Urinary tract infections (UTIs) remain an important and interesting topic: important as it is still one of the most common bacterial infections and causes of sepsis (urosepsis), and interesting as it is a great example of how new techniques contribute to our understanding of the pathogenesis of infectious diseases in general.
Even though clinical urine specimens have always been considered to be sterile when they do not yield uropathogens using standard operating procedures, this is now being questioned in several studies applying new sensitive methods such as DNA sequencing techniques.[1,2] Bacteria that are not routinely cultivated have been identified by new techniques in voided urine, urine collected by transurethral catheter and by suprapubic aspirate, regardless of whether the patients had urinary symptoms. Thus, a urine specimen that was previously assumed to be sterile appeared to obtain bacterial products. These findings have opened the discussion considering the accuracy of our current culture techniques, the presence of (uncultivated) bacteria in urine, the microbiome of the urogenital tract and the importance of the microorganism–host interaction.
These new insights, combined with the already known aspects of UTIs, are discussed in four reviews that are considered in this issue of Current Opinion of Infectious Diseases. As each review is focussing on a different patient population susceptible for UTIs, this issue provides a clear overview on where we stand at this moment concerning the epidemiology, pathophysiology, risk factors, consequences and treatment of UTIs.
The epidemiology of UTIs shows a great variety depending on the geographic location and patient population. An important aspect in this is the onset location, as the bacterial spectrum and resistance rates appear to greatly vary between community-associated UTIs and healthcare-associated UTIs. Tondogdu and Wagenlehner (pp. 73–79) describes the global epidemiology of UTIs, the dependence on geographic location and the influence of changes in our healthcare system. The shorter duration of hospital stay has led to an increase in admissions, which are shorter than the bacterial incubation time, resulting in the diagnosis of infection after hospital discharge. These infections are considered to be community onset healthcare-associated UTIs, a category that is not clearly defined yet. In addition to this, an increasing amount of patients are receiving healthcare (e.g. urinary catheter exchange) at home or are living in nursing homes. It is unclear whether the pathogens causing infections in these patients are comparable to the pathogens of the community or healthcare-associated UTIs, as many epidemiology studies have not taken this category into account. Research on this developing category is necessary to determine how we should categorize these patients to offer them an optimal treatment.
For UTI, different kinds or presentations exist. Schneeberger et al. (pp. 80–85) give an overview about febrile UTI and describe the most important conclusions of recent studies published considering the interaction between causative microorganism and the host. Eschericia coli remains the most common organism causing pyelonephritis and urosepsis. Less-virulent strains can cause infection in immunocompromised patients whereas more virulence factors are needed to infect the immune-competent host. Known risk factors are obstructive uropathy, diabetes and being older, although interestingly, neutropaenia was not a risk factor for lower UTI or urosepsis. This might be explained by the hypothesis that neutropenia as a result of chemotherapy does not affect immunoglobulin A-mediated genitourinary mucosal immunity. Another relevant reported finding was the isolation of E. coli isolates from hospitalized patients with pyelonephritis compared with outpatients with cystitis. The isolates from admitted patients with pyelonephritis showed higher bacterial gene diversity and more virulence factors. Further exploration on this field is necessary to determine whether these findings could contribute to the development of new therapeutic options.
A common presentation of UTI in men is bacterial prostatitis; nearly one in six men report a history of prostatitis. Furthermore, it is an inconvenient disease that also carries potential for serious morbidity from sepsis, abscess or fistula. In this issue of Current Opinion of Infectious Diseases, Gill and Shoskes (pp. 86–91) provide an overview and discussion of bacterial prostatitis, its categorization by type and a summary of recent findings.
The results of a large epidemiological study could not identify clear clinical risk factors for chronic prostatitis, but recurrence rates were high, suggesting incomplete eradiation. A possible explanation is found in the demonstration of the presence of bacteria that are able to produce a biofilm; the presence of these bacteria showed a positive association with symptom severity and a negative correlation with improvement after treatment. In another study, typical chronic prostatitis organisms were cultured in aspirated seminal vesicle fluid that suggests that the seminal vesicles may sequester bacteria causing recurrent prostatic infections. These new findings contribute to the understanding of the pathogenesis behind the recurrence of prostatic infections.
For patients with chronic prostatitis/pelvic pain syndrome and asymptomatic inflammatory prostatitis, to date no data support the role of ‘hidden infection’ or atypical organisms. Despite the strong belief of many patients that a true infection exists, there is currently no role for antibiotic treatment in these patients and there is often no explanation for their complaints [Gill and Shoskes (pp. 86–91)]. An important question in this matter is whether our current diagnostic tools are accurate. New sensitive techniques such as DNA sequencing might change our perspective on these patient categories.
In a certain way, the same counts for women with persistent urinary symptoms despite negative urine cultures. This issue is addressed in the review by Lakeman and Roovers (pp. 92–97). Interestingly, they discuss that in women with overactive bladder symptoms (OABs), low-count bacteriuria [at least 102 colony-forming unit (CFU)/ml] is more prevalent and the hypothesis is that intracellular bacterial communities in the bladder wall are present. Furthermore, several studies have reported relevant differences in the microbioma in women with OAB compared with healthy controls. These women are reporting symptoms and with the current diagnostic tools in most cases, no causative factor is found. Especially in these patients, the accuracy of our current diagnostic tools is being questioned, because the current screening methods are mostly using a detecting threshold of at least 105 CFU/ml, are mainly targeted at detecting uropathogenic E. coli and are failing to detect intracellular bacterial communities.
This detection threshold of at least 105 CFU/ml is based on studies performed by Kass in the 1940s and 1950s. However, after the publication of the results by Hooton et al. in 2013, which showed that in women with symptoms of cystitis, the sensitivity increased from 81 to 91% and the negative predictive value from 78 to 94% by using a threshold of at least 102 CFU/ml, the European Urology guidelines have been adapted [Lakeman and Roovers (pp. 92–97), Grabe et al.]. Nevertheless, in clinical practice, the most common used threshold is still a bacterial count of at least 105 CFU/ml. Whether low-count bacteriuria is the cause of OAB and should be treated with antibiotic therapy remains unclear. Studies evaluating the effect of antibiotic treatment for women with low-count bacteriuria are scarce and future research on this topic is required to determine which patients potentially benefit from antibiotic therapy.
In the recent years, the composition and diversity of particular microbiomes can be assessed far more accurately by molecular tools such as the quantitative PCR of 16S rRNA genes. Next to the identification of pathogens causing disease, these techniques have also led to new insights in the diversity of the bacterial communities that colonize the human body in the physiologic situation. The sequence analysis of 16S rRNA has now been used to determine the microbiota composition in numerous tissues in the healthy human body. The Human Microbiome Projects (http://commonfund.nih.gov/hmp/) have begun to catalogue the core microbial composition of the healthy human body to determine whether changes in this microbial community affect health.
These new insights into the characteristics of bacterial communities residing within our body are, however, also pointing out our rather limited capacity to cultivate microorganisms.
After reading this issue of Current Opinion of Infectious Diseases, it is evident that our perspective on UTIs is changing. The sequencing of bacterial DNA is providing us an increasing amount of information on bacteria and species residing within the urinary tract and their interaction with the host. Our current screening methods to detect UTIs might not be sufficient. However, there is little known about the effectiveness of treatment of patients with low-count bacteriuria and OAB symptoms. Future research is necessary to determine whether undetected bacteria are contributing to currently poorly understood symptoms and complaints of the urinary tract.
We would like to thank all authors for their contributions. After reading their interesting findings, it is clear that a lot of research questions remain, especially considering the human urinary microbiome, bacterial characteristics in UTIs and the microorganism–host interaction.