Reducing overall antibiotic exposure is likely to have an impact on decreasing antimicrobial resistance.
Antimicrobial stewardship programs often focus on antimicrobial de-escalation, while less emphasis is placed on treating for the shortest possible duration. Although both are important stewardship initiatives, there is a paucity of data suggesting that de-escalation directly results in less emergence of resistance. Furthermore, antimicrobial de-escalation has not been found to reduce duration of antibiotic therapy.1 However, reducing overall antibiotic exposure is likely to have an impact on decreasing antimicrobial resistance. By shortening antibiotic duration, overall antibiotic exposure is lessened, reducing selection pressure for multidrug resistant organisms. This finding is supported by studies of ventilator-associated pneumonia (VAP).2,3
Hospital-acquired pneumonia (HAP) and VAP account for significant antibiotic use and together are one of the most common health care—associated infections,4 These infections have a large economic burden and are associated with poor patient outcomes, including prolonged mechanical ventilation and hospital stay.5 A justified concern about the poor outcomes that HAP/ VAP cases represent drives antibiotic use because even suspected-but-unproven cases receive broad-spectrum antibiotic therapy.
Historically, HAP and VAP were treated for 14 to 21 days. Although treating these infections for an extended duration seems logical given their severity and the negative patient outcomes, prolonged courses of antibiotics have not been found to improve outcomes but instead to increase the rate of recurrent infections with multidrug- resistant (MDR) organisms.2,3 In 2005, guidelines by the Infectious Diseases Society of America (IDSA) for the treatment of HAP/VAP recommended treating for either 7 days or 14 days, depending on the causative pathogen. When the causative organism was a non-lactose fermenting gram-negative bacillus (NLFGNB), longer durations were recommended.6
Two well-designed, randomized controlled studies have been conducted to evaluate the equivalence of shorter and longer courses of antibiotic therapy in VAP, both comparing 8 and 15 days. The first study was in 401 patients with microbiologically proven VAP. Patients were randomized to 8 days or 15 days of antibiotics selected at the discretion of the treating clinicians. The rate of recurrence and 28-day mortality were similar in both groups: 18.8% (8 d) versus 17.2% (15 d) and 28.9% (8 d) versus 26% (15 d), respectively, and 8 days was found to be noninferior. A finding of this study was that in patients with NLFGNB VAP, there was a higher rate of infection recurrence in the 8-day group (40.6% vs 25.4%), yet there was no difference in in-hospital mortality: 34.4% and 41.3% for 8 and 15 days, respectively. Not only were clinical outcomes similar between groups, but reducing overall antibiotic exposure resulted in a significant reduction of recurrent MDR pulmonary infections. Recurrent pulmonary infections with an MDR organism occurred less frequently in the shorter-treatment group: 42% and 62% for 8 and 15 days, respectively.7 In the second study, 225 patients with early-onset VAP were randomized to receive 8 or 15 days of a beta-lactam plus an aminoglycoside. This study also did not find a difference in clinical cure between the two treatment durations. Clinical cure was 85.3% in the 8-day group and 84.4% in the 15-day group. Similarly, there was no difference in 21-day mortality, which was 8.6% in the 8-day group and 8.3% in the 15-day group.8 One surprising finding was that the rate of secondary infections was higher in the 8-day treatment group (35.3% vs 19.3%; P <.01).
Although secondary infections were more common with short courses, there was no difference in mortality. This difference was explained by the fact that since patients were followed for 21 days, the 8-day group had more time to develop a secondary infection because they were off antibiotics more than twice as long as the 15- day group.8 Patients were not followed after the 21-day period to evaluate for further resistant infections associated with antibiotic use, such as C. difficile infection. A finding that was not surprising in both studies was that there were more antibiotic-free days in the short-course groups, taking into account antibiotics used for the treatment of secondary infections.7,8 In these 2 studies, a shorter-course of antibiotics did not negatively impact patient outcomes or mortality. In fact, there was the added benefit of less overall antibiotic exposure and although patients were not assessed for the development of a C. difficile infection, presumably this should decrease the risk of infection.
There have been other attempts to shorten antibiotic therapy that do not fall into the prospective, randomized studies included above. Using objective measures to guide antibiotic duration instead of arbitrarily deciding treatment durations would be optimal for clinicians to individualize treatment based on the stability of the patient, resulting in earlier antibiotic discontinuation for many patients. One study used clinical pulmonary infection scores (CPIS) in patients with radiographic findings suggestive of nosocomial pneumonia to stratify patients at low risk of complications due to short courses of antibiotics. In this study, patients with a low CPIS were randomized to receive either ciprofloxacin for 3 days or standard care (antibiotics for 10-21 days). More severely ill patients were excluded. Clinical outcomes were similar between the 2 groups. As can be expected, in the short-course group, there were significantly more antibiotic-free days, and in the long-course group, there was more resistance and/or super-infections.9 This study was performed in 2000 and is now dated, but the principle is still relevant.
In suspected VAP, using ventilator settings as a surrogate for degree of illness may also be a method to transition patients off empiric antibiotics earlier. In a retrospective study, patients with stable and minimal ventilator settings empirically treated for suspected VAP who received less than 3 days of antibiotics had similar duration of mechanical ventilation, mortality, and time to discharge compared with a longer-duration group (greater than 3 days). However, the majority of patients in this study most likely did not truly have VAP, and therefore this strategy should be used only for stable patients for whom the diagnosis of VAP is uncertain.10
Biomarkers such as procalcitonin have also been of interest as a marker to guide antimicrobial duration. There is no clear consensus on the use of procalcitonin to diagnose bacterial infections and data conflicts on whether strategies that utilize this biomarker successfully result in shorter antibiotic durations. A metaanalysis of 14 studies including 4211 patients with acute respiratory infections found that overall, the duration of antibiotics was shorter in the procalcitonin-guided group (7 days vs 10 days; P <.0001). Treatment duration was also shorter in the subset of patients with VAP (11 days vs 14 days; P = .02). Although procalcitonin may have the potential to help guide antimicrobial decisions in HAP/ VAP, these durations of therapy are longer than recommended. It is unclear whether procalcitonin is as useful for antibiotic decision making at earlier time points.11
Shortening antimicrobial duration and thereby reducing overall antibiotic exposure is an effective antimicrobial stewardship strategy to slow the emergence of resistance. Current IDSA HAP/VAP guidelines support short-course treatment regardless of the organism and instead favor individualizing treatment and extending therapy when needed based on imaging and/or clinical findings.12 Overall, shorter-treatment durations are associated with more antibiotic-free days and fewer MDR secondary or recurrent infections and do not negatively affect clinical outcomes.3 In this era of antibiotic resistance, reducing overall antibiotic exposure may be a more easily implemented than programs focused on preventing broad-spectrum antibiotic use.
Dr. Hanretty is a clinical pharmacist in infectious diseases at St. Christopher’s Hospital for Children in Philadelphia, Pennsylvania. She completed a PGY1 Pharmacy Residency at UCSF Medical Center in San Francisco, and a PGY2 Residency in Infectious Diseases Pharmacy at Temple University. She is an active member of SIDP.
Dr. Gallagher is a clinical professor at Temple University School of Pharmacy and clinical pharmacy specialist in infectious diseases at Temple University Hospital, both in Philadelphia. He also is the director of the PGY2 Residency in Infectious Diseases Pharmacy at Temple. He is an active member of SIDP.