Article, Cardiology

A novel ECG parameter for diagnosis of acute pulmonary embolism: RS time

a b s t r a c t

Objectives: Pulmonary embolism (PE) is one of the leading causes of cardiovascular mortality worldwide. Electro- cardiography (ECG) may provide useful information for patients with acute PE. In this study, we aimed to inves- tigate the diagnostic value of the QRS duration and RS time in inferolateral leads in patients admitted to the emergency department, and pre-diagnosed with acute PE.

Methods: We retrospectively enrolled 136 consecutive patients, admitted to the emergency department, pre- diagnosed with the clinical suspicion of acute PE, and underwent computerized tomographic pulmonary angiog- raphy (CTPA) to confirm the PE diagnosis. The study subjects were divided into two groups according to the pres- ence or absence of PE, and the independent predictors of PE were investigated.

Results: Sixty-eight patients (50%) had PE. Patients with PE had a longer RS time. Among the ECG parameters, only RS time was an independent predictor of PE (OR: 1.397, 95% CI: 1.171-1.667; p b 0.001). The ROC curve anal- yses revealed that the cut-off value of RS time for predicting acute PE was 64.20 ms with a sensitivity of 85.3% and a specificity of 79.4% (AUC: 0.846, 95%CI: 0.749-0.944; p b 0.001). In the Correlation analyses; the RS time was correlated with RV end-diastolic diameter (r = 0.422; p b 0.001), RV/Left ventricle ratio (r = 0.622; p b 0.001), and systolic pulmonary artery pressure (SPAP) (r = 0.508; p b 0.001).

Conclusion: As a novel ECG parameter, RS time could be measured for each patient. A longer RS time can be a very useful index for diagnosing acute PE as well as for estimating the RV end-diastolic diameter and SPAP.

(C) 2018

Introduction

Pulmonary embolism (PE) is one of the leading causes of cardiovas- cular mortality worldwide [1]. PE, which presents itself with various symptoms and clinical severities, sometimes is detected incidentally, while other times may cause sudden death [2]. Several postmortem studies consistently show a high percentage of PE going undiagnosed in the antemortem period: 55.2% in a study conducted in 2001 [3], 60% in a study conducted in 2007 [1], and 68% in a study conducted in 2013 [4]. The heterogeneous clinical presentation of PE may lead to dif- ficulties in recognition, and thus delays in the initiation of life-saving treatments. In order to overcome these difficulties, several Risk scoring

* Corresponding author.

E-mail address: [email protected] (I. Rencuzogullari).

systems and diagrams have been developed, based on history, physical examination, and laboratory findings, to assess the clinical probability of PE. [5,6]. However, considering the missed rate of diagnosis, it appears the management of the diagnostic process is not satisfactory, and most likely due to nonspecific nature of the symptoms.

Electrocardiography is known to provide useful information for diagnosis of acute PE, as well as several ischemic, inflammatory, and arrhythmic heart diseases. In patients with acute PE, several ECG changes can be observed [7]. Although the most common findings are tachycardia (including Atrial tachyarrhythmias) (42.2%) and any ST- segment and/or T-wave changes (68.2%) [8], the literature on prognos- tic effect of tachycardia and nonspecific ST-T wave changes is contradic- tory [9]. A less common ECG finding, the Right axis deviation (RAD), is found in 3% of the low risk PE patients, but in 27.8% of the high risk PE patients [10]. Both complete and incomplete right bundle branch

https://doi.org/10.1016/j.ajem.2018.09.010

0735-6757/(C) 2018

block (RBBB) are more commonly seen in severe acute PE patients. These parameters have a good specificity (99% and 97%, respectively) for predicting which patients are likely to have severe acute PE, how- ever, they have a poor sensitivity for overall acute PE patients [11]. S1Q3T3 pattern, which is considered as the pathognomonic and spe- cific (97%) ECG finding for acute PE, may not be seen in all acute PE patients. However, as previously shown, this pattern may have a pre- dictive value for RV strain or other cardiovascular events in acute PE patients [11].

Right ventricle overloading and dilatation due to embolus may lead to a late QRS vector directed towards the right and posterior, and delay in electrical conductivity. Even through the RBBB, the RAD, S1Q3T3 pattern, and clock-wise rotation on the horizontal axis are an indirect sign of the right and posterior direction of QRS vector due to the right ventricular dilatation and the increase in duration of QRS in the ECG in acute PE patients [12]; an ECG pa- rameter that directly measures this delay has not been demon- strated yet.

The delay in electrical conductivity exhibits itself in stretching the S- wave in inferolateral leads, causing an increase in the elapsed time from the beginning of QRS to the peak of the S-wave (RS time). We hypoth- esized that RS time could be quantified to provide insight into PE diag- nosis. In this study, we assessed the relationship between diagnosis of acute PE with QRS duration and prolonged RS time measured from the inferolateral leads.

Materials and methods

Study population

This is a retrospective study over the records of 145 consecutive patients that have come to the emergency department between the dates of January 2016 and March 2017, that were preliminarily diagnosed with PE by the emergency department physician, and subsequently were referred to the computerized tomographic pul- monary angiography (CTPA) for confirmation of PE. CTPA of all pa- tients had been evaluated by an experienced radiologist blinded to the patients’ characteristics.

As part of our study, the ECGs of these patients were evaluated by two experienced cardiologists, who were also blinded to the patients CTPA. Two of the files were excluded due to the radiologist deeming the CTPA results as non-conclusive, leaving 143 patients. Of these 143 patients, seven were excluded due to poor quality ECG images. Finally, the remaining 136 patients constituted our study population (Fig. 1).

Demographic features, clinical features such as heart rate, blood pressure, respiration rate, mental status, oxygen saturation under noninvasive monitoring, as well as biochemical and hematological values, such as complete blood count, blood Biochemical parameters, troponin-I, and d-dimer that had been measured on admission were ob- tained from patients’ records. All the patients already had their revised Geneva and Wells scores calculated in their files, to assess the clinical likelihood of acute PE. We obtained these scores from patients’ records as well. We also obtained left ventricle (LV) ejection fraction, RV dilata- tion, degree of tricuspid regurgitation (TR), and systolic pulmonary ar- tery pressure (SPAP) which had been recorded by the consulting cardiologist at the time of arrival to the emergency department, through Bedside echocardiography.

Multislice CTPA (SOMATOM Sensation 64; Siemens, Erlangen, Germany) had been performed in all patients using a standard CTPA protocol for PE. PE had been defined as a partial and/or complete endoluminal filling defect in the pulmonary artery system in at least two consecutive computed tomography (CT) sections.

The study protocol was reviewed and approved by the Local Ethics Committee of Kafkas University in accordance with the principles of the Declaration of Helsinki.

Electrocardiography

Digital 12-lead Standard ECGs with paper speed of 25 mm/s and 10 mm/mV had been performed on each patient, upon admission to the emergency department, and the strips had been added to each patient’s record. As part of our study, these ECG strips were obtained from patients’ records, scanned, and were subjected to analysis using digital image processing software (imagej.nih.gov/ij/). All these mea- surements were performed by two experienced cardiologists blinded to other information of the patients. As a result of these measurements heart rate, rightward deviation of frontal QRS axis, complete or incom- plete RBBB, QRS fragmentation on at least two contiguous leads, T- wave inversion on precordial leads, ST-segment depression on at least two contiguous leads, ST-segment elevation on at least two con- tiguous leads, STE in V1 (STEV1), STE in AVR (STEAVR) leads, prominent S-wave in D1 lead, presence of Q/q wave, T-wave inversion in D3 lead, and S1Q3T3 pattern were investigated.

The QRS duration was defined as the interval from the start of the QRS complex until the J-point, and was measured from the lead with the longest duration. As part of QRS, the RS time was defined as the in- terval from the beginning of the QRS complex until the nadir of S or S’ wave (Fig. 2). In our study RS time was calculated from the inferolateral leads, which is represented by leads D1, AVL, D2, D3, AVF, V4, V5, and V6. Similar to QRS, the RS time was measured from the lead with the longest duration from among the above-mentioned leads. All durations are presented in milliseconds (msec).

Statistical analyses

Statistical analyses were performed using the SPSS version 22.0 (IBM, Chicago, Illinois). Normality of the data was analyzed using the Kolmogorov-Smirnov test. Numerical variables with a normal distribu- tion are presented in terms of mean +- standard deviation (SD) values, while the non-normally distributed variables are presented as median and interquartile range values. Frequencies were calculated for the cat- egorical variables (numbers and percentages [%]). The continuous vari- ables of both the groups were compared using the Student t-test or the Mann Whitney U test. The categorical data were compared using the Chi-Square test or the Fisher exact test. Statistical significance was de- fined as a P value b0.05. The correlation of RS time with RV dimension, RV/LV ratio, and SPAP was assessed using Pearson correlation analyses. Multivariate logistic regression analyses were performed to identify the independent predictors of acute PE using the variables with marginal association with it in the univariate analyses. The RS time with the best specificity and sensitivity for predicting acute PE was calculated using receiver-operating characteristic (ROC) curve analysis. Effect size (Cohen’s d) and power value (1-?) of RS time comparison between patients with and without PE were calculated using G*Power software (version 3.1.9.2). The alpha level used for this analysis was b0.05.

Results

The study population comprised 136 patients who had underwent CTPA to confirm the diagnosis of acute PE (mean age: 60 +- 17 years; 52.9% [n = 72] women). Sixty-eight (50%) patients had been diagnosed with acute PE, whereas PE had been excluded in the remaining 68 (50%). Characteristics of the study population are summarized in two tables: Baseline demographics, and clinical characteristics are in Table 1, and laboratory characteristics are in Table 2.

Patients with acute PE had an increased respiratory rate, higher d- dimer and Troponin-I levels, and decreased oxygen saturation than those without PE. There was no statistical difference in the age and sex of the patients with and without acute PE. Although the median re- vised Geneva score was significantly higher in acute PE patients than in those without PE (3.0 [3.0-4.5] vs. 3.5 [1.0-6.0]; p = 0.002), there was

Between January 2016 and March 2017,based on symptoms, laboratory results, demographic, and risk scoring schemes, 145 consecutive patients had been preliminarily diagnosed with PE, by the emergency department physician. All these patients had subsequently been referred to Computerized tomographic pulmonary angiography

Fig. 1. Flow chart. Enrollment of study patients.

CTPA results of two of these patients had been deemed inconclusive in determining whether they had PE, by the attending radiologist. We excluded those from our study, therefore, leaving 143 patients

PE had been excluded from 73 of the patients

All CTPA results were evaluated by an experienced radiologist

Seventy (70) patients had been diagnosed with PE

The ECG results of all these patients were analyzed, retrospectively, as part of our study.

The remaining 136 patients (68 PE(+), and 68 PE(-)) were included in our study

Seven of these patients were excluded from our study due to having poor-quality ECG images. Of these, 2 had been diagnosed with PE, while 5 had been deemed PE-free, leaving 68 PE (+), and 68 PE (-) patients.

no statistical difference between the two groups in terms of the Wells score. While more of the patients with PE had underwent surgery, or were immobilized within the previous four weeks; had more symptoms of deep vein thrombosis; and had a history of hyperten- sion than those without PE, these differences did not reach statistical significance.

A comparison of the echocardiographic and tomographic find-

ings revealed that in acute PE patients, right ventricular dilatation, increased systolic pulmonary artery pressure, and the presence of more than a mild degree of tricuspid regurgitation were more fre- quent than in those without PE. Based on the thorax CT findings, patients with PE had a larger RV end-diastolic diameter and a greater RV/LV ratio than those without PE. Echocardiographic and tomographic findings of the study patients are listed in Table 3.

An analysis of the ECG parameters revealed that the patients with PE had a longer RS time (69.65 +- 7.63 vs. 59.72 +- 6.35; p b 0.001) and a longer QRS duration (101.38 +- 16.88 vs. 92.15 +- 18.99; p = 0.038). The effect size and power value of RS time comparison were 1.41 and

0.99, respectively. Also, PE patients had a more frequent Prominent S wave in D1 lead than the patients without PE. Although the frequency of increased heart rate, right axis deviation of the QRS axis, complete or incomplete RBBB, atrial fibrillation, and S1Q3T3 pattern were higher in patients with acute PE, these differences were not statistically signif- icant (Table 3). To eliminate the effect of complete and incomplete RBBB on QRS expansion or RS time, the comparison was repeated with the ex- clusion of patients with RBBB. While the QRS duration was similar in pa- tients with and without PE (n: 56, 96 +- 15 vs. n: 48, 90 +- 20; p = 0.197); the RS time was significantly longer in patients with PE than in those without PE (66.4 +- 6.1 vs. 58.7 +- 6.2; p b 0.001). Ten patients were treated with thrombolytic therapy due to massive PE; even after excluding these patients from the analyses, the RS time was still longer in patients with PE than in those without PE (n: 58, 68.92 +- 7.12 vs. n: 68, 59.72 +- 6.35; p b 0.001).

multivariate regression analysis was used to determine the indepen- dent predictors of acute PE, using parameters associated with acute PE in the univariate analysis (RS time, QRS duration, Geneva score, d- dimer level N 500 ug/L, troponin I level, oxygen saturation, and

Fig. 2. Identification of RS time on ECG strip.

respiratory rate). RS time (OR: 1.397, 95%CI: 1.171-1.667; p b 0.001), Geneva score (OR: 1.486, 95%CI: 1.142-1.933; p = 0.003), and respira- tory rate (OR: 1.307, 95%CI: 1.053-1.621; p = 0.015) were found to be independent predictors of acute PE (Table 4).

The ROC curve analysis revealed that the cut-off value of RS time for predicting acute PE was 64.20 msec with a sensitivity of 85.3% and a specificity of 79.4% (AUC: 0.846, 95% CI: 0.749-0.944; p b 0.001). The

cut-off RS time value of 60 msec (1,5 mm on ECG) for predicting acute PE had a 88.2% sensitivity and a 41.2% specificity (AUC: 0.647, 95% CI: 0.522-0.759; p = 0.004). In the ROC curve comparison, the AUC value of RS time was higher than that of the QRS duration (AUC: 0.685 95% CI: 0.561-0.792; p = 0.007) (Fig. 3).

In the correlation analyses; in addition to the weak-moderate corre- lation between RS time and RV end-diastolic diameter (r = 0.422; p b 0.001), there was a moderate correlation between the RS time and the RV/LV ratio (r = 0.622; p b 0.001) as well as between the RS time and SPAP (r = 0.508 p b 0.001) (Table 5).

Discussion

This study demonstrated that a prolonged RS time was associated with acute PE. Further, to our knowledge, this is the first study to report that prolonged RS time is an independent predictor of acute PE. Further- more, the RS time was correlated with RV end-diastolic diameter, RV/LV ratio, and SPAP.

PE is challenging to diagnose because of the non-specific nature of its clinical findings and symptoms. Patients with a suspicion of PE should be evaluated in terms of Detailed history of Predisposing factors and specific laboratory indices in addition to a validated risk scoring system that assesses the clinical probability [6,13]. Subjects of this study had been found to have a high percentage of d-dimer positivity and a high clinical probability of PE based on the modified Geneva and Wells score (with modified Geneva score being more distinguishing than Wells score). CTPA is recommended for patients with a high clinical probability of developing PE or those with a low or intermediate clinical

Table 1 Demographic, clinical characteristics and risk scores of all patients, patients with and without pulmonary embolism, with p value. Abbreviations: PE: pulmonary embolism; DVT: Deep vein thrombosis.

All patients (n:136) Patients without PE (n:68)

Patients with PE (n:68) p value

Age (years)

60

+-17

57

+-15

63

+-18

0.089

Female gender, n (%)

72

(52.9)

34

(50.0)

38

(55.9)

0.627

Diabetes Mellitus, n (%)

14

(10.3)

8

(11.8)

6

(8.8)

0.689

Hypertension, n (%)

42

(30.9)

14

(20.6)

28

(41.2)

0.066

Smoking, n (%)

62

(45.6)

30

(44.1)

32

(47.1)

0.808

History of pulmonary disease, n (%)

22

(16.2)

10

(14.7)

12

(17.6)

0.742

Coronary artery disease, n (%)

26

(19.1)

10

(14.7)

16

(23.5)

0.355

Heart Failure, n (%)

14

(10.3)

8

(11.8)

6

(8.8)

0.690

History of pulmonary embolism, n (%)

4

(2.9)

4

(5.9)

0

(0.0)

0.151

History of DVT, n (%)

6

(4.4)

2

(2.9)

4

(5.9)

0.555

History of malignancy, n (%)

8

(5.9)

2

(2.9)

6

(8.8)

0.303

Surgery or immobilization within past four weeks, n (%)

28

(20.6)

10

(14.7)

18

(26.5)

0.230

DVT symptoms, n (%)

28

(20.6)

8

(11.8)

20

(29.4)

0.072

Hemoptysis, n (%)

8

(5.9)

2

(2.9)

6

(8.8)

0.303

Systolic blood pressure, mm Hg

128.82

+-20.46

129.41

+-19.84

128.24

+-21.35

0.815

Body temperature ?C

36.40

36.20-36.85

36.40

36.20-36.90

36.40

36.20-36.70

0.558

Respiratory rate (/min)

20.0

18.0-24.0

20.0

18.0-22.0

20.5

20.0-28.0

0.033

Oxygen saturation (%)

93.0

86.5-96.0

94.0

91.0-96.0

91.0

81.0-94.0

0.004

Geneva score

5.00

3.00-8.00

3.50

1.00-6.00

6.00

4.00-9.00

0.002

Geneva score N3, n (%)

96

(70.6)

34

(50.0)

62

(91.2)

b0,001

Wells score

3.50

3.00-5.75

3.00

3.00-4.50

4.50

3.00-7.00

0.170

Wells score N1, n (%)

114

(83.8)

58

(85.3)

56

(82.3)

0.742

Table 2

Laboratory characteristics of all patients, patients with and without pulmonary embolism, with p value. Abbreviations: PE: pulmonary embolism; WBC: White Blood Cell.

All patients (n:136) Patients without PE (n:68) Patients with PE (n:68) p value

Blood urea nitrogen (mg/dl)

40.21

+-16.38

39.62

+-15.18

40.79

+-17.72

0.770

Alanine transaminase (U/L)

25.68

+-17.77

28.71

+-20.32

22.65

+-14.48

0.161

Aspartate transaminase(U/L)

27.01

+-15.33

30.21

+-16.71

23.82

+-13.30

0.086

Potassium (mmol/L)

4.27

+-0.47

4.28

+-0.48

4.27

+-0.47

0.894

Sodium (mmol/L)

137.63

+-4.39

136.79

+-5.00

138.47

+-3.57

0.116

Calcium (mg/dL)

8.92

+-0.54

9.03

+-0.57

8.82

+-0.49

0.105

WBC (/1000)

9.57

+-4.40

9.34

+-4.91

9.80

+-3.87

0.672

Hemoglobin (g/dL)

13.79

+-2.17

13.91

+-1.82

13.68

+-2.51

0.653

Platelet (10,000/uL)

230.22

+-79.92

231.74

+-74.42

228.71

+-86.18

0.877

D-dimer (ug/L)

2000.50

+-1444.24

1546.15

+-914.72

2454.85

+-1724.17

0.008

D-dimer N500 ug/L, n (%)

61

(89.7)

28

(82.4)

33

(97.1)

0.046

Troponin I level (ng/mL)

0.02

0.0-0.12

0.01

0.0-0.11

0.02

0.01-0.15

0.032

Troponin I N 0.04, n (%)

24

(35.3)

10

(29.4)

14

(41.2)

0.310

risk of PE with d-dimer positivity [6]. In the present study, PE diagnosis had been confirmed using CTPA, and PE had been found in 50% of the patients.

The 12-lead ECG remains one of the initial tests performed upon ad- mission to the hospital for patients complaining of chest pain and/or dyspnea. To date, numerous ECG findings, including sinus tachycardia, S1Q3T3 pattern, atrial tachyarrhythmias, Q wave in lead III, S1Q3T3 pat- tern, complete or incomplete RBBB, T wave inversion, and fragmenta- tion in QRS have been associated with PE [8,14]. Several possible underlying mechanisms have been proposed for the PE-related ECG changes, such as rapid RV pressure overload and RV enlargement, re- duction in electrical conduction due to mechanical compression of the increased cavity pressure and cellular hypoxia, damage in the overall perfusion of cardiac myocardium caused by RV infarction and/or de- crease in the preload of the left ventricle due to RV dysfunction, and probable cellular ischemia induced by mediators (such as catechol- amines or histamine) [15-20]. However, a majority of the these ECG findings have been found not to have sufficient specificity and

sensitivity for diagnosing PE [21,22]. In our study, in accordance with the aforementioned literature, ECG parameters including RAD, RBBB, clockwise rotation, S1Q3T3, and S1S2S3 patterns were also more fre- quent in the PE group; however, this increase in frequency did not reach statistical significance. The reason for these ECG parameters not reaching statistical significance is perhaps due to our study subjects, with and without PE, had already been deemed high-risk for the pur- poses of PE.

In a healthy heart, the latest portion of the QRS complex, the S-wave in leads V4-V6, occurs due to the right ventricular and the septal electri- cal forces being directed towards the base, while the left ventricular electrical forces have a more posterior direction [12]. In patients with chronic obstructive pulmonary disease, the spatial orientation of the heart and the insulating effect of the over-aerated lungs induce a late QRS vector oriented superiorly, and to the right; resulting in a wide, slurred S-wave in leads I, II, III, V4, V5, and V6 [23-26]. In the course of PE, some changes in the heart’s position are observed. The dilatation and possible ischemia of the RV could cause delayed activation of the

Table 3

Echocardiographic, computerized tomographic and electrocardiographic characteristics of all patients, patients with and without pulmonary embolism, with p value. Abbreviations: PE: pulmonary embolism; LVEF: left ventricle ejection fraction; TR: Tricuspid regurgitation; RV: Right ventricle; LV: Left ventricle; RBBB: Right bundle branch block; ECG: Electrocardiography.

All patients (n:136) Patients without PE (n:68)

Patients with PE (n:68) p value

echocardiographic findings LVEF

58.26

+-8.77

58.50

+-9.12

58.03

+-8.54

0.827

TR more than mild degree, n (%)

62

(45.6)

20

(29.4)

42

(61.8)

0.007

RV dilatation, n (%)

60

(44.1)

14

(20.6)

46

(67.6)

b0.001

Pulmonary artery systolic pressure

39.60

+-12.34

30.38

+-6.43

48.83

+-9.61

b0.001

Computerized tomography findings RV end-diastolic diameter (mm)

39.13

+-9.00

34.49

+-6.10

43.77

+-9.11

b0.001

LV end-diastolic diameter (mm)

37.90

+-7.21

38.95

+-6.43

36.85

+-7.86

0.231

RV/LV ratio

1.05

+-0.27

0.89

+-0.10

1.22

+-0.29

b0.001

electrocardiographic findings Atrial fibrillation, n (%)

18

(13.2)

4

(5.9)

14

(20.6)

0.074

Heart rate, (bpm)

90.96

+-22.64

87.53

+-27.18

94.38

+-16.67

0.215

Right axis deviation, n (%)

16

(11.8)

4

(5.9)

12

(17.6)

0.132

Clockwise rotation, n (%)

38

(27.9)

12

(17.6)

26

(38.2)

0.059

Complete or incomplete RBBB, n (%)

32

(23.5)

12

(17.6)

20

(29.4)

0.253

Fragmentation in QRS, n (%)

30

(22.1)

10

(14.7)

20

(29.4)

0.144

Precordial T wave negativity, n (%)

52

(38.2)

22

(32.4)

30

(44.1)

0.318

ST depression in any ECG derivation, n (%)

26

(19.1)

12

(17.6)

14

(20.6)

0.758

ST elevation in two or more contiguous leads, n (%)

50

(36.8)

28

(41.2)

22

(32.4)

0.451

ST elevation in V1 lead, n (%)

24

(17.6)

14

(20.6)

10

(14.7)

0.525

ST elevation in AVR lead, n (%)

26

(19.1)

12

(17.6)

14

(20.6)

0.758

Prominent S-wave in D1 Lead, n (%)

50

(36.8)

16

(23.5)

34

(50.0)

0.024

Prominent Q-wave in D3 Lead, n (%)

38

(27.9)

16

(23.5)

22

(32.4)

0.417

T wave negativity in D3 Lead, n (%)

38

(27.9)

16

(23.5)

22

(32.4)

0.421

S1Q3T3 pattern, n (%)

10

(7.4)

2

(2.9)

8

(11.8)

0.163

S1S2S3 pattern, n (%)

24

(17.6)

6

(8.8)

18

(26.5)

0.056

Presence of ST depression in V4-6, n (%)

24

(17.6)

14

(20.6)

10

(14.7)

0.525

QRS duration (msec)

96.76

+-18.38

92.15

+-18.99

101.38

+-16.88

0.038

RS time (msec)

64.68

+-8.58

59.72

+-6.35

69.65

+-7.63

b0.001

Table 4

Univariate and multivariate logistic regression analysis of demographic, clinical, labora- tory and electrocardiographic characteristics that predict pulmonary embolism.

Table 5

The correlation analysis between RS time and QRS duration and RV end-diastolic diameter, RV/LV ratio and systolic pulmonary artery pressure. Abbreviations: RV: Right ventricle; LV: Left ventricle.

Univariate analysis of PE Multivariate analysis of PE

Odds ratio

95% C.I. p

value

Odds ratio

95% C.I. p

value

RV end-diastolic diameter (mm)

RV/LV

ratio

Systolic pulmonary artery pressure

RS time, msec 1.225 1.109-1.353 b0.001 1.397 1.171-1.667 b0.001

RS time (msec)

QRS duration, msec

3.436 1.025-11.520 0.045 – – –

r value 0.422 0.622 0.508

p value b0.001 b0.001 b0.001

Geneva score 1.266 1.072-1.495 0.005 1.486 1.142-1.933 0.003

D-dimer N500 ug/L 7.071 0.803-62.311 0.078 – – –

Troponin I level Oxygen saturation

(%)

Respiratory rate (/min)

RV, resulting in RBBB, RAD, prominent S-wave in D1, S1S2S3 pattern, and clockwise rotation of horizontal axis [9,11,15,20]. We hypothesized that the PE induced delay in electrical conductivity, and the right and posteriorly oriented QRS vector could induce prolongation of the RS time measured from inferolateral leads, and therefore prolongation of the QRS duration. In fact, we observed that the prolongation of QRS, and the prolongation of RS intervals were significantly associated with PE; and the prolongation of RS time was an independent predictor of PE. To our knowledge, in an ECG, the RS time is the first-identified inter- val that shows the delay of electrical conductivity due to PE and/or RV loading.

In this study, aside from observing that the prolongation of RS time was an independent predictor of PE, we found that, RS time was longer in PE patients, even in those with no RBBB and/or massive PE. Moreover, the RS time was significantly correlated with RV’s dimensions, and RV/ LV ratio. Based on this result, we concluded that the prolongation of the last part of the QRS complex is present, even in patients without mas- sive PE causing obvious RBBB, RAD, and clockwise rotation. Further- more, the RS time could be used to quantify this observation. Given the sensitivity and specificity for predicting acute PE of defined ECG parameters that are established previously [16,27], irrespective of com- plete or incomplete RBBB, RS time appears to be an effective diagnostic tool.

We established a prolonged RS time has a predictive value for PE di- agnosis. Our ROC curve analysis has shown an RS time of N64.20 msec has the best predictive value. However, it begs the question how a clini- cian would measure a time of that precise in an ECG strip. In order to simplify such evaluation by a clinician, we considered using 60 msec

100

QRS

RS

80

n 136 136 136

QRS duration (msec)

3.654

0.581-22.958

0.167

r value

0.292

0.367

0.296

0.914

0.846-0.986

0.020

p value

0.016

0.002

0.014

1.116

1.004-1.240

0.041

1.307

1.053-1.621

0.015

n

136

136

136

(width of 1.5 squares on an ECG strip) as the basis for prediction. We ob- served using 60 msec also had good predictive value.

Our study has certain limitations. First, our study had a retrospective design and was based on patient file analyses. Second, the sample size is relatively low. Furthermore, due to the low number of in-hospital mor- tality, prognostic data could not be introduced. Third the bedside echo- cardiography that had been conducted by the consultant cardiology physician at the emergency department lacks some measurements. For instance, right ventricular dimensions were not measured, and right ventricular dilatation was reported only as present or absent. Fourth, subjects included in this study do not represent a homogeneous group, since patients with a history of various cardiovascular and/or pulmonary diseases, as well as those with no other diagnosed cardio- vascular disease were included in the same study. As such, we do not know the effect of presence of (or lack thereof) other cardiovascular dis- eases, and how they affect the results of our study. Finally, in patients with a high clinical risk, PE might have been underdiagnosed, especially considering the sensitivity and specificity of CTPA. In these patients, conventional pulmonary angiography may be a more suitable option.

Conclusion

ECG is a non-invasive, fast, inexpensive, and easily accessible diag- nostic tool. Recognition of specific ECG changes that may indicate pres- ence of PE, could help lead to an earlier diagnosis of a potentially deadly disease. We are mindful of the difficulty of diagnosing PE with a single ECG parameter, and this study’s premise is not that RS time alone is suf- ficient to diagnose PE. However, our findings indicate prolonged RS time on surface ECG is a novel and effective parameter that can be very useful index (along with other parameters) for diagnosing acute PE in the emergency room.

Funding

The authors received no financial support for the research, author- ship, and/or publication of this article.

Sensitivit

60

40

20

0

0 20 40 60 80 100

100-Specificity

Acknowledgements

Thanks to www.metastata.com for their contributions to statistical analysis and trial design.

Conflicts of interest

None declared.

References

  1. Cohen AT, Agnelli G, Anderson FA, Arcelus JI, Bergqvist D, Brecht JG, et al. VTE impact

    Fig. 3. ROC graphics to detect the cut-off value of RS time for pulmonary embolism prediction and ROC curves comparison between QRS duration and RS time to detect the better electrocardiographic parameter in predicting pulmonary embolism.

    assessment group in europe (VITAE). venous thromboembolism in Europe. The number of VTE events and associated morbidity and mortality. Thromb Haemost 2007;98:756-64.

    Bougouin W, Marijon E, Planquette B, Karam N, Dumas F, Celermajer DS, et al. Sud- den death expertise center. Factors associated with pulmonary embolism-related sudden cardiac arrest. Circulation 2016;134:2125-7.

  2. Pineda LA, Hathwar VS, Grant BJ. Clinical suspicion of fatal pulmonary embolism. Chest 2001;120:791-5.
  3. Sweet 3rd PH, Armstrong T, Chen J, Masliah E, Witucki P. Fatal pulmonary embolism update: 10 years of autopsy experience at an academic medical center. JRSM Short Rep 2013;4:2042533313489824.
  4. Borohovitz A, Weinberg MD, Weinberg I. Pulmonary embolism: care standards in 2018. Prog Cardiovasc Dis 2018;60:613-21.
  5. Konstantinides SV, Torbicki A, Agnelli G, Danchin N, Fitzmaurice D, Galie N, et al. Task Force for the Diagnosis and Management of Acute Pulmonary Embolism of the European Society of Cardiology (ESC). 2014 ESC guidelines on the diagnosis and management of acute pulmonary embolism. Eur Heart J 2014;35:3033-69.
  6. Qaddoura A, Digby GC, Kabali C, Kukla P, Zhan ZQ, Baranchuk AM. The value of elec- trocardiography in prognosticating clinical deterioration and mortality in acute pul- monary embolism: a systematic review and meta-analysis. Clin Cardiol 2017;40: 814-24.
  7. Rodger M, Makropoulos D, Turek M, Quevillon J, Raymond F, Rasuli P, et al. Diagnos- tic value of the electrocardiogram in suspected pulmonary embolism. Am J Cardiol 2000;86:807-9.
  8. Digby GC, Kukla P, Zhan ZQ, Pastore CA, Piotrowicz R, Schapachnik E, et al. The value of electrocardiographic abnormalities in the prognosis of pulmonary embolism: a consensus paper. Ann Noninvasive Electrocardiol 2015;20:207-23.
  9. Ermis N, Ermis H, Sen N, Kepez A, Cuglan B. QT dispersion in patients with pulmo- nary embolism. Wien Klin Wochenschr 2010;122:691-7.
  10. Stein PD, Matta F, Sabra MJ, Treadaway B, Vijapura C, Warren R, et al. Relation of electrocardiographic changes in pulmonary embolism to right ventricular enlarge- ment. Am J Cardiol 2013;112:1958-61.
  11. Adult electrocardiography, chapter 1; normal electrocardiogram: origin and descrip- tion. In: Surawicz B, Knilans TK, editors. Chou’s electrocardiography in clinical prac- tice: adult and pediatric. 6th ed. Philadelphia: Saunders, Elsevier Inc.; 2008. p. 9-10.
  12. Meyer G. Effective diagnosis and treatment of pulmonary embolism: improving pa- tient outcomes. Arch Cardiovasc Dis 2014;107:406-14.
  13. Sinha N, Yalamanchili K, Sukhija R, Aronow WS, Fleisher AG, Maguire GP, et al. Role of the 12-lead electrocardiogram in diagnosing pulmonary embolism. Cardiol Rev 2005;13:46-9.
  14. Sreeram N, Cheriex EC, Smeets JL, Gorgels AP, Wellens HJ. Value of the 12-lead elec- trocardiogram at hospital admission in the diagnosis of pulmonary embolism. Am J Cardiol 1994;73:298-303.
  15. Punukollu G, Gowda RM, Vasavada BC, Khan IA. Role of electrocardiography in iden- tifying right ventricular dysfunction in acute pulmonary embolism. Am J Cardiol 2005;96:450-2.
  16. Yoshinaga T, Ikeda S, Nishimura E, Shioguchi K, Shikuwa M, Miyahara Y, et al. Serial changes in negative T wave on electrocardiogram in acute pulmonary thromboem- bolism. Int J Cardiol 1999;72:65-72.
  17. Ferrari E, Imbert A, Chevalier T, Mihoubi A, Morand P, Baudouy M. The ECG in pul- monary embolism. Predictive value of Negative T waves in precordial leads-80 case reports. Chest 1997;111:537-43.
  18. Kosuge M, Kimura K, Ishikawa T, Ebina T, Hibi K, Tsukahara K, et al. Prognostic sig- nificance of inverted T waves in patients with acute pulmonary embolism. Circ J 2006;70:750-5.
  19. Stein PD, Dalen JE, McIntyre KM, Sasahara AA, Wenger NK, Willis 3rd PW. The elec- trocardiogram in acute pulmonary embolism. Prog Cardiovasc Dis 1975;17:247-57.
  20. Brown G, Hogg K. Best evidence topic report. diagnostic utility of electrocardiogram for diagnosing pulmonary embolism. Emerg Med J 2005;22:729-30.
  21. Raghav KP, Makkuni P, Figueredo VM. A review of electrocardiography in pulmonary embolism: recognizing pulmonary embolus masquerading as ST-elevation myocar- dial infarction. Rev Cardiovasc Med 2011;12:157-63.
  22. Burch GE, Depasquale NP. The electrocardiographic diagnosis of pulmonary heart disease. Am J Cardiol 1963;11:622-38.
  23. Selvester RH, Rubin HB. New criteria for the electrocardiographic diagnosis of em- physema and cor pulmonale. Am Heart J 1965;69:437-47.
  24. Shmock CL, Pomerantz B, Mitchell RS, Pryor R, Maisel JC. The electrocardiogram in emphysema with and without chronic airways obstruction. Chest 1971;60:328-34.
  25. Silver HM, Calatayud JB. Evaluation of QRS criteria in patients with chronic obstruc- tive pulmonary disease. Chest 1971;59:153-9.
  26. Kukla P, McIntyre WF, Fijorek K, Dlugopolski R, Mirek-Bryniarska E, Bryniarski KL, et al. T-wave inversion in patients with acute pulmonary embolism: prognostic value. Heart Lung 2015;44:68-71.

Leave a Reply

Your email address will not be published. Required fields are marked *