MOF emerged in the early 1970s as a result of advances in ICU technology that enabled patients to survive single organ failure1
. Its epidemiology continues to change as our management strategies evolve from ongoing research (). In the late 1970s, reports by Polk, Eiseman, and Fry promoted the belief that MOF was the ’fatal expression of uncontrolled infection‘, and focused research on infection-driven MOF1
. Over half of MOF cases were associated with intra abdominal infection (IAI). Throughout the 1980s, IAI became less problematic due to advances in surgery and intensive care including: a) better initial management of abdominal trauma and IAI, b) more potent and appropriately dosed perioperative antibiotics, c) earlier diagnosis of postoperative IAI with advanced computerized axial tomography, and d) effective percutaneous abscess drainage using interventional radiology.
The Evolving Clinical Epidemiology of Multiple Organ Failure and PICS
A series of reports from Europe by Faist, Goris, and Waydhas convincingly demonstrated that MOF was a frequent occurrence in blunt trauma patients without infection2–4
. The term ’sepsis syndrome‘ became the vernacular to describe patients who appeared to be septic but had no obvious source of infection. The central question became: ’What is the driving mechanism of this deleterious inflammation?’ Given that shock was a consistent early event, whole body ischemia-reperfusion was an attractive explanation. Alternatively, epidemiologic studies by Border and others strongly implicated the gut as the occult source of bacteria that drove the sepsis syndrome5
. Indeed, prospective randomized controlled trials (PRCTs) testing selective gut decontamination and early enteral nutrition found decreased rates of nosocomial infection (principally pneumonia) with these gut-directed therapies6
. These clinical observations were supported by the experimental work of Alexander and others that persuasively focused attention on bacterial translocation as a unifying mechanism that characterized non infectious induced MOF7
In the late 1980s, Shoemaker popularized an alternative hypothesis that ’early unrecognized flow dependent oxygen consumption‘ was a prime cause of non infectious MOF8
. Pushing oxygen delivery (DO2
) to ‘supranormal’ levels during initial resuscitation became standard of care. Simultaneously, trauma system triage, Advanced Trauma Life Support (ATLS) and ’damage control‘ surgery were universally adopted. As a result, severely injured patients were triaged to designated trauma centers, underwent damage control surgery and in the ICU, received supranormal DO2 resuscitation. Fewer patients exsanguinated and more survived to ICU admission. Many developed abdominal compartment syndrome (ACS), which emerged as an epidemic in the mid 1990s and was subsequently shown to be another deadly MOF phenotype9
By the mid 1990s, ‘Sepsis syndrome’ had evolved into the systemic inflammatory response syndrome (SIRS). SIRS was presumed to be inherently beneficial; however, if exaggerated or perpetuated, SIRS could precipitate early MOF, independent of infection. It was not until the first decade of the new millennium that Polly Matzinger provided an explanation for SIRS in absence of obvious microbial infection: the host responds to noninfectious insults and tissue injury by releasing endogenous mediators that are ‘danger signals’10
. Tissue damage, per se, releases ‘alarmins’ and ‘damage-associated molecular pattern’ (DAMP) molecules that can stimulate innate immunity through TLR receptors or other sensing systems11, 12
. Thus, noninfectious insults can elicit exaggerated inflammation through the same pathway(s) as microbial pathogens and produce similar SIRS.
Also in the mid 1990s, analysis of the Denver MOF database revealed that MOF occurred ‘early’ or ‘late’ in the surgical ICU course13
. Two different patterns of SIRS induced early MOF. The ‘one hit’ model, i.e. a massive initial insult culminating in early fulminant SIRS and MOF, or the ‘two hit’ model, i.e. resuscitating severely injured patients with SIRS, followed by an early second inflammatory insult (e.g. pulmonary aspiration, blood transfusion, or early orthopedic intervention) amplifying SIRS to induce early MOF1
. This was believed due to priming and activation of innate immune response (principally PMN mediated). SIRS was followed by delayed immunosuppression, often leading to late infection, which, in turn, appeared to precipitate late MOF after 7 to 10 days.
By the early 2000s, ongoing research revealed that many time honored ICU interventions (e.g. high tidal volume mechanical ventilation, high volume crystalloid resuscitation, liberal blood transfusions, early TPN, intermittent hemodialysis) were actually promoting nosocomial infection and late MOF. Over the last decade, with more consistent delivery of evidence-based care to minimize these practices, ACS has become rare, mortality from traumatic shock-induced MOF has decreased substantially, and late MOF has disappeared14–16
However, this decrease in MOF incidence has not been observed with sepsis. Despite tremendous research efforts, sepsis remains a leading cause of MOF and prolonged ICU stay 17
. With the advancing age of our population, the incidence of sepsis is increasing, and mortality remains prohibitively high (>40%) for patients who are allowed to progress into septic shock18
Early sepsis is often difficult to recognize, and for most patients presenting early sepsis, the initial diagnosis is delayed or missed. Many interventions are known to have an impact on outcome, but are often administered in haphazard fashion. Optimal management strategies require assessment and implementation of current evidence-based standard operating procedures (SOPs). Such approaches have been successfully demonstrated by the recent ’Glue Grant‘ experience19
, the ‘Surviving Sepsis Campaign,’20
ARDSNET, and other evidence-based guidelines (EBGs). A widely recognized challenge is bedside implementation of EBGs in daily intensive care of the individual patient. Recent audits have shown surprisingly low compliance with widely accepted EBGs and substantial improvement in outcomes by strategies that only modestly improve compliance16, 21
One approach that has consistently demonstrated improved implementation of evidence-based care and patient outcome is computerized clinical decision support (CCDS). With rule sets devised by the ICU clinician team, computerized protocol systems standardize clinical decision making, ensure high compliance, and consistently out-perform expert ICU clinicians22
. EBG implementation not only improves outcome, but also provides a more stable platform for multicenter prospective randomized clinical studies, because confounding effects of variable care are controlled and ongoing process improvement is stimulated.
Implementation of CCDS has resulted in a surprisingly large decrease in mortality rate from severe sepsis (). Mortality rate from early fulminant SIRS has decreased significantly because intervention occurs prior to development of septic shock. Similar results were found with the Glue Grant experience in patients with severe blunt trauma using multi-site, consensus derived, evidence based standardized SOPs ()19
. Glue Grant participants found modest compliance with agreed evidence-based SOPs to be associated with dramatic reduction of overall mortality rate from approximately 22% pre SOP to 11% after SOP implemenation.
Mortality Rate for Severe Sepsis-Septic Shock and Trauma-The Methodist Hospital (TMH) Surviving Sepsis Campaign and the Glue Grant Experiences
However, there remain a large number of patients who linger in the ICU with manageable organ dysfunctions, but who usually do not meet established criteria for late MOF. Their clinical course is characterized by ongoing protein catabolism with poor nutritional status, poor wound healing, immunosuppression and recurrent infections. These patients are commonly discharged to long term acute care (LTAC) facilities, but due to excessive loss of lean body mass and prolonged immunosuppression, often develop secondary nosocomial infections and rarely rehabilitate or return to functional life.
We propose that this ‘persistent inflammation - immunosuppression catabolism syndrome (PICS)’ is the predominant phenotype that has replaced late occurring MOF in surgical ICU patients who fail to recover. Clinical management of these patients as they progress irreversibly toward indolent death is perhaps the most challenging problem currently confronting surgical intensive care.