Microbiology Liaison Reducing Prescription of Antibiotics in ICU

The Children's Hospital at Westmead

Abstract

A weekday liaison round of the Intensive Care Unit (ICU) by a Clinical Microbiologist aimed to improve communication and interpretation of laboratory results, appropriateness of antibiotic therapy and infection control. The service was highly acceptable to users (ICU Consultants) and involved little Microbiologist time.

Despite increased unit bed occupancy and throughput, a reduction of over 80% in the level (p<0.001) and a favourable change in slope (p=0.03) of third generation cephalosporin use was achieved. This antibiotic class has the highest risk of driving bacterial resistance. Overall broad spectrum antibiotic use in the ICU was significantly reduced, despite a modest compensatory increase in the use of antibiotics of lower resistance potential. No deaths were attributable to bacterial infections acquired in the ICU.

Aim

Prescribing in the ICU drives bacterial resistance throughout the Hospital. We aimed to improve antibiotic use, reduce the use of high-risk drugs and reduce the harm caused by ICU infections.

Nature of the Problem

Antibiotic overuse is a major contributor to bacterial resistance. Up to 60% of prescriptions for ICU pneumonia are inappropriate (Singh, 2000). For years our busy 18-bed ICU had no formal microbiology liaison process and antibiotic prescribing lacked continuity. Laboratory results were provided without clinical correlation or follow-up. Infectious disease consultation was requested infrequently. Cephalosporin antibiotics, known to drive antibiotic resistance (de Man, 2000; Graffunder, 2005) were commonly prescribed. In 2006, a patient died from fungal infection after a prolonged admission. Deficiencies in microbiology liaison were highlighted during the mortality review process.

Extent of the Problem

Antibiotic selection in the ICU is complex. Spiralling escalation in antibiotic use, with adverse resistance consequences, is common. Conversely, delays in starting adequate therapy are associated with higher mortality (Kumar, 2006) and length of stay. The number of new antibiotics in development is at its lowest point since the 1960s (Talbot, 2006). To balance the desire for broad antibiotic cover in seriously ill patients against the need to preserve bacterial susceptibility to current antibiotics, a strategy of ‘laboratory-guided de-escalation’ was proposed (Neiderman, 2006). This involves early broad spectrum treatment based on knowledge of bacteria seen in the unit, with modification after two to three days, once laboratory results are available. Our ICU consultants are only rostered for two to three days at a time, interfering with such strategies. Cephalosporin use has been high since 2006. MRSA was a major problem in 2005, followed by ESBL-producing (multi-resistant) bacteria in early 2007.

Strategic Importance

Multi-Resistant Organism (MRO) control is a NSW Health priority. The recent policy directive, PD2007_084, emphasises the importance of appropriate antimicrobial use and of antibiotic stewardship programs in hospitals. Reduced MRO transmission in the ICU will benefit the whole Hospital.

The project positively impacts on key performance measures of NSW Health’s state-wide program of reducing healthcare associated infections.

The project makes the most effective use of hospital microbiologist time - a precious resource - and thus is consistent with NSW Health’s Strategic Direction ‘Make smart choices about the costs and benefits of health services’.

Planning and Implementing Solutions

The Intensive Care, Microbiology and Infectious Diseases Departments met to discuss the scope and mechanism of the liaison service. A regular weekday round was timed to occur after daily laboratory results became available, so these could be integrated into prescribing decisions. It was decided the service should not replace the existing infectious diseases consultation service, since complex patients still benefit from in-depth bedside review by Infectious Diseases Specialists. Daily visits were felt to be important in assisting with the continuity of diagnostic and therapeutic plans. Direct consultant level dialogue was planned to avoid communication errors.

The round involves a 15 minute chart review by the Clinical Microbiologist, then a 15 minute discussion with the day’s intensivist. Junior medical staff from both services are frequently present, so the service also has a strong educational focus. The interaction is both result- and problem-based, and results in non-prescriptive, pragmatic treatment suggestions, rather than binding therapeutic decisions. The program is not a restrictive stewardship service - the disputed drug is not withheld when there is disagreement.

Typical recommendations include:

• cease antimicrobial prophylaxis after an appropriate period
• target identified pathogens
• avoid redundant therapy
• IV to oral switch
• appropriate treatment duration
• appropriate use of laboratory diagnostics
• infection control interventions, such as patient isolation and screening

Underlying the recommendations is a belief that agents of recognised high resistance potential, such as cephalosporins, should be used only when more suitable alternatives are unavailable, and then for the shortest possible duration. Patient care in ICU frequently involves additional (medical or surgical) teams, so the Microbiology-ICU discussions frequently require the engagement of these teams to reach a consensus on diagnostic or therapeutic plans.

Outcomes and Evaluation

Mean bed occupancy during the first year was 14.3, compared with 14.0 and 13.9 in previous years. Patient separations were 1175 (1070 and 1026 previously). ICU staff expressed high satisfaction. No deaths due to ICU-acquired infection occurred and unit length of stay declined. Two ESBL isolates were detected, similar to preceding years, and rapid infection control responses prevented further spread.

Data regarding monthly antimicrobial supply were analysed by segmented regression in Stata (Statacorp, College Station, Texas). An overall measure of ‘workhorse’ broad spectrum antibiotic use adjusted timentin, tazocin, cefepime, ceftazidime and broad spectrum cephalosporins for their differing standard doses.

We observed an immediate fall of approximately 80% in broad spectrum cephalosporin use (p<0.001, Figure 1) and an ongoing change in slope (p=0.003). A compensatory increase in penicillin-inhibitor agents (mostly timentin, an antibiotic of much lower resistance potential) was observed at the program’s initiation (p=0.003, Figure 2). Aggregate ‘workhorse’ broad spectrum antimicrobial use fell immediately (p=0.012, Figure 3) indicating the replacement of cephalosporin use by other agents was incomplete. These changes were sustained over time and, in some cases, became more pronounced.

Sustaining Change

A program like this is inherently dependent on the enthusiasm and commitment of participants. The educational and cultural change aspects of the program mean that effects can be sustained, even if the formal liaison process ceased. Regularly observing critically ill patients getting better on well-selected narrow spectrum therapy, doing well after IV-oral switch and not requiring prolonged antibiotic therapy has powerful messages for all staff.

Strong relationships built between ICU, Microbiology and Infection Control staff should provide some assurance that co-operation would continue, even if the daily round ceases.

Future Scope

Similar programs are possible in any hospital with a diagnostic microbiology laboratory and should benefit all participants as well as patients.

Clinical Microbiologists are well-placed to help interrupt the cycle of increasing broad spectrum antibiotic use and bacterial resistance. This may encourage a reversal of the trend to centralise diagnostic pathology services, as the regular presence and building of trust and relationships between diagnostic and clinical services has many positive outcomes.

References

• de Man, P, Verhoeven, BA, Verbrugh, HA, Vos, MC, van den Anker, JN 2000, “An antibiotic policy to prevent emergence of resistant bacilli”. Lancet, Vol. 355, No. 9208, pp. 973-8.
• Graffunder, EM, Preston, KE, Evans, AM, Venezia, RA 2005, “Risk factors associated with extended spectrum beta-lactamase organisms at a tertiary care hospital”. Journal of Antimicrobial Chemotherapy, Vol. 56, No. 1, pp. 139-45.
• Kumar, A, Roberts, D, Wood, KE, Light, B, Parrillo, JE, Sharma, S, Suppes, R, Feinstein, D, Zanotti, S, Taiberg, L, Gurka, D, Kumar, A, Cheang, M 2006. “Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock”. Critical Care Medicine, Vol.  34, No. 6, pp.1589-96.
• Niederman, MS 2006, “De-escalation therapy in ventilator-associated pneumonia”. Current Opinion in Critical Care. Vol. 12, No. 5, pp. 452-7.
• Singh, N, Rogers, P, Atwood, CW, Wagener, MM, Yu, VL 2000, “Short course empiric antibiotic therapy for patients with pulmonary infiltrates in the intensive care unit. A proposed solution for indiscriminate antibiotic prescription”. American Journal of Respiratory and Critical Care Medicine, Vol. 162, no. 2, pt. 1, pp. 505-511.
• Talbot, GH, Bradley, J, Edwards, JE Jr, Gilbert, D, Scheld, M, Bartlett, JG 2006, “Bad bugs need drugs: an update on the development pipeline from the Antimicrobial Availability Task Force of the Infectious Diseases Society of America”. Clinical Infectious Diseases, Vol. 42, No. 5, pp. 657-68. 

Contact

This project was entered in the 2008 NSW Health Awards, Make Smart Choices about the Costs and Benefits of Health Services category.

More about the 2008 NSW Health Awards