Use of drugs for hyperlipidaemia and diabetes and risk of primary and secondary brain tumours: nested case-control studies using the UK Clinical Practice Research Datalink (CPRD)

OBJECTIVES

Previous studies have suggested that fibrates and glitazones may have a role in brain tumour prevention. We examined if there is support for these observations using primary care records from the UK Clinical Practice Research Datalink (CPRD).

DESIGN

We conducted two nested case-control studies using primary and secondary brain tumours identified within CPRD between 2000 and 2016. We selected cases and controls among the population of individuals who had been treated with any anti-diabetic or anti-hyperlipidaemic medication to reduce confounding by indication.

SETTING

Adults older than 18 years registered with a general practitioner in the UK contributing data to CPRD.

RESULTS

We identified 7496 individuals with any brain tumour (4471 primary; 3025 secondary) in total. After restricting cases and controls to those prescribed any anti-diabetic or anti-hyperlipidaemic medication, there were 1950 cases and 7791 controls in the fibrate and 480 cases with 1920 controls in the glitazone analyses. Longer use of glitazones compared with all other anti-diabetic medications was associated with a reduced risk of primary (adjusted OR (aOR) 0.89 per year, 95% CI 0.80 to 0.98), secondary (aOR 0.87 per year, 95% CI 0.77 to 0.99) or combined brain tumours (aOR 0.88 per year, 95% CI 0.81 to 0.95). There was little evidence that fibrate exposure was associated with risk of either primary or secondary brain tumours.

CONCLUSIONS

Longer exposure to glitazones was associated with reduced primary and secondary brain tumour risk. Further basic science and population-based research should explore this finding in greater detail, in terms of replication and mechanistic studies.

Accuracy of potential diagnostic indicators for coeliac disease: a systematic review protocol

INTRODUCTION

Coeliac disease (CD) is a systemic immune-mediated disorder triggered by gluten in genetically predisposed individuals. CD is diagnosed using a combination of serology tests and endoscopic biopsy of the small intestine. However, because of non-specific symptoms and heterogeneous clinical presentation, diagnosing CD is challenging. Early detection of CD through improved case-finding strategies can improve the response to a gluten-free diet, patients' quality of life and potentially reduce the risk of complications. However, there is a lack of consensus in which groups may benefit from active case-finding.

METHODS AND ANALYSIS

We will perform a systematic review to determine the accuracy of diagnostic indicators (such as symptoms and risk factors) for CD in adults and children, and thus can help identify patients who should be offered CD testing. MEDLINE, Embase, Cochrane Library and Web of Science will be searched from 1997 until 2020. Screening will be performed in duplicate. Data extraction will be performed by one and checked by a second reviewer. Disagreements will be resolved through discussion or referral to a third reviewer. We will produce a narrative summary of identified prediction models. Studies, where 2×2 data can be extracted or reconstructed, will be treated as diagnostic accuracy studies, that is, the diagnostic indicators are the index tests and CD serology and/or biopsy is the reference standard. For each diagnostic indicator, we will perform a bivariate random-effects meta-analysis of the sensitivity and specificity.

ETHICS AND DISSEMINATION

Results will be reported in peer-reviewed journals, academic and public presentations and social media. We will convene an implementation panel to advise on the optimum strategy for enhanced dissemination. We will discuss findings with Coeliac UK to help with dissemination to patients. Ethical approval is not applicable, as this is a systematic review and no research participants will be involved.

PROSPERO REGISTRATION NUMBER

CRD42020170766.

The utility of diagnostic selective nerve root blocks in the management of patients with lumbar radiculopathy: a systematic review

OBJECTIVE

Lumbar radiculopathy (LR) often manifests as pain in the lower back radiating into one leg (sciatica). Unsuccessful back surgery is associated with significant healthcare costs and risks to patients. This review aims to examine the diagnostic accuracy of selective nerve root blocks (SNRBs) to identify patients most likely to benefit from lumbar decompression surgery.

DESIGN

Systematic review of diagnostic test accuracy studies.

ELIGIBILITY CRITERIA

Primary research articles using a patient population with low back pain and symptoms in the leg, SNRB administered under radiological guidance as index test, and any reported reference standard for the diagnosis of LR.

INFORMATION SOURCES

MEDLINE (Ovid), MEDLINE In-Process & Other Non-Indexed Citations, EMBASE, Science Citation Index, Biosis, LILACS, Dissertation abstracts and National Technical Information Service from inception to 2018.

METHODS

Risk of bias and applicability was assessed using the QUADAS-2 tool. We performed random-effects logistic regression to meta-analyse studies grouped by reference standard.

RESULTS

6 studies (341 patients) were included in this review. All studies were judged at high risk of bias. There was substantial heterogeneity across studies in sensitivity (range 57%-100%) and specificity (10%-86%) estimates. Four studies were diagnostic cohort studies that used either intraoperative findings during surgery (pooled sensitivity: 93.5% [95% CI 84.0 to 97.6]; specificity: 50.0% [16.8 to 83.2]) or 'outcome following surgery' as the reference standard (pooled sensitivity: 90.9% [83.1 to 95.3]; specificity 22.0% [7.4 to 49.9]). Two studies had a within-patient case-control study design, but results were not pooled because different types of control injections were used.

CONCLUSIONS

We found limited evidence which was of low methodological quality indicating that the diagnostic accuracy of SNRB is uncertain and that specificity in particular may be low. SNRB is a safe test with a low risk of clinically significant complications, but it remains unclear whether the additional diagnostic information it provides justifies the cost of the test.