Pulmonary embolism of the elderly in 50 cases in the short-stay geriatric unit of the Fann University Hospital in Dakar, Senegal

¹Geriatrics Department of the National University Hospital of Fann, Dakar, Senegal, ²Faculty of Medicine of Cheikh Anta Diop University of Dakar, Dakar, Senegal

^&Corresponding author
Assane Sall, Geriatrics Department of the National University Hospital of Fann, Dakar, Senegal

Introduction    

Venous thromboembolic disease ranks 3^rd with an overall annual incidence of 100-200 per 100,000 inhabitants [1]. Pulmonary Embolism (PE) is defined as the partial or total obstruction of the pulmonary arteries or their branches by emboli, most often of fibrin-cruoric origin. Its prevalence increases exponentially with aging, which modifies the cardiovascular structure and function. Thus, the incidence increased from 3.5/1000 inhabitants between 65 and 74 years of age to reach 6/1000 inhabitants after 75 years of age [2]. It is responsible for a high mortality in geriatrics with a death rate of 15.7% after 30 days and 54.6% after 3 years of a PE [3]. Indeed, PE often occurs in a geriatric profile marked by frailty, polypathology, and polymedication. Consequently, the clinical expression can be atypical (delirium, fall, malaise, etc.) and certain diagnostic and therapeutic modalities are limited by frequent comorbidities (severe nephropathies, hemorrhagic gastritis, etc.). This makes geriatrician management complex despite the existence of general recommendations [4]. Thus, the individualization of each situation is necessary and must be based on a global assessment of the health and fragility of the patient. In Africa, over the past 20 years, scientific knowledge on the issue has made considerable progress on the main questions that clinicians ask themselves when faced with a suspected PE in an emergency. In Senegal, several studies have been carried out on the issue but in a cardiological setting on a young population without taking into account geriatric dimensions [5,6]. It is in this context that we conducted this study in the geriatrics department of the National Hospital University Center (NHUC) of Fann, the general objective of which was to describe the particularities of PE in the elderly in order to optimize management.

Methods     

Study framework: this study took place at the short-stay unit of the Geriatrics Department of the NHUC of Fann in Dakar the capital city of Senegal. It is at the top of the national health pyramid of the health system. The geriatrics department was created there in 2015 from a geriatrics department. Its sectors include acute geriatrics in a unit with a capacity of 16 beds spread over two buildings, a geriatric consultation and assessment unit, and an intra- and extra-hospital mobile unit. Extension work is underway, offering in a few months, the largest geriatrics department in the sub-region with a capacity of 60 beds with additional sectors such as long-term geriatric stay and post-care and rehabilitation care.

Type and period of study: this was a descriptive, retrospective study on a population of people aged 65 or over hospitalized in the short-stay unit of the service over the period from July 1, 2019, to October 1, 2023.

Study population: all patient records with a diagnosis of PE based on thoracic computed tomography (CT) angiography were included. Any patient with incomplete and unusable records was excluded.

Collections and study variables: the data were collected from a collection sheet after free and informed consent from the elderly person or their legal representative. The following variables were collected and studied: socio-demographic characteristics; clinical (comorbidities, history, clinical manifestations, standardized geriatric assessment, etiological risk factors); paraclinical including biology (D-dimers, troponinemia, blood count, C-reactive protein, urea-creatinine, blood ionogram) and imaging (surface electrocardiogram, transthoracic echocardiography, frontal chest X-ray, thoracic CT angiography); therapeutic and evolutionary (average length of stay and patient outcomes). Geriatric syndromes were also systematically sought using scales validated in the elderly: Katz activities of daily living (ADL) (functional autonomy), mini-nutritional assessment (nutritional status), getriatric depression scale (mood) and Senegal test (cognitive function), fried criteria (frailty).

Input and analysis: the collected data were entered and analyzed using Epi.info software version 7.2.4 for Windows. Quantitative variables were expressed in frequency (%) and mean with their standard deviation; qualitative variables in frequency.

Ethical approval and consent to participate: for our study, the need for ethical approval was not necessary according to Senegalese national regulations. Permission was granted by the head of the department to conduct this study and access patient information. Access to the data is limited to the study team and confidentiality of the information has been ensured.

Results     

Epidemiological aspects: of 900 patients hospitalized during the study period, 50 patients developed a pulmonary embolism, i.e. a prevalence of 5.55 %. Despite systematic preventive heparin therapy on admission, 16% of cases occurred during hospitalization with a mean time to onset of 9 days± 4 and extremes of 4 and 20 days. The mean age was 80 years ± 9 years (extreme of 65 and 101 years) and the age group of 75-79 years was more representative (28%) (Table 1). A female predominance (60%) was found, i.e. a sex ratio M/F of 0.66. The mode of admission was dominated by transfer by the National Emergency Medical Assistance Service (28%) followed by the geriatric outpatient consultation and other intra-hospital services with 26% each. The hospital emergency reception service and home consultation each represented 10%.

Diagnostic aspects: the mean number of comorbidities was 3±1 with polypathology (≥3 comorbidities) noted in 74% of patients. Comorbidities were dominated by high blood pressure (HBP) with 70% followed respectively by type 2 diabetes (38%), heart disease, and chronic major neurocognitive disorders with 20% each (Table 1). A personal history of thromboembolic disease was found in 18% of patients consisting of PE (14%) and deep vein thrombosis (4%). A notion of active smoking was found in 4% and active alcoholism in 2% of patients. Polymedication (≥ 5 drugs/day) at admission was found in 24% of patients with an average of 5 drugs/day per patient. Cardiovascular drugs were most commonly found (58%), mainly antihypertensives (60%) followed by anticoagulants (36%) and antibiotics (34%) (Table 1). The clinical manifestations were dominated by general signs (42.3%) mainly delirium (58%) and deterioration of the general condition (40%); followed by physical signs (41.6%) represented dominated by pulmonary condensation syndrome (54%) and respiratory distress (30%). Functional signs were less represented (16%) dominated by dyspnea at rest (38%) followed by cough (18%) and chest pain (18%). Geriatric syndromes were dominated by loss of dependency for ADL (94%) followed by frailty (88%) and dementia (32%) (Table 2). The mean modified Geneva PE probability score was 7.6±3.5 (range 1-17) with clinical probability distributed as follows: intermediate (82%), high (14%), and low (2%). D-Dimer (DD) assay was performed in 16% and returned positive with a mean of 4218.87±1160.98 mg/ml after adjustment for age. The elevation of C-reactive protein was constant (100%). The abnormalities of the blood count were dominated by anemia (80%) followed by hyperleukocytosis (48%), thrombocytopenia, and thrombocytosis with 20% each. The glomerular filtration rate according to MDRD was on average 67.13 ml/kg/1.73 m2 (range 17.78 and 144.12) with renal failure in 72% of cases. Ionic disorders were found in the following proportions: hyponatremia (30%), hypokalemia (22%), hypernatremia (12%) and hyperkalemia (2%).

An increase in ultrasensitive troponin in 1 of the 2 beneficiaries, i.e. 4% of patients. The electrocardiogram performed in 82% of patients mainly found tachycardia (70%) followed by right bundle branch block and left ventricular hypertrophy with 29.3% each. Echocardiography was performed in 46% of patients and mainly found right ventricular dilation (RVD) (34.8%) and pulmonary arterial hypertension (PAH) (30.4%). A chest X-ray was performed in 28% of cases and mainly revealed a fluid pleural effusion (35.7%) and pulmonary edema (28.6%). CT angiography mainly found right PE (44%), lobar (34%), and non-massive (88%) associated with parenchymal lesions dominated by interstitial syndrome (24%) and frosted glass appearance (18%) (Table 3). The mean PESI score was 158.38±48.49 (range: 84 to 256) with a risk of death within 30 days considered high in the large majority (80%) of cases distributed as follows: class IV (12%), class V (68%). The risk of death within 30 days was moderate (class III) in 16% and low in 4% including 2% in class I and 2% in class II. The mean number of risk factors was 3±2 with 88% having at least 2 factors. Etiological factors were dominated by immobilization of more than 3 weeks (86%), infections (40%) followed by strokes (24%), and active cancer (16%) (Figure 1).

Therapeutic aspects: anticoagulation was the main therapeutic option with curative heparin therapy (88%) distributed as follows: low molecular weight heparin (86%) and unfractionated heparin (2%). It is replaced by Direct Oral Anticoagulation (DOAs)/Rivaroxaban in 40% of cases and Vitamin K antagonists (VKAs)/ Acenocoumarol in 4%. No patient had benefited from thrombolysis.

Evolutionary aspects: the average length of hospitalization was 12.88±8 days (range: 3 to 45 days). The outcome was favorable in 50% with return home and outpatient follow-up. A transfer to intensive care was made in 3 patients (6%) and a mortality of 44% was recorded.

Discussion     

Our series reports a prevalence of pulmonary embolism of the order of 5.72% close to that of Harot et al. [7] who found a prevalence of 8.7% in a reception service in a French geriatric population (≥65 years). Despite systematic preventive heparin therapy on admission, 16% of cases occurred during hospitalization. This observation is consistent with that of Harrot et al. [7] who found an absence of statistical association between the taking of all anticoagulants and a lower prevalence of PE. In Senegal, Ndiaye et al. [6] found 2.9% in a cardiology unit on a younger population (52.6±16.2). Advanced age is established as a risk factor for PE. The variability of the data can be explained by the inclusion criteria for patients differing from one study to another. The mean age of our series was 80±9 years with a female predominance (60%) by most literature data. In Israel, Aharoni et al. [3] found a similar mean age of 79±5.7 years with 64.7% women. In the internal medicine department of the Besançon University Hospital in France, Tisserand et al. [8] found a higher mean age of 82 years in a cohort of patients over 75 years with 56% women. Our high averages compared to studies conducted in Senegal and Africa could be explained by the geriatric vocation of the department. In addition, advanced age is a risk factor for developing PE in venous aging but also in pathologies and situations favoring the occurrence of thrombosis. The female predominance can be explained on the one hand by postmenopausal estrogen deficiency and on the other hand, the feminization of the Senegalese geriatric population and the greater attendance of our geriatric service by women 54.6%.

The classic manifestations of PE consisted mainly of dyspnea (38%) and hypoxemia (38%) followed by cough (28%) and chest pain (18%). These proportions are much lower than the literature data except for pain which is almost always weakly expressed in PE in the elderly. In 2017, Bordage et al. [9] in a French population over 80 years old found dyspnea (74%) and hypoxemia (58%). In the same vein, Tisserand et al. [8]in their comparative study identified dyspnea (47% in those over 75 years old vs. 54.5% in those under 75 years old; p = 0.79) however chest pain was associated with younger age (36.5% in those under 75 years old vs. 7.1% in those over 75 years old; p = 0.04). We found a high frequency of mental confusion (54%), and deterioration of the general condition (40%). Zoe Faure et al. [10] identified malaise and deterioration of the general condition in the respective proportions of 18% and 20%. In addition, the existence of deterioration of the general condition on admission was significantly associated with 1-year mortality (p = 0.003) [10]. Mental confusion or delirium is an acute and potentially reversible cerebral decompensation related to often serious acute aggression. It is associated with heavy morbidity and mortality in geriatrics. It is established as an atypical presentation of PE in the literature. In a cohort of 215 adults who received PE, 5 cases (2.3%), all aged over 72 years, were revealed by hyperactive mental confusion in the foreground [11]. Indeed, frequent hypoxia in PE is a factor that can decompensate the brain function of the elderly person weakened by the effects of aging and comorbidities including dementia (20%) and stroke (16%) among others. In the series of Tisserant et al. [8], syncope predominated in the elderly subject (33%) with a statistically significant difference (p = 0.04).

The differences observed with these studies demonstrate the heterogeneity of clinical symptoms on the one hand and the limitation to common signs of PE in the collection in most studies underestimating the geriatric dimensions. Besides ADL dependency, the other syndromes found were frailty (88%), cognitive disorders (32.2%), and malnutrition (22%). Frailty is defined as a decline in functional reserves and impaired ability to adapt to even minor stress. It is the result of a complex interaction leading to a decline in one or more areas of physical, psychological, and/or social functioning. Aharoni et al. [3] identified dementia in 12% and it was associated with mortality at 1 year (p = 0.011). The existence of cognitive disorders can mask the expression of symptoms such as pain and dyspnea, thus delaying management. The study "The Contemporary Management of PE" (COPE) of a population with a mean age of 70 years ±16 years found cognitive impairment in 11%. A recent American study [12] identified malnutrition in 5.4% and advanced age as a risk factor for in-hospital mortality (p<0.001), complications and other adverse outcomes, including acute renal failure (p<0.001), sepsis (p<0.001), shock (p<0.001), acidosis (, p<0.001) and need for mechanical ventilation (p<0.001) [12].

The mean of the modified Geneva probability scores for PE was 7.6±3.5 with an intermediate clinical probability in 82% consistent with the study of Harrot et al. [7] who noted the predominance of intermediate (62.3%) and low (30.5%) probability and suggested a decline in performance in geriatrics. Indeed, special attention should be paid to elderly people using bradycardic drugs and anticoagulants that can alter diagnostic performance. The high prevalence in our series could be explained by our inclusion criterion requiring diagnosis by CT angiography. Per current recommendations [4] suggesting CT angiography with intermediate or high clinical probabilities. CT angiography mainly found unilateral PE (64%) with a right predominance (44%), lobar (34%), segmental (32%), massive (12%) associated with parenchymal lesions dominated by interstitial syndrome (24%) and ground glass appearance (18%). Our observation is corroborated by several studies. In the COPE study [13], the diagnosis was made in 96.3% with segmental (26.4%), subsegmental (4.4%), and bilateral (25.2%) topography. Perrier et al. [14] in a younger population (mean age: 60±19) showed the excellent diagnostic performance of CT angiography (97%) with the demonstration of PE of the pulmonary trunk (32%), lobar (35%) and segmental (26%). The frequency of ground-glass lesions in our series could be justified by the COVID-19 pandemic that frames our study period. Ground-glass lesions are among the characteristic lesions of SARS-CoV-2 pneumonia.

Echocardiography carried out in our study (46%) found mainly (RVD) (34.8%) and PAH (30.4%). Zoe F et al. [10] found RVD in 39% of elderly patients in institutions. Another study found a frequency of RVD ≥ 25% in PE and its interest in risk stratification has been recommended [15]. It has a high specificity (98%) and a low sensitivity (50%) in the detection of PE. PAH has a high sensitivity (83%) and a positive predictive value (86%) [16]. It comes from mechanical obstruction of the pulmonary arteries and humoral factors secreted due to hypoxia. In addition to risk stratification, echocardiography is of major interest due to the frequency of contraindication of CT angiography due to frequent severe renal failure and, on the other hand, the often multiple and complex differential diagnoses of semiological atypia and polypathology. The etiologies of PE were multifactorial (on average 3±2 factors) dominated by immobilization of more than 3 weeks (86%), infections (40%), and active cancer (16%). A large disparity in the literature was found. The COPE study [13] found a distribution in the same order of magnitude with prolonged immobilization (21.9%) dominated by active neoplasia (16.8%). Elsewhere, Tisserand et al. [8] had a different distribution with a history of thromboembolic venous disease (38%), neoplasia (26%), and COPD (16.7%). Pulmonary Embolism (PE) is a common complication of infections particularly COVID-19, the pandemic of which covered our study period. The triggering cause of PE following acute infection, especially respiratory, is thought to be local inflammation causing local activation of coagulation and vasoconstriction. Endothelial dysfunction, a state of hypercoagulation with activation of platelets and leukocytes is linked to the increased risk of thromboembolic events in COVID-19 [17].

Curative heparin therapy (88%) followed by DOAC/Rivaroxaban (40%) was the main therapeutic choice following the COPE study with heparin therapy in (92.1%) followed upon discharge from the hospital by DOAC (75.6%) and vitamin K antagonists (6.7%) [13]. Heparin therapy is recommended in the acute phase of PE according to guidelines [4]. DOACs have revolutionized the treatment of PE concerning their lower constraint of use. A large clinical trial including 27,023 patients with VTE demonstrated the non-inferiority of DOACs compared to VKAs for the prevention of symptomatic or fatal VTE with a significant reduction in major bleeding rates [18]. Another study showed a significantly lower mortality (p = 0.035) for DOACs compared to other anticoagulants of any kind (37%). Thrombolysis and other modalities are not used as in other studies [13]. This could be related to a lack of access and especially the frequency of contraindications in our fragile and polypathological population for the most part.

A hospital mortality of 44% was recorded under the COPE study [13] which found PE as a cause of death in 48.6% of hospitalized patients. Faure Z et al. [10] noted a mortality of 16% within 30 days, however, Aharoniet al. [3] found mortality after admission in the following proportions: 15.7% within 30 days, 30.5% within 6 months, 40.6% within 1 year and 54.6% within 3 years. The high mortality in our series could be explained by the high frequency of poly pathology (74%), frailty (88%), the complexity of handling anticoagulants, high PESI score at 158.38±48.49 with a risk of death within 30 days considered high in the vast majority (80%) and the absence of codification of the management of the fragile geriatric population. The increasing mortality rates in parallel with the PESI and sPESI scores were corroborated in the literature [19]. In the COPE study, in-hospital and 30-day mortality in patients with high-risk or massive PE exceeded 20% [13]. In the series of Aharoni et al. [3], the risk factors for death at 1 year were: advanced age (p=0.03), dementia (p=0.011), neoplasia (p<0.001), heart failure (p<0/001) and diabetes (p=0.017).

Limitations: our study has limitations related to information and selection bias. This was a retrospective study with missing data. In addition, using the diagnosis of CT angiography as an inclusive criterion may underestimate our prevalence because it excludes patients who cannot benefit from CT angiography (end-stage renal failure, cost, instability, etc.). It is also a small single-center study that does not reflect the real epidemiology of PE in older people in Senegal.

Conclusion     

Pulmonary Embolism (PE) is a common pathology in geriatric settings. Its clinical presentation is polymorphic with frequent semiological atypicalities and geriatric syndromes. The etiological factors are often multiple and intertwined with frequent immobilization requiring a preventive approach associated with anticoagulation and early wheelchair placement. PE is often at high risk of severity in geriatrics with a high mortality rate. Collaboration between specialists (cardiologists, pulmonologists, geriatricians, etc.) and the establishment of a cardio-geriatric sector would allow optimal diagnostic and therapeutic management.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

Assane Sall put forward the idea and conceptualization of the study as well as the collection and analysis of data under the supervision of Mamadou Coumé. Rokhaya Djajheté, Massamba Bâ, and Dalahata Bâ participated in the development of this work. Guarantor of the study: Assane Sall. All authors read and approved the final version of the manuscript.

Tables and figure     

Table 1: distribution of patients according to general characteristics

Table 2: distribution of patients according to clinical manifestations of PE

Table 3: distribution according to imaging abnormalities

Figure 1: distribution of patients according to etiologies: AF; atrial fibrillation, DVT; deep vein thrombosis; COPD; chronic obstructive pulmonary disease, HF: heart failure, AID; autoimmune disease, IBD; inflammatory bowel disease

References