Limits...
Likelihood of infection in patients with presumed sepsis at the time of intensive care unit admission: a cohort study.

Klein Klouwenberg PM, Cremer OL, van Vught LA, Ong DS, Frencken JF, Schultz MJ, Bonten MJ, van der Poll T - Crit Care (2015)

Bottom Line: Yet, the accuracy of categorizing critically ill patients presenting to the intensive care unit (ICU) as being infected or not is unknown.We studied a cohort of critically ill patients admitted with clinically suspected sepsis to two tertiary ICUs in the Netherlands between January 2011 and December 2013.A higher likelihood of infection does not adversely influence outcome in this population.

View Article: PubMed Central - PubMed

Affiliation: Department of Intensive Care Medicine, University Medical Center Utrecht, Room F06.149, P.O. Box 85500, 3508, GA, Utrecht, The Netherlands. p.m.c.kleinklouwenberg@umcutrecht.nl.

ABSTRACT

Introduction: A clinical suspicion of infection is mandatory for diagnosing sepsis in patients with a systemic inflammatory response syndrome. Yet, the accuracy of categorizing critically ill patients presenting to the intensive care unit (ICU) as being infected or not is unknown. We therefore assessed the likelihood of infection in patients who were treated for sepsis upon admission to the ICU, and quantified the association between plausibility of infection and mortality.

Methods: We studied a cohort of critically ill patients admitted with clinically suspected sepsis to two tertiary ICUs in the Netherlands between January 2011 and December 2013. The likelihood of infection was categorized as none, possible, probable or definite by post-hoc assessment. We used multivariable competing risks survival analyses to determine the association of the plausibility of infection with mortality.

Results: Among 2579 patients treated for sepsis, 13% had a post-hoc infection likelihood of "none", and an additional 30% of only "possible". These percentages were largely similar for different suspected sites of infection. In crude analyses, the likelihood of infection was associated with increased length of stay and complications. In multivariable analysis, patients with an unlikely infection had a higher mortality rate compared to patients with a definite infection (subdistribution hazard ratio 1.23; 95% confidence interval 1.03-1.49).

Conclusions: This study is the first prospective analysis to show that the clinical diagnosis of sepsis upon ICU admission corresponds poorly with the presence of infection on post-hoc assessment. A higher likelihood of infection does not adversely influence outcome in this population.

Trial registration: ClinicalTrials.gov NCT01905033. Registered 11 July 2013.

No MeSH data available.


Related in: MedlinePlus

Patient outcomes for various sites of infection stratified by plausibility of infection. Data are crude associations. The length of ICU stay (LoS) is shown as median. ICU-acquired infections (ICU-AI) were defined as infections that started >48 hours after admission with a plausibility of infection of at least possible. Acute kidney injury (AKI) and adult respiratory distress syndrome (ARDS) that were present at or occurred during ICU admission were taken into account. Whiskers indicate the 95 % CI. p values indicate the results of the Cochran-Armitage chi-square test for trend. Urinary tract and skin/soft tissue infections are not shown because of relatively small subgroups after stratification
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4562354&req=5

Fig3: Patient outcomes for various sites of infection stratified by plausibility of infection. Data are crude associations. The length of ICU stay (LoS) is shown as median. ICU-acquired infections (ICU-AI) were defined as infections that started >48 hours after admission with a plausibility of infection of at least possible. Acute kidney injury (AKI) and adult respiratory distress syndrome (ARDS) that were present at or occurred during ICU admission were taken into account. Whiskers indicate the 95 % CI. p values indicate the results of the Cochran-Armitage chi-square test for trend. Urinary tract and skin/soft tissue infections are not shown because of relatively small subgroups after stratification

Mentions: Figure 3 shows various patient outcomes in the whole population, and stratified by infection likelihood and the most prevalent presumed sources of infection. The plausibility of infection was not associated with mortality either in the entire patient population admitted with a sepsis diagnosis (21 %, 18 %, 20 %, and 20 % mortality in patients with infection likelihoods of none, possible, probable, and definite, respectively) or in any of the main subgroups of presumed infection sites except for the lungs. Figure 4 displays the cumulative incidence functions of mortality for the none–possible vs. probable–definite classes of infection plausibility. The confidence intervals for all four categories overlap, meaning that in this crude survival analysis plausibility of infection was also not associated with mortality (p = 0.73; crude SHR 1.05; 95 % confidence interval (CI) 0.88–1.25). In the multivariable analysis, however, a higher plausibility of infection (probable/definite) was associated with a lower mortality (SHR 0.81; 95 % CI 0.67–0.97). This means that patients with a confirmed infection diagnosis actually have a lower mortality rate than patients with an unconfirmed infection or an alternative diagnosis. Cause-specific analysis revealed that this reduction was caused by a direct effect on death (CSHR 0.73; 95 % CI 0.61–0.89), and not by the indirect effect on a longer ICU length of stay (CSHR 0.93; 95 % CI 0.85–1.02). In subgroup analyses, the mortality hazard for each hospital was similar (hospital A: SHR 0.80, 95 % CI 0.62–1.03; hospital B: SHR 0.85, 95 % CI 0.63–1.13). These estimates were similar when restricting our analysis to cases with none or definite infections only (SHR 0.75, 95 % CI 0.55–1.01). Furthermore, the prevalence of the adult respiratory distress syndrome, the prevalence of acute kidney injury, and the length of stay significantly increased with greater infection likelihoods (p <0.001), whereas the occurrence of ICU-acquired infections did not (p = 0.36) (Fig. 3). In the main subgroups of presumed infection sites, the infection plausibility was not associated with outcome parameters in this crude analysis, except for pulmonary infections.Fig. 3


Likelihood of infection in patients with presumed sepsis at the time of intensive care unit admission: a cohort study.

Klein Klouwenberg PM, Cremer OL, van Vught LA, Ong DS, Frencken JF, Schultz MJ, Bonten MJ, van der Poll T - Crit Care (2015)

Patient outcomes for various sites of infection stratified by plausibility of infection. Data are crude associations. The length of ICU stay (LoS) is shown as median. ICU-acquired infections (ICU-AI) were defined as infections that started >48 hours after admission with a plausibility of infection of at least possible. Acute kidney injury (AKI) and adult respiratory distress syndrome (ARDS) that were present at or occurred during ICU admission were taken into account. Whiskers indicate the 95 % CI. p values indicate the results of the Cochran-Armitage chi-square test for trend. Urinary tract and skin/soft tissue infections are not shown because of relatively small subgroups after stratification
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4562354&req=5

Fig3: Patient outcomes for various sites of infection stratified by plausibility of infection. Data are crude associations. The length of ICU stay (LoS) is shown as median. ICU-acquired infections (ICU-AI) were defined as infections that started >48 hours after admission with a plausibility of infection of at least possible. Acute kidney injury (AKI) and adult respiratory distress syndrome (ARDS) that were present at or occurred during ICU admission were taken into account. Whiskers indicate the 95 % CI. p values indicate the results of the Cochran-Armitage chi-square test for trend. Urinary tract and skin/soft tissue infections are not shown because of relatively small subgroups after stratification
Mentions: Figure 3 shows various patient outcomes in the whole population, and stratified by infection likelihood and the most prevalent presumed sources of infection. The plausibility of infection was not associated with mortality either in the entire patient population admitted with a sepsis diagnosis (21 %, 18 %, 20 %, and 20 % mortality in patients with infection likelihoods of none, possible, probable, and definite, respectively) or in any of the main subgroups of presumed infection sites except for the lungs. Figure 4 displays the cumulative incidence functions of mortality for the none–possible vs. probable–definite classes of infection plausibility. The confidence intervals for all four categories overlap, meaning that in this crude survival analysis plausibility of infection was also not associated with mortality (p = 0.73; crude SHR 1.05; 95 % confidence interval (CI) 0.88–1.25). In the multivariable analysis, however, a higher plausibility of infection (probable/definite) was associated with a lower mortality (SHR 0.81; 95 % CI 0.67–0.97). This means that patients with a confirmed infection diagnosis actually have a lower mortality rate than patients with an unconfirmed infection or an alternative diagnosis. Cause-specific analysis revealed that this reduction was caused by a direct effect on death (CSHR 0.73; 95 % CI 0.61–0.89), and not by the indirect effect on a longer ICU length of stay (CSHR 0.93; 95 % CI 0.85–1.02). In subgroup analyses, the mortality hazard for each hospital was similar (hospital A: SHR 0.80, 95 % CI 0.62–1.03; hospital B: SHR 0.85, 95 % CI 0.63–1.13). These estimates were similar when restricting our analysis to cases with none or definite infections only (SHR 0.75, 95 % CI 0.55–1.01). Furthermore, the prevalence of the adult respiratory distress syndrome, the prevalence of acute kidney injury, and the length of stay significantly increased with greater infection likelihoods (p <0.001), whereas the occurrence of ICU-acquired infections did not (p = 0.36) (Fig. 3). In the main subgroups of presumed infection sites, the infection plausibility was not associated with outcome parameters in this crude analysis, except for pulmonary infections.Fig. 3

Bottom Line: Yet, the accuracy of categorizing critically ill patients presenting to the intensive care unit (ICU) as being infected or not is unknown.We studied a cohort of critically ill patients admitted with clinically suspected sepsis to two tertiary ICUs in the Netherlands between January 2011 and December 2013.A higher likelihood of infection does not adversely influence outcome in this population.

View Article: PubMed Central - PubMed

Affiliation: Department of Intensive Care Medicine, University Medical Center Utrecht, Room F06.149, P.O. Box 85500, 3508, GA, Utrecht, The Netherlands. p.m.c.kleinklouwenberg@umcutrecht.nl.

ABSTRACT

Introduction: A clinical suspicion of infection is mandatory for diagnosing sepsis in patients with a systemic inflammatory response syndrome. Yet, the accuracy of categorizing critically ill patients presenting to the intensive care unit (ICU) as being infected or not is unknown. We therefore assessed the likelihood of infection in patients who were treated for sepsis upon admission to the ICU, and quantified the association between plausibility of infection and mortality.

Methods: We studied a cohort of critically ill patients admitted with clinically suspected sepsis to two tertiary ICUs in the Netherlands between January 2011 and December 2013. The likelihood of infection was categorized as none, possible, probable or definite by post-hoc assessment. We used multivariable competing risks survival analyses to determine the association of the plausibility of infection with mortality.

Results: Among 2579 patients treated for sepsis, 13% had a post-hoc infection likelihood of "none", and an additional 30% of only "possible". These percentages were largely similar for different suspected sites of infection. In crude analyses, the likelihood of infection was associated with increased length of stay and complications. In multivariable analysis, patients with an unlikely infection had a higher mortality rate compared to patients with a definite infection (subdistribution hazard ratio 1.23; 95% confidence interval 1.03-1.49).

Conclusions: This study is the first prospective analysis to show that the clinical diagnosis of sepsis upon ICU admission corresponds poorly with the presence of infection on post-hoc assessment. A higher likelihood of infection does not adversely influence outcome in this population.

Trial registration: ClinicalTrials.gov NCT01905033. Registered 11 July 2013.

No MeSH data available.


Related in: MedlinePlus