Echocardiography delays in suspected infective endocarditis: a single-centre retrospective database analysis

Infective endocarditis (IE) carries significant morbidity and mortality, with reported in-hospital mortality ranging from 15 to 30%. Timely echocardiography is crucial for early diagnosis and optimal management. Current UK and European Society of Cardiology (ESC) guidelines recommend echocardiography within 24 hours for suspected IE, particularly in high-risk cases. We aimed to evaluate adherence to these guidelines at a UK district general hospital (DGH), to assess the relationship between delayed echocardiography and patient outcomes, and to assess the reasons why it may be delayed.

We performed a retrospective audit of 78 patients who were inpatients, undergoing echocardiography for suspected IE. Patient demographics, clinical features, day of the week and timing of echocardiography from initial clinical suspicion, and outcomes (30-day mortality, IE diagnosis per modified Duke criteria) were recorded. Echocardiography delays were defined as imaging beyond 24 hours from the echo being requested. Primary outcome was adherence to imaging timing. Secondary outcome was 30-day mortality.

Among the 78 patients audited, delays exceeded on-time scans (47/78, 60.3%, p=0.044, 95% confidence interval [CI] 48.5 to 71.2%). Day-of-week was associated with <24-hour breach on logistic modelling (p=0.031), worst on Thursday/weekends (100% delayed) vs. Wednesday (60%). Using modified Duke criteria, 32% of patients were deemed to have definite or possible IE. The overall 30-day mortality rate in the audit population was 20.5%. However, echocardiography delays alone did not significantly impact mortality (p=0.594).

In conclusion, despite clear guidelines, delays in echocardiography were prevalent. While delayed imaging alone did not significantly predict mortality, patients diagnosed with IE had significantly higher mortality. The day of the week of the echo request had a statistically significant impact on delays. These results highlight the importance of timely echocardiography to confirm IE diagnosis promptly, and potentially improve outcomes.

Introduction

Infective endocarditis (IE) is a severe, life-threatening infection affecting the cardiac endothelium, carrying a substantial risk of morbidity and mortality.^1,2 In-hospital mortality typically ranges between 15 and 30%, increasing to nearly 40% at one year.^2,3

IE is diagnosed with echocardiography, a non-invasive imaging modality that uses ultrasound waves to create a detailed image of the structure of the heart. Transthoracic echocardiography (TTE) is typically first line, with a transducer being placed on the chest. The more invasive second-line investigation, transoesophageal echocardiogram (TOE) makes use of a specialised probe that is inserted into the oesophagus to help assess the structure of the heart. Using echocardiography to promptly diagnose IE is vital to improving overall patient outcomes and reducing complications, such as systemic embolisation and heart failure.^3,4

Although international guidelines recommend echocardiography be performed within 24 hours in all patients with suspected IE,¹ in real-world clinical practice this can often be very difficult to deliver, especially in district general hospitals (DGH) with smaller echocardiography departments (and, therefore, more limited resources) than those working in large university teaching hospitals. We undertook this retrospective audit to identify the frequency with which there are delays to delivering echocardiography as per published guidelines, and aimed to understand the reasons for this in real-world practice.

Background

We undertook our audit at a busy DGH based in East Lancashire. It provides a full acute take (emergency department [ED], acute medical unit [AMU], general medicine and surgery) with on-site cardiology and a physiologist-led echocardiography service. Echocardiography imaging is available nine to five, Monday to Friday. Extremely pressing or clinically urgent requests can be performed over the weekend, by the on-call cardiologists, however, this would not be via the same request/triage system that this audit is evaluating, and, therefore, for the purposes of this audit, were excluded from analysis. Echocardiography requests arise from ED, acute/medical wards and surgical specialties. There are 15 sonographers that perform echocardiography, as well as six consultant cardiologists. However, all imaging undertaken on the medical wards outside of cardiology and the intensive care unit (ICU) are performed by the sonographers. Echo requests are triaged by the managing sonographer, and the sonographers on the day. There is no consultant input regarding the triage process. Blackburn is characterised by a high-throughput, multi-ethnic population with significant comorbidity burden, making timely imaging pathways operationally important and clinically consequential.

Method

Study design and population

We conducted a retrospective audit, assessing all adult patients on general medicine wards who underwent echocardiography for clinically suspected IE from November 2022 to February 2025. Patients were identified using echocardiography logs, clinical coding, and medical records. Inclusion criteria included any patient who was clinically suspected to have IE and underwent at least one echocardiographic examination (TTE initially, followed by TOE if clinically indicated). We excluded patients who were admitted under cardiology, as we were aiming to get an insight into the wider hospital population, and cardiology inpatients, we believed, may be subject to bias.

Data collection

Data were systematically collected using set criteria. Data were gathered retrospectively by looking at individual echo requests and filtering out those indicated for IE. The data set was then expanded to include patient identification variables, such as demographics (age, gender), clinical presentation (fever, murmur, embolic events, heart failure symptoms), and microbiology results from blood cultures. These were matched against echocardiography request and completion times, and clinical outcomes including 30-day mortality.

Echocardiography delay was specifically defined as any imaging performed more than 24 hours after initial suspicion or documented echocardiography request, as per British Society of Antimicrobial Chemotherapy (BSAC) and European Society of Cardiology (ESC) guidelines.^1,2

Definitions and outcomes

Each patient’s IE diagnosis was categorised according to modified Duke criteria into definite, possible, or rejected IE. The primary outcome was adherence to timing as per guidelines. We also aimed to identify whether there was a relationship between the day of request and the delay in getting the echo. Finally, we included 30-day mortality in our analysis, and aimed to identify whether there was a significant relationship between delays in our cohort and poor outcomes.

Statistical analysis

Data were analysed using SPSS version 26 (IBM, Armonk, NY). Descriptive statistics summarised patient characteristics. Logistic regression and chi-square tests were used to determine the association between echocardiography delays, IE diagnosis, and mortality. Statistical significance was defined as p<0.05.

Results

Patient cohort characteristics

The audit identified 78 patients (median age 64 years, range 24–91). The majority (59%) were male. Common IE risk factors included prosthetic valves (18%), intravenous drug use (14%), and immunosuppression (9%). Clinical presentation typically included fever (83%), with less common presentations, such as new murmurs or embolic events, being much less prevalent (table 1).

Adherence to guidelines

Only 40% (31 patients) underwent echocardiography within the recommended 24-hour period following clinical suspicion (table 2). Most (60%, 47 patients) experienced delays, with median imaging delays of 48 hours, and a range extending up to six days (144 hours). These results were statistically significant (47/78, 60.3%, one-sided exact binomial p=0.044, 95% confidence interval [CI] 48.5 to 71.2%). Of the 11 IE positive patients (confirmed findings on echocardiography), six waited >24 hours for their echo, with a median wait time of 28 hours.

Relationship between day of request and delays

In this cohort of 78 in-hours, non-bank holiday, echocardiography requests, target breach (<24 hours) varied by day of week in a logistic model (likelihood ratio [LR] χ²(6)=13.91, p=0.031), with the poorest performance on Thursday and at weekends (100% delayed) and the lowest delay on Wednesday (60%). However, supporting tests were weaker (Pearson χ² p=0.084; Kruskal–Wallis on continuous hours p=0.127), and a prespecified weekend versus weekday contrast was directionally worse, but not significant (Fisher’s p=0.195), likely reflecting small weekend samples and complete separation.

Clinical outcomes and mortality

Overall 30-day mortality was 20.5% (16 patients). IE-positive patients (definite IE) experienced significantly higher mortality (36%) compared with IE-negative cases (11%, odds ratio [OR] 5.22, p=0.0127) (table 3). Of the three IE positive patients whose scans were delayed for more than 24 hours, two died within 30 days (30-day mortality, 67%). The 30-day mortality in those with echocardiography delays alone were not directly associated with mortality (p=0.594), though indirect associations through delayed interventions could not be excluded.

Discussion

Our audit identified a significant delay in obtaining echocardiography for patients suspected to have IE at our DGH, with 60% of patients experiencing delays beyond the recommended 24-hour timeframe.⁴

Echo timeliness varied across the week even when restricting to in-hours, non-holiday requests: delays were consistently worse on Thursdays and at weekends, and relatively better mid-week. The overall weekday effect was statistically significant, but day-specific estimates, especially for weekends, are imprecise because of small numbers and several 100%-delayed cells (as expected, as only the most urgent echoes are performed over the weekend). Practically, this pattern suggests capacity/scheduling shortfalls late in the week and on weekends; larger samples and separation-robust modelling should confirm these signals before definitive service changes.

Previous studies support the hypothesis of early echocardiography reducing complications, notably embolic events and valvular damage.^3,5 For instance, delayed diagnosis may allow vegetations to enlarge, increasing the risk of systemic embolism, which was observed in 9% of our patients. As expected, and proved by several previous audits, the increased mortality rate in patients diagnosed with IE (36% vs. 11%, p=0.0127) highlights the clinical significance of early detection and management. Despite the lack of a direct correlation between imaging delays and mortality in our cohort, previous studies have shown that early echocardiography facilitates faster initiation of targeted therapy and timely surgical intervention, both of which are associated with better survival outcomes.³

One key factor affecting delays was access to TOE. Our data showed that only one patient underwent TOE, and those requiring it often faced additional delays due to limited on-site availability, as well as logistical issues. This highlights an opportunity for improving referral pathways to regional centres with better access to advanced imaging, alongside better support from the imaging departments to the ward team to help facilitate an efficient procedure.

Another key factor is simply workload. Many times, it was documented that requested echoes were having to be repeatedly chased by the ward team, and were pushed back day after day several times due to clinical pressures. This highlights the necessity of devising a technique to reduce the number of unnecessary requests and streamlining them, to reduce overall workload in the echocardiography department.

The implementation of a consultant-led triage service

One interesting technique identified during our research, was one implemented by Williams et al., in Doncaster and Bassetlaw Teaching Hospitals.⁶ They evaluated the impact of consultant triage for TTE requests for suspected IE. In their research they found several key findings:

TTE requests and approvals: over the six-month period, 103 TTE requests for suspected IE were reviewed. Only 39 patients (38%) met criteria and were approved for a scan, of which four cases of IE were actually confirmed by TTE.

Rejections and patient outcomes: 64 requests (62%) were rejected (no TTE performed) after consultant review. None of these rejected patients were subsequently found to have endocarditis (i.e. no missed IE cases) on follow-up. This indicates that inappropriate requests were safely filtered out.

Resource savings: this triage reduced the TTE scan workload by 62% (avoiding 64 unnecessary scans). The authors note this translated to approximately £4,380 in cost savings and freed approximately 64 hours of sonographer time that could be reallocated to better uses.

This highlights a key factor for our audit. Of the 78 echocardiograms performed and included in our data gathering, 53 were then negative for Duke criteria, with also no radiological evidence of IE as per Duke-ISCVID (International Society for Cardiovascular Infectious Diseases) criteria and, thus, deemed negative for IE. Currently, in this hospital, requests are not triaged by a consultant cardiologist. Rather, they are triaged based on what they are requested for, as well as availability on the day, and this decision lies with the sonographers and admin team. The quality of the request has a large impact on the triage process. In our data gathering, no echo request was rejected, even if they may have been deemed inappropriate requests. Having a consultant triage these requests prior to booking by the admin team, factoring in the history, clinical presentation of the patient, investigations, such as bloods/cultures, and whether there are other likely sources of infection, may help filter out unnecessary requests, and, thus, save time and money.

The role of risk stratification in optimising echocardiography requests

A critical challenge in managing suspected IE is distinguishing high-risk patients who require urgent echocardiography from those who may not benefit from immediate imaging. The VIRSTA score can help prioritise high-risk Staphylococcus aureus bacteraemia cases for urgent TOE, and safely defer imaging in low-risk cases.^7,8 Patients with a low VIRSTA score (<2) have a very low probability of IE, suggesting that unnecessary echocardiograms could be avoided in this group, freeing resources for those at higher risk. Implementing such risk stratification models in routine practice could optimise echocardiography allocation and reduce unnecessary delays for high-risk patients. VIRSTA could also be used alongside Duke criteria, in those patients who have Staph. aureus bacteraemia, to further help clinicians prioritise echo’s based on clinical need.

Another important consideration is that IE remains primarily a clinical diagnosis, with echocardiography serving as a supportive, rather than definitive, diagnostic tool. Several studies have demonstrated that a normal echocardiogram does not exclude IE, particularly in cases where vegetations are small. As many as 12% of IE presentations had negative echo findings.⁹ This reinforces the need for repeat imaging in high-suspicion cases, and highlights the potential benefit of using adjunct imaging techniques, such as cardiac computed tomography (CT), particularly in prosthetic valve infections where standard echocardiography may be insufficient. It also highlights the importance of not being overly reliant on the echo findings, as in nearly all cases reviewed, a negative echo automatically was assumed to be negative for IE, whereas this may not have been the case. This is again highlighted in the case study mentioned, where the initial echo was negative, and the subsequent echo was positive.

Study limitations and areas for improvement

This audit has several limitations that should be acknowledged. First, it is a single-centre, retrospective study, which limits the generalisability of findings. The relatively small sample size (78 patients) may have reduced statistical power, particularly in assessing mortality differences related to imaging delays. A multi-centre audit with a larger patient cohort would provide stronger evidence on this issue.

Second, the definition of ‘clinical suspicion’ used to measure echocardiography delays was based on clinical documentation, which may not have been consistently recorded. Variability in clinician decision-making, particularly in patients presenting with atypical symptoms, such as stroke or sepsis, may have introduced inconsistencies in how promptly echocardiography was ordered.

We did not assess long-term outcomes beyond 30-day mortality. While in-hospital mortality is an important metric, IE has significant long-term implications, including repeat infection, need for late surgical intervention, and complications, such as heart failure. Future studies should include six-month and one-year follow-up data to better understand how delays in echocardiography impact overall prognosis.

Because some day-groups had 100% delays (complete separation), standard logistic estimates were unstable; we, therefore, interpret day-of-week effects cautiously and recommend confirmation with penalised models and larger samples.

Another limitation is the lack of standardised echocardiography protocols at our institution. Some patients may have undergone TTE only, despite strong indications for TOE, due to resource limitations. Ensuring that all patients with Staph. aureus bacteraemia, prosthetic valves, or persistent bacteraemia undergo TOE, as per guidelines, would likely improve diagnostic accuracy and outcomes.

To improve the study design in future audits, a prospective data collection model with predefined time points for suspicion, imaging, and intervention would provide more precise insights into the effects of echocardiography delays.

Clinical implications and recommendations for improvement

The findings of this audit emphasise the need for system-level changes to optimise the timeliness of echocardiography in suspected IE. Several key strategies could be implemented to address delays.

Consultant-led triage service

The majority (68%) of echoes performed in this audit were negative for IE. Implementing a consultant-led triage service may help to filter out inappropriate requests, thus, freeing up slots for more high-risk patients, and reducing cost and workload for the department.

Implementation of an endocarditis pathway

Hospitals with dedicated endocarditis teams have demonstrated improved outcomes by ensuring rapid diagnostic and management pathways.^10,11 Regular multi-disciplinary team (MDT) meetings involving infectious disease specialists, cardiologists, microbiologists, and cardiac surgeons can facilitate faster decision-making.

An alert system for positive blood cultures, particularly Staph. aureus, could prompt automatic echocardiography requests.

Weekend and out-of-hours echocardiography access

A common reason for delays was the lack of weekend and overnight echocardiography services. A formalised weekend scanning rota, or referral pathway to tertiary centres with 24/7 echocardiography availability, could improve access.

Use of risk-stratification tools (VIRSTA) to reduce unnecessary imaging

The VIRSTA score can help prioritise patients who need urgent TOE while avoiding unnecessary echocardiograms in low-risk cases. By incorporating this tool into local protocols, resources could be better allocated, ensuring high-risk patients receive timely imaging. This could also be used alongside Duke criteria, to help decide which patients actually require an echo, and whether IE is really a likely diagnosis. However, it is important to note that VIRSTA is only indicated for Staph. aureus positive patients

Incorporating alternative imaging options in cases of diagnostic uncertainty

Cardiac CT scanning has been shown to increase the diagnostic accuracy of prosthetic valve endocarditis when echocardiographic findings are inconclusive, or in cases out of hours, where time is of the essence and TTE/TOE are not available or not appropriate.

Education and awareness among clinicians

Improved clinician education on the importance of early echocardiography, and the limitations of initial negative imaging, would help ensure that repeat studies are performed in high-suspicion cases. Teaching sessions emphasising clinical red flags for IE, such as embolic signs, fever of unknown origin, and persistent bacteraemia, can improve early recognition.

Prospective data collection and re-audit

A follow-up audit after implementing changes can assess whether these interventions improve adherence to timely echocardiography guidelines and whether they impact patient outcomes.

Conclusion

This audit highlights a significant gap in guideline adherence, with 60% of patients experiencing delayed echocardiography. While delays were not significantly linked to mortality, patients diagnosed with IE did have significantly worse outcomes, reinforcing the need for timely imaging and intervention. Optimising echocardiography workflows, reducing unnecessary imaging via risk stratification and sensible triage, and improving weekend/urgent access to TOE could help improve outcomes for patients with suspected IE.

By implementing structured pathways, such as dedicated endocarditis teams, VIRSTA risk stratification, implementing a consultant-led triage service and improving weekend echocardiography services, hospitals can work toward reducing avoidable delays and ensuring prompt, high-quality care for patients with suspected IE.

Key messages

• Echocardiography for suspected infective endocarditis (IE) should be performed within 24 hours, but this audit found that 60% of inpatient requests breached this target, highlighting a significant gap between guidelines and real-world district general hospital practice
• Delays varied by day of request, with poorer performance late in the week and at weekends, suggesting that service capacity and scheduling pressures contribute to delayed imaging
• Patients with definite or possible IE had substantially higher 30-day mortality than IE-negative patients, reinforcing the importance of early recognition, prompt imaging, and escalation
• Consultant-led triage, structured endocarditis pathways, and risk stratification tools such as VIRSTA in Staphylococcus aureus bacteraemia may help prioritise high-risk patients and reduce unnecessary echocardiography requests

Conflicts of interest

None declared.

Funding

None.

Study approval

Formal study approval was not required as this was an audit, local audit approval was obtained.

References

1. Gould FK, Denning DW, Elliott TS et al. Guidelines for the diagnosis and antibiotic treatment of endocarditis in adults: a report of the working party of the British Society for Antimicrobial Chemotherapy. J Antimicrob Chemother 2012;67:269–89. https://doi.org/10.1093/jac/dks039

2. Wahadat AR, Deckers JW, Budde RPJ et al. Implementation of the 2015 European Society of Cardiology guidelines for the management of infective endocarditis in the Netherlands. Neth Heart J 2020;28:628–36. https://doi.org/10.1007/s12471-020-01489-9

3. Young WJ, Jeffery DA, Hua A et al. Echocardiography in patients with infective endocarditis and the impact of diagnostic delays on clinical outcomes. Am J Cardiol 2018;122:650–5. https://doi.org/10.1016/j.amjcard.2018.04.039

4. Habib G, Lancellotti P, Antunes MJ et al.; ESC Scientific Document Group. 2015 ESC guidelines for the management of infective endocarditis: the Task Force for the Management of Infective Endocarditis of the European Society of Cardiology (ESC). Endorsed by: European Association for Cardio-Thoracic Surgery (EACTS), the European Association of Nuclear Medicine (EANM). Eur Heart J 2015;36:3075–128. https://doi.org/10.1093/eurheartj/ehv319

5. Thuny F, Di Salvo G, Belliard O et al. Risk of embolism and death in infective endocarditis: prognostic value of echocardiography: a prospective multicenter study. Circulation 2005;112:69–75. https://doi.org/10.1161/CIRCULATIONAHA.104.493155

6. Williams J, Rawcliffe M, Mills MT, Warriner DR. Investigating infective endocarditis: teaching hospitals to choose wisely. Br J Cardiol 2023;30:157. https://doi.org/10.5837/bjc.2023.032

7. Tubiana S, Duval X, Alla F et al. The VIRSTA score, a prediction score to estimate risk of infective endocarditis and determine priority for echocardiography in patients with Staphylococcus aureus bacteremia. J Infect 2016;72:544–53. https://doi.org/10.1016/j.jinf.2016.02.003

8. Peinado-Acevedo JS, Hurtado-Guerra JJ, Hincapié C et al. Validation of VIRSTA and Predicting Risk of Endocarditis Using a Clinical Tool (PREDICT) scores to determine the priority of echocardiography in patients with Staphylococcus aureus bacteremia. Clin Infect Dis 2021;73:e1151–e1157. https://doi.org/10.1093/cid/ciaa1844

9. Murdoch DR, Corey GR, Hoen B et al.; International Collaboration on Endocarditis-Prospective Cohort Study (ICE-PCS) Investigators. Clinical presentation, etiology, and outcome of infective endocarditis in the 21st century: the International Collaboration on Endocarditis-Prospective Cohort Study. Arch Intern Med 2009;169:463–73. https://doi.org/10.1001/archinternmed.2008.603

10. Davierwala P, Marin-Cuartas M, Misfeld M, Borger M. The value of an “Endocarditis Team”. Ann Cardiothorac Surg 2019;8:621–9. https://doi.org/10.21037/acs.2019.09.03

11. Roy AS, Hagh-Doust H, Abdul Azim A et al. Multidisciplinary teams for the management of infective endocarditis: a systematic review and meta-analysis. Open Forum Infect Dis 2023;10:ofad444. https://doi.org/10.1093/ofid/ofad444