Machine Learning-Based Prediction of Hemoglobinopathies Using Complete Blood Count Data.

BACKGROUND: Hemoglobinopathies, the most common inherited blood disorder, are frequently underdiagnosed. Early identification of carriers is important for genetic counseling of couples at risk. The aim of this study was to develop and validate a novel machine learning model on a multicenter data set, covering a wide spectrum of hemoglobinopathies based on routine complete blood count (CBC) testing. METHODS: Hemoglobinopathy test results from 10 322 adults were extracted retrospectively from 8 Dutch laboratories. eXtreme Gradient Boosting (XGB) and logistic regression models were developed to differentiate negative from positive hemoglobinopathy cases, using 7 routine CBC parameters. External validation was conducted on a data set from an independent Dutch laboratory, with an additional external validation on a Spanish data set (n = 2629) specifically for differentiating thalassemia from iron deficiency anemia (IDA). RESULTS: The XGB and logistic regression models achieved an area under the receiver operating characteristic (AUROC) of 0.88 and 0.84, respectively, in distinguishing negative from positive hemoglobinopathy cases in the independent external validation set. Subclass analysis showed that the XGB model reached an AUROC of 0.97 for ß-thalassemia, 0.98 for α0-thalassemia, 0.95 for homozygous α+-thalassemia, 0.78 for heterozygous α+-thalassemia, and 0.94 for the structural hemoglobin variants Hemoglobin C, Hemoglobin D, Hemoglobin E. Both models attained AUROCs of 0.95 in differentiating IDA from thalassemia. CONCLUSIONS: Both the XGB and logistic regression model demonstrate high accuracy in predicting a broad range of hemoglobinopathies and are effective in differentiating hemoglobinopathies from IDA. Integration of these models into the laboratory information system facilitates automated hemoglobinopathy detection using routine CBC parameters.

Diagnostic performance of machine learning models using cell population data for the detection of sepsis: a comparative study.

OBJECTIVES: To compare the artificial intelligence algorithms as powerful machine learning methods for evaluating patients with suspected sepsis using data from routinely available blood tests performed on arrival at the hospital. Results were compared with those obtained from the classical logistic regression method. METHODS: The study group consisted of consecutive patients with fever and suspected infection admitted to the Emergency Department. The complete blood counts (CBC) were acquired using the Mindray BC-6800 Plus analyser (Mindray Diagnostics, Shenzhen, China). Cell Population Data (CPD) were also recorded. The ML and artificial intelligence (AI) models were developed; their performance was evaluated using several indicators, such as the area under the receiver operating curve (AUC), calibration plots and decision curve analysis (DCA). RESULTS: Overall, all the tested approaches obtained an AUC>0.90. The logistic regression (LR) performed well compared to the ML/AI models. The naïve Bayes and the K-nearest neighbour (KNN) methods did not show good calibration properties. The multi-layer perceptron (MLP) model was the best in terms of discrimination, calibration and clinical usefulness. CONCLUSIONS: The best performance in the early detection of sepsis was achieved using the ML and AI models. However, external validation studies are needed to strengthen model derivation and procedure updating.

Recent advances in laboratory hematology reflected by a decade of CCLM publications.

On the occasion of the 60th anniversary of Clinical Chemistry and Laboratory Medicine (CCLM) we present a review of recent developments in the discipline of laboratory hematology as these are reflected by papers published in CCLM in the period 2012-2022. Since data on CCLM publications from 1963 to 2012 are also available, we were able to make a comparison between the two periods. This interestingly revealed that the share of laboratory hematology papers has steadily increased and reached now 16% of all papers published in CCLM. It also became evident that blood coagulation and fibrinolysis, erythrocytes, platelets and instrument and method evaluation constituted the 'hottest' topics with regard to number of publications. Some traditional, characteristic CCLM categories like reference intervals, standardization and harmonization, were more stable and probably will remain so in the future. With the advent of important newer topics, like new coagulation assays and drugs and cell population data generated by hematology analyzers, laboratory hematology is anticipated to remain a significant discipline in CCLM publications.

Analytical evaluation and quality assessment of the ARKRAY ADAMS A1c HA-8190V for Hb A1c.

ARKRAY ADAMS A1c HA-8190V is a fully automated high-pressure liquid chromatography (HPLC) system for the measurement of HbA1c. The runtime is 58 s per sample in the Variant mode and 24 s per sample in the Fast mode. We evaluated the analytical performance of this analyzer in the Variant mode, to verify the quality of analysis, according to the consensus criteria established for this measurand. Reproducibility, trueness, carry-over, linearity, interferences and comparison with ARKRAY ADAMS A1c HA-8180V and HA-8180T analyzers were evaluated. Total CVs were 0.76% (IFCC units) 0.55 (NGSP units) at low HbA1c concentration and 0.85% (IFCC units) 0.68 (NGSP units) at high HbA1c concentration. Mean difference with the target values was -1.25 mmol/mol (-0.119%) and total error at medium level was 2.83% (IFCC units), 2.46% (NGSP units) Carry-over was 0%. Linearity was shown in the range 27-122 mmol/mol (4.6-13.3%). The results were not affected in the range of total Hb 59-199 g/L, labile fraction up to 5.5%, carbamylated up to Hb 6.3% nor acetylated Hb up to 5.3%. Method comparison demonstrated good concordance between the methods. The analyzer demonstrated a high analytical performance adequate for routine clinical use in laboratories with high workloads.

Analytical evaluation and quality assessment of the ARKRAY ADAMS A1c Lite HA-8380V for HbA1c.

INTRODUCTION: ARKRAY ADAMS A1c Lite HA-8380V is a fully automated high- performance liquid chromatography system for the measurement of HbA1c. We aimed to evaluate its analytical performance using the Variant Mode. METHODS: Carry-over, linearity, imprecision, trueness, interferences and comparison against ARKRAY ADAMS A1c HA-8180V and HA-8180T analyzers were studied. RESULTS: Total CVs 0.93% (IFCC units), 0.63% (NGSP units) at low concentration and 1.01% (IFCC) 0.74% (NGSP), at high concentration. Mean difference with the target values was -0.44 mmol/mol (IFCC) -0.04% (NGSP). Carry-over, linearity and method comparison were excellent.The results were not affected in the range of total Hb 39-199 g/L, labile fraction 5.7%, carbamylated Hb 9.1% nor acetylated Hb 7.8%, effect of common variants was negligible. CONCLUSIONS: the analyzer demonstrated very good analytical performances, according to the consensus criteria established for HbA1c; it is adequate for laboratories with medium-low workloads.

Harmonized D-dimer levels upon admission for prognosis of COVID-19 severity: Results from a Spanish multicenter registry (BIOCOVID-Spain study).

Coagulopathy is a key feature of COVID-19 and D-dimer has been reported as a predictor of severity. However, because D-dimer test results vary considerably among assays, resolving harmonization issues is fundamental to translate findings into clinical practice. In this retrospective multicenter study (BIOCOVID study), we aimed to analyze the value of harmonized D-dimer levels upon admission for the prediction of in-hospital mortality in COVID-19 patients. All-cause in-hospital mortality was defined as endpoint. For harmonization of D-dimer levels, we designed a model based on the transformation of method-specific regression lines to a reference regression line. The ability of D-dimer for prediction of death was explored by receiver operating characteristic curves analysis and the association with the endpoint by Cox regression analysis. Study population included 2663 patients. In-hospital mortality rate was 14.3%. Harmonized D-dimer upon admission yielded an area under the curve of 0.66, with an optimal cut-off value of 0.945 mg/L FEU. Patients with harmonized D-dimer ≥ 0.945 mg/L FEU had a higher mortality rate (22.4% vs. 9.2%; p < 0.001). D-dimer was an independent predictor of in-hospital mortality, with an adjusted hazard ratio of 1.709. This is the first study in which a harmonization approach was performed to assure comparability of D-dimer levels measured by different assays. Elevated D-dimer levels upon admission were associated with a greater risk of in-hospital mortality among COVID-19 patients, but had limited performance as prognostic test.

Cardiac troponin and COVID-19 severity: Results from BIOCOVID study.

BACKGROUND: Myocardial injury is a common finding in COVID-19 strongly associated with severity. We analysed the prevalence and prognostic utility of myocardial injury, characterized by elevated cardiac troponin, in a large population of COVID-19 patients, and further evaluated separately the role of troponin T and I. METHODS: This is a multicentre, retrospective observational study enrolling patients with laboratory-confirmed COVID-19 who were hospitalized in 32 Spanish hospitals. Elevated troponin levels were defined as values above the sex-specific 99th percentile upper reference limit, as recommended by international guidelines. Thirty-day mortality was defined as endpoint. RESULTS: A total of 1280 COVID-19 patients were included in this study, of whom 187 (14.6%) died during the hospitalization. Using a nonspecific sex cut-off, elevated troponin levels were found in 344 patients (26.9%), increasing to 384 (30.0%) when a sex-specific cut-off was used. This prevalence was significantly higher (42.9% vs 21.9%; P < .001) in patients in whom troponin T was measured in comparison with troponin I. Sex-specific elevated troponin levels were significantly associated with 30-day mortality, with adjusted odds ratios (ORs) of 3.00 for total population, 3.20 for cardiac troponin T and 3.69 for cardiac troponin I. CONCLUSION: In this multicentre study, myocardial injury was a common finding in COVID-19 patients. Its prevalence increased when a sex-specific cut-off and cardiac troponin T were used. Elevated troponin was an independent predictor of 30-day mortality, irrespective of cardiac troponin assay and cut-offs to detect myocardial injury. Hence, the early measurement of cardiac troponin may be useful for risk stratification in COVID-19.

Leukocyte differential and reactive lymphocyte counts from Sysmex XN analyzer in the evaluation of SARS-CoV-2 infection.

BACKGROUND: Leukocyte differential present certain features in COVID 19 patients. RE-LYMP (reactive lymphocytes) is an extended inflammation parameter (EIP) reported by XN analyzer (Sysmex Corporation, Kobe, Japan) reflect the activation of lymphocytes triggered by infections. We aimed to assess the clinical utility of these parameters as biomarkers for the rapid detection of COVID 19. METHODS: The study group included 200 COVID 19 and 167 patients with other infections at admission. Differences of leukocyte differential, neutrophil/lymphocyte ratio (NLR) and EIP among groups were assessed with the Kruskal-Wallis test; parameters statiscally different in the groups were tested with Receiver operating characteristic (ROC) curve analysis to assess their diagnostic performance in distinguishing SARS-CoV-2 infections. The reliability of the selected parameters was evaluated in a validation group of 347 patients (160 COVID 19 and 187 other infections). RESULTS: NLR performed well to discard viral infections, area under curve (AUC) 0.988 (95%CI 0.973 - 0.991) and RE-LYMP was useful to distinguish COVID 19 and bacterial infections AUC 0.920 (95%CI 0.884 - 0.948); the two conditions NLR> 3.3 RE-LYMP> 0.6% was applied to the validation group and 153 out of 160 COVID 19 patients were correctly distinguished (95.6%). CONCLUSIONS: Early diagnosis of SARS-CoV-2 infection is critical for better caring of patients and to reduce the threat of nosocomial infection for professionals. Leukocyte differential and RE-LYMPH could assist in a preliminary differential diagnosis of the disease.

Characteristics and laboratory findings on admission to the emergency department among 2873 hospitalized patients with COVID-19: the impact of adjusted laboratory tests in multicenter studies. A multicenter study in Spain (BIOCOVID-Spain study).

Identification of predictors for severe disease progression is key for risk stratification in COVID-19 patients. We aimed to describe the main characteristics and identify the early predictors for severe outcomes among hospitalized patients with COVID-19 in Spain. This was an observational, retrospective cohort study (BIOCOVID-Spain study) including COVID-19 patients admitted to 32 Spanish hospitals. Demographics, comorbidities and laboratory tests were collected. Outcome was in-hospital mortality. For analysis, laboratory tests values were previously adjusted to assure the comparability of results among participants. Cox regression was performed to identify predictors. Study population included 2873 hospitalized COVID-19 patients. Nine variables were independent predictors for in-hospital mortality, including creatinine (Hazard ratio [HR]:1.327; 95% Confidence Interval [CI]: 1.040-1.695, p = .023), troponin (HR: 2.150; 95% CI: 1.155-4.001; p = .016), platelet count (HR: 0.994; 95% CI: 0.989-0.998; p = .004) and C-reactive protein (HR: 1.037; 95% CI: 1.006-1.068; p = .019). This is the first multicenter study in which an effort was carried out to adjust the results of laboratory tests measured with different methodologies to guarantee their comparability. We reported a comprehensive information about characteristics in a large cohort of hospitalized COVID-19 patients, focusing on the analytical features. Our findings may help to identify patients early at a higher risk for an adverse outcome.

Role of RDW in mathematical formulas aiding the differential diagnosis of microcytic anemia.

Many mathematical formulas containing simple red blood cell parameters have been proposed for differentiating between iron deficiency and thalassemia in patients with microcytic anemia. Approximately half of these formulas do include red cell distribution width (RDW), along with other red cell parameters. In the present study we investigated the role of RDW, expressed in relative or in absolute units in relation with the formulas' discriminant performance. We used a database containing over 2200 subjects with microcytic anemia, for whom a final diagnosis (iron-deficiency anemia, thalassemia, both or other) was available. Performance of the discriminant formulas was assessed by Receiver Operator Curve analysis. Substitution of relative by absolute RDW resulted in statistically significant performance increase (area under the ROC curve) in 16 out of 23 formulas, predominantly due to increased specificity. Relevant performance deterioration was seen in only three formulas that had low initial performance already with the original relative RDW. For optimal differential diagnostic performance, an RDW-based formula for distinguishing thalassemia from iron-deficiency anemia in microcytic anemia should contain 'absolute' instead of relative RDW.

Reticulocyte hemoglobin and hypochromic erythrocytes in the study of erythropoiesis in patients with inflammatory bowel disease.

Anemia due to lack of iron (absolute or functional deficiency) is a common complication of inflammatory bowel disease (IBD). We investigated the potential utility of reticulocyte hemoglobin content (Ret-He) and percentage of hypochromic red cells (Hypo-He) in the assessment of erythropoiesis. We recruited 123 anemic outpatients with IBD. Biochemical tests for iron deficiency were carried out. Full blood counts were performed on a XN 20 system (Sysmex Diagnostics). Differences among groups were studied using analysis of variance and post-hoc tests, considering p < .05 to be significant. Receiver operating characteristic analysis was used to assess the diagnostic performance of Ret-He and Hypo-He for detecting iron-deficient erythropoiesis. The gold standard used for diagnosing iron deficiency was soluble transferrin receptor (sTfR), with a cut-off of >52 nmol/L. Overall, 60 patients had iron deficiency anemia (IDA), 27 anemia of chronic disease (ACD) and 36 mixed ACD + IDA. Ret-He showed the best performance, with an area under curve (AUC) of 0.858 (95% CI 0.816-0.952), considering a cut-off of 30.0 pg, sensitivity of 76.8% and specificity of 99.8% (vs. AUC 0.727 [95% CI 0.624-0.814], considering a cut-off of 4.0%, sensitivity 72.0% and specificity 72.5% for Hypo-He). Ret-He and Hypo-He can be used to assess iron supply for erythropoiesis in patients with IBD, to evaluate long-term (Hypo-He) and short-term (Ret-He) periods.

NEUTROPHIL SCATTERING DATA DRIVEN PRE-MICROSCOPIC FLAGGING OF ACUTE LEUKEMIC CASES.

BACKGROUND: The hematology analyzer, Sysmex XN-1000, generates white blood cell count with varying scattering intensities during a complete blood count (CBC) analysis. OBJECTIVES: The objectives of the study were to study the predictive role of median and coefficient of variation of neutrophil scattering items in blood samples for differentiation of leukemic subjects. METHODS: We evaluated six neutrophil scattering parameters: neutrophil side scatter mean intensity, neutrophil side fluorescence light (SFL) mean intensity, neutrophil forward scatter mean intensity, neutrophil side scatter area distribution width (NE-WX), neutrophil SFL area distribution width (NE-WY), and neutrophil forward scatter area distribution width (NE-WZ), measured in white blood cell differential scattergram generated by the hematology analyzer (Sysmex XN-1000) at an academic medical center. RESULTS: We collected 433 blood samples from acute myeloid leukemia (AML) and acute lymphoid leukemia (ALL) cases and normal controls. AML group showed highly significant differences in the mean values compared with the control group. Out of six neutrophil scattering items, NE-WX, NE-WY, and NE-WZ showed high efficiency, with area under the curve (AUC) values of 0.764, 0.748, and 0.757, respectively, to differentiate AML from ALL cases and control groups. When comparing combined acute leukemia cases (AML plus ALL) with the control group, NE-WX, NE-WY, and NE-WZ generated highly significant AUC values (0.840, 0.884, and 0.801, respectively). CONCLUSION: The neutrophil scattering parameters generated during CBC analysis provide a new tool for the prediction of acute leukemia and its lineage.

Improvement in detecting sepsis using leukocyte cell population data (CPD).

Background The cell population data (CPD) parameters reported by XN analyzers (Sysmex Corporation, Kobe, Japan) reflect the size and internal structure of leukocytes. We explored whether CPD values could contribute to recognize those patients with fever at risk to develop sepsis. A profile of sepsis was developed combining CPD parameters and other markers. Methods We recruited 295 patients at the onset of fever, with infection confirmed by positive cultures. We studied the diagnostic performance of the CPD parameters in the differential diagnosis of sepsis vs. non-systemic bacterial infection using receiver operating characteristic (ROC) curve analysis. Additionally, the K-means unsupervised clustering method was applied. Once the clusters had been defined, the relationship between them and the CPD parameter values was assessed with the non-parametric Wilcoxon test. Lastly, the relationship between the clusters obtained and the categorical variables was examined with the χ2-test (or Fisher's exact test). Results ROC analysis demonstrated that NE-FSL, NE-WY, NE-WZ and MO-WZ had areas under the curve (AUCs) >0.700 for predicting sepsis. Using the K-means clustering algorithm, 80 patients (66.67%) were assigned to Cluster 1 and the others to Cluster 2. Out of 80 of patients in Cluster 1, 45 (56.25%) presented a PCT value ≥2 ng/mL, whereas almost 80% of Cluster 2 patients had a PCT <2 ng/mL. Cluster 1 was characterized by high NE-SFL, NE-WY, MO-X, MO-WX and MO-Z values (p<0.05). Conclusions CPD related to monocyte complexity and neutrophil activation were found to be significant, with high values suggesting sepsis.

Assessment of iron-restricted erythropoiesis in chronic renal disease: evaluation of Abbott CELL-DYN Sapphire mean reticulocyte hemoglobin content (MCHr).

We studied the performance of mean reticulocyte hemoglobin content (MCHr), measured on Abbott CELL-DYN Sapphire analyzer for the detection of functional iron deficiency. Patients with anemia and with renal disease were prospectively selected from the outpatient population of our hospital, 512 subjects were studied. Diagnoses and other medical data were retrieved from the hospital information system. MCHr was measured using a CELL-DYN Sapphire hematology analyzer (Abbott Diagnostics). Standard laboratory and statistical tests were used. Receiver operating characteristic (ROC) analysis was used to establish the diagnostic performance of MCHr for detecting iron-restricted erythropoiesis, with transferrin saturation <20% as the gold standard. Patients in the iron deficiency anemia group all had a state of iron depletion and iron-restricted erythropoiesis (median MCHr 25.4 pg). Patients with anemia of chronic disease showed also low MCHr, median 27.6 pg, but not statistically different from the iron deficient group (p = .0585). Renal patients with iron restriction (n = 66) had significantly lower MCHr (p < .0001) than those receiving adequate iron supply (n = 100): median MCHr were 27.9 and 32.5 pg, respectively. ROC analysis gave sensitivity 84.4% and specificity 80.1% with area under curve 0.863 (95% CI 0.823-0.902) at an MCHr cut-off 30.0 pg. MCHr on CELL-DYN Sapphire has equivalent clinical performance for detecting absolute or functional iron deficiency in patients with chronic kidney disease as previously published for another type of analyzer.

Analytical evaluation of the ADAMS™ A1c HA8180T analyzer for the measurement of HbA1c.

BACKGROUND: The ADAMS™ A1c HA-8180T (ARKRAY, Inc., Kyoto, Japan) is a high pressure liquid chromatography system designed for separation and quantitation of HbA1c, while detecting hemoglobin (Hb) variants and HbA2 , with an analysis time of just 3.5 min. The aim was to evaluate the analytical performance of this analyzer for routine HbA1c quantitation. METHODS: Trueness, imprecision, carry over, linearity, the effect of anemia and correlation with ADAMS™ A1c HA-8180V (in Variant mode) were assessed according to guidelines of the Clinical and Laboratory Standards Institute (CLSI) and manufacturer. The presence of coexisting interfering chemically modified Hb variants was also assessed. RESULTS: The results were found to be accurate (total error of 1.74%) and precise: within-run, between-run, between-day and total coefficients of variation were 0.71%, 0.77%, 0.36%, and 0.93% at low concentrations, and 0.43%, 0.29%, 0%, and 0.48% at high concentrations of HbA1c. Carry over, linearity, and correlation of methods were excellent. HbA1c measurement was independent of total Hb concentration in a range 40-214 g/L, and shows no interference from Hb chemically modified commonly observed in this type of analysis, while common Hb variants are detected. CONCLUSION: The analyzer is a suitable system for glycemic control in patients with diabetes and the detection of Hb variants.

Critical appraisal of discriminant formulas for distinguishing thalassemia from iron deficiency in patients with microcytic anemia.

BACKGROUND: Many discriminant formulas have been reported for distinguishing thalassemia trait from iron deficiency in patients with microcytic anemia. Independent verification of several discriminant formulas is deficient or even lacking. Therefore, we have retrospectively investigated discriminant formulas in a large, well-characterized patient population. METHODS: The investigational population consisted of 2664 patients with microcytic anemia: 1259 had iron deficiency, 1196 'pure' thalassemia trait (877 ß- and 319 α-thalassemia), 150 had thalassemia trait with concomitant iron deficiency or anemia of chronic disease, and 36 had other diseases. We investigated 25 discriminant formulas that only use hematologic parameters available on all analyzers; formulas with more advanced parameters were disregarded. The diagnostic performance was investigated using ROC analysis. RESULTS: The three best performing formulas were the Jayabose (RDW index), Janel (11T), and Green and King formulas. The differences between them were not statistically significant (p>0.333), but each of them had significantly higher area under the ROC curve than any other formula. The Jayabose and Green and King formulas had the highest sensitivities: 0.917 both. The highest specificity, 0.925, was found for the Janel formula, which is a composite score of 11 other formulas. All investigated formulas performed significantly better in distinguishing ß- than α-thalassemia from iron deficiency. CONCLUSIONS: In our patient population, the Jayabose RDW index, the Green and King formula and the Janel 11T score are superior to all other formulas examined for distinguishing between thalassemia trait and iron deficiency anemia. We confirmed that all formulas perform much better in ß- than in α-thalassemia carriers and also that they incorrectly classify approximately 30% of thalassemia carriers with concomitant other anemia as not having thalassemia. The diagnostic performance of even the best formulas is not high enough for making a final thalassemia diagnosis, but in countries with limited resources, they can be helpful in identifying those patients who need further examinations for genetic anemia.

Clinical Value of Hypochromia Markers in the Detection of Latent Iron Deficiency in Nonanemic Premenopausal Women.

BACKGROUND: Iron deficiency (ID) is the most common cause of anemia in fertile women and hemoglobin (Hb) within the reference interval does not exclude ID. The consequence of an imbalance between the iron requirements and supply is a reduction of red-cell Hb content, which causes hypochromic cells. The aim of this study was to assess the reliability of new parameters low Hb density (LHD%), reticulocyte Hb equivalent (RetHe), and percentage of hypochromic erythrocytes (%HypoHe) in the detection of latent ID (LID), defined as depletion of iron stores without anemia. METHODS: Two hundred fifty consecutive nonanemic women in fertile age (18-40 years, mean 33.5 years), whose analyses had been requested by general practitioners, were included. Independent samples t-test, receiver operating characteristic (ROC) curve analysis (gold standard for detecting LID ferritin <30 µg/l), and Cohen's kappa index were applied. RESULTS: One hundred fifty-three women had ferritin within the reference range and Hb >120 g/L; 97 (38.8%) had LID. The results were as follows: %HypoHe-AUC 0.934, cutoff 1.6%, sensitivity 85.7%, specificity 92.1%; RetHe-AUC 0.914, cutoff 29.9 pg, sensitivity 86.8%, specificity 85.7%; LHD%-AUC 0.898, cutoff 5.0%, sensitivity 85.9%, specificity 84.1%. Applying those cutoffs, agreement between ferritin and %HypoHe was κ 0.61 and 0.56 for RetHe and LHD%. CONCLUSIONS: LHD%, %HypoHe, and RetHe emerge as reliable tests for the investigation of LID and could improve the ability to detect ID before anemia is present.