PrognosticPredictionValueofqSOFA,SOFA,andAdmission...

Research ArticlePrognostic Prediction Value of qSOFA SOFA and AdmissionLactate in Septic Patients with Community-AcquiredPneumonia in Emergency Department

Haijiang Zhou 1 Tianfei Lan2 and Shubin Guo 1

1Department of Emergency Medicine Beijing Chao-yang Hospital Capital Medical UniversityBeijing Key Laboratory of Cardiopulmonary Cerebral Resuscitation Beijing China2Department of Allergy Beijing Shijitan Hospital Capital Medical University Beijing China

Correspondence should be addressed to Shubin Guo gsbchaoyangsinacn

Received 19 December 2019 Accepted 17 February 2020 Published 6 April 2020

Academic Editor Robert Derlet

Copyright copy 2020Haijiang Zhou et al)is is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Background Community-acquired pneumonia (CAP) is a leading cause of sepsis and common presentation to emergency department(ED) with a high mortality rate )e prognostic prediction value of sequential organ failure assessment (SOFA) and quick SOFA(qSOFA) scores in CAP in EDhas not been validated in detail)e aim of this research is to investigate the prognostic prediction value ofSOFA qSOFA and admission lactate compared with that of other commonly used severity scores (CURB65 CRB65 and PSI) in septicpatients with CAP in EDMethods Adult septic patients with CAP admitted between Jan 2017 and Jan 2019 with increased admissionSOFAge 2 from baseline were enrolled)e primary outcome was 28-daymortality)e secondary outcome included intensive care unit(ICU) admission mechanical ventilation and vasopressor use Prognostic prediction performance of the parameters above wascompared using receiver operating characteristic (ROC) curves KaplanndashMeier survival curves were compared using optimal cutoffvalues of qSOFA and admission lactate Results Among the 336 enrolled septic patients with CAP 89 patients died and 247 patientssurvived after 28-day follow-up )e CURB65 CRB65 PSI SOFA qSOFA and admission lactate levels were statistically significantlyhigher in the death group (Plt 0001) qSOFA and SOFA were superior and the combination of qSOFA+ lactate and SOFA+ lactateoutperformed other combinations of severity score and admission lactate in predicting both primary and secondary outcomes Patientswith admission qSOFAlt 2 or lactatele 2mmolL showed significantly prolonged survival than those patients with qSOFAge 2 orlactategt2mmolL (log-rank χ2 59825 Plt 0001) )e prognostic prediction performance of the combination of qSOFA and ad-mission lactate was comparable to the full version of SOFA (AUROC0833 vs 0795Z 1378P 0168 in predicting 28-daymortalityAUROC 0868 vs 0895 Z 1022 P 0307 in predicting ICU admission AUROC 0868 vs 0845 Z 0921 P 0357 in predictingmechanical ventilation AUROC 0875 vs 0821 Z 212 P 0034 in predicting vasopressor use)Conclusion qSOFA and SOFAweresuperior to CURB65 CRB65 and PSI in predicting 28-day mortality ICU admission mechanical ventilation and vasopressor use forseptic patients with CAP in ED Admission qSOFA with lactate is a convenient and useful predictor Admission qSOFAge 2 orlactategt2mmolL would be very helpful in discriminating high-risk patients with a higher mortality rate

1 Introduction

Community-acquired pneumonia (CAP) a major cause ofsepsis and the 8th leading cause of death is a commonrespiratory tract infection encountered in the emergencydepartment (ED) [1] CAP is caused by virus bacteria orfungi and its symptoms include cough chest pain fever anddyspnea Diagnosis of CAP is determined by symptoms

physical examinations laboratory findings and chest ra-diographs as well as etiologic agent culture Severity eval-uation is of vital importance for treatment location selectionempirical antimicrobial initiation and adjunctive andsupportive treatment adoption [2] For CAP severity as-sessment the CURB65 (confusion ureagt 7mmolL respi-ratory ratege 30min blood pressurelt 90mmHg systolicandor le60mmHg diastolic and agege 65 years) and

HindawiEmergency Medicine InternationalVolume 2020 Article ID 7979353 11 pageshttpsdoiorg10115520207979353

pneumonia severity index (PSI) score are most widely usedworldwide [3 4] Preliminary research studies demonstratedthat there were no significant differences in overall testperformance between CURB65 and PSI [5]

CAP accounts for substantial mortality worldwide witha high risk of developing respiratory failure and septic shock)e third international consensus definitions for sepsis andseptic shock generated new definitions and updated theclinical criteria [6] Sepsis was defined as life-threateningorgan dysfunction caused by a dysregulated host response toinfection and organ dysfunction was defined as an increasein the sequential organ failure assessment (SOFA) score of 2or higher [6] highlighting the clinical implication of (SOFA)score and qSOFA (respiratory ratege 22min altered men-tation and systolic blood pressurege 100mmHg) score [6]Lactate has been utilized as a marker to evaluate the acid-base homeostasis and perfusion status of patients Higherlactate levels are reported to be associated with higherhospital mortalities and longer length of ED and hospitalstay in unselected patients [7 8] However the prognosticvalue of SOFA score qSOFA score and lactate in patientswith CAP in the emergency department (ED) has not beenfully elucidated

In the present study we explored the prognostic pre-diction value of SOFA qSOFA and admission lactate inpatients with CAP in ED and the results were comparedwith that of other commonly used CAP severity scores(CURB65 CRB65 PSI)

2 Patients and Methods

)is was a single-center retrospective cohort study carriedout in Beijing Chao-yang Hospital Capital Medical Uni-versity which is a tertiary referral hospital located in thenorthern region of China with approximately 250000 an-nual ED visits per year )e Institutional Review Board andMedical Ethics Committee have approved this study andwritten informed consent was waived because of the ret-rospective design of this study

Adult patients with the discharge diagnosis of CAPadmitted between Jan 2017 and Jan 2019 were screened)einclusion criteria were agege 18 years new infiltrates on chestradiograph and two or more symptoms including coughfever dyspnea sputum production breathlessness andorchest pain [9] In accordance with sepsis 30 criteria we onlyenrolled CAP patients with increased SOFA scorege 2 frombaseline )e following patients were excluded from thisstudy (1) agelt 18 years (2) patients with metastatic tumoracquired immunodeficiency syndrome (AIDS) active tu-berculosis previous transplantation immunosuppressivetherapy and pregnancy (3) patients transferred from otherhospitals or diagnosed with hospital acquired pneumonia(4) patients with incomplete clinical laboratory or radio-graphic records (5) patients with increased admission SOFAscorelt 2 from baseline

Demographic characteristics of all enrolled patients onED arrival were collected and recorded by trained triagenurses on admission Past history comorbidities and vitalsigns (blood pressure heart rate breath rate and state of

consciousness) were also recorded Laboratory parameterson admission including full blood count hemoglobin level(HGB) hemocrit (HCT) platelet level (PLT) albumin(ALB) hepatic function (aspartate aminotransferase (AST)alanine aminotransferase (ALT) total bilirubin (TBIL)direct bilirubin (DBIL)) renal function (creatinine (CREA)blood urea nitrogen (BUN)) electrolytes and arterial bloodgas including lactate level were assessed and collectedCURB65 CRB65 PSI SOFA and qSOFA scores on ad-mission for each patient were calculated according to in-ternational criteria and analyzed utilizing data collected onED arrival

All patients were followed up for 28 days throughmedical records and 28-daymortality was the primary studyend point According to their prognosis after 28-day ad-mission patients were divided into death group and survivalgroup )e secondary outcome included ICU admissionmechanical ventilation use and vasopressor use

All analyses were performed using SPSS 220 statisticalsoftware package (SPSS Inc Chicago IL USA) Data withnormal distribution were described as meanplusmn standarddeviation and compared using Studentrsquos t-test Data withskewed distribution were expressed as median (interquartilerange) and compared using the MannndashWhitney U non-parametric test )e categorical variables were described aspercentages and compared using the chi-squared test orFisherrsquos exact test Receiver operating characteristics (ROC)curves for each predictor were constructed and the areaunder the curve (AUC) was determined to assess theirpredictive values Combination models of severity score andlactate were established using several logistic regressions tosave the predicted probabilities ROC curve analysis wasperformed using the saved probabilities as a new indicatorComparisons of each predictor were conducted usingMedCalc 150 Software (Acacialaan Ostend Belgium) AZ-testwas used for comparing the AUCs between different curvesFor comparison of the AUCs Z (A1minusA2)

(SE1)2 + (SE2)

21113969

was used the test values being Z005196 and Z001 258Based on the cutoff values sensitivity specificity positivepredictive value (PPV) and negative predictive value (NPV)were also calculated KaplanndashMeier survival curves were drawnusing cutoff values of qSOFA and lactate A two-tailed value ofPlt 005 was considered statistically significant

3 Results

A total of 561 patients were screened at recruitment and 225patients were excluded (Figure 1) A total of 336 patientswere finally included in our study group Of the 225 ex-cluded patients 16 patientsrsquo agelt 18 years old 74 patientswere transferred from other hospitals 10 patients were witha history of previous transplantation 5 patients were finallydiagnosed with pulmonary tuberculosis 3 patients were withpulmonary thromboembolism 6 patients were with lungcancer 4 patients were with HIV 57 patients were withincomplete medical records 30 patients were with admis-sion SOFA scorelt 2 from baseline and 20 patients were lostto follow-up with unknown prognosis

2 Emergency Medicine International

Of the 336 patients 89 patients were dead and 247patients survived after 28-day follow-up and the totalmortality rate was 265 (Table 1))e total mean age was 76(61 84) years and the male-to-female ratio was 173 1Comorbidities of enrolled patients include chronic ob-structive pulmonary disease (COPD) (119) cardiovas-cular disease (CDVD) (143) cerebrovascular disease(CBVD) (262) diabetes (229) chronic renal disease(CRD) (89) and hepatobiliary disease (HBD) (71))ere were no significant differences between the death andsurvival groups in age (P 0136) and male-female ratio(P 0914) CBVD was the most common comorbidity anda previous history of diabetes mellitus was more commonamong nonsurvivors (Table 1) 61 patients were admitted inICU and their age was older than those who were notadmitted in ICU (Table 1) As to laboratory parameters theALB level was significantly lower while the creatine andblood urea nitrogen (BUN) levels were significantly higherin the death group the patients admitted in ICU and thepatients who use mechanical ventilation or vasopressorsrespectively (Plt 005) (Tables 1 and 2) )e vital signs of thenonsurvivors the patients admitted in ICU and the patientsused mechanical ventilation or vasopressors were moreunstable (Plt 005) (Tables 1 and 2) )e CURB65 CRB65PSI SOFA qSOFA and admission lactate levels were sig-nificantly higher in the death group the ICU admissiongroup the mechanical ventilation group and the vaso-pressors use group (Plt 005) (Tables 1 and 2)

In predicting 28-day mortality ROC curve comparisonsshowed that the AUROC of qSOFA (0807) was the highestamong single predictors (Table 3 Figure 2) followed bySOFA (0795) PSI (0768) CURB65 (0744) CRB65 (0737)and admission lactate (0679) Moreover the qSOFA scorewas with the highest sensitivity (91) and negative pre-dictive value (NPV) (943) which highlighted the pre-diction performance of qSOFA However among thecombinations of severity score and admission lactate theAUROC of qSOFA+ lactate (0833) was the highest fol-lowed by the combination of SOFA+ lactate (0803)CRB65 + lactate (0776) PSI + lactate (0774) andCURB65 + lactate (0772) Multiple pairwise comparisons

among single predictors showed that there were no signif-icant differences between qSOFA and SOFA (Z 0373P 0709) qSOFA and PSI (Z 1296 P 0195) CURB65and PSI (Z 0940 P 0347) CURB65 and SOFA(Z 1499 P 0134) and PSI and SOFA (Z 0835P 0404) )ere were significant differences betweenqSOFA and CURB65 (Z 2333 P 0020) qSOFA andCRB65 (Z 2504 P 0012) qSOFA and lactate (Z 3246P 0001) and SOFA and lactate (Z 3143 P 0002)whereas pairwise comparisons among combinations of se-verity scores and lactate demonstrated that the combinationof qSOFA+ lactate demonstrated superiority over othercombinations except the combination of SOFA+ lactate)ere were no significant differences betweenCURB65 + lactate and CRB65 + lactate (Z 0244P 0807) CURB65 + lactate and PSI + lactate (Z 0062P 0951) CURB65+ lactate and SOFA+ lactate (Z 1054P 0292) PSI + lactate and SOFA+ lactate (Z 0993P 0321) and SOFA+ lactate and qSOFA+ lactate(Z 1187 P 0235)

)e ability to predict ICU admission was higher whenthe SOFA was used rather than either CURB65 CRB65qSOFA or PSI (Table 4 Figure 3) )ese differences werestatistically significant SOFA achieved the highest AUROC(0895) in predicting ICU admission followed by PSI(0837) qSOFA (0822) CURB65 (0774) lactate (0742)and CRB65 (0738) As to the comparisons of combinationsof severity scores and lactate SOFA+ lactate achieved thehighest AUROC (0902) followed by qSOFA+ lactate(0868) PSI + lactate (0840) CURB65 + lactate (0818) andCRB65 + lactate (0806) )ere were no significant differ-ences between SOFA+ lactate and qSOFA+ lactate(Z 1476 P 0140) and SOFA+ lactate and PSI + lactate(Z 1949 P 0051)

In predicting mechanical ventilation (Table 5 Figure 4)SOFA achieved the highest AUROC (0845) followed byqSOFA (0838) PSI (0780) CURB65 (0771) and CRB65(0758) SOFA and qSOFA had similar prediction perfor-mance (Z 0215 P 0830) As to the combinations ofseverity scores and lactate SOFA+ lactate (AUROC 0851)and qSOFA+ lactate (AUROC 0868) also had similarprediction value (Z 0746 P 0456) and demonstratedsuperiority over other combinations Among the singlepredictors in predicting vasopressor use (Table 6 Figure 5)the AUROC of SOFA (0821) and qSOFA (0820) was higherthan that of PSI (0767) CURB65 (0759) CRB65 (0746)and lactate (0758) while the difference between SOFA andqSOFA was not statistically significant (Z 0046P 0964) In terms of the combinations of severity scoresand lactate SOFA+ lactate (AUROC 0848) and qSO-FA+ lactate (AUROC 0875) also had similar predictionvalue (Z 1426 P 0154) and demonstrated superiorityover other combinations

As to the comparison of the prediction ability of SOFAand the combination of qSOFA and admission lactate nosignificant differences were found between them in pre-dicting 28-day mortality (0795 vs 0833 Z 1378P 0168) ICU admission (0895 vs 0868 Z 1022P 0307) and mechanical ventilation (0845 vs 0868

561 patients screened at recruitment

16 patientsrsquo age lt18 years old74 patients transferred from other hospitals

10 patients with previous history of transplantation5 patients diagnosed with tuberculosis

3 patients diagnosed with pulmonary embolism6 patients diagnosed with lung cancer

4 patients diagnosed with HIV

57 patients with incomplete medical records30 patients with admission SOFAlt2 from baseline

20 patients lost to follow-up

336 eligible septic patients with community-acquired pneumonia

89 nonsurvivors247 survivors

Figure 1 Flow chart of patients enrolled in our study

Emergency Medicine International 3

Z 0921 P 0357) while qSOFA+ lactate was signifi-cantly superior to SOFA in predicting vasopressor use (0821vs 0875 Z 212 P 0034)

)e mortality of CAP patients in ED who had aqSOFAge 2 or lactatege 2mmolL was 377 which was sig-nificant higher than those patients who had qSOFAlt 2 orlactatelt 2mmolL (52) (Table 7) For the secondary out-comes significant difference was also found between the twogroups in terms of ICU admission (273 vs 09) me-chanical ventilation (364 vs 26) and vasopressor use(464 vs 52) Moreover KaplanndashMeier survival curveanalyses revealed that the CAP patients in ED with admissionqSOFAlt 2 or lactatele 2mmolL showed significantly pro-longed survival than those patients with qSOFAge 2 or lac-tategt 2mmolL (Log-rank χ2 59825 Plt 0001) (Figure 6)

4 Discussion

CAP is one of themost common infections seen in EDwith awide spectrum of severity and pathogens Delayed treatmentmay result in serious consequences and even death)erefore timely diagnosis assessment of severity andtreatment can improve prognosis Severity evaluation is ofvital significance in the initial management of CAP Variousscoring systems of CAP exist )ey differ from each otherand they are used as tools to aid clinical diagnosis andtreatment however doctors should take clinical experienceinto consideration [2]

As far as the components are concerned the CURB65score is very similar to qSOFA score It comprises of fivevariables and divided patients into three broad risk groups as

Table 1 Baseline characteristics of enrolled patients with CAP in ED

All cohort Death Survival P ICU admission Non-ICU admission P

N () 336 89 (265) 247 (735) 61 (182) 275 (818)Age (yrs) 76 (61 84) 78 (67 85) 75 (61 83) 0136 81 (70 88) 74 (61 83) 0003Male n () 213 (634) 56 (629) 157 (636) 0914 40 (656) 173 (629) 0696Comorbidities n ()COPD 40 (119) 12 (135) 28 (113) 0592 16 (262) 24 (87) lt0001CDVD 48 (143) 18 (202) 30 (121) 0062 14 (230) 34 (124) 0033CBVD 88 (262) 28 (315) 60 (243) 0187 18 (295) 70 (255) 0515Diabetes 77 (229) 31 (348) 46 (186) 0002 24 (393) 53 (193) 0001CRD 30 (89) 9 (101) 21 (85) 0648 8 (131) 22 (80) 0205HBD 24 (71) 5 (56) 19 (77) 0515 4 (66) 20 (73) 0844Healthy 28 (83) 6 (67) 22 (89) 0526 8 (131) 20 (73) 0135

Laboratory resultsWBC (times109L) 100 (67 141) 110 (67 159) 98 (68 139) 0406 117 (75 159) 97 (66 139) 0046HGB (gL) 126 (115 137) 123 (111 134) 127 (117 138) 0125 123plusmn 22 127 (117 138) 0367HCT () 369 (318 408) 347plusmn 91 373 (329 409) 0037 355plusmn 93 373 (325) 0368PLT (times109L) 184 (131 251) 163 (123 249) 191 (136 251) 0311 155 (123 230) 191 (137 254) 0116ALB (gL) 362 (321 391) 340plusmn 59 367 (33 393) 0001 343plusmn 60 365 (325 391) 0044CREA (μmolL) 805 (608 1143) 92 (636 140) 759 (604 1058) 0011 955 (746 150) 759 (602 105) lt0001BUN (mmolL) 74 (53 110) 87 (60 134) 68 (51 104) 0010 104 (67 161) 68 (51 104) lt0001AST (UL) 30 (20 56) 33 (23 68) 30 (20 53) 0174 36 (23 64) 30 (20 54) 0240ALT (UL) 20 (14 37) 17 (12 35) 21 (14 37) 0225 16 (11 32) 21 (14 38) 0167TBIL (μmolL) 148 (96 230) 153 (103 235) 143 (95 227) 0318 151 (96 236) 147 (98 229) 0778DBIL (μmolL) 64 (40 106) 68 (44 113) 61 (38 102) 0089 61 (43 115) 64 (40 104) 0503K+ (mmolL) 38 (35 42) 39 (35 44) 38 (35 42) 0252 40 (36 44) 38 (35 42) 0050PaO2FiO2 303 (262 352) 285 (246 328) 313 (273 363) lt0001 286 (241 328) 310 (268 361) 0007

Vital signsSBP (mmHg) 132 (119 139) 122 (97 135) 134 (125 143) lt0001 122 (96 135) 132 (122 142) lt0001DBP (mmHg) 68 (65 76) 63 (55 70) 70 (66 78) lt0001 65plusmn 11 69 (65 76) lt0001HR (timesmin) 86 (78 94) 87 (79 106) 86 (78 94) 0009 86 (79 106) 86 (78 94) 0024

Severity scoresCURB65 2 (2 3) 3 (3 4) 2 (1 3) lt0001 3 (3 4) 2 (2 3) lt0001CRB65 2 (1 3) 3 (2 3) 2 (1 2) lt0001 3 (2 3) 2 (1 2) lt0001PSI 128plusmn 40 157plusmn 35 120plusmn 38 lt0001 174 (150 191) 121plusmn 36 lt0001SOFA 3 (2 5) 6 (4 8) 3 (2 4) lt0001 7 (6 9) 3 (2 4) lt0001qSOFA 2 (1 2) 3 (2 3) 1 (1 2) lt0001 3 (2 3) 1 (1 2) lt0001Lactate (mmolL) 14 (11 21) 18 (13 29) 13 (11 18) lt0001 25 (16 42) 13 (11 18) lt0001

Data are presented as n n () or median (QL QU) CAP community-acquired pneumonia ED emergency department SCAP severe community-acquiredpneumonia NSCAP nonsevere community-acquired pneumonia COPD chronic obstructive pulmonary disease CDVD cardiovascular disease CBVDcerebrovascular disease CRD chronic renal disease HBD hepatobiliary disease WBC white blood cell HGB hemoglobin HCT hematocrit PLT plateletALB albumin CREA creatinine BUN blood urea nitrogen AST aspartate aminotransferase ALT alanine aminotransferase TBIL total bilirubin DBILdirect bilirubin SBP systolic blood pressure DBP diastolic blood pressure HR heart rate CURB65 confusion ureagt 7mmolL respiratory ratege 30minblood pressurelt 90mmHg systolic andor le60mmHg diastolic and agege 65 years CRB65 confusion respiratory ratege 30min blood pressurelt 90mmHgsystolic andor le60mmHg diastolic and agege 65 years PSI pneumonia severity index SOFA sequential organ failure assessment qSOFA quick sequentialorgan failure assessment


follows Scores 0-1 low risk for 30-day mortality Score 2intermediate risk of 30-day mortality Scores 3ndash5 high riskof 30-day mortality [3] It was primarily designed to predictmortality and identify low-risk patients potentially suitablefor ambulatory management and has been widely utilized inpatients with CAP [10] )e CURB65 score has been ex-tensively validated and performed similarly to the PSI scorein predicting 30-day mortality of CAP patients [11] al-though previous study revealed that CURB65 may be moresuitable for identifying high-risk patients while PSI hadadvantage in the identification of low-risk patients [5] )esimplicity of calculation of CURB65 demonstrated superi-ority over other complex severity scores utilized in crowded

emergency rooms Furthermore the CRB65 score whichdoes not require a blood urea level is more suitable for use ingross-roots hospitals Previous research studies demon-strated that CURB65CRB65 does not incorporate an as-sessment of oxygenation and thus underestimated the risk ofdeath and severity of influenza pneumonia [2 12 13]

Comparatively the PSI score consisted of 20 differentparameters with different weights including demographicscomorbidities and clinical and laboratory findings [4] )ePSI score is heavily weighted by age and comorbiditieswhich shows that it underestimates severity of young CAPpatients and it is not advised to guide intensive care unitadmission [13 14] Moreover the underlying health


All cohorts MV NMV PUse of

vasopressorsNonuse ofvasopressors P

N () 336 83 (247) 253 (753) 108 (321) 228 (679)Age (yrs) 76 (61 84) 78 (67 84) 76 (61 84) 0584 78 (66 87) 75 (61 83) 0065Male n () 213 (634) 54 (607) 159 (644) 0535 71 (657) 142 (623) 0539Comorbiditiesn ()COPD 40 (119) 14 (157) 26 (105) 0194 15 (139) 25 (110) 0440CDVD 48 (143) 14 (157) 34 (138) 0650 18 (167) 30 (132) 0391CBVD 88 (262) 20 (225) 68 (275) 0352 30 (278) 58 (254) 0649Diabetes 77 (229) 15 (169) 62 (251) 0112 35 (324) 42 (184) 0004CRD 30 (89) 9 (101) 21 (85) 0648 12 (111) 18 (79) 0334HBD 24 (71) 6 (67) 18 (73) 0864 4 (37) 20 (88) 0092Healthy 28 (83) 7 (79) 21 (85) 0852 5 (46) 23 (101) 0091

Laboratory resultsWBC (times109L) 100 (67 141) 104 (67 158) 97 (68 139) 0244 108 (69 160) 98 (66 139) 0180HGB (gL) 126 (115 137) 126 (115 137) 126 (116 137) 0781 126 (113 136) 127 (117 138) 0314HCT () 369 (318 408) 354plusmn 92 372 (327 407) 0356 349plusmn 91 373 (329 409) 0092PLT (times109L) 184 (131 251) 155 (117 241) 192 (141 251) 0109 162 (123 245) 193 (141 252) 0082ALB (gL) 362 (321 391) 34plusmn 6 367 (328 393) 0002 342plusmn 57 368 (330 394) lt0001CREA (μmolL) 805 (608 1143) 925 (712 1492) 755 (599 104) lt0001 92 (69 1454) 754 (60 1014) lt0001BUN (mmolL) 74 (53 110) 95 (65 150) 68 (50 103) lt0001 90 (60 136) 68 (51 100) lt0001AST (UL) 30 (20 56) 36 (23 81) 29 (20 53) 0097 33 (23 75) 29 (20 53) 0122ALT (UL) 20 (14 37) 17 (12 45) 21 (14 36) 0694 18 (12 40) 21 (14 37) 0220TBIL (μmolL) 148 (96 230) 161 (106 25) 14 (95 223) 0164 160 (102 259) 139 (95 219) 0089DBIL (μmolL) 64 (40 106) 7 (44 123) 61 (39 97) 0037 70 (44 120) 60 (38 96) 0020K+ (mmolL) 38 (35 42) 4 (36 45) 38 (35 41) 0006 40 (35 43) 38 (35 42) 0220PaO2FiO2 303 (262 352) 285 (244 326) 313 (272 363) lt0001 287 (246 333) 315 (274 363) 0002

Vital signsSBP (mmHg) 132 (119 139) 122 (100 135) 134 (125 143) lt0001 113 (97 133) 135 (126 145) lt0001DBP (mmHg) 68 (65 76) 65 (55 71) 70 (66 78) lt0001 62 (55 69) 72 (68 78) lt0001HR (timesmin) 86 (78 94) 86 (79 104) 86 (78 94) 0033 89 (79 106) 86 (78 92) lt0001

Severity scoresCURB65 2 (2 3) 3 (3 4) 2 (1 3) lt0001 3 (25 4) 2 (1 3) lt0001CRB65 2 (1 3) 3 (2 3) 2 (1 2) lt0001 3 (2 3) 2 (1 2) lt0001PSI 128plusmn 40 161plusmn 37 120plusmn 36 lt0001 156plusmn 38 118plusmn 36 lt0001SOFA 3 (2 5) 6 (5 9) 3 (2 4) lt0001 6 (4 8) 3 (2 4) lt0001qSOFA 2 (1 2) 3 (2 3) 1 (1 2) lt0001 2 (2 3) 1 (1 2) lt0001Lactate (mmolL) 14 (11 21) 19 (13 42) 14 (11 18) lt0001 23 (14 41) 13 (10 16) lt0001

Data are presented as n n () or median (QL QU) CAP community-acquired pneumonia MV mechanical ventilation NMV nonmechanical ventilationED emergency department SCAP severe community-acquired pneumonia NSCAP nonsevere community-acquired pneumonia COPD chronic ob-structive pulmonary disease CDVD cardiovascular disease CBVD cerebrovascular disease CRD chronic renal disease HBD hepatobiliary disease WBCwhite blood cell HGB hemoglobin HCT hematocrit PLT platelet ALB albumin CREA creatinine BUN blood urea nitrogen AST aspartate ami-notransferase ALT alanine aminotransferase TBIL total bilirubin DBIL direct bilirubin SBP systolic blood pressure DBP diastolic blood pressure HRheart rate CURB65 confusion ureagt 7mmolL respiratory ratege 30min blood pressurelt 90mmHg systolic andor le60mmHg diastolic and agege 65years CRB65 confusion respiratory ratege 30min blood pressurelt 90mmHg systolic andor le60mmHg diastolic and agege 65 years PSI pneumoniaseverity index SOFA sequential organ failure assessment qSOFA quick sequential organ failure assessment


conditions may strongly influence mortality based onpneumonia severity in aged populations [15] A study byZhang et al demonstrated that PSI performed better thanCURB65 for mortality prediction while its discriminativepower decreased with advancing age [16] Regrettably thecomplexity of PSI limited its clinical application in ED

)e stratified and prognostic performance of SOFA andqSOFA in CAP has not yet been evaluated in detail and onlyfew studies have investigated its application in CAP in EDOur previous research proved that SOFA is superior toCURB65 PSI qSOFA and procalcitonin in predicting 28-day mortality with an AUROC of 0913 [17] while Kimreported AUROC of SOFA and qSOFA to be 083 and 081respectively [18] Comparatively qSOFA only requires a fewitems and vital signs Previous research findings proved thatit presented better clinical usefulness as prompt tools for EDor nonrespiratory specialists [19 20]

Serum lactate is a well-known prognostic marker forpatients with sepsis and initial ED lactate is reported to be auseful marker to risk-stratify critically ill patients presentingto ED [7] Sepsis 30 guidelines recommend lacta-tegt 20mmolL with the requirement of vasopressor use tomaintain a mean arterial blood pressure of 65mmHg as thenew definition of septic shock [6] Obtaining blood formeasuring lactate is recommended by surviving sepsiscampaign bundle update and if lactategt 2mmolL it shouldbe remeasured within 2ndash4 hours to guide resuscitation [21]Nonetheless the prognostic role of lactate on CAP patientshas not been well studied Gwak et al enrolled 397 CAPpatients and proved that the initial lactate level is inde-pendently associated with mortality in hospitalized patientswith CAP [22] Chen and Li enrolled 1641 patients andinvestigated the predictive performance of lactate CURB65and the combination of lactate and CURB65 for predicting

Table 3 ROC curve comparisons between CAP severity scores and admission lactate in predicting 28-day mortality

AUC (95 CI) P value Cutoff value Sensi () Speci () PPV () NPV ()CURB65 0744 (0685ndash0802) lt0001 30 753 619 416 874CRB65 0737 (0675ndash0799) lt0001 30 562 830 544 840PSI 0768 (0711ndash0824) lt0001 131 775 640 437 888SOFA 0795 (0740ndash0850) lt0001 40 843 644 460 919qSOFA 0807 (0754ndash0859) lt0001 20 910 538 415 943Lactate 0679 (0612ndash0745) lt0001 20 483 781 443 807CURB65 + lactate 0772 (0718ndash0827) lt0001 018 831 595 425 907CRB65 + lactate 0776 (0719ndash0833) lt0001 025 685 777 525 873PSI + lactate 0774 (0719ndash0828) lt0001 021 809 619 433 900SOFA+ lactate 0803 (0749ndash0857) lt0001 023 742 765 532 892qSOFA+ lactate 0833 (0786ndash0881) lt0001 029 640 879 656 871ROC area under the curve CAP community-acquired pneumonia Sensi sensitivity Speci specificity PPV positive predictive value NPV negativepredictive value CURB65 confusion ureagt 7mmolL respiratory ratege 30min blood pressurelt 90mmHg systolic andor le60mmHg diastolic andagege 65 years CRB65 confusion respiratory rate ge30min blood pressurelt 90mmHg systolic andor le60mmHg diastolic and agege 65 years PSIpneumonia severity index SOFA sequential organ failure assessment qSOFA quick sequential organ failure assessment

0

20

40

60

80

100

Sens

itivi

ty

80 10040200 60100 ndash specificity

CURB65CRB65PSI

SOFAqSOFALactate

(a)

0

20

40

60

80

100

Sens

itivi

ty


CURB65-lactate CRB65-lactatePSI-lactateSOFA-lactateqSOFA-lactate

(b)

0

20

40

60

80

100

Sens

itivi

ty80 10040200 60

100 ndash specificity

SOFAqSOFASOFA-lactateqSOFA-lactate

(c)

Figure 2 ROC curve comparisons of (a) CURB65 CRB65 PSI SOFA qSOFA and admission lactate in predicting 28-day mortality (b)different combinations of CAP severity scores and admission lactate in predicting 28-day mortality and (c) SOFA qSOFA SOFA+ lactateand qSOFA+ lactate in predicting 28-day mortality


Table 5 ROC curve comparisons between CAP severity scores and admission lactate in predicting mechanical ventilation

AUC (95 CI) P value Cutoff value Sensi () Speci () PPV () NPV ()CURB65 0771 (0713ndash0829) lt0001 30 783 621 404 897CRB65 0758 (0697ndash0820) lt0001 30 590 830 532 861PSI 0780 (0722ndash0837) lt0001 150 639 806 519 872SOFA 0845 (0794ndash0895) lt0001 50 759 787 539 909qSOFA 0838 (0789ndash0887) lt0001 20 940 538 400 965Lactate 0698 (0628ndash0768) lt0001 20 494 779 423 824CURB65 + lactate 0804 (0752ndash0857) lt0001 026 639 806 519 872CRB65 + lactate 0809 (0755ndash0864) lt0001 021 759 767 517 907PSI + lactate 0796 (0741ndash0850) lt0001 023 711 751 484 888SOFA+ lactate 0851 (0801ndash0902) lt0001 021 795 779 541 921qSOFA+ lactate 0868 (0826ndash0911) lt0001 026 711 889 678 904ROC area under the curve CAP community-acquired pneumonia Sensi sensitivity Speci specificity PPV positive predictive value NPV negativepredictive value CURB65 confusion ureagt 7mmolL respiratory rate ge 30min blood pressurelt 90mmHg systolic andor le60mmHg diastolic andagege 65 years CRB65 confusion respiratory ratege30min blood pressurelt 90mmHg systolic andor le60mmHg diastolic and agege 65 years PSIpneumonia severity index SOFA sequential organ failure assessment qSOFA quick sequential organ failure assessment

Table 4 ROC curve comparisons among severity scores and admission lactate in predicting ICU admission

AUC (95 CI) P value Cutoff value Sensi () Speci () PPV () NPV ()CURB65 0774 (0709ndash0840) lt0001 30 836 600 317 943CRB65 0738 (0667ndash0810) lt0001 30 590 796 391 897PSI 0837 (0777ndash0896) lt0001 155 738 836 499 935SOFA 0895 (0846ndash0943) lt0001 60 770 902 635 946qSOFA 0822 (0769ndash0874) lt0001 20 967 505 302 986Lactate 0742 (0669ndash0816) lt0001 20 557 771 350 887CURB65 + lactate 0818 (0759ndash0877) lt0001 021 672 822 456 919CRB65 + lactate 0806 (0741ndash0870) lt0001 016 770 749 405 936PSI + lactate 0840 (0782ndash0898) lt0001 025 721 862 537 933SOFA+ lactate 0902 (0857ndash0947) lt0001 018 820 876 595 956qSOFA+ lactate 0868 (0823ndash0912) lt0001 014 836 782 460 956ROC area under the curve CAP community-acquired pneumonia SCAP severe community-acquired pneumonia Sensi sensitivity Speci specificity PPVpositive predictive value NPV negative predictive value CURB65 confusion ureagt 7mmolL respiratory ratege 30min blood pressurelt 90mmHg systolicandor le60mmHg diastolic and agege 65 years CRB65 confusion respiratory ratege 30min blood pressurelt 90mmHg systolic andor le60mmHg diastolicand agege 65 years PSI pneumonia severity index SOFA sequential organ failure assessment qSOFA quick sequential organ failure assessment

CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100

Sens

itivi

ty

200 60 80 10040100 ndash specificity

(a)


0

20

40

60

80

100

Sens

itivi

ty


(b)


0

20

40

60

80

100

Sens

itivi

ty200 60 80 10040


(c)

Figure 3 ROC curve comparisons of (a) CURB65 CRB65 PSI SOFA qSOFA and admission lactate in predicting ICU admission (b)different combinations of CAP severity scores and admission lactate in predicting ICU admission and (c) SOFA qSOFA SOFA+ lactateand qSOFA+ lactate in predicting ICU admission


mortality and ICU admission in pneumonia patients in EDwhile results indicated that lactate is superior to CURB65 inpredicting mortality hospitalization and intensive care unit(ICU) admission and lactate-CURB65 combination im-proves the predictive value of single CURB65 [10] A similarstudy by Frenzen and colleagues concluded that admissionlactate levels significantly improved the prognostic value(need for mechanical ventilation vasopressors ICU ad-mission or hospital mortality) of CRBCURB65 scores inCAP patients with an optimal cutoff value of 18mmolL[23] Song et al enrolled 443 patients with CAP in ED andresults showed that the AUROC of qSOFA and SOFA forprediction of mortality was 0720 and 0845 while thecombination of qSOFA and lactate was not significantlydifferent from SOFA [24]

To the best of our knowledge not many studies haveexplored the prognostic prediction performance of qSOFA

score and admission lactate in septic CAP patients in EDbefore Our results revealed that qSOFA has the highestsensitivity and negative predictive value in predicting bothprimary (28-day mortality) and secondary outcomes (ICUadmission mechanical ventilation and vasopressor use)qSOFA outperformed other single predictors (CURB65CRB65 PSI SOFA and lactate) in predicting 28-daymortality although the AUROC of qSOFA was not sig-nificantly different from SOFA Similar to previous studiesthe combination of qSOFA+ lactate was proved comparableto full version of SOFA in predicting primary and secondaryoutcomes in our research [25 26] Moreover qSO-FA+ lactate was not statistically significant fromSOFA+ lactate (Z 1187 P 0235) Regrettably unlike theresults of previous study [24] our research indicated thatqSOFA+ lactate and SOFA+ lactate did not improve theprognostic prediction performance of single qSOFA

Table 6 ROC curve comparisons between CAP severity scores and admission lactate in predicting vasopressor use

AUC (95 CI) P value Cutoff value Sensi () Speci () PPV () NPV ()CURB65 0759 (0705ndash0814) lt0001 30 750 649 503 846CRB65 0746 (0688ndash0803) lt0001 30 537 851 631 795PSI 0767 (0712ndash0823) lt0001 1500 602 838 638 816SOFA 0821 (0772ndash0871) lt0001 40 843 684 558 902qSOFA 0820 (0772ndash0868) lt0001 20 907 575 503 929Lactate 0758 (0700ndash0817) lt0001 20 574 846 687 813CURB65 + lactate 0824 (0778ndash0871) lt0001 024 843 671 548 900CRB65 + lactate 0830 (0782ndash0877) lt0001 031 750 803 643 871PSI + lactate 0820 (0773ndash0867) lt0001 028 769 763 606 875SOFA+ lactate 0848 (0801ndash0894) lt0001 029 750 825 670 874qSOFA+ lactate 0875 (0835ndash0915) lt0001 038 694 899 765 861ROC area under the curve CAP community-acquired pneumonia Sensi sensitivity Speci specificity PPV positive predictive value NPV negativepredictive value CURB65 confusion ureagt 7mmolL respiratory ratege 30min blood pressurelt 90mmHg systolic andor le60mmHg diastolic andagege 65 years CRB65 confusion respiratory ratege 30min blood pressurelt 90mmHg systolic andor le60mmHg diastolic and agege 65 years PSIpneumonia severity index SOFA sequential organ failure assessment qSOFA quick sequential organ failure assessment

CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100Se

nsiti

vity

20 40 60 80 1000100 ndash specificity

(a)

0

20

40

60

80

100

Sens

itivi

ty



(b)

0

20

40

60

80

100

Sens

itivi

ty



(c)

Figure 4 ROC curve comparisons of(a) CURB65 CRB65 PSI SOFA qSOFA and admission lactate in predicting mechanical ventilation(b) different combinations of CAP severity scores and admission lactate in predicting mechanical ventilation and (c) SOFA qSOFASOFA+ lactate and qSOFA+ lactate in predicting mechanical ventilation


(Z 1886 P 0059) or single SOFA (Z 1066 P 0286)respectively )e relatively small sample size may explain theresult heterogeneity between our research and previousstudies Nonetheless in view of the simplicity and conve-nience of qSOFA it can be a better choice as tools forprognostic evaluation of septic CAP patients in ED Fur-thermore in predicting both primary and secondary out-comes (28-day mortality ICU admission mechanicalventilation and vasopressor use) the optimal cutoff valuefor qSOFA was 2 and optimal cutoff value for admissionlactate was 2mmolL Our study revealed that septic CAPpatients in ED with qSOFAlt 2 or lactatele 2mmolLdemonstrated significantly prolonged survival than those

patients with qSOFAge 2 or admission lactategt 2mmolLwhich could help physicians in ED enhance their awarenesswhen treating these kind of septic CAP patients A qSOFA of2 or more would be very helpful and useful in discriminatinghigh-risk patients with a high mortality rate

Our research is among the very few studies that haveexplored the prognostic performance of CURB65 CRB65PSI SOFA qSOFA and admission lactate at the same timeOur results highlighted the superiority of qSOFA and SOFAin predicting 28-day mortality ICU admission mechanicalventilation and vasopressor use However there are somelimitations in our study First the relatively small sample sizeand the retrospective and single-center design may result in

Table 7 Comparisons of severity scores and different outcomes in patients with CAP using qSOFA and lactate

qSOFAge 2 or lactategt 2P value

Yes (n 216) No (n 120)Severity scoresCURB65 3 (2 4) 2 (1 2) lt0001CRB65 2 (2 3) 1 (1 2) lt0001PSI 144plusmn 37 104plusmn 32 lt0001SOFA 4 (3 6) 2 (2 3) lt0001qSOFA 2 (2 3) 1 (1 1) lt0001Lactate 18 (12 28) 12 (10 14) lt0001

Primary outcome n ()28-day mortality 83 (384) 6 (50) lt0001

Secondary outcome n ()ICU admission 60 (278) 1 (08) lt0001Mechanical ventilation 80 (370) 3 (25) lt0001Use of vasopressors 102 (472) 6 (50) lt0001

Data are presented as n n () or median (QL QU) CAP community-acquired pneumonia CURB65 confusion ureagt 7mmolL respiratory ratege 30minblood pressurelt 90mmHg systolic andor le60mmHg diastolic and agege 65 years CRB65 confusion respiratory ratege 30min blood pressurelt 90mmHgsystolic andor le60mmHg diastolic and agege 65 years PSI pneumonia severity index SOFA sequential organ failure assessment qSOFA quick sequentialorgan failure assessment ICU intensive care unit

CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100Se

nsiti

vity


(a)


0

20

40

60

80

100

Sens

itivi

ty


(b)

SOFAqSOFA

SOFA-lactateqSOFA-lactate

0

20

40

60

80

100

Sens

itivi

ty


(c)

Figure 5 ROC curve comparisons of(a) CURB65 CRB65 PSI SOFA qSOFA and admission lactate in predicting vasopressor use (b)different combinations of CAP severity scores and admission lactate in predicting vasopressor use and (c) SOFA qSOFA SOFA+ lactateand qSOFA+ lactate in predicting vasopressor use


selection bias and do not allow for analysis of all clinical dataOur results should be verified by more multicenter pro-spective studies with larger sample size Second our researchonly enrolled septic patients with CAP in ED )ese patientsare of older age and with more complications which couldhave a highermortality rate andmay influence the final results

5 Conclusions

In conclusion we analyzed the prognostic prediction valueof CURB65 CRB65 PSI SOFA qSOFA and admissionlactate at the same time in patients with ED in our solitarycenter We found that qSOFA and SOFA were superior toCURB65 CRB65 and PSI in predicting 28-day mortalityICU admission mechanical ventilation and vasopressor usefor septic CAP patients in ED )e prediction performanceof the combination of qSOFA and admission lactate wascomparable to full version of SOFA qSOFA with admissionlactate could be a convenient valuable and practical tool forprognostic prediction Further multicenter studies withlarger sample size are needed to validate our results

Abbreviations

CAP Community-acquired pneumoniaED Emergency departmentPSI Pneumonia severity indexSOFA Sequential organ failure assessmentCURB65 Confusion ureagt 7mmolL respiratory ratege 30

min blood pressurelt90 mmHg systolic and orle60mmHg diastolic and agege 65 years

CRB65

Confusion respiratory ratege 30min bloodpressurelt90 mmHg systolic andor le60mmHgdiastolic and agege 65 years

HGB HemoglobinHCT HemocritPLT PlateletALB AlbuminAST Aspartate aminotransferaseALT Alanine aminotransferaseTBIL Total bilirubinDBIL Direct bilirubinCREA CreatinineBUN Blood urea nitrogenWBC White blood cellROC Receiver operating characteristicsAUC Area under the curveAUROC Area under the receiver operating characteristics

curvePPV Positive predictive valueNPV Negative predictive valueCOPD Chronic obstructive pulmonary diseaseCDVD Cardiovascular diseaseCBVD Cerebral-vascular diseaseCRD Chronic renal diseaseHBD Hepatobiliary diseaseICU Intensive care unit

Data Availability

)e data used to support the findings of this study areavailable from the corresponding author upon request

Log rank X2 = 53800 P lt 0001

CensoredCensored

qSOFA lt 2qSOFA ge 2

00

02

04

06

08

10

Cum

surv

ival

5 10 15 20 25 30 0Survival time (days)

(a)

Log rank X2 = 31969 P lt 0001

Lactate le 2mmolLLactate gt 2mmolLCensoredCensored

00

02

04

06

08

10

Cum

surv

ival


(b)

Log rank X2 = 59825 P lt 0001

qSOFA lt 2 or lactate le 2mmolLqSOFA ge 2 or lactate gt 2mmolLCensoredCensored

00

02

04

06

08

10

Cum

surv

ival


(c)

Figure 6 KaplanndashMeier survival curve comparison (a) between septic CAP patients with admission qSOFAlt 2 and septic CAP patientswith admission qSOFAge 2 (b) between septic CAP patients with admission lactatele 2mmolL and septic CAP patients with admissionlactategt 2mmolL (c) between septic CAP patients with admission qSOFAlt 2 or lactatele 2mmolL and septic CAP patients with ad-mission qSOFAge 2 or lactategt 2mmolL


Conflicts of Interest

)e authors declare that they have no conflicts of interest

References

[1] M Heron ldquoDeath leading causes for 2014rdquo National VitalStatistics Reports vol 65 no 5 pp 1ndash96 2016

[2] B Cao Y Huang D-Y She et al ldquoDiagnosis and treatment ofcommunity-acquired pneumonia in adults 2016 clinicalpractice guidelines by the Chinese )oracic Society ChineseMedical Associationrdquo-eClinical Respiratory Journal vol 12no 4 pp 1320ndash1360 2018

[3] W S Lim M M van der Eerden R Laing et al ldquoDefiningcommunity acquired pneumonia severity on presentation tohospital an international derivation and validation studyrdquo-orax vol 58 no 5 pp 377ndash382 2003

[4] M J Fine T E Auble D M Yealy et al ldquoA prediction rule toidentify low-risk patients with community-acquired pneu-moniardquo New England Journal of Medicine vol 336 no 4pp 243ndash250 1997

[5] J D Chalmers A Singanayagam A R Akram et al ldquoSeverityassessment tools for predicting mortality in hospitalised pa-tients with community-acquired pneumonia Systematic re-view and meta-analysisrdquo-orax vol 65 no 10 pp 878ndash8832010

[6] M Singer C S Deutschman C W Seymour et al ldquo)e thirdinternational consensus definitions for sepsis and septic shock(Sepsis-3)rdquo JAMA vol 315 no 8 pp 801ndash810 2016

[7] R Bou Chebl C El Khuri A Shami et al ldquoSerum lactate is anindependent predictor of hospital mortaltiy in critically illpatients in the emergency department a retrospective studyrdquoScandinavian Journal of Trauma Resuscitation and Emer-gency Medicine vol 25 no 1 p 69 2017

[8] Y J Park D H Kim S C Kim et al ldquoSerum lactate uponemergency department arrival as a predictor of 30-day in-hospital mortality in an unselected populationrdquo PLoS Onevol 13 no 1 Article ID e0190519 2018

[9] L A Mandell R G Wunderink A Anzueto et al ldquoInfectiousDiseases Society of AmericaAmerican )oracic Societyconsensus guidelines on the management of community-acquired pneumonia in adultsrdquo Clinical Infectious Diseasesvol 44 no Supplement_2 pp S27ndashS72 2007

[10] Y-X Chen and C-S Li ldquoLactate on emergency departmentarrival as a predictor of mortality and site-of-care in pneu-monia patients a cohort studyrdquo -orax vol 70 no 5pp 404ndash410 2015

[11] W S Lim S V Baudouin R C George et al ldquoPneumoniaguidelines committee of the BTS standards of care committeeBTS guidelines for the management of community acquiredpneumonia in adults updates 2009rdquo -orax vol 64no Suppl 3 pp iii1ndashiii55 2009

[12] S Mulrennan S S Tempone I T Ling et al ldquoPandemicinfluenza (H1N1) 2009 pneumonia CURB-65 score forpredicting severity and nasopharyngeal sampling for diag-nosis are unreliablerdquo PLoS One vol 5 no 9 Article IDe12849 2010

[13] J D Chalmers A Singanayagam and A T Hill ldquoPredictingthe need for mechanical ventilation andor inotropic supportfor young adults admitted to the hospital with community-acquired pneumoniardquo Clinical Infectious Diseases vol 47no 12 pp 1571ndash1574 2008

[14] J D Chalmers P Mandal A Singanayagam et al ldquoSeverityassessment tools to guide ICU admission in community-

acquired pneumonia systematic review and meta-analysisrdquoIntensive Care Medicine vol 37 no 9 pp 1409ndash1420 2011

[15] S Hamaguchi M Suzuki K Sasaki et al ldquoAdult pneumoniastudy groupndashJapan Six underlying health conditions stronglyinfluence mortality based on pneumonia severity in an ageingpopulation of Japan a prospective cohort studyrdquo BMC Pul-monary Medicine vol 18 no 1 p 88 2018

[16] Z Zhang Y Yong W Tan L Shen H Ng and K FongldquoPrognostic factors for mortality due to pneumonia amongadults from different age groups in Singapore and mortalitypredictions based on PSI and CURB-65rdquo Singapore MedicalJournal vol 59 no 4 pp 190ndash198 2018

[17] H Zhou S Guo T Lan S Ma F Zhang and Z Zhao ldquoRiskstratification and prediction value of procalcitonin andclinical severity scores for community-acquired pneumonia inEDrdquo -e American Journal of Emergency Medicine vol 36no 12 pp 2155ndash2160 2018

[18] M W Kim J Y Lim and S H Oh ldquoMortality predictionusing serum biomarkers and various clinical risk scales incommunity-acquired pneumoniardquo Scandinavian Journal ofClinical and Laboratory Investigation vol 77 no 7pp 486ndash492 2017

[19] O T Ranzani E Prina R Menendez et al ldquoNew sepsisdefinition (Sepsis-3) and community-acquired pneumoniamortality A validation and clinical decision-making studyrdquoAmerican Journal of Respiratory and Critical Care Medicinevol 196 no 10 pp 1287ndash1297 2017

[20] F Tokioka H Okamoto A Yamazaki A Itou and T Ishidaldquo)e prognostic performance of qSOFA for community-ac-quired pneumoniardquo Journal of Intensive Care vol 6 no 1p 46 2018

[21] M M Levy L E Evans and A Rhodes ldquo)e surviving sepsiscampaign bundlerdquo Critical Care Medicine vol 46 no 6pp 997ndash1000 2018

[22] M H Gwak S Jo T Jeong et al ldquoInitial serum lactate level isassociated with inpatient mortality in patients with com-munity-acquired pneumoniardquo -e American Journal ofEmergency Medicine vol 33 no 5 pp 685ndash690 2015

[23] F S Frenzen U Kutschan N Meiswinkel B Schulte-Hubbert S Ewig and M Kolditz ldquoAdmission lactate predictspoor prognosis independently of the CRBCURB-65 scores incommunity-acquired pneumoniardquo Clinical Microbiology andInfection vol 24 no 3 pp 306e1ndash306e6 2018

[24] H Song H G Moon and S H Kim ldquoEfficacy of quicksequential organ failure assessment with lactate concentrationfor predicting mortality in patients with community-acquiredpneumonia in the emergency departmentrdquo Clinical andExperimental Emergency Medicine vol 6 no 1 pp 1ndash8 2019

[25] K M Ho and N S H Lan ldquoCombining quick sequentialorgan failure assessment with plasma lactate concentration iscomparable to standard sequential organ failure assessmentscore in predicting mortality of patients with and withoutsuspected infectionrdquo Journal of Critical Care vol 38 pp 1ndash52017

[26] H Zhou T Lan and S Guo ldquoStratified and prognostic valueof admission lactate and severity scores in patients withcommunity-acquired pneumonia in emergency department asingle-center retrospective cohort studyrdquo Medicine (Balti-more) vol 98 no 41 Article ID e17479 October 2019


pneumonia severity index (PSI) score are most widely usedworldwide [3 4] Preliminary research studies demonstratedthat there were no significant differences in overall testperformance between CURB65 and PSI [5]

CAP accounts for substantial mortality worldwide witha high risk of developing respiratory failure and septic shock)e third international consensus definitions for sepsis andseptic shock generated new definitions and updated theclinical criteria [6] Sepsis was defined as life-threateningorgan dysfunction caused by a dysregulated host response toinfection and organ dysfunction was defined as an increasein the sequential organ failure assessment (SOFA) score of 2or higher [6] highlighting the clinical implication of (SOFA)score and qSOFA (respiratory ratege 22min altered men-tation and systolic blood pressurege 100mmHg) score [6]Lactate has been utilized as a marker to evaluate the acid-base homeostasis and perfusion status of patients Higherlactate levels are reported to be associated with higherhospital mortalities and longer length of ED and hospitalstay in unselected patients [7 8] However the prognosticvalue of SOFA score qSOFA score and lactate in patientswith CAP in the emergency department (ED) has not beenfully elucidated

In the present study we explored the prognostic pre-diction value of SOFA qSOFA and admission lactate inpatients with CAP in ED and the results were comparedwith that of other commonly used CAP severity scores(CURB65 CRB65 PSI)

2 Patients and Methods

)is was a single-center retrospective cohort study carriedout in Beijing Chao-yang Hospital Capital Medical Uni-versity which is a tertiary referral hospital located in thenorthern region of China with approximately 250000 an-nual ED visits per year )e Institutional Review Board andMedical Ethics Committee have approved this study andwritten informed consent was waived because of the ret-rospective design of this study

Adult patients with the discharge diagnosis of CAPadmitted between Jan 2017 and Jan 2019 were screened)einclusion criteria were agege 18 years new infiltrates on chestradiograph and two or more symptoms including coughfever dyspnea sputum production breathlessness andorchest pain [9] In accordance with sepsis 30 criteria we onlyenrolled CAP patients with increased SOFA scorege 2 frombaseline )e following patients were excluded from thisstudy (1) agelt 18 years (2) patients with metastatic tumoracquired immunodeficiency syndrome (AIDS) active tu-berculosis previous transplantation immunosuppressivetherapy and pregnancy (3) patients transferred from otherhospitals or diagnosed with hospital acquired pneumonia(4) patients with incomplete clinical laboratory or radio-graphic records (5) patients with increased admission SOFAscorelt 2 from baseline

Demographic characteristics of all enrolled patients onED arrival were collected and recorded by trained triagenurses on admission Past history comorbidities and vitalsigns (blood pressure heart rate breath rate and state of

consciousness) were also recorded Laboratory parameterson admission including full blood count hemoglobin level(HGB) hemocrit (HCT) platelet level (PLT) albumin(ALB) hepatic function (aspartate aminotransferase (AST)alanine aminotransferase (ALT) total bilirubin (TBIL)direct bilirubin (DBIL)) renal function (creatinine (CREA)blood urea nitrogen (BUN)) electrolytes and arterial bloodgas including lactate level were assessed and collectedCURB65 CRB65 PSI SOFA and qSOFA scores on ad-mission for each patient were calculated according to in-ternational criteria and analyzed utilizing data collected onED arrival

All patients were followed up for 28 days throughmedical records and 28-daymortality was the primary studyend point According to their prognosis after 28-day ad-mission patients were divided into death group and survivalgroup )e secondary outcome included ICU admissionmechanical ventilation use and vasopressor use

All analyses were performed using SPSS 220 statisticalsoftware package (SPSS Inc Chicago IL USA) Data withnormal distribution were described as meanplusmn standarddeviation and compared using Studentrsquos t-test Data withskewed distribution were expressed as median (interquartilerange) and compared using the MannndashWhitney U non-parametric test )e categorical variables were described aspercentages and compared using the chi-squared test orFisherrsquos exact test Receiver operating characteristics (ROC)curves for each predictor were constructed and the areaunder the curve (AUC) was determined to assess theirpredictive values Combination models of severity score andlactate were established using several logistic regressions tosave the predicted probabilities ROC curve analysis wasperformed using the saved probabilities as a new indicatorComparisons of each predictor were conducted usingMedCalc 150 Software (Acacialaan Ostend Belgium) AZ-testwas used for comparing the AUCs between different curvesFor comparison of the AUCs Z (A1minusA2)

(SE1)2 + (SE2)

21113969

was used the test values being Z005196 and Z001 258Based on the cutoff values sensitivity specificity positivepredictive value (PPV) and negative predictive value (NPV)were also calculated KaplanndashMeier survival curves were drawnusing cutoff values of qSOFA and lactate A two-tailed value ofPlt 005 was considered statistically significant

3 Results

A total of 561 patients were screened at recruitment and 225patients were excluded (Figure 1) A total of 336 patientswere finally included in our study group Of the 225 ex-cluded patients 16 patientsrsquo agelt 18 years old 74 patientswere transferred from other hospitals 10 patients were witha history of previous transplantation 5 patients were finallydiagnosed with pulmonary tuberculosis 3 patients were withpulmonary thromboembolism 6 patients were with lungcancer 4 patients were with HIV 57 patients were withincomplete medical records 30 patients were with admis-sion SOFA scorelt 2 from baseline and 20 patients were lostto follow-up with unknown prognosis





















4 Discussion




























0

20

40

60

80

100

Sens

itivi

ty


CURB65CRB65PSI

SOFAqSOFALactate

(a)

0

20

40

60

80

100

Sens

itivi

ty



(b)

0

20

40

60

80

100

Sens

itivi

ty80 10040200 60



(c)







CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100

Sens

itivi

ty


(a)


0

20

40

60

80

100

Sens

itivi

ty


(b)


0

20

40

60

80

100

Sens

itivi

ty200 60 80 10040


(c)








CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100Se

nsiti

vity


(a)

0

20

40

60

80

100

Sens

itivi

ty



(b)

0

20

40

60

80

100

Sens

itivi

ty



(c)












CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100Se

nsiti

vity


(a)


0

20

40

60

80

100

Sens

itivi

ty


(b)

SOFAqSOFA


0

20

40

60

80

100

Sens

itivi

ty


(c)




5 Conclusions


Abbreviations



CRB65




Data Availability


Log rank X2 = 53800 P lt 0001

CensoredCensored


00

02

04

06

08

10

Cum

surv

ival


(a)

Log rank X2 = 31969 P lt 0001


00

02

04

06

08

10

Cum

surv

ival


(b)

Log rank X2 = 59825 P lt 0001


00

02

04

06

08

10

Cum

surv

ival


(c)





References
















































4 Discussion




























0

20

40

60

80

100

Sens

itivi

ty


CURB65CRB65PSI

SOFAqSOFALactate

(a)

0

20

40

60

80

100

Sens

itivi

ty



(b)

0

20

40

60

80

100

Sens

itivi

ty80 10040200 60



(c)







CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100

Sens

itivi

ty


(a)


0

20

40

60

80

100

Sens

itivi

ty


(b)


0

20

40

60

80

100

Sens

itivi

ty200 60 80 10040


(c)








CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100Se

nsiti

vity


(a)

0

20

40

60

80

100

Sens

itivi

ty



(b)

0

20

40

60

80

100

Sens

itivi

ty



(c)












CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100Se

nsiti

vity


(a)


0

20

40

60

80

100

Sens

itivi

ty


(b)

SOFAqSOFA


0

20

40

60

80

100

Sens

itivi

ty


(c)




5 Conclusions


Abbreviations



CRB65




Data Availability


Log rank X2 = 53800 P lt 0001

CensoredCensored


00

02

04

06

08

10

Cum

surv

ival


(a)

Log rank X2 = 31969 P lt 0001


00

02

04

06

08

10

Cum

surv

ival


(b)

Log rank X2 = 59825 P lt 0001


00

02

04

06

08

10

Cum

surv

ival


(c)





References














































0

20

40

60

80

100

Sens

itivi

ty


CURB65CRB65PSI

SOFAqSOFALactate

(a)

0

20

40

60

80

100

Sens

itivi

ty



(b)

0

20

40

60

80

100

Sens

itivi

ty80 10040200 60



(c)







CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100

Sens

itivi

ty


(a)


0

20

40

60

80

100

Sens

itivi

ty


(b)


0

20

40

60

80

100

Sens

itivi

ty200 60 80 10040


(c)








CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100Se

nsiti

vity


(a)

0

20

40

60

80

100

Sens

itivi

ty



(b)

0

20

40

60

80

100

Sens

itivi

ty



(c)












CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100Se

nsiti

vity


(a)


0

20

40

60

80

100

Sens

itivi

ty


(b)

SOFAqSOFA


0

20

40

60

80

100

Sens

itivi

ty


(c)




5 Conclusions


Abbreviations



CRB65




Data Availability


Log rank X2 = 53800 P lt 0001

CensoredCensored


00

02

04

06

08

10

Cum

surv

ival


(a)

Log rank X2 = 31969 P lt 0001


00

02

04

06

08

10

Cum

surv

ival


(b)

Log rank X2 = 59825 P lt 0001


00

02

04

06

08

10

Cum

surv

ival


(c)





References

































CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100

Sens

itivi

ty


(a)


0

20

40

60

80

100

Sens

itivi

ty


(b)


0

20

40

60

80

100

Sens

itivi

ty200 60 80 10040


(c)








CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100Se

nsiti

vity


(a)

0

20

40

60

80

100

Sens

itivi

ty



(b)

0

20

40

60

80

100

Sens

itivi

ty



(c)












CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100Se

nsiti

vity


(a)


0

20

40

60

80

100

Sens

itivi

ty


(b)

SOFAqSOFA


0

20

40

60

80

100

Sens

itivi

ty


(c)




5 Conclusions


Abbreviations



CRB65




Data Availability


Log rank X2 = 53800 P lt 0001

CensoredCensored


00

02

04

06

08

10

Cum

surv

ival


(a)

Log rank X2 = 31969 P lt 0001


00

02

04

06

08

10

Cum

surv

ival


(b)

Log rank X2 = 59825 P lt 0001


00

02

04

06

08

10

Cum

surv

ival


(c)





References


































CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100Se

nsiti

vity


(a)

0

20

40

60

80

100

Sens

itivi

ty



(b)

0

20

40

60

80

100

Sens

itivi

ty



(c)












CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100Se

nsiti

vity


(a)


0

20

40

60

80

100

Sens

itivi

ty


(b)

SOFAqSOFA


0

20

40

60

80

100

Sens

itivi

ty


(c)




5 Conclusions


Abbreviations



CRB65




Data Availability


Log rank X2 = 53800 P lt 0001

CensoredCensored


00

02

04

06

08

10

Cum

surv

ival


(a)

Log rank X2 = 31969 P lt 0001


00

02

04

06

08

10

Cum

surv

ival


(b)

Log rank X2 = 59825 P lt 0001


00

02

04

06

08

10

Cum

surv

ival


(c)





References






































CURB65CRB65PSI

SOFAqSOFALactate

0

20

40

60

80

100Se

nsiti

vity


(a)


0

20

40

60

80

100

Sens

itivi

ty


(b)

SOFAqSOFA


0

20

40

60

80

100

Sens

itivi

ty


(c)




5 Conclusions


Abbreviations



CRB65




Data Availability


Log rank X2 = 53800 P lt 0001

CensoredCensored


00

02

04

06

08

10

Cum

surv

ival


(a)

Log rank X2 = 31969 P lt 0001


00

02

04

06

08

10

Cum

surv

ival


(b)

Log rank X2 = 59825 P lt 0001


00

02

04

06

08

10

Cum

surv

ival


(c)





References






























5 Conclusions


Abbreviations



CRB65




Data Availability


Log rank X2 = 53800 P lt 0001

CensoredCensored


00

02

04

06

08

10

Cum

surv

ival


(a)

Log rank X2 = 31969 P lt 0001


00

02

04

06

08

10

Cum

surv

ival


(b)

Log rank X2 = 59825 P lt 0001


00

02

04

06

08

10

Cum

surv

ival


(c)





References































References





























PrognosticPredictionValueofqSOFA,SOFA,andAdmission...

Documents

Transcript of PrognosticPredictionValueofqSOFA,SOFA,andAdmission...