Problems with the urine culture test

Urine culture tests for UTI are inaccurate and based on science that is out of date. They miss 90% of infections.

Recent studies have found mid-stream urine culture tests used by GPs and hospital consultants to diagnose miss 90% of infections. (Note – these studies use the thresholds and criteria used by laboratories, study design, and population). (i)

Relying on flawed tests have serious consequences for patients. When cultures fail to identify an infection, doctors and urologists can dismiss bacterial infection and look for other causes. With an infection ruled out, patients often then undergo painful and invasive investigations to find another cause. These can worsen existing undiagnosed infections.

So why do cultures fail to establish an infection in up to 90% of samples?

Laboratory analysis and protocols

Research carried out in the 1950s remains the global diagnostic basis for interpreting a urine culture. To confirm a positive infection, laboratory analysis results must show growth of ≥10 ⁵ bacterial colony forming units in a millilitre of urine. This means that there must be a level of 100,000 colony forming units of bacteria per millilitre of cultured urine to acknowledge infection.

This diagnostic criteria is based on the research studies by E.H Kass of 1956/7.(i) This study involved 74 women with an acute kidney infection and 335 symptomatic patients used as a control. He then used a further sample of pregnant women with kidney infection to represent severe UTI infection in another study to establish a diagnostic positive bacterial threshold, which he set at >10 ⁵ cfu/ml. This threshold is very specific, limiting the ability of clinicians to consider bacterial infections at lower levels. Counts below this threshold are usually considered contamination of the sample, the reported ‘mixed growth’, ‘low growth’ or ‘no significant growth’ that people often receive via their GP surgery on return of the urine culture.

Urine culture – the case against

These research studies have been challenged for the following reasons

The usage and study of pregnant women with acute kidney infections rather than those suffering from lower urinary tract infections of the bladder and urethra. Kass’s test was never, in its creation, validated for use in lower urinary tract infections for the bladder or urethra. But it has been readily adopted globally for all infections of the urinary and renal tracts and has not been revised or updated since the 1950s.
Kass’s belief that the normal bladder is sterile and bacteria introduced via the urethra are the cause of infection. Researchers have refuted that belief in recent years now more is understood about the urinary microbiome (the community of microbes in the urinary tract). Read more about the urinary microbiome. Are infections solely attributable to bacteria and in particular one bacteria introduced into the bladder? What part do viruses, fungi or other microbes in this urinary soup play?
Kass also noted that sufferers of a UTI tend to over-dilute their urine by drinking too much water. This means that any sample is not sufficiently concentrated to identify the growth of a single bacterial species let alone multiple infection causing bacteria.
The failure to recognise polymicrobial infections in Kass’s research (an infection comprising of one or more bacteria) with his belief that the urinary tract infection is caused by a single bacteria. Because of this, samples with multiple organisms or lactobacillus are labelled as ‘contaminated’ or ‘mixed growth’ as it is assumed that these bacteria came from other sources. In reality, these bacteria are not contaminants and may well be contributors to the infection, as in some cases infections can be polymicrobial.
The culture numbers are likely to depend on the ease of growth of bacteria in a laboratory. Bacterial species react differently to an oxygenated environment. The bacteria residing in your bladder have a limited oxygen source but when a urine sample is placed on a petri dish in the laboratory, the oxygen they are exposed to increases. This means that bacteria which flourish beneficially in this type of environment will grow. An example of this is E-coli whereas those that prefer the ecosystem and oxygen levels of the bladder will grow in limited numbers and are often dismissed as contamination.
The standard urine culture also does not pick up on viral or fungal causes of infection, which have been proven to be the cause for certain symptomatic patients with negative culture tests, even in patients that are not immunocompromised.
The E. coli focused design of a standard culture could explain our current E. coli centric view of UTI we have today, since so many other organisms remain undetected. Indeed in one study that analysed 157,000 urine samples using a different technology, E.coli was the dominant species in only 28% of cases.

Infections can also be multi-bacterial

This laboratory analysis has also relied on the culture of a single pure-growth bacterium based on the postulates formulated by Robert Koch and Friedrich Loeffler first set out in the late 1800s. These are:

that the micro-organism or other pathogen must be present in all cases of the disease.
the pathogen can be isolated from the diseased host and grown in pure culture – that is one single bacterial organism.
the pathogen from the pure culture must cause the disease when inoculated into a healthy, susceptible laboratory animal
and finally that the pathogen must be re-isolated from the new host and shown to be the same as the originally inoculated pathogen.

Research since the 1980s now challenges this current single, causal pure bacterial diagnosis as it has been demonstrated that infections can be multi-bacterial. (iii) However, infection with mixed bacteria is currently considered contamination by vaginal or vulval matter or other sources. Research studies now show evidence of the bladder being the origin for these mixed microbial growth which could include several strains of bacteria, fungi, moulds or viruses, a combination of which could potentially cause infection. (iii).

These multi-bacterial infections may require different treatment interpretations. One antibiotic or natural antimicrobial may be insufficient to deal with more than one pathogen in the urine.

A 2018 research study using urine specimens obtained via the clean-catch mid stream urine method (MSU) and employing the current UK microbiological protocols which are tailored to acute UTI, failed to detect a variety of bacterial species, including recognized urinary bacteria.

The diagnostic MSU culture was unable to differentiate between patients with lower urinary tract symptoms (LUTS) and controls i.e. those with no symptoms of a UTI. (iv)

The role of epithelial cells as indicators of infection

The immune system response to an upsurge in bacteria is by shedding bladder wall epithelial cells to prevent the attachment of bacteria to the bladder wall and setting up colonies of infection. When a UTI has taken hold, more epithelial cells are present. They can be reported as ‘few’, ‘moderate’ or ‘many’ on a patient’s laboratory report. A report showing ‘few’ is critical because it shows inflammation, yet at present unless there is positive bacterial growth alongside these epithelial cells, it is likely the sample will dismissed and the cell counts noted as contamination.

The skin cells shed from the vulva or vagina and kidney epithelial cells are each different in shape to those of the bladder wall helping to identify a kidney infection or vulval contamination as opposed to a UTI. With the automation of laboratory processes to increase the number of samples analysed daily, this detailed differentiation is missed if numbers are low and often a negative bacterial test has meant that these cells are dismissed as contamination of the sample. However, this clinical research study published in The International Urogynaecology Journal in 2018 demonstrates bacterial infection based on bladder wall epithelial cell analysis even in patients with low counts.

Sample Contamination

1 in 4 urine samples sent for analysis are rejected due to “contamination” and at present there is no established protocol or standard for collection of urine by Public Health England other than the recommendation of a clean catch, mid-stream urine sample.

For women this can prove to be a particularly difficult process. It requires starting the urine flow whilst parting the labia, stopping urination to position a small tube to collect the urine whilst keeping the labial folds open, removing the sample pot and finishing urinating. Spillage or splashing must be avoided and all of this often in a cramped GP surgery toilet or urgent care centre. More often or not there are little or no instructions from the GP or Nurse as to how to collect a “clean catch” sample.

Contamination is usually caused by flora and bacteria entering the urine specimen from genital skin or the hands. This haphazard method of collection can cause a mixed growth laboratory result (which can be noted as contamination), which results in a repeat sample request a few days later by the GP delaying treatment for a worsening infection.

In 2020, researchers examined the methods of sample collection to understand the issue of sample contamination. The aim of this study was to compare the occurrence of contamination and the quality of critical urinary sediment which is necessary for determining whether infection is present – that is through the shedding of urinary epithelial cells. They obtained samples from four different collection methods from patients suffering from lower urinary tract symptoms. The methods used were mid-stream urine (MSU), catheter specimen urine (CSU), a commercial MSU collecting device (Peezy) and a natural void from the first stream of urine. They found that the natural void was best at collecting bladder urinary sediment, with the majority of epithelial cells present derived from the urinary tract. CSU sampling missed much of the urinary sediment and showed sparse culture results. Finally, the MSU collection methods did not capture much of the bladder sediment. You can read the study, published in The International Urogynaecology Journal here.

Peezy Midstream, the collection device mentioned in the study above have also undertaken research in 2018 in association with Loyola University using the expanded quantitive urine culture (EQUC). In this study, they used EQUC to obtain specimens by standard clean catch urine collection or the Peezy midstream device from healthy women with no history of recurrent or chronic UTI. They hypothesized that voided urine obtained using Peezy would include less microbes from the urethra, periurethral, vulva, and vagina than voided samples obtained by typical clean catch techniques. Their results found that the use of the Peezy device provided a step toward a “cleaner” catch but noted that additional studies and methods are needed to validate their study conclusions particularly in the usage of “healthy” controls rather than those with a history of UTI. The study published in Female Pelvic Medicine and Reconstructive Surgery can be read here.

The storage and delivery of urine samples to the laboratory – how this affects white blood cells

Even if a sample is sent off to the laboratory for analysis, studies have found white blood cells (Leukocyte esterase) show significant degradation in the hours after a urine sample is collected. (v)

A positive test for leukocyte esterase on either dipstick or in laboratory analysis normally indicates the presence of inflammation of the urinary tract and may indicate infection.

The agreed total number of leukocytes to be present in a millilitre of urine to establish the baseline for infection remains unchanged since the work of C Dukes in 1928. He proposed a threshold of less than 10 white blood cells (WBC) per millilitre of urine as the upper limit of normal white blood cell excretion by the body in urine. Above that limit and the sample may be positive for infection if bacteria are grown that meet the existing criteria for confirmation of infection. Below that limit and with no bacterial growth, then the sample is considered “normal”. (vi)

A further point to note is that the time between sample collection and laboratory analysis may well exceed four hours especially if samples are only collected once a day from GP surgeries or if a hospital laboratory is very busy. This will lead to the sample WBC count decreasing significantly before laboratory analysis.

Commercial and hospital laboratories analyse around 65 million urine samples annually in the UK within the National Health Service (NHS) (vii). One UK hospital trust noted their pathology laboratory receives 900 samples a day for analysis.(viii) Thus immediate processing of the sample is unlikely.

A study found that after four hours post-collection, WBC loss was 60% without the use of boric acid as a preservative (x).

What to avoid if you need to provide a urine sample:

Dilution of the sample by drinking too much liquid (the first urination of the day when you wake up or acidic, concentrated urine is best for analysis). If this is not possible, limit fluids for around three hours before providing your sample so it isn’t too dilute.
Usage of previous antibiotics to treat a recent infection. Ideally you need to be off antibiotics for around 7-10 days to clear the medication from your system before providing another urine sample. Usage of antibiotics may inhibit bacterial growth resulting in a negative result even if you still have symptoms.
If you are using D-Mannose or other natural herbal supplements to treat your infections stop at least 48 hours before testing.
Urinary painkillers such as Azo should be stopped 24 hours before a sample as this medication type can affect results.
Diuretics should also be stopped as well as Vitamin C – all can lead to false negative results.

Urine Test References

(i) O’Brien K, Hillier S, Simpson S, et al. An Observational Study of Empirical Antibiotics for Adult Women with Uncomplicated UTI in General Practice. J Antimicrob Chemother 2007; 59(6): 1200–1203.

Little P, Merriman R, Turner S, et al. Presentation, pattern, and natural course of severe symptoms, and role of antibiotics and antibiotic resistance among patients presenting with suspected Uncomplicated urinary tract infection in primary care: observational study. BMJ 2010; 340: b5633.

McIsaac WJ, Low DE, Biringer A, et al. The impact of empirical management of acute cystitis on unnecessary antibiotic use. Arch Intern Med 2002; 162(5): 600–605.

Nazareth I, King M. Decision making by general practitioners in diagnosis and management of lower urinary tract symptoms in women. BMJ 1993; 306(6885): 1103–1106.

Vellinga A, Cormican M, Hanahoe B, et al. Antimicrobial management and appropriateness of treatment of urinary tract infection in general practice in Ireland. BMC Fam Pract 2011; 12(1): 108.

(ii) Kass, E.H. Asymptomatic infections of the urinary tract. Transactions of the Association of American Physicians Volume 69, 1956, Pages 56-64

(ii) Kass EH. Bacteriuria and pyelonephritis of pregnancy. Arch. Intern. Med. 105, 194–198 (1960).

(iii) Paul C. Schreckenberger, Travis K. Price, Evann E. Hilt, Tanaka Dune, Cynthia Brincat, Linda Brubaker, Elizabeth R. Mueller, Alan J. Wolfe. Detecting clinically relevant micro-organisms: we can do better 2015

(iii) Evann E. Hilt, Kathleen McKinley, Meghan M. Pearce, Amy B. Rosenfeld, Michael J. Zilliox, Elizabeth R. Mueller, Linda Brubaker, Xiaowu Gai, Alan J. Wolfe, Paul C. Schreckenberger, Urine is Not Sterile: Use of Enhanced Urine Culture Techniques to Detect Resident Bacterial Flora in the Adult Female Bladder. Journal of Clinical Microbiology. 2014 52(3):871-6

(iii) Natasha Curtissa, Aswini Balachandrana, Louise Krskab, Claire Peppiatt-Wildmanb, Scott Wildmanb, Jonathan Ducketta, A case control study examining the bladder microbiome. 2017

(iii) Lisa Karstens, Mark Asquith, Sean Davin, Patrick Stauffer, Damien Fair, W.Thomas Gregory, James T.Rosenbaum, Shannon K.McWeeney and Rahel Nardos. Does the Urinary Microbiome Play a Role in Urgency Urinary Incontinence and its Severity? 2016

(iii) Ahmed Moustafa, Harinder Singh, Weizhong Li, Kelvin J. Moncera, Manolito G. Torralba, Yanbao Yu, Oriol Manuel, William Biggs, J. Craig Venter, Karen E. Nelson, Rembert Pieper, Amalio Telenti. Microbial Metagenome of Urinary Tract Infection. 2017

(iii) Marcus J. Drake, Nicola Morris, Apostolos Apostolidis, Mohammad S. Rahnama and Julian R. Marchesi. The Urinary Microbiome and its Contribution to LUTS. 2015

(iii) Travis K. Price, Tanaka Dune, Evann E. Hilt, Krystal J. Thomas-White, Stephanie Kliethermes, Cynthia Brincat, Linda Brubaker, Alan J. Wolfe, Elizabeth R.Mueller, and Paul C. Schreckenberger. The Clinical Urine Culture: Enhanced Techniques Improve Detection of Clinically Relevant Micro-organisms. Journal of Clinical Microbiology. May, 2016

(iv) Stamm WE, Counts GW, Running KR, Fihn S, Turck M, Holmes KK. Diagnosis of coliform infection in acutely dysuric women. N. Engl. J. Med. 307(8), 463–468 (1982).

(iv) Latham RH, Stamm WE. Urethral syndrome in women. Urol. Clin. North Am. 11(1), 95–101 (1984).

(v) Sathiananthamoorthy S, et al J. Reassessment of Routine Midstream Culture in Diagnosis of Urinary Tract Infection. 2018, J. Clin. Microbiol., doi:10.1128/JCM.01452-18

(vi) Dukes C., Some Observations on Pyuria. Proc R Soc Med 1928; 21: 1179–83

(vii) Forte Medical FOI Request April 2016 to each UK NHS Hospital Trust

(viii) https://www.wyevalley.nhs.uk/visitors-and-patients/county-hospital-(acute)/a-z-departments/pathology.aspx

(x) Kupelian AS, Horsley H, Khasriya R, Amussah RT, Badiani R, Courtney AM, Chandhyoke NS, Riaz U, Savlani K, Moledina M, Montes S, O’Connor D, Visavadia R, Kelsey M, Rohn JL, Malone-Lee J. Discrediting microscopic pyuria and leucocyte esterase as diagnostic surrogates for infection in patients with lower urinary tract symptoms: Results from a clinical and laboratory evaluation. BJU International 2013

Rabinovitch A. Urinalysis and collection, transportation, and preservation of urine specimens: approved guideline. 2d ed. Wayne, Pa.: National Committee for Clinical Laboratory Standards, 2001. NCCLS document GP16-A2.