Hello. I'm Randeep Jawa. I'm a trauma critical care surgeon at Stony Brook Medicine in Long Island, New York. Thank you for joining us on this session on who said they were septic. This lecture will focus on sepsis definitions and their impact on care organization in low and middle income countries. As a bit of background, the true burden of sepsis incidents and outcomes is not known. Much of our data is based on that which was extrapolated from high income countries. Those data suggested that there were approximately 31.5 million cases annually with 5.1 million annual deaths. A recent study using vital registry data indicated that there were 49 million cases of sepsis annually across the globe with an 11 million potentially avoidable deaths. Likely these are underestimates and reasons for this include lack of population based data, especially from low and middle income countries, limitations of ICD coding, quality of data, issues with documentation, and bias for that there's limited global surveillance systems. It should be noted that in 2017 sepsis accounted for 20% of all cause global mortality. There appears to be an inverse relationship between income level and sepsis incidence and between income level and mortality. As seen in the graph on the left, lower and middle income countries have higher incidence of sepsis and sepsis related mortality. Indeed, 85% of all sepsis cases and 84.8% of all sepsis related deaths occur in low and middle income countries. Because sepsis is actually a syndrome, there are a variety of definitions that have been espoused. The consensus definitions for sepsis have changed over time. In 1992, they consisted of four categories, SERS, sepsis, severe sepsis, and septic shock. At sepsis two meeting in 2001, which consisted primarily of North America and European attendees, the definitions were kept constant, but the list of signs and symptoms were increased. Finally, at the third international consensus conference in 2016, only sepsis and septic shock were retained. Let's examine sepsis two from the 2001 international consensus conference a little bit closer. It had the systemic inflammatory response syndrome, followed by sepsis, where sepsis was defined as SERS plus infection. Next category was severe sepsis, which was defined as sepsis plus organ dysfunction. And the final category was septic shock, which was defined as systolic blood pressure less than 90 millimeters from mercury, or a greater than 40 millimeter decrease from baseline, despite volume resuscitation. The authors noted that these definitions did not allow for precise staging, and the attendees, as previously indicated, were based primarily in the US, Canada, and Western Europe. A more detailed description of the various diagnostic criteria is included in the table at the right side of the page. And of note, it includes general hemodynamic, inflammatory, organ dysfunction, and tissue perfusion variables. Sepsis three introduced the notion of the QSOFA, or quick SOFA, as a screening tool. Its components were respiratory rate, mental status, and systolic blood pressure. It also advanced the notion of using the SOFA score to assess organ dysfunction. Components of the sequential organ failure assessment score included PaO2 to FiO2, Glasgow Coma score, main arterial pressure, vasopressor type and dose, serum creatinine or urine output, bilirubin, and platelet count. Sepsis was therefore defined as a QSOFA score greater than or equal to 2, and a SOFA score greater than or equal to 2 in a patient with suspected infection. Meanwhile, septic shock was defined as a patient requiring vasopressors to maintain a MAP greater than or equal to 65 millimeters of mercury and a serum lactate greater than 2 millimeters per liter, despite adequate fluid resuscitation. The concept of severe sepsis was eliminated. Sepsis three definitions were based primarily on studies from the U.S., Canada, and Western Europe. The question becomes, are these consensus sepsis definitions generalizable? These definitions were based on findings in high-income countries, such as the USA, Canada, and Western Europe. In these countries, there's a lower prevalence of HIV, tuberculosis, malaria, and dengue. These definitions focus on bacterial infection and on a dysregulated host immune response. Protozoan and viral infections present and behave differently in that they can cause direct damage. For example, falciparum malaria can block the microcirculation via the red blood cell biomass, and dengue viral proteins can directly damage the endothelial glycocalyx. The World Health Organization published in 2012 the Integrated Management of Adult Illness District Clinician Manual, where severe sepsis or septic shock were defined as suspected infection plus hypotension with a systolic blood pressure less than 90 and one or more of the following, namely pulse greater than or equal to 100, respiratory rate greater than or equal to 24, temperature less than 36, or temperature greater than 38. The European Society for Intensive Care Medicine Working Group on Global Intensive Care made several recommendations for sepsis recognition in low and middle-income countries. Sepsis was defined as an acute infection with any two of the following parameters, namely a respiratory rate greater than or equal to 22, a systolic blood pressure less than 100, and or an acute change in mental status. Septic shock they recommended as being defined as having greater than or equal to two clinical indicators of systemic hypoperfusion independent of hypotension because hypotension was typically a pre-terminal event. The work group recommended separate definitions for sepsis from malaria and dengue as their presentation management differs and they further indicated that the recognition of sepsis in children be based on different indicators. A universal vital assessment score for mortality prediction was developed in Sub-Saharan Africa that was based on all clinical parameters and no laboratory data was required. Its components included respiratory rate, systolic blood pressure, Glasgow coma score, temperature, heart rate, oxygen saturation, and HIV status. The score performed well in terms of mortality prediction for patients with infectious and undiagnosed causes of mortality. It outperformed the modified early warning score as well as Q-SELFA in predicting risk of death in hospitalized patients regardless of underlying condition. A subsequent study in patients with sepsis confirmed these findings. Utilization of coding data to identify sepsis in administrative databases is inherently biased as sepsis-specific codes will underestimate prevalence whereas the combination of sepsis-specific codes with indirect codes such as those for infection will overestimate the prevalence. The problem is amplified in neonates where there are no validated sepsis definitions. Furthermore, these codes may not correlate with findings on chart review by clinicians. The question therefore naturally arises as to what is the applicability of these various sepsis definitions when labs, fluids, vasopressors, or other monitoring equipment is not available. For example, how would the SOFA score be calculated? Recognizing the problematic nature of sepsis definitions, let's now tackle the challenges in identifying and managing sepsis in low- and middle-income countries. There are multiple issues in sepsis recognition and management. The first being that there are limited ICUs and emergency rooms. Labs, biomarkers, and imaging are not available. There are few ICU physicians. There are also limited primary care physicians. Hence, sepsis may be diagnosed by community health workers, midwives, and nurses. As indicated above, they would be relying on clinical variables. Let's first examine what defines an intensive care unit or a critical care bed and their availability. The World Federation of Societies of Intensive and Critical Care Medicine proposed a three-tiered ICU classification system whereby level 1 ICUs are those that provide oxygen, non-invasive monitoring, and more intensive nursing care than a regular ward. Level 2 ICUs have the short-term ability to provide invasive monitoring and basic life support. Level 3 ICUs have the full spectrum of monitoring and life support technologies, and they serve as a regional resource for the critically ill and are also active in research and education. ICU bed availability differs widely across the globe, with North America and European high-income countries having approximately 5 to 30 ICU beds per 100,000 population, whereas much of Asia and Africa have 0.1 to 2.5 ICU beds per 100,000 population. China has 3.9, Mongolia has 11.7. This low ICU bed capacity limits patient access. It forces the providers to frequently make triage decisions and therefore likely increases preventable mortality. Further, the expenses for ICU care are covered largely by patients and or their families in low and some middle-income countries. Hence, there's the potential for denial of ICU admission to poor patients, and furthermore, there's the potential for premature withdrawal of life-saving interventions. These factors likely contribute to the disproportionate mortality in low and middle-income countries. Let's consider whether health systems in low and middle-income countries have the additional necessary resources. Low and middle-income countries face a variety of constraints that may limit their ability to provide intensive care. For example, electricity was only reliably available in 35.1% of 12 African countries that were surveyed. One-fourth of hospitals in resource-limited settings lack oxygen. Hand hygiene mechanisms are often lacking, and there are limitations in terms of availability of equipment, such as that for invasive or non-invasive monitoring, ventilators, and renal replacement therapy machines. Additional constraints are the availability of laboratory studies and imaging. Key or even basic labs are often not available. Even if they are available, there are issues with cost and payment, infrastructure, and insufficient trained personnel. Microbiologic studies are often not available as well. Neither are advanced imaging facilities. To this end, while 100% of high-income countries have at least one CAT scan per 1 million people, only 15% of low-middle-income countries have one CT scan. Given these constraints, is the surviving sepsis campaign feasible? A survey of 44 anesthesia providers in high-income countries and 263 anesthesia providers in Africa at the 2009 Pan-African Anesthesia Conference indicated that only 1.5% of African respondents had resources to meet the surviving sepsis campaign guidelines continually. This number was reduced to 1.2% of sub-Saharan African respondees. The resources were available to implement about three-fourths of the surviving sepsis guidelines in Africa. However, there was a wide variability. Of note, 16 out of the 74 guidelines were do not use or required no intervention. The study also indicated that the respondents were biased in that they primarily came from private institutions or larger referral centers. The survey demonstrated a multifaceted lack of resources in that 25% of hospitals lacked an intensive care unit, 15% lacked an emergency room. A variety of drugs were lacking. Of note, piperacillin was only available consistently in 26.4%, nor adrenaline was only consistently available in 46.7%. Further limiting access was the out-of-pocket payment requirements. A survey examining maternal sepsis in Malawi and lower-middle-income countries in Africa was published in 2018. It evaluated the 2012 surviving sepsis guidelines. They noted that antibiotic were always available in only 61.5% of Malawi hospitals. They noted that the basic infrastructure was always available in only 53.8% of Malawi hospitals. As seen in this slide from the survey, all surviving sepsis campaign recommendations were only implementable in approximately 30% of Malawi hospitals and low-income country hospitals, which was substantially less than the lower-middle-income country hospitals. Additional constraints are ICU staffing. There is limited ICU physician staffing. Indeed, in sub-Saharan Africa, non-physicians with training in some aspects of anesthesia run and staff the ICUs in cooperation with MDs. There's limited nurse staffing. Also, allied health and pharmacy staffing are limited. Furthermore, there is variable training of these providers. A study from an upper-middle-income country, namely Turkey, indicated that knowledge of sepsis bundles ranged from 12% for registrars and 16% for specialists for targeting central venous oxygen saturation of greater than 70% in severe sepsis to 88% and 90% respectively for obtaining blood cultures within three hours prior to antibiotic use. So why does this lack of resources matter? Well, it's because it limits the ability of providers to identify and treat sepsis. Furthermore, treatment paradigms may differ. For example, malnutrition results in hypoalbuminemia and edema, and volume resuscitation paradigms may need to be adjusted, as was demonstrated in pediatric sepsis with the FEAST trial. Let's shift gears a little bit and consider who pays for health care. According to the World Health Organization, high-income countries account for 80% of global health care spending, where the U.S. alone accounts for 40%. Indeed, the U.S. ICU costs are approximately 1% of the gross domestic product. In high-income countries, government spending accounts for approximately 70% of health care expenditure, and the average spending per capita is more than four times the average GDP per capita of low-income countries. In middle- and high-income countries, the share of health spending via domestic public sources has increased over the past 20 years, with a compensatory decrease in out-of-pocket spending. In low-income countries, out-of-pocket spending comprises approximately 44%, and there has been decreased governmental spending as external aid has increased. The external aid accounts for approximately 29% of spending, but it's not purely additional. Let's consider sepsis hospitalization costs in high-income countries. A 2000 study of U.S. inpatient hospital stays determined that sepsis was the most expensive condition with an aggregate cost of over $23 billion, whereas the average cost in a 2017 review of high-income countries and some upper-middle-income countries indicated the cost was approximately U.S. $32,000. Let's evaluate sepsis spending in several additional countries. A study from Brazil indicated that the mean cost for hospitalization for sepsis was $624, and for hospitalization for sepsis requiring ICU care it was $1,708. It should be noted that over 75% of the population exclusively uses the S.U.S. Health Service, and the remaining has access to private health services. Meanwhile, in a lower-middle-income country, namely Pakistan, the total health care budget was $150 million in 2020-2021. 70% of the health care costs were paid out-of-pocket. This high cost restricts ICU access. A further note is that vasopressor costs increased by nearly 60-fold between 2015 and 2017. In Indonesia, another lower-middle-income country, in 2018 universal health care coverage was provided by a single national payer for 203 million people, and it was being expanded to reach all 264 million people. A study demonstrated that the mean cost per sepsis case was approximately $1,253. However, they noted that there was a wide range of costs for sepsis, and it depended on the type of infection and location thereof. They noted that sepsis and focal infections are not coded together in their national disease price calculations, which may lead to under-budgeting for sepsis. In Nepal, another lower-middle-income country, 6.3% of the GDP is spent on health care, and out-of-pocket spending pays for 55.4% of total health care costs. The Nepalese government has free newborn care that covers up to NPR 8,000 for hospitalization costs. In neonates, the median cost exceeded this free newborn care cap in approximately 69% of cases at a tertiary care hospital. The authors recommended a revision of the maximum limit for free newborn care to avoid catastrophic expenditure. Without a generalizable definition of sepsis, it's difficult to determine the burden of disease, organize and develop health care systems, perform quality improvement, determine costs, and reimbursement thereof. In conclusion then, it's important that we define sepsis in a manner that is generalizable, as having a definition that can be applied across a wide array of settings will facilitate recognition of sepsis, and therefore its management. Costs are of course tied to recognition and management, and payment is again tied to recognition, management, and cost thereof. Obviously, government support and financing will depend on disease prevalence, namely if we recognize the presence of the disease and the cost of treatment. This in turn will impact the availability of the necessary resources to care for septic patients, and also the organization of the health care system to facilitate care of these patients. Thank you for attending, and if any questions please feel free to email me at randeep.jawa at stonybrookmedicine.edu. Thank you.

sepsis

definitions

impact

healthcare

low-income countries

middle-income countries

Research

Clinical Research Design

Presentation

Select

2022

Foundational

Intermediate

Advanced

Sepsis Resources