scholarly journals Establishing Classifiers With Clinical Laboratory Indicators to Distinguish COVID-19 From Community-Acquired Pneumonia: Retrospective Cohort Study

10.2196/23390 ◽  
2021 ◽  
Vol 23 (2) ◽  
pp. e23390
Author(s):  
Wanfa Dai ◽  
Pei-Feng Ke ◽  
Zhen-Zhen Li ◽  
Qi-Zhen Zhuang ◽  
Wei Huang ◽  
...  

Background The initial symptoms of patients with COVID-19 are very much like those of patients with community-acquired pneumonia (CAP); it is difficult to distinguish COVID-19 from CAP with clinical symptoms and imaging examination. Objective The objective of our study was to construct an effective model for the early identification of COVID-19 that would also distinguish it from CAP. Methods The clinical laboratory indicators (CLIs) of 61 COVID-19 patients and 60 CAP patients were analyzed retrospectively. Random combinations of various CLIs (ie, CLI combinations) were utilized to establish COVID-19 versus CAP classifiers with machine learning algorithms, including random forest classifier (RFC), logistic regression classifier, and gradient boosting classifier (GBC). The performance of the classifiers was assessed by calculating the area under the receiver operating characteristic curve (AUROC) and recall rate in COVID-19 prediction using the test data set. Results The classifiers that were constructed with three algorithms from 43 CLI combinations showed high performance (recall rate >0.9 and AUROC >0.85) in COVID-19 prediction for the test data set. Among the high-performance classifiers, several CLIs showed a high usage rate; these included procalcitonin (PCT), mean corpuscular hemoglobin concentration (MCHC), uric acid, albumin, albumin to globulin ratio (AGR), neutrophil count, red blood cell (RBC) count, monocyte count, basophil count, and white blood cell (WBC) count. They also had high feature importance except for basophil count. The feature combination (FC) of PCT, AGR, uric acid, WBC count, neutrophil count, basophil count, RBC count, and MCHC was the representative one among the nine FCs used to construct the classifiers with an AUROC equal to 1.0 when using the RFC or GBC algorithms. Replacing any CLI in these FCs would lead to a significant reduction in the performance of the classifiers that were built with them. Conclusions The classifiers constructed with only a few specific CLIs could efficiently distinguish COVID-19 from CAP, which could help clinicians perform early isolation and centralized management of COVID-19 patients.

2020 ◽  
Author(s):  
Wanfa Dai ◽  
Pei-Feng Ke ◽  
Zhen-Zhen Li ◽  
Qi-Zhen Zhuang ◽  
Wei Huang ◽  
...  

BACKGROUND The initial symptoms of patients with COVID-19 are very much like those of patients with community-acquired pneumonia (CAP); it is difficult to distinguish COVID-19 from CAP with clinical symptoms and imaging examination. OBJECTIVE The objective of our study was to construct an effective model for the early identification of COVID-19 that would also distinguish it from CAP. METHODS The clinical laboratory indicators (CLIs) of 61 COVID-19 patients and 60 CAP patients were analyzed retrospectively. Random combinations of various CLIs (ie, CLI combinations) were utilized to establish COVID-19 versus CAP classifiers with machine learning algorithms, including random forest classifier (RFC), logistic regression classifier, and gradient boosting classifier (GBC). The performance of the classifiers was assessed by calculating the area under the receiver operating characteristic curve (AUROC) and recall rate in COVID-19 prediction using the test data set. RESULTS The classifiers that were constructed with three algorithms from 43 CLI combinations showed high performance (recall rate >0.9 and AUROC >0.85) in COVID-19 prediction for the test data set. Among the high-performance classifiers, several CLIs showed a high usage rate; these included procalcitonin (PCT), mean corpuscular hemoglobin concentration (MCHC), uric acid, albumin, albumin to globulin ratio (AGR), neutrophil count, red blood cell (RBC) count, monocyte count, basophil count, and white blood cell (WBC) count. They also had high feature importance except for basophil count. The feature combination (FC) of PCT, AGR, uric acid, WBC count, neutrophil count, basophil count, RBC count, and MCHC was the representative one among the nine FCs used to construct the classifiers with an AUROC equal to 1.0 when using the RFC or GBC algorithms. Replacing any CLI in these FCs would lead to a significant reduction in the performance of the classifiers that were built with them. CONCLUSIONS The classifiers constructed with only a few specific CLIs could efficiently distinguish COVID-19 from CAP, which could help clinicians perform early isolation and centralized management of COVID-19 patients.


2019 ◽  
Vol 5 (1) ◽  
pp. 6
Author(s):  
Arvyl Jan L. Andaya ◽  
Prince Duncan C. Maylas ◽  
Ma. Estrella H. Sales

This study was designed to determine whether Averroha bilimbi extract can be used as an alternative anticoagulant for manual complete blood count (CBC) in the hematology clinical laboratory instead of Ethylenediaminetetraacetic acid (EDTA), the recommended anticoagulant for CBC. Blood from 15 volunteers was extracted and placed in EDTA-anticoagulated tubes and tubes with Averrhoa bilimbi extract. Samples from both tubes were tested for CBC. Using independent t-test the study revealed that there is no difference in the red blood cell (RBC) count, white blood cell (WBC) count, hemoglobin, hematocrit, and a 3-part differential of EDTA anticoagulated blood and blood with Averrhoa bilimbi extract as anticoagulant. The morphology of lymphocytes and monocytes were not affected, however, the granulocytes showed cytoplasmic distortion and vacuolation in the Averrhoa bilimbi extract.


Author(s):  
E. O. Aribo ◽  
O. E. Ofem ◽  
M. E. Moses

Background: Ocimum gratissimum is an ethnopharmacological plant with numerous healing potentials. But there is paucity in scientific literature on its impact on the blood cells. Aim and Objective: The present study was intended to evaluate the dose and time-dependent effect of aqueous leaf extract of Ocimum gratissimum on some haematological in albino Wistar rats. Methods: Twenty four rats were randomly assigned to four equal groups. Group 1 was control, groups 2, 3 and 4 were given 450 mg/kg, 800 mg/kg and 1800 mg/kg respectively of the extract daily for 30 days. All rats had free access to water and rat chow. Results: The result showed a significantly lower RBC count in groups 3 and 4 when compared with control (p<0.05 and p<0.01 respectively). PCV was significantly lower in group 4 than control (p<0.05). Platelet count was significantly lower in group 4 than control (p<0.05) and group 2 (p<0.05). A significant increase in WBC count was observed in groups 2 and 3 when compared with control (p<0.05 and p<0.01 respectively). A significant reduction in neutrophil count was observed in groups 2, 3 and 4 compared with control (p<0.05; p< 0.05 and p<0.01 respectively). There were significant increases in lymphocyte counts in groups 2, 3 and 4 when compared with control (p<0.05, p<0.01 and p<0.01 respectively). There were no significant changes in MCV, MCH and MCHC in the groups. Conclusion: In conclusion, aqueous leaf extract of Ocimum gratissimum causes reduction in RBC count, PCV, platelet count and neutrophil count but causes increases in total WBC count and lymphocyte count.


2021 ◽  
Vol 99 (Supplement_1) ◽  
pp. 168-169
Author(s):  
Cynthia Jinno ◽  
Braden T Wong ◽  
Martina Kluenemann ◽  
Xunde Li ◽  
Yanhong Liu

Abstract The objective of this experiment was to investigate the effects of dietary supplementation of Bacillus amyloquefaciens on total and differential blood cell count in weaned pigs experimentally infected with a pathogenic E. coli. A total of 50 weaned pigs (7.41 ± 1.35 kg) were individually housed in disease containment rooms and randomly assigned to one of the 5 treatments: sham control (CON-), sham B. amyloquefaciens (BAM-), challenged control (CON+), challenged B. amyloquefaciens (BAM+) and challenged carbadox (CAR+). The experiment lasted 28 days with 7 days’ adaptation and 21 days after the first E. coli inoculation. The doses of F18 E. coli inoculum were 1010 CFU/3 mL oral dose daily for 3 consecutive days. Whole blood samples were collected from all pigs on d -7, and d 0, 7, 14, and 21 post infection (PI) to measure total and differential blood cell count by complete blood count (CBC) analysis. Supplementation of BAM or CAR increased (P &lt; 0.05) either the percentage or the number of lymphocytes on d 0 before E. coli inoculation. E. coli challenge increased (P &lt; 0.05) white blood cell (WBC) count on d 7 and 21 PI, while supplementation of BAM tended (P &lt; 0.10) to have low WBC on d 7 PI and had lower (P &lt; 0.05) WBC on d 21 PI compared with CON+. Pigs in BAM+ also had lower (P &lt; 0.05) neutrophil count on d 14 PI, pigs fed with CAR had lower (P &lt; 0.05) neutrophil count on d 14 and 21 PI, compared with pigs in CON+. No difference was observed in red blood cell profile among all treatments throughout the experiment. In conclusion, pigs fed with B. amyloquefaciens have similar systemic immune response to pigs in antibiotic group and have relatively lower systemic inflammation caused by E. coli compared with control group.


2018 ◽  
Vol 12 (10) ◽  
pp. 922-925 ◽  
Author(s):  
Tuğba Kula Atik ◽  
Emre Ispir ◽  
Aykut Aytekin ◽  
Ufuk Turhan ◽  
Esma Yeniiz ◽  
...  

Moxifloxacin is a fourth generation widely used fluoroquinolone antibiotic. There are three cases of moxifloxacin-induced neutropenia reported in the literature and we report the fourth case. A 26-year-old man with pneumonia was treated with moxifloxacin because of penicillin allergy. On the second day of therapy, leukopenia [White blood cell (WBC) count 2.7×10³/μL] and neutropenia (neutrophils 1.21×10³/μL) occurred. Rothia mucilaginosa was isolated in sputum culture. On the fourth day of hospitalization moxifloxacin treatment was stopped and clarithromycin 500 mg PO twice daily was started. Leukopenia and neutropenia resolved one day after discontinuation of moxifloxacin that WBC and neutrophil count rose 4.5×10³/μL and 1.97×10³/μL, respectively. On the sixth day of hospitalization, WBC and neutrophil count was 4.3×10³/μL and 2.29×10³/μL, respectively. The immunomodulatory effects of moxifloxacin may result in the changes of WBC count like leukopenia with neutropenia. Moxifloxacin induced neutropenia may be more common and is an important adverse effect. More observational studies about safety profiles of moxifloxacin are needed.


Author(s):  
Yunbao Pan ◽  
Guangming Ye ◽  
Xiantao Zeng ◽  
Guohong Liu ◽  
Xiaojiao Zeng ◽  
...  

AbstractBackgroundThe clinical presentation of 2019 Novel Coronavirus (2019-nCov) infected pneumonia (NCIP) resembles that of other etiologies of community-acquired pneumonia (CAP). We aimed to identify clinical laboratory features to distinguish NCIP from CAP.MethodsWe compared the ability of the hematological and biochemical features of 84 patients with NCIP at hospital admission and 316 patients with CAP. Parameters independently predictive of NCIP were calculated by multivariate logistic regression. The receiver operating characteristic (ROC) curves were generated and the area under the ROC curve (AUC) was measured to evaluate the discriminative ability.ResultsMost hematological and biochemical indexes of patients with NCIP were significantly different from patients with CAP. Nine laboratory parameters were identified to be highly predictive of a diagnosis of NCIP by multivariate analysis. The AUCs demonstrated good discriminatory ability for red cell distribution width (RDW) with an AUC of 0.88 and Hemoglobin (HGB) with an AUC of 0.82. Red blood cell (RBC), albumin (ALB), eosinophil (EO), hematocrit (HCT), alkaline phosphatase (ALP), and white blood cell (WBC) had fair discriminatory ability. Combinations of any two parameters performed better than did the RDW alone.ConclusionsRoutine laboratory examinations may be helpful for the diagnosis of NCIP. Application of laboratory tests may help to optimize the use of isolation rooms for patients when they present with unexplained febrile respiratory illnesses.


2019 ◽  
Vol 12 (3) ◽  
pp. 449-453 ◽  
Author(s):  
Okereke Henry Nnamdi ◽  
Udegbunam Rita Ijeoma ◽  
Nnaji Theophilus Okaforx

Aim: The effect of citrate phosphate dextrose adenine-1 on the hematological parameters of stored Nigerian indigenous dog's blood with plastic vacutainer was investigated. This was done in view of determining the viability and stability of the studied parameters for blood banking purpose. Till date, there is no literature on the stability of whole blood of Nigerian indigenous dogs for blood banking purposes. Materials and Methods: A total of six apparently adult healthy dogs were sampled, and their blood was stored at 4°C and analyzed for their packed cell volume (PCV), hemoglobin (Hgb) concentration, red blood cell (RBC) count, total and differential white blood cell (WBC) count, platelet count (PC), mean corpuscular values (mean corpuscular Hgb [MCH], MCH concentration, and mean corpuscular volume [MCV]), blood potency of hydrogen (pH), and erythrocyte sedimentation rate (ESR) over a period of 14 days. Results: Significant changes were observed in some of the studied parameters. Of the 14 days' study period, PCV, Hgb concentration, total WBC count, PC, and neutrophil count showed no significant (p≥0.05) difference until day 2 post-storage (PS), while RBC count, ESR, MCV, and lymphocyte count showed no significant (p≥0.05) variation up until day 3 PS. Blood pH showed no significant (p≥0.05) variation within 24-h PS but was significantly (p≤0.05) higher than obtained values on days 1, 3, 7, 10, and 14 PS. Conclusion: Based on our finding, we could suggest that hematological laboratories and hemotherapists could use canine blood stored in a plastic vacutainer for 2-3 days.


2018 ◽  
Vol 30 (5) ◽  
pp. 678-687 ◽  
Author(s):  
Nicolas Herman ◽  
Catherine Trumel ◽  
Anne Geffré ◽  
Jean-Pierre Braun ◽  
Marion Thibault ◽  
...  

In order to develop bovine hematology reference intervals (RIs) in accordance with new international recommendations, we analyzed 156 blood specimens of healthy adult dairy and beef cows from 32 farms with a Sysmex XT-2000iV analyzer, and by manual scoring of platelet clumps and white blood cell (WBC) differential. We established RIs by the nonparametric method, and examined effects of age, production type (beef vs. dairy), and stage of lactation. RIs could not be determined for platelet count and indices because clumps were observed in 80% of specimens. Optical and impedance red blood cell (RBC) counts were similar, although statistically different. RIs for analyzer and manual WBC differentials were not different except for lymphocyte concentration, the subpopulations of which were counted manually. Hematocrit was higher in beef than dairy cattle, and hemoglobin was lower in early lactation. Increases in RBC count, mean corpuscular volume, mean corpuscular hemoglobin, and RBC distribution width were noted with increasing age, along with decreases in WBC count, neutrophils, and lymphocytes. Most RIs in our study, with the exception of neutrophils, were similar to those previously reported using a flow cytometry analyzer.


2020 ◽  
Vol 7 (8) ◽  
pp. 3916-3920
Author(s):  
Manigatta Doddagowda Shilpa ◽  
Raju Kalyani ◽  
P N Sreeramulu

Introduction: The most common malignancy which endangers the quality of life in females is breast cancer. It is also the major public problem in society and despite advances in treatment strategies, it still leads to high mortality. In all breast cancer patients, the first assessment done before surgery is the complete blood count (CBC). There exists a strong relationship between immune response of the body and clinical staging of breast cancer. This study aimed to determine the most reliable hematological markers for the prognosis of breast cancer, and to evaluate the correlation between hematological parameters and disease staging. Methods: This study was a case control study conducted for a period of 3 years from Jan 2016 to December 2018 in the Department of Pathology (R.L. hospital and Research Center attached to Sri Devaraj Urs medical college, Tamaka, Kolar, Karnataka). The comparison of the hematological parameters was done between the study group (70 cases) and the control group (20 cases). The relation between the hematological parameters and the various stages of breast cancer was analyzed before initiating the treatment. Results: The majority of the breast cancer patients were of stage 2. The hemoglobin concentration, red blood cell (RBC) count, lymphocyte count, and were reduced in the study group, with a significant difference (p < 0.05) when compared to the controls. Hemoglobin concentration, RBC count, hematocrit, white blood cell (WBC) count, and absolute lymphocyte count were inversely proportional to the stage of breast cancer. The absolute count, ratio (NLR), and platelet-to-lymphocyte ratio (PLR) were directly proportional to the progression of disease. Conclusion: Hemoglobin, RBC count, WBC count (including differential count), NLR, and PLR are the key hematological indicators which predict the severity and mortality risk of breast carcinoma.  


2018 ◽  
Vol 56 (12) ◽  
pp. 2125-2131 ◽  
Author(s):  
Giuseppe Lippi ◽  
Gian Luca Salvagno ◽  
Simona Lampus ◽  
Elisa Danese ◽  
Matteo Gelati ◽  
...  

Abstract Background: Although it is known that glucose concentration exhibits a time-dependent decay in uncentrifuged serum and lithium-heparin blood tubes, no evidence exists on how this variation may depend on blood cell counts (CBC) and volumes. Methods: Venous blood was drawn from 30 non fasting healthy volunteers into three serum and three lithium-heparin tubes. One serum and lithium-heparin tubes were centrifuged within 15 min after collection and glucose was measured with a hexokinase assay. The second and third serum and lithium-heparin tubes were maintained at room temperature for 1 and 2 h after the first tubes were centrifuged. These other tubes were then centrifuged and glucose was measured. CBC was performed in the first lithium-heparin tube, before centrifugation. Results: The mean decrease of glucose was higher in lithium-heparin plasma than in serum (0.33 vs. 0.24 mmol/L/h; p<0.001). Glucose concentration decreased by 7% and 5% per hour in lithium-heparin plasma and serum, respectively. In univariate analysis, the absolute decrease of glucose concentration was associated with sex (higher in men than in women), red blood cell (RBC) count, hematocrit, white blood cell (WBC) count, neutrophils and monocytes in both lithium-heparin plasma and serum. In multivariate analysis, the decrease of glucose concentration remained independently associated with RBC, WBC, neutrophils and monocytes in both sample matrices. No significant association was found with platelet number and erythrocyte or platelet volume. Conclusions: Glucose concentration decrease in uncentrifuged lithium-heparin and serum tubes depends on the baseline number of RBC, WBC, neutrophils and monocytes within the tubes.


Sign in / Sign up

Export Citation Format

Share Document