Fibrinolysis and clinical outcomes in acute pulmonary embolism. Madras medical college pulmonary embolism (M-PER) registry from India

Abstract

Background

Acute pulmonary embolism (APE) is the third most common cause of vascular death. Data on APE from India and other low-and middle-income countries is sparse.

Objectives

Study the clinical characteristics, prognostic factors, in-hospital mortality (IMH) and 12 months mortality of patients with APE in India.

Methods

We prospectively enrolled 186 consecutive patients diagnosed with APE between November 2016 and November 2021 in Madras Medical College Pulmonary Embolism Registry (M-PER). All patients had electrocardiography and echocardiography. High risk patients and selected intermediate risk patients underwent fibrinolysis.

Results

75 % of our patients were below 50 years of age. 35 % were women. The mean time to presentation from symptom onset was 6.04 ± 10.01 days. 92 % had CT pulmonary angiography. Intermediate risk category (61.3 %) was the more common presentation followed by high risk (26.9 %). Electrocardiography showed S1Q3T3 pattern in 56 %. 76 % had right ventricular dysfunction and 12.4 % had right heart thrombi(RHT) by echocardiography. 50.5 % received fibrinolysis. Patients with RHT received fibrinolysis more frequently (78.3 % vs 46.6 %; p = 0.007). In-hospital mortality (IHM) was 15.6 %. Systemic arterial desaturation and need for mechanical ventilation independently predicted IHM. Ten patients (5.3 %) were lost to follow up. One year mortality was 26.7 % (47/176). One year mortality of patients discharged alive was similar among high, intermediate and low risk groups(14.8 % vs 1.9 % vs 10.5 %; p = 0.891).

Conclusions

Patients with PE are often young and present late in India. The in-hospital and 12 months mortality were high. Low and intermediate risk groups had a high post discharge mortality similar to high risk patients.

Graphical abstract

Main findings of Madras Medical College Pulmonary embolism Registry.

Abbreviations

PE	Pulmonary Embolism
APE	Acute Pulmonary Embolism
DVT	Deep Vein thrombosis
CTPA	Computerized tomography pulmonary angiogram
CTEPH	Chronic thromboembolic pulmonary hypertension
PESI	Pulmonary Embolism Severity Index
IHD	Ischemic Heart Disease
IHM	In Hospital Mortality
RHT	Right Heart Thrombus
PERT	Pulmonary Embolism Response Team

1. Introduction

Pulmonary embolism (PE), a potentially lethal form of Venous thromboembolism (VTE) continues to be an important cause of mortality and morbidity across the world, despite advances in diagnosis and treatment. It is the third most common cause of vascular death after myocardial infarction and stroke and is a leading preventable cause of death in hospitalized patients.^1,2 The reported annual incidence rates of PE is approximately 39–115 per 100,000 inhabitants.³ The actual figure could be still higher because of underreporting and undetected silent PE in patients with deep vein thrombosis (DVT).⁴ Studies have shown that annual hospitalization for Acute PE(APE) is on the increase.⁵ Approximately one-third of patients die from APE within the first hours of presentation, often before the diagnosis can be confirmed and therapy initiated or because the diagnosis was overlooked.⁶ Mortality rate of treated PE ranges from 3 to 8 %, but increases to 30 % in untreated PE.⁷ Age-standardized annual PE-related mortality rates in Europe has decreased linearly from 12.8 to 6.5 deaths per 100,000 between 2000 and 2015(8).

In contrast to the relatively large amount of data on arterial thromboses, data on VTE is sparse, even though they substantially contribute to the global burden of disease.⁹ The majority of the data on the incidence and mortality of VTE comes from high-income countries, especially North America, Western Europe, Australasia, and Southern Latin America. VTE has traditionally been considered to be rare in Asia. This may be attributed to a paucity of published literature on VTE from Asian countries. The data available from India is limited, except for few small studies, case series and retrospective studies.10, 11, 12, 13 We therefore sought to study the clinical characteristics, prognostic factors, in-hospital mortality and one year mortality of patients with APE in India.

2. Methods

2.1. Ethical considerations

The study protocol was designed in accordance with the Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments involving humans and was approved by the “Institutional Ethics committee” of Madras Medical College (Approval number: 01012017/ECR/270/Inst.TN/2013). All patients gave written informed consent in vernacular language. Participants were made aware that participation in the study is voluntary and that refusal to enrol in the study, will not affect the service they obtained.

2.2. Study setting and design

We conducted this prospective observational hospital based study [M-PER – Madras medical college pulmonary embolism registry] at Madras Medical College, the oldest public teaching hospital in South India. We enrolled all consecutive patients above 18 years diagnosed with APE admitted to the intensive care unit of our institution from October 2016 to November 2021.

2.3. Clinical evaluation

All patients underwent a detailed clinical evaluation. Baseline variables collected include age, gender, active malignancy (newly diagnosed or undergoing treatment), recent hospitalisation <4 weeks, recent trauma, recent immobilisation ≥4 days within 2 months, recent major surgery within 2 months, coronary and lifestyle risk factors (hypertension, diabetes, obesity, smoking and alcohol use) and other comorbid conditions (chronic kidney disease, chronic obstructive pulmonary disease, heart failure, previous VTE).

Presenting symptoms (chest pain, dyspnoea, syncope, cough, haemoptysis, leg swelling) and their duration, New York heart association functional status at admission, cardiovascular vitals (heart rate, blood pressure, respiratory rate, oxygen saturation), pedal oedema, varicose veins and any evidence suggesting deep venous thrombosis were also captured. Well's score, Pulmonary Embolism Severity Index (PESI) and simplified PESI (sPESI) scores were calculated for all the patients.

2.4. Electrocardiography and echocardiography

All patients had 12-lead electrocardiogram (ECG) and detailed echocardiography on admission and as necessary subsequently. Presence of sinus tachycardia, S1Q3T3 pattern, QRS axis, bundle branch blocks, right ventricular strain pattern were specifically looked for and documented. Right ventricular dilatation, right ventricular hypokinesia, tricuspid annular plane systolic excursion, right ventricular S prime, tricuspid regurgitation pressure gradient, pulmonary acceleration time and left ventricular ejection fraction were measured according to standard guidelines.¹⁴

2.5. Other investigations

Routine biochemical investigations including, blood sugar, cholesterol, blood urea, serum creatinine, complete blood count, liver function tests, serum electrolytes were done. Cardiac troponins, computerized tomography pulmonary angiogram (CTPA) and venous doppler evaluation of the lower limbs were done whenever possible.

2.6. Diagnosis and definitions

Diagnosis of APE was made with suggestive clinical features and supportive evidence from either CTPA or echocardiogram. Patients were classified as having as suggested by European Society of Cardiology guidelines into high risk, intermediate risk and low risk PE based on hemodynamic status, sPESI score, echocardiogram and cardiac troponins.³ Cardiogenic shock was defined as an admission BP of ≤90 mm of Hg with signs of systemic hypoperfusion. Thrombi in the main pulmonary artery trunk and/or in the right or left pulmonary arteries constituted central PE. Large embolus overlying the pulmonary trunk bifurcation was labelled as saddle embolus.

2.7. Treatment

All patients received anticoagulation with unfractionated heparin with overlap long term therapy with vitamin K antagonists. High-risk APE patients were considered for fibrinolysis with either streptokinase or Tenecteplase. Fibrinolysis in the intermediate risk group was decided based on elevated cardiac troponins, echocardiographic right heart dysfunction, suboptimal response to anticoagulation/clinical deterioration as assessed by tachycardia, tachypnea and systemic desaturation.

2.8. Follow up

Discharged patients were on regular follow-up every 3 months for 12 months, and twice a year thereafter. Clinical status of follow-up defaulters was collected by telephonic calls and home visits.

2.9. Outcomes

Primary outcome was in-hospital mortality (IHM). Secondary outcome was all-cause mortality at 12 months.

Funding

Our study was an investigator initiated, non-funded research project.

3. Analysis

Differences in baseline demographics, clinical parameters, reperfusion therapy, in-hospital course and complications between the patients who died in-hospital and those who were discharged alive were analysed to identify predictors in IHM and 12 months mortality. Similar comparison was done between the three European Society of Cardiology (ESC) risk group categories. In addition, an analysis of 12 months mortality was done to identify predictors of post discharge mortality after excluding the patients who had IHM and those who were lost to follow up.

4. Statistical analysis

Categorical variables are presented as frequencies and percentages, and compared using Pearson's Chi-squared test or Fisher's Exact Test. Continuous variables are presented as mean with standard deviation (SD), median with interquartile range and compared using student's t-test (normal distribution) or Mann–Whitney test (non-normal). One way Anova was used to compare differences in the baseline characterises and outcomes among the three ESC risk group. Covariates with p < 0.10 on univariable analysis were planned to be included in the final multivariable model to identify predictors of mortality. Multivariable analysis was done with binary logistic regression and results expressed as odds ratio (OR) with confidence interval (CI) and p-values. Significance was assumed at a two-sided value of p < 0.05.

The Kaplan Meier method was used to analyse the difference in the 12 months survival between the comparison groups (ESC risk groups, lysis vs no lysis and right heart thrombus vs no right heart thrombus) with equality of distribution, tested by the log-rank test. Lost to follow up patients were censored either at discharge or at their last known follow up. Hazard ratios were calculated with Cox proportional hazard model at 12 months, with propensity score matching done to account for the potential confounders. The proportional hazards assumption was evaluated by examining log (‐log) survival curves. Analyses were performed using SPSS version 28.0 (SPSS for Mac, Version 28.0. Armonk, NY: IBM Corp Released 2021.

5. Results

5.1. Clinical presentation

186 patients diagnosed with APE at our institution between November 2016 to November 2021 were included in the study. Selected baseline characteristics including clinical features, investigations done and treatment received are shown in Table-1. Study population was young with 74.7 % of patients ≤50 years of age (mean age 42.6 ± 13.3 years). 34.9 % were women. A third were alcohol users and one fourth were current smokers. Breathlessness (97.6 %) was the commonest symptom followed by chest pain (36.9 %). Distribution of NYHA symptom class, ESC risk groups, PESI and sPESI scores are shown in Fig. 1 A to 1D. Most patients had NYHA III symptoms at presentation (50 %). The mean time to presentation from symptom onset was 6.04 ± 10.01 days. Salient ECG and echocardiographic features are presented in Fig. 1E and F. Sinus tachycardia (82 %) was the most common ECG finding, while S1Q3T3 pattern was observed in 56 %. 93 % had dilated right heart, 76 % had right ventricular dysfunction and 12.4 % had echocardiographic thrombus in the right heart. Of the 169 (90.8 %) patients who had CTPA, 55.6 % (94/169) had thrombi in the central pulmonary arteries. Intermediate risk (114; 61.3 %) was the predominant mode of presentation of APE in our study, followed by high-risk(50; 26.9 %). Baseline features according to the ESC risk group at presentation are show in Table 2. History of immobilization was more common in the high-risk group, while previous PE/DVT was more common in the low-risk group. Dyspnea, tachycardia, tachypnea, hypotension and systemic desaturation were more commonly observed in the high-risk group. S1Q3T3 pattern in ECG, echocardiographic RV dilatation and RV dysfunction were seen more often in the higher risk group. Use of thrombolytic therapy, use of mechanical ventilator and use of vasopressor agents were significantly more in high-risk group compared to the other two groups.

Table 1.

Baseline demographic and clinical features, investigations and management.

Baseline features	N	%	Mean(±SD)
Age			42.31 ± 13.45
Female	65	34.9 %
Risk factors and comorbid conditions
Current smoking	47	25 %
Alcohol use	63	33.9 %
Obesity	14	7.5 %
Immobilisation	26	14 %
Recent surgery	19	10 %
Prior history of pulmonary embolism (PE)	6	3 %
Prior history of deep vein thrombosis (DVT)	20	10.7 %
Previous history of both DVT and PE	7	3.8 %
Trauma	15	8 %
Active malignancy	9	4.8 %
Diabetes mellitus	32	17 %
Systemic hypertension	19	10 %
Chronic obstructive pulmonary disease	6	3 %
Coronary artery disease	7	3.76 %
Chronic kidney disease	1	0.5 %
COVID related (4 Post COVID, 1 during COVID)	5	2.7 %
Pregnancy (antenatal-2, postnatal-4)	6	3.0 %
Other hypercoagulable states	11	5.9 %
Signs and symptoms
Breathlessness	182	97.8 %
Chest pain	68	36.6 %
Syncope	28	15 %
Cough	61	32.8 %
Pedal edema	68	36.6 %
Haemoptysis	22	11.8 %
Pulse(rate/min)			112.46 ± 15.434
Pulse rate >100/min	152	81.7 %
Systolic blood pressure ≤90mmhg	50	26.9 %
Respiratory rate/min			29.02 ± 5.878
Pulse Oximetry saturation <90 %	46	24.7 %
Varicose vein	9	4.8 %
Deep vein thrombosis - Clinical	53	28.5 %
Well's Score ≥5 (Pulmonary embolism likely)	171	92 %
Well's score 0–4 (Pulmonary embolism less likely)	15	8 %
Tricuspid annular plane systolic excursion (TAPSE)			13.41 ± 3.57 mm
Troponin
Positive	35	18.8 %
Negative	85	45.7 %
Not done	66	35.5 %
Computerised Tomographic Pulmonary angiography
Central	94	50.5 %
Peripheral	75	40.3 %
Saddle thrombus	17	9.1 %
Not done	17	9.1 %
Doppler study of leg veins
Deep vein thrombosis present	75	40.3 %
Deep vein thrombosis not present	78	41.9 %
Doppler study - not done	33	17.7 %
Treatment
Fibrinolysis	94	50.5 %
Vasopressor use	31	16.7 %
Ventilator use	18	9.7 %

Fig. 1.

Baseline clinical status, electrocardiographic and echocardiographic changes observed in the MPER.

Legend: A-New York Heart Association (NYHA) class of symptoms at admission, B-European society of cardiology risk classification of admitted patients, C-Pulmonary Embolism Severity Index (PESI) score, D- Simplified PESI score, E− Electrocardiographic observations, F- Echocardiographic observations.

Table 2.

Baseline features according to ESC Risk Categories.

Characteristics No (%)	High Risk (n = 50)	Intermediate Risk (n = 114)	Low risk (n = 22)	p Value
Age	40.86 ± 13.72	43.17 ± 13.21	41.18 ± 14.32	0.552
Female (65)	20 (40 %)	40 (35.1 %)	5 (22.7 %)	0.366
Risk factors
Smoking (47)	12 (24 %)	31 (27.2 %)	4 (18.2 %)	0.653
Alcohol (63)	17 (34 %)	38 (33.3 %)	8 (36.4 %)	0.963
Obesity14	4 (8 %)	10 (8.8 %)	0 (0 %)	0.357
Diabetes32	11 (22 %)	18 (15.8 %)	3 (13.6 %)	0.559
Hypertension19	6 (12 %)	11 (9.6 %)	2 (9.1 %)	0.885
Immobilisation26	12 (24 %)	10 (8.8 %)	4 (18.2 %)	0.029
Surgery in past 4 weeks19	8 (16 %)	9 (7.9 %)	2 (9.1 %)	0.283
Trauma15	5 (10 %)	8 (7 %)	2 (9.1 %)	0.798
Active malignancy9	4 (8 %)	4 (3.5 %)	1 (4.5 %)	0.466
Previous PE/DVT33	4 (8.0 %)	22 (19.3 %)	7 (31.8 %)	0.040
Symptoms and Signs
Breathlessness (182)	50 (100 %)	113 (99.1 %)	19 (86.4 %)	≤0.001
Chest pain (68)	16 (32 %)	42 (36.8 %)	10 (45.5 %)	0.548
Cough (61)	14 (28 %)	40 (35.1 %)	7 (31.8 %)	0.669
Hemoptysis22	4 (8 %)	13 (11.4 %)	5 (22.7 %)	0.199
Syncope28	15 (30 %)	13 (11.4 %)	0 (0 %)	<0.001
Leg swelling (68)	17 (34 %)	41 (36 %)	10 (45.5 %)	0.635
Pulse (per minute)	121.36 ± 13.56	111.9 ± 14.01	95.14 ± 10.36	≤0.001
Respiratory rate/min	31.88 ± 6.13	28.83 ± 5.22	23.50 ± 4.27	≤0.001
Oxygen saturation	87.54 ± 8.25	92.86 ± 5.35	96.27 ± 1.35	<0.001
Systolic BP(mm hg)	81.80 ± 8.96	117.19 ± 15.6	123.45 ± 23.25	≤0.001
Cardiogenic shock (50)	50 (100 %)	0 (0 %)	0 (0 %)	<0.001
Wells score	97.62 mean	92.13 mean	91.23 mean	0.808
ECG
Sinus tachycardia (152)	49 (98 %)	96 (84.2 %)	7 (31.8 %)	<0.001
S1Q3T3 (104)	32 (64 %)	68 (59.6 %)	32 (18.2 %)	<0.001
RBBB (60)	21 (42 %)	36 (31.6 %)	3 (13.6 %)	0.058
ECHO
RA/RV dilation (172)	50 (100 %)	109 (95.6 %)	13 (59.1 %)	<0.001
TR pressure gradient	43.62 ± 13.18	42.50 ± 19.67	36.23 ± 27.13	0.302
RV dysfunction (141)	48 (96 %)	93 (81.6 %)	0 (0 %)	<0.001
Right heart thrombus (RA/RV)23	4 (8 %)	19 (16.7 %)	0(0 %)	0.052
Clot in PAs (57)	20 (40 %)	33 (28.9 %)	4 (18.2 %)	0.148
Saddle thrombus17	7 (14 %)	8 (7 %)	2 (9.1 %)	0.360
DVT present (53)	12 (24 %)	33 (28.9 %)	8 (36.4 %)	0.555
PESI score (median 85)	148.23 mean	80.42 mean	36.89 mean	<0.001
Fibrinolysis (94)	46 (92 %)	48 (42.1 %)	0 (0 %)	<0.001
Ventilator18	16 (32 %)	1 (0.9 %)	1 (4.5 %)	<0.001
Vasopressor31	31 (62 %)	0 (0 %)	0 (0 %)	<0.001

5.2. Treatment

All patients received intravenous heparin. Half of the patients (94/196; 50.5 %) received systemic fibrinolysis. 92.4 % (46/50) of the high-risk patients and 42 % (48/114) of the intermediate risk patients received fibrinolysis (Table 2). Among the intermediate risk group, 70 % (14/20) of the intermediate-high risk group and 27 % (17/63) of the intermediate-low risk group received fibrinolysis (Fig. 3B). Streptokinase (88/94; 93.6 %) was the predominant lytic agent used. One patient underwent catheter directed thrombolytic therapy. One patient underwent successful surgical embolectomy. All patients discharged alive were sent on oral anticoagulants.

Fig. 3.

A. Benefit of thrombolysis in acute pulmonary embolism according to the ESC risk groups

Legend: Figure on the left shows the total number of patients enrolled in each group and on the right display the mortality in percentage in each of these groups.

B: Benefit of thrombolysis in acute pulmonary embolism in the ESC intermediate risk subgroups – high and low.

5.3. In hospital mortality (IHM)

IHM was 15.6 % (29/186). More patients in the high-risk group had in-hospital mortality compared to the intermediate and low risk groups (44 % vs 5.3 % vs 4.5 %; p < 0.001) (Table-3). IHM was similar in men and women (Table 4a). Active malignancy, poor NYHA class at presentation, tachycardia, tachypnea, hypotension, systemic desaturation, hypotension and cardiogenic shock, RV dysfunction in echocardiogram, high PESI and sPESI scores, use of vasopressors and mechanical ventilation were the univariate predictors of IHM. However, in the multivariable analysis, only low systemic arterial oxygen saturation and use of mechanical ventilation predicted IHM (Table 4b).

Table 3.

Mortality analysis among three ESC risk groups.

Mortality Analysis	Total	High Risk	Intermediate Risk	Low risk	p Value
In-hospital	29/186 (15.6 %)	22/50 (44 %)	6/114 (5.3 %)	1/22 (4.5 %)	<0.001
12 months	47/176 (26.7 %)	26/49 (53.1 %)	18/107 (16.8 %)	3/41 (7.3 %)	<0.001
Post discharge	18/147 (12.2 %)	4/27 (14.8 %)	12/101 (11.9 %)	2/19 (10.5 %)	0.891

Table 4a.

Predictors of in-hospital mortality - Univariate analysis (n = 186).

Characteristics No (%)	Total (n = 186)	Live (n = 157)	Death n = 29 (15.6 %)	p Value
Age	42.31 ± 13.45	42.62 ± 13.35	40.62 ± 14.13	0.463
Sex
Female	65 (34.9 %)	51 (32.5 %)	14 (48.3 %)	0.101
Male	121 (65.1 %)	106 (67.5 %)	15 (51.7 %)
Risk factors
Smoking	47(25.3 %)	39(24.8 %)	8(27.6 %)	0.755
Alcohol	63(33.9)	54(34.4 %)	9 (31 %)	0.725
Obesity	14(7.5 %)	12(7.6 %)	2(6.9)	0.889
Diabetes	32(17.2 %)	26(16.6 %)	6(20.7 %)	0.588
Hypertension	19(10.2 %)	17(10.8 %)	2(6.9 %)	0.521
Immobilisation	26(14 %)	19(12.1 %)	7(24.1 %)	0.086
Surgery in past 4 weeks	19(10.2 %)	15(9.6 %)	4(13.8 %)	0.489
Trauma	15(8.1 %)	13(8.3 %)	2(6.9 %)	0.801
Active malignancy	9(4.8 %)	5(3.2 %)	4(13.8 %)	0.014
Previous PE/DVT	33(17.7 %)	29(18.5 %)	4(13.8 %)	0.545
Symptoms and Signs
Dyspnea
NYHA I	4 (2.2 %)	4 (2.5 %)	0 (0 %)
NYHA II	16 (8.6 %)	16 (10.2 %)	0 (0 %)
NYHA III	93 (50 %)	87 (55.4 %)	6 (20.7 %)
NYHA IV	73 (39.2 %)	50 (31.8 %)	23 (79.3 %)	<0.001
Duration of Symptoms (days)	6.04 ± 10.01	6.47 ± 10.68	3.9 ± 3.949	0.203
Chest pain	68(36.6 %)	62(39.5 %)	6(20.7 %)	0.053
Cough	61(32.8 %)	53(33.8 %)	8(27.6 %)	0.515
Hemoptysis	22(11.8 %)	20(12.7 %)	2(6.9 %)	0.371
Syncope	28(15.1 %)	24(15.3 %)	4(13.8 %)	0.836
Leg swelling	68(36.6 %)	61(38.9 %)	7(24.1 %)	0.131
Clinical DVT	53(28.5 %)	47(29.9 %)	6(20.7 %)	0.311
Pulse (per minute)	112.5 ± 15.4	111.2 ± 15.3	119.5 ± 14.9	0.007
Respiratory rate/min	29.0 ± 5.9	28.5 ± 5.6	32.1 ± 6.6	0.002
Oxygen saturation	91.8 ± 6.6	92.9 ± 5.3	86.2 ± 9.9	<0.001
Systolic blood pressure	10.8.4 ± 22.3	112.3 ± 20.6	87.2 ± 18.9	<0.001
Cardiogenic shock	50 (26.9 %)	28(17.8 %)	22(75.9 %)	≤0.001
Wells score >4 (High probability)	171(91.9 %)	143(91.1 %)	28(96.6 %)	0.320
Total hospitalisation in days	5.61 ± 3.36	5.72 ± 2.35	5.03 ± 6.57	0.583
ECG
Sinus tachycardia	152(81.7 %)	125(79.6 %)	27(93.1 %)	0.084
S1Q3T3	104(55.9 %)	86(54.8 %)	18(62.1 %)	0.467
RBBB	60(32.3 %)	49(31.2 %)	11(37.9 %)	0.477
ECHO
RA/RV dilation	172(92.5 %)	143(91.1 %)	29(100 %)	0.094
RV dysfunction	141(75.8 %)	114(72.6 %)	27(93.1 %)	0.018
Clot in RA/RV/PA	70 (37.6 %)	60 (38.2 %)	10 (34.5 %)	0.432
Right heart thrombus (RA/RV)	23 (12.4 %)	21 (13.4 %)	2 (6.9 %)	0.330
Clot in MPA/LPA or RPA	57 (30.6 %)	48 (30.6 %)	9 (31 %)	0.961
Saddle thrombus	17(9.1 %)	16(10.2 %)	1(3.4 %)	0.247
PESI score	85.8 ± 30.6	81.2 ± 28.2	110.9 ± 31.6	<0.001
S-PESI score ≥1	138 (74.2 %)	112 (71.3 %)	26 (89.7 %)	0.038
Fibrinolysis	94 (50.5 %)	72 (45.9 %)	22 (75.9 %)	0.003
Ventilator	18(9.7 %)	3(1.9 %)	15(51.7 %)	<0.001
Vasopressor	31(16.7 %)	13(8.3 %)	18(62.1 %)	<0.001

Table 4b.

Predictors of in hospital mortality: Multivariable Analysis (n = 186).

Variables	In hospital Death
	Exp (B)	CI (95 %)	p Value
Active malignancy	3.876	0.306-49.11	0.296
NYHA class	2.417	0.774-7.549	0.129
Pulse rate	1.008	0.965-1.042	0.730
Systolic blood pressure	1.034	1.000–1.069	0.053
Respiratory rate	0.951	0.842-1.074	0.418
Low systemic oxygen saturation	1.096	1.024-1.192	0.010
Right ventricular dysfunction	1.289	0.227-7.318	0.774
Ventilator use	23.100	4.477-119.17	<0.001

5.4. 12 months mortality

12 months follow up data was available for 176 of 186 patients (94 %). 18 patients had died during 12 months follow up. All-cause mortality at 12 months was 26.7 % (47/176). The univariable and multivariable predictors of 12 months mortality, were almost similar to the predictors of in-hospital mortality(Table 5a, Table 5bb). Presence of RHT and fibrinolysis did not influence the survival at 12 months. The high-risk group has a significantly higher likelihood of dying at 12 months since admission compared to the low-risk group (HR 5.14 (95 % CI 1.56–16.99; p < 0.001) after adjusting for the baseline risk factors and treatment allotment (Figure-2). Among the patients who were discharged alive, those who died during the follow up, had higher mean age compared to the survivors (41.25 ± 12.4 Vs 49.83 ± 15.4; p < 0.00 9) (Supplementary Table-1). Age was the only predictor of post discharge mortality. Among the three ESC risk groups, mortality was not significantly different in the patients discharged alive (14.8 %, 11.9 % & 10.5 %, p = 0.89) (Table-3)

Table 5a.

Predictors of mortality at 12 months - Univariable analysis (n = 176).

Characteristics No (%)	Total (n = 176)	Live (n-129)	Death (n = 47)	p Value
Age	42.02 ± 13.27	41.25 ± 12.48	44.15 ± 15.17	0.20
Female	60(34.1 %)	40 (31.0 %)	20(42.6 %)	0.15
Risk factors
Smoking	42(23.9 %)	30(23.3 %)	12(25.5 %)	0.75
Alcohol	59(33.5 %)	46(35.7 %)	13(27.7 %)	0.32
Obesity	14(8 %)	9(7 %)	5(10.6 %)	0.42
Diabetes	30(17 %)	19(14.7 %)	11(23.4 %)	0.17
Hypertension	19(10.8 %)	13(10.1 %)	6(12.8 %)	0.67
Immobilisation	24(13.6 %)	14(10.9 %)	10(21.3 %)	0.07
Surgery in past 4 weeks	17(9.7 %)	12(9.3 %)	5(10.6 %)	0.79
Trauma	14(8 %)	10(7.8 %)	4(8.5 %)	0.86
Active malignancy	8(4.5 %)	4(3.1 %)	4(8.5 %)	0.12
Previous PE/DVT	33(18.8 %)	26(20.2 %)	7(14.9 %)	0.42
Symptoms and Signs
Breathlessness	172(97.7 %)	125(96.9 %)	47(100 %)	0.22
Duration of Symptoms (days)	6.06 ± 10.24	5.2 ± 6.02	8.4 ± 17.04	0.213
Chest pain	67 (38.1 %)	56 (43.4 %)	11 (23.4 %)	0.016
Cough	58(33 %)	40(31 %)	18(38.3 %)	0.36
Hemoptysis	22(12.5 %)	18(14.0 %)	4(8.5 %)	0.33
Syncope	26(14.8 %)	18(14 %)	8(17 %)	0.61
Leg swelling	66(37.5 %)	50(38.8 %)	16(34 %)	0.60
Clinical DVT	52(29.5 %)	41(31.8 %)	11(23.4 %)	0.28
Pulse (per minute)	112.85 ± 15.33	111.16 ± 15.218	119.52 ± 14.916	0.007
Respiratory rate >20/min	158(89.8 %)	112(86.8 %)	46(97.9 %)	0.032
Oxygen saturation <90 %	60(34.1 %)	36(27.9 %)	24(51.1 %)	0.004
Cardiogenic shock	49(27.8 %)	23(17.8 %)	26(55.3 %)	≤0.001
Wells score >4	161(91.5 %)	116(89.9 %)	45(95.7 %)	0.36
ECG
Sinus tachycardia	145(82.4 %)	103(79.8 %)	42(89.4 %)	0.14
S1Q3T3	100(56.8 %)	74(57.4 %)	26(55.3 %)	0.80
RBBB	57(32.4 %)	40(31.0 %)	17(36.2 %)	0.51
ECHO
RA/RV dilation	162(92.0 %)	115(89.1 %)	47(100 %)	0.019
RV dysfunction	134(76.1 %)	93(72.1 %)	41(87.2 %)	0.03
Right Heart thrombi (RA/RV)	20(11.4 %)	15(11.6 %)	5(10.6 %)	0.85
Clot in PAs	54(30.7 %)	40(31 %)	14(29.8 %)	0.87
Saddle thrombus	17(9.1 %)	16(10.2 %)	1(3.4 %)	0.247
PESI Score	86.17 ± 30.84	80.32 ± 29.03	102.23 ± 30.22	<0.001
S-PESI score ≥1	130(73.9 %)	90(69.8 %)	40(85.1 %)	0.040
Fibrinolysis	90(51.1 %)	60(46.5 %)	30(63.8 %)	0.042
Ventilator	18(10.2 %)	3(2.3 %)	15(31.9 %)	<0.001
Vasopressor	30(17 %)	11(8.5 %)	19(40.4 %)	<0.001

Table 5b.

Predictors of mortality at 12 months - Multivariable analysis (n = 176).

Variables	1 year follow up death
	Exp (B)	CI (95 %)	p Value
Pulse rate	1.013	0.982-1.045	0.421
Systolic blood pressure	1.011	0.990-1.031	0.308
Respiratory rate	0.974	0.893-1.061	0.540
Oxygen saturation	1.080	1.010-1.155	0.024
Right ventricular dysfunction	1.266	0.431-3.714	0.668
Ventilator use	14.025	3.302-59.564	<0.001

Fig. 2.

Survival analysis at 12 months.

Legend: A-Kaplan-Meier curves for overall 12 months survival for all patients, B-Cox-regression analysis for prediction of mortality according to ESC risk classification, adjusted for active malignancy and thrombolytic status, C-Cox-regression analysis for prediction of mortality according to thrombolytic status, adjusted for baseline risk profile including ESC risk group stratification, D-Cox-regression analysis for prediction of mortality according to presence of right heart thrombus (RHT), adjusted for baseline risk profile including ESC risk group stratification and thrombolytic status.

5.5. Thrombolytic therapy and mortality

In the logistic regression analysis, after adjusting for the confounding factors, thrombolytic therapy was not associated with mortality benefit (adjusted OR = 0.820; 95 % CI 0.912–3.492; p = 0.79). Benefit of thrombolytic therapy was separately analyzed in high and intermediate ESC risk groups. There was no significant improvement in mortality in either high risk or the intermediate risk or both the groups put together (Fig. 3A). 83 of the 114 intermediate risk group patients, could be subclassified into intermediate-high and intermediate-low risk groups based on availability of serum troponin levels and RV function evaluation by imaging. There was no benefit of thrombolytic therapy in either of these two groups (Fig. 3B). Thrombolytic therapy during index hospitalization, was not associated with any mortality benefit in the 12 months follow up (Supplementary Table-2a). Among the intermediate risk patients who were discharged alive, fibrinolysed patients had a lower mortality (9.5 % vs 13.5 %; p = 0.536), at 12 months, which however was not statistically significant (Supplementary Table-2b).

5.6. Thrombi and bleeding

Echocardiography showed clot in the central pulmonary arteries in 57 (30.6 %) of patients of whom 17 (9.1 %) had saddle thrombus. 23 (12.4 %) had thrombi in the right heart of whom 4 were in high-risk group and 19 in intermediate risk group. The proportion of patients who received lysis was higher in patients with RHT compared to those without RHT (78.3 % vs 46.6 %; p = 0.007). However, neither the presence of thrombi in the pulmonary artery nor RHT predicted mortality (Table 4a, Table 4b, Table 5a, Table 5b). Minor bleeding was observed in 4 %, while major bleed requiring blood transfusion in one patient.

5.7. COVID-19 and in other hypercoagulable states

The COVID-19 pandemic, spanning from April 2020 to November 2021, accounted for 32.8 % (20 out of 61 months) of the total enrolment duration. However, only 18.3 % (34 out of 186) of the total patients were enrolled during this period, indicating a lower enrolment rate. Furthermore, the IHM of the patients enrolled during the pandemic period was high compared to those enrolled in the pre-pandemic period [23.5 % (8/34); vs 13.8 % (21/152)]. However, this difference was not statistically significant (p < 0.190). Five patients developed APE within 6 weeks of COVID-19 infection. Three of them were lysed, of whom 2 had IHM.

Six APEs occurred around pregnancy (two antenatal and four postnatal). One antenatal APE was successfully thrombolysed. The other antenatal patient had IHM. All four postnatal APE patients (3 lysed, 1 not lysed) survived and were discharged alive. 11 patients had thrombophilia (5 APLA positive, 3 hyper-homocysteinemia) of whom 5 were given fibrinolysis. There were 2 IHD one each in the lysed and non-lysed group.

6. Discussion

To the best of our knowledge, this is the largest prospective study on APE from this part of the world, evaluating the clinical outcomes, outcome predictors with one year follow up. Our study has the following key findings.

6.1. Young age of onset

APE develops at an earlier age in India than in western population, affirming the view that it should not be considered a disease of the old age.⁹ The mean age of our study population (42.6 years) was lower than the mean age (56–66 years) reported in the EMPEROR, ICOPER and RIETE registries.^15,16 Whether the reported higher atherosclerotic cardiovascular disease risk, atherogenic dyslipidaemia and metabolic syndrome in South Asians compared with other populations17, 18, 19 could play a role in venous thrombotic disorders as well, needs further evaluation.

6.2. High in-hospital mortality

The in-hospital mortality of 15.6 % is high. This could be because our study enrolled higher proportion of high risk patients. 26.9 % of our patients were hypotensive, compared to the 3.0 %, 4.2 % and 8.0 % in the EMPEROR, ICOPER and RIETE studies respectively.^15,16,20 This in-hospital mortality of the intermediate risk group (5.3 %) in our study is higher than reported in other studies, involving intermediate risk PE,^21,22 the reason for which is unclear. The mortality in our study remained high despite, immediate use of heparin in all our patients once APE is suspected. Early heparin use has been shown to be associated with a better clinical outcome.23, 24, 25

6.3. High post discharge mortality for low and intermediate risk APE

The 12 months mortality of low and intermediate risk APE patients discharged alive, was high similar to those of the high risk groups. This highlights the need for equally vigorous follow up care in low and intermediate risk patients similar to the high risk patients, focussing on drug compliance and risk factor modification.

6.4. Gender differences in enrolment and mortality

Most (65.1 %) of our patients were male, unlike the balanced gender distribution in most studies.^{13,15,16,21,26,27} Male preponderance has been reported from a retrospective registry from India.²⁸ Further studies are warranted to explore the possible gender differences in access to healthcare for patients with VTE in resource limited health care settings. However, we did not observe any gender differences in mortality as few studies have reported^8,29

6.5. Fibrinolysis and mortality

A high proportion of patients received fibrinolysis (92 % in high risk and 42 % in intermediate risk) in our study compared to other studies (2.4 % in EMPEROR and 13 % in ICOPER).^15,20 Guidelines have given a class I indication for lysis in APE based on the available literature^3,30, 31, 32 despite absence of large-scale mortality studies of thrombolytic therapy in APE. Fibrinolysis rates have been lower, even among high risk and hypotensive patients (11 % in RIETE registry, 31.7 % in ICOPER)^26,33 in earlier studies. However, despite high use of fibrinolysis, we did not observe any significant mortality benefit of fibrinolysis in APE. The lack of benefit could be due to the mixed group of APE with high, intermediate and low risk patients being included in the MPER. In addition, our study is not adequately powered to detect benefit of fibrinolysis. Further our patents presented with a median delay of 6 days after symptom onset, which is longer than generally reported in the literature.³⁴ The clinical presentation delay has been reported to be associated suboptimal benefit of fibrinolysis in APE.³⁵ Streptokinase (93.6 %) was the predominant lytic agent used in our study while others have used alteplase^10,15 and tenecteplase.^21,22

In a retrospective analysis of 140 patients with APE from India, Muralidharan et al reported a lower mortality rate among 25 out of 40 patients with massive APE who received fibrinolysis compared to the 15 patients with massive APE who did not receive fibrinolysis.¹¹ In our study, though high-risk APE patients, who underwent fibrinolysis had a lower IHM compared to those not fibrinolysed (41.3 % vs 75 %), this was not statistically significant, due to the smaller number of patients in the non-lysed group (4 of 50 patients). In a large nationwide study, in the United States of America, thrombolytic therapy was associated with significantly reduction in the all-cause case mortality in hemodynamically unstable patients with APE (15 % vs 47 %; p < 0.0001). However unlike our study, only 30 % of the unstable patients received fibrinolytic therapy,³⁶ highlighting the significant underusage of fibrinolysis. Our study did not show mortality benefit for fibrinolysis in the 103 intermediate risk group patients who had a 42 % fibrinolysis rate. This was true even in the 20 patients who could be sub-classified into intermediate high risk group. Earlier studies on intermediate risk PE patients also have shown similar lack of benefit with fibrinolysis.^21,22 Earlier studies have shown, that fibrinolytic therapy did not affect long-term mortality, residual dyspnoea or RV dysfunction in the intermediate high risk patients.³ However, it is possible that fibrinolytic therapy may still be beneficial in selected patients with intermediate risk PE, preferably administered in a lower dose to avoid bleeding complications, which is being currently tested in the PEITHO-3 trial.³⁷ The recommendation for thrombolytic therapy for intermediate risk PE, is still evolving, and more prospective data is needed.^3,38 Ultrasound-facilitated catheter-directed fibrinolysis (OPTALYSE PA trial) and Catheter-Directed Mechanical Thrombectomy (CDT) have been studied for Intermediate-risk APE with favourable initial results.^39,40

6.6. Right heart thrombi and clinical outcome

The reported incidence of RHT detected by echocardiography in unselected APE varies from 2.6 % to 18 %.41, 42, 43, 44 It has been also reported that right heart thrombi are an indicator of adverse outcome.^42,45 However in our study, the 23 patients (12.4 %) who had RHT had an outcome similar to those without RHT. Our study concurs with the earlier view that the mortality in APE is related more to clinical and hemodynamic impairment than to the presence of thrombi in right heart.^43,46

Our study has also underscored the significance of bedside oxygen saturation in independently predicting IHM in APE. This finding has a potential utility, especially in resource-limited settings.

6.7. Strengths and limitations

Our study has the strength of being the largest prospective study of APE from India with high rate of echocardiography and CT pulmonary angiography used in diagnosis. However out study has a number of limitations. Being a single-centre teaching hospital based study, enrolling patients arriving to the intensive care unit, the results of this study may not apply to other community based VTE management. We have not used low molecular weight heparins and direct oral anticoagulants in our center, in view of the limited and variable availability of these drugs. It is possible that mechanical circulatory support and catheter based therapies for APE could have altered the outcome in some of our cases. We did not have a routine echocardiographic or functional capacity assessment during follow up. So we could not comment on possible transition to chronic thromboembolic pulmonary hypertension during follow up. Further as we could not ascertain the cause of death for all patients who had died during follow up, particularly during the COVID pandemic, cardiovascular mortality during follow up could not be provided.

7. Conclusion

Patients with APE presenting in hospital in this South Indian city were young, present late after symptom onset, with a high in-hospital and 12 months mortality. Thrombolytic therapy was not beneficial in a broad unselected population of patients with APE. Presence of RHT was not associated with increased mortality. Intermediate and low risk patients who are discharged alive have a high mortality at 12 months comparable to that of high risk patients. Close follow up of patients with acute pulmonary embolism is necessary after discharge irrespective of the risk category during index hospitalisation.

What is already known in this subject
	•Thrombolytic therapy is recommended in high risk APE patients
	•Benefit of thrombolytic therapy in intermediate risk APE patients is unclear
What new information this study adds for Indian population
	•Routine thrombolytic therapy for all patients with intermediate risk APE, may not be beneficial
	•Presence of visible thrombus in right heart is not necessarily an indication of thrombolytic therapy
	•Low and intermediate risk patients with acute pulmonary embolism, have a high mortality risk at 12 months similar to high risk patients

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article: