Uncaptured rates of postpartum venous thromboembolism: a US national analysis

OBJECTIVE

To quantify the proportion of postpartum venous thromboembolism (VTE) readmissions, including those that occur at different hospitals from index admission, and describe risk factors for this outcome.

DESIGN

Retrospective observational study.

SETTING

US hospitals included in the Nationwide Readmissions Database.

SAMPLE

A total of 3 719 238 patients >14 years of age with a delivery-associated hospitalisation in 2014.

METHODS

Univariate analysis was performed to identify patient and hospital factors associated with readmissions. Significant factors were included in multivariate logistic regression to identify independent risk factors. Results were weighted for national estimates.

MAIN OUTCOME MEASURES

Readmission with VTE to both index and different hospitals at 30, 60 and 90 days.

RESULTS

The VTE cumulative readmission rate was 0.053% (n = 1477), 0.063% (n = 1765) and 0.069% (n = 1938) at 30, 60 and 90 days, respectively. Patients were readmitted to different hospitals 31% of the time within 90 days. Risk factors for different hospital VTE readmission were unique and included younger age and initial admission to a small/medium-sized hospital. Initial admission to a for-profit hospital increased the likelihood of readmission to a different hospital.

CONCLUSIONS

Nearly one in three postpartum VTEs are missed by the current quality metrics, with significant implications for outcomes and quality. For-profit hospitals have a significant portion of their VTE readmissions hidden, falsely lowering their readmission rates relative to public hospitals.

TWEETABLE ABSTRACT

US analysis shows 1 in 3 readmissions for postpartum venous thromboembolism currently missed.

Analytical evaluation of the ADAMS™ A1c HA 8180 thalassemia mode high-pressure liquid chromatography analyser for the measurement of HbA2 and HbF

BACKGROUND

ADAMS^™ A1cHA-8180T is a HPLC system; within 3.5 min, it quantifies HbF, HbA₂ , and HbA₀ and flags abnormal peaks. We evaluate its analytical performance for routine estimation of HbA₂ and HbF, and critical tests were performed for identifying β-thalassemia carriers.

METHODS

Trueness imprecision, carry over, linearity, and effect of anemia were evaluated according to ICLH, ICLS, or manufacture's guidelines. Comparison (ADAMS^™ A1c HA-8160T) was performed by running 400 samples from healthy subjects, 30 alpha and 80 beta carriers (range: 1.9-5.7 %).

RESULTS

Trueness - HbA₂ 2.7 %, bias 0.81 %; HbA₂ 5.8 %, bias 0.38 %. HbA₂ 4.0% is not affected by Hb in the range 221-40 g/L. Carry over was negligible. Within run: normal control - CV 1.5 %, high control - CV 0.9 %.Within laboratory: normal control - total CV% 1.59%; high control - 0.92 %. Linearity - y = 1.034x - 0.17, R²  = 0.998 (range: 2.8-4.8%).Method comparison - y = 0.93x + 0.22, R²  = 0.997. HbF imprecision CVs between 0.66 and 1.24% and trueness between 0 and 2.8%. Linearity - y = 1.088x - 0.27, R²  = 0.999 (0.1-5.7%).

CONCLUSIONS

ADAMS^™ A1c HA-8180T provides a rapid and reliable separation of HbA₂ . The measurement is accurate and reproducible, which is needed because of the slight difference between normal and pathological values. The gap in HbA₂ values between normal subjects and β-thalassemia carriers makes this an appropriate method for rapid screening for carriers.

Local verification between the hematological analyzers Sysmex XN-series and XE-5000

INTRODUCTION

The aim of this study was to perform a verification of the hematology analyzer Sysmex XN-2000 by comparing with the previous XE-5000. This study assessed the precision and carryover on the XN-2000 and the systematic error between the both counters according to desirable biological variability criterion and a flag comparison study.

METHODS

Within-run precision and between-batch precision were measured according to the ICSH guidelines. A comparative study was performed analyzing two hundred and six samples of peripheral blood from patients. The statistical study was conducted using the Passing-Bablok and Bland-Altman analyses. The leucocyte flag comparison was made by measuring the efficiency rate.

RESULTS

Between-batch precision was lower than that recommended by the biological variability criterion and manufacturer specifications. The comparison gave nonagreement results for neutrophil and basophil counts according to the criterion of biological variability. Erythroblasts and immature granulocytes showed nonagreement, but there is no available biological variation database for these parameters to compare with. Nevertheless, excellent absolute agreement was found for red blood cell parameters, and for platelet, lymphocyte, monocyte, and eosinophil counts.

CONCLUSIONS

The global results obtained for the precision, comparability, and efficiency provide a satisfactory integration of the XN-2000 in the core laboratory routine and accomplish an optimal reliability.

Low hemoglobin density potential marker of iron availability

INTRODUCTION

Low hemoglobin density (LHD%) is a new parameter provided by Beckman-Coulter derived from the mean cell hemoglobin concentration, using the mathematical sigmoid transformation LHD% = 100×√(1-(1/(1 + e(1.8(30-MCHC)))). This study investigated the reliability of LHD% for the assessment of iron status in the presence of inflammation.

METHODS

Healthy subjects (n = 90) and patients with iron deficiency (IDA, n = 110), chronic kidney disease (CKD, n = 65) and anemia of chronic disease (ACD, n = 85; 24 were iron deficient and 61 were iron sufficient) were analyzed on a LH 780 analyzer (Beckman Coulter Inc., Miami, FL, USA). Independent samples U test and receiver operating characteristic (ROC) curve analysis were applied. To determine the concordance between LHD% and soluble transferrin receptor (sTrR) Cohen's κ index was calculated.

RESULTS

LHD % values showed no statistical difference in patients with IDA and patients with ACD accompanied with IDA (P = 0.6427); LHD% values in these patients were significantly different (P < 0.0001) compared with the iron-sufficient patients with ACD. ROC analysis for LHD% in the detection of iron deficiency showed the following: area under curve 0.903; cut off 5.5%, sensitivity 88.6%, specificity 76.9%; κ index, 0.65.

CONCLUSION

LHD% is a reliable parameter for the detection of iron deficiency in patients with anemia in the presence of inflammation.

Clinical utility of the new Beckman-Coulter parameter red blood cell size factor in the study of erithropoiesis

CHr has been used as a diagnostic tool, together with biochemical markers, to distinguish IDA from ACD, and is incorporated to NKF-K/DOQI guidelines for the monitoring of rHuEPO therapy. The measurement of CHr has been restricted to the analysers of a single manufacturer, Siemens. Red blood cell size factor (RSf) is a new parameter provided by Beckman-Coulter which joins together the volume of erythrocytes and the volume of reticulocytes Rsf = square root (MCV x MRV). The aims of the study were to establish the values of RSf in normal population and in different types of anemia to investigate its clinical usefulness in the study of erythropoiesis and its correlation with CHr. Samples from 449 patients (learning group) were run sequentially on both LH 750 (Beckman-Coulter) and Advia 2120 (Siemens) analysers. Good correlation between CHr and RSf was observed, r(2) = 0.85. Receiver operating characteristic curve analysis for RSf and the diagnosis of restricted erythropoiesis. [table: see text] The diagnostic usefulness of RSf was evaluated on a validation group which included 220 consecutive patients with anemia. This study shows a very good level of agreement between RSf and CHr. Both are suitable parameters for the study of erythropoiesis.