Comparative analysis of survival between elderly and non-elderly severe sepsis and septic shock resuscitated patients

Objective To compare outcomes between elderly (≥65 years old) and non-elderly (<65 years old) resuscitated severe sepsis and septic shock patients and determine predictors of death among elderly patients. Methods Retrospective cohort study including 848 severe sepsis and septic shock patients admitted to the intensive care unit between January 2006 and March 2012. Results Elderly patients accounted for 62.6% (531/848) and non-elderly patients for 37.4% (317/848). Elderly patients had a higher APACHE II score [22 (18-28) versus 19 (15-24); p<0.001], compared to non-elderly patients, although the number of organ dysfunctions did not differ between the groups. No significant differences were found in 28-day and in-hospital mortality rates between elderly and non-elderly patients. The length of hospital stay was higher in elderly compared to non-elderly patients admitted with severe sepsis and septic shock [18 (10-41) versus 14 (8-29) days, respectively; p=0.0001]. Predictors of death among elderly patients included age, site of diagnosis, APACHE II score, need for mechanical ventilation and vasopressors. Conclusion In this study population early resuscitation of elderly patients was not associated with increased in-hospital mortality. Prospective studies addressing the long-term impact on functional status and quality of life are necessary.


INTRODUCTION
Severe sepsis and septic shock are major reasons for intensive care unit (ICU) admission worldwide and they are associated with high morbidity and mortality rates, despite intense efforts towards early diagnosis and treatment. (1)(2)(3) Rivers et al. (4) proposed the concept of early goaldirected therapy for the treatment of severe sepsis and septic shock patients in 2001. This principle has been incorporated in the Surviving Sepsis Campaign International Guidelines for Management of Severe Sepsis and Septic Shock. (5) Accordingly, early identification, broad spectrum antibiotic administration and hemodynamic stabilization have been the cornerstone of severe sepsis and septic shock management. (6) The number of elderly patients (age ≥65 years old according to the World Health Organization) with severe sepsis and septic shock has been increasing steadily. (7) This population of elderly patients is characterized by an increased prevalence of chronic illness, comorbidities, frailty and functional impairment. (8,9) Nevertheless, while recent evidence has demonstrated that elderly patients submitted to complex therapeutic interventions during hospitalization showed benefits in long-term survival, (10) observational studies have shown that increased age is an independent predictor of death among septic and non-septic patients. (11)(12)(13) We postulated that elderly patients with severe sepsis or septic shock resuscitated following the Surviving Sepsis Campaign Guidelines have similar short-term mortality rates when compared to non-elderly patients with severe sepsis or septic shock.

OBJECTIVE
To perform a retrospective, single-center cohort study to compare the outcomes between elderly (≥65 years) and non-elderly (<65 years) severe sepsis and septic shock resuscitated patients and to determine the main predictors of death among elderly patients.

METHODS
This study was approved and the informed consent waived by the Hospital Israelita Albert Einstein Institutional Review Board (protocol 716,880 and CAAE: 32786114. 1.0000.0071). This study was conducted in a 41-bed medical-surgical ICU of a tertiary care at a private hospital in São Paulo, Brazil.

Patients
According to the institutional protocol for severe sepsis and septic shock resuscitation, all patients admitted to the emergency department or those in hospital who had been seen by the rapid response team and fulfilled the criteria for severe sepsis and septic shock were admitted to the ICU. All adult patients with severe sepsis or septic shock admitted to the ICU between January 2006 and December 2012 were included in this study. A case manager followed these patients until hospital discharge and their data were recorded.
The criteria for admission to hospital floor, intermediary care and ICU for patients with no diagnosis of severe sepsis and septic shock was based on the clinical judgment of the attending physician. However, severe sepsis and septic shock patients coming from the emergency department, or those who had been screened by the rapid response team were necessarily admitted to the ICU.
Septic shock was defined as sepsis-induced hypotension (systolic blood pressure <90mmHg or mean arterial blood pressure <65mmHg or a drop of >40mmHg in the absence of another cause of hypotension) despite adequate fluid resuscitation. Elderly patients were defined according to the World Health Organization as those aged ≥65 years.

Early goal-directed therapy
All patients were resuscitated following the institutional protocol for severe sepsis and septic shock. The onset of treatment was defined as the time of severe sepsis and septic shock diagnosis. Once a patient was diagnosed with severe sepsis or septic shock, the 6-hour resuscitation bundle was initiated. This included blood sampling with measurement of arterial lactate level, collection of blood cultures before antibiotics administration, broadspectrum antibiotics administration within 1 hour of the onset and a fluid load with crystalloids (20mL/kg) or equivalent doses of colloids. (5) The early goal-directed therapy was applied to patients with severe sepsis associated with arterial lactate levels ≥4.0mmol/L or those who remained hypotensive (systolic blood pressure <90mmHg or MAP <65mmHg) despite fluid resuscitation with crystalloids (20mL/kg) or equivalent doses of colloids. After the diagnosis of severe sepsis or septic shock, the following therapeutic goals were targeted during the first 6-hours of resuscitation: central venous pressure between 8 and 12mmHg (12 to 15mmHg in mechanically ventilated patients), MAP ≥65mmHg, central venous oxygen saturation (SvcO 2 ) or mixed venous (SvO 2 ) ≥70% and 65%, and diuresis ≥0.5mL/kg/h.

Variables collected
Demographic data, number of comorbidities, location before ICU admission, number of new organ dysfunctions at severe sepsis and septic shock diagnosis, source of infection, Acute Physiology and Chronic Health Evaluation II (APACHE) score, (15) need for vasopressors, invasive mechanical ventilation, amount of fluids administered, in-hospital and ICU length of stay, inhospital and mortality at day 28 were collected.

Statistical analysis
Categorical variables were presented as absolute and relative frequencies. Continuous variables were presented as mean and standard deviation (SD) when normally distributed and as median and interquartile range (IQR) when not normally distributed (tested by the Kolmogorov-Smirnov test).
Patients were divided into two groups according to the age: elderly patients (≥65 years) and non-elderly patients (<65 years). Categorical data were compared between elderly and non-elderly patients with the χ 2 test or Fisher's exact test when appropriate. Continuous data were compared with the independent t test when normally distributed and with the Mann-Whitney U test in the case of non-normal distribution.
A univariate logistic regression analysis was first performed to identify which factors or predictors were associated with in-hospital mortality in all study patients and then only among the elderly patients. Predictors that showed a p value ≤0.20 in the univariate analysis were entered into the multivariate analysis. A multivariate logistic regression analysis with a backward elimination procedure was undertaken to obtain an adjusted odds ratio (OR) along with 95% confidence interval (95%CI) and define which variables were independently associated with inhospital mortality between all study patients and then only among the elderly patients. Statistical tests were 2-sided and p<0.05 was considered statistically significant. Statistical analyses were performed using IBM™ Statistical Package for the Social Science (SPSS™) version 20.0 for Windows.
The main source of infection in elderly and nonelderly patients was the respiratory tract (57.8% versus 45.4%, for elderly and non-elderly patients; p<0.001) while intra-abdominal infections were more common in non-elderly patients.

Site of diagnosis
The most common patient location at severe sepsis and septic shock diagnosis was the emergency department, with no difference between elderly and non-elderly patients (50.8% versus 48.9%; p=0.619) ( Table 1).
A large proportion of non-elderly patients were diagnosed on the hospital floor (33.8% versus 19.6%, for non-elderly and elderly patients; p<0.001), while intermediary care was the most frequent site of diagnosis for elderly patients in comparison to nonelderly patients (17.5% versus 5.7%; p<0.001) ( Table 1).

Administered treatments
Compliance with the institutional protocol for severe sepsis and septic shock resuscitation has been published elsewhere. (3) The elderly patients received less fluid [median (IQR)] during the initial 6-hours of resuscitation than the non-elderly patients [1.8 (1.0 to 2.5) versus 2.0 (1.4 to 3.0) liters, for elderly and non-elderly patients; p=0.001]. The need for vasopressors (58.8% versus 58.7, for elderly and non-elderly patients; p=0.981) and mechanical ventilation (38.4% versus 38.2%, for elderly and non-elderly patients; p=0.943) did not differ between the groups.

Outcomes
In-hospital mortality and mortality at day 28 did not differ between elderly and non-elderly severe sepsis and septic shock patients ( Table 3). The median (IQR) length of hospital stay was higher in elderly compared to non-elderly patients admitted with severe sepsis [15 (8-34) versus 12 (6-24) days; p=0.027] or septic shock [21 (11-47) versus 18 (9-36) days; p=0.016]. The median length of ICU stay did not differ between elderly and non-elderly severe sepsis and septic shock patients.

Predictors of death
The univariate and multivariate logistic regression analysis addressing the predictors of death in all septic patients and only among the elderly patients are presented in tables 4 and 5, respectively.

DISCUSSION
This study demonstrated that in-hospital and 28-day mortality rates were not different between elderly and non-elderly patients submitted to early goal-directed therapy for the treatment of severe sepsis and septic shock. However, increased age was an independent predictor of in-hospital death among elderly patients. Therapeutic goals included in the first 6 hours of resuscitation from the initial diagnosis of severe sepsis and septic shock were achieved similarly in both groups. Nevertheless, less fluid for hemodynamic stabilization was administered to the elderly patients. Traditionally, elderly patients receive less intensive treatment compared to non-elderly patients, probably due to the possibility of deleterious effects of an aggressive therapy and fear of fluid overload. (16) Recently, increased acceptance of complex ICU interventions in older patients was associated with greater intensity of treatment and improved survival. (10) This is perhaps the result of increased experience with the care of elderly patients over the years and technical improvements such as protocols associated with hemodynamic monitoring tools and continuous renal replacement therapy, representing a true evolution in practice throughout the times. Our results confirm these findings, as the proportion of elderly patients receiving mechanical ventilation and vasopressors was not different from that in younger patients.
Our findings confirm the tolerance of elderly patients to an early goal-directed therapy algorithm for severe sepsis and septic shock resuscitation, with no differences in mortality compared to non-elderly patients. These results have important clinical implications, since there is an increasing demand for ICU admission of elderly patients, which is often associated with high costs and limited availability of ICU beds worldwide. (7,17,18) Our results support the concept that ICU admission and early goal-directed therapy implementation should not be denied to elderly patients showing severe sepsis and septic shock.
The impact of age itself on higher mortality rates due to sepsis is not uniformly observed in epidemiological investigations. (11,19,20) Other retrospective analysis demonstrated that age was associated with a significantly increased risk of death in elderly patients with severe sepsis or septic shock. (11,18) Nonetheless, these studies rely on administrative databases for sepsis diagnosis, which may be inaccurate and lacking important aspects such as adherence to the proposed treatment. Similarly, in our study, after adjustments for baseline patient's characteristics in a multivariable logistic regression  The site of diagnosis, the presence of liver cirrhosis, APACHE II score, the arterial blood lactate level, the number of organ dysfunction and the need for mechanical ventilation were independently associated with increased risk of in-hospital death among severe sepsis and septic shock patients. Increased age, the site of diagnosis, APACHE II score, the need for mechanical ventilation and vasopressor administration were independently associated with increased risk of inhospital death among elderly patients. model, age was independently associated with increased risk of in-hospital mortality in elderly patients with severe sepsis or septic shock.
Most studies addressing mortality in septic patients have focused on short-term endpoints. (4,12) Few observational studies have addressed the long-term prognosis of elderly severe sepsis and septic shock patients submitted to early goal-directed therapy. Lemay et al. reported a 1-year mortality rate of 31% in elderly patients with severe sepsis. (20) Nevertheless, adherence to specific therapeutic goals besides antibiotics administration was not reported and the study relied on administrative database for sepsis diagnosis. Similarly, Wang et al. described a 1-year mortality rate of 23% in a population of adults aged 45 or older, with sepsis defined as hospitalization or treatment in the emergency department for a serious infection with the presence of two or more systemic inflammatory response criteria, with no mention of ICU patients with severe sepsis or septic shock. (21) Recently, it was shown that long-term survival in elderly patients with circulatory failure (including sepsis) is poor, with mortality rates of 92 and 97% after 6 and 12 months, respectively. (13) These findings support the hypothesis that excess longterm mortality persists among those suffering from sepsis, probably because sepsis triggers an independent pathophysiological process leading to early death.
Respiratory infections accounted for the most sepsis cases in elderly patients, whereas abdominal infections were the most common cause in younger patients, a finding which has not been confirmed by other authors, where respiratory infection was the major source of infection in both elderly and non-elderly patients. (11) A possible explanation for this interesting finding is the greater incidence of liver cirrhosis and solid organ transplantation in younger patients, probably reflecting a high proportion of patients with spontaneous bacterial peritonitis.
Our study has limitations. First, we were unable to evaluate the functional status before and after ICU discharge. Functional status has been related to pre-existing underlying factors, and it plays a greater role than chronological age in the outcome of elderly patients with severe sepsis and septic shock. (22) Second, this was a single center and retrospective study, which potentially limits the generalizability of our findings. Lastly, we used 65 as the cut-off age following the definition of elderly by World Health Organization. However, as pointed out earlier, the chronological age is not always representative of the functional condition of the patients.

CONCLUSION
In this study population of severe sepsis and septic shock patients, early resuscitation of elderly patients was not associated with increase in mortality. Elderly patients with severe sepsis or septic shock may benefit from aggressive resuscitation and advanced treatment modalities. However, prospective studies are warranted to address long-term impact of resuscitation maneuvers on functional status and quality of life.