Cancer patients are susceptible to overwhelming invasive infections that present as severe sepsis or septic shock (Critical Care 2004;8(5):R291-298). Barrier deficiencies and immunosuppression in cancer patients lead to increased risk of bacteremia and development of septic shock. Early initiation of effective antibiotics is essential, and dramatic improvements have been made in care of patients with sepsis due to international campaigns. Recent publication of the consensus recommendations from an expert panel (JAMA 2016;315:801-810) provides new tools to health systems seeking to improve sepsis-related health outcomes. The Centers for Medicare & Medicaid Services (CMS) has initiated a quality improvement program requiring compliance to a bundle of treatments. These advances afford an opportunity to improve treatment to patients with this devastating condition.
At Seattle Cancer Care Alliance, which unites researchers and cancer specialists from Fred Hutchinson Cancer Research Center, Seattle Children's, and UW Medicine to treat cancer patients, we have streamlined an approach to deliver a care bundle of rapidly administered lifesaving medicines designed to improve sepsis-related outcomes in cancer patients.
There have been many versions of the definition of sepsis, with the most recent revision made by the Sepsis-3 expert panel process and published (JAMA 2016;315:801-810). Prior definitions of sepsis relied on the systemic inflammatory response syndrome (SIRS criteria) and suspected infection. Severe sepsis was defined by sepsis and organ dysfunction while septic shock occurred in those with persistent hypotension despite adequate fluid resuscitation. In the revised definitions (Sepsis-3), the expert panel of the Society of Critical Care Medicine and the European Society of Intensive Care Medicine determined that the term “severe sepsis” is redundant and re-defined sepsis as evidence of infection plus life-threatening organ dysfunction, caused by a dysregulated immune system. These new definitions essentially eliminate the entity of “severe sepsis” and reclassify this condition as “sepsis.” The Sepsis-3 definitions provide more specific criteria to meet the sepsis definition (with acute increase in Sequential Organ Failure Assessment [SOFA] score), and for determination of those at risk for sepsis advocate the use of a higher-performing and easier to calculate “quick SOFA” (qSOFA) to replace the non-specific SIRS criteria (JAMA 2016;315:762-774).
Emergence of Sepsis
Barrier deficiencies are common in cancer patients. The skin and lining of the gut encompass the body's normal barrier to invasive infections and regimen-related toxicities (RRT) such as mucositis, an inflammation and breakdown of the rapidly dividing cells of the mucosal epithelium of the mouth and gut, are common in patients administered cytotoxic and conditioning chemotherapy. Graft versus host disease is a reaction of the hematopoietic stem cell transplant (HSCT) graft against the host tissues and involvement of the gut is common, leading to increased permeability and microbial translocation of gut flora.
Furthermore, many patients have indwelling central venous catheters for administration of chemotherapy, which breach the skin and provide a direct portal for skin flora to enter the systemic circulation. Ability to clear bacteremia in cancer patients is compromised by immunodeficiency. Neutropenia is the most common and serious immunodeficiency in cancer patients, and especially in those with hematologic malignancies. Neutrophils are the most abundant white blood cell and an essential part of the innate immune system that combats acute bacterial and fungal infections. Neutropenia is a lack of circulating neutrophils and is defined by an absolute neutrophil count (ANC) below 500 cells/uL of blood. Neutropenia occurs due to underlying hematologic malignancies such as acute myeloid leukemia or as a consequence of treatments such as cytotoxic chemotherapy for hematologic malignancies and solid tumors or due to conditioning therapy prior to HSCT.
Care outcomes in cancer patients with septic shock have been improving over time, similar to trends seen in non-cancer patients with severe sepsis and septic shock, likely due to improvements in targeted and supportive therapies. Studies underscore the point that while related conditions on a spectrum of disease severity, these entities are in fact quite different in terms of prognosis.
Improved survival for neutropenic cancer patients with initiation of early empiric antibiotic therapy has long been recognized, with results that have been consistently reproduced. Appropriate antibiotics to cover the specific etiologic microbe are also important: a study of bacteremic hospitalized patients showed therapy in those with severe neutropenia (ANC< 100 cell/uL) with a delay in initiation of appropriate antimicrobial >24 hours resulted in 18-fold higher risk of mortality when compared to those with earlier appropriate antibiotics. In a series of studies assessing time to antibiotic administration in the UK, considerable delays were noted, even after quality improvement processes were initiated. As a best practice performance standard, oncology guidelines recommend the prompt administration of empiric antibiotics within one hour from the triage of a febrile neutropenic patient.
Appropriate empiric antimicrobial therapy for cancer patients with neutropenic sepsis is controversial with monotherapy generally accepted as superior for uncomplicated neutropenic fever, but combination antimicrobial therapy may be superior for those with severe disease. Typical practices include risk-stratifying patients with neutropenic fever to low or high risk to be managed as outpatient or inpatients, respectively. In higher risk patients, it is generally accepted to give an antipseudomonal beta-lactam antibiotic infusion as monotherapy when fever is uncomplicated and without localizing symptoms. In cancer patients with neutropenic fever, a Cochrane meta-analysis demonstrated that initial empiric beta-lactam monotherapy was superior compared to beta-lactam and aminoglycoside combination therapy both in decreased infection-related mortality and in less toxicity.
A similar meta-analysis from the same group found no difference in non-cancer patients with sepsis treated with beta-lactam monotherapy versus beta-lactam and aminoglycoside combination therapy. However, these studies did not focus on the most severely ill patients, i.e., those with severe sepsis and septic shock. In a large multicenter propensity matched retrospective study, researchers showed lower mortality with combination therapy when compared to monotherapy. This same group also performed a meta-analysis showing that, in the highest risk patients with septic shock, combination antimicrobial therapy improved outcomes.
Recognizing the lack of specific data, the Infectious Diseases Society of America and National Comprehensive Cancer Network guidelines extrapolate from other patient populations to suggest addition of aminoglycosides, fluoroquinolones, and/or glycopeptide (vancomycin) for complications such as hypotension, pneumonia or where high local prevalence of antimicrobial resistance exists. Confirming common practices of modifying and de-escalating therapy when results of microbial cultures are available, a recent trial showed de-escalation of empirical antimicrobial therapy in those who respond to empirical regimens was not associated with worse prognosis.
International efforts aimed at improving sepsis outcomes have ensued following a landmark clinical trial of bundled care for SS/SSh. The Surviving Sepsis Campaign guidelines have provided operational advice and data analyses have validated the bundled approach to care (Intensive Care Medicine 2015;41:1620-1628, Critical Care Medicine 2015;43:3-12). The National Quality Forum recommended and the CMS has adopted a new reporting measure called the Early Management Bundle, Severe Sepsis/Septic Shock (SEP-1). This complex reporting mechanism requires hospital compliance with bundle elements within three and six hours of sepsis presentation. Bundle elements deemed to be essential for care of the septic patient include fluid resuscitation, blood cultures and lactic acid obtained, and antibiotic administration.
Seattle Cancer Care Alliance sees many cancer patients with hematologic malignancies, including those who undergo HSCT and are otherwise immunocompromised. Understanding the need for emergent administration of effective antibiotics and other supportive care in our cancer patients who develop sepsis and septic shock, we developed a care bundle that standardizes best practices to deliver life-supporting interventions in an emergent fashion while the patient is still in the outpatient clinic or hospital floor and prior to transfer to higher level care. Antibiotics on a pre-printed order form are “on-the-shelf” ready, infusion compatible for simultaneous administration and rely on standardized, non-weight based dosing. Other supportive care includes IV fluid and stress-dose steroid administration and cultures obtained prior to antibiotic infusion.
Cancer patients are at high risk of sepsis and septic shock, and tools are available to reduce risk and improve care for sepsis. Clinics, emergency rooms, and acute care or intensive care floors that take care of cancer patients should harness the efforts of hospital quality improvement departments working on CMS sepsis bundle reporting to standardize care processes in algorithmic ways for cancer patients. An algorithmic approach to care of sepsis, a medical emergency in cancer patients, is likely to yield excellent results in reducing this condition.