Effects of Sepsis on Morbidity and Mortality in Critically Ill Patients 2 Years After Intensive Care Unit Discharge

Livia Biason, MD; Cassiano Teixeira, MD, PhD; Jaqueline Sangiogo Haas, RN, MS; Cláudia da Rocha Cabral, RN; Gilberto Friedman, MD, PhD


Am J Crit Care. 2019;28(6):424-432. 

In This Article


Our study showed that compared with patients without sepsis, patients with sepsis have higher mortality in the ICU and within 2 years of discharge. Patients with sepsis also have more pain, more hospital readmissions, and worse functional decline during this period.

The septic process and excessive long-term morbidity and mortality rates appear to be directly related.[4,5,18–21] We compared only patients with sepsis and patients without sepsis, and we believe that results of previous studies may not be accurate because both the need for ICU care and the presence of several comorbidities translate to an increased risk of death.[6,22] In agreement with previous studies, our data show that the presence of sepsis, increased age, and higher APACHE II score substantially increase the risk of death after ICU discharge.[20–23] Mortality related to sepsis is particularly high in Brazil, and our data are consistent with those of previous studies conducted in developing countries.[1,24,25] A lack of recognition of sepsis by trained professionals, lack of adherence to prevention measures, and consequent health care–associated infections and shortage of ICU beds are possible explanations for the high mortality rate.[1,26]

Evidence for the best time to start the follow-up monitoring of these patients (upon ICU admission or upon discharge) is conflicting.[3] For this study, we constructed 2 Kaplan-Meier survival curves, one with mortality assessed after ICU admission and the other with mortality assessed after ICU discharge. Mortality rates were similar regardless of the onset of follow-up monitoring.

Regarding morbidity after ICU discharge, patients with sepsis appear to experience significant impairments.[3,5] Our results support these findings by demonstrating functional decline in patients with sepsis (as measured by the Karnofsky and IADL scales). Functional worsening seems to depend highly on the population studied.[3] Our results showed a similar functional decline between patients with sepsis and those without sepsis, but when adjusted for confounding factors, the decline evaluated with the IADL scale remained significantly related to sepsis. This finding may suggest that sepsis is an independent factor for post-ICU functional decline. Patients with sepsis tend to have decreased functional status and quality of life before ICU admission,[22] which could contribute to a functional decline after ICU discharge.

Another important finding was the higher prevalence of pain in patients with sepsis, which was also reported in previous studies.[23,27] Baumbach et al,[28] however, found no difference in the prevalence of pain between patients with sepsis and patients without sepsis 6 months after ICU discharge. Evidence suggests that proinflammatory cytokines may be somewhat implicated in increasing pain intensity in this group of patients.[29] Battle et al[30] argued that immobilization and lack of rehabilitation could lead to chronic pain.

Hospital readmission rates were higher among patients with sepsis, which is in agreement with results of previous studies.[31,32] According to a recent study,[33] immune system changes, residual organ damage, and accelerated progression of preexisting chronic conditions by the septic process, in addition to quality of patient care during ICU and hospital stay and post-ICU syndrome, could explain the need for readmission, highlighting the importance of post-ICU care.

In this context, the importance of a multidisciplinary team to care for patients with complex medical conditions after discharge from the ICU and the hospital has been studied. These patients experience various types of physical and psychological deterioration, so a multidisciplinary team is most qualified to follow them.[34–38]

A major strength of our study was the choice of comparison population. Interpreting mortality and quality-of-life studies in critically ill patients is very difficult because of heterogeneity among studies.[19] Most studies compare patients with sepsis with the general population. However, this comparison hampers data interpretation because of the difficulty of distinguishing findings caused by critical illness from findings specifically caused by sepsis.[6,39] In an attempt to isolate the effects of sepsis, we designated a group of critically ill patients without sepsis as the control group, as in the studies conducted by Korosec et al[40] and by Linder et al.[4]

Some limitations of this study should be noted. First, we did not review the patients' cause of death. Published reports display a wide gap regarding the hospitalization of patients with sepsis and their cause of death in the long term. Second, although the Karnofsky and Lawton IADL scales are routinely used, neither of these scales nor any other scale used in previous studies has been validated for use in critically ill patients or in patients with sepsis. In this study, we used 2 scales in an attempt to more accurately and comprehensively measure the degree of dependence, aiming to demonstrate through both scales the quality of life of these patients. We acknowledge that degree of dependence is an extremely broad concept, but we still lack an instrument capable of measuring it. Third, our findings regarding the presence of pain were obtained through nonvalidated, subjective questionnaires. Finally, although ICU mortality risk stratification models exist, none of them were used for the populations under study.