Egyptian Journal of Chest Diseases and Tuberculosis xxx (2017) xxx–xxx Contents lists available at ScienceDirect Egyptian Journal of Chest Diseases and Tuberculosis journal homepage: www.sciencedirect.com Prognostic values of pneumonia severity index, CURB-65 and expanded CURB-65 scores in community-acquired pneumonia in Zagazig University Hospitals Samah M Shehata a,⇑, Ashraf E Sileem a, Noha E Shahien b a b Chest Department, Faculty of Medicine, Zagazig University, Egypt Tropical Medicine Department, Faculty of Medicine, Zagazig University, Egypt a r t i c l e i n f o Article history: Received January 2017 Accepted 11 January 2017 Available online xxxx Keywords: Community-acquired pneumonia Expanded CURB-65 CURB-65 Pneumonia severity index Prognosis a b s t r a c t Introduction: Assessment of severity and site of care decisions for community-acquired pneumonia patients (CAP) are very important for patients’ safety and optimal use of resources Late admission to the intensive care unit (ICU) leads to increase the rate of mortality in CAP We aimed to evaluate the effectiveness of the new expanded CURB-65 score in comparison with other pneumonia severity scoring systems (PSI, CURB-65) in predicting CAP patients’ outcomes Methods: a prospective study included 250 consecutive patients hospitalized for CAP at Chest and Tropical medicine Departments and ICUs at Zagazig University Hospitals, Egypt in the period between May 2016 and November 2016 Results: The mean age of patients was 59.17 ± 14.04 years, 56% of all patients had comorbid diseases As regards patients, outcomes (ICU admission rate, the median length of hospital stay, the overall 30-day mortality rate and need for invasive mechanical ventilation) were 29.6%, days 11.2% and 23.6% respectively COPD and liver cirrhosis were significantly associated with increased the 30-day mortality in our CAP patients Mortality rate increased with the severity of liver cirrhosis In the multivariate analysis (age P 65 years, LDH > 230 u/L, Albumin < 3.5 g/dL, Platelet count < 100 Â 109/L, SBP < 90 mmHg or DBP 60 mmHg, septic shock and Confusion) were the independent predictors of the 30-day mortality Expanded CURB-65 was correlated with severity of liver disease guided by Child Pugh score (r (0.34), pvalue (0.01) The 30-day mortality was lower in expanded CURB-65 score (1–4) about 4.1% than PSI class (I–III) and CURB-65 score (0–2) While, the 30-day mortality was higher in expanded CURB-65 score (5–8) about 25.9% than PSI class (IV–V) 14.2% and lastly CURB-65 score (3–5) 18.4% Expanded CURB-65 score (5–8) was associated with more frequent ICU admission about 49.4% than other two scores The expanded CURB-65 scoring system was the best predictor of 30-day mortality, ICU admission and need for mechanical ventilation in CAP patients as it had the highest sensitivity, negative predictive value and largest area under the ROC curve Conclusions: Expanded CURB-65 score is simple, objective and more accurate scoring system for evaluation of CAP severity and can improve the efficiency of predicting the mortality in CAP patients, better than CURB-65 and PSI scores Also, Expanded CUEB-65 may generate new therapeutic and prognostic modality in CAP especially in patients with liver cirrhosis Ó 2017 The Egyptian Society of Chest Diseases and Tuberculosis Production and hosting by Elsevier B.V This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-ncnd/4.0/) Introduction Peer review under responsibility of The Egyptian Society of Chest Diseases and Tuberculosis ⇑ Corresponding author at: Chest Department, Faculty of Medicine, Zagazig University, El-Sharkia, Egypt E-mail address: Sama7she7ata2000@yahoo.com (S.M Shehata) Community acquired pneumonia (CAP) is one of the most common infectious causes of death in the world with a mortality rate of 1% in outpatient settings and higher than 50% in hospitalized patients The suboptimal management is one of the most important reasons of the high mortality rate, regarding antibiotic treatment, or the identification of individuals who required intensive http://dx.doi.org/10.1016/j.ejcdt.2017.01.001 0422-7638/Ó 2017 The Egyptian Society of Chest Diseases and Tuberculosis Production and hosting by Elsevier B.V This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) Please cite this article in press as: S.M Shehata et al., Prognostic values of pneumonia severity index, CURB-65 and expanded CURB-65 scores in community-acquired pneumonia in Zagazig University Hospitals, Egypt J Chest Dis Tuberc (2017), http://dx.doi.org/10.1016/j.ejcdt.2017.01.001 S.M Shehata et al / Egyptian Journal of Chest Diseases and Tuberculosis xxx (2017) xxx–xxx care unit admission Early identification of risk factors in these patients allows earlier intervention and, thus, improvement of the outcomes [1,2] During last decades, several scoring systems as PSI, CURB-65 and SMART-COP have been developed to assess pneumonia severity PSI consists of 20 variables and it is accurate in predicting the 30-day mortality, but its complexity decrases its clinical application [3] On the other hand, CURB-65 features are simple [4] However, the age and complications of the patient in both PSI and CURB65 carry heavier weight, underestimating the severity in young patients and falsely referring the elderly CAP patients as severe [5] Moreover, both PSI and CURB-65 cannot detect patients who need to be referred to the ICU, while SMART-COP can compensate this function [6] The SMART-COP, (Systolic blood pressure, Multilobar infiltrates, Albumin, Respiratory rate, Tachycardia, Confusion, Oxygen and pH) can give better accuracy for prediction of the need for intensive respiratory or vasopressor support, but it is still a complicated process to calculate multiple points for different variables [7] Liu et al [8] developed a new simpler and more effective scoring system, named expanded-CURB-65 including: Confusion, Urea >7 mmol/L, Respiratory rate P30/min, low systolic ( 230 u/L, Albumin < 3.5 g/dL, Platelet count < 100 Â 109/L It expands the independent risk factors into variables in assessing CAP severity, significantly increases high-risk patients identification, through decreasing the relative weight of age and blood pressure, and excluding the use of imaging and comorbid illnesses in the calculation Aim of the work To evaluate effectiveness of the new expanded CURB-65 score in comparison with other pneumonia severity scoring systems (PSI, CURB-65) in predicting CAP patients0 outcomes (ICU admission, Length of hospital stay, 30-day mortality and need for invasive mechanical ventilation) in Zagazig University Hospitals Patients and methods This was a prospective study included 250 consecutive patients hospitalized for CAP at Chest and Tropical medicine Departments and ICUs at Zagazig University Hospitals, Egypt in the period between May 2016 and November 2016 Informed consent was taken from all patients before being enrolled in the study Inclusion criteria: Patients who were 18 years or more, admitted from the community, had two or more clinical signs and symptoms related to pneumonia (fever >38 °C, cough, dyspnea, chest pain or crackles on auscultation), and presented by new infiltration on chest radiography were included in the study CAP was diagnosed according to Infectious Disease Society of America (IDSA)/American Thoracic Society (ATS) guidelines on the management of community-acquired pneumonia in adults [9] Patients were excluded if: they were immunocompromised (using immunosuppressive drugs, having a human immunodeficiency virus infection, malignancy), had been admitted to hospital and/or used any antibiotic in last two weeks All the data were collected from each subject, including demographic data, co-morbidity diseases, physical examination, laboratory and radiologic findings The laboratory findings were analyzed within 24 h after admission including (CBC, Liver and kidney function tests, ABG, Fasting blood glucose, Serum Na+, Serum LDH and blood culture) The diagnosis of comorbid diseases was done: COPD was diagnosed by history, clinically, radiologically and spirometry if possi- ble Liver cirrhosis was diagnosed clinically, laboratory and radiologically, cirrhotic patients were classified according to Child Pugh scores, into grade A, B, C [10] Congestive heart failure was confirmed by history, clinically and echocardiogram Cerebrovascular disease diagnosed by history, clinically, CT and/or MRI Direct admission to ICU was done if the patient had one major or minor of the criteria according to IDSA/ATS guidelines [9] The diagnosis of sepsis was done according to the International Consensus Criteria published in 2003 [11] The scores of PSI, CURB-65, and expanded CURB-65 were calculated for all patients Patients were subsequently classified into two levels of risk groups: (a) PSI: Non-severe risk group (classes I-III); and severe risk group (class IV, V); (b) CURB-65: Nonsevere risk group (scores 0–2) and severe risk group (scores 3–5) (c) Expanded CURB-65: Non-severe risk group (score 0–4) and severe risk group (score 5–8) [12–14] The final outcomes: Length of stay (LOS) in hospital, ICU admission, 30-day mortality after hospital admission and need for invasive mechanical ventilation were compared Statistical analysis All data were collected, tabulated and statistically analyzed using SPSS 20.0 for windows (SPSS Inc., Chicago, IL, USA) and MedCalc 13 for windows (MedCalc Software bvba, Ostend, Belgium) Quantitative data were expressed as median (IQR), and qualitative data were expressed as absolute frequencies (number) & relative frequencies (percentage) Percent of categorical variables were compared using Chi-square test or Fisher’s exact test when appropriate To determine predictors for 30 day-mortality, univariate logistic regression was done Multivariate logistic regression analysis model was done by enter method Receiver operating characteristic (ROC) curve analysis was used to identify optimal cut-off values of PSI, CURB-65 and expanded CURB-65 with maximum sensitivity and specificity for prediction of ICU admission, 30 day-mortality, LOS > days and need for invasive MV Area under Curve (AUROC) was also calculated, criteria to qualify for AUC were as follows: 0.90–1 = excellent, 0.80–0.90 = good, 0.70– 0.80 = fair; 0.60–0.70 = poor; and 0.50–0.6 = fail The optimal cutoff point was established at point of maximum accuracy All tests were two sided P < 0.05 was considered statistically significant (S), p < 0.001 was considered highly statistically significant (HS), and p P 0.05 was considered non statistically significant (NS) Results Patients’ characteristics and outcomes of the CAP patients enrolled in the study were summarized in (Table 1) The mean age for overall patients was 59.17 ± 14.04 years, 62.8% of patients were males and 37.2% were females 56% of all patients had comorbid diseases: COPD (11.6%), congestive heart failure (10%), DM (16.4%), chronic renal diseases (4.8%), cerebrovascular diseases (6.4%) and liver cirrhosis patients were 17 (6.8%) who classified according to child Pugh into Grade A: patients (23%), Grade B: patients (47%), Grade C: patients (30%) (Data not shown) Some laboratory findings which were done within 24 h of hospital admission: Platelet count 230 U/L was present in 112 patients As regards patients0 outcomes: Intensive care unit (ICU) admission rate was 29.6%, the median length of hospital stay was days and the overall 30-day mortality rate was 11.2% 23.6% of patients needed invasive mechanical ventilation Please cite this article in press as: S.M Shehata et al., Prognostic values of pneumonia severity index, CURB-65 and expanded CURB-65 scores in community-acquired pneumonia in Zagazig University Hospitals, Egypt J Chest Dis Tuberc (2017), http://dx.doi.org/10.1016/j.ejcdt.2017.01.001 S.M Shehata et al / Egyptian Journal of Chest Diseases and Tuberculosis xxx (2017) xxx–xxx Table Patients’ characteristics and outcomes of CAP patients enrolled in the study CAP patients (N = 250) No % Age Mean ± SD (years) Age P 65 years 59.17 ± 14.04 110 44% Sex Male Female 157 93 62.8% 37.2% Comorbidities Congestive heart failure COPD Chronic renal diseases Chronic liver diseases Cerebrovascular diseases Diabetes mellitus 140 25 29 12 17 16 41 56% 10% 11.6% 4.8% 6.8% 6.4% 16.4% Physical and laboratory findings at hospital admission Confusion Respiratory rate P 30/min Heart rate P 125/min SBP < 90 mmHg or DBP 60 mmHg LDH > 230 U/L Platelet count < 100 Â 103/mm3 Albumin < 3.5 g/dl 39 155 72 92 112 57 83 15.6% 62% 28.8% 36.8% 44.8% 22.8% 33.2% Patients0 outcomes LOS: Median (IQR) days 30-day mortality ICU admission Need for invasive mechanical ventilation (6–12) 28 74 59 11.2% 29.6% 23.6% N = Total number of patients; Quantitative data were expressed as mean ± SD; Qualitative data were expressed as number (percentage) Analysis of some Risk factors (measured immediately on hospital admission) and their relation to 30-day mortality using univariate and multivariate regression analyses (Table 2): univariate analysis showed that the following risk factors (age P 65 years, LDH > 230 u/L, Albumin < 3.5 g/dL, Platelet count < 100 Â 109/L, Confusion, Respiratory rate P 30/min, SBP < 90 mmHg or DBP 60 mmHg, HR P 125 bpm, septic shock, some comorbid diseases e.g COPD and liver cirrhosis) were significantly associated with increased the 30-day mortality in our CAP patients The number of CAP patients died from COPD and liver cirrhosis were (8 patients (27.6%), P-value 0.008) and (7 patients (41.2%), P-value 0.001) respectively Mortality rate increased with the severity of liver cirrhosis (3 patients and patients) were died in Child Pugh grades B, C respectively (Data not shown) While the multivariate analysis showed that (age P 65 years, LDH > 230 u/L, Albumin < 3.5 g/dL, Platelet count < 100 Â 109/L, SBP < 90 mmHg or DBP 60 mmHg, septic shock and Confusion) were the independent predictors of the 30-day mortality in the studied patients Patients0 outcomes including (ICU admission, LOS, need for mechanical ventilation and 30-day mortality) in relation to different severity scoring systems were presented in (Table 3): There was statistical significant difference between both subgroups of Expanded CURB-65 score as regards median LOS (P-value 0.001) The 30-day mortality was lower in expanded CURB-65 score (1– 4) about 4.1% than PSI class (I–III) was 4.9% and lastly CURB-65 score (0–2) which was 6.1% While, the 30-day mortality was higher in expanded CURB-65 score (5–8) about 25.9% than PSI class (IV–V) was 14.2% and lastly CURB-65 score (3–5) which was 18.4% Also, there was very high significance difference between both subgroups of Expanded CURB-65 score as regards 30-day mortality (P-value < 0.001) expanded CURB-65 score (5–8) was associated with more frequent ICU admission about 49.4% than PSI class (IV-V) which was 32.5% and lastly CURB-65 score (3–5) was 39.8%, with statistical significant difference between subgroups of both expanded CURB-65 and PSI scores (P-value < 0.001) Lastly, as regards the need for mechanical ventilation there was no statistical significance difference between either subgroups of all the three scores Expanded CURB-65 was correlated with severity of liver disease guided by child pugh score (r (0.34) p-value (0.01) (Data not shown) At AUC of ROC curve, Sensitivity and negative predictive values for prediction of 30-day mortality in Expanded CURB-65, PSI and CURB-65 scores were (75%, 60.71% and 53.57%) and (95.9%, 94.4% and 93.2%) respectively The expanded CURB-65 scoring system was the best predictor of 30-day mortality in CAP patients as it had the largest AUC (0.793) p-value < 0.0001, which was significantly higher than PSI > (AUC 0.740,) and CURB-65 > (AUC 0.706) (Table and Fig 1) At AUC of ROC curve, the sensitivity of the three scoring systems for prediction of ICU admission (expanded CURB-65, PSI and the CURB-65 score) were 45.45%, 21.21% and 30.30% respectively (Table 5) AUC of expanded CURB-65 score was 0.631 (p-value 0.0003), which was significantly higher than PSI (AUC 0.578) and CURB-65 score (AUC 0.545) (Fig 2) At AUC of ROC curve, the sensitivity of the three scoring systems for prediction of invasive mechanical ventilation (expanded CURB-65, PSI and the CURB-65) were 30.51%, 20.34% and 32.20% respectively (Table 6) AUC of expanded CURB-65 score was 0.588 (p-value 0.0459), which was significantly higher than PSI (AUC 0.561) and CURB-65 score (AUC 0.521) (Fig 3) Table Univariate and multivariate analyses of some risk factors associated with 30-day mortality in CAP patients Univariate analysis Age P 65 years Confusion RR P 30/min SBP < 90 mmHg or DBP 60 mmHg HR P 125/min Septic shock LDH > 230 U/L Platelet count < 100 Â 103/mm3 Albumin < 3.5 g/dl Chronic liver diseases COPD Constant b OR (95%CI) 0.981 2.064 1.142 1.108 1.206 2.189 3.023 2.144 0.659 1.683 0.861 3.01 7.87 3.13 3.02 3.34 8.92 20.55 8.53 2.01 5.38 2.38 Multivariate analysis (1.87–4.85) (3.36–18.42) (1.14–8.54) (1.35–6.79) (1.49–7.44) (3.55–22.44) (4.75–88.81) (3.67–19.79) 1.36–3.06 (1.24–23.33) (1.48–3.84) p B OR (95%CI)