Development and Evaluation of a Natural Language Processing System for Curating a Trans-Thoracic Echocardiogram (TTE) Database

BACKGROUND

Although electronic health records (EHR) provide useful insights into disease patterns and patient treatment optimisation, their reliance on unstructured data presents a difficulty. Echocardiography reports, which provide extensive pathology information for cardiovascular patients, are particularly challenging to extract and analyse, because of their narrative structure. Although natural language processing (NLP) has been utilised successfully in a variety of medical fields, it is not commonly used in echocardiography analysis.

OBJECTIVES

To develop an NLP-based approach for extracting and categorising data from echocardiography reports by accurately converting continuous (e.g., LVOT VTI, AV VTI and TR Vmax) and discrete (e.g., regurgitation severity) outcomes in a semi-structured narrative format into a structured and categorised format, allowing for future research or clinical use.

METHODS

135,062 Trans-Thoracic Echocardiogram (TTE) reports were derived from 146967 baseline echocardiogram reports and split into three cohorts: Training and Validation (n = 1075), Test Dataset (n = 98) and Application Dataset (n = 133,889). The NLP system was developed and was iteratively refined using medical expert knowledge. The system was used to curate a moderate-fidelity database from extractions of 133,889 reports. A hold-out validation set of 98 reports was blindly annotated and extracted by two clinicians for comparison with the NLP extraction. Agreement, discrimination, accuracy and calibration of outcome measure extractions were evaluated.

RESULTS

Continuous outcomes including LVOT VTI, AV VTI and TR Vmax exhibited perfect inter-rater reliability using intra-class correlation scores (ICC = 1.00, p < 0.05) alongside high R2 values, demonstrating an ideal alignment between the NLP system and clinicians. A good level (ICC = 0.75-0.9, p < 0.05) of inter-rater reliability was observed for outcomes such as LVOT Diam, Lateral MAPSE, Peak E Velocity, Lateral E' Velocity, PV Vmax, Sinuses of Valsalva and Ascending Aorta diameters. Furthermore, the accuracy rate for discrete outcome measures was 91.38% in the confusion matrix analysis, indicating effective performance.

CONCLUSIONS

The NLP-based technique yielded good results when it came to extracting and categorising data from echocardiography reports. The system demonstrated a high degree of agreement and concordance with clinician extractions. This study contributes to the effective use of semi-structured data by providing a useful tool for converting semi-structured text to a structured echo report that can be used for data management. Additional validation and implementation in healthcare settings can improve data availability and support research and clinical decision-making.

Rare presentation of Fabry disease as 'burnt-out' hypertrophic cardiomyopathy

We herein report the case of a 53-year-old man who was historically diagnosed with hypertrophic cardiomyopathy (HCM) and was lost to follow-up, before presenting with end-stage heart failure. This was initially suspected as dilated cardiomyopathy and then 'burnt-out phase' of HCM but subsequently the underlying diagnosis was Fabry disease. Fabry disease is an uncommon lysosomal-storage disease due to reduced or absent activity of the alpha-galactosidase A enzyme. Cardiac involvement most frequently comprises left ventricular hypertrophy. Early treatment of the underlying condition with enzyme replacement therapy may prevent the progression to end-stage heart failure. Fabry disease should be considered in all patients presenting with a clinical phenotype of HCM and a historical diagnosis should be re-evaluated in light of new diagnostic tools. Untreated Fabry can progress to a 'burnt out' phase, whereby initial hypertrophy undergoes eccentric remodelling to a dilated, severely impaired left ventricle.

SAT0399 RISK OF OSTEOARTHRITIS IN AN INCIDENT COHORT OF PEOPLE WITH PSORIATIC ARTHRITIS: A POPULATION-BASED COHORT STUDY

Background:

Psoriatic arthritis (PsA) is a chronic inflammatory arthritis that causes pain, stiffness and swelling around the joints. PsA is reported to affect between 10 and 40% of individuals with psoriasis1and in the majority of patients presents after, or synchronously with, psoriasis onset.2Osteoarthritis (OA) is a common form of non-inflammatory arthritis related to joint degeneration and typically commences late in the fifth decade. PsA and OA have long been considered two distinct arthropathies, however they do have some overlapping features and symptoms and in certain circumstances it can be difficult to differentiate between them, particularly in the small joints of the hands or spine.3

Objectives:

To determine the risk of a diagnosis of osteoarthritis in psoriatic arthritis patients compared to patients with psoriasis and a general population cohort.

Methods:

Incident PsA patients aged 18-89 years at diagnosis were identified from the UK Clinical Practice Research Datalink between 1998 and 2014. All PsA patients were matched to a cohort of patients with psoriasis and a general population cohort (with no psoriasis or PsA) at a 1:4 ratio based on index date, year of birth, sex and general practice. The baseline prevalence of OA of any site was calculated as a percentage for each study cohort and then those prevalent cases were excluded from the numerators and denominators of the incident calculations. The incidence of OA was calculated and relative risks (RRadj), adjusting for body mass index (BMI), were calculated using conditional Poisson regression.

Results:

In total, 6,783 incident PsA patients were identified. The baseline prevalence of OA ranged from 22.1% (CI9521.1-23.1) in the PsA cohort to 12.6% (CI9512.2-13.0) and 11.0% (CI9510.6-11.3) in the psoriasis and general population cohorts respectively. The incidence of OA was significantly higher in the PsA cohort compared to the psoriasis and general population cohorts after adjusting for BMI (RRadj1.68 CI951.46-1.93 and RRadj1.86 CI951.62-2.14 respectively) (Tables 1 and 2).

Conclusion:

An increased risk of OA was observed in patients with PsA compared to patients with psoriasis alone and those in the general population. Further work is needed to determine whether this reflects a true increase in OA risk or misdiagnosed PsA and the extent to which it can be explained by differences in the opportunity for OA diagnosis between cohorts.

Table 1.

Incidence of osteoarthritis in the PsA, psoriasis and general population cohorts

CasesMedian age (years) at diagnosis (IQR)Person yearsIncidence rate per10,000 person yearsCI95Any OAϮGeneral population137459 (51 - 66)125,798109.2(103.5 - 115.0)Psoriasis143259 (53 - 67)122,279117.1(111.0 - 123.2)PsA46459 (51 - 66)28,574162.4(147.6 - 177.2)

Ϯincluding spondylosis

References:

[1]Ogdie A, Weiss P. Rheum Dis Clin North Am. 2015;41(4):545-568.

[2]Tillett Wet al.Rheumatol. 2017;56(12):2109-2113.

[3]McGonagle Det al. Rheumatology. 2015;54(1):29-38.

Table 2.

Risk of osteoarthritis in patients with PsA compared with patients in the general population and patients with psoriasis

PsA compared with a general population cohortPsA compared with a psoriasis cohortUnadjustedAdjusted†UnadjustedAdjusted†RRCI95PRRCI95PRRCI95PRRCI95POA1.871.67-2.10<0.00011.861.62-2.14<0.00011.681.50-1.88<0.00011.681.46-1.93<0.0001

†adjusted for BMI taken as the closest entry within 3 years of the index date

Acknowledgments:

This report is independent research funded by the National Institute for Health Research, Programme Grants for Applied Research [Early detection to improve outcome in patients with undiagnosed PsA (‘PROMPT’), RP-PG-1212-20007]. The views expressed are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care.

Disclosure of Interests:

None declared

Effectiveness and cost-effectiveness of serum B-type natriuretic peptide testing and monitoring in patients with heart failure in primary and secondary care: an evidence synthesis, cohort study and cost-effectiveness model

BACKGROUND

Heart failure (HF) affects around 500,000 people in the UK. HF medications are frequently underprescribed and B-type natriuretic peptide (BNP)-guided therapy may help to optimise treatment.

OBJECTIVE

To evaluate the clinical effectiveness and cost-effectiveness of BNP-guided therapy compared with symptom-guided therapy in HF patients.

DESIGN

Systematic review, cohort study and cost-effectiveness model.

SETTING

A literature review and usual care in the NHS.

PARTICIPANTS

(a) HF patients in randomised controlled trials (RCTs) of BNP-guided therapy; and (b) patients having usual care for HF in the NHS.

INTERVENTIONS

Systematic review: BNP-guided therapy or symptom-guided therapy in primary or secondary care. Cohort study: BNP monitored (≥ 6 months' follow-up and three or more BNP tests and two or more tests per year), BNP tested (≥ 1 tests but not BNP monitored) or never tested. Cost-effectiveness model: BNP-guided therapy in specialist clinics.

MAIN OUTCOME MEASURES

Mortality, hospital admission (all cause and HF related) and adverse events; and quality-adjusted life-years (QALYs) for the cost-effectiveness model.

DATA SOURCES

Systematic review: Individual participant or aggregate data from eligible RCTs. Cohort study: The Clinical Practice Research Datalink, Hospital Episode Statistics and National Heart Failure Audit (NHFA).

REVIEW METHODS

A systematic literature search (five databases, trial registries, grey literature and reference lists of publications) for published and unpublished RCTs.

RESULTS

Five RCTs contributed individual participant data (IPD) and eight RCTs contributed aggregate data (1536 participants were randomised to BNP-guided therapy and 1538 participants were randomised to symptom-guided therapy). For all-cause mortality, the hazard ratio (HR) for BNP-guided therapy was 0.87 [95% confidence interval (CI) 0.73 to 1.04]. Patients who were aged < 75 years or who had heart failure with a reduced ejection fraction (HFrEF) received the most benefit [interactions (p = 0.03): < 75 years vs. ≥ 75 years: HR 0.70 (95% CI 0.53 to 0.92) vs. 1.07 (95% CI 0.84 to 1.37); HFrEF vs. heart failure with a preserved ejection fraction (HFpEF): HR 0.83 (95% CI 0.68 to 1.01) vs. 1.33 (95% CI 0.83 to 2.11)]. In the cohort study, incident HF patients (1 April 2005-31 March 2013) were never tested (n = 13,632), BNP tested (n = 3392) or BNP monitored (n = 71). Median survival was 5 years; all-cause mortality was 141.5 out of 1000 person-years (95% CI 138.5 to 144.6 person-years). All-cause mortality and hospital admission rate were highest in the BNP-monitored group, and median survival among 130,433 NHFA patients (1 January 2007-1 March 2013) was 2.2 years. The admission rate was 1.1 patients per year (interquartile range 0.5-3.5 patients). In the cost-effectiveness model, in patients aged < 75 years with HFrEF or HFpEF, BNP-guided therapy improves median survival (7.98 vs. 6.46 years) with a small QALY gain (5.68 vs. 5.02) but higher lifetime costs (£64,777 vs. £58,139). BNP-guided therapy is cost-effective at a threshold of £20,000 per QALY.

LIMITATIONS

The limitations of the trial were a lack of IPD for most RCTs and heterogeneous interventions; the inability to identify BNP monitoring confidently, to determine medication doses or to distinguish between HFrEF and HFpEF; the use of a simplified two-state Markov model; a focus on health service costs and a paucity of data on HFpEF patients aged < 75 years and HFrEF patients aged ≥ 75 years.

CONCLUSIONS

The efficacy of BNP-guided therapy in specialist HF clinics is uncertain. If efficacious, it would be cost-effective for patients aged < 75 years with HFrEF. The evidence reviewed may not apply in the UK because care is delivered differently.

FUTURE WORK

Identify an optimal BNP-monitoring strategy and how to optimise HF management in accordance with guidelines; update the IPD meta-analysis to include the Guiding Evidence Based Therapy Using Biomarker Intensified Treatment (GUIDE-IT) RCT; collect routine long-term outcome data for completed and ongoing RCTs.

TRIAL REGISTRATION

Current Controlled Trials ISRCTN37248047 and PROSPERO CRD42013005335.

FUNDING

This project was funded by the NIHR Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 21, No. 40. See the NIHR Journals Library website for further project information. The British Heart Foundation paid for Chris A Rogers' and Maria Pufulete's time contributing to the study. Syed Mohiuddin's time is supported by the NIHR Collaboration for Leadership in Applied Health Research and Care West at University Hospitals Bristol NHS Foundation Trust. Rachel Maishman contributed to the study when she was in receipt of a NIHR Methodology Research Fellowship.

Protocol for a systematic review and individual participant data meta-analysis of B-type natriuretic peptide-guided therapy for heart failure

BACKGROUND

Several aggregate data meta-analyses suggest that treatment guided by the serum concentration of natriuretic peptides (B-type natriuretic peptide (BNP) or its derivative N-terminal pro-B-type natriuretic peptide (NT-BNP)) reduces all-cause mortality compared with usual care in patients with heart failure (HF). We propose to conduct a meta-analysis using individual participant data (IPD) to estimate the effect of BNP-guided therapy on clinical outcomes, and estimate the extent of effect modification for clinically important subgroups.

METHODS

We will use standard systematic review methods to identify relevant trials and assess study quality. We will include all randomized controlled trials (RCTs) of BNP-guided treatment for HF that report a clinical outcome. The primary outcome will be time to all-cause mortality. We will collate anonymized, individual patient data into a single database, and carry out appropriate data checks. We will use fixed-effects and random-effects meta-analysis methods to combine hazard ratios (HR) estimated within each RCT, across all RCTs. We will also include a meta-analysis and meta-regression analyses based on aggregate data, and combine IPD with aggregate data if we obtain IPD for a subset of trials.

DISCUSSION

The IPD meta-analysis will allow us to estimate how patient characteristics modify treatment benefit, and to identify relevant subgroups of patients who are likely to benefit most from BNP-guided therapy. This is important because aggregate meta-analyses have suggested that clinically relevant subgroup effects exist, but these analyses have been unable to quantify the effects reliably or precisely.

TRIALS REGISTRATION

PROSPERO 2013: CRD42013005335.

HIV-related neurocognitive impairment screening: the patient's perspective on its utility and psychological impact

Despite ever improving advances in antiretroviral therapy, neurocognitive impairments such as asymptomatic and mild neurocognitive impairment remain a significant problem for the HIV-positive population. We distributed a post-neurocognitive impairment screening service evaluation questionnaire to assess satisfaction and anxiety. Subjects were HIV positive and aged 18-50. They were screened using the Brief Neurocognitive Score and International HIV Dementia Score as well as undergoing screening for anxiety (Generalised Anxiety Disorder Assessment [GAD-7]), depression (Participant Health Questionnaire Mood Scale [PHQ-9]) and memory (Everyday Memory Questionnaire [EMQ-R]). On completion, they were either reassured that the tests were normal or were referred for further investigation. Following assessment, subjects were asked to complete an anonymous satisfaction survey; 101 surveys were analysed. Forty-nine per cent of participants stated that they "felt better" following screening, 43% said it "made no difference", 6% stated it "worried me" and 1% "did not understand". On a scale of 0-10 of helpfulness, the mean score was 7.53. Forty-seven subjects indicated that they were referred for further investigation and 46 subjects that nothing else was needed; 8 reported they did not know. Those referred on rated satisfaction at a mean of 7.54/10 and those with normal screen as 7.09/10 (p = 0.46). Of the groups that were referred for further investigation, 6% said the test "worried them" compared to 4% in the non-referred group. Forty-nine per cent said they "felt better" despite an abnormal result compared to 50% in a normal screening result (p = 0.76). The results of this survey show that screening for neurocognitive impairment by this method is acceptable and helpful to participants. It did not lead to an increase in anxiety and there was no correlation between referred for further investigations and anxiety suggesting concerns about creating undue anxiety by screening and referral are unfounded.

Translational examination of changes in baroreflex function after renal denervation in hypertensive rats and humans

Renal denervation has shown promise in the treatment of resistant hypertension, although the mechanisms underlying the blood pressure (BP) reduction remain unclear. In a translational study of spontaneously hypertensive rats (n=7, surgical denervation) and resistant hypertensive human patients (n=8; 5 men, 33-71 years), we examined the relationship among changes in BP, sympathetic nerve activity, and cardiac and sympathetic baroreflex function after renal denervation. In humans, mean systolic BP (SBP; sphygmomanometry) and muscle sympathetic nerve activity (microneurography) were unchanged at 1 and 6 months after renal denervation (P<0.05). Interestingly, 4 of 8 patients showed a 10% decrease in SBP at 6 months, but sympathetic activity did not necessarily change in parallel with SBP. In contrast, all rats showed significant and immediate decreases in telemetric SBP and lumbar sympathetic activity (P<0.05), 7 days after denervation. Despite no change in SBP, human cardiac and sympathetic baroreflex function (sequence and threshold techniques) showed improvements at 1 and 6 months after denervation, particularly through increased sympathetic baroreflex sensitivity to falling BP. This was mirrored in spontaneously hypertensive rats; cardiac and sympathetic baroreflex sensitivity (spontaneous sequence and the Oxford technique) improved 7 days after denervation. The more consistent results in rats may be because of a more complete (>90% reduction in renal norepinephrine content) denervation. We conclude that (1) renal denervation improves BP in some patients, but sympathetic activity does not always change in parallel, and (2) baroreflex sensitivity is consistently improved in animals and humans, even when SBP has not decreased. Determining procedural success will be crucial in advancing this treatment modality.

Hemodynamic evaluation and midterm outcome of aortic valve replacement with size 19 Perimount prosthetic valve

BACKGROUND

We sought to investigate the effect of patient prosthesis mismatch on hemodynamic profile using dobutamine stress echocardiography, and to evaluate midterm survival of patients undergoing aortic valve replacement with 19-mm Perimount (Baxter Healthcare, Santa Ana, California) aortic prosthetic valves.

METHODS

Between December 1, 1999, and August 17, 2005, 147 patients (mean age, 76.8 +/- 5.51 years) had aortic valves replaced with 19-mm Perimount prostheses. Dobutamine stress echocardiography was performed in a subgroup of 24 patients (mean age, 76.6 +/- 5.60 years). Univariable predictors of peak transprosthetic gradient (PTG) under maximum stress, adjusted for resting PTG, were investigated by regression. Survival in the whole cohort was described, and univariable predictors of survival were investigated by Cox regression.

RESULTS

In the stress echocardiography subgroup, cardiac output (p < 0.0001), PTG (p < 0.0001), and effective orifice area index increased significantly (p = 0.002) under stress. Peak transprosthetic gradient under stress was strongly associated with PTG at rest (p < 0.0001). After controlling for PTG at rest, no other variables were associated with PTG under stress. In the whole cohort, mean duration of follow-up was 2.21 years; 23 patients died. Neither body surface area nor effective orifice area index was significantly associated with survival.

CONCLUSIONS

The 19-mm Perimount aortic prosthesis has acceptable hemodynamic performance. Transvalvular gradients were within a clinically acceptable range, both at rest and under stress. These findings suggest that patient-prosthesis mismatch is unlikely to cause a clinically important problem when the prosthesis is used, which is consistent with survival experience in the whole cohort.

Accident and emergency nurses as researchers: exploring some of the ethical issues when researching sensitive topics

The breadth and complexity of accident and emergency (A&E) services is potentially a fruitful area for nursing and interdisciplinary research. A significant proportion of potential areas for research in A&E settings could be considered as 'sensitive topic areas'. Sensitive topic research raises many challenges; in particular, it raises specific ethical issues. This paper defines and identifies examples of sensitive topic area research in A&E and discusses some of the ethical issues to be considered before embarking on such research in A&E settings.

Antidepressant versus placebo for depressed elderly

BACKGROUND

Depression warranting intervention is found in ten percent of people over the age of 60. Older depressed people are more likely to die than non-depressed. Relatively few receive therapeutic interventions, and those that do, tend to receive low dose antidepressant therapy. Depression in older people is thought to differ in terms of aetiology, presentation, treatment and outcome than in younger people. Concomitant physical illness and increasing social, physical and neurophysiological diversity are associated with the ageing process. Consequently drug treatment of older patients is often carried out in institutions and on patients suffering from multiple physical problems.

OBJECTIVES

To determine the efficacy of antidepressant medication compared with placebo in the treatment of depression in older patients.

SEARCH STRATEGY

The search strategy incorporated: electronic literature searches of databases held by the Cochrane Collaboration Depression, Anxiety and Neurosis Review Group (CCDAN) (see Collaborative Review Group Search Strategy). Reference lists of related reviews and references of located studies. Contact was made with authors working in the field.

SELECTION CRITERIA

All randomised, placebo controlled trials using antidepressants in the treatment of the presenting episode of depression in patients described as elderly, geriatric senile or older adult.

DATA COLLECTION AND ANALYSIS

Two types of data were extracted (if available) from each study. The first type of data was dichotomous data, this consisted of recovered/not recovered. The second, continuous data,included: Hamilton Depression Rating Scale (HAM-D), Montgomery-Asberg Rating Scale (MADRS) and other depression rating scale scores. An analysis using Peto Odds ratios for the dichotomous data and weighted mean difference for continuous data was performed using RevMan 3.1. The presence of heterogeneity of treatment effect was assessed.

MAIN RESULTS

Seventeen trials contributed data to the analyses comparing the efficacy of antidepressant treatment and placebo. Analyses of efficacy was based on 245 patients treated with TCAs (223 with placebo), 365 patients treated with SSRIs (372 with placebo) and 58 patients treated with MAOIs (63 with placebo). The standardised effect size for the three groups respectively were: TCAs; OR: 0.32 (0.21,0.47), SSRIs; OR; 0.51 (0.36,0.72), MAOIs: 0.17 (0.07,0.39).

REVIEWER'S CONCLUSIONS

TCAs, SSRIs and MAOIs are effective in the treatment of older community patients and inpatients likely to have severe physical illness. At least six weeks of antidepressant treatment is recommended to achieve optimal therapeutic effect. There is little evidence concerning the efficacy of low dose TCA treatment. Further trials are required before low dose TCA treatment is routinely recommended.

Out-patient psychodynamic psychotherapy services: an analysis of patient discharge data

OBJECTIVE

To investigate the functioning of a specialist out-patient Psychodynamic Psychotherapy Service by studying demographic characteristics of patient referrals and referral sources. To investigate the effect of number of treatment sessions on outcome; and to discuss the results in the light of some prevailing stereotypes about psychotherapeutic treatment.

DESIGN

Analysis of discharge data on a cohort of patients discharged from a psychotherapy clinic.

SETTING

An out-patient psychotherapy clinic serving the West Sector of the City of Glasgow.

SUBJECTS

All patients discharged between mid 1993 and 1998.

RESULTS

The clinic served all General Practices within eight postal code areas in West Glasgow (catchment population 260,000). Over a five year period 1648 patients were discharged from the clinic; 68% of referrals were from General Practitioners, 18% from Psychiatrists and the remainder from Clinical Psychologists, Consultant Physicians and Surgeons. A wide range of diagnostic conditions were encountered with personality disorder being diagnosed in more than one-third of patients seen. Successful outcome, assessed by patients' subjective ratings on a four point scale at termination of treatment, showed a 50% improvement rate achieved between 7-12 treatment sessions in the Non Personality Disorder group, and between 12-20 treatment sessions in the Personality Disorder group. In total the proportion of not improved and improved ratings was 61% and 39% respectively. Four hundred (73%) of the not improved group received, five sessions or less.

CONCLUSIONS

The results indicated that the Psychotherapy Service treated a large number of patients, many from middle to high areas of deprivation, and that improvement could be effected over relatively short time periods. The data did not conform to existing stereotypes such as psychotherapy being confined to the treatment of "worried well", affluent groups. It took twice as many sessions to achieve a 50% improvement rate in patients with a diagnosis of personality disorder compared with patients without such a diagnosis.

The monitoring of contaminated impervious planar surfaces

The responses of several types of surface contamination monitor have been investigated for three typical working surfaces contaminated by three radionuclides commonly used in diagnostic or analytical biomedical applications. Of the three radionuclides, 99mTc is an intermediate-energy gamma emitter, 125I decays by electron capture, emitting low-energy x-rays, and 32P is a moderate-energy beta emitter. For each radionuclide monitored values have generally been found to be within 30% of projected response, this being based on calibrations which make use of standard radionuclide sources and conditions. The several types of contaminated non-absorbent surface that have been investigated produce monitored values which for a given type of monitor and radionuclide cannot be differentiated from each other.