a b s t r a c t

Background: Pulmonary embolism (PE) is a common emergency department (ED) diagnosis with a wide range in mortality rates. Methods to identify and risk-stratify PE, including measuring right ventricular strain (RVS) by echocardiography (echo), are essential in providing effective and efficient care. A limited echo examination aims to expedite and increase availability of RVS determination through assessment at the bedside by the ED provider.

Objective: The objective is to determine the level of agreement and test characteristics of right ventricular dilation (RVd), as a marker of RVS, on limited echo compared with consultative echo.

Methods: This is a retrospective cohort study of consecutive ED patients undergoing limited echo examinations for chest pain, dyspnea, or hypotension and a subsequent consultative echo within 72 hours. ? values and test characteristics were calculated to determine the level of agreement and accuracy between the limited echo examination and consultative echo for RVd and RVS.

Results: There were 411 focused examinations performed by 69 different providers over a 12-month period (median, 5 examinations per provider). The prevalence of RVS on limited echo examination was 6.2% (n = 25). The ? value for the level of agreement between limited and consultative echo for RVd was 0.44 (95% confidence intervals [CI], 0.27-0.61). The specificity of RVd on limited echo for RVS was 0.98 (95% CI, 0.96- 0.99) with 6 false-positive categorizations, whereas the sensitivity was 0.26 (95% CI, 0.16-0.37).

Conclusions: In this retrospective cohort study, limited echo demonstrated moderate agreement with consultative echo for RVd. Right ventricular dilation on limited echo was highly specific for RVS but had low sensitivity.

(C) 2014

Background

As many as 300,000 people in the United States die annually from pulmonary embolism (PE), and the diagnosis is frequently missed [1]. Echocardiography (echo) is the primary imaging method for determining whether right ventricular strain (RVS) is present in PE and plays a vital role in both the rapid identification of PE and risk stratification with implications for prognosis and therapy [2]. However, determination of RVS by consultative echo is inconsistently available in the emergency department (ED) and not feasible to perform on all patients presenting with symptoms suggestive of PE or with a PE diagnosis [3].

In a recent consensus statement by the American Society of Echocardiography and American College of Emergency Physicians, limited cardiac ultrasound performed by emergency physicians (EPs)

? Prior Presentations: None.

?? Funding Sources/Disclosures: None.

* Corresponding author. Tel.: +1 860 402 1276; fax: +1 203 785 4580.

E-mail addresses: [email protected] (R.A. Taylor), [email protected] (C.L. Moore).

is acknowledged as a “fundamental tool” [4]. Along with identifying pericardial effusion, determining global left ventricular function, and assessing volume status, identification of “right ventricular (RV) enlargement” is identified as important to “prioritize further testing, alter differential diagnosis assessments, and assist with Treatment decisions” for patients with possible PE [4]. While EP limited echo for 3 of the 4 elements (pericardial effusion, left ventricular function, and volume status) has been well described, the accuracy of EPs of various skill levels for determining the presence of RVS has not been systematically investigated.

The objective of this study was to determine the level of agreement and test characteristics of right ventricular dilation (RVd), a marker of RVS, identified by EP limited echo when compared with consultative echo.

Methods

Study design

This is a retrospective cohort of ED patients undergoing a limited EP echo with assessment for RVd and a subsequent consultative echo within 72 hours. The study protocol was approved, and the institutional review board waived the requirement for informed consent.

0735-6757/$ – see front matter (C) 2014 http://dx.doi.org/10.1016/j.ajem.2013.12.043

372 R.A. Taylor, C.L. Moore / American Journal of Emergency Medicine 32 (2014) 371–374

Study setting and population

Urban, academic ED with over 90,000 annual adult visits and a well-established emergency ultrasound program, including an emergency ultrasound fellowship. Identification of RVd (defined as a right ventricle as large or larger than the left ventricle) has been taught as part of the program since 2002 and is routinely recorded as present or absent when limited echo is performed. All patients in the Ultrasound imaging database between December 2006, and January, 2008 who had an adequate limited echo for nontraumatic symptoms of potential cardiac etiology (primarily chest pain, dyspnea, and/or hypotension) and a subsequent consultative echo within 72 hours were included. Limited echoes were performed using ultrasound machines (Philips HD11XE and Envisor HD, Phillips USA, Andover, MA) with phased array probes with images recorded on digital video disk. Patient identification, reason for examination, and limited echo findings were recorded on templated paper log sheets during the study period.

Methods and measurement

Data abstraction and record review

A reviewer blinded to consultative echo abstracted data from the ultrasound log and ED patient record. A second observer (blinded to limited echo findings) abstracted the results of consultative echo from the dictated report.

Measures

Right ventricular dilation was noted as present, absent, or indeterminate as recorded on the ultrasound log sheet. This is typically determined by qualitative assessment of the ratio of the size of the right ventricle to left ventricle (RV:LV), ideally in the apical 4 chamber view. At our institution, an RV:LV ratio greater than or equal to 1 is considered enlarged for limited echo. On consultative echo, RVS was considered present if any note was made of RVd or other markers of RV dysfunction (eg, RV hypokinesis, paradoxical septal motion).

Data analysis

Statistical analysis was performed using JMP (version 9; SAS Institute, Cary, NC) and VassarStats (http://faculty. vassar.edu/ lowry/VassarStats.html). Normality for continuous variables was evaluated by visual inspection of histograms, normal probability plots, and evaluation of numeric statistics. ? values with 95% confidence intervals (CI) were calculated to determine agreement for all providers and the subgroups of junior residents (PGY1-2), senior residents (PGY3-4), and fellowship-trained providers. Sensitivity, specificity, and positive and negative likelihood ratios of RVd for RVS were calculated with 95% CI using consultative echo as the reference standard.

Results

During the study period, 411 patients were identified who had both a limited echo and consultative echo within 72 hours. Of these, 4 patients were excluded due to inadequate Image quality on limited echo. Limited echoes were performed by 69 different providers (median no. of exams, 5). The prevalence of RVS was 6.2% on limited echo and 18% on consultative echo (Table 1).

The ? value for the level of agreement between limited and consultative echo for RVd was 0.44 (95% CI, 0.27-0.61; Table 2). Subgroup analysis by junior residents, senior residents, and fellowship-trained providers demonstrated ? values of 0.41 (95%

Table 1

Limited and consultative echo examination characteristics (N = 407)

Patient characteristics

Age, mean (+-SD) 62 +- 18

Male, n (%) 206 (50)

Provider skill level^a, n (%)

PGY-1 116 (29)

PGY-2 23 (6)

PGY-3 96 (24)

PGY-4 61 (15)

Fellow 62 (15)

Attending 43 (11)

Other^b 10 (2)

RVS, n (%) Limited echo Consultative echo

Present 25 (6) 73 (18) LVEF^c, n (%)

Normal 270 (66) 288 (71)

Moderate 66 (16) 58 (14)

Severe 66 (16) 46 (11)

Hyperdynamic 4 (1) 11 (2.7)

Inadequate 1 (0.2) 0 (0) Effusion^d, n (%)

None 342 (84) 359 (88)

Small 49 (12) 32 (7.9)

Moderate 9 (2) 8 (2)

Large 7 (2) 8 (2)

Abbreviations: PGY, postgraduate year; LVEF, left ventricular ejection fraction.

^a Only applies to limited echo.

^b Physician assistants, nurse practitioners, medical students.

^c Left ventricular ejection fraction visually estimated with normal, 50%-70%; moderate, 30%-50%; severe, b30%; hyperdynamic N 70%.

^d Effusion size visually estimated with small, b 1 cm; moderate, 1-2 cm; large, N 2 cm.

CI, 0.13-0.68), 0.54 (95% CI, 0.27-0.81), and 0.55 (95% CI, 0.11-

1.00), respectively.

The specificity and positive likelihood ratio (LR+) of RVd on limited echo for RVS on consultative echo was 98% (95% CI, 96%-99%) and 14 (95% CI, 6-35) with 6 false-positive categorizations. The sensitivity and negative likelihood ratio of limited echo for RVS was 26% (95% CI, 16%-37%) and 0.75 (95% CI, 0.65-0.86).

Discussion

Previous studies have demonstrated that determination of RVS by experts can be of diagnostic value in patients with possible PE and is of prognostic and therapeutic importance in patients diagnosed with PE [5,6]. To the best of our knowledge, this is the first study examining emergency providers of various skill levels performing limited echo examinations for the assessment of RVd as a marker of RVS.

Comparison of limited echo to consultative echo demonstrated moderate agreement (? = 0.44) for RVd. In examining the subgroups of junior residents, senior residents, and attending/fellows, Experience level did not significantly change the level of agreement. A prior study assessing cardiologist-interpreted echoes for RVd demonstrated excellent agreement among observers with ? values of 0.89 to 0.96 [7].

There are several possible explanations for the reduced level of agreement in our study. First, EPs in general may not be as skilled at identifying more subtle signs of RVS on limited echo, and our cutoff of a ratio of 1:1 or greater was chosen to maximize specificity at the expense of sensitivity and overall agreement. While lower ratios may be considered enlarged in the literature, we teach this approach for several reasons including (1) ease of learning; 2) alliance with previous binary measures (eg, “yes” or “no” for presence of effusion) for limited echo; and 3) to limit false-positive examinations that may result from measures of RVS that are more difficult to assess (eg, RV hypokinesis) and have lower levels of agreement, even among experts [7,8].

R.A. Taylor, C.L. Moore / American Journal of Emergency Medicine 32 (2014) 371–374 373

Table 2

Two-by-two tables comparing limited vs consultative echocardiography

In the cardiology study that demonstrated excellent agreement, images were acquired and then subsequently read by the 2 observers, eliminating any chance for disagreement because of image acquisition. In addition, in our study, the examinations were not performed contemporaneously, and RV function may have improved or worsened between the time of the initial limited echo and the consultative echo.

In patients with massive PE, approximately two-thirds of deaths occur within the first hour [9]. As a consequence, rapid identification of massive PE is of considerable importance. Computed tomography is often the definitive diagnostic test for PE but takes time to be performed and interpreted and removes the patient from the resuscitation area. As an alternative, previous literature has demonstrated that, in patients with hemodynamic instability, echo signs of RVS are invariably present when PE is the cause [10]. In addition, the presence of RVS is used to qualify hemodynamically stable PEs as “submassive,” which have higher

morbidity and mortality than “small” or “minor” PEs, and may benefit from more aggressive therapy [2]. In our study, while false positives did occur, the specificity and LR+ of RVd on limited echo for RVS was high (specificity, 98%; LR+, 14 x). An LR+ of

10 or greater is typically considered clinically useful, suggesting that when RVd is identified using limited echo, it should be considered important and may help to expedite further evalua- tion, treatment, or consultation.

In contrast, the sensitivity and LR- of RVd on limited echo for RVS by providers of various skill levels was poor (sensitivity, 26%; LR-, 0.75). Such a low sensitivity and LR- implies that a negative limited echo for RVd should not be used alone for ruling out RVS in suspected or diagnosed PE. Other markers of RVS could be incorporated into limited echo, including a lower ratio of RV:LV as indicating RVd and other signs of RVS such as RV hypokinesis and septal changes, but these must be balanced by the reduction in specificity for RVS.

374 R.A. Taylor, C.L. Moore / American Journal of Emergency Medicine 32 (2014) 371–374

Limitations

This study was limited by its retrospective approach. As a sample from a single institution with a well-established ultrasound program, these results may be limited in generalizability. As mentioned above, RV assessment by limited echo and consultative echo was not performed contemporaneously, and results may have been different if they were done more closely together.

Conclusion

Determination of RV dilation as marker of RVS using limited echo by EPs of various skill levels demonstrated moderate agreement with consultative echo. The presence of RVd on limited echo was highly specific and had a high LR+ for RVS, but the absence of RV dilation on limited echo does not rule out RVS on consultative echo.

References

1. Tapson VF. Acute pulmonary embolism. N Engl J Med 2008;358:1037-52.
2. Jaff MR, McMurtry MS, Archer SL, et al. Management of massive and submassive pulmonary embolism, iliofemoral deep vein thrombosis, and chronic thrombo- embolic pulmonary hypertension. Circulation 2011;123:1788-830.
3. Moore CL, Molina AA, Lin H. Ultrasonography in community emergency departments in the United States: access to ultrasonography performed by consultants and status of emergency physician-performed ultrasonography. Ann Emerg Med 2006;47:147-53.
4. Labovitz AJ, Noble VE, Bierig M, et al. Focused cardiac ultrasound in the emergent setting: a consensus statement of the American Society of Echocardiography and American College of Emergency Physicians. J Am Soc Echocardiogr 2010;23: 1225-30.
5. Kucher N, Rossi E, De Rosa M, Goldhaber SZ. prognostic role of echocardiography among patients with acute pulmonary embolism and a systolic arterial pressure greater than or equal to 90 mm Hg. Circulation 2005;112:U648-9.
6. Volpicelli G, Lamorte A, Tullio M, et al. Point-of-care multiorgan ultrasonography for the evaluation of undifferentiated hypotension in the emergency department. Intensive Care Med 2013;39:1290-8.
7. Kline JA, Steuerwald MT, Marchick MR, Hernandez-Nino J, Rose GA. Prospective evaluation of right ventricular function and functional status 6 months after acute submassive pulmonary embolism: frequency of persistent or subsequent elevation in estimated pulmonary artery pressure. Chest 2009;136:1202-10.
8. Chua S, Levine RA, Yosefy C, et al. Assessment of right ventricular function by real- time three-dimensional echocardiography improves accuracy and decreases Interobserver variability compared with conventional two-dimensional views. Eur J Echocardiogr 2009;10:619-24.
9. Gifford Jr RW, Groves LK. Limitations in the feasibility of pulmonary embolectomy. A clinicopathologic study of 101 cases of massive pulmonary embolism. Circulation 1969;39:523-30.
10. McConnell MV, Solomon SD, Rayan ME, Come PC, Goldhaber SZ, Lee RT. Regional right ventricular dysfunction detected by echocardiography in acute pulmonary embolism. Am J Cardiol 1996;78:469-73.