Blood Cultures in Suspected Simple Cystitis vs Pyelonephritis

Written by: Em Wessling, MD (PGY-2) Edited by: Will Ford, MD (NUEM ‘19) Expert commentary by: Justin Morgenstern, MD

---

While the Joint Commission historically focused their urinary interests on CAUTI protection, there is room for improvement in how we care for those with community acquired urinary tract infections. In many emergency departments, patients with suspected pyelonephritis are having blood cultures drawn to screen for bacteremia. However, these cultures may be unnecessary and costly for most patients.

The Choosing Wisely campaign from the ABIM Foundation started in 2012 with the goal to reduce “the overuse [of medical testing] that does not add value for patients” (1). In 2015, Choosing Wisely, in collaboration with the Society for Healthcare Epidemiology of America, recommended that blood cultures should not be performed unless there are appropriate symptoms due to false positives leading to over treatment (2).  When looking at emergency department data, the lack of utility of blood cultures in general holds true. A study from Glasgow found that only 1.4% of all blood cultures drawn in the emergency department were true positives. Of these, less than 15% (or less than 0.2% of all cultures drawn) were used to guide clinical treatment, regardless of the suspected source of infection (3). 

Similarly, Choosing Wisely Australia recommends avoiding blood cultures if patients are not systemically septic and have a “direct specimen for culture,” including urine (4). However, researchers in Australia continue to debate if this Choosing Wisely recommendation is based on enough evidence to apply broadly, or if blood cultures would still be useful for specific, more complicated populations. 

In patients with pyelonephritis, we can see that blood cultures rarely add clinical value. In 2017, it was shown that less than 10% of patients who were hospitalized for community acquired acute pyelonephritis had positive blood cultures (5). In the same study, only 2.3% of the cases had differing blood cultures when compared to urine cultures that resulted in a change of care (5).   This was also demonstrated in a review article from 2005 that looked at the utility of blood cultures in immunocompetent, non-pregnant, adult patients and concluded that there was no use for blood cultures in this population (6). Blood cultures also have limited utility in predicting prognosis in patients with pyelonephritis. A recent study from Spain looked at all-cause mortality in pyelonephritis and urinary sepsis patients with bacteremia vs those without and found that here was no change (7). The same prospective study found no significant difference in length of stay and ICU transfers (7). 

Blood cultures do occasionally have a role to play in the treatment of pyelonephritis. While the average person with an uncomplicated UTI or pyelonephritis may not have an indication for blood cultures, there are select populations for whom blood cultures show a distinct benefit. Initially, it was postulated that those groups would include those with instrumentation of the Genito-urinary tract and those who are immunocompromised (6).  Recent studies suggest that blood cultures may also be helpful in patients recently treated with antibiotics, as they are at a higher risk for sterile urine culture but may still have a positive blood culture. Additionally, chronically ill patients may have polymicrobial urine cultures, for whom a single clinically relevant organism may be able to be isolated from a blood culture (8). 

While there is a plethora of research to demonstrate that in pyelonephritis for which a urine cultures is available, blood cultures are often not clinically significant, researchers are still trying to parse out which select groups would benefit from them.

---

Expert Commentary

This is an excellent post that clearly comes to the right conclusion: blood cultures are not necessary for most patients with pyelonephritis. (In fact, I think it’s likely that even urine cultures are overused.)

Whenever we order a test, we should consider: how will the results change my management? 

Blood cultures are occasionally used diagnostically (for endocarditis), but pyelonephritis is a clinical diagnosis. The results of the blood culture is not going to change our final diagnosis. Therefore, the only management change we could possible make based on the blood cultures is a change in antibiotics. 

Our initial antibiotics cannot be guided by cultures, but luckily our empiric antibiotics are incredibly effective. There are only a handful of bacterial species that routinely cause urinary tract infections, and we have a handful of commonly used antibiotics, so we choose correctly most of the time. Even when the chosen antibiotic is reported as resistant on the culture, you will frequently find that the patient is better clinically. (In vitro antibiotic resistance is not the same as in vivo resistance.)

Only a small number of patients will have a positive blood cultures. Only a smaller number will have a positive culture demonstrating resistance to the original antibiotic. And an even smaller number will still be sick at the time that the culture is reported. For this small minority of patients, the culture will guide our new antibiotic choice, but considering the limited menu of antibiotics we use for UTIs, we probably could have made the same decision empirically, and we would be right most of the time. (Even in the era of multidrug resistance and ESBL, you should have a general sense of what antibiotics work in your community.)

However, that entire line of logic is unnecessary if you already took a urine culture. (The same line of reasoning can demonstrate why urine cultures are probably also overused, but I will admit that although I never send cultures in simple UTIs, I still send them in pyelonephritis.) Considering that it is the actual source of the infection, the urine culture is far more likely to grow the causative organism. So if you already have a test that will guide your antibiotic change in the case of resistance, the blood culture is completely redundant. It cannot help. So we should stop sending them. 

Justin Morgenstern, MD

Emergency Medicine

Toronto, Canada

---

References

1. Levinson, Wendy, et al. "‘Choosing Wisely’: a growing international campaign." BMJ Qual Saf 24.2 (2015): 167-174.

2. Society for Healthcare Epidemiology of America. “Don’t Perform Urinalysis, Urine Culture, Blood Culture or C. Difficile Testing Unless Patients Have Signs or Symptoms of Infection. Tests Can Be Falsely Positive Leading to over Diagnosis and Overtreatment.” Choosing Wisely - An Initiative of the ABIM Foundation, ABIM Foundation, 1 Oct. 2015, www.choosingwisely.org/clinician-lists/shea-urinalysis-urine-culture-blood-culture-or-c-difficile-testing/.

3. Howie, Neil, Jan F. Gerstenmaier, and Philip T. Munro. "Do peripheral blood cultures taken in the emergency department influence clinical management?." Emergency Medicine Journal 24.3 (2007): 213-214.

4. Denny, Kerina J., and Gerben Keijzers. "Culturing conversation: How clinical audits can improve our ability to choose wisely." Emergency Medicine Australasia 30.4 (2018): 448-449.

5. Kim Y, Seo MR, Kim SJ, Kim J, Wie SH, Cho YK, Lim SK,
Lee JS, Kwon KT, Lee H, Cheong HJ, Park DW, Ryu SY,
Chung MH, Pai H. Usefulness of blood cultures and radiologic imaging studies in the management of patients with community-acquired acute pyelonephritis. Infect Chemother 2017;49:22-30.

6. Mills, Angela M., and Suzanna Barros. "Are blood cultures necessary in adults with pyelonephritis?." Annals of emergency medicine 46.3 (2005): 285-287.

7. Artero, Arturo, et al. "The clinical impact of bacteremia on outcomes in elderly patients with pyelonephritis or urinary sepsis: A prospective multicenter study." PloS one 13.1 (2018): e0191066.

8. Karakonstantis, Stamatis, and Dimitra Kalemaki. "Blood culture useful only in selected patients with urinary tract infections–a literature review." Infectious Diseases 50.8 (2018): 584-592.

---

How to Cite This Post

[Peer-Reviewed, Web Publication] Wessling, E, Ford, W. (2020, Mar 26). Blood Cultures in Suspected Simple Cystitis vs Pyelonephritis. [NUEM Blog. Expert Commentary by Morgenstern, J]. Retrieved from https://www.nuemblog.com/blog/bcx-cystitis

---

The UTI that isn’t: Why a common condition presents such a diagnostic challenge.

Written by: Ashley Amick, MD (NUEM Alum ‘18) Edited by: Michael Macias, MD (NUEM Alum ‘17) Expert commentary by: Alexander Lo, MD

---

This is Part 2 of the blog post on the diagnosis of UTIs. Check out Part 1 here

Urinary tract infection (UTI) is the most common commonly diagnosed infection in the United States.  However, a high incidence of diagnoses does not render those diagnoses appropriate.  Increasing evidence suggests that this common condition poses a serious diagnostic challenge.  Erroneously identified UTIs frequently result in inappropriate treatment, as well as delays in management of the true underlying pathology.  In an era where ever more terrifying multi-drug resistant organisms continue to emerge, increasing emphasis is placed on evidence-based practice and antimicrobial stewardship.   In the acute care setting, where information is limited and time is scarce, guideline-based management can aid the Emergency Physician (EP) in improving both individual and community-level outcomes.

Despite increased awareness of UTI’s role in antimicrobial stewardship and cost-effective care, leading interest groups have failed to create a consensus definition of UTI.  (For an interesting experiment ask your colleagues what they consider diagnostic criteria for UTI, and prepare for wide variability).   Generally speaking, UTI is a diagnosis arrived at by two core features: 1) laboratory testing suggestive of infection, of which urine culture is considered gold standard; and 2) clinical symptomatology. 

Herein lies a major quandary for the Emergency Physician EP – culture data is not available in a timely fashion, and determining what defines a “symptom” of a UTI is, at best, elusive.  In the absence of culture data, the EP must rely upon a urinalysis (UA), with or without microscopy, as a surrogate.  Certain elements of the UA are thought to be particularly predictive of a true infection, including leukocyte esterase, nitrite, white blood cells, red blood cells, and bacteria.  However, when considered either alone or in combination, there is variable sensitivity and specificity of nearly all elements of a dipstick or UA.  Even when both leukocyte esterase and nitrite are present, the sensitivity and specificity is too poor to definitively diagnose or exclude a UTI.

Part of the poor predictive performance of UAs may be attributed to poor collection techniques and the presence of chronic bactiuria.  Obtaining a clean-catch sample in the emergency department setting can be a formidable challenge.  Studies suggest less than 10% of ED patients use proper midstream clean-catch techniques.  Concerningly, 50% of patients with a contaminated urine sample receive inappropriate intervention and antibiotics.  Proper education on sampling techniques as well as and in and out catheterization when appropriate, should be routinely employed. 

Despite adequate sample collection, UA interpretation is frequently confounded by the presence of asymptomatic bactiuria (ASB).  While definitions vary, the Infectious Disease Societies of America (IDSA) define ASB as isolation of a specified quantitative count of bacteria (105 cfu/ml from clean catch specimens) in a patient without symptoms or signs referable to urinary infection, such as frequency, urgency, dysuria, or suprapubic pain.  ASB is common in the geriatric population, and prevalence increase with age and in institutionalized patients.  ASB, like UTI, will frequently yield a UA positive for bacteria, LE, nitrate, and pyuria, therefore rending the UA of little use in differentiating between these two conditions. Given these considerations, the clinical symptoms become the most important factor in making the correct diagnosis.

When considering the diagnosis of UTI, beginning with an assessment of patient signs and symptoms seems not only rational, but intuitive. However, in the ever-increasing drive for efficiency, UAs are frequently drawn indiscriminately to expedite work-up.  In a recent study of patient treated for UTI in an ED population, 2/3 of patients diagnosed with a UTI had a UA collected as part of an order set, often before being evaluated by a clinician.  It was also found that antibiotics were administered inappropriately in 59% of those patients, due to lack of clinical signs or symptoms to substantiate a diagnosis of UTI.  Going about the diagnostic work-up in a backwards way invites not only anchoring bias when a UA is positive, but places pressure on the clinician to treat a UTI that isn’t.  Clinicians require discipline in looking beyond an abnormal UA, and work to objectively determine if the criteria for UTI are met based on symptomatology – or better yet – order UAs only when symptoms warrant further investigation.

Determining what constitutes a symptom – at least a symptom that should prompt a urinalysis – remains controversial.  According to the CDC and SHEA guidelines, symptoms consistent with a UTI include fever and lower genitourinary symptoms such as dysuria, urgency, frequency, suprapubic pain, and costovertebral angle discomfort.  Noteworthy is the omission of falls, altered mentation, and general malaise in the elderly in the absence of an indwelling catheter.  (See the related post: ‘delirium as a symptom of UTI, physiology or pseudoaxiom?’ for further discussion)

According to the most contemporary guidelines, these nonspecific symptoms without localizing symptoms or fever, are no longer sufficient to support the diagnosis of UTI.  This represents a shift in not only traditional clinical teaching, but a departure from prior guidelines.  This change results from a realization that both asymptomatic bactiuria and altered mentation are prevalent in the geriatric population, and there is a paucity of evidence supporting a causal link between these findings.  Despite these new recommendations, altered mentation, confusion, weakness, and falls are among the most frequent reasons for obtaining a UA in the geriatric population.  In a population where ASB is prevalent, and procuring a clean urine sample is challenging, geriatric patients are at high risk of morbidity from inappropriate antibiotic therapy and unnecessary testing.  Perhaps more concerning is that with a presumptive diagnosis of UTI, little thought may be devoted to other potential diagnoses – at least until the patient fails to improve.

---

Expert Commentary 

Over 50 million U.S. adults > 65 years of age (“older adults”), account for over 20 million Emergency Departments (ED) visits each year [1].  Many of these patients have unmet and complex underlying medical needs that are often understated by their chief complaints. The tempting application of traditional ‘one complaint; one algorithm’ approach taught to many emergency physicians, may often result in long-term, downstream, adverse outcomes.  One of those relevant to the accompanying blog, is the traditional “if grandma is delirious, look for and treat the UTI” doctrine.  A review of the literature proves that the evidence linking UTI’s to delirium in older adults is lacking [2]. Many older adults are bacteriuric; most do NOT have to be treated [3].  The delirium is not a reason to treat bacteriuria [4].  It is also just as likely that it is the other comorbid conditions causing the delirium, since 75% of older adults have two or more comorbid chronic conditions [5]. many of which have the potential to cause delirium at any time[6].   The patient may likely require admission for the delirium, but a more comprehensive investigation into its etiology is more helpful than treating the easy target of a contaminated urine sample

Alexander S Lo, MD, PhD

Assistant Professor of Emergency Medicine, Northwestern University 

---

How to Cite this Post

[Peer-Reviewed, Web Publication]  Amick A, Macias M (2018, November 26). The UTI that isn’t: Why a common condition presents such a diagnostic challenge [NUEM Blog. Expert Commentary by Lo A]. Retrieved from http://www.nuemblog.com/blog/uti-part2

---

Other Posts You May Enjoy

---

Resources

1. Little, P., et al. "Dipsticks and diagnostic algorithms in urinary tract infection: development and validation, randomised trial, economic analysis, observational cohort and qualitative study." Health Technol Assess 13.19 (2009): 1-73.

2. Van Nostrand, Joy D., Alan D. Junkins, and Roberta K. Bartholdi. "Poor predictive ability of urinalysis and microscopic examination to detect urinary tract infection." American journal of clinical pathology 113.5 (2000): 709-713.

3. Schulz, Lucas, et al. "Top Ten Myths Regarding the Diagnosis and Treatment of Urinary Tract Infections." The Journal of emergency medicine (2016).

4. Bent, Stephen, and Sanjay Saint. "The optimal use of diagnostic testing in women with acute uncomplicated cystitis." The American journal of medicine 113.1 (2002): 20-28.

5. Klausing, Benjamin T., et al. "The influence of contaminated urine cultures in inpatient and emergency department settings." American Journal of Infection Control (2016).

6. Gupta, Kalpana, et al. "International clinical practice guidelines for the treatment of acute uncomplicated cystitis and pyelonephritis in women: a 2010 update by the Infectious Diseases Society of America and the European Society for Microbiology and Infectious Diseases." Clinical infectious diseases 52.5 (2011): e103-e120.

7. Nicolle, Lindsay E., et al. "Infectious Diseases Society of America guidelines for the diagnosis and treatment of asymptomatic bacteriuria in adults." Clinical Infectious Diseases (2005): 643-654.

8. Detweiler, Keri, Daniel Mayers, and Sophie G. Fletcher. "Bacteruria and Urinary Tract Infections in the Elderly." Urologic Clinics of North America 42.4 (2015): 561-568.

9. Kiyatkin, Dmitry, Edward Bessman, and Robin McKenzie. "Impact of antibiotic choices made in the emergency department on appropriateness of antibiotic treatment of urinary tract infections in hospitalized patients." Journal of hospital medicine (2015).

10. Horan, Teresa C., Mary Andrus, and Margaret A. Dudeck. "CDC/NHSN surveillance definition of health care–associated infection and criteria for specific types of infections in the acute care setting." American journal of infection control 36.5 (2008): 309-332.

Delirium as a symptom of UTI: physiology or pseudoaxiom?

Written by: Ashley Amick, MD (NUEM alum '18) Edited by: Michael Macias, MD (NUEM alum '17) Expert commentary by: Alexander S Lo, MD, PhD

---

Asymptomatic bacteriuria (ASB) is a prevalent condition in the elderly population.  Bacterial colonization of the genitourinary (GU) tract increases with age and institutionalized status.  Though once thought to be pathogenic, randomized trials clearly demonstrate that treatment of ASB with antibiotics does not improve outcomes, except in pregnant patients and those undergoing GU procedures.  Emerging data even suggest there may be a protective effect of colonizing bacteria.  Conversely, there is increasing recognition of the dangers of inappropriate antibiotic use, both to the individual and the general population, and widespread agenda to limit unnecessary antimicrobial use. 

As the antibiotic stewardship movement marches forward, the treatment of ASB continues to be a central focus.  Recent guidelines emphasize that the presence of lower GU symptoms is the key distinction between UTI and ASB.  This strategy may be easily adopted in young otherwise healthy patients, but reaches a major stumbling block when considering the elderly population.  This is in part due to the fact that many clinicians believe that there is a causative relationship between UTI and delirium in the absence of other localizing symptoms or signs of systemic infection.  In other words, delirium is the symptom that substantiates a diagnosis of UTI in the presence of otherwise asymptomatic bacteriuria.  This concept, now generations old, is still taught in many medical school curricula.  The correlation between delirium and UTI is so well established in the minds of clinicians that many have never questioned whether this presumed association is rooted in data.

The concerning truth is that there is no reliable evidence to suggest that such a relationship between delirium and UTI exist.  A recent review of the literature found only five papers addressing this association primarily, all were observational and therefore lacked the ability to make conclusions about the degree of causation.  All studies were severely methodologically flawed, and none were case-control, cohort, or RTCs.  Additionally, there is no physiologic evidence or models to suggest that bacteriuria in the absence of systemic illness, results in cognitive dysfunction.  No known studies have ever shown that treatment of otherwise asymptomatic bacteriuria improves delirium outcomes.  Taking these data into account, the CDC and SHEA created guidelines specifically do not include delirium as a reason to treat potential UTIs in non-catheterized patients.  These represent a departure from earlier guidelines that included altered mental status as a symptom of UTI in the elderly.  The new SHEA recommendations have been tested in a large randomized trail and were found to be safe when compared to standard care.

Despite efforts to shift practice patterns in the direction of a more guideline-based management, ASB continues to be unnecessarily treated at high rates in the elderly.  One reason may be that anecdote is a powerful source of bias.  Many clinicians support their belief of a causative correlation between UTI and delirium by referencing cases where patient presented with confusion and were found to have a UTI.  The problem is, how was that “UTI” diagnosed?  The distinction is more than just semantics.  In the absence of GU symptoms and signs of systemic infection, then the clinician made the diagnosis solely on the basis of a UA and urine culture.  But as previously discussed, both a UA and culture will frequently be positive in both ABS and UTI, and cannot reliably distinguish between the two conditions. 

Many clinicians will cite the fact that the patients may improve following antibiotic administration, thereby confirming their suspicion of a presumed UTI-related delirium.  However, delirium frequently is short lived and self-resolving, therefore improvement is likely to be simply coincidental.  In addition, along with antibiotics administration patients also often receive intravascular volume, thereby improving hydration status, which is a frequent cause of delirium.  These factors confound the ability of the clinician to objectively interpret the causative relationship between the delirium and bacteriuria.  High quality randomized trials will be needed to further clarify these issues and assess is the high rate of concurrence of bacteriuria and delirium is due to causation or simply coincidence.

---

Expert Commentary

Over 50 million U.S. adults > 65 years of age (“older adults”), account for over 20 million Emergency Departments (ED) visits each year [1].  Many of these patients have unmet and complex underlying medical needs that are often understated by their chief complaints. The tempting application of traditional ‘one complaint; one algorithm’ approach taught to many emergency physicians, may often result in long-term, downstream, adverse outcomes.  One of those relevant to the accompanying blog, is the traditional “if grandma is delirious, look for and treat the UTI” doctrine.  A review of the literature proves that the evidence linking UTI’s to delirium in older adults is lacking [2]. Many older adults are bacteriuric; most do NOT have to be treated [3].  The delirium is not a reason to treat bacteriuria [4].  It is also just as likely that it is the other comorbid conditions causing the delirium, since 75% of older adults have two or more comorbid chronic conditions [5]. many of which have the potential to cause delirium at any time[6].   The patient may likely require admission for the delirium, but a more comprehensive investigation into its etiology is more helpful than treating the easy target of a contaminated urine sample

Alexander S Lo, MD, PhD

Assistant Professor of Emergency Medicine, Northwestern University 

---

Posts you may also enjoy

---

How to cite this post

[Peer-Reviewed, Web Publication]   Amick A, Macias M (2018, July 30). Delirium as a symptom of UTI: physiology or pseudoaxiom.  [NUEM Blog. Expert Commentary by Lo A]. Retrieved from http://www.nuemblog.com/blog/uti-part1

---

Resources

1. Pines JM, Mullins PM, Cooper JK, Feng LB, Roth KE. National trends in emergency department use, care patterns, and quality of care of older adults in the United States. Journal of the American Geriatrics Society. 2013;61(1):12-17.

2. Balogun SA, Philbrick JT. Delirium, a Symptom of UTI in the Elderly: Fact or Fable? A Systematic Review. Canadian geriatrics journal : CGJ. 2014;17(1):22-26.

3. Finucane TE. "Urinary Tract Infection"-Requiem for a Heavyweight. Journal of the American Geriatrics Society. 2017;65(8):1650-1655.

4. Ninan S. Don't assume urinary tract infection is the cause of delirium in older adults. Bmj. 2013;346:f3005.

5. Working Group on Health Outcomes for Older Persons with Multiple Chronic C. Universal health outcome measures for older persons with multiple chronic conditions. Journal of the American Geriatrics Society. 2012;60(12):2333-2341.

6. Kuluski K, Hoang SN, Schaink AK, et al. The care delivery experience of hospitalized patients with complex chronic disease. Health expectations : an international journal of public participation in health care and health policy. 2013;16(4):e111-123.

7. McKenzie, Robin, et al. "Bacteriuria in individuals who become delirious." The American journal of medicine 127.4 (2014): 255-257.

8. Balogun, Seki A., and John T. Philbrick. "Delirium, a symptom of UTI in the elderly: fact or fable? a systematic review." Canadian Geriatrics Journal 17.1 (2013): 22-26.

9. Nace, David A., Paul J. Drinka, and Christopher J. Crnich. "Clinical uncertainties in the approach to long term care residents with possible urinary tract infection." Journal of the American Medical Directors Association 15.2 (2014): 133-139.

10. Gau, Jen-Tzer, et al. "Interexpert agreement on diagnosis of bacteriuria and urinary tract infection in hospitalized older adults." J Am Osteopath Assoc 109.4 (2009): 220-226.

11. Juthani-Mehta, Manisha, et al. "Interobserver variability in the assessment of clinical criteria for suspected urinary tract infection in nursing home residents." Infection Control & Hospital Epidemiology 29.05 (2008): 446-449.

12. Schulz, Lucas, et al. "Top Ten Myths Regarding the Diagnosis and Treatment of Urinary Tract Infections." The Journal of emergency medicine (2016).