• 
• 

Three Part Question

In [febrile neonates] is [clean-catch or supra-pubic aspiration] better at [obtaining uncontaminated urine samples and so detecting true urinary tract infections]?

Clinical Scenario

A previously well, term 4 day old female neonate presents to the emergency department. On examination she is febrile with a temperature of 39.1 oC, tachycardic, tachypnoeic and has no focus for her fever. You decide to do a full septic screen and one of the things you need is a urine sample. You wonder whether a supra-pubic aspiration would be the best method for obtaining a sample in this neonate?

Search Strategy

Using the Medline database: 1950- week 1 June 2010 via OVID.
Embase database: 1980- week 23 2010 via OVID
CinAHL: CINAHL plus: 1937- June 2010
[clean catch.mp. OR clean catch$.mp] AND [suprapubic aspiration$.mp.OR bladder aspiration$.mp. OR urine aspiration$.mp.] AND [exp Urinary Tract Infections OR exp Urine OR urine sample.mp OR exp Urinalysis] LIMIT to [english language and humans and ("all infant (birth to 23 months)" or "newborn infant (birth to 1 month)"]

Search Outcome

Medline: 8 papers, 4 of which were relevant
Embase: 0 citations
CINAHL: 2 papers, no additional relevant papers

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Ramage IJ, Chapman JP, Hollman AS et al 1999 UK	49 infants (<24 months) in a paediatric teaching hospital ward and its associated neonatal unit with suspected UTIs.	Diagnostic study: Paired clean catch and supra-pubic aspiration samples were taken (with supra-pubic aspiration as the gold standard).	Sensitivity	Clean catch= 88.9% (65.3-98.6%)	There was no median age given therefore it is not known how many of the participants were under the age of 28 days. The recruitment strategy for the study was not stated and the sample size was small.
			Specificity	Clean catch= 95% (83.1%-99.4%)
			False positives	Clean catch= 5%
			False negatives	Clean catch= 12%
Hardy JD et al 1976 UK	30 unwell children (20 males and 10 females), all in-patients, <3 years old and 15 of which were <28 days old.	Diagnostic study: Compared clean catch, bag specimen and suprapubic aspiration of urine.	Mixed growth	clean catch= 73.3%, supra-pubic aspiration= 0%	Children were in-patients in paediatric ward having routine screening and only entered into the study if they had bacterial growth from this specimen this is a different population from neonates presenting in the ED with a different pre-test probability. No power study and small participant numbers. Also the study was not blinded.
			Pure growth (diagnostic levels)	clean catch= 6.67%, supra-pubic aspiration= 13.3%
			Pure growth (below diagnostic levels)	clean catch= 6.67%, supra-pubic aspiration= 0%
			No growth	clean catch= 13.3%, supra-pubic aspiration= 86.7%
Conn NK 1970 UK	7 children aged 2 weeks to 11 years in a paediatric ward (4 female and 3 male)	Diagnostic study	No growth	clean catch=0%, supra-pubic aspiration= 57.1%	A very small sample size. Unblinded study.
			Growth (diagnostic levels)	clean catch= 71.4%, supra-pubic aspiration= 42.9%
			Growth (below diagnostic levels)	clean catch= 28.6%, supra-pubic aspiration= 0%
Luco M et al 2006 Chile	1173 children aged 11 days to 14 years (mean age 3.6 years), (63% female) presenting to the paediatric emergency department with suspected UTI.	Retrospective cohort; over 1 year analysing all children ages under 15 undergoing urinary sampling	Sensitivity	Sterile methods (supra-pubic aspiration and catheter)= 83.3%, non-sterile (urine bad and clean catch)= 91.2%	Large age range with unknown number of participants less than 28 days old. The grouping together of urinary collection methods into sterile and non-sterile means supra-pubic aspiration and clean catch techniques can not be compared individually.
			Specificity	Sterile= 98.2%, non-sterile= 90.1%
			Positive predictive value	Sterile= 92.1%, non-sterile= 73%
			Negative predictive value	Sterile= 96%, non-sterile= 97.2%

Comment(s)

There is much debate in the available literature about the best technique for collecting urine specimens in children, however there is little evidence comparing the use of clean-catch and supra-pubic aspiration in the emergency department especially in neonates. Data comparing clan catch and supra-pubic aspiration is limited and there is wide variability in accuracy between studies, with most studies done with small numbers of older participants and in the in-patient setting. An explanation for this variability may be the technique of clean catch as in the Ramage et al (which had found clean catch a relatively accurate technique) a lot of effort was put in to educating those obtaining the sample. However in Hardy et al clean catch was found to be a very unreliable method of collection. All studies agreed that when clean catch was used, compared to supra-pubic aspiration, a significant number of false positives would be treated unnecessary.

Clinical Bottom Line

More research is needed in this field, especially in neonates and the emergency department setting. However, all the literature agreed that supra-pubic aspiration is the method of choice for obtaining a urine sample quickly and accurately when managing a sick neonate in the emergency department.

References

1. Ramage IJ, Chapman JP, Hollman AS, Elabassi M, McColl JH, Beattie JT Accuracy of clean-catch urine collection in infancy The Journal of Pediatrics December 1999, 765-767
2. Hardy JD, Furnell PM, Brumfitt W Comparison of sterile bag, clean catch and suprapubic aspriation in the diagnosis of urinary infection in early childhood British Journal of Urology 1976; 279-283
3. Conn NK A study of some of the methods of urinary collection in children Journal of Clinical Pathology 1970; 81-84
4. Luco M, Lizama M, Reichhard C, Hirsch T Urine microscopy as screen for urinary tract infections in a paediatric emergency unit in Chile Pediatric Emergency Care October 2006; 705-709