Sepsis: use of clinical criteria as well as novel diagnostic tests aim to improve patient outcomes

Sepsis, as defined by Stedman’s Medical Dictionary, is “the presence of various pathogenic organisms, or their toxins, in the blood or tissues.”¹ By necessity, the physician must understand sepsis in its clinical nature, and quickly learns to make the distinction between sepsis and bacteremia, which does not carry the same grave prognosis or demand the same urgency. It is true that, if left unchecked, bacteremia may bring about deleterious effects on the host, including endocarditis and osteomyelitis, and require extensive treatment; the term sepsis, however, refers to a syndrome characterized by an inappropriate host response provoked by an invasive infection. 

When a clinician is confronted with a patient who is possibly septic, multiple criteria and diagnostic tests are used to reach this diagnosis. As the clinical picture may shift rapidly, bringing about sudden decompensation of the patient, the clinician must be prepared to act even before the results of ordered tests are available. This empiric treatment is then adjusted accordingly based on both the patient’s response and the data as it is obtained. Here, we review findings common with sepsis, and the diagnostics that allow us to tailor the treatment and allow for the best possible outcome.

In 1991, and again ten years later, two large conferences were held to discuss sepsis and criteria necessary when making its diagnosis.^2,3 From these conferences, the terms SIRS (systemic inflammatory response syndrome) and PIRO (predisposition, infection/insult, response, and organ dysfunction) were generated, and they are used by many in grading the severity of a clinical picture complicated by sepsis.⁴ Regardless of the degree of sepsis, at least two or more of the following are required to make the diagnosis: fever, tachycardia, tachypnea, and leukocytosis. Sepsis is notably more severe when it is accompanied by organ dysfunction due to hypotension, which leads to hypoperfusion of organs. This end-organ injury may be made apparent as decreased urine output, altered mental status, or lactic acidosis, and in its more severe state, is associated with multiple organ failure (multiple organ dysfunction syndrome).

With more than 750,000 cases of sepsis occurring annually, and more than 200,000 deaths each year attributable to sepsis, it is important that this condition is promptly recognized.⁵ An inadequate approach to working up a septic patient may range from the clinician failing to recognize when infection is present, to administering inappropriate empiric antibiotics, to performing an inadequate examination, and even to sending off incorrect laboratory tests or failing to order the correct tests, such as gram stain/culture, which will aid in targeting empiric treatment choices. Equally serious and potentially devastating is the failure to perform surgical intervention in a timely manner when required for adequate treatment. 

Once a differential has been narrowed down to an infectious agent as the underlying cause for the patient’s presenting condition, an appropriate antibiotic may be selected based on a few generalities. In obstetrical or gynecological patients, for example, the anatomic location of the infection and the time of onset to recognition of the infection both play important roles. For abdominal incisions, e.g., following cesarean delivery or hysterectomy, gram-positive organisms such as staph aureus and strep pyogenes as well as streptococcus agalactiae are important to consider, along with gram-negative facultative anaerobes. We have seen in our institution an increase in the number of post-operative wounds infected with methicillin-resistant staphylococcus aureus (MRSA), and it is important to include coverage against this organism. It should also be noted that the importance of obtaining culture and gram stain of infected tissue prior to the initiation of empiric treatment cannot be overemphasized.⁶ 

Occasionally, infection may worsen in spite of the selection and administration of appropriate antibiotic therapy. The development of a pelvic abscess or involvement of an implant may not allow for adequate penetration of antimicrobial treatment to the site of the infection. Once bacterial counts reach 108-109 colony-forming units per gram of tissue, the growth rate of the infectious agent slows, and the organism becomes less susceptible to certain antibiotics. This has been termed the Eagle effect, after H. Eagle proposed this as a mechanism of treatment failure in a group of mice inoculated with Group A streptococcus and treated with penicillin.⁷ Once a patient becomes critically ill, and severe sepsis ensues, appropriate resuscitative efforts must be expeditiously undertaken, including a return to the operating room if deep infection is present. 

In order to assist the clinician in quickly distinguishing sepsis from a milder systemic inflammatory response generated against a noninfectious cause, several biomarkers have been proposed. Recently, procalcitonin has emerged as a suitable candidate for aiding in the ability to accurately diagnose sepsis. This 116-amino acid polypeptide is induced in a relatively short time after provocation by a bacterial stimulus, and its long half-life allows levels to be measured serially, which may allow for the tailoring of antibiotic treatment.⁸ Use of an individual biomarker may allow for a simpler mechanism for determining if sepsis is present than the use of panels of tests examining chemokine/interleukin profiles present in helper T-cell profiles (Th17).⁹ 

Equally important in determining when sepsis is present and its resolution or response to treatment is the rapid identification of the specific pathogen(s) present and their susceptibility to clinically relevant antimicrobials. While gram stain of infected fluid/tissue may be done rapidly and may allow for categorization of a suspected infectious agent, culture and then follow-up determination of antimicrobial resistance may take several days. Recently, the Infectious Diseases Society of America (IDSA) has led a strong push to foster the development of new diagnostic tests that will be easy to use and provide results within an hour.¹⁰ Caliendo et al. state that the ideal diagnostic test is accurate; relies on heat-stable reagents with an extended shelf-life; is easily portable; requires minimal technical skills; is rapid (<1 hr), sensitive, and specific; does not require being run in large batches; is cost-effective; and is suitable for use in a broad range of clinical samples. They say that currently PCR-based tests meet some but not all of these criteria.¹⁰ 

In work that my colleagues and I at the University of Texas Health Sciences Center have done on group B streptococcus (GBS), we have developed an antibody-based test (termed the N-Assay) which allows for the simultaneous identification of a microbial pathogen and determination of its antimicrobial susceptibility. By substituting antibodies specific for varying pathogens, one may selectively identify pathogens present in a patient. A similar assay that we developed in 2010 was tested earlier this year in more than 300 patients, and it was found that GBS could be reliably detected in much less time than is typically required for culture.¹¹ We have converted this test to an ELISA format, and as such, have seen preliminary results that allow for identification of GBS in under an hour. Furthermore, by changing the specificity of the capture antibody, we are able to detect vancomycin-resistant enterococcus (VRE), Neisseria gonorrhea, and E.coli, as well as Candida albicans. Initial results with VRE have consistently shown that susceptibility to selected antibiotics may be reliably determined after a brief period of culture. 

Every effort must be made to stem the development of sepsis. More rapid diagnostic tests will play an increasingly important role in aiding the clinician not only in making the diagnosis, but in tailoring the course of treatment. In addition to improved tests, the clinician will continue to play a central role in determining whether the course of action requires aggressive resuscitation and broad spectrum antibiotics, as well as operative management and aggressive debridement.