NHSN CAUTI as a surrogate of catheter harm

The NHSN CAUTI metric has often been used as a surrogate for patient harm related to the urinary catheter. In an attempt to reach a more accurate definition for symptomatic CAUTI, the NHSN surveillance definition was updated in January 2015 to exclude funguria and lower urine-culture colony counts. ⁹ These definitional changes resulted in marked reductions in reported NHSN CAUTI events.Reference Advani, Lee, Schmitz and Camins ^{5 ,} Reference Fakih, Groves, Bufalino, Sturm and Hendrich ^{10 ,} Reference Bardossy, Jayaprakash and Alangaden ¹¹ However, these reductions in surveillance CAUTI events do not reflect a decrease in catheter-associated bacteriuria (CA bacteriuria) or events treated inappropriately with antimicrobials.Reference Al-Qas Hanna, Sambirska, Iyer, Szpunar and Fakih ¹² Overall, the CAUTI metric may not adequately reflect infectious harm, and it is rarely associated with bacteremia.Reference Tambyah and Maki ^{13 ,} Reference Kizilbash, Petersen, Chen, Naik and Trautner ¹⁴

Due to the reliance on the presence of a positive urine culture and fever as integral elements of the surveillance definition, NHSN CAUTI events are more susceptible to changes in testing or ordering practices and the prevalence of fever.Reference Al-Qas Hanna, Sambirska, Iyer, Szpunar and Fakih ¹² Such practices may eclipse process improvement initiatives, centered on proper device need and care.Reference Mullin, Kovacs and Fatica ^{15 ,} Reference Bardossy, Williams and Jones ¹⁶ Additionally, the national focus has been on preventing NHSN CAUTI events rather than catheter-related harm. Noninfectious catheter complications, such as urethral injury, pain, or inadvertent catheter removal, have received limited attention.Reference Bardossy, Williams and Jones ^{16 ,} Reference Fakih, Gould and Trautner ¹⁷ The catheter also limits patient mobility and increases the risk of falling because it acts as a “1-point restraint.”Reference Saint, Lipsky and Goold ¹⁸ In a recent study of patient-reported outcomes after urinary catheter insertions, noninfectious complications were 5 times more likely than infectious complications. A high percentage of patients also reported restrictions of social activities and activities of daily living.Reference Saint, Trautner and Fowler ¹⁹

Promoting appropriate device utilization and care

To date, most interventions focused on CAUTI prevention have concentrated on judicious urinary catheter use. Interventions to reduce unnecessary catheter use through reminders and removal orders have decreased overall catheter complications.Reference Meddings, Rogers, Krein, Fakih, Olmsted and Saint ²⁶ Reducing device utilization is the most important strategy to prevent catheter harm, whether infectious or noninfectious. Early statewide CAUTI prevention work in Michigan reduced catheter use by providing alternatives to indwelling urinary catheters, avoiding unnecessary catheter use, and early removal.Reference Saint, Greene, Kowalski, Watson, Hofer and Krein ^{27 –} Reference Fakih, Watson and Greene ²⁹ Consequently, the AHRQ supported “On the CUSP—Stop CAUTI,” a national effort to prevent CAUTI focusing on optimizing catheter use and care.Reference Fakih, George, Edson, Goeschel and Saint ^{30 ,} Reference Saint, Greene and Krein ³¹ Other strategies that have been advocated include proper insertion and maintenance of catheters with adherence to aseptic technique, hand hygiene, and maintenance of a closed urinary collecting system.Reference Magill, Edwards and Bamberg ^{1 ,} Reference Hooton, Bradley and Cardenas ²⁴

Although decreasing catheter use represents one of the most important interventions, it may not lower NHSN CAUTI rates or the SIR in all settings. Some interventions that decrease catheter days (denominator) may select for a higher risk group of catheterized patients, thus resulting in higher rate of CAUTIs (numerator) or SIR in some settings (despite a reduction in the absolute number of events for the overall population).Reference Fakih, Greene and Kennedy ³² In a study by Wright et al,Reference Wright, Kharasch, Beaumont, Peterson and Robicsek ³³ decreasing the device utilization by 22% increased the CAUTI rate by 6%. They showed a reduction in their CAUTI rate using patient days as the denominator instead of catheter days. Hence, the current metric that calculates CAUTI rates using catheter days as a denominator may place hospitals that successfully reduce their catheter use at a disadvantage. This metric may fail to incentivize programs actively engaged in reducing device usage.

Urinalysis with reflex to cultures

Many hospitals are modifying urine culture ordering practices with “urinalysis with reflex to culture” or “reflex urine culture” protocols to avoid unnecessary use.Reference Epstein, Edwards and Halpin ³⁴ This restricts urine culture testing to when urinalysis is found to be positive for 1 or more parameters: leukocyte esterase, nitrite, white blood cells (WBCs), or bacteria. The reflex protocol has been well validated in the outpatient setting as an initial screening test in patients with UTI symptoms. The utility of this protocol in other settings is controversial. One of the challenges is the dearth of evidence on the urinalysis parameters that best predict infection. A normal urinalysis has a high negative predictive value for UTIs,Reference Stovall, Haenal and Jenkins ³⁵ whereas positive urinalysis parameters like pyuria and bacteriuria have a low positive predictive value, especially for CAUTIs.Reference Humphries and Dien Bard ^{36 –} Reference Trautner ³⁹

Reflex protocols focus on laboratory findings rather than trying to elicit informed decisions before placing the orders.Reference Tambyah and Maki ³⁷ A positive test for nitrite can indicate presence of gram-negative bacteriuria, but it cannot definitively diagnose a UTI in the absence of symptoms. Given the high rate of bladder and catheter colonization, reflex protocols are not preferred in catheterized patients unless they are targeted to patients with symptoms like flank pain or UTI symptoms after recent catheter removal. Future studies evaluating the impact of varying thresholds of pyuria in reflex protocols may help decrease unnecessary urine cultures rather than diagnose UTIs.Reference Humphries and Dien Bard ^{36 ,} Reference Fakih and Khatib ⁴⁰

Other challenges with reflex protocols include variability in specimen collection techniques. Evidence suggests that urine samples with >5–8 squamous epithelial cells may have poor yield. Current protocols do not adjust for urine contamination, quality of specimen, or transport time. The performance of a urinalysis differs by laboratory and depends on the testing standards and capacity.Reference Humphries and Dien Bard ³⁶

Stewardship of culturing

More recent interventions to reduce CAUTI have focused on “stewardship of culturing” or reducing inappropriate urine cultures in catheterized patients. This goal is driven by ongoing recognition that patients with an indwelling urinary catheter have a high incidence of colonization. The American College of Critical Care Medicine (ACCCM) and IDSA advocate that positive urine cultures in catheterized patients do not require treatment in most cases. In their joint guidance for workup of fever in a catheterized critically ill patient, they recommend urine cultures only in (1) kidney transplant recipients, (2) neutropenic patients, (3) patients with recent genitourinary surgery, and (4) patients with evidence of genitourinary obstruction.Reference O’Grady, Barie and Bartlett ⁴¹ Mullin et alReference Mullin, Kovacs and Fatica ¹⁵ showed >30% reduction in their CAUTI rates by decreasing inappropriate urine cultures in their intensive care units by 48% without a significant change in catheter utilization.Reference Fakih and Khatib ⁴⁰ This reduction in CAUTI rates may not reflect a decrease in true UTIs; rather, it highlights the sensitivity of the NHSN definition to rate of culturing.

Hence, we recommend a 2-pronged testing approach in catheterized patients.Reference Fakih and Khatib ⁴⁰ The first step is to develop an optimized process to decide when it is appropriate to obtain urinalyses and urine cultures in these patients. Examples include removal of urinalysis and culture from nonurologic preoperative screening, discouraging indiscriminate “pan-culturing” in response to fever, and excluding catheterized patients from reflex urine culture protocols. The second step is centered on physician engagement and education around the significance of bacteriuria and pyuria, and the risk of urine testing in the absence of signs and symptoms.Reference Mullin, Kovacs and Fatica ^{15 ,} Reference Fakih and Khatib ^{40 ,} Reference Keller, Feldman, Smith, Pahwa, Cosgrove and Chida ⁴² Urine cultures obtained as a part of pan-culturing can falsely elevate NHSN CAUTI events, even in the presence of an alternate source of infection due to high incidence of catheter colonization. Advani et alReference Advani, Smith and Fisher ⁴³ developed a urine culture algorithm for catheterized patients that focuses on directed evaluation of signs and symptoms before obtaining urine cultures (with exclusions for signs of pyelonephritis, obstruction or sepsis, and pregnant or neutropenic patients). Thoughtful culturing through algorithm-directed evaluation of fever based on signs and symptoms led to a 27% reduction in unnecessary urine culture orders and >30% reduction in CAUTIs at their medical center.Reference Advani, Smith and Fisher ⁴³

Under ideal circumstances, stewardship of culturing has downstream benefits on antibiotics use, decreasing antimicrobial resistance and/or resource utilization. However, caution is advised; diagnostic stewardship that limits urine cultures indiscriminately may drive providers to administer broad-spectrum antibiotics to treat for clinical suspicion alone, without the opportunity to de-escalate or discontinue antibiotic therapy in the absence of urine-culture results.Reference Morgan, Malani and Diekema ⁴⁴ Stewardship of culturing can be a clinically successful strategy in the setting of ongoing education and open communication and feedback.Reference Mullin, Kovacs and Fatica ^{15 ,} Reference Fakih and Khatib ⁴⁰ It should be integrated into daily activities of clinicians or hardwired into the electronic medical record using a computerized decision-support routine in daily work.Reference Keller, Feldman, Smith, Pahwa, Cosgrove and Chida ⁴² However, such testing approaches will be effective only if they are accepted and perceived as effective by healthcare workers. For long-term success, continued collaboration between different stakeholders and disciplines, with a supportive internal and external environment, is needed.Reference Krein, Damschroder, Kowalski, Forman, Hofer and Saint ⁴⁵

Alternatives to indwelling catheters

External collection devices (ECDs) have been advocated as alternatives to indwelling urinary catheters.Reference Hooton, Bradley and Cardenas ^{24 ,} Reference Gould, Umscheid, Agarwal, Kuntz and Pegues ⁴⁶ ECDs have been traditionally used for men, but they have also been developed for women. In the past, female ECDs had considerable challenges with containing urinary output, avoiding damage to the pubic skin or vaginal mucosa, and remaining in place. Currently, we do not have enough data about ECD indications, efficacy studies, or cost analyses of CAUTI prevention bundles that incorporate ECDs.Reference Gray, Skinner and Kaler ⁴⁷

Despite their better tolerability and design, the newer female ECDs are associated with significant cost and nursing burden from frequent replacement, which is required every 8–12 hours. Both male and female patients using ECDs are still susceptible to UTIs, even if these devices eliminate NHSN CAUTI events. Furthermore, these ECDs may be associated with adverse events such as an increased risk of skin breakdown, and they may not be suitable for patients undergoing high-volume diuresis. Using these ECDs will result in a decrease in catheter days (denominator), with a possibility of increasing CAUTI rates or SIR in some settings.

Device utilization and Standardized Utilization Ratio (SUR)

The risk of device harm is ongoing as long as the catheter is in use. Beyond focusing on reducing unnecessary catheter insertions and catheter days, utilization is a process and outcome measure that reflects the risk of infectious and noninfectious sequelae of catheter use.Reference Abrantes-Figueiredo, Ross and Banach ⁴⁸ Device utilization is an overall objective metric that captures potential catheter harm, but it does not measure appropriateness of use or duration of catheter use per patient. For example, the risk of CAUTI may be different in a unit with a single indwelling catheter for 10 days compared to a unit with 5 catheters for 2 days each, despite both units reporting the same device days (1 catheter ×10 days = 5 catheters × 2 days = 10 catheter days).Reference Fakih, Gould and Trautner ¹⁷ The recently introduced SUR helps identify whether a unit or a hospital uses more catheters than its comparators. The SUR is a ratio of observed to predicted device days and is compared to a national benchmark. ⁴⁹ The SUR adjusts for hospital- and unit-level factors and may be used as a surrogate for potential catheter harm.

Population Standardized Infection Ratio (SIR)

The SIR or “device SIR” is a ratio of observed to predicted NHSN CAUTI events and is based on the number of actual observed catheter days for a specific unit. Both device SIR and SUR provide valuable information about a patient’s risk related to a urinary catheter. However, a metric that represents population risk can combine both elements. This is calculated as the ratio of observed to predicted NHSN-CAUTI events based on the predicted device days (compared to device SIR which is based on actual device days). This metric, “population SIR,” accounts for the risk of device exposure and infection to all patients. Importantly, both device SIR and population SIR use the NHSN CAUTI definition. However, population SIR will entice teams to also focus on noninfectious complications because it rewards hospitals that focus on device utilization. Population SIR can also better account for both, changes in device utilization and CAUTI risk over time.Reference Fakih, Greene and Kennedy ³²

Catheter-Associated Bacteriuria

Catheter-associated bacteriuria is common in patients with indwelling catheters and, in its vast majority, is asymptomatic. Inappropriate antimicrobial exposure in patients with asymptomatic CA bacteriuria increases the risk for resistance, adverse drug events, and Clostridioides difficile colitis.Reference Hooton, Bradley and Cardenas ^{24 ,} Reference Trautner ^{39 ,} Reference Daneman, Bronskill and Gruneir ⁵⁰ Because the identification of CA bacteriuria is a key driver to inappropriate antibiotic use, including it as a performance metric may reduce catheter harm and support antimicrobial stewardship efforts.Reference Naik, Skelton, Amspoker, Glasgow and Trautner ^{51 ,} Reference Spivak, Burk and Zhang ⁵² Capturing CA bacteriuria can be fully automated from electronic medical records, and it can be risk adjusted similar to the standardized antibiotic administration ratio (SAAR).Reference van Santen, Edwards and Webb ⁵³ Tracking specific patients with CA bacteriuria that receive antimicrobial therapy may incorporate both diagnostic and antimicrobial stewardship, but it cannot distinguish whether antibiotics were prescribed for bacteriuria alone or an alternate source of infection, and this metric cannot be electronically captured.Reference Flanders and Saint ⁵⁴ Process measures like SUR and CA bacteriuria can bring a fresh perspective to those leading quality efforts in acute care and can motivate decision makers to prioritize resources toward avoiding overall device harm and promoting stewardship.