Search

The rapidly evolving field of machine learning (ML), along with artificial intelligence in a broad sense, is revolutionising many areas of healthcare, including laboratory medicine. The amalgamation of the fields of ML and patient-based real-time quality control (PBRTQC) processes could improve the traditional PBRTQC and error detection algorithms in the laboratory. This narrative review discusses published studies on using ML for the detection of systematic errors, non-systematic errors, and combinations of different types of errors in clinical laboratories. The studies discussed used ML for detecting bias, the requirement for re-calibration, samples contaminated with intravenous fluid or EDTA, delayed sample analysis, wrong-blood-in-tube errors, interference or a combination of different types of errors, by comparing the performance of ML models with human validators or traditional PBRTQC algorithms. Advantages, limitations, the creation of standardised ML models, ethical and regulatory aspects and potential future developments have also been discussed in brief.

Method-related variations in the measurement of hormones and the reference intervals used in the clinical laboratory can have a significant, but often under-appreciated, impact on the diagnosis and management of endocrine disorders. This variation in laboratory practice has the potential to lead to an errant approach to patient care and thus could cause harm. It may also be the source of confusion or result in excessive or inadequate investigation. It is important that laboratory professionals and clinicians know about these impacts, their sources, and how to detect and mitigate them when they do arise. In this review article, we describe the historical and scientific context from which inconsistency in the clinical laboratory arises. Examples from the published literature of the impact of the method, reference interval, and clinical decision threshold-related discordances on the assessment and monitoring of various endocrine disorders are discussed to illustrate the sources, causes, and effects of this variability. Its potential impact on the evaluation of growth hormone deficiency and excess, thyroid and parathyroid disorders, hyperandrogenism, hypogonadism, glucocorticoid excess and deficiency, and diabetes mellitus is elaborated. Strategies for assessment and mitigation of the discordance are discussed. The clinical laboratory has a responsibility to recognise and address these issues, and although a lot has been accomplished in this area already, there remains more to be done.

Purpose: Chemical adherence testing is a reliable method to assess adherence to antihypertensive drugs. However, it is expensive and has limited availability in clinical practice. To reduce the number and costs of chemical adherence tests, we aimed to develop and validate a clinical screening tool to identify patients with a low probability of non-adherence in patients with uncontrolled hypertension., Materials and Methods: In 495 patients with uncontrolled hypertension referred to the University Medical Centre Utrecht (UMCU), the Netherlands, a penalised logistic regression model including seven pre-specified easy-to-measure clinical variables was derived to estimate the probability of non-adherence. Non-adherence was defined as not detecting at least one of the prescribed antihypertensive drugs in plasma or urine. Model performance and test characteristics were evaluated in 240 patients with uncontrolled hypertension referred to the Heartlands Hospital, United Kingdom., Results: Prevalence of non-adherence to antihypertensive drugs was 19% in the UMCU and 44% in the Heartlands Hospital population. After recalibration of the model's intercept, predicted probabilities agreed well with observed frequencies. The c-statistic of the model was 0.63 (95%CI 0.53-0.72). Predicted probability cut-off values of 15%-22.5% prevented testing in 5%-15% of the patients, carrying sensitivities between 97% (64-100) and 90% (80-95), and negative predictive values between 74% (10-99) and 70% (50-85)., Conclusion: The combination of seven clinical variables is not sufficient to reliably discriminate adherent from non-adherent individuals to safely reduce the number of chemical adherence tests. This emphasises the complex nature of non-adherence behaviour and thus the need for objective chemical adherence tests in patients with uncontrolled hypertension.

Hypophosphatasia (HPP) is a group of inherited disorders characterised by the impaired mineralisation of bones and/or teeth and low serum alkaline phosphatase (ALP) activity. It is caused by a mutation in the ALPL gene encoding the tissue-non-specific isoenzyme of ALP (TNSALP) resulting in a loss of function. The disease is highly heterogenous in its clinical expression ranging from stillbirth without mineralised bone to the mild form of late adult onset with symptoms and signs such as musculoskeletal pain, arthropathy, lower-extremity fractures, premature loss of teeth or an incidental finding of reduced serum ALP activity. A classification based on the age at diagnosis and the presence or absence of bone symptoms was historically used: perinatal, prenatal benign, infantile, childhood, adult and odontohypophosphatasia. These subtypes are known to have overlapping signs and complications. Three forms of HPP distinguishable by their genetic characteristics have been described: severe, moderate and mild. Severe forms of HPP (perinatal and infantile severe) are recessively inherited, whereas moderate HPP may be dominantly or recessively inherited. The biochemical hallmark of HPP is persistently low serum ALP for age and increase in natural substrates of TNSALP, pyridoxal 5'-phosphate and phosphoethanolamine supported by radiological findings. The diagnosis is confirmed by ALPL sequencing. A multidisciplinary team of experts is essential for the effective management. Calcium restriction is recommended in infants/children to manage hypercalcaemia. A targeted enzyme replacement therapy for HPP has become available and correct diagnosis is crucial to allow early treatment., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2021. No commercial re-use. See rights and permissions. Published by BMJ.)

Background and Purpose: Subarachnoid haemorrhage (SAH) is a potentially life-threatening cause of acute headache. Current conventional practice in the United Kingdom (UK) is for head computed tomography (CT) followed by cerebrospinal fluid (CSF) xanthochromia analysis if the head CT is normal. However, with increasing radiological accuracy, head CT alone may be sufficient to exclude SAH without requiring CSF xanthochromia for confirmation. This study aims to determine whether CSF xanthochromia is still required to confirm exclusion of SAH after normal head CT within a tertiary referral centre., Methods: Data was obtained from Nottingham University Hospitals (NUH) NHS Trust databases on 999 patients presenting with symptoms suspicious of SAH from 2012 to 2015. All patients had CSF xanthochromia analysis when head CT was normal or inconclusive for suspected SAH., Results: A total of 179 patients were diagnosed with SAH (17.9%). 176 patients were diagnosed radiologically and 3 patients required further investigations. The 3 of the 802 patients who underwent lumbar puncture (LP) and were identified as having had SAH presented more than 24 hours after ictus. When stratified according to the time of presentation, a normal CT head was observed in those who presented within 24 hours from ictus (sensitivity of 100% [95% CI 92.5-100] and negative predictive value of 100% [97.2-100])., Conclusion: Within a tertiary referral centre for SAH, a normal head CT has a very high negative predictive value to exclude SAH when carried out within 24 hours from ictus provided a 3rd generation CT scanner is utilised, and the scan is reported by a neuroradiologist. CSF xanthochromia analysis in this cohort may still be indicated in those presenting with a high clinical suspicion of SAH and in hospital settings where neuroradiologists or 3rd generation CT scanners are not easily accessible.