Aging Clin Exp Res DOI 10.1007/s40520-017-0821-9
ORIGINAL ARTICLE
The value of C-reactive protein in infection diagnosis and prognosis in elderly patients Omer Karasahin1 · Pınar Tosun Tasar2 · Ozge Timur3 · Filiz Yıldırım3 · Dogan Nasır Binici3 · Sevnaz Sahin4
Received: 26 June 2017 / Accepted: 3 August 2017 © Springer International Publishing AG 2017
Abstract Background The aim of this study was to determine the value of C-reactive protein level in the diagnosis and prognosis of infection in elderly patients. Study population This prospective study included inpatients in the palliative care unit during the 1-year period between January 2016 and January 2017. Patients’ demographic data, Acute Physiology and Chronic Health Evaluation score, and Charlson Comorbidity Index were recorded. Results A total of 233 patients were included in the study. A total of 199 instances of infection were diagnosed in 175 of those patients; 75.3% of the infections were detected at admission and 24.7% during hospitalization. At a cut-off value of 4.82, CRP value had 81.0% specificity and 75.4% sensitivity in the diagnosis of infection. Among the patients with infection, there was no difference between those who died and those who survived in terms of baseline CRP level, but a significant difference emerged in CRP level at 48 and 96 h. Factors which were found to significantly reduce survival time were the presence of chronic kidney disease, chronic obstructive pulmonary disease, hypoxia and tachycardia at admission, APACHE-II score over 20.5, initial
* Pınar Tosun Tasar [email protected] 1
Erzurum Regional Training and Research Hospital, Infectious Diseases Clinic, Erzurum, Turkey
2
Erzurum Regional Training and Research Hospital, Geriatrics Clinic, Erzurum, Turkey
3
Erzurum Regional Training and Research Hospital, Internal Medicine Clinic, Erzurum, Turkey
4
Division of Geriatrics, Department of Internal Medicine, Ege University Hospital, Izmir, Turkey
albumin level below 2.44 g/dL, and serum CRP clearance rates of less than 11% at 48 h and 20% at 96 h. Conclusion In elderly patients with infection, the initial CRP value alone does not have prognostic value, but changes observed in serial CRP measurement are a valid indicator of prognosis. Keywords Infection · Elderly · C-reactive protein
Introduction Infection is the leading cause of acute inflammation in elderly and pediatric patients. Infectious diseases are the primary cause of death for one-third of elderly individuals, and constitute a major risk of mortality for the remaining two-thirds [1]. Chronic diseases and associated organ failure contribute to the development of life-threatening infections in the elderly [2]. Infection-related mortality in the elderly is also attributed to immunosenescence [2–5]. Immune aging involves changes in acquired immunity in particular. The number of naive T cells falls dramatically. There is also a substantial change in the number of B cells, and specific antibody production is generally reduced despite sustained immunoglobulin levels. Consequently, vaccine responses are diminished and the ability to react to new infections is limited [3, 6, 7]. In contrast, the production of proinflammatory cytokines including interleukin (IL)-1, tumor necrosis factor alpha (TNF-α), and interferon gamma (IFN-γ) during acute infection is generally undiminished in elderly individuals [3, 8]. Anti-inflammatory mechanisms in the elderly are unable to adequately control the induction of proinflammatory cytokines in response to septic stimuli. This immune dysfunction is accompanied by a distinctly more procoagulant state in older patients. These molecular events increase
13
Vol.:(0123456789)
the risk of mortality due to serious infection in the elderly [3, 5]. Moreover, common signs of infection in young adults, especially fever and leukocytosis, are less frequent or absent in older adults [9]. This makes diagnosing infection in the elderly difficult. However, early diagnosis and treatment are critical due to the high incidence of morbidity and mortality in these patients [1, 2]. For these reasons, biomarkers are utilized. Elderly individuals generally have higher acute IL-6 production compared to younger adults, but their response time is prolonged [3, 5]. C-reactive protein (CRP) is synthesized in the liver in response to IL-6 in particular, as well as IL-1 and TNF-α. CRP level is a convenient, repeatable, and inexpensive biomarker used in infection diagnosis, prognosis, and evaluation of treatment response [10]. The aim of this study was to determine the value of CRP level in the diagnosis and prognosis of infection in elderly patients.
Methods This prospective, observational study was approved by the ethics committee of the Erzurum Regional Training and Research Hospital. Study population Geriatric inpatients who were treated for 48 h or more in the palliative care unit of our hospital between January 2016 and January 2017 were included in the study. Patients who were younger than 65 years old, were hospitalized for less than 48 h, or who were repeatedly admitted to the palliative care unit were not included. Definitions and data collection Infection was defined as a pathologic process in a normally sterile tissue, fluid, or body cavity caused by a pathogenic microorganism and/or treated with antibiotic therapy due to clinical and radiologic findings consistent with infection despite negative microbiological results or inability to take culture sample. In addition, the Centers for Disease Control and Prevention (CDC) 2008 criteria were used to diagnose hospital infections [11]. Blood and urine samples were collected from all patients at admission, and sputum samples were obtained for culturing from patients with respiratory symptoms. Cultures were repeated after 48 h to determine microbiologic response according to the focus of infection. Clinical (fever, blood pressure, pulse, and oxygen saturation) and laboratory findings, demographic data, comorbid diseases, Acute Physiology and Chronic Health Evaluation (APACHE-II) score, and Charlson Comorbidity Index (CCI) at time of admission were recorded for all patients.
13
Aging Clin Exp Res
The clinical and laboratory findings at time of hospital admission are presented as initial (0-h) values. For patients diagnosed with infection, clinical (fever, blood pressure, pulse, and oxygen saturation) and laboratory findings were also recorded for the first 5 days after the initiation of antibiotic therapy. The first endpoint of the study was day 5, and the second endpoint was the day of discharge or death. CRP clearance (cCRP) was calculated based on the serum CRP level measured at time of infection diagnosis (0-h) and the 48- and 96-h serum CRP levels. The results were evaluated as percentage (%): (
48-h cCRP = (0-h CRP − 48 − hr CRP) / 0-h CRP × 100
) .
96-h cCRP = (0-h CRP − 96-h CRP) / 0-h CRP × 100
Statistical analysis The data were analyzed using SPSS version 21.0 statistical software package. Categorical descriptive data were expressed as frequency distribution and percentage; continuous variables were expressed as mean ± standard deviation and median (maximum–minimum). To assess the diagnostic value of the initial CRP level in infection, receiver-operating characteristic (ROC) analysis was used to determine measurement and cut-off points. The Youden index (J = Sensitivity + Specificity − 1) was utilized. Statistical comparisons were made between patients who developed infection and those who did not, and between infected patients who died and those who survived. Categorical data were compared using the Chi-squared test; the continuous data did not meet the assumptions of parametric hypothesis and were compared with the nonparametric Kruskal–Wallis and Mann–Whitney U tests. Kaplan–Meier analysis was done to determine risk factors affecting survival. Of these risk factors, ROC analysis and the Youden index were used to identify cut-off points for APACHE-II score, initial albumin values, and 48- and 96-h cCRP in predicting mortality. Cut-off values determined for mortality prediction were 20.5 for APACHE-II score, 2.44 g/dL for albumin, 11% for 48-h cCRP, and 20% for 96-h cCRP. Variables were grouped according to those above and those below the identified cut-off points. Kaplan–Meier analysis revealed the presence of chronic obstructive pulmonary disease (COPD), chronic kidney disease (CKD), initial hypoxia ( SO2 100/min), as well as APACHE-II score >20.5, initial albumin
ORIGINAL ARTICLE
The value of C-reactive protein in infection diagnosis and prognosis in elderly patients Omer Karasahin1 · Pınar Tosun Tasar2 · Ozge Timur3 · Filiz Yıldırım3 · Dogan Nasır Binici3 · Sevnaz Sahin4
Received: 26 June 2017 / Accepted: 3 August 2017 © Springer International Publishing AG 2017
Abstract Background The aim of this study was to determine the value of C-reactive protein level in the diagnosis and prognosis of infection in elderly patients. Study population This prospective study included inpatients in the palliative care unit during the 1-year period between January 2016 and January 2017. Patients’ demographic data, Acute Physiology and Chronic Health Evaluation score, and Charlson Comorbidity Index were recorded. Results A total of 233 patients were included in the study. A total of 199 instances of infection were diagnosed in 175 of those patients; 75.3% of the infections were detected at admission and 24.7% during hospitalization. At a cut-off value of 4.82, CRP value had 81.0% specificity and 75.4% sensitivity in the diagnosis of infection. Among the patients with infection, there was no difference between those who died and those who survived in terms of baseline CRP level, but a significant difference emerged in CRP level at 48 and 96 h. Factors which were found to significantly reduce survival time were the presence of chronic kidney disease, chronic obstructive pulmonary disease, hypoxia and tachycardia at admission, APACHE-II score over 20.5, initial
* Pınar Tosun Tasar [email protected] 1
Erzurum Regional Training and Research Hospital, Infectious Diseases Clinic, Erzurum, Turkey
2
Erzurum Regional Training and Research Hospital, Geriatrics Clinic, Erzurum, Turkey
3
Erzurum Regional Training and Research Hospital, Internal Medicine Clinic, Erzurum, Turkey
4
Division of Geriatrics, Department of Internal Medicine, Ege University Hospital, Izmir, Turkey
albumin level below 2.44 g/dL, and serum CRP clearance rates of less than 11% at 48 h and 20% at 96 h. Conclusion In elderly patients with infection, the initial CRP value alone does not have prognostic value, but changes observed in serial CRP measurement are a valid indicator of prognosis. Keywords Infection · Elderly · C-reactive protein
Introduction Infection is the leading cause of acute inflammation in elderly and pediatric patients. Infectious diseases are the primary cause of death for one-third of elderly individuals, and constitute a major risk of mortality for the remaining two-thirds [1]. Chronic diseases and associated organ failure contribute to the development of life-threatening infections in the elderly [2]. Infection-related mortality in the elderly is also attributed to immunosenescence [2–5]. Immune aging involves changes in acquired immunity in particular. The number of naive T cells falls dramatically. There is also a substantial change in the number of B cells, and specific antibody production is generally reduced despite sustained immunoglobulin levels. Consequently, vaccine responses are diminished and the ability to react to new infections is limited [3, 6, 7]. In contrast, the production of proinflammatory cytokines including interleukin (IL)-1, tumor necrosis factor alpha (TNF-α), and interferon gamma (IFN-γ) during acute infection is generally undiminished in elderly individuals [3, 8]. Anti-inflammatory mechanisms in the elderly are unable to adequately control the induction of proinflammatory cytokines in response to septic stimuli. This immune dysfunction is accompanied by a distinctly more procoagulant state in older patients. These molecular events increase
13
Vol.:(0123456789)
the risk of mortality due to serious infection in the elderly [3, 5]. Moreover, common signs of infection in young adults, especially fever and leukocytosis, are less frequent or absent in older adults [9]. This makes diagnosing infection in the elderly difficult. However, early diagnosis and treatment are critical due to the high incidence of morbidity and mortality in these patients [1, 2]. For these reasons, biomarkers are utilized. Elderly individuals generally have higher acute IL-6 production compared to younger adults, but their response time is prolonged [3, 5]. C-reactive protein (CRP) is synthesized in the liver in response to IL-6 in particular, as well as IL-1 and TNF-α. CRP level is a convenient, repeatable, and inexpensive biomarker used in infection diagnosis, prognosis, and evaluation of treatment response [10]. The aim of this study was to determine the value of CRP level in the diagnosis and prognosis of infection in elderly patients.
Methods This prospective, observational study was approved by the ethics committee of the Erzurum Regional Training and Research Hospital. Study population Geriatric inpatients who were treated for 48 h or more in the palliative care unit of our hospital between January 2016 and January 2017 were included in the study. Patients who were younger than 65 years old, were hospitalized for less than 48 h, or who were repeatedly admitted to the palliative care unit were not included. Definitions and data collection Infection was defined as a pathologic process in a normally sterile tissue, fluid, or body cavity caused by a pathogenic microorganism and/or treated with antibiotic therapy due to clinical and radiologic findings consistent with infection despite negative microbiological results or inability to take culture sample. In addition, the Centers for Disease Control and Prevention (CDC) 2008 criteria were used to diagnose hospital infections [11]. Blood and urine samples were collected from all patients at admission, and sputum samples were obtained for culturing from patients with respiratory symptoms. Cultures were repeated after 48 h to determine microbiologic response according to the focus of infection. Clinical (fever, blood pressure, pulse, and oxygen saturation) and laboratory findings, demographic data, comorbid diseases, Acute Physiology and Chronic Health Evaluation (APACHE-II) score, and Charlson Comorbidity Index (CCI) at time of admission were recorded for all patients.
13
Aging Clin Exp Res
The clinical and laboratory findings at time of hospital admission are presented as initial (0-h) values. For patients diagnosed with infection, clinical (fever, blood pressure, pulse, and oxygen saturation) and laboratory findings were also recorded for the first 5 days after the initiation of antibiotic therapy. The first endpoint of the study was day 5, and the second endpoint was the day of discharge or death. CRP clearance (cCRP) was calculated based on the serum CRP level measured at time of infection diagnosis (0-h) and the 48- and 96-h serum CRP levels. The results were evaluated as percentage (%): (
48-h cCRP = (0-h CRP − 48 − hr CRP) / 0-h CRP × 100
) .
96-h cCRP = (0-h CRP − 96-h CRP) / 0-h CRP × 100
Statistical analysis The data were analyzed using SPSS version 21.0 statistical software package. Categorical descriptive data were expressed as frequency distribution and percentage; continuous variables were expressed as mean ± standard deviation and median (maximum–minimum). To assess the diagnostic value of the initial CRP level in infection, receiver-operating characteristic (ROC) analysis was used to determine measurement and cut-off points. The Youden index (J = Sensitivity + Specificity − 1) was utilized. Statistical comparisons were made between patients who developed infection and those who did not, and between infected patients who died and those who survived. Categorical data were compared using the Chi-squared test; the continuous data did not meet the assumptions of parametric hypothesis and were compared with the nonparametric Kruskal–Wallis and Mann–Whitney U tests. Kaplan–Meier analysis was done to determine risk factors affecting survival. Of these risk factors, ROC analysis and the Youden index were used to identify cut-off points for APACHE-II score, initial albumin values, and 48- and 96-h cCRP in predicting mortality. Cut-off values determined for mortality prediction were 20.5 for APACHE-II score, 2.44 g/dL for albumin, 11% for 48-h cCRP, and 20% for 96-h cCRP. Variables were grouped according to those above and those below the identified cut-off points. Kaplan–Meier analysis revealed the presence of chronic obstructive pulmonary disease (COPD), chronic kidney disease (CKD), initial hypoxia ( SO2 100/min), as well as APACHE-II score >20.5, initial albumin
