A retrospective analysis was performed of a prospective database of 213 consecutive patients admitted to the coronary care unit of Coimbra University Hospitals with a diagnosis of intermediate-to-high risk PE, between May 13, 2000 and October 15, 2010. The diagnosis was confirmed by computed tomography (CT) angiography or ventilation/perfusion scintigraphy. Clinical data collected at admission enabled retrospective risk stratification based on current guidelines6; this was complemented by analysis of patients’ medical records. The decision to perform fibrinolysis with alteplase (Actylise®, Boehringer Ingelheim, Germany) was at the discretion of the attending physician, and most patients (n=183) received fibrinolytic therapy. Patients were initially anticoagulated with unfractionated heparin, together with warfarin. Inotropic, vasopressor and ventilatory support were instituted depending on clinical assessment.

The following risk factors for PE were analyzed: advanced age (≥80 years), obesity, lower limb varicose veins, recent immobilization, known thrombophilia, previous venous thromboembolism, pregnancy, use of oral contraceptives, known cancer, heart failure, and recent major trauma or major surgery. Other demographic, clinical, laboratory and imaging variables were also recorded. Patients with a body mass index (BMI) of ≥30 kg/m2 were considered obese. Median follow-up was three years for 187 (95.4%) of the 196 patients who were discharged from hospital.

Patients with PASP of ≥40 mmHg were considered to have PH, as estimated by transthoracic echocardiography through calculation of the pressure gradient of tricuspid regurgitation using the formula PASP=4(VTR)2+central venous pressure. Central venous pressure was estimated by analysis of the respiratory variation of the inferior vena cava.17 Median time to echocardiography was 22 months (interquartile range 13–38, maximum 137). Only echocardiograms performed in the same hospital at least three months after discharge were considered valid; of the 187 patients, 121 (56.8% of the total) met this requirement. For individuals who underwent serial echocardiography during follow-up, only the most recent exam was used in the analysis. Patients were divided into two groups: group A (PASP <40 mmHg) with 106 patients (87.6%), and group B (PASP ≥40 mmHg) with 15 patients (12.4%).

Continuous variables with normal distribution were expressed as means±standard deviation, while those showing an abnormal distribution were assessed by the Kolmogorov-Smirnov test and expressed as medians and interquartile range. Categorical variables were presented as absolute frequencies and percentages. Comparisons between groups were made using the Student's t test, Mann–Whitney U test or chi-square test, depending on the type of variable. In all tests, a p value of <0.05 was considered statistically significant. Parameters showing a significant association in univariate analysis with the development of PH during follow-up were included in a multivariate model. The best cut-off for each of these variables was determined by receiver-operator characteristic (ROC) curves. SPSS version 20.0 (Chicago, Illinois, USA) was used for the statistical analysis.

Of the 213 patients included in the study, 61.5% were women, and mean age was 61.1±18.1 years. Most (90.6%) presented at least one risk factor for PE, while 20 cases (9.4%) were considered idiopathic; there was no significant difference in this respect between those with or without PH after PE (13.2 vs. 6.7%, p=0.692) (Figure 1).

Figure 1.

Diagram of the study population. PASP: pulmonary artery systolic pressure: PE: pulmonary embolism.

(0.34MB).

There were significant differences in PE risk factors (Table 1) between the two groups in age and use of oral contraceptives. Less than 3% of group A were aged >79 years at the time of PE diagnosis, as opposed to 53.3% of group B, and group B patients were older than group A (70.7±11.8 vs. 54.7±17.3, p<0.001) at diagnosis (Figure 2). No patients in group B were taking oral contraceptives at the time of admission, while around a quarter of those in group A were. The prevalence of other risk factors was similar in the two groups. Group B had a higher proportion of men (60.0 vs. 31.1%, p=0.041).

Figure 2.

Comparison of age at admission between the two study groups. PASP: pulmonary artery systolic pressure.

(0.09MB).

Group B had higher BMI than group A (31.3 [27.3–37.8] kg/m2 vs. 27.6 [25.0–30.8] kg/m2, p=0.036) and thus a higher prevalence of obesity (66.67% vs. 29.63%, p=0.023) (Figure 2). At admission, group A patients had significantly lower systolic blood pressure than group B (113 [110–140] mmHg vs. 140 [120–152] mmHg, p=0.027) (Table 2). A minority (5.6%) of patients presented hypotension at admission, and were diagnosed with high-risk PE; of these, 50% died before follow-up echocardiography (three during hospital stay and three after discharge). None of the six survivors of high-risk PE developed PH during follow-up (Figure 3).

Figure 3.

Comparison of body mass index at admission between the two study groups. BMI: body mass index; PASP: pulmonary artery systolic pressure.

(0.09MB).

Analysis of arterial blood gas data showed no significant differences between the two groups, although group B had higher hemoglobin than group A (15.1 vs. 13.3 g/dl, p=0.014) (Table 3). There were no significant differences between the groups in imaging findings at admission or in therapeutic approach, including use of fibrinolysis. Recurrence of PE during follow-up was similar in the two groups, seven cases (6.6%) in group A and one (6.7%) in group B.

The following parameters were included in the multivariate analysis: age (p=0.001), gender (p=0.041), BMI (p=0.036), systolic blood pressure (p=0.027), hemoglobin (p=0.014), and use of oral contraceptives (p=0.022). Age and BMI were predictors of the development of PH, with hazard ratio of 1.09 (95% confidence interval [CI] 1.02–1.20, p=0.012) per year of age, and 1.19 (95% CI 1.04–1.36, p=0.013) per kg/m2 of BMI. Figure 4 shows the significant difference (p<0.001) in PASP between patients aged <68 years and BMI <30 kg/m2 (31.2±4.6 mmHg) and those aged ≥68 years and BMI ≥30 kg/m2 (48.8±5.0 mmHg).

Figure 4.

Pulmonary artery systolic pressure during follow-up according to age and body mass index. BMI: body mass index; PASP: pulmonary artery systolic pressure.

(0.1MB).

The retrospective nature of the present study means it has various limitations, of which the main one is the fact that a diagnosis of PH was based on echocardiographic assessment rather than direct catheterization, the gold-standard method.6 Thus, besides the possibility of some patients being false positives or false negatives, it was not possible to identify concomitant causes of PH such as left ventricular dysfunction. We would stress that only direct catheterization or pulmonary angiography (or at least ventilation/perfusion scintigraphy) can confirm a diagnosis of CTEPH, since CT angiography is associated with a false negative rate of 50%.25 Accordingly, we prefer to use the term PH after PE, since a diagnosis of CTEPH requires use of one of the diagnostic techniques mentioned above.

A further limitation is the lack of echocardiographic follow-up in part of the study population, which may have introduced bias in assessment of PH prevalence. Only use of a prospective protocol can mitigate this important limitation. Nevertheless, we feel that this first analysis of the incidence of PH after intermediate-to-high risk PE may call attention to the problem and that an incidence of around 10% at three years may prompt closer clinical follow-up of these patients, particularly among the elderly and/or obese. Naturally, differential diagnosis with other causes of PH remains crucial for the appropriate selection of patients for surgical or medical interventions.

The authors declare that no experiments were performed on humans or animals for this study.

The authors declare that they have followed the protocols of their work center on the publication of patient data and that all the patients included in the study received sufficient information and gave their written informed consent to participate in the study.

The authors have obtained the written informed consent of the patients or subjects mentioned in the article. The corresponding author is in possession of this document.