Point-of-care ultrasound for critically-ill patients: A mini-review of key diagnostic features and protocols

Table 1 Characteristics of basic critical care echocardiography

	Key features	Accuracy % (95%CI)	Clinical utility	Limitations
Pericardial effusion	Echo-free space between heart and the parietal layer of the pericardium. 15 mL: Minimum detectable by echocardiography; > 50 mL: Pathological. Nature of the fluid-non-echogenic space (serous fluid), echogenic fluid (blood, pus)	ED physicians using a combination of parasternal short and long axis, apical and subcostal views: (1) Sensitivity 96 (90.4-98.9); (2) Specificity 98 (95.7-98.7); (3) PPV 92.5 (85.8- 96.7); and (4) NPV 98.9 (97.3-99.7). Accuracy: 97.5 (95.7-98.7)[29]	Diagnostic, as a cause of dyspnea; Characterisation of fluid; Estimate size of effusion; Guide approach for pericardiocentesis	Pleural effusion, pericardial fat pad may be mistaken as pericardial effusion. Limited echo windows may affect the sensitivity and specificity of CCE. 4 standard views should be done to assess if the effusion is localised or global[30]
Pericardial tamponade	A pericardial effusion with: (1) Diastolic RV collapse; (2) Systolic RA collapse < 1/3 of cardiac cycle (earliest sign); (3) A plethoric IVC with minimal respiratory variation; and (4) Doppler: Exaggerated respiratory cycle changes in mitral and tricuspid valve in-flow velocities (peak E wave velocity will drop at least 25% (mitral) 40% (tricuspid) in expiration compared to inspiration (suggestive of pulsus paradoxus)	(1) Sensitivity 48-60; Specificity 75-90[31] (sensitivity and specificity improves as the severity increases); (2) RA collapse. Sensitivity 55-97; Specificity 33-100[31]. Absence of both RA systolic, RV diastolic collapse: NPPV 90; Sensitivity 95-97; Specificity 40; (3) Sensitivity 92% but not specific[32]; and (4) Pulsus paradoxus itself: Sensitivity 82% (95%CI: 72%–92%); in the presence of pericardial effusion, positive LR 3.3 (95%CI: 1.8-6.3) and negative LR 0.03 (95%CI: 0.01-0.24)[31]	Identifying tamponade as cause of shock. If found to be the cause of cardiac arrest, and had pericardiocentesis after diagnosis, survival to discharge increased by 15.4% (compared to 1.4% without POCUS)[33]	(1) Plethoric IVC may be caused by chronic lung disease, congestive cardiac failure, tricuspid regurgitation; (2) Patients on mechanical ventilation will not demonstrate plethora because inspiration is generated by positive pressure and hence IVC expands rather than collapses[34]; (3) Doppler techniques require more advanced practitioners of POCUS; and (4) Respiratory variation of the mitral and tricuspid inflows should not be used as a sole criterion for tamponade without the presence of chamber collapse, IVC dilation, or abnormal hepatic vein flows (blunting or reversal of diastolic flows in expiration)
Right ventricular dilation and dysfunction	(1) RV dilatation in PE: Diameter-> 42 mm (base), > 35 mm (mid-level). Longitudinal dimension > 86 mm[35]; (2) RV dysfunction in PE, TAPSE < 17.5 mm, indicated abnormal, RV systolic, function[36]; (3) RV hypokinesis; (4) Right heart thrombi; (5) Ventricular interdependence; (6) Leftward septal displacement; and (7) McConnell sign (Normal contraction or sparing of the RV apex with hypokinesis of midportion of the RV free wall)	(1) Enlargement of the RV compared to the LV. Sensitivity 55. Specificity 86[37]; (2) RV dysfunction indicated by abnormal TAPSE Sensitivity 87. Specificity 91. AUC 0.96 (95%CI: 0.87-1.00)[36]; (3) RV hypokinesis for diagnosis of PE. Sensitivity 70. Specificity 33. Predictor of 30-d mortality in PE. Sensitivity 52.4 (43.7-61.0). Specificity 62.7 (59.5-65.8). NPV 90.6 (88.1-92.7). PPV 16.1 (12.8-19.9)[38]; (4) –; (5) –; (6) –; and (7) Sensitivity 70%. Specificity 33; PPV 67; NNV 36[30]	To identify acute cor pulmonale or pulmonary embolism. Various echocardiographic signs can be used to rule in PE, but none can rule it out. This is due to the known variability of PE presentation, clot burden, and physiologic reserve that contribute to pulmonary vascular resistance and acute RH strain[36]. RV dysfunction in PE found to be predictor of early mortality[38]. Presence of right heart thrombi is associated with an increased risk of death in 30 d	Obtaining adequate RV views in critically ill patients may be challenging, especially post abdominal-surgery with a smaller subcostal window. There are numerous methods available to measure RV size and function, yet the parameter that is the most accurate in the critically ill is controversial[39]. McConnell’s sign may also be present in RV infarct and not just PE (i.e. Not specific for PE)
Left ventricular dysfunction[40]	(1) 2D Biplane; (2) Visual ejection fraction; (3) MAPSE < 12 mm; and (4) E-point septal separation > 7 mm	(1) -; (2) Predicts LVEF < 50%. AUROC 0.8 (0.70-0.90); (3) Predicts LVEF < 50% AUROC 0.73 (0.62-0.84); and (4) Predicts EF < 30%. Sensitivity 100 (95%CI: 62.9-100). Specificity 51.6 (95% CI: 38.6-64.5)[41]	(1) Allows more informed risk counselling, prognostication. Patients with no cardiac activity on PoCUS were much less likely to achieve ROSC, had shorter mean resuscitation times[42]; and (2) Relatively easy and rapid. Internal Medicine physicians were able to identify normal versus decreased LVSF with high sensitivity, specificity, and "good" interrater agreement compared to formal echocardiography after completing a training program[43]	(1) Requires optimal acquisition of endocardial borders, time consuming, requires training; (2) and (3) are rarely done
Variation of IVC diameter with respiration	(1) Collapsibility index, measured 4cm caudal to the right atrium, with a deep standardised inspiration; (2) Distensibility index during intermittent positive pressure ventilation; and (3) IVC collapse of > 50 %	(1) Fluid responsiveness: Depending on whether a standardised or non- standardised spontaneous breath was taken: Sensitivity 66-93 Specificity 99-98[44,45]; (2) Comparable to pulse pressure variation in predicting fluid responsiveness (AUROC 0.75 ± 0.07); (3) Cut off value of 16.5%. Sensitivity 71.4; Specificity 76.5[46]; and (4) In predicting CVP < 8 mmHg: PPV of 87, NPV of 96, AUROC 0.93	Assessment of fluid responsiveness to avoid unnecessarily fluid boluses. The degree to which the CVP falls during spontaneous inspiration depends upon 3 variables: Cardiac function; The drop in pleural pressure; Venous return	Requires a spontaneously breathing patient, able to cooperate and perform a standardised breath. Accuracy affected by point of measurement along the IVC and the angle of insonation, given the cylindrical nature of the IVC and especially for the use of M-Mode measurements. IVC may be dilated in valvulopathies, pulmonary hypertension or in highly trained athletes[25]. May not accurately indicate volume status because venous return can be affected by other factors e.g. vascular tone. IVC collapsibility may be confounded by pressure within the abdominal cavity e.g. Intra-abdominal hypertension, ascites, IPPV

AUROC: Area under receiver operating characteristic; CVP: Central venous pressure; ED: Emergency department; IPPV: Intermittent positive pressure ventilation; IVC: Inferior vena cava (plethoric IVC defined as diameter > 2.1 cm and < 50% inspiratory reduction); LR: Likelihood ratio; LV: Left ventricle; LVEF: Left ventricular ejection fraction; LVSF: Left ventricular systolic function; MAPSE: Mitral annular plane systolic excursion; NPV: Negative predictive value; PE: Pulmonary embolism; PPV: Positive predictive value; RA: Right atrial; ROSC: Return of spontaneous circulation; RV: Right ventricle; TAPSE: Tricuspid annular plane systolic excursion.