The relation between international normalized ratio and mortality in acute pulmonary embolism: A retrospective study

Abstract

Background

Acute pulmonary embolism (PE) is a serious clinical disease characterized by a high mortality rate. The aim of this study was to assess the prognostic value of international normalized ratio (INR) in acute PE patients not on anticoagulant therapy.

Methods

The study included 244 hospitalized acute PE patients who were not receiving previous anticoagulant therapy. Based on their 30‐day mortality, patients were categorized as survivors or non‐survivors. INR was measured during the patients’ admission, on the same day as the diagnosis of PE but before anticoagulation started.

Results

Thirty‐day mortality occurred in 39 patients (16%). INR was higher in non‐survivors than in survivors (1.3±0.4 vs 1.1±0.3, P=.003). In multivariate analysis, INR (HR: 3.303, 95% CI: 1.210–9.016, P=.020) was independently associated with 30‐day mortality from PE. Inclusion of INR in a model with simplified pulmonary embolism severity index (sPESI) score improved the area under the receiver operating characteristics (ROC) curve from 0.736 (95% CI: 0.659‐0.814) to 0.775 (95% CI: 0.701‐0.849) (P=.028). Also, the addition of INR to sPESI score enhanced the net reclassification improvement (NRI=8.8%, P<.001) and integrated discrimination improvement (IDI=0.043, P=.027).

Conclusion

Elevated INR may have prognostic value for 30‐day mortality in acute PE patients not on anticoagulation. Combining INR with sPESI score improved the predictive value for all‐cause mortality. However, further large‐scale studies are needed to confirm it's prognostic role.

1. Introduction

Acute pulmonary embolism (PE) is a serious cardiovascular disease with high mortality and morbidity rates.1, 2 Identifying prognostic factors in patients with pulmonary embolism is very important, particularly when trying to determine therapy or predict outcomes.3 Although hemodynamic instability is the strongest predictor of mortality, there are many others, including electrocardiographic changes,4 echocardiographic parameters,5 computed tomographic pulmonary angiography (CTPA) findings,6 and increased levels of cardiac biomarkers,7 which are released in response to right ventricular overload.

Various types of scoring systems are used to predict mortality in acute PE patients. The simplified Pulmonary Embolism Severity Index (sPESI) score is most commonly used to identify high‐risk patients.8 However, this score does not take into account any hemostatic parameters. Prothrombin time (PT) international normalized ratio (INR) has been accepted as a sensitive, reliable marker of coagulation abnormalities. Previous studies have shown that PT, fibrin degradation products, and D‐dimer levels correlated well with mortality in patients with traumatic brain injury.9, 10 Furthermore, an increased INR is associated with poor clinical outcomes in critically ill patients with hypoxic hepatitis.11, 12 However, data on the prognostic utility of coagulation parameters in acute PE patients not on anticoagulant therapy are scarce. Wong et al.13 study is the only study to date demonstrating the prognostic significance of admission INR in acute PE patients, and did not include assessing white blood cell (WBC) and right ventricular dysfunction. The present study validates and also addresses potential mechanisms of elevated INR in acute PE.

2. Materials and Methods

This retrospective study analyzed 272 patients with acute PE diagnosed by CTPA in our hospital between April 2008 and January 2015. Exclusion criteria included patients under age 18, those who had received previous anticoagulant therapy (warfarin, enoxaparin, and/or novel oral anticoagulants, including dabigatran, rivaroxaban, and apixaban), patients with any underlying disease limiting their estimated life expectancy to less than 1 month (major trauma), those with hematological disorders, infectious or inflammatory diseases, or serious renal and liver diseases, patients who were documented to be on immunosuppressant therapy (including steroids), and those with incomplete documentation. Twenty‐eight patients were excluded from the study, as follows: two patients had end‐stage cancer, two had serious renal disease, three had a history of serious liver failure, one had septic shock, one was on immunosuppressant therapy, one had a hematological disorder, one had a closed traumatic brain injury, seven had incomplete data, and 10 had received anticoagulant therapy. The remaining 244 patients (146 patients were diagnosed with acute PE in Emergency Department, 88 patients were diagnosed with acute PE in Ward/Intensive Care Unit during hospital stay) composed the final study population. Data on clinical and demographic characteristics and laboratory parameters were collected from medical records. Patients were grouped as survivors and non‐survivors. The study protocol was approved by the local ethics committee (the number of the local ethics committee: 321).

2.1. Definitions

Thirty‐day mortality was defined as death from any cause within 30 days after the diagnosis of acute PE. Major bleeding was defined as a decline in hemoglobin level of 20 g/L or more or transfusion of 2 or more units of red cells. Altered mental status was defined as signs of disorientation, lethargy, stupor, or coma. Syncope was defined as a transient, self‐limited loss of consciousness due to transient global cerebral hypoperfusion characterized by rapid onset, short duration, and spontaneous complete recovery.14 All patients’ sPESI scores, including age, heart rate, systolic blood pressure, arterial oxyhemoglobin saturation, presence of cancer, chronic heart failure, and chronic pulmonary disease, were calculated for clinical risk stratification.8 Myocardial necrosis was defined as cardiac troponin (cTnI) elevation >0.06 ng/mL.

2.2. Echocardiography and computed tomography imaging

During patients’ initial evaluations, echocardiographic imaging was performed with the Vivid 3 system (Vivid 3; GE Heathcare,Vingmed Ultrasound, Horten, Norway). Right ventricular (RV) dimensions were measured from the apical four‐chamber view in diastole at the mid‐ventricular level. Systolic pulmonary arterial pressure (sPAP) was calculated by adding the trans‐tricuspid pressure gradient to the mean right‐atrial pressure estimated from the inferior vena cava diameter and its degree of collapse at inspiration, as follows: mean right‐atrial pressure was estimated to be 5 mmHg if the collapse was complete, 10 mmHg if the collapse was >50%, 15 mmHg if the collapse was >50% with inspiration, and 20 mmHg if there was no visible collapse of a dilated inferior vena cava with inspiration.15

All patients underwent non‐ECG‐gated CTPA in the radiology clinic, using pulmonary embolism protocol (16‐slice multidetector CT scanner; Somatom Sensation; Siemens Medical Systems, Erlangen, Germany; field of view: 35 cm, section thickness: 3 mm, contrast material volume: 135 mL, contrast material injection rate: 4 mL/s). CTPA images were obtained while the patients were holding their breath or breathing shallowly, depending on the degree of dyspnea. Acute PE was diagnosed in cases of a complete or partial luminal filling defect in the main pulmonary artery (MPA) or its branches. All CTPA scans were analyzed independently by two radiologists. The clots were classified as central, peripheral, or both. A central clot was defined as the presence of filling defects within the main pulmonary artery to lobar pulmonary arteries, whereas a peripheral clot was defined as the presence of filling defects within the segmental and sub‐segmental arteries. Right ventricular (RV) and left ventricular (LV) short‐axis diameters (RVd and LVd) were measured on a four‐chamber view by identifying the maximal distance between the ventricular endocardium and the interventricular septum, perpendicular to the long axis of the heart. RVd/LVd ratio (RVd/LVd) was calculated, and ratios over 0.9 were considered as right ventricular dysfunction (RVD).16

2.3. Laboratory analysis

Blood samples were obtained at admission. Activated thromboplastin time (aPTT), PT, INR, D‐dimers, fibrinogen levels, and complete blood count (CBC) were measured from blood samples taken during the patient's admission, on the same day as the diagnosis of PE but prior to the initiation of anticoagulant treatment. Blood samples for INR/PT and aPTT measurements were collected in blood collection tubes containing 3.2% sodium citrate (0.5 mL citrate, 4.5 mL blood). Samples were immediately centrifuged for routine testing, and analysis was performed within 1 hour after sampling. INR was measured using the reagent, HemosIL RecombiPlasTin 2G (Instrumentation Laboratory, Bedford, MA, USA). aPTT measurement was made using the reagent, HemosIL SynthASil (Instrumentation Laboratory). CBC was determined via a Abbott Cell‐Dyn 3700 autoanalyzer using commercial assay kits (Abbott Diagnostic, Santa Clara, CA, USA). Biochemical measurements were performed using Siemens Healthcare Diagnostic Products kits and calibrators (Marburg, Germany).

2.4. Medications

In the present study, 17.6% of the patients received a thrombolytic treatment, including a tissue plasminogen activator or streptokinase. After the diagnosis of acute PE was ascertained, low‐molecular weight heparin (LMWH) was given subcutaneously at weight‐adjusted doses. After the thrombolytic treatment, intravenous unfractionated heparin (UFH) or subcutaneous LMWH was used. Warfarin therapy was given to patients without active cancer on the day of admission, except for those who were treated with thrombolytic therapy; in this case, they were treated with warfarin 24 hours after thrombolytic therapy. Patients who received warfarin treatment were discharged with appropriate INR levels of 2–3. Patients with cancer were prescribed weight‐adjusted LMWH at discharge.

2.5. Statistical analysis

Continuous variables were presented as mean values±SD or medians with ranges, and categorical variables were expressed as percentages. Variables were compared by a two‐tailed Student's t test for continuous variables of normal distribution or by the Mann‐Whitney U test for continuous variables of non‐normal distribution. A Chi‐square test was used for categorical variables. Correlation analyses were carried out with a Spearman's correlation test for variables showing non‐normal distribution. The effect of various variables on mortality was calculated by univariate cox regression analysis. In these analyses, variables with unadjusted P values <.1 were identified as confounding factors and included in multivariate cox regression analyses to determine the independent predictors of acute PE mortality. The predictive values of INR and a combination of INR and sPESI score were estimated by comparing the areas under the receivers operating characteristic (ROC) curve. DeLong's test was used to compare the AUC from each of models,17 which were analyzed by use of Analyze‐it software programme. Moreover, the increased discriminative value after the addition of INR to sPESI score was also estimated using the Net Reclassification Improvement (NRI) and Integrated Discrimination Improvement.18 Survival curves were generated using the Kaplan‐Meier method, and the log‐rank test was used to compare survival between the groups. All statistical tests were two‐tailed, and a P value <.05 was considered statistically significant. All analyses were performed using SPSS version 15 (SPSS, Inc.; Chicago, IL, USA).

3. Results

Table 1 shows the patients’ clinical characteristics. Their mean age was 63±18 years, and 54.5% were female. At the 30‐day follow‐up, 39 of the 244 patients had died. The survivors were significantly younger than the non‐survivors (60±18 vs 69±17 years, P=.006). Compared with the survivors, non‐survivors had lower systolic blood pressure (SBP) (100±26 vs 116±20 mmHg, P<.001). However, non‐survivors’ heart rates were higher (113±23 vs 96±23 beats/s, P<.001). Similarly, sPESI scores ≥1 were more common in non‐survivors than in survivors (100% vs 79%, P<.001). The non‐survivors had a higher incidence of altered mental status (10% vs 0%, P<.001). RVD was more frequent in non‐survivors than in survivors (95% vs 56%, P<.001). Two patients died due to major bleeding. There was no difference in minor bleeding between the groups (1% vs 3%, P=.397). Other clinical variables were comparable in both groups.

Table 1.

Baseline clinical characteristics of the study population

Variables	Survivors (n=205)	Non‐survivors (n=39)	P value
Age, years	60±18	69±17	.006
Gender (male/female)	115/90	18/21	.253
History of cancer, n (%)	43 (21)	14 (36)	.044
Hypertension, n (%)	69 (34)	15 (33)	.969
Diabetes Mellitus, n (%)	32 (16)	7 (18)	.715
COPD, n (%)	22 (11)	5 (13)	.703
Deep vein thrombosis, n (%)	61 (30)	10 (29)	.832
History of Stroke/TIA, n (%)	27 (13)	5 (13)	.990
Heart failure, n (%)	11 (5)	1 (3)	.458
CAD, n (%)	22 (11)	1 (3)	.117
Smoking, n (%)	19 (9)	5 (13)	.495
Recent surgery, n (%)	73 (36)	12 (31)	.561
Altered mental status, n (%)	0 (0)	4 (10)	<.001
Syncope, n (%)	12 (6)	1 (2)	.402
Heart rate, bpm	96±23	113±23	<.001
Systolic blood pressure, mmHg	116±20	100±26	<.001
Hypoxiaa, n (%)	117 (57)	31 (80)	.009
sPESI score ≥1, n (%)	161 (79)	39 (100)	.001
sPAP, mmHg	47±19	53±19	.251
RVD, n (%)	110 (56)	35 (95)	<.001
Location of PE
MPA, n (%)	76 (41)	20 (65)	.014
RV dimension, mm	40±7	43±5	.036
T‐wave inversion (V1‐3), n (%)	19 (8)	3 (6)	.692
RBBB, n (%)	11 (4)	6 (2)	.848
Thrombolytic treatment, n (%)	38 (19)	5 (13)	.390

CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease; TIA, transient ischemic attack; sPAP, Systolic pulmonary arterial pressure; sPESI, simplified pulmonary embolism severity index; MPA, main pulmonary arteries; PE, pulmonary emboli; RBBB, right bundle‐branch block; RVD, right ventricular dysfunction.

^aHypoxia was defined as arterial oxyhemoglobin saturation <90%.

Table 2 shows laboratory findings of the study groups. White blood cell (WBC) count was higher in non‐survivors than in survivors (13±5 vs 11±4 (10⁹/L), P=.008). Non‐survivors had significantly higher glucose levels at admission compared with survivors (175±87 vs 149±67 mg/dL, P=.040).

Table 2.

Laboratory findings of the study population

Variables	Survivors (n=205)	Non‐survivors (n=39)	P value
White blood cell (109/L)	11±4	13±5	.008
Hemoglobin (g/dL)	12±3	12±2	.463
Platelets (109/L)	251±96	251±118	.978
Glucose, mg/dL	149±67	175±87	.040
Creatinine, mg/dLa	0.8 (0.7‐1.1)	1 (0.7‐1.7)	.246
ALT U/La	24 (14‐42)	35 (16‐123)	.294
AST U/La	30 (20‐47)	33 (23‐67)	.380
Myocardial necrosis, n (%)	68 (33)	17 (44)	.211
D‐dimer, ng/mLa	1920 (832‐4134)	2475 (932‐4270)	.405
Fibrinogen (mg/dL)	422±147	354±112	.214
PT, second	13.3±2.8	15.2±3.7	<.001
aPTT, second	28±8	27±7	.657
INR	1.1±0.3	1.3±0.4	<.001

ALT, alanine aminotransferase; AST, aspartate transaminase; PT, prothrombin time; aPTT, activated prothrombin time; INR, international normalized ratio.

^aComparison was made using Mann‐Whitney U test at P<.05 and these values were described by median with inter‐quartile range (25th and 75th percentile).

INR was significantly higher in non‐survivors than in survivors (1.3±0.4 vs 1.1±0.3, P<.001). For the entire study population, Spearman's rank correlation coefficient showed a weak, but statistically significant, inverse correlation between INR and SBP as well as a positive, statistically significant correlation between INR and sPESI scores, alanine aminotransferase (ALT), WBC count, and RVd (Table 3).

Table 3.

Linear association between INR and other variables

Variables	Correlation co‐efficient	P value
sPAP	0.095	.247
ALT	0.194	.002
sPESI	0.203	.001
Right ventricular dimension	0.174	.019
WBC	0.183	.004
D‐dimer levels	−0.033	.660
Fibrinogen levels	−0.086	.516
Systolic BP	−0.202	.003

INR, international normalized ratio; sPAP; systolic pulmonary artery pressure; ALT, alanine aminotransferase; sPESI, simplified pulmonary embolism severity index; cTnI, cardiac troponin I; WBC, white blood cell.

PE accounted for the majority of both in‐hospital and 30‐day mortality in our study. Data for in‐hospital and 30‐day cause‐specific mortality were presented separately in Supplementary Table 1. Univariate analysis showed that glucose at admission, sPESI score, RVD, altered mental status, age, WBC count, SBP, heart rate, hypoxia, INR, and PT were significantly associated with 30‐day mortality. Among univariables, WBC, sPESI scores, INR, altered mental status, and RVD were independent predictors of mortality (Table 4). Multivariate analyses were performed separately to evaluate the ability of PT to predict mortality in patients with acute PE. PT was found to be a significant predictor of early mortality (HR: 1.103, 95% CI: 1.014–1.199, P=.022). The AUC was 0.716 for the prediction of 30‐day mortality and the best cut‐off value was 1.12, with sensitivity, specificity, positive predictive value, and negative predictive value, 82.1%, 54.6%, 25.6%, and 94.1%, respectively. A Kaplan‐Meier analysis revealed that INR >1.12 was significantly associated with increased 30‐day mortality (Figure 1).

Table 4.

Independent predictors of 30‐day mortality based on INR

Variables	Univariate		Multivariate
	HR (95% CI)	P value	HR (95% CI)	P value
Agea, per 1‐year increase	1.028 (1.007‐1.050)	.008
Gender	1.436 (0.765‐2.695)	.260
Cancer historya	1.910 (0.992‐3.675)	.053
Altered mental status	13.120 (4.579‐37.593)	<.001	7.278 (2.012‐ 26.362)	.002
Hypoxiaa	2.696 (1.239‐5.866)	.012
RVD	12.277 (2.952‐51.055)	<.001	10.304 (2.455‐43.245)	.001
sPESI score, per 1‐point increase	1.808 (1.404‐2.327)	<.001	1.364 (1.005‐1850)	.046
WBC, per 103/L increase	1.104 (1.030‐1.184)	.005	1.079 (1.003‐1.161)	.040
Glucose, per 1 mg/dL increase	1.004 (1.001‐1.007)	.034	1.003 (0.999‐.1.007)	.185
INR, per 0.1 increase	3.789 (1.906‐7.529)	<.001	3.303 (1.210‐9.016)	.020
Heart ratea, per 1 bpm increase	1.028 (1.013‐1.044)	<.001
Systolic BPa, per 1 mmHg increase	0.978 (0.967‐0.988)	<.001
Myocardial necrosis	1.487 (0.790‐2.800)	.219

HR, hazards ratio; CI, confidence interval; RVD, right ventricular dysfunction; sPESI, simplified pulmonary embolism severity index; WBC, white blood cell; INR, international normalized ratio; BP, blood pressure.

^aAge, cancer history, hypoxia, heart rate, and blood pressure were not entered to the multivariate model as these parameters are included in the sPESI score calculation.

Figure 1.

Kaplan‐Meier survival curves for all‐cause mortality based on international normalized ratio

Compared with the sPESI score alone, combined use of the INR and sPESI score was associated with significant improvements in the ability to predict 30‐day mortality (AUC: 775 vs 736, P=.028; Figure 2). The addition of INR to s PESI score was related to a significant NRI of 8.8% (P<.001), and an IDI of 0.043 (P=.027; Table 5).

Figure 2.

Receiver operating characteristic (ROC) curve analysis comparing the sPESI score and the new model with the inclusion of INR

Table 5.

Reclassification of acute PE patients who died or who were alive at 30‐day based on INR status at diagnosis

sPESI without INR	sPESI with INR			Total
	<3% Risk	3‐6% Risk	>6% Risk
Patients who died, no.
<3% Risk	0	0	0	0
3‐6% Risk	0	0	0	0
>6% Risk	0	0	39	39
Total no.	0	0	39	39
Patients who were alive, no.
<3% risk	0	0	0	0
3‐6% risk	9	33	2	44
>6% risk	0	11	150	161
Total	9	44	152	205

PE, pulmonary emboli; sPESI, simplified Pulmonary Embolism Severity Index; INR, international normalized ratio.

In subgroup analysis, we compared the prognostic value of INR between patients who were diagnosed with PE in Emergency Department and those were diagnosed with PE in ICU/Wards. The predictive value of INR was similar in the two groups [AUC:0.706 (95% CI: 0.611‐0.801) vs AUC:0.710 (95% CI: 0.539‐0.881), P=.97].

4. Discussion

The results of the present study found that INR was significantly higher in non‐survivors than in survivors. In addition, elevated INR in the absence of anticoagulant therapy was independently associated with 30‐day mortality, as were sPESI score, RVD, WBC, and altered mental status. Consistent with the results of the present study, prior studies have shown that elevated sPESI scores, altered mental status, and RVD were associated with increased mortality rates in acute PE.8, 19, 20

Although PT and aPTT poorly reflect in vivo hemostasis, they are commonly used to evaluate coagulation state.21 However, these tests are a convenient method to quickly estimate the concentration of one or at times multiple coagulation factors for which each test is sensitive.22 In general, global clotting test results become elevated when the levels of clotting factors decrease to less than 50%. This is relevant since the levels of coagulation factors, that are needed for adequate hemostasis, are somewhere between 25% and 50%.21 An elevated INR may result from factor VII deficiency, mild vitamin K deficiency, mild liver insufficiency, or lower doses of vitamin K antagonists. In the vast majority of critically ill patients, deficiencies in coagulation factors are acquired.23 In most other cases, deficiencies in coagulation factors are caused by impaired synthesis, massive loss, or increased turnover (consumption).

Activating the inflammatory and coagulation pathways plays an important role in the pathogenesis of vascular diseases.24 Growing evidence shows that these two systems have a cross‐talk relationship, in which one can trigger the other. In general, endothelial cells play a central role in the coagulation response to systemic inflammation.24 In acute thromboembolic events, thrombin can have a pro‐inflammatory effect.25 Activation of the coagulation system and increased inflammation may lead to consumptive coagulopathy. In addition, the WBC count is correlated with the levels of hemostatic parameters.26 Furthermore, it is likely that severe PE alone may cause an elevated INR by inducing consumptive coagulopathy or disseminated intravascular coagulation (DIC).27 Central pulmonary embolism with a higher thrombi burden could also lead to a higher degree of hypoxemia.28 This may activate the inflammatory pathways favoring the larger emboli.27 In the present study, INR was positively correlated with WBC count. In addition, the rate of the centrally lodged thrombi was higher in non‐survivors than survivors. Shock and hypoperfusion has been shown to induce anticoagulation system.29 Our study showed that INR was weakly related to SBP. These findings indicate that inflammation (WBC) triggers coagulation activation, resulting in consumptive coagulopathy, which may lead to increased INR. Furthermore, our findings support the hypothesis with the association of elevated INR with lower SBP in patients with acute PE.

Acute PE can cause right‐heart failure, which leads to hepatic insufficiency or hypoxic hepatitis.30 Elevated right‐side pressure, in turn, leads to hepatic congestion, which may affect the production of coagulation factors. A previous study categorized PE patients into four groups according to arterial partial oxygen pressure, presence of massive or sub‐massive PE, pre‐existing cardiopulmonary disease, and presence of RVD.31 The study compared liver enzymes and PT among the four groups and found that aspartat transaminase (AST) and ALT levels were higher in cases of massive PE than in non‐massive cases. In contrast, in a study by Rendina et al.,32 the PT did not differ between the RVD and non‐RVD groups. However, the present study did not routinely measure other hepatic tests, such as alkaline phosphatase (ALP), gamma‐glutamyltransferase (GGT), and bilirubin, but it did find a weak positive relationship between INR and ALT levels. Therefore, we believe that hepatic dysfunction may also be responsible for elevated INR in some PE patients.

Early death after PE is strongly associated with RVD or RV dilatation.19 Animal studies and human autopsies have shown that acute PE with pulmonary hypertension in both rats and humans has resulted in right ventricular myocyte lysis and infiltration by neutrophils, macrophages, and lymphocytes 33, 34 after which such inflammation may trigger injury, leading to increased cTnI levels associated with mortality.35 Elevated WBC and neutrophil counts may reflect PE‐related RVD.19 Also, higher WBC count is associated with mortality in a previous study by Venetz et al. 36 as found in our study. Furthermore, INR was positively correlated with RV dimensions in the present study.

A previous study showed that critically ill patients may have some hemostatic abnormalities, including prolonged PT, INR, and aPTT.23 Fei et al. 37 reported that PT/INR, and aPTT were more significantly prolonged in non‐survivors than in survivors who were hospitalized in an intensive care unit. They also found that a INR of 1.07 seconds was the cut‐off point for predicting in‐hospital mortality with a sensitivity of 85% and a specificity of 49%. Okada et al. 38 showed that increased INR was associated with future adverse events in acute, decompensated heart failure. Our findings are in agreement with the above‐mentioned studies’ results, in which prolonged PT and increased INR were associated with mortality.

Another study, which has been recently published first study on the prognostic significance of admission INR in acute PE, found that patients with acute PE who had INRs greater than 1.2 in the absence of anticoagulant use had higher mortality during a 6‐month follow‐up.13 In the same study, PT was an independent predictor of 6‐month mortality.17 The PE cohort in our study were high‐risk patients as reflected by the high proportions of patients with sPESI ≥1, the high proportions of clot burdens seen in the main pulmonary artery on CTPA, and 43 out of the 244 patients received thrombolytic therapy, whereas the study by Wong et al. included mostly submassive PE. That study did not evaluate RVD or RV dilatation, which is the most common cause of early death in cases of acute PE,19 WBC showing inflammation, or myocardial necrosis measured with higher plasma levels of troponin I, whereas, the present study examined these parameters because they are associated with mortality in acute PE.7, 36 In addition, that study did not investigate ALT levels representing hepatic injury or sPAP associated with elevated right‐side pressure leading to hepatic congestion and affecting the production of coagulation factors. In contrast, the present study evaluated the relationship of these parameters to INR in acute PE. Similarly, in contrast to the above‐mentioned study, the present study examined the results of fibrinogen and D‐dimer levels and investigated these parameters’ relationships to INR.

We believe that our findings have potentially important clinical implications. It has been considered that elevated baseline INR in the absence of anticoagulant therapy is associated with mortality in patients with acute PE. The prognostic relevance of INR may be across all spectrum of admitted PE patients, though cannot be explorated to outpatient‐managed PE patients. However, the mechanism of this association is unclear. An elevated INR by itself may be associated with early mortality. In addition, increased INR may indicate the severity of PE through systemic inflammation, increased RV dimensions, lower SBP, and sPESI scores. Furthermore, elevation of INR may represent hemostatic derangement.

4.1. Limitations

The present study had several limitations. First, it was retrospective. Second, the number of patients included was relatively small. Third, the study did not routinely measure commonly used inflammatory markers, such as C‐reactive protein and interleukins and, thus, did not compare INR with these inflammatory markers. Fourth, the study did not investigate hemostatic parameters in specific types of cancer and the relationship between those parameters and the extent of the cancer and its treatment modalities. Fifth, the study did not assay the important parameters of liver dysfunction, such as albumin, ALP, GGT, and levels of special coagulation factor. Finally, this observational study cannot show a causal relationship between INR and PE, although INR appears to be related to the severity of PE.

5. Conclusions

In patients with acute PE, elevation of INR in the absence of anticoagulation was independently associated with 30‐day mortality. The ease of measuring elevation of INR may make it useful for prognostication. However, further studies are needed to implement this measurement into clinical practice and to determine whether an increase in INR can be correlated with inflammatory markers and RVD in patients with acute PE.

Funding

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

Supporting information

 

Kırış T, Yazıcı S, Durmuş G, et al. The relation between international normalized ratio and mortality in acute pulmonary embolism: A retrospective study. J Clin Lab Anal. 2018;32:e22164 10.1002/jcla.22164