Anaesthesia 2013, 68, 1156–1160
doi:10.1111/anae.12422
Original Article Correlation between supra-sternal Doppler cardiac output (USCOM) measurements and chest radiological features L. Huang,1 L. A. H. Critchley,2 R. L. K. Lok3 and Y. Liu4 1 PhD Student, 2 Professor & Honorary Consultant, Department of Anaesthesia and Intensive Care, and 3 Resident, Department of Imaging and Interventional Radiology, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong, China 4 Attending Doctor, Department of Radiology, Peking University Third Hospital, Beijing, China
Summary Cardiac output can be measured non-invasively using supra-sternal Doppler (USCOM, Sydney, NSW, Australia). However, scanning can be difficult in practice in older patients, the reason for which has not been elucidated previously. Chest radiographs from 60 previously studied anaesthetised patients were reviewed and scored for aortic unfolding, enlargement and calcification, and cardiac enlargement. Corresponding supra-sternal Doppler scans were graded as easy or difficult using the Cattermole scoring system. Twenty patients who were difficult to scan, aged 60–88 years, had mean (SD) radiological scores of 5.9 (2.5) out of 12, while 20 adult controls, 40–60 years, and 20 older patients who were easy to scan, 60–80 years, had radiological scores of 0.9 (1.1) and 1.7 (1.4), respectively (p < 0.001). Over 75% of the patients who were difficult to scan had two or more radiological features suggestive of aortic unfolding and cardiac enlargement. Morphological or anatomical changes associated with ageing within the upper chest play an important part in the success of using supra-sternal Doppler in older patients. .................................................................................................................................................................
Correspondence to: L. Huang Email:
[email protected] Accepted: 13 August 2013
There is a growing interest among practising anaesthetists in point-of-care, minimally invasive cardiac output monitoring for high-risk surgery [1]. Several technologies that can provide continuous readings have recently been developed [2]. One of the more promising technologies that can be used in anaesthetised patients is external Doppler via the supra-sternal route. This technology may have advantages such as its reported ability to reflect accurately serial changes in cardiac output when used correctly, the fact that the probe does not enter the body and its low running costs [3]. However, supra-sternal Doppler does have its weaknesses, for example accuracy may depend on 1156
calibration, which uses body height to predict aortic valve diameter and assumes that all patients have a stereotypical body build [4]. In addition, skill and experience may be needed to insonate the aortic valve, focus the beam and capture the optimum flow signal. In some patients, access to the sternal notch, the window for aortic valve insonation, may be difficult because of the type of surgery being performed, and in patients over 50 years of age, it can be very difficult, and in some cases impossible, to detect a reliable flow signal [3, 5]. In 2013, the only commercially available supra-sternal Doppler cardiac output monitor was the USCOM © 2013 The Association of Anaesthetists of Great Britain and Ireland
Huang et al. | Correlation between USCOM and chest X-ray
(USCOM Ltd., Sydney, NSW, Australia). In a previously published study, we found that the quality of USCOM scans declined with the age of the patient studied [3]. In many of the more difficult cases, we noted that the chest X-ray of the patient showed significant morphological changes to the aorta and the heart. We therefore hypothesised that age-related aortic unfolding and calcification, and cardiac enlargement, could explain why older patients were often more difficult to scan using the USCOM. Thus, a retrospective review of chest X-rays from those patients over 60 years of age from this previous study was performed. A scoring system was devised to categorise aortic changes and cardiac enlargement, which was compared with the existing assessments of USCOM scan quality [6, 7].
Methods The Chinese University of Hong Kong and New Territories East Cluster Joint Clinical Ethics Committee approved the previous study and written informed consent was obtained [3]. This included permission to access and use clinical data for research and publication purposes pertaining to the study. Following publication of the original study of 180 patients, we decided to investigate the problems pertaining to difficult USCOM scans further, which included review of the chest X-rays of the original patients, and further ethical permission was deemed unnecessary. As age was the major predictor of unsuccessful or poor USCOM scan in our previous study, we chose to analyse radiographic data from the older and higher-risk groups. Sixty patients were selected from the previous 180 for further analysis. Patients were divided into three groups: 20 elderly patients (over 60 years) in whom USCOM scan had been difficult (Old/difficult); 20 elderly patients in whom the scan had been easy (Old/easy); and a control group of 20 younger patients (40–60 years of age) who were selected because USCOM scanning had been easy. The ability to insonate the aortic valve and obtain a reliable USCOM scan was recorded in the original study and assessed using the Cattermole scoring system, with a score of 5 or below indicating that the scan had been difficult [7]. The Cattermole score assesses the quality of the USCOM scan Doppler flow profiles [8]. Chest X-rays from the selected patients were studied by two specialist qualified radiologists (A and B). © 2013 The Association of Anaesthetists of Great Britain and Ireland
Anaesthesia 2013, 68, 1156–1160
Figure 1 Examples of chest X-rays from patients who were easy and dififcult to scan using the USCOM. In the former (left), the Cattermole score was 10/12 (excellent) and the combined X-ray score was 0/12 (no morphological changes reported). In the patient who was difficult to scan (right) patient, significant aortic enlargement and unfolding were present. The patient’s Cattermole score was < 5 (poor) and the X-ray score was 6/12. The pathway for the USCOM beam is clearly more tortuous. They were blinded to the patient’s age and USCOM scan results and quality. A simple six-point chest Xray score was developed to evaluate changes in the aorta and cardiac size (Fig. 1). One point was awarded for each of: left heart enlargement; right heart enlargement; aortic unfolding (two points if severe); aortic enlargement; and calcification of the aorta. Thus, a normal chest X-ray would score zero points, whereas one with severe aortic changes and cardiac enlargement would score 3–6 points. Each radiologist scored all chest X-rays independent of each other, and the two scores were added together producing a maximum score of 12 points for each patient. Statistical analysis was performed using the statistical package for the social sciences (IBM SPSS Statistics version 20, Armonk, NY, USA). Student’s t-test, ANOVA with Bonferroni correction, Mann–Whitney U-test, chi-squared test and linear regression with correlation were used as appropriate. The intraclass coefficient was used to compare the concordance between the two radiologists when scoring the chest X-rays.
Results Data from 60 patients were analysed (Table 1). The Control group was mainly composed of healthy patients with a predominance of female patients scheduled for gynaecological surgery, whereas the other two 1157
Anaesthesia 2013, 68, 1156–1160
Huang et al. | Correlation between USCOM and chest X-ray
Table 1 Baseline characteristics of patients who underwent USCOM scanning intra-operatively. Old/ difficult = age > 60 years and difficult to scan; Old/ easy = age > 60 years and easy to scan; Control = age 40–60 years and easy to scan. Values are mean (SD) or number (proportion).
Table 3 Detailed breakdown of X-ray assessment scores assigned by both radiologists (A, B). Each score was 0 or 1 except for unfolded aorta, which was 2 if severe. Old/difficult = age > 60 years and difficult to scan; Old/easy = age > 60 years and easy to scan; Control = age 40–60 years and easy to scan.
Old/difficult (n = 20) Age; years Men ASA 1 Height; cm Weight; kg BMI; kg.m 2
70 14 2 160 61 24
(8) (70%) (10%) (9) (9) (4)
Old/easy (n = 20) 70 12 1 158 60 24
Control (n = 20)
(6) (60%) (5%) (8) (8) (2)
50 5 13 158 61 24
(5) (25%) (65%) (9) (13) (6)
Table 2 Quality of USCOM scan scored using the Cattermole system and X-ray scores (sum of scores from two separate radiologists). Old/difficult = age > 60 years and difficult to scan; Old/easy = age > 60 years and easy to scan; Control = age 40–60 years and easy to scan. Values are mean (SD). Old/difficult (n = 20) Cattermole score X-ray score Radiologist A Radiologist B
Old/easy (n = 20)
Control (n = 20)
p value
3.8 (1.7)
8.7 (0.8)
9.2 (1.1)
< 0.001
5.9 (2.5) 3.3 (1.2)
1.7 (1.4) 1.1 (0.8)
0.9 (1.1) 0.4 (0.8)
< 0.001 < 0.001
2.6 (1.4)
0.6 (0.7)
0.5 (0.5)
< 0.001
groups were mainly men and had more comorbidities. There were no differences among the three groups with respect to height, weight or body mass index (Table 1). The ease of USCOM scanning was scored using the Cattermole scoring system, with a low score indicating a more difficult scan. Mean (SD) Cattermole scores were lower in the Old/difficult group compared with the Old/easy and Control groups. In the Old/ difficult group, the mean score was below the accepted 5-point threshold for reliable USCOM scan quality (Table 2). The scores based on chest X-ray evaluation were higher in the Old/difficult group compared with the Old/easy and Control groups (Table 2). Radiologist A scored the chest X-rays on average 0.35 points higher than her counterpart radiologist (p < 0.001). A more 1158
X-ray features Left ventricular enlargement Right ventricular enlargement Aorta dilated Aorta unfolded Aorta calcified Total score (A + B)
Old/difficult (n = 20)
Old/easy (n = 20)
Control (n = 20)
13, 13
5, 4
3, 8
9, 4
0, 0
0, 0
16, 11 20, 15 8, 6 115
4, 0 7, 3 6, 5 34
2, 0 3, 2 0, 0 18
detailed breakdown of the chest X-ray scoring by the two radiologists is provided (Table 3). Radiologist A was more likely to report abnormal aortic findings. The interclass coefficient between the two radiologist’s scores was r = 0.85. As the quality of the USCOM scan decreased, the radiologist’s X-ray assessment scores increased. There was a strong relationship between the Cattermole and chest X-ray score for all patients (R2 = 0.78, p < 0.001). To show this relationship to better effect, a box plot was drawn where the Cattermole scores of the USCOM scan were grouped into excellent (score of 10–12), good (score of 9), fair (score of 6–8) and poor (score of 5 and below). The Old/easy and Control group patients were all excellent to fair, while the Old/difficult group patients were all poor. In the latter, 15 patients (75%) had a combined X-ray score of 4 or more, whereas in the other two groups, no patient scored more than 4 (p < 0.001, Fig. 2).
Discussion Our main finding was that when USCOM scans were more difficult to perform, morphological changes were often found on the routine pre-operative chest X-ray, which indicated that the aorta was unfolded, enlarged and/or calcified and that the heart was enlarged. These chest X-ray changes were mainly present in patients over the age of 60 years, and provided an explanation as to why USCOM scanning becomes more difficult as patients become older when the supra-sternal window and aortic valve are used. © 2013 The Association of Anaesthetists of Great Britain and Ireland
Huang et al. | Correlation between USCOM and chest X-ray
Figure 2 Box-plot comparing the quality of the USCOM scan scored using the Cattermole score with the combined X-ray assessment scores from the two radiologists. Boxes are median and quartiles, whiskers are 95% CI and dots are outliers. The poor quality group includes all the difficult-to-scan patients. *p < 0.001. The USCOM may also be used to measure cardiac output by interrogating the pulmonary valve via the left parasternal window, and this may provide a useful alternative in problematic cases. However, left lateral positioning, as is required to obtain a good quality scan in this area, may not be feasible because of ongoing surgery, and scanning may be made more difficult if the edge of the lung overlies the area being interrogated. When planning this study, we could not find a published method for scoring aortic elongation, dilation and unfolding on a standard posterior–anterior chest X-ray. Hence, we developed our own scoring system based on morphological changes of the aorta and cardiac enlargement, as both these may alter the anatomical position of the aortic valve and thus make it more difficult to interrogate the aortic valve from the supra-sternal notch. To reduce discrepancies between the two radiologists, their scores were combined. Thus, if a radiological feature was marginal, it was more likely to be identified by at least one of the radiologists. We decided to use the Cattermole scoring system for USCOM scan quality as it has been shown to provide a more sensitive assessment compared with its predecessor, the 6-point Fremantle score [6, 7]. © 2013 The Association of Anaesthetists of Great Britain and Ireland
Anaesthesia 2013, 68, 1156–1160
Pathological changes can be found in almost all parts of the human body as it ages, and the cardiovascular system is no exception with very typical and well-recognised changes. When a person ages, the ascending aorta increases in length by approximately 12%, and its diameter by 3%, for each decade of life. This phenomenon is described as aortic unfolding and dilatation, and the pathological mechanism is the loss of elastic tissue from the wall of the aorta [9, 10]. It is also commonly associated with calcification of the aorta, and is rare in patients less than 45 years of age. However, above 60 years, the incidence of aortic calcification is 15%, and this increases sharply to 50% by the age of 80 [11]. Left ventricular hypertrophy and left atrial enlargement are also common with ageing [12–14]. It is well known from echocardiographic studies that the patient’s position and the movement of their lungs, especially during mechanical ventilation, can affect the quality of ultrasound scans of the heart [15–17]. The supra-sternal window for ultrasonic scanning of the aortic valve relies on an unobstructed passage of the ultrasound beam, and this is provided by the fat and thymus-filled anterior mediastinum in front of the trachea and the ascending part of the aorta as it passes upwards (superiorly) and before it curves backwards (posteriorly) over the left main bronchus. However, any intervening air-filled space or bone will interrupt and attenuate the ultrasound signal. Occasionally, if the trachea in the thoracic inlet is prominent and lies anteriorly, it may prove difficult to insert the USCOM probe deep enough into the supra-sternal notch to get a good signal, and its beam may be partially obstructed by the sternum. However, we propose another, more subtle, problem associated with ageing. As the aorta unfolds and the heart enlarges, this alters the position of the aortic valve in the mediastinum, making it more difficult to interrogate because of interference by other structures such as the lungs. This phenomenon becomes very obvious when computerised tomography (CT) scans of the upper thorax are studied in relation to the passage of the ultrasound waves. Calcification of the ascending aorta may also play a role in attenuating the USCOM signal. The correlation between the ability to perform successful supra-sternal USCOM scans and chest 1159
Anaesthesia 2013, 68, 1156–1160
radiographical evidence of aortic unfolding and heart enlargement shown by this study provides an explanation as to why USCOM scanning can be difficult in some elderly anaesthetised patients. Therefore, we recommend reviewing the chest X-ray before deciding whether to use the USCOM device in older patients. Users of the USCOM should be familiar with the radiological features that may cause problems and be able to relate them to probe use. Further useful applied anatomical data can be obtained from a CT or magnetic resonance imaging scan of the upper thorax if available. In conclusion, supra-sternal use of the USCOM is affected by age-related changes in the aorta (unfolding, enlargement and calcification) and cardiac enlargement. These changes can be identified on the routine pre-operative chest X-ray. Routinely reviewing the X-ray pre-operatively may help guide the use of USCOM during surgery.
Acknowledgements We thank Mr. Hui Chun-hung Freddie (Research Assistant) from our department for his help with obtaining consent from Cantonese-speaking patients. No external funding and no competing interests declared.
Huang et al. | Correlation between USCOM and chest X-ray
4.
5.
6.
7.
8.
9.
10.
11.
12.
13. 14.
References 1. Hamilton MA, Cecconi M, Rhodes A. A systematic review and meta-analysis on the use of preemptive hemodynamic intervention to improve postoperative outcomes in moderate and high-risk surgical patients. Anesthesia and Analgesia 2011; 112: 1392–402. 2. Marik PE. Noninvasive cardiac output monitors: a state-of theart review. Journal of Cardiothoracic and Vascular Anesthesia 2013; 27: 121–34. 3. Huang L, Critchley LA. Study to determine the repeatability of supra-sternal Doppler (ultrasound cardiac output monitor) during general anaesthesia: effects of scan quality, flow vol-
1160
15.
16.
17.
ume, and increasing age. British Journal of Anaesthesia 2013 Jul 23; doi:10.1093/bja/aet254. Nidorf SM, Picard MH, Triulzi MO, et al. New perspectives in the assessment of cardiac chamber dimensions during development and adulthood. Journal of the American College of Cardiology 1992; 19: 983–8. Huang L, Critchley LA. Accuracy and precision of the USCOM: does a meta-analysis provide the answer? Anaesthesia 2013; 68: 431–2. Dey I, Sprivulis P. Emergency physicians can reliably assess emergency department patient cardiac output using the USCOM continuous wave Doppler cardiac output monitor. Emergency Medicine Australasia 2005; 17: 193–9. Cattermole GN, Leung P, Tang CO, Smith BE, Graham CA, Rainer TH. A new method to score the quality of USCOM scans. Hong Kong Journal of Emergency Medicine 2009; 16: 288. Critchley LA, Huang L. USCOM – Window to the circulation: utility of supra-sternal Doppler in an elderly anaesthetized patient for a robotic cystectomy. Journal of Clinical Monitoring and Computing 2013 Aug 2; doi:10.1007/s10877-0139499-5. O’Rourke M, Farnsworth A, O’Rourke J. Aortic dimensions and stiffness in normal adults. JACC Cardiovascular Imaging 2008; 1: 749–51. Redheuil A, Yu WC, Mousseaux E, et al. Age-related changes in aortic arch geometry: relationship with proximal aortic function and left ventricular mass and remodeling. Journal of the American College of Cardiology 2011; 58: 1262–70. Teale C, Romaniuk C, Mulley G. Calcification on chest radiographs: the association with age. Age and Ageing 1989; 18: 333–36. Savage DD, Garrison RJ, Kannel WB, et al. The spectrum of left ventricular hypertrophy in a general population sample: the Framingham Study. Circulation 1987; 75: I26–33. Levy D. Left ventricular hypertrophy: Epidemiological insights from the Framingham Heart Study. Drugs 1988; 35: 1–5. Gerdts E, Oikarinen L, Palmieri V, et al. Correlates of left atrial size in hypertensive patients with left ventricular hypertrophy: the Losartan Intervention For Endpoint Reduction in Hypertension (LIFE) Study. Hypertension 2002; 39: 739–43. Cook CH, Praba AC, Beery PR, Martin LC. Transthoracic echocardiography is not cost-effective in critically ill surgical patients. Journal of Trauma 2002; 52: 280–4. Font VE, Obarski TP, Klein AL, et al. Transesophageal echocardiography in the critical care unit. Cleveland Clinic Journal of Medicine 1991; 58: 315–22. Parker MM, Cunnion RE, Parrillo JE. Echocardiography and nuclear cardiac imaging in the critical care unit. Journal of the American Medical Association 1985; 254: 2935–9.
© 2013 The Association of Anaesthetists of Great Britain and Ireland
