Prevalence, Clinical Correlates, and Prognostic Significance of Right Ventricular Dysfunction in Acute Pulmonary Embolism: A Retrospective Cohort Study from a Tertiary Referral Hospital in Somalia

1. Introduction

Acute pulmonary embolism (PE) is a common and potentially fatal cardiovascular emergency. Its clinical spectrum ranges from asymptomatic or mildly symptomatic disease to acute right-sided heart failure, obstructive shock, and death. Despite improvements in diagnostic imaging and anticoagulant therapy, PE remains an important cause of cardiovascular mortality and hospital admission worldwide [1,2,3].

The principal determinant of early clinical deterioration in acute PE is the effect of pulmonary arterial obstruction on right ventricular function. An abrupt increase in pulmonary vascular resistance produces right ventricular pressure overload, dilatation, impaired contractility, and reduced forward blood flow. Right ventricular dilatation may also result in interventricular septal displacement, impaired left ventricular filling, decreased cardiac output, systemic hypotension, and, ultimately, circulatory collapse [4,5,6].

Right ventricular dysfunction (RVD) is therefore a central component of contemporary PE risk stratification. It may be identified by transthoracic echocardiography or computed tomography pulmonary angiography (CTPA). Echocardiographic indicators include right ventricular dilatation, reduced tricuspid annular plane systolic excursion, impaired right ventricular systolic function, and interventricular septal flattening. CTPA may demonstrate an increased right-to-left ventricular diameter ratio, reflux of contrast into the inferior vena cava, or interventricular septal deviation [7,8,9,10].

Current international guidelines recommend that patients with confirmed PE undergo early assessment of hemodynamic status, clinical severity, right ventricular function, and cardiac biomarkers. Among patients without overt hemodynamic instability, evidence of RVD may identify an intermediate-risk subgroup that requires closer monitoring and, in selected cases, consideration of advanced treatment strategies [11,12].

The prognostic importance of RVD has been demonstrated in several observational studies and meta-analyses. In hemodynamically stable patients with acute PE, echocardiographic RVD has been associated with approximately a two-fold increase in short-term mortality [13,14]. Similarly, CTPA-derived right ventricular enlargement, particularly an increased right-to-left ventricular diameter ratio, has been associated with adverse outcomes, although the strength of this association varies according to the imaging threshold, patient population, and clinical risk category [15,16,17]. Recent evidence suggests that the presence of multiple abnormal right ventricular parameters may provide greater prognostic discrimination than any single measurement alone [18].

In addition to acute thromboembolic burden, underlying cardiovascular disease may influence the clinical consequences of PE. Hypertension, diabetes mellitus, obesity, and pre-existing heart failure may reduce cardiopulmonary reserve and increase susceptibility to acute right ventricular failure. Active malignancy may further increase the risk of thromboembolism, recurrent events, bleeding, and death [19,20,21]. The interaction between these comorbidities and RVD may be particularly important in settings where patients present late or access to critical care is limited.

The epidemiology of PE in sub-Saharan Africa remains incompletely characterized. Historically, PE was considered relatively uncommon in African populations; however, this perception may have reflected underdiagnosis caused by limited access to CTPA, echocardiography, D-dimer testing, and specialist care [22,23,24]. Hospital-based studies from African countries have reported considerable variation in the prevalence and mortality of PE, with outcomes often influenced by delayed presentation, limited diagnostic capacity, inadequate availability of anticoagulants, and restricted intensive care resources [25,26,27].

Somalia faces substantial challenges in the diagnosis and management of acute cardiopulmonary disease. Access to advanced imaging, echocardiography, cardiac biomarkers, and intensive care services remains limited in many healthcare facilities. At the same time, the country is experiencing an increasing burden of noncommunicable diseases, including hypertension, diabetes mellitus, and cardiovascular disease [28,29]. Despite these challenges, published data on acute PE, PE-associated RVD, and in-hospital mortality in Somalia are scarce.

The present retrospective cohort study aimed to determine the prevalence of RVD and describe its demographic, clinical, and comorbidity-related correlates among adults with CTPA-confirmed acute PE admitted to a tertiary referral hospital in Mogadishu, Somalia. We also evaluated the association between RVD and in-hospital mortality and identified independent predictors of death during hospitalization. We hypothesized that RVD would be common and independently associated with increased in-hospital mortality.

2. Methods

2.1. Study Design and Setting

We conducted a retrospective cohort study at Mogadishu Somali–Türkiye Recep Tayyip Erdoğan Training and Research Hospital, a tertiary referral and teaching hospital in Mogadishu, Somalia. The hospital provides emergency, pulmonary, cardiovascular, intensive care, diagnostic imaging, and specialist services to patients referred from Mogadishu and other regions of the country.

The study included consecutive adult patients admitted with acute PE between 1 January 2020 and 31 December 2023.

2.2. Study Population

Eligible participants were adults aged ≥18 years with acute PE confirmed by CTPA during the index hospitalization.

Inclusion criteria

• Age ≥18 years at presentation.
• Acute PE confirmed by CTPA.
• Admission to the study hospital during the study period.
• Availability of demographic, clinical, imaging, and outcome data.
• Documented in-hospital survival status.

Exclusion criteria

• Nondiagnostic or technically inadequate CTPA.
• Unverified radiological diagnosis of PE.
• Missing information regarding in-hospital survival.
• Documented chronic thromboembolic pulmonary hypertension.
• Previous chronic PE or another chronic pulmonary vascular thromboembolic condition.

After eligibility screening and data cleaning, 293 patients were included in the final analysis.

2.3. Data Collection

Data were extracted from electronic medical records, laboratory databases, radiology reports, and the hospital’s picture archiving and communication system. Information was recorded using a standardized electronic data-collection form.

Collected variables included:

• Demographic characteristics, including age and sex.
• Presenting symptoms, including dyspnea, chest pain, and hemoptysis.
• Clinical signs, including tachycardia, tachypnea, hypoxemia, syncope, and lower-extremity edema.
• Comorbidities, including hypertension, diabetes mellitus, pre-existing heart failure, active malignancy, and smoking history.
• Imaging findings, including CTPA and echocardiographic evidence of RVD.
• Treatment received, including low-molecular-weight heparin, direct oral anticoagulants, and systemic thrombolysis.
• Level of care, including intensive care unit admission.
• Major bleeding complications.
• In-hospital mortality.

Operational definitions

Acute pulmonary embolism

Acute PE was defined as a newly diagnosed intraluminal filling defect involving the pulmonary trunk, main pulmonary arteries, lobar arteries, or more distal pulmonary arterial branches on CTPA.

Right ventricular dysfunction

RVD was defined as structural or functional impairment of the right ventricle documented by transthoracic echocardiography or CTPA. Acceptable indicators included one or more of the following:

• Right-to-left ventricular diameter ratio >0.9 on CTPA or echocardiography.
• Right ventricular enlargement.
• Interventricular septal flattening.
• Tricuspid annular plane systolic excursion <16 mm.
• Contrast reflux into the inferior vena cava.
• Other radiologist- or cardiologist-documented evidence of right ventricular strain or dysfunction.

Because the study was retrospective, the presence of RVD was based on the findings documented in the clinical record rather than on a single uniform imaging protocol.

Hypoxemia

Hypoxemia was defined as an oxygen saturation <90% while breathing ambient air or a partial pressure of arterial oxygen <60 mmHg on arterial blood gas analysis.

Pre-existing heart failure

Pre-existing heart failure was defined as a documented clinical diagnosis of heart failure or a previous left ventricular ejection fraction <40%.

Active malignancy

Active malignancy was defined as an ongoing cancer diagnosis requiring treatment, metastatic disease, or a cancer diagnosis established within the preceding six months.

Hemodynamic instability

Hemodynamic instability was defined as a sustained systolic blood pressure <90 mmHg or a decrease in systolic blood pressure of ≥40 mmHg lasting more than 15 minutes, in the absence of another identifiable cause.

Major bleeding

Major bleeding was defined according to International Society on Thrombosis and Haemostasis criteria.

Treatment

Patients received anticoagulation or systemic thrombolysis according to the treating clinician’s assessment and the available institutional resources. Low-molecular-weight heparin was the principal initial anticoagulant. Direct oral anticoagulants were used when clinically appropriate and available.

In the absence of routinely documented formal ESC or AHA risk categories, systemic thrombolysis was generally reserved for patients with persistent hemodynamic instability or severe RVD accompanied by refractory hypoxemia and clinical deterioration. Accordingly, comparisons of treatment between groups were interpreted as descriptive because of the potential for confounding by indication.

Outcomes

The primary outcome was in-hospital mortality during the index admission. Secondary outcomes included the prevalence of RVD, clinical correlates of RVD, intensive care admission, treatment patterns, and major bleeding.

Statistical analysis

Statistical analyses were performed using Stata version 17.0 (StataCorp LLC, College Station, TX, USA). All tests were two-sided, and p<0.05 was considered statistically significant.

Continuous variables were assessed for distribution using the Shapiro–Wilk test. Normally distributed variables were summarized as means and standard deviations, whereas non-normally distributed variables were summarized as medians and interquartile ranges. Categorical variables were reported as frequencies and percentages.

Patients were compared according to RVD status and in-hospital survival. Pearson’s chi-square test was used for categorical variables, with Fisher’s exact test applied when expected cell counts were small. Continuous variables were compared using the independent-samples t-test or Mann–Whitney U test, as appropriate.

Multivariable logistic regression was used to identify independent predictors of in-hospital mortality. Candidate variables associated with mortality at p<0.25 in bivariate analyses were considered for inclusion in the initial model. Backward stepwise elimination was then performed, retaining variables with p<0.05 in the final model. Multicollinearity was assessed using variance inflation factors. All retained variables had variance inflation factors <2.1. Model calibration was assessed using the Hosmer–Lemeshow test, which indicated adequate fit (p=0.72). Results are presented as adjusted odds ratios with 95% confidence intervals.

Ethical approval

The study was approved by the Institutional Review Board of Mogadishu Somali–Türkiye Recep Tayyip Erdoğan Training and Research Hospital (approval reference: MSTH/18618). The requirement for informed consent was waived because the study used retrospective, de-identified clinical records.

3. Results

3.1. Baseline Characteristics

A total of 293 adults with CTPA-confirmed acute PE were included. The mean age was 59.2 years, and 201 participants (68.6%) were female. The largest age group was 60–79 years, comprising 127 patients (43.3%), followed by patients aged 18–39 years (24.6%), 40–59 years (21.2%), and ≥80 years (10.9%).

Hypertension was present in 114 patients (38.9%), diabetes mellitus in 80 (27.3%), pre-existing heart failure in 39 (13.3%), active malignancy in 26 (8.9%), and a smoking history in 18 (6.1%). Overall, 210 patients (71.7%) required intensive care admission.

Table 1. Baseline characteristics of patients with acute pulmonary embolism.

Characteristic	Category	n	%
Age group	18–39 years	72	24.6
	40–59 years	62	21.2
	60–79 years	127	43.3
	≥80 years	32	10.9
Sex	Female	201	68.6
	Male	92	31.4
Hypertension	Yes	114	38.9
Diabetes mellitus	Yes	80	27.3
Pre-existing heart failure	Yes	39	13.3
Active malignancy	Yes	26	8.9
Smoking history	Yes	18	6.1
Intensive care admission	Yes	210	71.7

Table 1. Baseline characteristics of patients with acute pulmonary embolism.

Characteristic	Category	n	%
Age group	18–39 years	72	24.6
	40–59 years	62	21.2
	60–79 years	127	43.3
	≥80 years	32	10.9
Sex	Female	201	68.6
	Male	92	31.4
Hypertension	Yes	114	38.9
Diabetes mellitus	Yes	80	27.3
Pre-existing heart failure	Yes	39	13.3
Active malignancy	Yes	26	8.9
Smoking history	Yes	18	6.1
Intensive care admission	Yes	210	71.7

3.2. Prevalence and Correlates of RVD

RVD was identified in 110 patients (37.5%). The prevalence of hypertension was higher among patients with RVD than among those with preserved right ventricular function (46.4% vs. 34.4%; p=0.042). Pre-existing heart failure was also more frequent in the RVD group (19.1% vs. 9.8%; p=0.024).

Although active malignancy was more common among patients with RVD than among those without RVD, the difference was not statistically significant (12.7% vs. 6.6%; p=0.081). Patients with RVD were substantially more likely to require intensive care admission than patients with preserved right ventricular function (92.7% vs. 59.0%; p<0.001).

Table 2. Baseline characteristics according to right ventricular dysfunction status.

Characteristic	Normal RV, n=183	RVD, n=110	p-value
Female sex	126 (68.9%)	75 (68.2%)	0.905
Hypertension	63 (34.4%)	51 (46.4%)	0.042
Diabetes mellitus	51 (27.9%)	29 (26.4%)	0.779
Pre-existing heart failure	18 (9.8%)	21 (19.1%)	0.024
Active malignancy	12 (6.6%)	14 (12.7%)	0.081
Smoking history	13 (7.1%)	5 (4.6%)	0.377
Intensive care admission	108 (59.0%)	102 (92.7%)	<0.001

Table 2. Baseline characteristics according to right ventricular dysfunction status.

Characteristic	Normal RV, n=183	RVD, n=110	p-value
Female sex	126 (68.9%)	75 (68.2%)	0.905
Hypertension	63 (34.4%)	51 (46.4%)	0.042
Diabetes mellitus	51 (27.9%)	29 (26.4%)	0.779
Pre-existing heart failure	18 (9.8%)	21 (19.1%)	0.024
Active malignancy	12 (6.6%)	14 (12.7%)	0.081
Smoking history	13 (7.1%)	5 (4.6%)	0.377
Intensive care admission	108 (59.0%)	102 (92.7%)	<0.001

3.3. Clinical Presentation, Treatment, and Outcomes

Dyspnea was the most frequent presenting symptom, occurring in 275 patients (93.9%), followed by chest pain in 251 (85.7%) and hemoptysis in 183 (62.5%). Hypoxemia was substantially more common among patients with RVD than among those with preserved right ventricular function (91.8% vs. 58.5%; p<0.001). Syncope was also more frequent in the RVD group (12.7% vs. 1.6%; p<0.001).

Lower-extremity swelling was less frequent among patients with RVD than among those without RVD (21.8% vs. 45.4%; p<0.001). This finding should be interpreted cautiously because peripheral edema may reflect chronic venous or cardiac disease rather than the severity of acute pulmonary vascular obstruction.

Low-molecular-weight heparin was administered to 234 patients (79.9%), direct oral anticoagulants to 47 (16.0%), and systemic thrombolysis with tissue plasminogen activator to 12 (4.1%). Treatment patterns differed significantly according to RVD status, with patients with RVD more likely to receive low-molecular-weight heparin or thrombolysis. This difference likely reflects treatment-selection bias because patients with more severe presentations were preferentially treated with intensive therapies.

Major bleeding occurred in 4 of 38 patients who received thrombolysis (10.5%) and in 2 of 255 patients who received anticoagulation alone (0.8%). The overall in-hospital mortality rate was 17.4%. Mortality was significantly higher among patients with RVD than among those without RVD (29.1% vs. 10.4%; p<0.001).

Table 3. Clinical characteristics and outcomes according to RVD status.

Variable	Normal RV, n=183	RVD, n=110	Total, N=293	p-value
Dyspnea	91.8%	97.3%	93.9%	0.059
Chest pain	83.1%	90.0%	85.7%	0.101
Hemoptysis	61.8%	63.6%	62.5%	0.747
Tachypnea	65.0%	62.7%	64.2%	0.691
Tachycardia	65.0%	69.1%	66.6%	0.475
Hypoxemia	58.5%	91.8%	71.0%	<0.001
Syncope	1.6%	12.7%	5.8%	<0.001
Lower-extremity swelling	45.4%	21.8%	36.5%	<0.001
Positive D-dimer	83.6%	88.2%	85.3%	0.284
Low-molecular-weight heparin	76.0%	86.4%	79.9%	<0.001*
Direct oral anticoagulant	23.5%	3.6%	16.0%
Systemic thrombolysis	0.6%	10.0%	4.1%
In-hospital mortality	10.4%	29.1%	17.4%	<0.001

*The p-value represents the overall comparison of treatment distribution across groups.

Table 3. Clinical characteristics and outcomes according to RVD status.

Variable	Normal RV, n=183	RVD, n=110	Total, N=293	p-value
Dyspnea	91.8%	97.3%	93.9%	0.059
Chest pain	83.1%	90.0%	85.7%	0.101
Hemoptysis	61.8%	63.6%	62.5%	0.747
Tachypnea	65.0%	62.7%	64.2%	0.691
Tachycardia	65.0%	69.1%	66.6%	0.475
Hypoxemia	58.5%	91.8%	71.0%	<0.001
Syncope	1.6%	12.7%	5.8%	<0.001
Lower-extremity swelling	45.4%	21.8%	36.5%	<0.001
Positive D-dimer	83.6%	88.2%	85.3%	0.284
Low-molecular-weight heparin	76.0%	86.4%	79.9%	<0.001*
Direct oral anticoagulant	23.5%	3.6%	16.0%
Systemic thrombolysis	0.6%	10.0%	4.1%
In-hospital mortality	10.4%	29.1%	17.4%	<0.001

*The p-value represents the overall comparison of treatment distribution across groups.

3.4. Factors Associated with In-Hospital Mortality

Fifty-one patients died during hospitalization. Compared with survivors, non-survivors were more likely to have required intensive care admission (88.2% vs. 68.2%; p=0.004), have RVD (62.8% vs. 32.2%; p<0.001), and present with hypoxemia (88.2% vs. 67.4%; p=0.003).

The distribution of age groups also differed between survivors and non-survivors (p=0.042). No statistically significant differences were observed according to sex, hypertension, pre-existing heart failure, active malignancy, or diabetes mellitus.

Table 4. Clinical characteristics according to in-hospital survival status.

Variable	Survivors, n=242	Non-survivors, n=51	p-value
Age 18–39 years	24.8%	23.5%	0.042
Age 40–59 years	18.2%	35.3%
Age 60–79 years	45.0%	35.3%
Age ≥80 years	12.0%	5.9%
Female sex	68.6%	68.6%	0.996
Hypertension	39.3%	37.3%	0.790
Pre-existing heart failure	12.4%	17.6%	0.312
Active malignancy	8.3%	11.8%	0.418
Diabetes mellitus	28.9%	19.6%	0.175
Intensive care admission	68.2%	88.2%	0.004
RVD	32.2%	62.8%	<0.001
Hypoxemia	67.4%	88.2%	0.003

Table 4. Clinical characteristics according to in-hospital survival status.

Variable	Survivors, n=242	Non-survivors, n=51	p-value
Age 18–39 years	24.8%	23.5%	0.042
Age 40–59 years	18.2%	35.3%
Age 60–79 years	45.0%	35.3%
Age ≥80 years	12.0%	5.9%
Female sex	68.6%	68.6%	0.996
Hypertension	39.3%	37.3%	0.790
Pre-existing heart failure	12.4%	17.6%	0.312
Active malignancy	8.3%	11.8%	0.418
Diabetes mellitus	28.9%	19.6%	0.175
Intensive care admission	68.2%	88.2%	0.004
RVD	32.2%	62.8%	<0.001
Hypoxemia	67.4%	88.2%	0.003

3.5. Multivariable Analysis

In the multivariable logistic regression model, RVD was the only variable independently associated with in-hospital mortality. Patients with RVD had more than twice the odds of dying during hospitalization compared with patients with preserved right ventricular function (aOR, 2.55; 95% CI, 1.30–5.04; p=0.007).

Hypoxemia was associated with mortality in the unadjusted analysis but was not independently associated with mortality after adjustment. Similarly, intensive care admission, hypertension, diabetes mellitus, age group, and sex did not retain statistical significance in the final model.

Table 5. Multivariable logistic regression analysis of in-hospital mortality.

Variable	Comparison	aOR	95% CI	p-value
RVD	RVD vs. normal RV	2.55	1.30–5.04	0.007
Hypoxemia	Yes vs. no	2.10	0.59–7.52	0.255
Intensive care admission	Yes vs. no	1.38	0.38–4.97	0.621
Hypertension	Yes vs. no	0.91	0.46–1.80	0.781
Diabetes mellitus	Yes vs. no	0.65	0.29–1.44	0.284
Age 40–59 years	vs. 18–39 years	1.87	0.78–4.48	0.160
Age 60–79 years	vs. 18–39 years	0.85	0.37–1.95	0.705
Age ≥80 years	vs. 18–39 years	0.58	0.14–2.32	0.442
Male sex	vs. female sex	0.98	0.49–1.97	0.964

Table 5. Multivariable logistic regression analysis of in-hospital mortality.

Variable	Comparison	aOR	95% CI	p-value
RVD	RVD vs. normal RV	2.55	1.30–5.04	0.007
Hypoxemia	Yes vs. no	2.10	0.59–7.52	0.255
Intensive care admission	Yes vs. no	1.38	0.38–4.97	0.621
Hypertension	Yes vs. no	0.91	0.46–1.80	0.781
Diabetes mellitus	Yes vs. no	0.65	0.29–1.44	0.284
Age 40–59 years	vs. 18–39 years	1.87	0.78–4.48	0.160
Age 60–79 years	vs. 18–39 years	0.85	0.37–1.95	0.705
Age ≥80 years	vs. 18–39 years	0.58	0.14–2.32	0.442
Male sex	vs. female sex	0.98	0.49–1.97	0.964

4. Discussion

In this retrospective cohort study of 293 adults with CTPA-confirmed acute PE, RVD was present in 110 patients, corresponding to a prevalence of 37.5%. Patients with RVD were more likely to have hypertension and pre-existing heart failure and were substantially more likely to present with hypoxemia, syncope, and a requirement for intensive care admission. The principal finding was that RVD was independently associated with in-hospital mortality. After adjustment for age, sex, hypertension, diabetes mellitus, hypoxemia, and intensive care admission, patients with RVD had 2.55-fold higher odds of in-hospital death than patients with preserved right ventricular function.

The prevalence of RVD observed in this study is comparable to estimates reported in international cohorts. A recent meta-analysis of echocardiographic studies reported RVD in approximately 37.8% of patients with acute PE [14]. Earlier meta-analyses also demonstrated that echocardiographic RVD was associated with an approximately two-fold increase in short-term mortality among hemodynamically stable patients [13,14]. The consistency of our findings with these reports suggests that acute right ventricular involvement is common across different healthcare systems and patient populations.

However, the prevalence of RVD should be interpreted in the context of the diagnostic methods used in this study. RVD was identified using either echocardiography or CTPA because both modalities were part of routine clinical practice at the study institution. Although this approach reflects real-world diagnostic practice in a resource-constrained setting, it may have introduced measurement heterogeneity. Differences in imaging quality, timing of assessment, reporting practices, and diagnostic thresholds may affect the observed prevalence of RVD. CTPA-derived right-to-left ventricular diameter ratio is useful for identifying right ventricular strain, but its prognostic performance may differ according to the selected cutoff and the clinical risk profile of the patient [15,16,17].

The association between RVD and mortality was clinically substantial. In-hospital mortality was 29.1% among patients with RVD compared with 10.4% among those with preserved right ventricular function. This finding is consistent with the established pathophysiology of acute PE. Sudden obstruction of the pulmonary vascular bed increases right ventricular afterload and may cause right ventricular dilatation, reduced contractility, impaired left ventricular filling, and systemic hypoperfusion. When compensatory mechanisms fail, patients may develop hypotension, obstructive shock, multiorgan dysfunction, and death [4,5,6].

Our adjusted findings are also consistent with previous evidence. Echocardiographic RVD has been associated with increased short-term mortality even among patients who are initially hemodynamically stable [13,14]. In addition, CTPA findings of right ventricular enlargement and septal displacement have been associated with mortality and adverse clinical outcomes [15,16,17,18]. The present study extends these observations to a Somali tertiary-care population and demonstrates that RVD remains prognostically important despite differences in diagnostic infrastructure, treatment availability, and patient characteristics.

Patients with RVD had higher prevalences of hypertension and pre-existing heart failure. The association with heart failure is clinically plausible because impaired baseline cardiac function may reduce the ability of the right ventricle to tolerate an acute increase in pulmonary vascular resistance. Even a moderate thromboembolic burden may therefore precipitate clinically important right-sided failure in patients with limited cardiopulmonary reserve. Hypertension may also represent a marker of the broader cardiovascular risk profile in this population.

Active malignancy was more common among patients with RVD, although the difference did not reach statistical significance. Malignancy is associated with venous thromboembolism through several mechanisms, including systemic inflammation, activation of coagulation pathways, endothelial dysfunction, immobility, and treatment-related risk [20,21]. The lack of statistical significance in the present study may be explained by the relatively small number of patients with active malignancy and limited statistical power. Larger multicenter studies are needed to determine whether malignancy modifies the relationship between RVD and mortality in Somali patients with PE.

Hypoxemia and syncope were strongly associated with RVD. Hypoxemia in acute PE may result from ventilation–perfusion mismatch, reduced pulmonary perfusion, impaired cardiac output, and coexisting pulmonary disease. Syncope may indicate a transient or sustained reduction in cerebral perfusion caused by severe right ventricular strain and reduced systemic blood flow. These clinical features may be particularly useful in resource-limited settings, where formal risk scores, cardiac biomarkers, and urgent echocardiography may not be consistently available.

Although hypoxemia was significantly associated with mortality in the unadjusted analysis, its association was attenuated after adjustment for RVD and other covariates. This may reflect the close pathophysiological relationship between hypoxemia and right ventricular strain. RVD may provide a more direct indication of the severity of pulmonary vascular obstruction and its effect on cardiac performance, whereas oxygen saturation is influenced by several pulmonary and systemic factors. Nevertheless, the confidence interval for the adjusted association with hypoxemia was wide, and limited statistical power or residual confounding cannot be excluded.

Intensive care admission was more common among patients with RVD and among those who died. However, intensive care admission was not independently associated with mortality in the adjusted model. This finding should not be interpreted as evidence that intensive care treatment is ineffective or harmful. Rather, intensive care admission is likely to be a marker of disease severity and may be influenced by clinical deterioration, bed availability, referral patterns, and treatment decisions. Adjustment for variables closely related to severity may also have attenuated its independent association with mortality.

Treatment comparisons should be interpreted cautiously. Patients with RVD were more likely to receive low-molecular-weight heparin or systemic thrombolysis. This pattern most likely reflects confounding by indication, because clinicians preferentially administered more intensive treatment to patients with severe clinical presentations. The observational design, the limited number of patients who received thrombolysis, and the absence of standardized treatment allocation prevent causal conclusions regarding treatment effectiveness or safety. The bleeding events observed in this study should therefore be regarded as descriptive findings.

The overall in-hospital mortality rate was 17.4%. This rate is clinically important and may reflect several interacting factors, including delayed presentation, limited access to early diagnostic imaging, restricted availability of echocardiography and cardiac biomarkers, limited intensive care capacity, and inconsistent use of formal PE risk-stratification tools. Mortality comparisons with high-income countries should nevertheless be made cautiously because outcomes are influenced by case mix, referral patterns, disease severity, diagnostic criteria, treatment availability, and the timing of presentation [22,23,24,25,26,27].

This study has several strengths. It provides novel data on PE-associated RVD from Somalia, includes a relatively large cohort for a single-center study in a resource-constrained environment, uses CTPA-confirmed PE as the inclusion criterion, and evaluates a clinically meaningful outcome. In addition, the study examines both the prevalence of RVD and its association with in-hospital mortality while accounting for several important demographic and clinical covariates.

Several limitations should be acknowledged. First, the retrospective design limits causal inference and creates the possibility of incomplete documentation and information bias. Second, RVD was assessed using echocardiography or CTPA, and the imaging protocols and reporting practices may not have been uniform. Third, standardized measurements such as TAPSE, right-to-left ventricular diameter ratio, right ventricular fractional area change, troponin, and natriuretic peptides may not have been available for every patient. Fourth, formal clinical risk scores, including the Pulmonary Embolism Severity Index and simplified Pulmonary Embolism Severity Index, were not routinely documented. Fifth, treatment decisions were individualized and influenced by disease severity, creating potential confounding by indication. Sixth, the study was conducted at a single tertiary referral hospital, limiting generalizability to rural facilities, secondary hospitals, and patients who died before reaching hospital care. Finally, the number of deaths was relatively small in relation to the number of candidate predictors, and the multivariable model may therefore have limited precision and potential for overfitting.

Despite these limitations, the findings have important practical implications. In resource-limited hospitals, focused cardiac ultrasound and targeted echocardiographic assessment may provide rapid prognostic information when formal risk scores, biomarkers, or advanced monitoring are unavailable. A systematic assessment of right ventricular size and function may facilitate earlier recognition of high-risk patients, guide decisions regarding intensive monitoring, and support more rational allocation of critical care resources. This approach should complement, rather than replace, assessment of hemodynamic status, clinical severity, comorbidities, laboratory findings, and available imaging.

Future multicenter studies in Somalia and other countries in the Horn of Africa should prospectively evaluate standardized echocardiographic parameters, cardiac biomarkers, validated PE risk scores, treatment strategies, and short- and long-term outcomes. Such research would help determine whether locally adapted risk-stratification pathways can improve the early recognition and management of high-risk PE.