Prognostic Value of ECG Among Patients with Acute Pulmonary Embolism and Normal Blood Pressure

CLINICAL RESEARCH STUDY

Prognostic Value of ECG Among Patients with Acute Pulmonary Embolism and Normal Blood Pressure Simone Vanni, MD, PhD, Gianluca Polidori, MD, Ruben Vergara, MD, Giuseppe Pepe, MD, PhD, Peiman Nazerian, MD, Federico Moroni, MD, Emanuele Garbelli, MD, Fabio Daviddi, MD, Stefano Grifoni, MD The Emergency Department, Azienda Ospedaliero-Universitaria Careggi, Florence, Italy.

ABSTRACT OBJECTIVE: To investigate the prognostic value of electrocardiography (ECG) alone or in combination with echocardiography in patients with acute pulmonary embolism and normal blood pressure. METHODS: Consecutive adult patients presenting to the emergency department at Azienda OspedalieroUniversitaria Careggi with the first episode of pulmonary embolism were included. Patients with systolic blood pressure less than 100 mm Hg were excluded. ECG and echocardiography were performed within 1 hour from diagnosis and evaluated in a blinded fashion. Right ventricular strain was diagnosed in the presence of one or more of the following ECG findings: complete or incomplete right ventricular branch block, S1Q3T3, and negative T wave in V1-V4. The main outcome measurement was clinical deterioration or death during in-hospital stay. The association of variables with the main outcome was evaluated by multivariate Cox survival analysis. RESULTS: A total of 386 patients with proved pulmonary embolism were included in the study; 201 patients (52%) had right ventricular dysfunction according to echocardiography, and 130 patients (34%) showed right ventricular strain. Twenty-three patients (6%) had clinical deterioration or died. At multivariate survival analysis, right ventricular strain was associated with adverse outcome (hazard ratio 2.58; 95% confidence interval, 1.05-6.36) independently of echocardiographic findings. Patients with both right ventricular strain and right ventricular dysfunction (26%) showed an 8-fold elevated risk of adverse outcome (hazard ratio 8.47; 95% confidence interval, 2.43-29.47). CONCLUSION: Right ventricular strain pattern on ECG is associated with adverse short-term outcome and adds incremental prognostic value to echocardiographic evidence of right ventricular dysfunction in patients with acute pulmonary embolism and normal blood pressure. © 2009 Elsevier Inc. All rights reserved. • The American Journal of Medicine (2009) 122, 257-264 KEYWORDS: Echocardiography; Electrocardiography; Prognosis; Pulmonary embolism; Right ventricular dysfunction

Pulmonary embolism has different clinical presentations, from mild complaints to shock. One of the main determinants of clinical presentation and prognosis is the extension of obstruction within the pulmonary circulation.1 Accord-

Funding: None. Conflict of Interest: None of the authors have any conflicts of interest associated with the work presented in this manuscript. Authorship: All authors had access to the data and played a role in writing this manuscript. Requests for reprints should be addressed to Simone Vanni, MD, PhD, Emergency Department, AOU-Careggi, Viale Pieraccini 17, 50139 Firenze, Italy. E-mail address: [email protected]

0002-9343/$ -see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.amjmed.2008.08.031

ingly, the presence of acute right ventricular dysfunction is an important prognostic factor.1-5 In these studies, right ventricular dysfunction was diagnosed by echocardiography. However, echocardiography may be difficult to perform in the emergency setting. Electrocardiography (ECG) has a widespread use and is easily performed and interpreted in the emergency department. Some electrocardiographic signs, indicated as “right ventricular strain” pattern, were found to be related to the extension of obstruction in the pulmonary circulation6 and to right ventricular pressure overload7,8 with potential prognostic implications. Although ECG shares simple applicability and interpretation, and low cost, limited data are available about its prognostic

258

The American Journal of Medicine, Vol 122, No 3, March 2009

value in patients with pulmonary embolism,9,10 and no ECG, arterial blood gas analysis, and echocardiography. study has specifically investigated patients with normal Intravenous unfractionated heparin was started as soon blood pressure. This patient population represents the largas pulmonary embolism was suspected with standard doest proportion of patients with pulmonary embolism and the ses,12 and thrombolysis (recombined tissue plasminogen main therapeutic challenge because of actual uncertainty activator, 100 mg intravenously for 2 hours) was instiabout best treatment.11 tuted in patients with pulmonary The present study was designed embolism and right ventricular to investigate the potential prognosdysfunction as deemed appropriCLINICAL SIGNIFICANCE tic value of ECG in patients with ate by the attending physician. an objectively confirmed pulmonary To establish the prognostic role ● Right ventricular strain pattern is assoembolism and normal blood presof ECG in normotensive patients ciated with increased risk of all-cause sure. Furthermore, we compared and to compare ECG prognostic death or clinical deterioration in the the prognostic relevance of elecvalue with echocardiography, both short term. trocardiographic evidence of right ECG and echocardiography were ● Right ventricular strain pattern adds inventricular strain with that of right performed prospectively at preventricular dysfunction as revealed sentation blinded to the patient’s cremental prognostic value to echocarby echocardiography. clinical history and other test diographic evidence of right ventricular results. dysfunction. The primary end point was a MATERIALS AND METHODS ● Patients with right ventricular strain composite of death for any cause should receive strict clinical surveillance and clinical deterioration (defined Setting and Selection of and accurate evaluation for reperfusion as progression to shock, mechanical Participants therapy as patients with echocardiograventilation, or cardiopulmonary reConsecutive adult patients who suscitation, or the need for infusion phic right ventricular dysfunction. presented from January 1998 to of a catecholamine, except for doJune 2006 to the emergency depamine infused at a rate ⱕ5 ␮g/kg/ partment of a third-level teaching min) during in-hospital stay. hospital (Careggi, Florence, Italy) with the clinical suspiThe sample size was calculated assuming an incidence of cion of pulmonary embolism were considered for the study. the primary end point in normotensive patients with pulmoExclusion criteria were systolic arterial pressure persistently nary embolism with or without right ventricular strain simless than 100 mm Hg, a prior documented episode of pulilar to that of those patients with and without right ventricmonary embolism, and a history of severe chronic obstruc4 ular dysfunction (⬃11% and 3%, respectively). When we tive pulmonary disease or echocardiographic signs of longset the type I error level at 5%, a sample size of 130 pairs term right ventricular overload (see “Echocardiographic of subjects offered 80% power to detect a difference from Examination”). Patients with life expectancy less than 6 the incidence of the primary end point in the exposed months also were excluded. The diagnosis of pulmonary compared with nonexposed patients. embolism was established by perfusion lung scan or spiral computed tomography.4,5 A direct scan of the lungs was obtained using the multislice computed tomographic scanner Siemens Sensation 4 (Siemens Medical Systems, Erlangen, Germany), with 1-mm slices every 20 mm and a standard sequential acquisition technique. After the injection of contrast material, adjacent 3-mm slices were obtained over the hilar region using the spiral acquisition technique. Total scan time ranged from 5 to 10 minutes. The diagnosis of pulmonary embolism was based on the direct visualization of partial or complete filling defects within the pulmonary arteries. Patients with proved pulmonary embolism were enrolled in the study and gave written consent to the use of their medical information for research purposes. The study was approved by the institutional review board.

A 12-lead ECG was obtained in all patients with an objectively confirmed pulmonary embolism within 1 hour from diagnosis. ECG was indicative of acute right ventricular strain when at least one of the following patterns was found: complete or incomplete right bundle branch block, S waves in lead I combined with Q waves in lead III with or without T inversion in lead III (S1Q3T3), or inverted T waves in precordial leads V1, V2, and V3. We chose only these 3 signs because we think they were the most frequently used in the everyday practice. The ECGs were examined by 2 expert independent readers. Interobserver agreement (K coefficient) was 0.88. In case of discordance, the ECG was examined by a third independent reader.

Management Strategies and Study Design

Echocardiographic Examination

Patients were managed as described.4 Briefly, the initial patient assessment in the emergency department included clinical history, physical examination, chest x-ray, 12-lead

Standard color 2-dimensional echocardiographic Doppler examinations were performed within 1 hour from the diagnosis of pulmonary embolism as previously described.4

Electrocardiography

Vanni et al Table 1

ECG and Pulmonary Embolism Prognosis

259

Features of Study Patients Based on Electrocardiography at Presentation All Patients (n ⫽ 386)

Age (y) Females Previous or concomitant disease Diabetes CVD Malignancy Risk factors for VTE Permanent Transient Idiopathic Clinical presentation Syncope Chest pain Dyspnea Tachycardia Cold sweating SAP (mm Hg) PAO2 (mm Hg) Thrombolysis RVD Troponin I (ng/mL)b ⬎0.15 ng/mL

RVS (n ⫽ 130)

No RVS (n ⫽ 256)

P Value

67 ⫾ 16 233 (60%)

a

71 ⫾ 14 83 (64%)

65 ⫾ 17 150 (59%)

.001 .325

37 (10%) 47 (12%) 86 (22%)

14 (11%) 21 (16%) 31 (24%)

23 (9%) 26 (10%) 55 (22%)

.586 .100 .607

105 (27%) 150 (39%) 131 (34%)

39 (30%) 45 (35%) 46 (35%)

66 (26%) 105 (41%) 85 (33%)

.415 .269 .733

33 (9%) 120 (31%) 257 (67%) 118 (31%) 54 (14%) 132 ⫾ 22 65 ⫾ 12 42 (11%) 201 (52%) 0.28 ⫾ 0.92 23 (29%)

14 (11%) 45 (35%) 95 (73%) 56 (43%)a 30 (23%)a 129 ⫾ 24a 61 ⫾ 12a 19 (15%) 102 (79%)a 0.27 ⫾ 0.37 14 (44%)

19 (7%) 75 (29%) 162 (63%) 62 (24%) 24 (9%) 134 ⫾ 20 67 ⫾ 11 24 (9%) 99 (39%) 0.28 ⫾ 0.37 9 (17%)

.335 .297 .067 ⬍.001 .001 .042 ⬍.001 .127 ⬍.001 .935 .045

RVS ⫽ right ventricular strain; RVD ⫽ right ventricular dysfunction; CVD ⫽ cardiovascular disease (previous coronary or cerebrovascular event); VTE ⫽ venous thromboembolism; PE ⫽ pulmonary embolism; SAP ⫽ systolic arterial pressure; PAO2 ⫽ arterial partial pressure of oxygen. a P ⬍.05 vs the no RVS group by Fisher’s exact test. b Troponin I was tested in 78 patients, 32 with and 46 without RVS.

Briefly, patients with at least one of the following findings were diagnosed with acute right ventricular dysfunction: right ventricular dilatation (end-diastolic diameter ⬎30 mm or right/left ventricular end-diastolic diameter ratio ⬎1 in apical 4-chamber view); paradoxical septal systolic motion; and/or pulmonary hypertension (Doppler pulmonary acceleration time ⬍90 ms or presence of a right ventricular/atrial gradient ⬎30 mm Hg). Patients with signs of right ventricular overload in the presence of right ventricular free wall hypertrophy (end-diastolic thickness ⬎6 mm) were excluded from the study.

Cardiac Troponin I Testing In a subset of patients (n ⫽ 78), starting from 2004, cardiac troponin I was determined on an ADVIA Centaur Analyzer (Bayer VitalGmbH, Fernwald, Germany) according to the manufacturer’s instructions. The investigator responsible for the measurements was unaware of the patients’ baseline parameters or clinical course. Reported values in normal healthy adults by our central laboratory are less than 0.15 ng/mL.

Statistical Analysis Data points are expressed as means ⫾ standard deviation. The unpaired Student t test or 1-way analysis of variance was used to compare normally distributed data where appropriate. The Fisher exact test was used for the comparison

of noncontinuous variables expressed as proportions. Kappa measurement was performed to examine the interobserver agreement for ECG signs. To investigate the prognostic relevance of the baseline parameters listed in Table 1, including right ventricular strain pattern, a multivariate Cox proportional-hazards regression model13 was applied to the end point, taking into account those variables that were expected to have an association with the main outcome and reached a probability value of less than .10 in the univariate analysis. Multivariate analysis was performed with a stepwise backward regression model. To exclude potential bias resulting from different treatment between patients with or without right ventricular strain, the analysis was repeated by adjusting for thrombolytic administration. In addition, to evaluate the role of a strategy combining ECG with echocardiography for risk stratification of pulmonary embolism, multivariate analysis compared the complication risk of 4 patient groups defined by the combination of ECG and echocardiography. No adjustments for other baseline parameters were made in this latter model. Survival curves were constructed according to the Kaplan–Meier method. P values are 2-sided, and a P value of less than .05 was considered to indicate statistical significance. Calculations were performed with the Statistical Package for the Social Sciences (version 14.0, SPSS Inc, Chicago, Ill).

260

The American Journal of Medicine, Vol 122, No 3, March 2009

Table 2

Events During In-Hospital Stay All Patients (n ⫽ 386)

Clinical deterioration Progression to shock Catecholamine infusionb Mechanical ventilation Cardiopulmonary resuscitation Death (all causes) Composite

15 10 12 5 4 12 23

(4%) (3%) (3%) (1%) (1%) (3%) (6%)

RVS (n ⫽ 130) 11 8 9 4 4 8 15

No RVS (n ⫽ 256)

a

(9%) (6%) (6%) (3%) (3%) (6%)a (12%)a

4 2 3 1 0 4 8

(2%) (1%) (1%) (1%) (0%) (2%) (3%)

P Value .002

.025 .002

RVS ⫽ right ventricular strain. a P ⬍.05 vs the no RVS group by Fisher’s exact test. b Except for dopamine infused at a rate ⱕ5 ␮g/kg/min.

RESULTS Patients and Management A total of 557 consecutive patients with an objectively confirmed acute pulmonary embolism were considered for the study from January 1998 to June 2006. Of these patients, 36 were excluded because of shock on admission (7%), whereas 123 additional patients (22%) were excluded because they had a documented previous episode of pulmonary embolism (50 patients), were affected by severe chronic obstructive pulmonary disease, had echocardiographic signs of long-standing right ventricular overload (63 patients), or lacked adequate acoustic window (14 patients). Twelve patients declined to participate. Therefore, 386 patients were included in the study. Patients had a mean age of 67 ⫾ 16 years (range 18-92 years), and 233 (60%) were female (Table 1). Of the overall patients, 131 (34%) had an idiopathic pulmonary embolism. The diagnosis of pulmonary embolism was performed by lung scan (162 patients), CT scan (195 patients), or pulmonary angiography (29 patients). A total of 234 patients (61%) had a concomitant deep vein thrombosis. Intravenous unfractionated heparin was started in all patients as soon as pulmonary embolism was suspected, and 43 patients (11%) were treated with thrombolytic agents. After the initial treatment, 375 patients (97%) received vitamin K antagonists and 11 patients (3%) received a vena cava filter because of permanent contraindications to anticoagulation. During in-hospital stay, 23 patients (6%) reached the combined end point of death or clinical deterioration, with 12 patients who died (3%) (Table 2). A total of 201 patients (52%) had echocardiographic signs of right ventricular dysfunction. As expected, right ventricular dysfunction was associated with clinical deterioration or death during inhospital stay (hazard ratio [HR] 4.24, P ⫽ .009) (Table 3).

Electrocardiography and Short-Term Outcome At presentation, ECG was indicative of acute right ventricular strain in 130 patients (34%) (Table 1). Patients with right ventricular strain more often had dyspnea or cold

sweating at presentation. In addition, they showed mean systolic blood pressure and partial oxygen pressure lower than those without right ventricular strain. During in-hospital stay, 15 patients (12%) with right ventricular strain reached the end point and 8 patients (6%) died (P ⫽ .002 and P ⫽ .025 vs patients without ECG signs of right ventricular strain, respectively, Table 2). These differences yielded a high negative predictive value (97%; 95% confidence interval [CI], 95-98) and a low, but statistically significant, positive predictive value (12%; 95% CI, 8-14) for adverse clinical outcome. At univariate Cox survival analysis, the presence of right ventricular strain was associated with death or clinical deterioration at 30 days (HR 3.75; 95% CI, 1.58-8.89) (Table 3, Figure 1). At multivariate analysis, right ventricular strain was still associated with the composite end point (HR 2.58; 95% CI, 1.05-6.36) independently of echocardiographic findings (Table 3). To exclude potential bias resulting from different treatment between patients with or without right ventricular Table 3 Results of Univariate and Multivariate Cox Proportional-Hazards Age-Adjusted Analysis of the Relation Between Baseline Clinical Variable and In-Hospital Outcome (Clinical Deterioration or Death) Univariate Variables RVS RVD Syncope Tachycardia Dyspnea PAO2 ⬍ 66 (mm Hg) Cold sweating

HR (95% CI) 3.75 4.24 4.55 2.54 3.20 2.58

Multivariate P Value

HR (95% CI)

(1.58-8.89) .003 2.58 (1.05-6.36) (1.44-12.53) .009 _____________ (1.86-11.13) ⬍.001 4.42 (1.19-9.16) (1.12-5.76) .026 _____________ (0.95-10.79) .061 _____________ (0.93-7.18) .067 _____________

2.38 (0.94-6.04)

.068 _____________

P Value .038 .178 .022 .266 .054 .211 .217

CI ⫽ confidence interval; RVS ⫽ right ventricular strain; RVD ⫽ right ventricular dysfunction; HR ⫽ hazard ratio; PAO2 ⫽ arterial partial pressure of oxygen. Dashes denote variables not included in the final model.

Vanni et al

ECG and Pulmonary Embolism Prognosis

Figure 1 In-hospital follow-up of patients with first episode of acute pulmonary embolism and normal blood pressure based on ECG pattern at presentation. ECG ⫽ electrocardiography.

strain, the analysis was repeated by adjusting for thrombolytic administration, and the result did not change (HR 2.92; 95% CI, 1.18-7.23).

Right Ventricular Strain Pattern and Echocardiographic Evidence of Right Ventricular Dysfunction for Risk Assessment in Pulmonary Embolism Among patients with right ventricular dysfunction on admission, 102 (51%) showed at least one ECG sign of right ventricular strain (P ⬍.001 vs patients without right ventricular dysfunction) (Table 4). Each ECG sign of right ventricular strain was significantly more frequent in patients with right ventricular dysfunction than in those without. S1Q3T3 or TnegV1-V3 were the most frequent ECG signs in the group with right ventricular dysfunction (6-fold more frequent than in the group with no right ventricular dysfunction).

Table 4 Electrocardiography Findings in Patients with Pulmonary Embolism With or Without Right Ventricular Dysfunction

Signs of RVS RBBB S1Q3T3 TnegV1-V3

All Patients (n ⫽ 386)

RVD (n ⫽ 201)

130 39 56 60

102 30 49 47

(34%) (10%) (15%) (16%)

(51%)a (15%)a (24%)a (23%)a

no RVD (n ⫽ 185)

P Value

28 9 7 13

⬍.001 .001 ⬍.001 ⬍.001

(15%) (5%) (4%) (7%)

RVS ⫽ right ventricular strain; RBBB ⫽ complete or incomplete right bundle branch block; S1Q3T3 ⫽ S waves in lead I combined with Q waves in lead III with or without T inversion in lead III; TnegV1-V3 ⫽ inverted T waves in precordial leads V1, V2, and V3. a P ⬍.05 vs the no RVD group by Fisher’s exact test.

261

Figure 2 Incidence of clinical end point during in-hospital stay in 386 normotensive patients with first episode of acute pulmonary embolism and different ECG and echocardiographic results at presentation. Echo ⫽ echocardiography; RVS ⫽ right ventricular strain; RVD ⫽ right ventricular dysfunction.

To investigate the potential usefulness of a new riskassessing strategy combining ECG and echocardiographic results, we categorized patients into 4 groups: patients without both right ventricular strain and right ventricular dysfunction (n ⫽ 157); patients without right ventricular strain and with right ventricular dysfunction (n ⫽ 99); patients with right ventricular strain and without right ventricular dysfunction (n ⫽ 28); and patients with both right ventricular strain and right ventricular dysfunction (n ⫽ 102). As shown in Figure 2, the incidence of death or clinical deterioration was not different in the first 2 groups (2% vs 5%, P ⫽ .38), suggesting that in patients without right ventricular strain pattern, echocardiography adds little prognostic information. Conversely, in patients with right ventricular strain the absence of right ventricular dysfunction identified a group of patients at low risk (4%, P ⫽ .29 vs no right ventricular strain patients), whereas the presence of right ventricular dysfunction sharply increased the risk of acute adverse events (cumulative incidence 14%, P ⬍.001 vs all other patients). Accordingly, Cox survival analysis showed that the presence of both right ventricular strain and right ventricular dysfunction was associated with an 8-fold elevated risk of an adverse outcome during the in-hospital phase of pulmonary embolism (HR 8.47; 95% CI, 2.4329.47) (Table 5).

Right Ventricular Strain Pattern and Cardiac Troponin I for Risk Assessment in Pulmonary Embolism Troponin I was tested in 78 of 386 patients (Table 1). Abnormal cardiac troponin I values (⬎0.15 ng/mL) were more often present in patients with right ventricular strain

262

The American Journal of Medicine, Vol 122, No 3, March 2009

Table 5 Combination of Right Ventricular Strain on Electrocardiography and Right Ventricular Dysfunction on Echocardiography for Risk Stratification of Patients with Pulmonary Embolism and Normal Blood Pressure Clinical Deterioration or Death Risk Group

HR (95% CI)

Patients without both RVS and RVD Patients without RVS and with RVD Patients with RVS and without RVD Patients with both RVS and RVD

______________

P Value

2.65 (0.63-11.08)

.182

1.83 (0.19-17.61)

.600

8.47 (2.43-29.47)

.266

CI ⫽ confidence interval; HR ⫽ hazard ratio; RVS ⫽ right ventricular strain; RVD ⫽ right ventricular dysfunction.

than in those without (P ⫽ .026). Patients who reached the end point (n ⫽ 8) had abnormal troponin I values (63%) more often than patients with uncomplicated course (26%) (P ⫽ .045). The derived negative predictive value (95%; 95% CI, 90-98) was comparable to that of right ventricular strain. The positive predictive value (22%; 95% CI, 11-30) was higher than that of right ventricular strain, although the difference between the 2 tests did not reach statistical significance (P ⫽ .122). Patients with both right ventricular strain and troponin I greater than 0.15 ng/mL had a nonsignificant increase in the incidence of the end point (14%) in comparison with those without (9%, P ⫽ .629).

DISCUSSION The present study demonstrates the association of right ventricular strain pattern showed by the presentation ECG with adverse short-term outcome (clinical deterioration or death) in patients with acute pulmonary embolism and normal blood pressure. In addition, right ventricular strain was found to add incremental prognostic value to echocardiographic evidence of right ventricular dysfunction. A large body of evidence has shown that although ECG may reveal several abnormalities in pulmonary embolism, it has a low sensitivity and specificity in pulmonary embolism diagnosis.14,15 Few studies have been done assessing the prognostic role of ECG in patients with pulmonary embolism, particularly regarding the risk of death or clinical deterioration during in-hospital phase.7,10,16 However, all these studies included patients with clinically evident hemodynamic impairment (persistent hypotension or shock) who are known to have a severe prognosis in the short term and who receive a clear benefit from aggressive treatment.17,18 No study has been focused on normotensive patients, who are the largest and most heterogeneous subgroup of patients with pulmonary embolism. This subgroup, dif-

ferent from that of patients with evident hemodynamic impairment, requires careful risk stratification in the acute phase with 2 main objectives: to identify patients who are at low risk of adverse events and can be safely managed in low-intensity beds, and to identify those patients who need strictly surveillance (monitoring) and may require more aggressive treatments than heparin alone.5,19 Recent echocardiographic evidence of right ventricular dysfunction was found to have prognostic relevance in this subset of patients.4,5 However, echocardiography may not be routinely available or may be difficult to perform in an emergency setting. Therefore, we look to a more simple way to stratify normotensive patients and to optimize the use of echocardiography in the emergency workup of patients with acute pulmonary embolism. The present study showed that the majority of normotensive patients with acute pulmonary embolism had no ECG signs of right ventricular strain at presentation (66%). The absence of right ventricular strain showed a high negative predictive value (97%) with respect to adverse shortterm outcome, similar to that of absence of right ventricular dysfunction at urgent echocardiography (98%), therefore identifying by itself a large group of patients at low risk of clinical deterioration. In this group of patients, urgent echocardiography seems to add little prognostic information (Figure 2). Although these data suggest that perhaps it is safe to postpone echocardiography in normotensive patients without ECG signs of right ventricular strain, caution is needed. First, as with all instrumental and humoral data, ECG must be interpreted together with other relevant clinical findings (Table 1). Second, sometimes a right ventricular strain pattern can become evident hours after patient presentation,9 so that serial ECG might be warranted, especially when echocardiography is postponed. The presence of at least one classic ECG sign of right ventricular strain (34% of all normotensive patients) is associated with increased risk of death or clinical deterioration (HR 2.58) in the short term, independently of other relevant clinical variables and most important of echocardiographic evidence of right ventricular dysfunction. Thus, normotensive patients with right ventricular strain on admission should receive strict clinical monitoring, similar to those with right ventricular dysfunction at echocardiography.4,5 Humoral markers were found to be related to prognosis in patients with acute pulmonary embolism.19-23 It would have been interesting to directly compare the prognostic yield of ECG with that of troponins or natriuretic peptides; unfortunately, at the beginning of present study these data were not available and we could analyze cardiac troponin I results in only 78 patients. On analyzing this subset of patients, we found that troponin I values were abnormal more often in those with clinical deterioration or death in comparison with those with uncomplicated course. According to other recent studies, troponin I showed a high negative predictive value (⬎90%) but a low positive predictive

Vanni et al

ECG and Pulmonary Embolism Prognosis

value (⬃20%) for short-term adverse outcome.21,22 Of note, the same limitation was shared by right ventricular strain, echocardiography, and natriuretic peptides.4,5,21,22 For this reason, strategies combining humoral and imaging tests have recently been investigated to better identify low- and high-risk patients.19,23 To our knowledge no previous studies investigated the prognostic relevance of combining ECG and echocardiography in acute pulmonary embolism. In our large cohort of normotensive patients, the combination of a right ventricular strain pattern with echocardiographic evidence of right ventricular dysfunction was able to identify a subgroup with an 8-fold increased risk of adverse short-term outcome, similar to that reported for the combination of N-terminal pro-brain natriuretic peptide plus echocardiography (12-fold) and of troponin I plus echocardiography (10-fold).19 Reasonably, these results provide the background for testing the possible benefits of early pulmonary reperfusion, either by thrombolytics or other techniques, in patients with both right ventricular strain and right ventricular dysfunction at presentation. On the other hand, we could not demonstrate the potential usefulness of combining troponin I and ECG results for pulmonary embolism prognostication, probably because of the small number of patients with troponin I results. Furthermore, other potential associations between right ventricular strain and anamnestic or clinical findings were investigated by multivariate analysis, but only syncope remained significant (Table 3). Although these findings may be of potential clinical interest, 9% of patients had syncope at presentation and approximately 4% of patients had both right ventricular strain and syncope in our cohort, largely reducing the clinical impact of this subgroup for risk stratification. The main limitation in the use of right ventricular strain as a prognostic tool in patients with pulmonary embolism is that approximately 10% of the normal population may show these signs, in particular incomplete or complete right bundle branch block.15 Accordingly, in our series 7% of patients without right ventricular dysfunction had at least one sign of right ventricular strain on admission. However, as revealed by univariate and multivariate analyses, right ventricular strain at ECG was not the only variable associated with the combined end point (Table 3). These data again suggest that a right ventricular strain pattern must be interpreted together with other simple clinical variables that contribute to recognize the prognostic significance of ECG findings. Moreover, in patients with uncertain origin of right ventricular strain pattern (acute right ventricular overload or not), echocardiography may erase any doubt. Another important limitation is that the physicians caring for the patients were aware of ECG and echocardiographic results, thus introducing potential bias, in particular because of differences in management and treatment. For this reason, we repeated the multivariate analysis by adjusting for thrombolytic treatment, and

263 right ventricular strain remained significantly associated with the end point.

CONCLUSIONS The present study of normotensive patients with pulmonary embolism revealed that right ventricular strain pattern at ECG is associated with short-term clinical deterioration and death. When combined with echocardiography, right ventricular strain was better able to recognize normotensive patients at high risk of clinical deterioration. These patients might benefit by strict surveillance during hospital stay and careful evaluation for pulmonary reperfusion therapy.

References 1. Tapson VF. Acute pulmonary embolism. N Engl J Med. 2008;358: 1037-1052. 2. Kasper W, Konstantinides S, Geibel A, et al. Prognostic significance of right ventricular afterload stress detected by echocardiography in patients with clinically suspected pulmonary embolism. Heart. 1997;77: 346-349. 3. Goldhaber SZ, Visani L, De Rosa M. Acute pulmonary embolism: clinical outcomes in the International Cooperative Pulmonary Embolism Registry (ICOPER). Lancet. 1999;353:1386-1389. 4. Grifoni S, Olivotto I, Cecchini P, et al. Short-term clinical outcome of patients with acute pulmonary embolism, normal blood pressure, and echocardiographic right ventricular dysfunction. Circulation. 2000; 101:2817-2822. 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 of 90 mm Hg or higher. Arch Intern Med. 2005;165:1777-1781. 6. Iles S, Le Heron CJ, Davies G, et al. ECG score predicts those with the greatest percentage of perfusion defects due to acute pulmonary thromboembolic disease. Chest. 2004;125:1651-1656. 7. Kucher N, Walpoth N, Wustmann K, et al. QR in V1 an ECG sign associated with right ventricular strain and adverse clinical outcome in pulmonary embolism. Eur Heart J. 2003;24:1113-1119. 8. Punukollu G, Gowda RM, Vasavada B, et al. Role of electrocardiography in identifying right ventricular dysfunction in acute pulmonary embolism. Am J Cardiol. 2005;96:450-452. 9. Daniel KR, Courtney DM, Kline JA. Assessment of cardiac stress from massive pulmonary embolism with 12-lead ECG. Chest. 2001; 120:474-481. 10. Geibel A, Zehender M, Kasper W, et al. Prognostic value of the ECG on admission in patients with acute major pulmonary embolism. Eur Respir J. 2005;25:843-848. 11. Wan S, Quinlan DJ, Agnelli G, Eikelboom JW. Thrombolysis compared with heparin for the initial treatment of pulmonary embolism: a meta-analysis of the randomized controlled trials. Circulation. 2004; 110:744-749. 12. Hyers TM, Agnelli G, Hull RD, et al. Sixth ACCP Consensus Conference on antithrombotic therapy. Antithrombotic therapy for venous thromboembolic disease. Chest. 2001;119:176S-193S. 13. Cox DR. Regression models and life tables (with discussion). J R Statist Soc B. 1972;34:187-220. 14. Manganelli D, Palla A, Donnamaria V, Giuntini C. Clinical features of pulmonary embolism. Doubts and certainties. Chest. 1995;107 (1 Suppl):25S-32S. 15. Richman PB, Loutfi H, Lester SJ, et al. Electrocardiographic findings in emergency department patients with pulmonary embolism. J Emerg Med. 2004;27:121-126. 16. Kosuge M, Kimura K, Ishikawa T, et al. Prognostic significance of inverted T waves in patients with acute pulmonary embolism. Circ J. 2006;70:750-755.

264 17. Task Force on Pulmonary Embolism, European Society of Cardiology. Guidelines on diagnosis and management of acute pulmonary embolism. Eur Heart J. 2000;21:1301-1336. 18. British Thoracic Society guidelines for the management of suspected acute pulmonary embolism. Thorax. 2003;58:470-483. 19. Binder L, Pieske B, Olschewski M, et al. N-terminal pro-brain natriuretic peptide or troponin testing followed by echocardiography for risk stratification of acute pulmonary embolism. Circulation. 2005; 112:1573-1579. 20. Konstantinides S, Geibel A, Heusel G, et al; Management Strategies and Prognosis of Pulmonary Embolism-3 Trial Investigators. Hepa-

The American Journal of Medicine, Vol 122, No 3, March 2009 rin plus alteplase compared with heparin alone in patients with submassive pulmonary embolism. N Engl J Med. 2002;347: 1143-1150. 21. Meyer T, Binder L, Hruska N, et al. Cardiac troponin I elevation in acute pulmonary embolism is associated with right ventricular dysfunction. J Am Coll Cardiol. 2000;36:1632-1636. 22. Kucher N, Goldhaber SZ. Cardiac biomarkers for risk stratification of patients with acute pulmonary embolism. Circulation. 2003;108:2191-2194. 23. Kucher N, Wallmann D, Carone A, et al. Incremental prognostic value of troponin I and echocardiography in patients with acute pulmonary embolism. Eur Heart J. 2003;24:1651-1656.