This was a single-center descriptive and analytical observational study of a series of cases received at the ROCHER Medical Imaging Center (CIMR) in Kinshasa over a period of 24 months, going from January 2019 to January 2021 and coming from emergency department of Saint Joseph Hospital in Kinshasa.

The population of this study consisted of any adult patient consulting the emergency department of Saint Joseph Hospital and having performed an AST for suspected pulmonary embolism during the study period at the CIMR imaging department.

Inclusion criteria: Adult patients presenting to the emergency department of the Saint Joseph hospital in Kinshasa and having benefited from a chest CT-angiography at the CIMR, whatever the indication, but by which a diagnosis of EPA could be confirmed, were included. Non-inclusion criteria: An AST with chronic PE, A normal AST.

Ethical and benevolent principles were observed. The use of the results of the present study was limited to the strict exploitation related to the objective of the study. The protocol of our study was presented to the ethics local committee and we obtained a favorable opinion.

For the statistical analysis, a value of p < 0.05 was retained as significant for the interpretation of the results. The quantitative data were compared by ANOVA analysis of covariance then two-by-two student test. The normality of the distributions was checked using the Shapiro-Wilk test. The quantitative data were then compared by Student’s t test and were correlated using Spearman’s correlation. Qualitative data were studied using Fisher’s exact test. All statistics were carried out using SPSS software version 21.0 for Windows. Equation of the linear regression line: y = 0.6953x + 8.0679.

From 2019 to 2021, the Le Rocher imaging department received 246 patients with suspected PE coming from the HSJ emergency department, among whom 63 had an acute pulmonary embolism on CT-angiography (25.6%). Most patients had no morbid history or risk factors. Of a total of 63 patients, 55 presented with non-massive embolism (87.3%), while two patients died (3.2%). The average Qanadli score was 46.6% ± 12%, or lower in 73.1% of cases. The mean age of patients with massive pulmonary embolism was similar to that of patients with non-massive pulmonary embolism as were many women and men. All patients with non-massive pulmonary embolism were stable (alive), while 2 patients with massive pulmonary embolism died. The majority of patients with massive pulmonary embolism had a history of thromboembolic disease as a risk factor in a higher proportion compared to patients with non-massive embolism. According to the PESI score, classes IV and V were most frequently found in patients with massive pulmonary embolism, while classes I, II and III were found in patients with non-massive pulmonary embolism (Table 1).

The Qanadli score was on average higher (57.6 ± 12) in patients with massive embolism compared to patients with non-massive embolism (45 ± 11.2), the comparison of proportions showed a statistically significant difference (p = 0.005). On the other hand, when comparing the averages of the mean LV diameters, the average of patients with massive embolism was low compared to patients with non-massive embolism (p = 0.005). Unlike the pulmonary artery measurement which was larger in patients with massive embolism (p = 0.007) compared to patients with non-massive embolism. The same for the measurement of the aorta (p = 0.035) (Table 2).

Of a total of 63 patients followed in our study, 2 died, representing a mortality of 3.2%. Below, Table 3 provides us with information on the general characteristics of patients according to survival. Deceased patients were older than living patients (p = 0.0392). A history of DVT and heart failure was found in 50% of these patients, respectively. They were both classified V according to the PESI score. While living patients had a PESI score ranging from I to III in 32.8%, 29.5%, and 19.7% respectively. Most of them had no history (49.5%) or risk factors (62.3%). (Table 3).

*Standard deviation.

*standard deviation.

Out Qanadli score was higher in deceased patients (64.5% ± 2.1%) compared to living patients (46% ± 11.8%). All deceased patients had a Qanadli score greater than or equal to 60% with a statistically significant difference (p = 0.0313). These patients had a larger diameter pulmonary artery (36.3 ± 5.4) compared to the deceased patients (29.7 ± 3.3) with a statistically significant difference (p = 0.0085). The mean RV diameter of deceased patients was higher compared to living patients (p = 0.0384). While the average LV in living patients was higher in the living than in the deceased (p = 0.213). The RV/LV ratio was higher in patients with massive PE compared to stable patients (1.7 versus 1.1 p = 0.001). The sensitivity of an RV/LV ratio >1 for the occurrence of massive PE was 91% with a specificity of 59% (Table 4).

In our 4-year series, 246, including 63, had acute pulmonary embolism on CT. The incidence of acute pulmonary embolism in this population is therefore 25.6%. Our rate is close to the numerous rates reported in the majority of studies, which is around 25% [13] and even up to 62% in the study by Nural et al. [14] .

Indeed, the latest European recommendations have highlighted spiral pulmonary angiography in the diagnostic but also prognostic strategy for PE [15] . Due to advances in imaging, certain anomalies (dilatation of the right cavities and paradoxical septum) can also be visualized with a multi-slice scanner, and the aim of our work is to study their prognostic nature. Indeed, valvular flows, the analysis of which is done by ultrasound and which cannot be studied directly on CT angiography (in particular tricuspid insufficiency), have little place for the prognosis of the EP at present [15] .

In our series, the mean age of patients with massive pulmonary embolism was similar to that of patients with non-massive pulmonary embolism (p = 0.062) as much as the number of women and men. The average age of the patients was 66.5 ± 9.3 years with a F/M sex ratio of 1.6. Our results are similar to those reported by Bounameaux et al. [16] . Not only is age a risk factor for PE, it is reported in the literature that several factors for PE occur with age. In our series, the majority of patients with massive pulmonary embolism had a history of thromboembolism as a risk factor in a higher proportion compared to patients with non-massive embolism (p = 0.024) in whom the average age was statically higher. According to the PESI score, classes IV and V were most frequently found in patients with massive pulmonary embolism, while classes I, II and III were found in patients with non-massive pulmonary embolism (p = 0.000). Note that this score becomes pejorative with age.

Of a total of 63 patients followed in our study, 2 died, representing a mortality of 3.2%. Deceased patients were older than living patients (p = 0.0392). A history of DVT and heart failure was found in 50% of these patients, respectively. They were both classified V according to the PESI score. While living patients had a PESI score ranging from I to III in 32.8%, 29.5%, and 19.7% respectively. Most of them had no history (49.5%) or risk factors (62.3%).

Qanadli score was higher in deceased patients (64.5% ± 2.1%) compared to living patients (46% ± 11.8%). All deceased patients had a Qanadli score greater than or equal to 60% with a statistically significant difference (p = 0.0313). These patients had a larger diameter pulmonary artery (36.3 ± 5.4) compared to the deceased patients (29.7 ± 3.3) with a statistically significant difference (p = 0.0085). The mean RV diameter of deceased patients was higher compared to living patients (p = 0.0384). While the average LV in living patients was higher in the living than in the deceased (p = 0.213).

In fact, hospital mortality from pulmonary embolism is of the order of 10% [16] [17] [18] [19] , but this figure varies significantly depending on the area concerned and hemodynamic tolerance, but also depending on the authors. PE-related deaths most often occur within an hour of the onset of symptoms: 2/3 of patients with massive PE will die within this time, according to Wood et al. [17] . This delay was verified in our study, the 2 deaths occurred among patients with massive PE, a few hours after their arrival in the emergency room.

The PESI score another severity score based on exclusively clinical elements was developed by Aujesky [17] , with 30-day mortality as the endpoint: the Pulmonary Embolism Severity Index (PESI). It was developed retrospectively from a large North American hospital registry and was subject to internal validation and external validations in various independent, particularly European, cohorts [17] [18] . The score increases with age, male gender, number of comorbidities and clinical abnormalities. The patients are divided into five classes, and mortality increases from class I to V. It is of the order of 2.5% in classes I and II, it is 7% in class III, 11% in class IV and 24% in class V. In our series, according to the PESI score, classes IV and V were most commonly found in patients with massive pulmonary embolism, while classes I, II and III were found in patients with non-massive pulmonary embolism (p = 0.000). Note that this score becomes pejorative with age.

The RV/LV ratio: The RV/LV ratio was higher in patients with massive PE compared to stable patients (1.7 versus 1.1 p = 0.001). The sensitivity of an RV/LV ratio > 1 for the occurrence of massive PE was 91% with a specificity of 59%. In the study by Mbozo’o et al. [20] 11.11% of pulmonary embolisms were severe, i.e., an RV/LV ratio > 1.5. A pulmonary artery trunk diameter value greater than 27.5 mm was correlated with an RV/LV ratio > 1 (Sensitivity: 73%, Specificity: 77%). This author reports that the lethality of pulmonary embolism was correlated with an RV/LV ratio greater than 1.5 (Sensitivity = 100%, Specificity = 98%). Indeed, according to the new recommendations of the European Society of Cardiology published in 2008, the severity criteria for a pulmonary embolism are as follows: RV/LV ratio greater than 1.5 which is a reflection of right ventricular dysfunction and Diameter of the pulmonary artery and TAP/ Aorta ratio greater than 1 [15] .

Due to advances in imaging, certain anomalies (dilatation of the right cavities and paradoxical septum) can also be visualized on a multi-slice scanner, and the aim of our work is to study their prognostic nature. Indeed, valvular flows, the analysis of which is done by ultrasound and which cannot be studied directly on CT angiography (in particular tricuspid insufficiency), have little place for the prognosis of the EP at present. The parameters studied on ultrasound for the prognosis of PE can therefore technically be replaced by a CT study. The latter indirectly evaluates right ventricular dysfunction by measuring the RV/LV ratio, the presence or absence of a paradoxical septum, or the presence of contrast product reflux in the IVC and VSH. Its prognostic evaluation would be all the more interesting as it is the reference examination for the diagnosis of PE; all patients therefore benefit from it, unless there is a contraindication. CT signs of right ventricular dysfunction: Signs of right ventricular dysfunction have been validated on echocardiography as being interesting prognostic markers. With advances in imaging, these signs can be transposed to CT, and this is all the more interesting as this examination is the gold standard for the diagnosis of PE. A single examination therefore makes it possible to make the diagnosis and stratify the prognostic risk of patients with PE. Indirect evaluation of right heart function (ratio of RV/LV diameters in short axis section, position of the interventricular septum, blood reflux in the IVC or VSH, size of the pulmonary artery), direct pulmonary ischemic complications of the PE (presence of a pulmonary infarction), and the vascular obstruction index can be interesting prognostic markers, and are easy to assess on a CT angiogram.

The increase in the RV/LV ratio and the deformation of the interventricular septum, assessed by echocardiography, have been known for a long time for their short-term prognostic value in PE, and numerous studies have shown their equivalence as a prognostic value. to the scanner. Two studies, by Lim et al. [18] showed that the signs of right ventricular failure on CT, defined by an RV/LV ratio > 1 or a bulging of the interventricular septum, had a sensitivity between 78% and 92%, a specificity of 100%, and a value 100% positive predictive using transthoracic echocardiography as a reference examination. The scanner therefore seems reliable in detecting these signs, it is correlated with ultrasound for the detection of signs of right ventricular failure. The scanner is more interesting, especially since the measurements to be carried out are easy, quick, and reproducible, and the diagnosis and prognosis are carried out in a single examination in current practice. Arterial Obstruction Score: The Qanadli score was on average higher (57.6 ± 12) in patients with massive embolism compared to patients with non-massive embolism (45 ± 11.2), the comparison of proportions showed a statistically significant difference (p = 0.005). Comparison of mean RV diameters did not show a statistically significant difference. On the other hand, when comparing the averages of the mean LV diameters, the average of patients with massive embolism was low compared to patients with non-massive embolism (p = 0.005).

Unlike the pulmonary artery measurement which was larger in patients with massive embolism (p = 0.007) compared to patients with non-massive embolism. The same for the measurement of the aorta (p = 0.035).

In our study, the Qanadli score was on average higher (57.6 ± 12) in patients with massive embolism compared to patients with non-massive embolism (45 ± 11.2), the comparison of proportions showed a statistically significant difference (p = 0.005). Indeed, the diagnosis of PE previously relied on pulmonary angiography, and, among the vascular obstruction scores calculated to quantify the importance of the embolism, that of Miller was widely used. PE was considered serious from 50% - 60% vascular obstruction. Mansencal et al. [21] show a good correlation between the degree of vascular obstruction on angiography (using the Miller score) and the impact on the right cavities defined by an RV/LV ratio > 0.6 in apical 4-chamber ultrasound section. But this link was not found with the measurements on parasternal long axis sections: on these sections, the dilation of the right cavities was not correlated with vascular obstruction. Mansencal et al. [21] therefore showed the existence of a link between RV dilation and embolic load of the EP only on 4-chamber ultrasound sections. The literature is quite discordant regarding the link between the importance of PE according to vascular obstruction scores and clinical outcome, but this may be due to very different study populations. Some studies find a correlation between early mortality and the importance of the embolic load on CT-scan, others not, but the embolic load is nevertheless higher in hemodynamically unstable patients than in stable patients: obstruction scores vascular scans would therefore be good indicators of the clinical severity of PE. The Mastora score focuses on vessels up to the division’s segments, and the Qanadli score to the subsegmental arteries. Furthermore, these scores do not take into account the underlying situation, in particular chronic respiratory insufficiency, already responsible for an increase in pulmonary resistance.