Investigating infective endocarditis: teaching hospitals to choose wisely

Transthoracic echocardiography (TTE) is used to assess for evidence of infective endocarditis (IE). Inappropriate patient selection leads to significant burden on healthcare services. We aimed to assess effectiveness of cardiology consultant vetting of TTE requests for suspected IE in reduction of unnecessary scans. All inpatient TTE requests querying IE over a six-month period were vetted. Clinical information and pathology results were reviewed, and requests were either accepted, deferred, or rejected. A total of 103 patients had TTE requested: 39 (38%) were accepted for scan; four cases of IE were confirmed on TTE. There were 62% of patient requests rejected and not scanned, and no cases of IE subsequently diagnosed. Thus, consultant vetting of TTE requests for suspected IE is an effective way to safely reduce unnecessary scans and enables cost-effective streamlining of echocardiography services.

Introduction

Infective endocarditis (IE) is a potentially fatal infection of the endothelial lining of cardiac structures, with an estimated incidence between 42 and 67 cases per million in England.¹ Mortality without treatment approaches 100%.² The 1994 Duke diagnostic criteria for IE merge clinical, echocardiographic, and microbiological information, and the European Society of Cardiology (ESC) recommend use of the Duke criteria in conjunction with multi-disciplinary team (MDT) opinions of cardiologists and microbiologists.^3,4 Initial investigations include three sets of blood cultures and transthoracic echocardiography (TTE), which remains the ESC recommended first-line imaging for IE. In the UK, TTE use is rising by 6% each year.3 The COVID-19 pandemic caused unprecedented strain on NHS resources, including echocardiography services, with 20 times as many patients waiting over six weeks for outpatient TTE.⁴ The Doncaster and Bassetlaw Teaching Hospitals NHS Foundation Trust (DBTH) cardiology department sought to assess the effectiveness of consultant vetting of TTE requests for suspected IE in reducing unnecessary scans.

Method

Between 1 June and 31 December 2020, all DBTH TTE requests querying endocarditis were vetted by a consultant cardiologist. Local microbiologists offered criteria (table 1) advising if TTE was mandatory, strongly recommended, or to be considered. Requests were accepted, deferred, or rejected. For rejected requests, a letter was sent to referring clinicians. Common reasons for rejection included an alternative clear source of infection on another imaging modality (such as computed tomography), or preliminary cultures showing no significant or contaminant growth following discussion with microbiology. Deferred requests were accepted or rejected after gathering further information. All scans were to the level of the British Society of Echocardiography minimum dataset. All patients were followed up for 16–23 (median 19) months. In deceased patients medical certificate of cause of death (MCCD) data were collected. Data were analysed on Microsoft Excel 2020 (Microsoft, Washington) and assessed for normality: non-parametric data are presented as median (interquartile range, IQR); parametric data are presented as mean ± standard deviation (SD).

Results

There were 128 TTE requests for 103 individuals. Of those, 39 patients were accepted for scan and 64 patients were not scanned. There were 15 scan requests deferred until a final decision was made. Unscanned accepted patients were either discharged or died prior to scan. Rejected but scanned patients had scans organised before request cancellation. Table 2 demonstrates the distribution of results for the accepted and rejected requests. There were four patients diagnosed with acute IE during the study period: all had Staphylococcus aureus bacteraemia, of whom two died with IE listed on MCCD.

Discussion

Our results demonstrate consultant vetting reduced the number of patients requiring echocardiography by 62% (64 patients not scanned). No cases of IE were subsequently diagnosed in rejected requests. This reduction confers approximate savings of £4,380, enabling 64 re-allocated sonographer hours. Furthermore, rationalising scans decreases the burden of overdiagnosis.⁵

A higher number of blood cultures were collected in accepted patients. While three separate sets of cultures can be difficult to obtain and cause patient distress, the low diagnostic yield of TTE necessitates combination with positive cultures to raise the index of suspicion.⁶ Deferring requests allows such supporting evidence to become available. However, patients with difficult intravenous access may have delayed diagnosis and subsequent inferior outcomes.7

In 128 requests, nine scans had evidence of IE; four cases were diagnosed. Of the five individuals with possible vegetations, four (80%) patients were alive at 18 months. The deceased patient had serial equivocal TTE with MCCD stating multi-organ failure from urosepsis. In two (40%) patients, Lambl’s excrescences were seen but clear alternative diagnoses were made. One patient had previously been treated for IE elsewhere, with our repeat TTE only showing possible evidence of IE. One patient had serial TTE (trans-oesophageal scans being unavailable due to COVID-19), one scan with a definite vegetation, repeated negative cultures and ultimately an alternative diagnosis. The four other individuals with definite vegetations, seen on five scans, all died. They had Staph. aureus bacteraemia accounting for the high mortality.8 IE was the cause of death in two (50%); hospital-acquired pneumonia and liver failure in the remainder.

The mortality rate was high; 9 (23.1%) of 39 accepted individuals and 21 (32.8%) of 64 rejected individuals died, possibly a result of small study effect or selection bias. In unscanned deceased patients, no MCCD listed IE. Theoretically, request rejection could cause failure to detect fatal IE in unscanned patients. However, considering untreated mortality, those surviving beyond follow-up could not have had IE.

We recognise the limitations of our study. Specifically, lack of pre-pandemic data, as COVID-19 may have influenced presentation and management. Vetting clinicians could only review available information and deferring requests may delay diagnosis and treatment. The cost savings should be offset with consultant vetting time and requesting time for deferred/repeated requests.

Conclusion

Cardiologist vetting of TTE requests for suspected IE demonstrates a safe way to reduce unnecessary scans, provided there is clear communication between all members of the IE multi-disciplinary team.

Key messages

• Transthoracic echocardiogram (TTE) is a cornerstone of diagnosis for infective endocarditis, however, TTE is often requested inappropriately or with inadequate information
• We investigated whether the vetting of requests for TTE to look for infective endocarditis by experienced consultant cardiologists could provide a safe and effective way of reducing the burden of investigations
• Vetting TTE requests reduced the number of scans performed by 62%; no patients that were not scanned died from infective endocarditis at follow-up
• Experienced cardiologists, along with support from microbiologists, can reduce the need for TTE by reviewing requests, provided there are adequate clinical data available

Conflicts of interest

None declared.

Funding

None.

Study approval

This study was internally registered. Local research ethics committee approval was sought and deemed unnecessary.

Acknowledgements

We recognise the contributions of Catherine Wright, Alex Smith, Howard Briggs, Naomi Bairstow, Sarah Ritzmann and Dr Abraham Chacko from DBTH cardio-respiratory department, and Dr Bala Subramanian from the DBTH microbiology department.

References

1. Quan TP, Muller-Pebody B, Fawcett N et al. Investigation of the impact of the NICE guidelines regarding antibiotic prophylaxis during invasive dental procedures on the incidence of infective endocarditis in England: an electronic health records study. BMC Med 2020;18:84. https://doi.org/10.1186/s12916-020-01531-y

2. Cabell CH, Abrutyn E, Karchmer AW. Bacterial endocarditis: the disease, treatment, and prevention. Circulation 2003;107:e185–e187. https://doi.org/10.1161/01.CIR.0000071082.36561.F1

3. Richards M. Diagnostic: recovery and renewal. Report of the independent review of diagnostic services for NHS England. London: NHS England, 2020. Available from: https://www.england.nhs.uk/wp-content/uploads/2020/11/diagnostics-recovery-and-renewal-independent-review-of-diagnostic-services-for-nhs-england-2.pdf

4. Blake I. Record numbers waiting over six weeks for vital heart scans. British Heart Foundation, 16 November 2021. Available at: https://www.bhf.org.uk/what-we-do/news-from-the-bhf/news-archive/2021/november/record-numbers-waiting-over-six-weeks-vital-heart-scans [accessed November 2022].

5. Welch HG, Black WC. Overdiagnosis in cancer. J Natl Cancer Inst 2010;102:605–13. https://doi.org/10.1093/jnci/djq099

6. Sivak JA, Vora AN, Navar AM et al. An approach to improve the negative predictive value and clinical utility of transthoracic echocardiography in suspected native valve infective endocarditis. J Am Soc Echocardiogr 2016;29:315–22. https://doi.org/10.1016/j.echo.2015.12.009

7. Rajani R, Klein JL. Infective endocarditis: a contemporary update. Clin Med 2020;20:31–5. https://doi.org/10.7861/clinmed.cme.20.1.1

8. Khan MZ, Khan MU, Syed M, Balla S. Trends in microbiology data and association with mortality in infective endocarditis (2002–2017). Am J Cardiol 2021;142:155–6. https://doi.org/10.1016/j.amjcard.2020.12.055