Potential cost savings by minimisation of blood sample delays on care decision making in urgent care services

Data-driven rapid detection of Helicobacter pylori infection through machine learning with limited laboratory parameters in Chinese primary clinics

Background

Helicobacter pylori (H. pylori) is a significant global health concern, posing a high risk for gastric cancer. Conventional diagnostic and screening approaches are inaccessible, invasive, inaccurate, time-consuming, and expensive in primary clinics.

Objective

This study aims to apply machine learning (ML) models to detect H. pylori infection using limited laboratory parameters from routine blood tests and to investigate the association of these biomarkers with clinical outcomes in primary clinics.

Methods

A retrospective analysis with three ML and five ensemble models was conducted on 1409 adults from Hubei Provincial Hospital of Traditional Chinese Medicine. evaluating twenty-three blood test parameters and using theC 14 urea breath test as the gold standard for diagnosing H. pylori infection.

Results

In our comparative study employing three different feature selection strategies, Random Forest (RF) model exhibited superior performance over other ML and ensemble models. Multiple evaluation metrics underscored the optimal performance of the RF model (ROC = 0.951, sensitivity = 0.882, specificity = 0.906, F1 = 0.906, accuracy = 0.894, PPV = 0.908, NPV = 0.880) without feature selection. Key biomarkers identified through importance ranking and shapley additive Explanations (SHAP) analysis using the RF model without feature selection include White Blood Cell Count (WBC), Mean Platelet Volume (MPV), Hemoglobin (Hb), Red Blood Cell Count (RBC), Platelet Crit (PCT), and Platelet Count (PLC). These biomarkers were found to be significantly associated with the presence of H. pylori infection, reflecting the immune response and inflammation levels.

Conclusion

Abnormalities in key biomarkers could prompt clinical workers to consider H. pylori infection. The RF model effectively identifies H. pylori infection using routine blood tests, offering potential for clinical application in primary clinics. This ML approach can enhance diagnosis and screening, reducing medical burdens and reliance on invasive diagnostics.

Reducing blood sample hemolysis in the emergency department using S-Monovette® in aspiration mode

Background

Blood sample hemolysis continues to be a significant problem in clinical practice. In vitro hemolysis rates up to 77% have been reported in literature. The use of manual aspiration techniques for blood sampling has previously been shown to reduce the burden of erythrocyte injury in the pre-analytical phase compared to the vacuum collection technique. This study compares the hemolysis rates between two blood sampling methods: 5.0 ml BD Vacutainer® SST™ (BDV) and 4.9 ml S-Monovette® serum gel tubes in aspiration mode (SMA).

Methods

This was a prospective randomised controlled study conducted in an Emergency department (ED). A convenience sample of 191 adult patients, aged 18-90 years old, presenting at the ED and requiring blood samples for serum electrolyte was included in the study. Paired blood samples were obtained through an intravenous cannula from each patient with randomised order of blood draw using SMA or BDV. Patient data was obtained and hemolysis index (HI), serum lactate dehydrogenase (LDH), and serum potassium (K) levels measured.

Results

The adjusted mean HI (35.2 vs 21.5 mg/dL, p < 0.001), serum K (4.38 vs 4.16 mmol/L, p < 0.001) and LDH levels (259.6 vs 228.4 U/L, p < 0.001) were significantly higher in blood samples taken using BDV compared to SMA. The frequency of severely hemolyzed (>150 mg/dL) samples was also higher in blood collected using BDV (16.2%) compared to SMA (0%).

Conclusions

The burden of hemolysis in blood samples taken from IV cannulae can be effectively reduced with the use of manual aspiration using the S-Monovette® blood collection system as compared to BD-Vacutainer.

Preanalytical Errors in Clinical Biochemistry Laboratory and Relationship With Hospital Departments and Staff: A Record-Based Study

OBJECTIVE

Despite many efforts to improve the overall quality of clinical laboratory service, errors that compromise patient safety and increase healthcare costs continue to occur even infrequently. We aimed to determine the causes of preanalytical errors and related factors by evaluating the laboratory records of a tertiary hospital.

METHODS

In this retrospective study, 7,762,981 requests that were registered in the biochemistry laboratory records of Ondokuz Mayıs University Health Practice and Research Center during 2019 were examined. All rejected samples were analyzed according to the department where they were collected and the reasons for rejection.

RESULTS

Of the total sample rejections, 99,561 (74.8%) were preanalytical and 33,474 (25.2%) belonged to the analytical phase. The preanalytical rejection rate was 1.28%, with the highest rejection rate in the inpatients (2.26%) and the lowest in the outpatients (0.02%). Insufficient sample (43.7%), clotted sample (35.1%), and inappropriate sample (11.1%) were the first 3 rows of rejection reasons. It was determined that sample rejection rates were low during routine working hours and high during nonworking hours.

CONCLUSIONS

Preanalytical errors were most common in inpatient wards and were most likely due to incorrect phlebotomy techniques. Education of health personnel on good laboratory practices, systematic monitoring of errors, and the development of quality indicators will play an important role in reducing the vulnerability of the preanalytical phase.

Searching for the Optimal Method of Financing Hospital Emergency Departments-Comparison of Polish and Selected European Solutions

Hospital emergency departments are units of the State Medical Rescue system in Poland, which was established to help people in a state of a health emergency. The aim of this study is to develop an optimal method of financing emergency departments in Poland. The study used Polish data from 2016–2019 on the financing of services at the Clinical Emergency Department of the University Clinical Center in Gdańsk. For benchmarking and mathematical modeling, data for the Czech Republic, Germany and Latvia was used. The results of the analysis shows significant differences, to the disadvantage of Clinical Emergency Department, between the potential contract values in the tested models and the actual amounts of funds transferred by the National Health Fund Pomeranian Voivodeship Branch for the activities of Clinical Emergency Department under the concluded contracts. The introduction of co-payment on the part of patients reporting to the emergency departments with minor ailments that do not require hospitalization generates financial revenues, but does not significantly improve the financial results of the analyzed ward. However, it may be educational for patients in terms of raising their awareness of the correct place to seek assistance in the event of a sudden illness.

Managing hemolyzed samples in clinical laboratories

Hemolysis is conventionally defined as membrane disruption of red blood cells and other blood cells that is accompanied by subsequent release of intracellular components into the serum or plasma. It accounts for over 60% of blood sample rejections in the laboratory and is the most common preanalytical error in laboratory medicine. Hemolysis can occur both in vivo and in vitro. Intravascular hemolysis (in vivo) is always associated with an underlying pathological condition or disease, and thus careful steps should always be taken by the laboratory to exclude in vivo hemolysis with confidence. In vitro hemolysis, on the other hand, is highly preventable. It may occur at all stages of the preanalytical phase (i.e. sample collection, transport, handling and storage), and may lead to clinically relevant, yet spurious, changes in patient results by interfering with laboratory measurements. Hemolysis interference is exerted through several mechanisms: (1) spectrophotometric interference, (2) release of intracellular components, (3) sample dilution and (4) chemical interference. The degree of interference observed depends on the level of hemolysis and also on the assay methodology. Recent evidence shows that preanalytical practices related to detection and management of hemolyzed samples are highly heterogeneous and need to be standardized. The Working Group for Preanalytical Phase (WG-PRE) of the European Federation of Clinical Chemistry and Laboratory Medicine (EFLM) has published many recommendations for facilitating standardization and improvement of this important preanalytical issue. Some key EFLM WG-PRE publications related to hemolysis involve: (i) a call for more transparency and some practical recommendations for improving the harmonization of the automatic assessment of serum indices and their clinical usefulness, specifically the hemolysis index (H-index), (ii) recommendations on how to manage local quality assurance of serum or plasma hemolysis/icterus/lipemia-indices (HIL-indices) and (iii) recommendations on how to detect and manage hemolyzed samples in clinical chemistry testing. In this review we provide a comprehensive overview of hemolysis, including its causes and effects on clinical laboratory assays. Furthermore, we list and discuss the most recent recommendations aimed at managing hemolyzed samples in everyday practice. Given the high prevalence of hemolyzed blood samples, the associated costs, the great heterogeneity in how hemolysis is handled across healthcare settings, countries and continents, and increasing patient cross-border mobility, standardization and quality improvement processes aimed at combatting this important preanalytical problem are clearly warranted.