Patient (Parent/Care Giver)? It’s very important that we get a clean catch urine sample from your infant. We want to make sure the urine we get does not have any contamination from the skin. This means if we get a sample with no signs of infection we know the fever is most likely not due to a UTI. If there are signs of infection in the sample we will need to put a catheter inside your infant’s bladder to get a better sample of urine. This can be uncomfortable for your infant. This is why we really like to try to get a clean catch sample.
However, waiting to get a clean catch sample can be a long process. But there is a new method called the Quick-Wee method that can increase the chance we get some urine in less than five minutes. It involves rubbing over their bladder area with a cold gauze pad. There is no harm to your infant other than they might get fussy and cry. So, if you are happy with this we can give it a wee go!
Keener Kontest: Last weeks’ winner was Timothy Dekoninck a PGY1 Emergency Medicine resident from Wright State University. Tim knew the first membrane oxygenator was developed in 1885.
Listen to the SGEM podcast on iTunes to hear this weeks’ keener question. If you know the answer, then send an email to TheSGEM@gmail.com with “keener” in the subject line. The first correct answer will receive a cool skeptical prize.
Other FOAMed Resources:
- St Emlyns – JC: The Quick-Wee Method for Faster Clean Catch Urine Collection
- PEMLit – What are WEE Waiting for? The Quick-Wee Method for Faster Clean Catch Urine Collection
- PEMPlaybook – Urine Trouble
Remember to be skeptical of anything you learn, even if you heard it on the Skeptics’ Guide to Emergency Medicine.