Recognising Kawasaki disease in UK primary care: a descriptive study using the Clinical Practice Research Datalink

Integration and validation of host transcript signatures, including a novel 3-transcript tuberculosis signature, to enable one-step multiclass diagnosis of childhood febrile disease

BACKGROUND

Whole blood host transcript signatures show great potential for diagnosis of infectious and inflammatory illness, with most published signatures performing binary classification tasks. Barriers to clinical implementation include validation studies, and development of strategies that enable simultaneous, multiclass diagnosis of febrile illness based on gene expression.

METHODS

We validated five distinct diagnostic signatures for paediatric infectious diseases in parallel using a single NanoString nCounter® experiment. We included a novel 3-transcript signature for childhood tuberculosis, and four published signatures which differentiate bacterial infection, viral infection, or Kawasaki disease from other febrile illnesses. Signature performance was assessed using receiver operating characteristic curve statistics. We also explored conceptual frameworks for multiclass diagnostic signatures, including additional transcripts found to be significantly differentially expressed in previous studies. Relaxed, regularised logistic regression models were used to derive two novel multiclass signatures: a mixed One-vs-All model (MOVA), running multiple binomial models in parallel, and a full-multiclass model. In-sample performance of these models was compared using radar-plots and confusion matrix statistics.

RESULTS

Samples from 91 children were included in the study: 23 bacterial infections (DB), 20 viral infections (DV), 14 Kawasaki disease (KD), 18 tuberculosis disease (TB), and 16 healthy controls. The five signatures tested demonstrated cross-platform performance similar to their primary discovery-validation cohorts. The signatures could differentiate: KD from other diseases with area under ROC curve (AUC) of 0.897 [95% confidence interval: 0.822-0.972]; DB from DV with AUC of 0.825 [0.691-0.959] (signature-1) and 0.867 [0.753-0.982] (signature-2); TB from other diseases with AUC of 0.882 [0.787-0.977] (novel signature); TB from healthy children with AUC of 0.910 [0.808-1.000]. Application of signatures outside of their designed context reduced performance. In-sample error rates for the multiclass models were 13.3% for the MOVA model and 0.0% for the full-multiclass model. The MOVA model misclassified DB cases most frequently (18.7%) and TB cases least (2.7%).

CONCLUSIONS

Our study demonstrates the feasibility of NanoString technology for cross-platform validation of multiple transcriptomic signatures in parallel. This external cohort validated performance of all five signatures, including a novel sparse TB signature. Two exploratory multi-class models showed high potential accuracy across four distinct diagnostic groups.

Kawasaki disease: Shedding light on a mysterious diagnosis

Kawasaki disease is an acute systemic febrile vasculitis of medium and small arteries, most often occurring in children under age 5 years. This condition is the most common cause of acquired heart disease in children in the developed world. The cause is unclear but is thought to be a hyperimmune reaction to an infectious agent. Diagnosis is clinical; the classic presentation includes persistent fever, lymphadenopathy, oral mucosal changes, conjunctivitis, and rash. Although the disease technically is self-limiting, treatment with IV immunoglobulin (IVIG) and high-dose aspirin is necessary to prevent cardiac complications, such as coronary artery aneurysm, pericarditis, or myocarditis. This article reviews the pathophysiology, clinical presentation, diagnosis, and treatment of Kawasaki disease.

Challenges in early diagnosis of Kawasaki disease in the pediatric emergency department: differentiation from adenoviral and invasive pneumococcal disease

Early recognition and distinction of Kawasaki disease (KD) from other febrile infectious diseases is one of the biggest challenges in pediatric emergency departments (PED). The aim of this study was to assess the utility of clinical findings and routinely used laboratory parameters for early discrimination between KD, invasive pneumococcal disease (IPD) and adenovirosis (AdV). A retrospective, cross-sectional study of children aged 3-36 months consecutively admitted to the PED and diagnosed with either KD (n = 110), AdV (n = 440) or IPD (n = 122) was conducted. At first presentation to the PED, 56.3% of KD patients had none or only one clinical criterion, 31% of patients with AdV and 11% with IPD had > 2 criteria. The levels of platelets (Plt), aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were higher and white blood cells (WBC) significantly lower in KD than in IPD and AdV group. The WBC < 20 ×10⁹/l showed a sensitivity of 80.9% and specificity of 79.7% in comparison to AdV. The ROC curve showed a significant, but low sensitivity for AST, ALT and Plt. The erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) did not show any significant diagnostic accuracy. Significant association between incomplete KD and rash, WBC < 20 ×10⁹ and Plt > 400 ×10⁹/L compared to AdV and conjuctivitis, rash and Plt > 400 × 10⁹/L, was found. Due to the time delay and nonspecific early presentation, differentiating KD from IPD and AdV is challenging. Tools used for identification of patients at risk for severe bacterial infections in PED lack sensitivity for identification of KD cases. New biomarkers are warranted for distinction of KD from IPD or AdV.

Kawasaki disease incidence in children and adolescents: an observational study in primary care

BACKGROUND

Kawasaki disease is reported to be increasing in incidence and is the commonest childhood cause of acquired heart disease in the Western world.

AIM

To determine the current UK incidence of Kawasaki disease across childhood and adolescence; and investigate trends over time and season.

DESIGN AND SETTING

An observational, descriptive study in the UK.

METHOD

The Health Improvement Network (THIN) database of primary healthcare records was searched for codes or text indicating Kawasaki disease. Identified records were compared with a study case definition and a date of onset was assigned to cases. The incidence, age/sex distribution, and trend in seasonal and temporal distribution were estimated (2008-2012).

RESULTS

A total of 110 episodes of Kawasaki disease in 109 children were identified from 3.9 million person-years observation. The incidence of Kawasaki disease was 2.8 per 100 000 person-years (95% confidence interval [CI] = 2.3 to 3.4) when aged <20 years; 9.1 (95% CI = 7.3 to 11.2) aged <5 years, and 3.0 per 100 000 (95% CI = 2.0 to 4.3) across the age groups when possible cases were included. More cases were identified in males (55%) with one-fifth of cases diagnosed after 5 years of age. There was no statistically significant trend in incidence over the study years (P = 0.10 adjusted for sex and month), or between seasons (P = 0.65 adjusted for year and sex).

CONCLUSION

Although the incidence of Kawasaki disease remains low and has stabilised in the UK, GPs should recognise that the condition occurs throughout childhood and across the seasons.

Early Appearance of Principal Symptoms of Kawasaki Disease is a Risk Factor for Intravenous Immunoglobulin Resistance

It is difficult to accurately predict treatment resistance in Kawasaki disease (KD). Patients considered to be low-risk cases often develop resistance to intravenous immunoglobulin (IVIG). We herein examined whether information from the clinical course of KD could improve the prediction accuracy of a previously reported risk score. We retrospectively reviewed the clinical records of 100 KD patients. The clinical characteristics and laboratory data were compared between IVIG-sensitive and IVIG-resistant patients and also between patients with and without coronary artery aneurysm (CAA). The total incidence of IVIG resistance and CAA development was 34 and 13 %, respectively. Multiple regression analysis identified the early appearance of principal symptoms (≤day 2 of the illness) as a risk factor for IVIG resistance (OR 2.88, 95 % CI 1.11-7.44, p = 0.0041), whereas delayed IVIG administration (≥day 6) (OR 2.23, 95 % CI 0.66-7.64, p = 0.018) and IVIG resistance (OR 9.05, 95 % CI 2.27-36.10, p = 0.015) were independent predictors for CAA development. The addition of the first appearance day of principal symptoms into a previously reported scoring system improved its prediction accuracy for IVIG resistance. KD patients who had presented with any principal symptoms within 2 days of fever onset were at a high risk for IVIG resistance regardless of previously reported risk score. A careful medical history-taking that is focused on the clinical course enables a better prediction of IVIG resistance.

The rise of big clinical databases

BACKGROUND

The routine collection of large amounts of clinical data, 'big data', is becoming more common, as are research studies that make use of these data source. The aim of this paper is to provide an overview of the uses of data from large multi-institution clinical databases for research.

METHODS

This article considers the potential benefits, the types of data source, and the use to which the data is put. Additionally, the main challenges associated with using these data sources for research purposes are considered.

RESULTS

Common uses of the data include: providing population characteristics; identifying risk factors and developing prediction (diagnostic or prognostic) models; observational studies comparing different interventions; exploring variation between healthcare providers; and as a supplementary source of data for another study. The main advantages of using such big data sources are their comprehensive nature, the relatively large number of patients they comprise, and the ability to compare healthcare providers. The main challenges are demonstrating data quality and confidently applying a causal interpretation to the study findings.

CONCLUSION

Large clinical database research studies are becoming ubiquitous and offer a number of potential benefits. However, the limitations of such data sources must not be overlooked; each research study needs to be considered carefully in its own right, together with the justification for using the data for that specific purpose.