Diagnostic accuracy of multiorgan point-of-care ultrasound compared with pulmonary computed tomographic angiogram in critically ill patients with suspected pulmonary embolism.

BACKGROUND: Critically ill patients have a higher incidence of pulmonary embolism (PE) than non-critically ill patients, yet no diagnostic algorithm has been validated in this population, leading to the overuse of pulmonary artery computed tomographic angiogram (CTA). This study aimed to comparatively evaluate the diagnostic accuracy of point-of-care ultrasound (POCUS) combined with laboratory data versus CTA in predicting PE in critically ill patients.
METHODS: A prospective diagnostic accuracy study. Critically ill patients with suspected acute PE undergoing CTA were prospectively enrolled. Demographic and clinical data were collected from electronic medical records. Blood samples were collected, and the Wells and revised Geneva scores were calculated. Standardized multiorgan POCUS and CTA were performed. The discriminatory power of multiorgan POCUS combined with biochemical markers was tested using ROC curves, and multivariate analysis was performed.
RESULTS: A total of 88 patients were included, and 37 (42%) had PE. Multivariate analysis showed a relative risk (RR) of PE of 2.79 (95% CI, 1.61-4.84) for the presence of right ventricular (RV) dysfunction, of 2.54 (95% CI, 0.89-7.20) for D-dimer levels >1000 ng/mL, and of 1.69 (95% CI, 1.12-2.63) for the absence of an alternative diagnosis to PE on lung POCUS or chest radiograph. The combination with the highest diagnostic accuracy for PE included the following variables: 1- POCUS transthoracic echocardiography with evidence of RV dysfunction; 2- lung POCUS or chest radiograph without an alternative diagnosis to PE; and 3- plasma D-dimer levels >1000 ng/mL. Combining these three findings resulted in an area under the curve of 0.85 (95% CI, 0.77-0.94), with 50% sensitivity and 96% specificity.
CONCLUSIONS: Multiorgan POCUS combined with laboratory data has acceptable diagnostic accuracy for PE compared with CTA. The combined use of these methods might reduce CTA overuse in critically ill patients.

Introduction

Acute pulmonary embolism (PE) is the sudden occlusion of the pulmonary artery or its branches [1], and patients admitted to the intensive care unit (ICU) are at increased risk of PE [2, 3]. Diagnostic algorithms and prediction scores, such as the Wells and revised Geneva scores, are available to guide the diagnostic approach to PE in outpatient settings and are occasionally used in the ICU [4]. However, critical illness makes it difficult to diagnose PE [5], and these prediction scores remain unacceptably inaccurate in critically ill patients [6].

The pulmonary computed tomographic angiogram (CTA) is the gold standard for diagnosing PE [7] and has been widely used in critically ill patients. However, the potential overuse of CTA unnecessarily exposes patients to ionizing radiation, iodinated contrast media, and transfer risks. In addition, many patients with PE have severe hypoxemia and hemodynamic instability that precludes CTA use [8], requiring alternative methods for diagnosis.

The use of point-of-care ultrasound (POCUS) in the ICU has emerged as an excellent diagnostic tool [9-11]. The investigation of PE with multiorgan ultrasound (US), involving cardiac, pulmonary, and lower limb venous scans, has shown promising results in emergency departments [12, 13], suggesting its potential role in PE detection in critically ill patients.

PE is associated with high mortality and has a nonspecific clinical presentation. Therefore, alternative diagnostic strategies to help with timely decisions are important in critically ill patients. The purpose of this study was to comparatively evaluate the diagnostic accuracy of multiorgan POCUS combined with laboratory data versus CTA in predicting PE in critically ill patients.

Materials and methods

This prospective diagnostic accuracy study is reported according to the Standards for Reporting of Diagnostic Accuracy Studies (STARD) guidelines [14] and involved comparing the results of the index test (multiorgan POCUS combined with laboratory data) and reference standard (CTA), defined as the best available method for establishing the presence or absence of the condition of interest (PE) (S1 File).

All study procedures followed the tenets of the Declaration of Helsinki of 1975 and were performed in accordance with the guidelines of the Research Ethics Committee of the Hospital de Clínicas de Porto Alegre, Brazil, for research involving human subjects. Our study was reviewed and approved by the Research Ethics Committee of the Hospital de Clínicas de Porto Alegre, Brazil, in September 2018 (number 2018–0282). Written informed consent was obtained from each study participant or legal representative, and the patient was only included in the study after this consent.

The patients were enrolled from September 2018 to February 2020. Critically ill adult patients with suspected acute PE for whom the attending physician ordered pulmonary artery CTA were included.

Demographic and clinical data were collected from electronic medical records. Blood samples were collected for D-dimer, troponin, and N-terminal pro-B-type natriuretic peptide (NT-proBNP) level measurements at study entry. The Wells score and the revised Geneva score were calculated by the same researcher for all patients. The study protocol included standardized multiorgan POCUS and CTA. Multiorgan POCUS involved performing lower limb deep vein US, lung US, and transthoracic echocardiography (TTE).

POCUS examinations were performed independently by two critical care fellows with advanced training in POCUS who had undergone three months of training in image acquisition specifically for the study protocol. All images were saved and reviewed by a senior physician if necessary. The researchers performing the POCUS examinations were blinded to the CTA results. POCUS examinations and CTA were performed no more than 24 hours apart.

Lung US was performed with a convex probe. An A- or B-pattern was defined on the anterior chest. The presence or absence of posterolateral alveolar pleural syndrome (PLAPS) was determined on the posterolateral chest [15]. Images of the anterosuperior and anteroinferior quadrants were acquired on the midclavicular line at the second and fifth intercostal spaces, respectively. Images of the posterolateral chest were obtained between the mid-and posterior axillary lines, with the inferior posterolateral quadrant located at the thoracoabdominal transition. The presence of one or more of the following lung abnormalities was considered an alternative diagnosis to PE on lung US: 1- the absence of pleural slip suggesting the presence of pneumothorax; 2- the presence of a hypoechoic pleural-based lesion suggesting consolidation or pulmonary atelectasis; 3- the presence of three or more B lines in an intercostal space in non-dependent lung areas suggesting alveolar-interstitial edema; 4- the presence of homogeneous anechoic area in a dependent lung area suggesting pleural effusion [15, 16].

Lower limb venous US was performed with a linear probe to scan the femoral and popliteal veins bilaterally. Three points were examined in each extremity, two in the femoral vein and one in the popliteal vein. Deep vein thrombosis (DVT) was defined as the absence of complete compression of the vessel wall with slight probe compression, with or without visualizing hyperechogenic areas within the vessel.

TTE was performed with a cardiac probe. The right ventricle (RV) to left ventricle (LV) diameter ratio (RV/LV ratio) and the tricuspid annular plane systolic excursion (TAPSE) in the apical four-chamber view were assessed. An RV/LV ratio ≥ 1.0 or a TAPSE < 1.7 cm were considered abnormal [1].

The following parameters were evaluated: epidemiological and clinical data, electrocardiograms, chest radiographs, and laboratory tests, including ultrasensitive troponin I, D-dimers, and NT-proBNP. D-dimer levels were quantified by latex agglutination assay and considered elevated if >400 ng/mL (laboratory reference value). Ultrasensitive troponin I levels were determined by chemiluminescent microparticle immunoassay, and levels >52 pg/mL suggest myocardial injury (laboratory reference value). NT-proBNP levels were determined by electrochemiluminescence and considered elevated if >125 pg/mL (laboratory reference value).

Chest radiograph and electrocardiogram findings were classified as normal or abnormal. Chest radiograph abnormalities included consolidation, bilateral diffuse infiltrates, pleural effusion, or a combination of those. Electrocardiogram abnormalities included sinus bradycardia, atrioventricular block, sinus tachycardia, atrial fibrillation, atrial flutter, and S1Q3T3 pattern (a finding suggestive of RV overload).

All patients underwent CTA for the diagnosis of PE. CTA images were obtained with a 16-channel CT scanner in helical scan mode. The images were considered positive or negative for PE according to the report of two independent radiologists, one of them with expertise in thoracic radiology. In case of disagreement, a third radiologist analyzed the images. In patients with PE, thrombus location was defined according to the caliber of the affected vessel and classified as follows: main artery, lobar, segmental, or subsegmental. Scans were considered negative in the presence of adequate opacification of the pulmonary artery without filling defects [17].

The sample size was calculated using PEPI software (version 11.65, 2016). The calculation was based on a previously reported maximum sensitivity of multiorgan POCUS (lower limbs, lung, and heart) of 90% for PE detection [13] and incidence of 30% of PE on CTA in critically ill patients [6]. To detect a 10% difference in the sensitivity to detect PE between the combined findings of multiorgan POCUS and CTA, with an accuracy of 12% and an alpha error of 0.05, a sample size of 85 patients was necessary.

Categorical variables were expressed as percentages. Quantitative data were expressed as mean and standard deviation if normally distributed or as median and interquartile range if not normally distributed. Interobserver agreement of CTA readings was assessed by kappa (κ) statistics, with κ values of 0.4–0.6 indicating moderate agreement, 0.61–0.8 high agreement, and 0.81–1.0 very high agreement. As appropriate, groups were compared by Student’s t-test, Mann-Whitney U test, or chi-square test. Variables with p <0.1 in univariate analysis and those with clinical relevance were included in multivariate analysis.

The diagnostic accuracy of POCUS and laboratory tests was analyzed using receiver operating characteristic (ROC) curves when the variables were continuous. ROC curves were also constructed by combining categorical variables after logistic regression. A p <0.05 was considered statistically significant. All statistical analyses were performed using SPSS software, version 21.0 (Chicago, IL, USA).

Results

Eighty-eight critically ill patients were included in the study. Two patients had a history of previous PE or DVT, but none of them were being treated for PE or DVT in the last six months. The mean patient age was 58 ± 15 years. Most patients were women (53%), and the mean Simplified Acute Physiology Score 3 (SAPS 3) was 60 ± 15. The main reason for ICU admission was acute respiratory failure (63%), followed by cardiopulmonary arrest (17%) and circulatory shock (12%). ICU mortality and 30-day hospital mortality were 36% and 43%, respectively.

Of 88 patients, 37 (42%) had PE detected on pulmonary CTA examination. Of these, 12 patients (32%) had PE in the main artery, three (8%) in the lobar branch, 18 (49%) in the segmental branch, and four patients (11%) in the subsegmental branch. The agreement in PE diagnosis between the two radiologists was very high (κ = 0.89).

Table 1 shows the main characteristics of the 88 patients. Patients were divided into two groups according to the presence or absence of PE. Patients without PE had a higher SAPS 3, more commonly developed chronic kidney disease and sepsis, and more often required mechanical ventilation. The Wells score was significantly higher in patients with PE than in those without PE (4.2 ± 2.5 vs. 2.9 ± 1.9; p  = 0.012), but the Geneva score was not. The groups with and without PE did not differ significantly in mortality (37% vs. 47%; p  = 0.5), length of ICU stay (5 [3-16] vs. 9 [5-20] days; p  = 0.28), or length of hospital stay (21 [12-39] vs. 18 [13-44] days; p  = 0.68).

Table 1

Characteristics of the 88 patients included in the study.

Characteristics	All patients (n = 88)	With PE	Without PE	P
		(n = 37)	(n = 51)
Demographics
Age (years)	58 ± 15	57 ± 15	59 ± 15	0.73
Male gender (n,%)	41 (47%)	15 (40%)	26 (50%)	0.45
SAPS 3 score	60 ± 15	57 ± 16	62 ± 14	0.047
Comorbidities
Hypertension (n,%)	45 (51%)	19 (51%)	26 (50%)	1
Diabetes (n,%)	24 (27%)	7 (19%)	17 (33%)	0.2
Chronic kidney disease (n,%)	10 (11%)	1 (2%)	9 (18%)	0.04
Malignancy (n,%)	22 (25%)	9 (24%)	13 (25%)	1
Signs and symptoms
Tachycardia1 (n,%)	44 (50%)	20 (54%)	24 (47%)	0.66
Tachypnea1 (n,%)	77 (87%)	33 (89%)	44 (86%)	0.75
Chest pain (n,%)	21 (24%)	14 (38%)	7 (14%)	0.018
Severity of illness
Presence of sepsis (n,%)	40 (45%)	10 (27%)	30 (58%)	0.006
Use of vasopressors (n,%)	30 (34%)	10 (32%)	20 (39%)	0.33
Need for MV (n,%)	55 (62%)	18 (48%)	37 (72%)	0.03
Need for CVC (n,%)	57 (65%)	20 (54%)	37 (72%)	0.1
Immobilization2 (n,%)	32 (36%)	14 (38%)	18 (35%)	0.9
Prediction rules
Wells score	3.4 ± 2.3	4.2 ± 2.5	2.9 ± 1.9	0.012
Wells score >4 (n,%)	41 (46%)	21 (57%)	20 (31%)	0.15
Revised Geneva score (n)	5.5 ± 3.1	6.2 ± 3.4	5.1 ± 2.8	0.16
Revised Geneva score >7 (n,%)	30 (34%)	15 (40%)	15 (29%)	0.39

CVC: central venous catheter; MV: mechanical ventilation; PE: pulmonary embolism; SAPS 3: Simplified Acute Physiology 3.

1Tachycardia was defined as heart rate >100 bpm and tachypnea as respiratory frequency >20 rpm.

2Immobilization was defined as bed restriction >48 hours.

The mean time interval between multiorgan POCUS examinations and CTA was 10 ± 7.5 hours. Of the 37 patients with positive findings for PE on CTA, 22 (60%) had RV dysfunction on POCUS TTE, with 63% sensitivity and 85% specificity. All patients were considered to have acute RV dysfunction. Twenty-eight patients (76%) had no alternative diagnosis to PE on lung POCUS, with 78% sensitivity and 39% specificity. Seven patients (19%) with positive findings for PE on CTA had DVT on lower limb deep vein POCUS, with 19% sensitivity and 94% specificity. The three POCUS findings (RV dysfunction, absence of an alternative diagnosis to PE on lung POCUS, and presence of lower limb DVT) occurred simultaneously in six patients (16%).

Of 88 patients, 10 (11%) had evidence of lower limb DVT on POCUS, with no significant difference between the groups with and without PE. Patients without PE tended to have an alternative pulmonary diagnosis based on lung POCUS more frequently than those with PE (35% vs. 22%; p  = 0.06). Regarding POCUS TTE, patients with PE had a higher RV/LV ratio than those without PE, both in absolute values (0.98 ± 0.2 vs. 0.82 ± 0.17; p  = 0.01) and when compared to the cut-off value of 1 (38% vs. 8%; p  = 0.001). Patients with PE also had lower TAPSE values (1.7 ± 0.5 vs. 2.0 ± 0.4 cm; p  = 0.027), indicating more significant RV dysfunction in these patients. Comparing the groups by TAPSE cut-off value above or below 1.7 cm, patients with PE had a TAPSE <1.7 cm (47% vs. 10%; p  =  0.001) more frequently than those without PE (Table 2). When evaluating the performance of each POCUS examination separately, TTE with the presence of RV dysfunction showed the highest diagnostic accuracy.

Table 2

Results of biochemical tests, imaging studies and point-of-care findings.

Characteristics	All patients	With PE	Without PE	p
	(n = 88)	(n = 37)	(n = 51)
Biochemical and imaging results
P/F ratio	187 (118–245)	152 (117–244)	180 (129–220)	0.73
D-dimers (ng/mL)	3054 (1672–4950)	4600 (2775–5001)	2205 (1220–3640)	0.001
N-terminal pro-BNP (pg/mL)	1147 (269–3858)	1420 (253–2752)	1013 (299–4114)	0.69
Ultrasensitive troponin I (pg/mL)	51 (27–164)	96 (28–297)	43 (24–90)	0.02
Abnormal electrocardiogram (n,%)	50 (57%)	24 (65%)	26 (51%)	0.14
Abnormal chest X-ray (n,%)	65 (74%)	21 (57%)	44 (86%)	0.02
Point-of-care ultrasound findings
Alternative diagnosis to PE on lung ultrasound (n,%)	27 (31%)	8 (22%)	19 (37%)	0.08
Presence of DVT in lower limbs (n,%)	10 (11%)	7 (19%)	3 (6%)	0.08
Qualitative RV dysfunction (n,%)	9 (10%)	7 (19%)	2 (4%)	0.034
RV/LV ratio	0.9 ± 0.2	1 ± 0.2	0.8 ± 0.2	0.01
RV/LV ratio >1 (n,%)	19 (20%)	14 (38%)	4 (8%)	0.001
TAPSE (cm)	1.9 ± 0.5	1.7 ± 0.5	2.0 ± 0.4	0.027
TAPSE <1.7cm (n,%)	23 (26%)	18 (47%)	5 (10%)	0.001

DVT: deep vein thrombosis; LV: left ventricle; NT-proBNP: N-terminal pro-BNP (brain natriuretic peptide); PE: pulmonary embolism; P/F ratio: ratio between partial arterial oxygen pressure and inspired oxygen fraction; RV: right ventricle; TAPSE: tricuspid annular plane systolic excursion.

Patients with PE had higher D-dimer levels than patients without PE (4600 [2775-5001] vs. 2205 [1220-3640] ng/mL; p  = 0.001). All patients with D-dimer levels below the cut-off value of 400 ng/mL did not have PE on CTA.

There was no significant difference between the groups with and without PE concerning NT-proBNP levels. Patients with PE had higher, but not clinically relevant, ultrasensitive troponin I levels than patients without PE (96 [28-297] vs. 43 [24-90] pg/mL, p  = 0.02).

The single laboratory parameter that improved accuracy when combined with POCUS was plasma D-dimer. Ultrasensitive troponin I and NT-proBNP did not perform well, whether alone or combined with other parameters (S2 File).

Multivariate analysis showed that the presence of RV dysfunction, assessed by the presence of RV/LV ratio ≥1 or TAPSE <1.7 cm or qualitative RV dysfunction, had a relative risk (RR) of 2.79 (95% CI, 1.61–4.84) for the presence of PE. The absence of an alternative pulmonary diagnosis, based on lung POCUS or normal chest radiograph findings, had an RR of 1.69 (95% CI, 1.12–2.63) for the presence of PE, and D-dimer levels >1000 ng/mL had an RR of 2.54 (95% CI, 0.89–7.20) for the presence of PE (Table 3).

Table 3

Multivariate analysis of predictors of pulmonary embolism by clinical variables and multiorgan point-of-care ultrasound.

Variables	Odds ratio	95% CI
Presence of RV dysfunction*	2.79	1.61–4.84
No alternative pulmonary diagnosis**	1.69	1.12–2.63
D-dimers >1000 ng/mL	2.54	0.89–7.20

LV: left ventricle; RV: right ventricle; TAPSE: tricuspid annular plane systolic excursion.

95% CI: 95% confidence interval.

*RV/LV ratio >1 or TAPSE <1.7 cm or qualitative RV dysfunction.

**By lung point-of-care ultrasound or chest X-ray.

Combining POCUS with laboratory data provided optimal diagnostic accuracy for PE. The combination with the highest accuracy included the following three variables: 1– POCUS TTE with evidence of RV dysfunction (TAPSE <1.7 cm or RV/LV ratio ≥1 or qualitative RV dysfunction); 2– lung POCUS or chest radiograph without an alternative diagnosis to PE; and 3– plasma D-dimer levels >1000 ng/mL. Together, this combination had an area under the ROC curve (AUC) of 0.85 (95% CI, 0.77–0.94), with 50% sensitivity and 96% specificity (Fig 1).

Fig 1

ROC curves demonstrating the performance of the combination of the following three variables: 1- transthoracic echocardiography with right ventricular dysfunction; 2- lung abnormalities on lung ultrasound or chest radiograph; and 3- D-dimer levels > 1000 ng/mL in predicting pulmonary embolism.

Twenty patients (23%) presented the three positive findings concomitantly (1- presence of RV dysfunction in TTE; 2- no alternative pulmonary diagnosis; and 3- D-dimer levels >1000 ng/mL), and 18 of them had a positive PE diagnosis on CTA, with sensitivity, specificity, positive and negative predictive values of 50%, 96%, 90%, and 73%, respectively. Conversely, seven patients (8%) had the absence of the three findings concomitantly, all of them without PE on pulmonary CTA. The diagnostic accuracy of each ultrasound and combined ultrasounds for the diagnosis of PE is shown in Table 4.

Table 4

Diagnostic accuracy of single-organ ultrasound and combined ultrasound scans for the diagnosis of pulmonary embolism.

	Sens % (95% IC)	Spec % (95% IC)	PPV % (95% IC)	NPV (95% IC)	+LR (95% IC)	-LR (95% IC)
Cardiac ultrasound	0.63 (0.45–0.79)	0.85 (0.72–0.94)	0.76 (0.56–0.90)	0.76 (0.62–0.87)	4.31 (2.08–2.95)	0.43 (0.28–0.68)
Lung ultrasound	0.78 (0.61–0.90)	0.39 (0.25–0.54)	0.48 (0.35–0.62)	0.70 (0.50–0.86)	1.27 (0.96–1.69)	0.57 (0.28–1.16)
Limb ultrasound1	0.19 (0.08–0.36)	0.94 (0.83–0.99)	0.70 (0.35–0.93)	0.61 (0.49–0.72)	3.18 (0.88–11.4)	0.86 (0.72–1.02)
Multiorgan ultrasound2	0.16 (0.06–0.32)	1,00 (0.93–1.00)	1.00 (0.54–1.00)	0.62 (0.51–0.73)	2.28 (0.70–10.0)3	0.84 (0.73–0.97)
Cardiac and lung ultrasound plus D-dimers >1000ng/mL	0.50 (0.33–0.67)	0.96 (0.86–1.00)	0.90 (0.68–0.99)	0.73 (0.60–0.83)	12.5 (3.09–50.5)	0.52 (0.37–0.73)

Sens: sensitivity; Spec: specificity; PPV: positive predictive value; NPV: negative predictive value; +LR: positive likelihood ratio; -LR: negative likelihood ratio

1 Lower limb venous ultrasound

2 Evidence of RV dysfunction and absence of an alternative pulmonary diagnosis and presence of deep venous thrombosis of the lower limbs on point-of-care ultrasound (POCUS)

3 Calculation performed with specificity at the lower limit of 95% CI

Discussion

Our findings showed that the combination of 1- POCUS TTE with evidence of RV dysfunction (TAPSE <1.7 cm or RV/LV ratio ≥1), 2- the absence of an alternative diagnosis to PE on lung POCUS or chest radiograph, and 3- D-dimer levels >1000 ng/mL has good diagnostic accuracy in predicting PE compared with CTA.

In the present study, the incidence of PE in patients undergoing pulmonary CTA for suspected PE was 42%. In non-critically ill hospitalized patients, it is estimated that 30% of CTA scans will be positive in patients with a high probability of PE [18].

Twenty patients had the combination of RV dysfunction, D-dimers >1000 ng/mL, and the absence of an alternative pulmonary diagnosis, and 18 (90%) of them had PE on pulmonary CTA. Furthermore, seven patients with the absence of RV dysfunction in TTE, D-dimers <1000 ng/mL, and the presence of an alternative pulmonary diagnosis had negative CTA for PE. Based on our findings, 25 patients (28%) would avoid undergoing CTA with a more comprehensive pre-test probability assessment, integrating POCUS findings and laboratory data.

TTE with signs of RV dysfunction was the component of multiorgan POCUS with the highest diagnostic accuracy. Lower limb US had the lowest impact on improving the accuracy of our results. It is estimated that 30% of patients with PE have lower limb DVT [19]. In our sample, however, 19% of patients with PE had DVT. The possibility that thrombi may originate from the upper circulation in critically ill patients, mainly associated with the presence of central venous catheters, has already been suggested in a previous study [20]. Still, this association was not found in our population of patients. Thrombi may be lodged in the iliac veins or the inferior vena cava and have already completely detached and carried into the pulmonary circulation. Our results are consistent with those of a recent study involving critically ill patients that demonstrated a lower incidence of lower limb DVT in patients with PE [21].

D-dimer testing has been used only occasionally in critically ill patients because of its low specificity in this population [22]. However, our results suggest that D-dimer levels, when combined with multiorgan POCUS, improve diagnostic accuracy for PE and may be helpful, especially if the cut-off value is adjusted for severity of disease and age [23]. In our sample, a D-dimer cut-off value >1000 ng/mL maintained excellent sensitivity, with a high negative predictive value. All patients with D-dimer levels <400 ng/mL did not have PE on pulmonary CTA, reinforcing the idea that D-dimers are useful for ruling out PE, even in critically ill patients.

Nazerian et al. [13] investigated patients with suspected PE in the emergency department and showed that multiorgan POCUS has higher diagnostic accuracy than single-organ POCUS. Our results demonstrate that the combination of laboratory data with multiorgan POCUS further improves diagnostic accuracy for PE in critically ill patients. The presence of RV dysfunction in TTE, absence of pulmonary differential diagnosis on lung POCUS or chest radiograph, and elevated D-dimer levels above 1000 ng/mL provided the best combination for the diagnosis of PE compared with CTA. Conversely, variables classically associated with PE, such as age [23], troponin and NT-proBNP [24], and the revised Geneva score [5], did not perform well in our population, in agreement with our previous retrospective results [6]. In this prospective study, combining the Wells and Geneva prediction scores with multiorgan scans did not increase diagnostic accuracy.

In a meta-analysis published in 2021, Falster et al. [16] reported that the use of POCUS in patients with suspected PE has findings with high specificity, and its use should be encouraged to select which patients should be referred for CTA. In our study protocol, we did not assess some of the findings that have high specificity for PE, such as the McConnell’s sign and the presence of pulmonary infarctions. We understand that these findings suggest PE more strongly than just RV dysfunction or absence of an alternative pulmonary diagnosis, but they are less frequent findings and require greater operator expertise for their identification. In addition, some protocols for pulmonary and cardiac assessment require specific positioning of the patient in bed, which is difficult to perform in critically ill patients with high severity.

This study has limitations. First, 11 potentially eligible patients were not included in the study due to logistic reasons. The maximum time interval of 24 hours allowed between POCUS and CTA was exceeded. Still, there is no reason to believe that these patients would differ from the included ones. Second, we defined the absence of PE on pulmonary CTA as a negative result, which is currently the preferred imaging study for diagnosing PE. However, we did not follow up the patients who survived for three months, which is considered the gold standard to exclude PE, nor did we perform autopsies on the patients who died. Third, a formal exclusion of patients with previous RV dysfunction was not performed. However, no patients were suspected to have previous pulmonary hypertension based on previous echocardiography data and clinical data. Fourth, this is a single-center study.

Conclusion

In conclusion, multiorgan POCUS combined with laboratory data has acceptable diagnostic accuracy for PE compared with CTA and might be helpful in the ICU setting, providing an alternative method when performing a CTA is too risky and reducing unnecessary tests when the probability of disease is too high. The possibility of a 25% reduction in the need for CTA in critically ill patients with suspected PE is encouraging. Considering that multiorgan POCUS assessment of critically ill patients with suspected PE is an easy-to-perform, low-cost and low-risk technique, this diagnostic method combined with laboratory data could be easily implemented in clinical practice.

STARD diagram.

(PDF)

Click here for additional data file.

ROC curves demonstrate the performance of biochemical parameters in predicting pulmonary embolism.

A: Troponin. B: NT-pro-BNP.

(PDF)

Click here for additional data file.

Database with information on the 88 patients included in the final analysis.

(XLSX)

Click here for additional data file.

8 Aug 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Patricia Rezende do Prado

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

Reviewer #3: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: No

**********

5. Review Comments to the Author

Reviewer #1: • Prediction scores as Wells and Geneva for pretest probability of PE validated in outpatient settings are occasionally used in the ICU. However, none of these clinical prediction rules have been validated in the ICU population. Therefore, more researches are needed to identify features that help to accurately identify patients with PE in this setting, and to develop a suitable prediction rule for critically ill patients with suspected PE. I think that in this study I do not see the relevance of considering these scores to analyze the usefulness of the multiorgan pocus. Please, comment about it.

• In table 2, define and explain better what it means “Alternative diagnosis on lung ultrasound”.

• Lung US performed by authors was based on diagnosis of acute respiratory failure method published on 2008. I would like to suggest that you make a more current and didactic classification shown in systematic review by Falster C et al

Reviewer #2: This is an interesting study that looked into utilization of POCUS with clinical and lab values to diagnosed PE in comparison with CTPA.

"A noninvasive diagnostic strategy to help with timely decisions is important in critically ill patients" I think this statement should be modified as CTPA is already noninvasive, I would consider changing it.

Definition of altered lung result on the basis of 3 or more altered fields may miss focal lung abnormalities like consolidation or atelectasis and this definition may under-diagnose alternative diagnoses to PE.

The authors did not specify if RV dysfunction was acute or chronic as presence of previous RV dysfunction may confound the findings significantly. Also, I believe all patients with previous RV dysfunction should have been excluded from the study.

There was no mention of patients with history of previous PE or being treated for PE as these also may confound the findings.

Were the images acquired by the fellows saved and reviewed by more experienced physician to concur with accuracy of ultrasound image acquisition given that ultrasound is operator dependent?

The authors mentioned specificity of 100% in the abstract and then mentioned specificity of 96% in the results section, this should be corrected and clarified.

Given sensitivity of 50%, I think it is really difficult to state in the conclusion the diagnostic accuracy in high at this point. I would be more cautious with such a statement as the data presented does not fully support this conclusion.

Reviewer #3: I had a very difficult time determining what proportion of PEs were sub-segmental in nature. The paper also requires significant language editing and the abstract is lacking key information i.e. the sample size included in the study.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: Yes.

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

29 Aug 2022

Thank you for reviewing the above referenced manuscript. The comments were very useful and the manuscript was amended accordingly. All changes made are highlighted in yellow in the revised version of the manuscript. Please see below the answers for all queries.

Do not hesitate to contact us if you require any further information.

Journal Requirements:

Comment 1: Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming.

Answer 1: We thoroughly revised instruction to authors and made all arrangements to the files and figures to meet PLOS ONE's style requirements.

Comment 2: We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match. When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

Answer 2: As recommended, we provide the correct grant numbers for the awards in the ‘Funding Information’ section.

Reviewer #1

Thank you for your kind review. We incorporated your suggestions in the manuscript as follows:

Comment 1: Prediction scores as Wells and Geneva for pretest probability of PE validated in outpatient settings are occasionally used in the ICU. However, none of these clinical prediction rules have been validated in the ICU population. Therefore, more researches are needed to identify features that help to accurately identify patients with PE in this setting, and to develop a suitable prediction rule for critically ill patients with suspected PE. I think that in this study I do not see the relevance of considering these scores to analyze the usefulness of the multiorgan pocus. Please, comment about it.

Answer 1: This is an interesting point and we completely agree with the reviewer. The Wells and Geneva scores are used in an intensive care unit, but they are not validated in critically ill patients. There is no robust literature on the diagnostic investigation of PE in this population of patients. A previous study by our group (Ref 6, Wells and Geneva Scores Are Not Reliable Predictors of Pulmonary Embolism in Critically Ill Patients: A Retrospective Study) demonstrated that these scores were not good risk predictors for acute PE in critically ill patients, but the study was retrospective and we would like to confirm these findings in a prospective study. Then, we initially designed the study to include these scores in the analysis. However, our prospective results also showed low accuracy of these scores in critically ill patients with suspected pulmonary embolism (PE). Combining Wells and Genebra scores to mutiorgan ultrasound did not result I increased accuracy beyond the accuracy of multiorgan ultrasound. Therefore, we did not recommend to combine these scores with multiorgan ultrasound (please refer to Table 3 and Figure 1). We clarify this point in Discussion section (page 17, paragraph 1), as follows:

“Conversely, variables classically associated with PE, such as troponin and NT-proBNP [24], age [23], and the revised Geneva score [5], did not perform well in our population, in agreement with our previous retrospective results [6]. In this prospective study, combining the Wells and Geneva prediction scores with multiorgan scans did not increase diagnostic accuracy.”

Comment 2: In Table 2, define and explain better what it means “Alternative diagnosis on lung ultrasound”.

Answer 2: To better clarify what “Alternative diagnosis on lung ultrasound” means, we included a statement in Methods section (page 6, paragraph 2), as follows:

“The presence of one or more of the following lung abnormalities was considered an alternative diagnosis to PE on lung US: 1- the absence of pleural slip suggesting the presence of pneumothorax; 2- the presence of a hypoechoic pleural-based lesion suggesting consolidation or pulmonary atelectasis; 3- the presence of three or more B lines in an intercostal space in non-dependent lung areas suggesting alveolar-interstitial edema; 4- the presence of homogeneous anechoic area in a dependent lung area suggesting pleural effusion [15,16].”

Comment 3: Lung US performed by authors was based on diagnosis of acute respiratory failure method published on 2008. I would like to suggest that you make a more current and didactic classification shown in systematic review by Falster C et al.

Answer 3: Thank you for this important note. We used the paper by Falster et al. (Ref 16, please refer to References) to define the item "differential diagnosis" in pulmonary ultrasound. This point was not clearly stated in our article, but it is now better clarified as presented in the answer above (comment and answer 2) and in Methods section (page 6, paragraph 2). Unfortunately, some relevant points from the meta-analysis by Falster et al. were not incorporated into our POCUS analysis, for instance, the identification of pulmonary infarctions and McConnell’s sign. Despite the high specificity of these two findings, their recognition would require long training and senior expertise, but we chose a more easily reproducible approach.

Reviewer #2

This is an interesting study that looked into utilization of POCUS with clinical and lab values to diagnosed PE in comparison with CTPA.

Comment 1: "A noninvasive diagnostic strategy to help with timely decisions is important in critically ill patients" I think this statement should be modified as CTPA is already noninvasive, I would consider changing it.

Answer 1: We agree with the reviewer and changed the sentence as follows (page 4, paragraph 4):

“Therefore, alternative diagnostic strategies to help with timely decisions are important in critically ill patients.”

Comment 2: Definition of altered lung result on the basis of 3 or more altered fields may miss focal lung abnormalities like consolidation or atelectasis and this definition may under-diagnose alternative diagnoses to PE.

Answer 2: We agree with the reviewer. In all patients, we examined eight lung fields and collected two different variables in pulmonary ultrasound: 1- the presence of abnormal pulmonary findings; and 2- the presence of alternative pulmonary diagnosis (as better clarified in Methods section - page 6, paragraph 2). The absence of an alternative diagnosis on lung ultrasound (all lung fields considered) showed better performance and provided a more accurate information than the results of abnormal lung findings. Then we presented data on pulmonary ultrasound in a more objective way, referring only to the presence of alternative pulmonary diagnosis, which considered all lung fields examined.

The presence or absence of differential diagnosis on pulmonary ultrasound showed better performance in our analysis, being included in our final model of better diagnostic accuracy (please refer to Table 2). The definitions of alternative pulmonary diagnosis used in the article are the following: 1 - absence of pleural slip suggesting the presence of pneumothorax; 2 - presence of a hypoechoic pleural-based lesion suggesting consolidation or atelectasis; 3 - presence of three or more B lines in an intercostal space in no dependent lung areas suggesting alveolar-interstitial edema; 4 - presence of homogeneous anechoic area in dependent lung area suggesting effusion". Among these items, consolidations and atelectasis might be present in only one lung field, but due to its clinical relevance was considered an alternative pulmonary diagnosis. In this sense, we do not believe we have missed or under-diagnosed alternative lung diagnosis to PE.

Comment 3: The authors did not specify if RV dysfunction was acute or chronic as presence of previous RV dysfunction may confound the findings significantly. Also, I believe all patients with previous RV dysfunction should have been excluded from the study.

Answer 3: This is a very important point to clarify. We consider all patients to have acute RV dysfunctions. Few patients had previous echocardiography, none of them with right ventricular dysfunction or pulmonary hypertension. We assumed that patients without previous echocardiography would not have right ventricular dysfunction or pulmonary hypertension. We included a statement in Results, to better clarify the topic, as follows (page 11, paragraph 1 ):

“Of the 37 patients with positive findings for PE on CTA, 22 (60%) had RV dysfunction on POCUS TTE, with 63% sensitivity and 85% specificity. All patients were considered to have acute RV dysfunction.”

Comment 4: There was no mention of patients with history of previous PE or being treated for PE as these also may confound the findings.

Answer 4: This is a very important point in the evaluation of patients with suspected PE. Information on previous history of PE or VTD was analyzed in all patients, as this is one of the items used to calculate the Wells score, which was performed for all patients. In our sample, only two patients had previous PE or VTD and none of them were on treatment for PE or DTV in the last six months. This information was included in Results section (page 9, paragraph 1 of the results):

“Eighty-eight critically ill patients were included in the study. Two patients had a history of previous PE or DVT, but none of them were being treated for PE or DVT in the last six months.”

Comment 5: Were the images acquired by the fellows saved and reviewed by more experienced physician to concur with accuracy of ultrasound image acquisition given that ultrasound is operator dependent?

Answer 5: When we design the paper, we had a specific concern about this point. Critical care fellows were specifically trained for ultrasound image acquisition for the protocol for three months before to start the inclusion of patients. All images were saved and revised by a senior physician (author R.B.A). This point is clarified in Material and Methods (page 6, paragraph 1), as follows:

“POCUS examinations were performed independently by two critical care fellows with advanced training in POCUS who had undergone three months of training in image acquisition specifically for the study protocol. All images were saved and reviewed by a senior physician if necessary.”

Comment 6: The authors mentioned specificity of 100% in the abstract and then mentioned specificity of 96% in the results section, this should be corrected and clarified.

Answer 6: Thanks for pointing this error. The correct value for specificity is 96%. The information was corrected accordingly in Abstract (page 3, paragraph 1):

“Combining these three findings resulted in an area under the curve of 0.85 (95% CI, 0.77-0.94), with 50% sensitivity and 96% specificity.”

Comment 7: Given sensitivity of 50%, I think it is really difficult to state in the conclusion the diagnostic accuracy in high at this point. I would be more cautious with such a statement as the data presented does not fully support this conclusion.

Answer 7: We agree with the reviewer. The statement was changed as follows (page 18, paragraph 1, and in Abstract, page 3):

In Conclusion section:

“In conclusion, multiorgan POCUS combined with laboratory data has acceptable diagnostic accuracy for PE compared with CTA and might be helpful in the ICU setting, providing an alternative method when performing a CTA is too risky and reducing unnecessary tests when the probability of disease is too high.”

In Abstract section:

“Multiorgan POCUS combined with laboratory data has acceptable diagnostic accuracy for PE compared with CTA. The combined use of these methods might reduce CTA overuse in critically ill patients.”

Reviewer #3

Comment 1: I had a very difficult time determining what proportion of PEs were sub-segmental in nature.

Answer 1: As stated in Material and Methods (page 8, paragraph 1), thrombus location was defined according to the caliber of the affected vessel and classified as follows: main artery, lobar, segmental, or subsegmental. Scans were considered negative in the presence of adequate opacification of the pulmonary artery without filling defects, according to Moores et al. (included Ref 17, please refer to References). We had four patients with subsegmental PE (11% of patients with PE). This information was added to Results (page 9, paragraph 2), as follows:

“Of 88 patients, 37 (42%) had PE detected on pulmonary CTA examination. Of these, 12 patients (32%) had PE in the main artery, three (8%) in the lobar branch, 18 (49%) in the segmental branch, and four patients (11%) in the subsegmental branch.”

Comment 2: The paper also requires significant language editing.

Answer 2: Thanks for this important contribution to the final quality of the manuscript. A new round of language editing was performed. Paraphrasing that did not change the meaning of sentences is not highlighted in yellow. (Attached certificate in "response to reviewers" file)

Comment 3: Abstract is lacking key information i.e. the sample size included in the study., in order to provide in the abstract an informative and balanced summary of what was done and what was found in the study, in accordance with STROBE statement.

Answer 3: Abstract was fully revised and all key information (including sample size) was amended accordingly with STROBE statement, in order to provide in the abstract an informative and balanced summary of what was done and what was found in the study. Please refer to Abstract section (page 2).

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

19 Sep 2022

29 Sep 2022

Dear reviewers, we thank you for revising our manuscript. We make changes as requested. Answer follows below:

Thank you for reviewing the above referenced manuscript. All changes made are highlighted in yellow in the revised version of the manuscript. Please see below the answers for all queries.

Journal Requirements:

Comment 1: Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Answer 1: We thoroughly revised the reference list. In our last version of the manuscript, we added reference 17, which defines findings suggestive of PE on pulmonary computed tomographic angiogram. Also, we modified the position of reference 16 (previously reference 23), mentioned now in Materials and Methods and Discussion sections. In the current version of the manuscript, the references are all complete and correct. No reference cited that have been retracted. Reference list meets PLOS ONE style requirements.

Reviewer #2

I would like to thank the authors for addressing the comments on the manuscript.

Comment 1: I think it would be appropriate that the authors acknowledge the lack of exclusion of previous RV dysfunction as weakness in the discussion section.

Answer 1: A statement was added to the Discussion section as follows (page 18, paragraph 1):

“Third, a formal exclusion of patients with previous RV dysfunction was not performed. However, no patients were suspected to have previous pulmonary hypertension based on previous echocardiography data and clinical data. Fourth, this is a single-center study.”

Reviewer #3

Comment 1: I am quite satisfied with the author's revisions of this manuscript. In particular, the attention paid to the inclusion of the STROBE statement elevates the project considerably.

Answer 1: We thank the reviewer for spending his time to improve your manuscript.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

2 Oct 2022

Diagnostic accuracy of multiorgan point-of-care ultrasound compared with pulmonary computed tomographic angiogram in critically ill patients with suspected pulmonary embolism.

PONE-D-22-18103R2

Dear Dr. Girardi Adriana

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Patricia Rezende do Prado

Academic Editor

PLOS ONE

10 Oct 2022

PONE-D-22-18103R2

Diagnostic accuracy of multiorgan point-of-care ultrasound compared with pulmonary computed tomographic angiogram in critically ill patients with suspected pulmonary embolism.

Dear Dr. Girardi:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Patricia Rezende do Prado

Academic Editor

PLOS ONE