📚 Guideline basis
Hahn RT, et al. Guidelines for Performing a Comprehensive Transesophageal Echocardiographic Examination (JASE 2013;26:921–64); Mayo PH, et al. on TEE safety in the ICU (Chest 2017, PMID 26252890); ASE/AIUM guidelines for probe reprocessing and infection control; Zoghbi WA, et al. ASE 2024 prosthetic valve guideline (modality selection).
The decision to escalate
TEE is not "better echocardiography". It is a different examination with different strengths, a small but real procedural risk, and a resource cost. The escalation decision should be protocol-driven rather than discretionary, and rests on two questions: can TTE answer this question at all, and will the answer change management within hours?
Indications where TEE is decisively superior
Indication | Why TTE fails |
Inadequate transthoracic windows — the commonest ICU indication | Ventilation, hyperinflation, dressings, drains, subcutaneous emphysema, post-sternotomy anatomy, obesity, inability to position |
Prosthetic mitral valve assessment | Acoustic shadowing places the entire left atrium behind the prosthesis; TTE cannot exclude prosthetic MR (Chapter 23) |
Infective endocarditis | TTE sensitivity ~50–70% for native vegetations, substantially lower for prosthetic valves and perivalvular abscess (Chapter 24) |
Aortic dissection | TTE cannot exclude type A dissection; TEE approaches CT sensitivity for the ascending aorta and arch (Chapter 16) |
Left atrial appendage thrombus | Not visualised transthoracically |
Post-cardiotomy tamponade | Loculated, posterior, clot rather than fluid; frequently invisible on TTE (Chapter 36) |
ECMO and MCS cannula positioning | Cannula tips and recirculation are poorly seen transthoracically (Chapter 35) |
Intracardiac shunt, PFO, source of embolism | Bubble study far more sensitive; interatrial septum better resolved |
Imaging during CPR | Only modality that images continuously without interrupting compressions (Chapter 31) |
Intraoperative and intraprocedural guidance | Structural interventions require real-time 2D/3D guidance |
Contraindications
Absolute | Relative |
Known oesophageal perforation | Recent oesophageal or gastric surgery |
Oesophageal stricture, obstruction, or tumour | Known oesophageal varices |
Active upper gastrointestinal bleeding requiring endoscopy first | Severe coagulopathy or thrombocytopenia |
Patient refusal (where capacity exists) | Cervical spine instability |
Prior radiotherapy to the mediastinum, Zenker's diverticulum | |
Recent oesophageal food impaction or dysphagia of unknown cause |
Relative contraindications are weighed against diagnostic necessity. In a patient with refractory shock and no transthoracic windows, a relative contraindication rarely outweighs the diagnostic imperative — but the risk-benefit reasoning must be documented, and a gastroenterology opinion is appropriate where time permits.
Safety
Reported serious complication rates in ICU cohorts are consistently below 1%. The principal events, in rough order of frequency:
Complication | Notes |
Transient hypotension or hypertension | Sedation and sympathetic response |
Arrhythmia | Usually transient supraventricular |
Oropharyngeal or dental injury | Bite guard prevents most probe damage and dental injury |
Displacement of the endotracheal tube | The most commonly under-anticipated event; verify tube position and cuff pressure after insertion |
Bleeding (oropharyngeal, oesophageal, gastric) | Higher with coagulopathy and varices |
Oesophageal or gastric perforation | Rare (well under 0.1% in most series) but the event that dominates the risk discussion; mortality is high |
Pressure injury from prolonged probe contact | Relevant particularly for indwelling miniaturised probes (Chapter 27) |
Risk in the critically ill is concentrated in the unstable patient, not the probe. The most frequent adverse sequence in ICU TEE is not oesophageal injury but haemodynamic deterioration during sedation for the procedure in a patient with marginal reserve. Plan sedation with the same care as an induction: pre-optimise volume and vasopressor support, have a vasopressor bolus drawn up, and do not sedate a patient with critical aortic stenosis, severe pulmonary hypertension, or tamponade without anticipating the pressure drop.
Technique
Preparation
- Confirm the indication and the specific question. Write it down; the exam is structured around it.
- Check contraindications, anticoagulation, platelet count, and any history of dysphagia or oesophageal disease.
- Consent where possible; document the risk-benefit reasoning where the patient cannot consent.
- Fasting where circumstances permit; in emergency use, decompress the stomach with a nasogastric tube if one is in place.
- Verify airway security. In the non-intubated patient with respiratory compromise, intubate first rather than sedating for TEE.
- Assemble: bite guard, suction, sedation and vasopressor, monitoring, and a leak-tested, high-level-disinfected probe.
Insertion in the intubated patient
- Ensure adequate sedation and, where indicated, neuromuscular blockade — an inadequately sedated patient is the commonest cause of a failed or traumatic insertion.
- Head in neutral or slightly flexed position.
- Lubricate generously. Insert the probe in the midline with the tip unlocked and neutral.
- Jaw thrust displaces the tongue and opens the hypopharynx; this single manoeuvre resolves most failed blind insertions.
- Advance gently. Resistance at 15–20 cm is the cricopharyngeus; wait, reposition, re-attempt.
- Never force against resistance. If two gentle attempts fail, use direct or video laryngoscopy to visualise the introitus and guide the probe posterior to the endotracheal tube. Forcing is the mechanism of perforation.
- After insertion, re-verify endotracheal tube position, cuff pressure and ventilation.
During the study
- Keep the probe unlocked when not actively imaging.
- Avoid sustained maximal anteflexion or retroflexion in one position; reposition periodically to limit mucosal pressure.
- Limit total imaging time to what the question requires.
- Monitor haemodynamics continuously; the operator is not the person watching the monitor.
Withdrawal
Unlock all controls and return the tip to neutral before withdrawing. Withdrawing a flexed or locked probe is a described mechanism of oesophageal injury. Inspect the probe for blood on removal and document it.
Probe reprocessing
TEE probes are semi-critical devices and require high-level disinfection between every use. Leak testing before and after each use is mandatory — a probe with a breached seal is both an electrical hazard and a vector for contamination, and a failed leak test removes the probe from service. Probe tracking (patient, operator, disinfection cycle) is a governance requirement, not an administrative nicety.
ICU-specific practicalities
Situation | Consideration |
Prone positioning | TEE is feasible in the proned patient with the head turned and has been described in ARDS cohorts; probe manipulation is restricted and views are limited |
During CPR | Insert during a compression cycle; imaging continues uninterrupted; also allows verification of the compression point (Chapter 31) |
Nasogastric tube in situ | Can be left, but may impede insertion; consider withdrawing it to the oesophagus or removing it |
Cervical spine precautions | Maintain in-line stabilisation; consider laryngoscope-assisted insertion |
Repeated studies expected | Consider a miniaturised indwelling probe (Chapter 27) rather than repeated instrumentation |
Coagulopathy | Correct where feasible; use the gentlest possible technique; accept a shorter, more targeted exam |
🛑 Critical pitfall: Forcing a probe past resistance in a sedated patient who cannot report pain. This is how oesophageal perforation happens. Two failed gentle attempts mandate laryngoscope-assisted insertion or abandonment.
🛑 Critical pitfall: Withdrawing the probe while it remains flexed or locked. Always return to neutral first.
🛑 Critical pitfall: Sedating a patient with tamponade, critical aortic stenosis, or severe pulmonary hypertension for TEE without anticipating cardiovascular collapse. Have vasopressor drawn up and consider performing the study with minimal sedation, or after stabilisation.
- 💡 Clinical pearl: The single most effective manoeuvre for a failed blind insertion is a firm jaw thrust by an assistant, not more force on the probe.
- 💡 Clinical pearl: Re-check the endotracheal tube after insertion. Tube displacement during TEE is common and easily missed while attention is on the images.
- 💡 Clinical pearl: Write the clinical question on the report before starting. An ICU TEE performed to answer "is there a paravalvular leak?" is a ten-minute study; the same study performed as an undirected survey takes forty and answers less.
References
- Hahn RT, Abraham T, Adams MS, et al. Guidelines for performing a comprehensive transesophageal echocardiographic examination: recommendations from the ASE and the SCA. J Am Soc Echocardiogr 2013;26:921–64.
- Mayo PH, Narasimhan M, Koenig S. Transesophageal echocardiography in the intensive care unit. Chest 2017. PMID 26252890.
- Expert Round Table on Echocardiography in ICU. International consensus statement on training standards for advanced critical care echocardiography. Intensive Care Med 2014;40:654–66.
- Zoghbi WA, Jone PN, Chamsi-Pasha MA, et al. Guidelines for the evaluation of prosthetic valve function with cardiovascular imaging. J Am Soc Echocardiogr 2024;37:2–63.