Sepsis National Hospital Inpatient Quality Measure (SEP-1): Multistakeholder Work Group Recommendations for Appropriate Antibiotics for the Treatment of Sepsis

Abstract

The Center for Medicare and Medicaid Services adopted the Early Management Bundle, Severe Sepsis/Septic Shock (SEP-1) performance measure to the Hospital Inpatient Quality Reporting Program in July 2015 to help address the high mortality and high cost associated with sepsis. The SEP-1 performance measure requires, among other critical interventions, timely administration of antibiotics to patients with sepsis or septic shock. The multistakeholder workgroup recognizes the need for SEP-1 but strongly believes that multiple antibiotics listed in the antibiotic tables for SEP-1 are not appropriate and the use of these antibiotics, as called for in the SEP-1 measure, is not in alignment with prudent antimicrobial stewardship. To promote the appropriate use of antimicrobials and combat antimicrobial resistance, the workgroup provides recommendations for appropriate antibiotics for the treatment of sepsis.

According to the Healthcare Cost and Utilization Project Statistical Brief number 204, sepsis was the most expensive condition treated across all payers and the second most common reason for hospitalization in 2013 [1]. Antibiotic-resistant infections have added to the challenges of optimally managing sepsis, causing 2 million illnesses and approximately 23000 deaths each year, in the United States [2]. Given the costs as well as the high mortality rate (182242 sepsis deaths in 2014), a focused effort is needed to improve sepsis outcomes.

In a 2001 landmark study, Rivers et al demonstrated that early goal-directed therapy for the treatment of severe sepsis and septic shock in the emergency department made a significant impact on mortality [3]. Resultant of the findings and with support from the Surviving Sepsis Campaign, Dr Rivers and his respective institution, Henry Ford Hospital, collaborated with the Society of Critical Care Medicine (SCCM), the Infectious Diseases Society of America (IDSA), and emergency physicians to develop the Severe Sepsis and Septic Shock: Management Bundle (SEP-1) quality measure [4]. With the purpose of supporting the efficient, effective, and timely delivery of high-quality sepsis care, the SEP-1 quality measure was added to the Hospital Inpatient Quality Reporting (IQR) Program in July 2015 by the Centers for Medicare and Medicaid Services (CMS), with the performance period starting in October 2016 [5].

SEP-1 is a chart-abstracted composite measure that calculates performance based off completion of multiple interventions according to the SEP-1 measure specifications detailed in the Specifications Manual for National Hospital Inpatient Quality Measures published by the Joint Commission. The SEP-1 measure specifications include, among other vital aspects of sepsis care, the timely administration of a specific listed broad-spectrum antibiotic [6]. This requirement is a major point of concern for multiple medical specialty societies and led to an August 2015 joint letter by the American College of Emergency Physicians (ACEP), SCCM, and IDSA to CMS [7] that highlighted the unintended consequences of requiring the administration of specific listed broad-spectrum antibiotics to severe sepsis patients who present with a known infection that would respond better to a specific antibiotic or antibiotics not in the included tables. Addressing those concerns, in July 2016, CMS made amendments to SEP-1’s “Broad Spectrum or Other Antibiotic Administration Selection” data element, to include the following language:

“If an IV antibiotic from Table 5.0 or an appropriate combination of IV antibiotics from Table 5.1 is not started or given within the 3 hours following presentation of severe sepsis, but there is a lab report or physician/APN/PA documentation indicating the causative organism and susceptibility is known (see exception for Clostridium difficile) and an IV antibiotic identified as appropriate to treat the causative organism is given within 3 hours following presentation of severe sepsis, choose Value ‘1’.” [6]

We appreciate the action CMS has taken to allow a process for physicians to exercise individual clinical judgment in the treatment of septic patients. We agree with the need for SEP-1 and the requirement therein for the timely administration of appropriate antibiotics to patients with sepsis or septic shock. We believe, however, that not all of the antibiotics outlined in SEP-1 (Table 5.0: Monotherapy Antibiotics, Sepsis, Combination Antibiotic Therapy Table; Table 5.1: Antibiotic Generic/Trade Name Crosswalk) are appropriate, and that the use of antibiotics outlined in the SEP-1 measure are not consistently in alignment with prudent antimicrobial stewardship. We also recognize the difficult situation CMS must address to decrease the high mortality rate and costs associated with sepsis. There is a delicate balance to strike between utilizing clinical quality measures to standardize care according to best practices while also allowing physicians to provide apply their clinical expertise to address unique clinical situations they face.

While this exception is necessary and does not penalize hospitals or physicians for administering an antibiotic not listed in the SEP-1 antibiotic tables, further effort is needed to promote the practice of antimicrobial stewardship to combat the growing issues of antibiotic resistance, adverse drug events, and increased risk for C. difficile infections. This need has been highlighted by the National Action Plan for Combating Antibiotic-Resistant Bacteria developed by the White House in March 2015 [8] as well as the proposed rule for the Medicare and Medicaid Hospital Conditions of Participation published in June 2016 [9].

MODIFICATION OF ANTIBIOTIC TABLES FOR SEP-1 WORKGROUP PROCESS AND RESULTS

Recognizing the need to promote the judicious use of antibiotics in the treatment of sepsis, IDSA convened a multistakeholder workgroup in May 2016 to address the aforementioned concerns. The Modification of Antibiotic Tables for SEP-1 Workgroup (MATS WG) was composed of 1 representative from ACEP, 2 from SCCM, 1 from the Society of Hospital Medicine, and 2 from IDSA with the goal to strengthen patient safety by promoting the optimal use of antibiotics. The specific objectives of the MATS WG were to evaluate and revise as necessary the content of the antibiotic tables outlined in SEP-1 to enable the appropriate empiric treatment of sepsis.

To evaluate the current antibiotic tables specified for SEP-1 (Table 5.0: Monotherapy Antibiotics, Sepsis, Combination Antibiotic Therapy Table; Table 5.1: Antibiotic Generic/Trade Name Crosswalk, Sepsis), the MATS WG members were asked to complete a survey providing their approval or disapproval for each antibiotic listed in Table 5.0 and each antibiotic drug class combination listed in the Combination Antibiotic Therapy Table. In addition to disseminating the survey to the MATS WG, feedback was sought from interested non-workgroup stakeholders within each respective professional society.

A total of 12 respondents completed the survey. For Table 5.0, a total of 14 antibiotics were reviewed. Of the 14 monotherapy antibiotics, 9 received majority approval by the respondents, 3 antibiotics were not approved (ampicillin-sulbactam, levofloxacin, moxifloxacin), and 2 antibiotics (ertapenem, ticarcillin-clavulanate) received an equal amount of votes for approval and disapproval. The survey respondents cited a common concern of large amounts of antibiotic resistance as to why ampicillin-sulbactam, levofloxacin, and moxifloxacin were not approved. Escherichia coli resistance to ampicillin-sulbactam has been observed in numerous studies [10–13]. Levofloxacin has been shown to be efficacious against many pathogens, namely, pneumonia caused by the Haemophilus species [14]; however, levofloxacin resistance has been observed in Haemophilus influenzae [15, 16], Pseudomonas aeruginosa, and E. coli [17]. Similarly, moxifloxacin has been shown to have very high rates of resistance in C. difficile isolates [18]. As for the antibiotics that resulted in a draw, the MATS WG was polled again to reach consensus on only ertapenem, as ticarcillin-clavulanate (Timentin) was discontinued due to safety issues [19]. Ertapenem was ultimately approved for inclusion by the workgroup. Respondents were also asked to provide suggestions for monotherapy antibiotics not listed in Table 5.0. The MATS WG proposed 2 antibiotics for inclusion in the revised Table 5.0 (ceftazidime-avibactam, ceftolozane-tazobactam).

The Combination Antibiotic Therapy Table allows for a total of 21 antibiotic class combinations. Of the total combinations, 10 were approved by the majority of respondents. The antibiotic class combinations not approved for the empiric treatment of sepsis consisted of any combinations with macrolides or ciprofloxacin, as well as the specific combinations of aztreonam with cephalosporins and aminoglycosides with intravenous clindamycin. Macrolide antibiotics, predominately administered as the first line of treatment for community-acquired respiratory tract infections caused by Mycoplasma pneumoniae, have been observed to have high rates of resistance in M. pneumoniae in Asia and moderate levels of resistance in Europe and the United States [20]. Furthermore, resistance to macrolides in Streptococcus pneumoniae has been steadily increasing since the 1990s [21]. Last, the macrolide antibiotic telithromycin (Ketek), specified in Table 5.1 for SEP-1, has been discontinued due to business reasons [22]. Ciprofloxacin has also seen an increase in clinical ineffectiveness, particularly in nosocomial gram-negative bacteremia [23]. Additionally, as part of the larger antibiotic class of fluoroquinolones, ciprofloxacin resistance has remained high among many isolates, including methicillin-resistant Staphylococcus aureus, P. aeruginosa, E. coli, anaerobes, and other pathogens found in intensive care units [24]. With the efficacy of available antibiotics diminishing and the dearth of development of new antibiotics, restricting the use of effective antibiotics with relatively low resistance rates is imperative to preserve an efficacious antibiotic pipeline. Concerns regarding increasing resistance and adequate coverage of pathogens were cited as reasons why the combinations of aztreonam [25] with cephalosporins [26] and aminoglycosides [27] with intravenous (IV) clindamycin [28] were not approved by the working group. Moreover, the combination of aminoglycosides and clindamycin IV has been suggested to be less effective in the treatment of intra-abdominal infections when compared to β-lactam antibiotics as well as concerns with possible adverse effects that include nephrotoxicity, ototoxicity, and pseudomembranous colitis [29].

MODIFICATION OF ANTIBIOTIC TABLES FOR SEP-1 WORKGROUP RECOMMENDATIONS

Resulting from the work of the MATS WG, we respectfully provide our recommendations for Table 5.0: Monotherapy Antibiotics in Table 1, the Combination Antibiotic Therapy Table in Table 2, and Table 5.1: Antibiotic Generic/Trade Name Crosswalk in Table 3. The MATS WG recommendations are provided next to the SEP-1 antibiotic specifications detailed in the Specifications Manual for National Hospital Inpatient Quality Measures Discharges 1 July 2016 (3Q16) through 31 December 2016 (4Q16) version 5.1.

Revisiting the amendments CMS has made to the SEP-1 measure specifications, we believe the addition of cefotetan to the Table 5.1 (Antibiotic Generic/Trade Name Crosswalk) is not appropriate. Cefotetan is mainly administered as a surgical prophylactic and has resulted in high rates of surgical site infections [30]; in addition, the IDSA has recommended that cefotetan not be used to treat community-acquired intra-abdominal infections in adults due to increasing resistance [31]. As a result, we respectfully recommend that CMS remove cefotetan from Table 5.1.

We appreciate the effort CMS has made to address the serious issue of sepsis and hope that consideration will be given to our proposed modifications. We welcome further discussion with CMS and other stakeholders on these matters related to the SEP-1 measure.

Note

Potential conflicts of interest. C. M. C. was president of the Society of Critical Care Medicine in 2015 and a payment was made from this organization to Emory University for time spent in this role. The authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

References