Sepsis continues to be a worldwide, global problem associated with a mortality rate as high as 60%. It is estimated that severe sepsis affects over 750,000 people in the U.S. alone, each year resulting in annual costs of almost $17 billion.1 Sepsis claims more lives than breast, colorectal, pancreatic and prostate cancers combined and ranks as the leading cause of death in the non-coronary ICU.
Sepsis remains a common and expensive problem in developing in nations as well, where it is known to affect up to 30 of every 1,000 live births. It has been primarily focused on four patient populations: infants, the elderly, the immuno-compromised and critically ill patients.2 Most recently a new population has been recognized: "Baby Boomers of World War II will become the 'Sepsis Boomers' of 2010 and beyond."3
Collateral Damage
In an earlier publication of Executive Insight, we focused on the annual hospital cost of healthcare associated infections by site of infection.4 Central line-associated blood stream infections totaled 248,000 with an approximate cost of $36,000 per infection. Community- acquired infections resulted in a 13% mortality rate compared to 38 percent for hospital-acquired infections. Further, inadequate initial therapy increased the mortality rate to potentially 62 percent.
Importantly, the organisms involved have not changed either nationally or internationally, but their Mechanisms of Resistance (MOR) have, and it is the collateral damage due to Multi-Drug Resistance Organisms (MDROs) that now represents the nidus of the increase of collateral damage.
Table 1 contains a list of the top five pathogenic organisms. Remarkably, they have remained relatively unchallenged and stable over the last 20 years, globally as well as nationally. The top two are Gram positive organisms with well-established track records. Three and four are the gram negative rods that have clearly shown the greatest change and caused the greatest concern in the last 24 months. The multiplicity and means by which Gram negatives have become resistant is astounding and their MOR even more so. The new names are ESBLs, CRKPs, KPCs, Amp C, CTM-X and CRAM.
The ever-present Candida albican represents a significantly different problem in that it is a eukaryote and its 18S RNA is not detectable by the standard molecular methods that employ a 16S RNA for prokaryotes for bacterial genus/species detection. Hence, laboratory methods of molecular detection do not recover C. albicans or other species in the genera which is a significantly growing problem for the immuno-suppressed, the elderly and chronically ill.
Sepsis & SIRS
Your medical staff must recognize that sepsis is the end-stage of a complicated cascade associated primarily with immune function/dysfunction and can occur rapidly or over a longer period of time. This complex interaction of the innate immune response to inflammation has been termed the Systemic Inflammatory Response Syndrome (SIRS). Sepsis is defined as SIRS resulting from an infection of bacterial, viral, fungal or parasitic origin. Severe sepsis is associated with at least one acute organ dysfunction, hypoperfusion or hypotensive event. SIRS is defined by the presence of more than one of the four clinical criteria:
1. Body temperature greater than 38° C or less than 36° C,
2. Heart rate greater than 90 beats per minute,
3. Respiratory rate greater than 20 breaths per minute or hyperventilation with a PaCO2 < 32 mm Hg and/or
4. White blood cell count greater than 120,000 mm3 or less than 4,000 mm3 or with less than 10 percent immature neutrophils.5
Septic shock is severe sepsis with acute circulatory failure characterized by persistent arterial hypotension, unexplained by other causes. The treatments for these two overlapping syndromes are very different-antibiotics should not be given to all patients with SIRS, but only those with clinical or laboratory evidence of infection. The biomarkers listed in Table 4 utilize the SIRS cascade.
Emergence of New Platforms
Bacteremia, or the recovery of the bacterium from a blood culture, is not required for the diagnosis of sepsis. In fact, about 50 percent of the patients who are "septic" and treated for a microbial etiology never have a positive culture. There are a multitude of reasons, not the least of which is that duplicating the growth and physiology of an infected blood stream in an in vitro mechanical device is not easy.
Furthermore, bacteremic patients shower the blood stream at different times with varying concentrations, almost a sigmoid curve, and if blood cultures are drawn after the initiation of antibiotics, the recovery potential is greatly reduced.
The unusual or non-traditional microbes associated with infections of a global nature are unique, their growth requirements are different and require special adaptation.
Additionally, bacteremic patients are often colonized with organisms from a catheter or an indwelling device and these infections seed the blood stream from biofilms. The phenotype of the biofilm organism within the blood is difficult to recover; hence, the concept of Viable But Non-Culturable (VBNC) evolved and serves as a reminder of the limitations of the methodologies presently in use (Table 2).
The standard methodology has traditionally utilized growth of an organism and its metabolic convergence of sub-traits as potential targets for detection/differentiation. The most common of these are the Biomerieux VITEK®, the BACT/ALERT® and the BD BACTEC® used in more than 80 percent of U.S. laboratories. The detection time for positive cultures has been reduced using a variety of amplification steps to highlight the organism's metabolic pathway or its optical growth. In general terms, detection is within 24-48 hours, often with 8 hours, depending on the bio-burden within the patient and the prior use of antimicrobials. The costs of the instruments shown in Table 2 can be extensive, but various recovery mechanisms are available.
To enhance detection, molecular and microscopic methods have been added to compliment recovery and provide more rapid identification and potential susceptibility patterns. Coupling the automated microbiology systems with the Walk Away System Processing (WASP), recovery, identification and susceptibility can be obtained from a positive blood culture in less than 24 hours.
The newest platforms are addressing a new term, "zero-pass" detection, which is a hybrid concept that employs both advanced molecular and microscopic methods. The molecular methods are advanced, specific gene detection, while the microscopic assessment is based on Peptide Nucleic Acid Fluorescence In Situ hybridization (PNA-FISH) by AdvanDX. This FDA-approved system can directly detect Staphylococcus aureus, Enterococcus faecalis or Candida albicans in positive blood cultures.
Another method of detection is to use a "signature" or "fingerprint" of each organism within each institution, determine its bimodal population and the percent positive for a particular organism population over 72 hours. This allows us not only to identify the frequency and the time with which an organism optimally will be detected, but also the time at which blood cultures, when negative, can be used to rule out the infectious agent, allowing antibiotic therapy to be discontinued or de-escalated.
Lastly, recent research utilizing two types of mass spectrometry (MS) has given credence to the use of a tandem platform .6 MS has had a significant global impact in Asia, the Pacific Rim and Europe, but is only recently gaining acceptance in the U.S.
Costly Clinical Challenges
Recovering an organism does not always suggest its pathogenicity. Laboratories cannot differentiate a colonizer from a pathogen; the entire cascade of SIRS and sepsis needs to be correlated with bacteremia since transient bacteremia is always a possibility. Furthermore, with these detection methods, multiple isolates recovered from the same blood source will be a more common outcome. It is often difficult to determine which isolate in a mixed species infection is a significant pathogen.
Therapeutic guidelines exist and should reflect the definition of sepsis and incorporate a Rapid Response Protocol.5
The following three steps should be taken if sepsis is present:
1. Determine serum lactic values.
2. Obtain blood cultures prior to antibiotics administration.
3. Start broad spectrum antibiotics within 3 hours.
Treatment, Interventions
Antibiotics have become a "necessary evil" in selected case management. The public is aware of the consequenc es of resistance and the overuse of antibiotics. Antibiotic use is based on the initial hypothesis driven by Louis Pasteur and the principles that he developed: eliminate an organism, eliminate a disease. Unfortunately, most diseases today from the ill and/or ICU patients are associated with normal flora and the acquisition of one's own microbial flora. And this flora is a reflection of previous antibiotic therapy. Hence, many hospitals are now looking at the "green management" of many infectious diseases: no antibiotics.
An international initiative to improve the diagnosis, management and treatment of sepsis was resulted in the creation of the Surviving Sepsis Campaign (SSC). The grim statistics described earlier in this report led to the creation of SSC in 2002 with the goal of reducing sepsis-related mortality by 25 percent during the subsequent five years. First, comprehensive guidelines were developed in 2004 and subsequently updated in 2008 when SSC partnered with the Institute for Health Care Improvement (IHI) to develop a Sepsis Care Quality and Program utilizing treatment "bundles" analogous to the IHI Guidelines for Ventilator-Associated Pneumonia (VAP).7 The Sepsis Resuscitation Bundle was based on key elements of the SSC's evidence- based sepsis guidelines and included the pre-key elements described briefly about collection of blood culture specimens, serum lactate, and initiation of broad spectrum antibiotics.8
The SSC-IHI Guidelines are still unrecognized. At the 2010 Critical Care Nurses Annual Meeting, a survey showed that 25 percent of critical care nurses were unaware of any hospital guidelines for rapid blood stream detection and the majority of nurses were unaware of the newer instrumentations and methods.
Table 3 highlights some comparisons between standard and emerging therapeutic methods. Historically standard antibiotic therapy was based on individual physician assessment of the individual patient. With the significant resistance problems, patient care in the ICU involved guidance from the pharmacy and therapeutics committees. Ultimately hospitals developed protocol-based, antimicrobial therapy by disease for each unit.
In our institution, we added a "crop rotation" based on the development of resistance within the ICU. Antibiotics were prescribed based on a bundle for six months, then rotated based on the disease and the primary organisms within the unit.
On the Horizon
Table 4 lists biomarkers that have evolved and identifies emerging biomarkers specific for bacterial, viral and fungal antigens. The five most commonly studied biomarkers of sepsis include: C-reactive protein, pro-calcitonin, serum Amyloid A, mannan/anti-mannan and IFN-Y-Inducible Protein 10. All have had their champions. Historically, C-reactive protein, a general acute phase reactant (measured in concentrations up to a thousand-fold in blood) has been the marker of choice from the array of non-specific indices used in medical practice, both for prediction and/or assessment of management. It has a significant track record.1
The biomarker that has generated the most enthusiasm has been pro-calcitonin (PCT). PCT is a precursor to the thyroid hormone calcitonin. It is normally produced by special ells in the thyroid gland called C-cells. It is reported to be present in very low concentrations (0.033 ng/ml in the serum of healthy individuals) and is known to increase up to a thousand-fold under inflammatory conditions, in particular systemic bacterial infections. It has been reported to rise within 2-4 hours of infection and peak 6 to 8 hours later. While persistent elevated levels are indicative of the continued presence of an infectious agent and/or sepsis. It has significant champions in a number of defined reports using specific markers using specific methodology. PCT has also been used to rule out fungal and viral infections.
PCT was not incorporated into the SSC-IHI Guidelines and has not been widely implemented in clinical practice for a number of reasons. Physicians have used it as a 'rule out' rather than a 'rule-in' biomarker. PCT has also been found to be a biomarker for lower respiratory infections often associated with VAP. Investigators have reported that decreasing levels of PCT are predictive of survival in the ICU for mechanically-ventilated patients. However, there is a continuing controversy about the methodology employed and the variation in ranges established for these various methods. Until that is resolved, controversy will surround the use of PCT as an effective biomarker.1
Serum amyloid A is an aprolith lipoprotein, reported to have potential for diagnosing sepsis. SAA is expressed up to levels of a thousand times higher within 8 to 24 hours from onset of sepsis. Compared to CRP levels, SAA levels are reported to rise faster and higher after onset and remain higher over a longer period of time. Mannan (M) and Anti-Mannan (AM) antibodies are used exclusively to diagnose invasive fungal infections due to the presence of M in the cells walls of fungal organisms. They have been evaluated in studies of Candidiasis and Aspergillosis but have not obtained the clinical impact predicted by the original investigations. A high rate of false positives and negatives for the M-assay has reduced its popularity.
Lastly, IFN-Y-inducible protein 10 (IP-10), a pro-inflammatory chemokine, is a promising biomarker for the diagnosis of viral infections. Several studies have demonstrated its ability to differentiate bacterial from viral infections and its value for monitoring viral response. But its newness and limited clinical trials have reduced its use.1
John Thomas is a professor, Pathology, West Virginia University School of Medicine, Morgantown; Linda Corum is an assistant professor, Medical Laboratory Science at WVU; and Beverley Orr is a technical supervisor, Boston Medical Center.