The Persistent Inflammation, Immunosuppression, and Catabolism Syndrome (PICS) Ten Years Later

Abstract

With the implementation of new intensive care unit (ICU) therapies in the 1970s, multiple organ failure (MOF) emerged as a fulminant inflammatory phenotype leading to early ICU death. Over the ensuing decades, with fundamental advances in care, this syndrome has evolved into a lingering phenotype of chronic critical illness (CCI) leading to indolent late post-hospital discharge death. In 2012, the University of Florida (UF) Sepsis Critical Illness Research Center (SCIRC) coined the term Persistent Inflammation, Immunosuppression, and Catabolism Syndrome (PICS) to provide a mechanistic framework to study CCI in surgical patients. This was followed by a decade of research into PICS-CCI in surgical ICU patients in order to define the epidemiology, dysregulated immunity, and long-term outcomes after sepsis. Other focused studies were performed in trauma ICU patients and emergency department sepsis patients. Early deaths were surprisingly low (4%); 63% experienced rapid recovery. Unfortunately, 33% progressed to CCI, of which 79% had a poor post-discharge disposition and 41% were dead within one year. These patients had biomarker evidence of PICS, and these biomarkers enhanced clinical prediction models for dismal one-year outcomes. Emergency myelopoiesis appears to play a central role in the observed persistent immune dysregulation that characterizes PICS-CCI. Older patients were especially vulnerable. Disturbingly, over half of the older CCI patients were dead within one year and older CCI survivors remained severely disabled. Although CCI is less frequent (20%) after major trauma, PICS appears to be a valid concept. This review will specifically detail the epidemiology of CCI, PICS biomarkers, effect of site of infection, acute kidney injury, effect on older patients, dysfunctional high-density lipoproteins, sarcopenia/cachexia, emergency myelopoiesis, dysregulated erythropoiesis, and potential therapeutic interventions.

Social Media Information

A review of UF SCIRC’s research efforts characterizing CCI, PICS biomarkers, effect of site of infection, acute kidney injury, effects on older patients, dysfunctional high-density lipoproteins, sarcopenia/cachexia, emergency myelopoiesis, and dysregulated erythropoiesis.

I. INTRODUCTION

Multiple organ failure (MOF) emerged as a deadly syndrome in surgical intensive care units (ICUs) in the early 1970s. Sepsis and trauma were the primary inciting events. With tremendous the advances in care over the ensuing four decades, the epidemiology of MOF evolved from a fulminant phenotype of progressive organ failure leading to early death into a lingering phenotype of chronic critical illness (CCI) leading to indolent death.(1, 2) CCI was first described in a 1985 article entitled “to save or let die.”(3) This was followed by reports in the 1990s describing CCI in ventilator-dependent patients who were discharged to long-term acute care (LTAC) facilities for ventilator weaning.(4) These reports focused on long-term functional disability using descriptive terms including “polyneuropathy of critical illness,” “myopathy of critical illness,” and “ICU-acquired weakness.”(5) It was subsequently recognized that CCI affected other systems.(6) Most recently, the CCI literature has popularized the term “post-intensive care unit syndrome,” adding that ICU delirium contributes to long-term cognitive impairments with depression and posttraumatic stress disorders.(7) These reports have largely come from heterogeneous medical ICU patients and implicate different risk factors depending upon the patient population, but offer no unifying underlying pathobiology for CCI.

In a 2012 review article, the University of Florida (UF) Sepsis Critical Illness Research Center (SCIRC) coined the term Persistent Inflammation, Immunosuppression, and Catabolism Syndrome (PICS) to describe underlying pathobiology of the CCI phenotype that is now commonly seen in surgical ICU survivors.(8) This term was proposed to provide a mechanistic paradigm in which to study CCI in surgical ICU patients who are now surviving previously lethal inflammatory insults (e.g. trauma, sepsis, burns, and pancreatitis). The purpose of this review article is to summarize the shift from MOF into PICS-CCI in surgical patients.

II. VALIDATION OF PICS-CCI PARADIGM

Figure 1 summarizes the timeline of the evolving syndrome of MOF. This has been discussed in other articles.(2, 8) In brief, with the introduction of new ICU therapies in the 1970s, MOF emerged as a highly lethal syndrome. Seminal case series (of predominately penetrating trauma) published in the late-1970s concluded that MOF was the “fatal expression of uncontrolled infection.” The two recognized phenotypes were uncontrolled sepsis and septic “auto-catabolism” that led to unremitting MOF with an ICU mortality of 50–80%. Research efforts led to improved use of antibiotics, advances in computerized tomography/interventional radiology, and nutritional interventions.

Figure 1:

Timeline of evolving epidemiology of multiple organ failure (MOF). The upper portion (a) depicts the major advances in care from 1970 through early 2010s and the lower portion depicts the resulting series of predominant clinical presentations (or phenotypes) that emerged over these four decades. The bottom of the figure (b) lists corresponding research topics that have been pursued. ICU, intensive care unit; TPN, total parenteral nutrition; IAI, intra-abdominal infection; PAC, pulmonary artery catheter; CT, computerized tomographic; IR, interventional radiology; ATLS, advanced trauma life support; ACS, abdominal compartment syndrome; SIRS/CARS, systemic inflammatory response syndrome/compensatory anti-inflammatory response syndrome; TIC, trauma-induced coagulopathy; MTPs, massive transfusion protocols; FAST, focused abdominal sonography of trauma; SOPs, standard operating procedures; PICS-CCI, persistent inflammation, immunosuppression catabolism syndrome-chronic critical illness; VO₂, oxygen consumption; PMN, neutrophil; IEDs, immune-enhancing diets; DAMPs, damage-associated molecular patterns; PAMPs, pathogen-associated molecular patterns; CTAA, chronic trauma-associated anemia. Adapted from Moore FA, Moore EE, Billiar TR, Vodovotz Y, Banerjee A, Moldawer LL. The role of NIGMS P50 sponsored team science in our understanding of multiple organ failure. J Trauma Acute Care Surg. 2017;83(3):520–531. Copyright© 2017 Lippincott Williams.

In the mid-1980s, however, studies demonstrated that victims of severe blunt trauma were developing MOF without identifiable sites of infection. The term “sepsis syndrome” was popularized to describe how both infectious and non-infectious insults resulted in similar auto-destructive inflammatory responses that cause MOF. There was considerable debate concerning underlying pathobiology and many popular alternative research hypotheses were pursued through the 1980s.(1) Additionally, through the 1980s and early 1990s there were tremendous advances in trauma and surgical critical care. Victims of major trauma were being triaged into specialized trauma centers where they were promptly treated based on Advanced Trauma Life Support (ATLS) guidelines, damage control surgery, and aggressive ICU resuscitation.(9) Early deaths from bleeding decreased, but an epidemic of abdominal compartment syndrome (ACS) emerged as a disturbing phenotype of early post-injury MOF.(10)

The first consensus conference on sepsis and organ failure convened in 1992 introduced the “systemic inflammatory response syndrome” (SIRS) into ICU vernacular to replace sepsis syndrome.(11) However, it was not until the late 1990s that the “danger model” and “endogenous alarmins” were proposed to explain how infectious and non-infectious insults could cause similar SIRS. Infectious insults release pathogen associated molecular patterns (PAMPs, such as lipopolysaccharides) and non-infectious insults release damage associated molecular patterns (DAMPs, such as HMGB1).(12, 13) Both DAMPs and PAMPs bind to the same pattern recognition receptors to activate a similar innate immune response. Additionally, with improved ICU care, MOF became a bimodal phenomenon. One third of the cases occurred early (within 3 days) due to the inciting event, while the remaining two thirds occurred later (average 7 days) and were associated with nosocomial infections.(14) Consistent with this observation, it was proposed that SIRS was followed by a compensatory anti-inflammatory response syndrome (CARS) which set the stage for immunosuppression, nosocomial infection, and late MOF.(15) By the late-1990s, SIRS/CARS became the accepted paradigm to explain the immunologic trajectory leading to early versus late MOF, with subsequent research focusing on characterizing the multiple defects in adaptive immunity that characterize CARS.(16)

By the early 2000s, with fundamental changes in early trauma care, the ACS epidemic disappeared.(17) Additionally, with the emphasis on consistent implementation of evidence-based standard operating procedures (SOPs) for ICU care [e.g., Glue Grant (GG) trauma experience and the Surviving Sepsis Campaign (SCC)], early ICU deaths decreased and late MOF deaths disappeared. As a result, by the early 2010s, the predominant MOF phenotype had become lingering CCI. In the early 2000s, the National Institute of General Medical Sciences (NIGMS) funded the GG program entitled “Injury and the Host Response to Injury” to characterize the genomic response after severe blunt trauma. A 2012 GG publication entitled “A Genomic Storm in Critically Injured Humans” reported that 75% of genes were upregulated (predominantly innate immunity) or downregulated (predominantly adaptive immunity).(18) This study demonstrated simultaneous (not sequential) SIRS and CARS. It concluded that failure to restore immune homeostasis characterized a complicated ICU course.

In 2012, the PICS-CCI paradigm was proposed. This was based on a) the recognition of CCI as the predominant “new” MOF phenotype in surgical ICUs, b) the results of the GG studies c) an analysis of Dr. Moore’s sepsis MOF database from Houston and, d) Dr. Moldawer’s ongoing research regarding dysregulated immunity in chronic models of sepsis and trauma.(19–23) We proposed that, after an inflammatory insult, the host response can provoke intense SIRS leading to fulminant MOF and early death trajectory. Fortunately, early insult recognition and consistent implementation of ICU SOPs have reduced this trajectory’s fatal expression. If severely insulted patients do not die of early MOF, there are two alternative clinical outcomes. They can experience rapid recovery (RAP) or progress into PICS-CCI. The clinical trajectory of PICS-CCI patients is characterized by persistent inflammation with an acute phase response and ongoing protein catabolism. Despite aggressive nutritional intervention, there is substantial loss of lean body mass and a proportional decrease in functional status, as well as poor wound healing. PICS-CCI patients remain immunosuppressed and suffer from recurrent nosocomial infections. They are largely discharged to LTACs and skilled nursing facilities (SNFs), and commonly experience sepsis recidivism requiring re-hospitalization, failure to rehabilitate, and an indolent death.

a. Validation of the PICS-CCI Paradigm

Figure 2 depicts the updated PICS-CCI paradigm.(24) The sections that follow highlight key research findings. These were observed mostly in patients enrolled in a five-year prospective cohort study in surgical ICU sepsis patients. Other studies were done in trauma ICU patients and emergency department (ED) sepsis patients. We primarily chose to study sepsis because of its abundance in the surgical ICUs. We had also established sepsis screening and management SOPs embedded into the electronic medical record and demonstrated that high compliance with the SSC guidelines substantially reduced mortality.(25) Over 4 years, we enrolled 363 patients with a previously published study design.(26) Of note, post discharge assessments were performed at 3, 6, and 12 months for mortality, Health Related Quality of Life [HRQOL-measured by EuroQol (EQ)-5D-3L and Short Form-36 (SF-36)], cognitive function [by the Hopkins Verbal Learning Test (HVLT)], physical function [by Short Physical Performance Battery (SPPB) and handgrip strength] and performance status (by Zubrod Performance Status). Baseline HRQOL and performance status were obtained by patient/proxy reported 4-week recall assessment as soon as possible after sepsis onset.

Figure 2:

Bone marrow changes associated with the clinical trajectories of sepsis. BM, bone marrow; EMVs, exosomes and microvesicles; PICS, persistent inflammation-immunosuppression and catabolism syndrome; LTAC, long-term acute care; MOF, multiple organ failure; SIRS, systemic inflammatory response syndrome; HSCs, hematopoietic stem cells; Tregs, regulatory T-cells; sPDL-1, soluble programmed death-ligand 1; MDSC, myeloid-derived suppressor cells. Adapted from: Rincon JC, Efron PA, Moldawer LL. Immunopathology of chronic critical illness in sepsis survivors: Role of abnormal myelopoiesis. J Leukoc Biol. 2022; 112(6):1525–34. Copyright© 2022 Society for Leukocyte Biology.

b. Epidemiology of CCI

To characterize the epidemiology of CCI in surgical sepsis, study patients were categorized into three predefined clinical trajectories(27): 1) early death (within 14 days of sepsis onset) was surprisingly low at 4%, 2) RAP (discharged from the ICU within 14 days with resolution of organ dysfunction) occurred in 63%, and, 3) CCI (ICU stay greater than or equal to 14 days with evidence of persistent organ dysfunction) developed in 33%. Not surprisingly, the CCI and RAP patients were different at baseline. CCI patients were more likely to be male, older and have more comorbidities. More CCI patients presented in shock (44% vs 15%) with higher APACHE II scores (22 vs 14). Expectedly, CCI patients experienced significantly more organ dysfunction. MOF occurred in 36% of CCI versus 2% in RAP. CCI patients also experienced more nosocomial infections (3.1 vs 1.5 per 100 hospital days). Most CCI patients (~80%) had a “poor” post-discharge disposition (LTAC, SNF, hospice, transfer back to referring hospital, late hospital deaths) compared to the RAP patients (~80%) who had a “good” post-discharge disposition (home or rehabilitation). While early mortality of the overall study population was low, the CCI cohort experienced significant ongoing post-hospital discharge mortality with a 12-month mortality of 41% compared to 4% in RAP (see Figure 3a). The primary causes of late CCI deaths were progressive MOF and recurrent sepsis. At pre-sepsis baseline, there was no difference in performance status in CCI versus RAP, but at 3, 6, and 12 months patients with CCI performed much worse (see Figure 3b). Additionally, at 3 and 6 months, CCI patients had significantly lower levels of HRQOL (by EQ-5D-3L) and physical function (based on SPPB, hand grip, and Zubrod at 3 months, and SPPB and Zubrod at 6 months). By 12 months, the CCI group had significantly lower levels of cognition (by HVLT), HRQOL, and physical function on all measures.(28)

Figure 3:

Comparison of outcomes over 1-year after sepsis for CCI versus RAP sepsis cohorts. (A) One-year mortality probability and (B) Twelve-month Zubrod scores. Data presented as mean ± standard error with statistical significance set at p < 0.05. Adapted from: Brakenridge SC, Efron PA, Cox MC, Stortz JA, Hawkins RB, Ghita G, et al. Current Epidemiology of Surgical Sepsis: Discordance Between Inpatient Mortality and 1-year Outcomes. Ann Surg. 2019; 270(3):502–10. Copyright© 2019 Wolters Kluwer Health, Inc.

As a corollary, a prospective study was undertaken in 135 blunt trauma patients (using GG entry criteria) to determine the incidence and characteristics of CCI.(29) Early deaths were low at 2%, whereas 79% exhibited RAP and 19% progressed to CCI. Patients who developed CCI were older, presented in more severe shock, and experienced greater organ failure severity. CCI patients developed more infectious complications (84% vs 35%) and were more likely to be discharged to a long-term care setting (56% vs 34%) than to a rehabilitation facility or home. At four-months, CCI patients had significantly higher mortality (16% vs 2%) and lower mean SF-36 general health scores (43 vs 65).

c. PICS Biomarkers in Surgical Sepsis

To validate the concept of PICS-CCI in our sepsis cohort, serial in-hospital biomarker profiles (out to 28 days) were compared.(30–32) It was found that the CCI cohort had biomarkers reflecting persistent inflammation [increased interleukin (IL)-6, IL-8, interferon gamma-induced protein 10 (IP-10), monocyte chemoattractant protein 1 (MCP1), and granulocyte-macrophage colony-stimulating factor (GM-CSF)]; immunosuppression [decreased absolute lymphocyte count (ALC) with increased soluble programmed death ligand-1 (sPDL-1) and IL-10]; stress metabolism [increased CRP and glucagon-like peptide 1 (GLP-1)]; catabolism [increased urinary 3-methylhistidine, and decreased insulin-like growth factor binding protein-3 (IGFBP3)]; and anti-angiogenesis [decreased vascular endothelial growth factor (VEGF), and increased soluble vascular endothelial growth factor receptor-1 (sFlt-1), angiopoietin-2 (Ang-2), and stromal cell-derived factor-1 (SDF-1)]. Multivariate logistic regression (MLR) prediction models were created using PIRO clinical variables (reflecting predisposition, insult, response, and organ dysfunction). Significant biomarkers were then added to determine if they strengthened the clinical predictions. Clinical PIRO models on day 4 accurately predicted CCI (AUC=0.89) and on day 7 accurately predicted 1-year Zubrod 4/5 (dismal outcomes, AUC=0.80). IL-10 and IP-10 levels on day 4 minimally improved prediction of CCI (AUC=0.90). However, IL-10, IL-6, IL-8, MCP1, IP-10, Ang2, GLP-1, sPDL-1, and SDF-1 on day 7 considerably improved the prediction of 1-year Zubrod 4/5 status (AUC=0.88, see Figure 4). This improved prediction of 1-year Zubrod 4/5 strongly suggests that PICS plays a role underlying pathobiology in the dismal long-term outcomes of CCI after sepsis.

Figure 4:

Area under the receiver operator curve (AUC) for MLR prediction models for Zubrod 4/5 (reflecting dismal long-term outcomes) at day 7 with clinical variables without biomarkers (AUC=0.80) and with biomarkers (AUC=0.88). Adapted from: Darden DB, Brakenridge SC, Efron PA, Ghita GL, Brumback B, Mohr AM, et al. Biomarker Evidence of the Persistent Inflammation, Immunosuppression and Catabolism Syndrome (PICS) in Chronic Critical Illness (CCI) after Surgical Sepsis. Ann Surg 2021;274(4):664–73. Copyright© 2021 Wolters Kluwer Health, Inc.

d. Site of Infection

To determine the effect of the role of site of infection in surgical sepsis, we compared the baseline predisposition, insult characteristics, serial biomarkers, hospital outcomes, and long-term outcomes of patients to five sites of infection.(33) These include intra-abdominal (incidence 44%), pulmonary (19%), skin/soft tissue (S/ST, 17%), genitourinary (GU, 12%), and vascular (7%). Most intra-abdominal infections (IAIs) were present on admission and required source control. They had more persistent inflammation (increased IL-6 and IL-8), immunosuppression (decreased ALC and increase sPDL-1), and persistent organ dysfunction (by serial SOFA scores). The long-term outcomes were poor with 37% CCI, 49% poor discharge dispositions, and 30% 1-year mortality. Most pulmonary infections were hospital-acquired. They had similar protracted inflammation and organ dysfunction, but immunosuppression normalized. Long-term outcomes were similarly poor (54% CCI, 47% poor disposition, 32% 1-year mortality). S/ST and GU infections occurred in younger patients with fewer comorbidities, less perturbed immune responses, and faster resolution of organ dysfunction. Comparatively, S/ST had better long-term outcomes (23% CCI, 39% poor disposition, 13% 1-year mortality) and GU had the best (10% CCI, 20% poor disposition, 10% 1-year mortality). Vascular sepsis patients were older males with more comorbidities. Inflammation was blunted with baseline immunosuppression and persistent organ dysfunction. They had the worst long-term outcomes (38% CCI, 67% poor disposition, 57% 1-year mortality). We conclude that further clarification of the underlying pathobiology of different subgroups is needed to develop precise interventions and design future trials. Additionally, IAI best reflected the PICS-CCI endotype and may prove to be the subset in which to study precision interventions.(34) The challenge will be to identify IAI patients (roughly one third) who are likely to progress into CCI and thus potentially benefit from novel multimodality interventions directed at the underlying pathobiology of PICS.

III. OLDER SEPSIS PATIENTS

To characterize the effect of age in surgical sepsis, we categorized our patients by age into young (≤ 45 years, 20%), middle-aged (46–64 years, 40%), and older (≥ 65 years, 40%) patients.(35) Compared to young and middle-aged patients, older patients had significantly more comorbidities, IAIs (37% vs 14% and 37%, respectively), septic shock (36% vs 12% and 25%) and organ dysfunctions by serial SOFA scores. Additionally, older patients had a higher 30-day mortality (17% vs 6% and 4%, respectively), fewer ICU-free days and more progression to CCI (42% vs 22% and 34%) with higher poor disposition discharge to non-home destinations (62% vs 19% and 40%). Long-term outcomes of older patients were dismal with higher 12-month mortality (33% vs 11% and 14%, respectively). Older survivors were severely disabled at 3 months and did not improve at 6- and 12-month follow-up. In contrast, the younger and middle-aged cohorts showed moderate disabilities at 3 months with improvements in their Zubrod performance status as well as objectively measured physical and cognitive functions at later time points.

The high rate of CCI and lack of recovery in older survivors is disturbing. Multiple contributing factors are likely involved including immuno-senescence, “inflammaging,” comorbidities, lack of physiologic reserve, pre-existing disability, and epigenetic changes.(36, 37) We compared serial PICS biomarkers in a) older (versus young) and b) older CCI (versus older RAP) patients (see Figure 5).(38) We found that both the older (versus young) and older CCI (versus older RAP) patients had more persistent aberrations in biomarkers reflecting inflammation (increased IL-6 and IL-8), immunosuppression (decreased ALC with increased sPDL-1), stress metabolism (increased GLP-1), catabolism (decreased albumin and IGFBP-3), and anti-angiogenesis (decreased VEGF and increased sFLT-1) over 14 days after sepsis. Based on these data, we conclude PICS plays a greater role in CCI in older sepsis patients. This is consistent with the GG trauma studies where blood neutrophil genome-wide expression analysis revealed older CCI patients had an attenuated early transcriptomic response; this attenuated gene response was consistent with the patients’ lower plasma cytokine and chemokine concentrations. Later, these older patients demonstrated amplified and persistent gene expression changes consistent with simultaneous inflammation and immunosuppression.(39)

Figure 5:

Comparisons of selected biomarkers. The left panel shows data for older (≥ 65 yrs) vs younger (≤ 45 yrs) patients and the right panel shows data for older CCI vs older RAP patients. A) shows increased interleukin-6 (IL-6) reflecting inflammation; B) shows increased soluble programmed death ligand 1 (sPDL-1) reflecting immunosuppression; C) shows increased glucagon-like peptide 1 (GLP-1) reflecting stress metabolism; D) decreased albumin catabolism. Perforated line represents biomarker levels in matched control subjects. Shaded area represents published normal reference ranges. Data presented as median (25% and 75%) with statistical significance set at p < 0.05. Adapted from Mankowski RT, Anton SD, Ghita GL, Brumback B, Darden DB, Bihorac A, et al. Older adults demonstrate biomarker evidence of the persistent inflammation, immunosuppression and catabolism syndrome (PICS) after sepsis. J Gerontol A Biol Sci Med Sci. 2022; 77(1):188–96. Copyright© 2022 Oxford University Press.

a. Sarcopenia, Cachexia, and Functional Disability

It is well recognized that loss of lean body mass in patients requiring prolonged ICU stays contributes to poor functional outcomes.(40, 41) There are multiple inciting factors and a discussion of them is beyond the scope of this review.(5) However, PICS-induced inflammation drives catabolism and blocks anabolism, and likely plays an important role in resulting cachexia. This occurs despite aggressive ICU nutritional support.(42, 43) Moreover, 40% of our cohort are elderly and the impact of pre-existing sarcopenia is poorly understood. To gain insight into these issues, we performed a focused subset study of 47 IAI patients (with baseline abdominal CT scans) to determine the impact of sarcopenia and acute muscle wasting on clinical outcomes.(44) Repeat abdominal CT scans were obtained at 3- and 12 month follow up and CT morphometric analysis quantified loss of torso skeletal muscle. Overall, these patients exhibited acute and persistent muscle wasting with an average 8% decrease at 3 months (that persisted at 12 months). Of note, half of the cohort were sarcopenic (SAR) at baseline. Sarcopenic (SAR) and non-sarcopenic (NSAR) groups were similar regarding age and comorbidity burden. SAR patients, however, had greater acute physiologic derangement (APACHE II, 18 vs 12), higher incidence of MOF (57% vs 17%), longer hospital (21 vs 12 days) and ICU length of stays (13 vs 4 days), and higher inpatient mortality (17% vs 0%) as well as 1-year mortality (35% vs 4%). Baseline muscle mass was a strong predictor of early death or CCI as well as poor functional status at 12 months. Additionally, persistent muscle mass loss at 3 months in SAR patients was associated with decreased HRQOL and SF-36 physical function domains.

b. Acute Kidney Injury

Acute kidney injury (AKI) is surprisingly common after sepsis and plays an important role in MOF inter-organ crosstalk.(45, 46) In our study, 62% of surgical sepsis patients developed AKI.(47) In one-third of those patients, AKI rapidly reversed within 48 hours. By the time of discharge or death, ~60% of persistent AKI patients did not recover renal function, and ~20% were dependent on renal replacement therapy. Overall, 24% of sepsis patients developed persistent AKI without renal recovery at discharge and their 1-year mortality was 43% (compared to 7% for rapidly reversed AKI and 8% with no AKI). These patients had persistently increased biomarkers reflecting PICS. The most pronounced elevations were observed in IL-8 and sPDL-1. Similarly, elevated biomarkers of anti-angiogenesis (i.e., Ang2, and sFlt-1) were observed. We also reported on a whole-genome analyses of RNA isolated from the urinary cells within 12 hours of sepsis onset.(48, 49) A set of 239 genes was identified that shows excellent effectiveness in classifying septic patients from those with chronic systemic disease in both internal and independent external validation cohorts. Functional analysis indexes revealed disrupted biological pathways in early sepsis and key molecular networks driving its pathogenesis. This approach can complement blood transcriptomic approaches for sepsis diagnosis and prognostication.

c. Dysfunctional High-Density Lipoprotein

Lipoproteins and cholesterol levels are some of the strongest predictors of clinical outcomes among sepsis patients.(50) In a series of clinical studies, we explored the functional aspects of dysregulated lipid metabolism and both their predictive ability and ability to discriminate biological sepsis subtypes. First, we showed that dysfunctional high-density lipoprotein (Dys-HDL) was increased in patients experiencing adverse outcomes (in-hospital death or discharge to nursing home or hospice).(51) Dys-HDL is both pro-inflammatory and pro-oxidant and, rather than protecting low density lipoprotein (LDL) from oxidation, propagates inflammation. Second, we showed that HDL-mediated “cholesterol efflux” was reduced in older adults with sepsis compared to controls.(52) Cholesterol efflux is one of the major mechanisms by which bacterial endotoxins are eliminated from circulation. Third, we showed that Dys-HDL levels correlated closely with SOFA score at enrollment (first 24 hours), 48 hours later, and were significantly higher in 28-day non-survivors.(53) Dys-HDL levels at enrollment were also found to be predictive of 48-hour SOFA score in a linear regression model. Fourth, we sought to study HDL’s relationship to functional outcomes and observed that enrollment total cholesterol and HDL-C levels were significantly lower in patients experiencing poor physical functional outcomes between 3 and 12 months after sepsis.(54) Fifth, in our recent work, we sought to derive and validate a new lipoprotein-based sepsis phenotype.(55) Using machine learning, we defined a “hypolipoprotein” (HYPO) cluster of patients exhibiting lower levels of HDL-C, LDL-C, antioxidant proteins apolipoproteinA-I (apoA-I), paraoxonase-1 (PON-1) with higher indicators of endothelial dysfunction (ICAM-1). Patients in the HYPO cluster exhibited significantly worse organ failure and higher mortality compared to patients with a normolipoprotein (NORMO) phenotype. In future work, we will use lipidomics and transcriptomics to characterize differences in patients exhibiting these HYPO vs NORMO phenotypes, and establish mechanisms and identify key pathways for drug targeting.

IV. EMERGENCY MYELOPOIESIS AS A DRIVING INFLUENCE ON PICS-CCI

Chronic murine models of sepsis and trauma have been developed to better reflect PICS-CCI and have identified expansion of myeloid-derived suppressor cells (MDSCs) as a central mechanism for the observed persistent immune dysregulation.(23) Our focused translational studies in severe sepsis/septic shock (SS/SS) patients confirmed the clinical relevance of these laboratory observations. It demonstrated that the percentage of MDSCs dramatically increases in SS/SS patients over the 28-day study period.(56, 57) Moreover, MDSC expansion correlated with predefined clinical trajectories. Early death patients had the highest initial expansion of MDSCs at both 12 and 24 hours when compared to patients who experienced RAP and CCI. RAP and CCI patients had similar significant elevations at these early timepoints and both experienced similarly decreased MDSCs on day 4. Thereafter, MDSC numbers normalized in RAP patients, while CCI patients continued to demonstrate increased MDSC numbers at later timepoints. It was subsequently shown that sustained MDSC expansion was a strong predictor of nosocomial infections [odds ratio (OR) =7.33] and poor post-discharge disposition (OR=8.40). More interestingly, almost all MDSCs were granulocytic with a gene expression profile reflective of a highly inflammatory and immunosuppressive transcriptome. With regard to MDSC suppressor activity, MDSCs obtained prior to day 7 were not immunosuppressive, while MDSCs after day 7 suppressed T cell proliferative responses and potently suppressed stimulated T-cell production of pro-inflammatory cytokines. These data provide rationale for the use of immunotherapy that modulate MDCSs as in advanced malignancies to achieve durable response rates. Finally, studies implicate subsets of MDSCs having unique roles in lymphocyte suppression; however, differences in the numbers and expression patterns of different MDSC subsets after sepsis remain controversial. Importantly, not only has our human research identified unique epigenetic profiles for MDSCs from septic patients, but our preliminary data may indicate the MDSCs are somewhat unique than those identified in cancer patients.(58, 59) This reinforces the importance of precision/personalized medicine in our approach to improving outcomes to PICS-CCI, and not making similar past errors of a “silver bullet” therapy for MDSCs and/or sepsis.

a. Dysregulated Erythropoiesis

Chronic trauma-associated anemia (CTAA) is a unique subtype of anemia in critical illness that is characterized by a prolonged hyper-catecholamine state mediating bone marrow (BM) dysfunction that manifests as decreased red blood cell production with low reticulocyte counts despite adequate iron stores and erythropoietin levels in the peripheral blood, as well as increased mobilization of hematopoietic progenitors from the BM (see Figure 6).(60–62) After trauma, circulating inflammatory mediators and catecholamines impair iron metabolism and inhibit erythropoiesis.(63) Norepinephrine release activates bone marrow macrophages to release HMGB-1, which leads to granulocyte colony stimulating factor (G-CSF) release, downregulates expression of stromal adhesion molecules, thus mobilizing hematopoietic progenitors.(64) These circulating progenitors migrate to sites of injury for tissue repair; however, in CCI, this hematopoietic progenitor cell mobilization is prolonged and leads to aberrant wound healing and decreased bone marrow cellularity.(65, 66) Preclinical studies suggest that interruption of hyper-catecholamine pathways and the neuroendocrine stress response via non-selective beta-adrenergic receptor blockade or by sympathetic nervous system outflow inhibition with the alpha-2 agonist clonidine can restore bone marrow production of erythroid progenitors and alleviate persistent injury-associated anemia.(62, 67) Subsequent investigations have suggested that the pathophysiology of sepsis-associated anemia is similar to that of injury-associated anemia. Specifically, persistent elevation of inflammatory cytokines among critically ill septic patients have been associated with iron-restricted anemia occurring in the absence of systemic iron deficiency, independent of endogenous erythropoietin.(68) Despite receiving more transfusions, CCI patients have persistent anemia that was associated with systemic inflammation and poor functional outcomes.(69) Finally, observations that transcriptomic changes occur within human bone marrow after severe trauma suggest potential for targeted gene therapies.(69) Unfortunately, clinical therapies have been difficult to establish, partly due to heterogeneity in practice patterns and the dangers of administering medications that blunt hyper-catecholamine pathways and the neuroendocrine stress response via non-selective beta-adrenergic receptor blockade or by sympathetic nervous system outflow inhibition. Patients who have developed (or are at risk for developing) shock depend upon catecholamines and the neuroendocrine stress response to maintain hemodynamic stability.

Figure 6:

The pathophysiology of chronic trauma associated anemia: Systemic inflammation, neuroendocrine activation, and allogenic blood product transfusions propagate iron dysregulation, mobilization of hematopoietic progenitors to injured tissues, and suppression of erythroid progenitor growth (NE: norepinephrine, IL-6: interleukin 6, G-CSF: granulocyte colony-stimulating factor, EPO: erythropoietin, EPO-R: erythropoietin receptor, TfR: transferrin receptor, CSF1/2-R: colony-stimulating factor 1 receptor and colony-stimulating factor 2 receptor). Reprinted with permission of the American Thoracic Society. Copyright © 2023 American Thoracic Society. All rights reserved. Originally published by Loftus TJ, Mira JC, Miller ES, Kannan KB, Plazas JM, Delitto D, Stortz JA, Hagen JE, Parvateneni HK, Sadasivan KK, Brakenridge SC, Moore FA, Moldawer LL, Efron PA, Mohr AM. 2018. The Postinjury Inflammatory State and the Bone Marrow Response to Anemia. Am J Respir Crit Care Med. Volume 198, pages 629–638. The American Journal of Respiratory and Critical Care Medicine is an official journal of the American Thoracic Society.

V. POTENTIAL THERAPEUTIC INTERVENTIONS AND FUTURE DIRECTIONS

PICS-CCI represents the new ‘endotype’ for ICU survivors who fail to fully recover and experience persistent debilitating inflammation and lean tissue wasting. An endotype is defined by its distinct pathobiological mechanisms, independent of the external causes. In the case of PICS-CCI, this can be sepsis, severe trauma, or burn injury. Importantly, this differs from a phenotype, which is defined as any observable characteristic or trait of a disease without any implication of a mechanism. However, as an endotype, PICS-CCI shares several characteristics with other persistent terminal inflammatory diseases, including cancer, chronic renal disease and cardiac cachexia. Currently, there are no proven therapies for PICS-CCI, but shared qualities with other chronic diseases may indicate that treatment modalities in these other conditions may apply to PICS-CCI.

a. Nutrition and Exercise

Current therapeutic approaches to treat PICS-CCI are in their infancy. The “septic auto-catabolism” phenotype of MOF in the 1980s provided the rationale for early total parenteral nutrition (TPN).(70) Subsequently, clinical trials failed to show that early TPN in the ICU was beneficial. Early enteral nutrition was shown to be superior (primarily by reducing nosocomial infections). However, neither early enteral nor TPN prevent the early catabolism that characterizes PICS-CCI.(71) However, extrapolating from other patient populations who experience chronic low-grade inflammation and cachexia, anabolic nutritional support should be effective. These interventions include high protein (1.5 to 2.0 grams/kg) and leucine supplementation with the addition of anabolic agents (such as propranolol and oxandrolone). Fish oil-derived specialized pro-resolving mediators (i.e. resolvins) are another nutritional strategy that may curb ongoing inflammation.(72) Other potential interventions from the aging literature include nicotinamide riboside, cocoa, and senolytics. These nutritional interventions should be combined with ICU physical therapy and post-discharge institution-based or remotely delivered resistant exercise training to promote anabolism.

b. Immunotherapeutics

PICS-CCI pathobiology at its foundation is an immunologic “dyscrasia.” Thus, immunotherapy (as in cancer) should be effective.(73, 74) Unfortunately, previous efforts utilizing a host of promising therapeutic agents to modulate SIRS in early sepsis were uniformly ineffective. In large part, this was due to patient heterogeneity and the over-assumption in the existence of a “silver bullet” agent. These remain issues with PICS-CCI patients. PICS-CCI frequently occurs in older patients with pre-existing comorbidities and is inherently intertwined with the patient’s prior functional status. Additionally, multiple immunologic derangements are operational to variable degrees and multimodality therapy will likely be needed.

We have demonstrated that MDSC expansion plays a role in later PICS-CCI pathobiology. As there is no way to remove MDSCs without making the patient granulocytopenic (which PICS-CCI patients would not tolerate), other methods of altering MDSC production, differentiation, and immunosuppression will be required. This could include, but not be limited to, the use of epigenetics and/or exosomes.(58, 75)

c. Precision Medicine and Artificial Intelligence

Precision medicine has emerged as an attractive approach to incorporate different factors (epigenomic, metabolomic, and proteomic) to identify specific pathologic endotypes.(76, 77) Readily available tests to determine the functional immune status of individual patients is critically important in determining personalized options. “Big Data” may prove to be particularly useful in PICS-CCI patients to quickly endotype a patient and prognosticate.(78) We have shown recently that the leukocyte transcriptome within 48 hours post-trauma is highly predictive of outcomes.(79) Similarly, we are exploring multivalent transcriptomic metrics in sepsis.(80) These techniques may be further improved using machine-learning algorithms and deep-learning technologies.

VI. LIMITATIONS TO THIS MANUSCRIPT

Several limitations must be noted by the reader. Firstly, this is a single institutional pilot proof of concept study. In addition, associations do not prove causation and interventional trials are required. Finally, there were limitations to some of the biological data obtained (e.g. biomarkers were only obtained while patients were in the hospital) and this aspect of the data requires expansion of the collection times, etc.

VII. SUMMARY AND LESSONS LEARNED

As a result of tremendous advances in patient care over the past five decades, MOF’s predominant phenotype has evolved into a lingering PICS-CCI endotype. In a 2012 review article, we proposed PICS as a mechanistic framework in which to study CCI in surgical ICU patients. A series of studies were performed to clarify its epidemiology, pathobiology, and long-term outcomes. Early ICU deaths were surprisingly low (<5%), but ~33% surgical sepsis patients and ~20% of trauma patients admitted to the ICU progress into CCI. These patients have biomarker evidence of PICS during hospitalization. Roughly 40% are dead at 1 year and most survivors remain severely disabled. Older patients are especially vulnerable. We observed phenotypic heterogeneity by site of infection. Persistent AKI, dysregulated myelopoiesis/erythropoiesis, Dys-HDL, and sarcopenia/cachexia were important contributing factors to the dismal long-term outcomes of PICS-CCI patients. While there are no proven therapies for PICS-CCI, we believe that anabolic nutrition as SIRS subsides, aggressive ICU PT with post-discharge exercise, and precision immunomodulation of specific defects are likely strategies to improve the long-term outcomes of PICS-CCI patients.

Advice from our laboratory for other institutions for similar type studies are numerous, but some key points are as follows. Firstly, work closely with your local institutional review board to determine solutions for studies in which the patient cannot consent for themselves and other individuals who can are not immediately available. In addition, family and legal authorized representative may be a vulnerable population when the patient is initially so ill. Thus, we were able to obtain approval for a 96-hour waiver of consent for our work. Secondly, we determined that weekly adjudication meetings with all staff, and importantly, all primary investigators and physicians, were key to confirming sepsis, its severity and site of infection. This meeting and the importance of attendance cannot be overemphasized for the success of such a project.