Introduction
As a result of demographic changes with increased life expectancy and further improvements in medical care, the average age of emergency care patients is expected to increase and people aged ≥ 80 years will represent an increasing proportion of patients [
1]. Therefore, these patients should receive special consideration in studies. However, it is notable that older patients tend to be underrepresented in clinical trials [
2].
The prevalence of pulmonary artery embolism (PE) correlates positively with age [
3]. Hence, due to demographic changes, more patients with suspected PE are expected to present to the emergency department (ED). As part of the diagnostic process, it is recommended that a pretest probability (PTP) is determined prior to computed tomography pulmonary angiography (CTPA) with, for example, the Wells representing a well-established diagnostic tool, regardless of age [
4]. Using the two-stage classification, PE can be expected in about 8.4% of cases in the “PE unlikely” category and in 34.4% in the “PE likely” category [
5]. However, it is problematic that particularly in the group of patients with low PTP, only a few patients actually will have confirmed PE [
6‐
8], leading to frequent unnecessary CTPA in this patient group. Efforts should therefore be made to reduce unnecessary CTPA in primary assessment. In addition to the PTP, a blood gas analysis (BGA) is regularly performed in clinical practice, and although it does not have a specific role in determining PTP [
4] it often influences further diagnostics. In a previous publication [
9], we were able to show that a significant number of CTPA can be avoided by combining the Wells score with standardized PaO
2 as a calculated product of the BGA even with increased D‑dimer threshold without missing a pulmonary artery embolism. However, as the physiological aspects of the lungs also change with increasing age [
10], the effectiveness of this modified algorithm in elderly patients remains unclear. Therefore, the objective of this study was to evaluate the clinical presentations of patients with pulmonary artery embolism stratified by age, as well as to analyze blood gases and a BGA-optimized PTP in elderly patients with suspected pulmonary artery embolism.
Discussion
With this study we can make an important contribution to an optimized ED management of patients aged ≥ 80 years with suspected or confirmed PE who are currently underrepresented in clinical trials [
2]. First, we characterize clinical findings and prognostic parameters in octogenarians with confirmed PE compared to younger patients. Overall, elderly patients with PE are characterized by higher clinical risk markers and elevated mortality rates compared to younger patients. Moreover, to our knowledge, this is the first study to investigate a BGA-optimized Wells score in elderly patients using a standardized PaO
2 in order to reduce the number of unnecessary CTPAs. With this study, we were able to demonstrate that in patients with suspected PE but low PTP a significant number of unnecessary CTPAs could be avoided by using an BGA-optimized pretest algorithm even in elderly patients.
The comparison of all baseline characteristics reveals that patients aged ≥ 80 years are significantly more likely to be female. This is likely due to the longer life expectancy of female patients, a finding consistent with other studies of very old patients with PE [
14]. Gómez et al. [
14] found similar results irrespective of the PTP. However, the study also found significant differences in dyspnea, chest pain and hemoptysis, which could not be confirmed by the analysis of our cohort. Nevertheless, these results do not contradict our findings, but rather emphasize the heterogeneity of the clinical presentation of patients with PE [
15]. Our trial data also showed that older patients had less tachycardia, more cardiovascular comorbidity and more oxygen administration. This appears plausible in the context of the increasing number of comorbidities with advancing age [
16] and the higher number of medications in this patient cohort [
17]. Additional documentation of medication upon admission was not conducted in this retrospective study, which would have further strengthened the validity of the results. However, it is not uncommon for elderly patients ≥ 65 years of age to be prescribed β‑blockers [
17], which could explain the reduced heart rate observed in older patients compared to younger ones. Furthermore, elderly patients also have a higher risk profile. Patients ≥ 80 years of age were significantly more likely to have signs of right heart dysfunction in CTPA or echocardiography, elevated cardiac markers (troponin and nt-pro-BNP) and a higher sPESI and were, therefore, assigned to a higher risk class. Accordingly, the in-hospital mortality was also significantly higher in these patients, although the significantly different cohort size and the lack of documentation of the cause of death must also be considered when analyzing the data. Study data also show similar or even higher mortality rates within 30 days in elderly patients ≥ 80 years, even after pulmonary artery embolism has been excluded [
18]. Overall, these results are consistent with previous studies showing that elderly patients with acute pulmonary embolism have higher in-hospital mortality than younger patients [
19].
There is limited data available on very elderly patients with PE. A cut-off of 65 years is often used as a definition for elderly [
15,
20], but this threshold needs to be reconsidered in the context of demographic changes [
21] and the continuously improving medical care. Additionally, diagnostics are challenging regardless of age [
14]. The lack of significant differences in diagnostic and clinical findings between older and younger patients with confirmed PE further emphasizes these difficulties. For example, there was no significant difference in chest pain, dyspnea or ECG documentation when comparing the two age cohorts. Therefore, other parameters like blood gases should be considered. Notably, there was no significant difference in BGA separated by age cohorts. A number of factors appear to be involved in this process. First of all, the group of patients aged ≥ 80 years is relatively small, which must be taken into account during statistical evaluation. Additionally, the age-associated decline in standardized PaO
2, although not statistically significant, is noticeable. This trend is also illustrated by linear regression slopes in the scatter plot, indicating an impact on the effectiveness of the blood gas-optimized algorithm. Follow-up studies with larger cohorts are necessary to verify this trend.
On the other hand, the absence of a statistical difference in PaO
2 should be recognized as an indicator for this parameter’s limitations, as it does not account for potential hypocapnic hyperventilation. This limitation is overcome by the use of standardized PaO
2, i.e., the value of paO
2 adjusted for partial pressure of carbon dioxide. Standardized PaO
2 better reflects the respiratory efforts in hypoxemic conditions and may have greater diagnostic significance in the assessment of suspected PE. The reduction in CTPA examinations by using our BGA-optimized algorithm published previously was lower in the present subanalysis. Given that the current PaO
2 is used to calculate the standardized PaO
2, and this value declines with increasing age due to physiological changes [
22,
23], the significance of using standardized PaO
2 within our algorithm is diminished. Nonetheless, 23.5% of CTPAs could still be avoided in patients over 80 years, compared to 33.2% in those younger than 80 years. Thus, younger patients particularly benefit from reduced exposure to unnecessary X‑rays. However, avoiding nearly one in four CTPA examinations in this elderly patient cohort still represents a relevant resource-efficient reduction of overall CTPA examination.
Limitations
Several aspects should be considered when interpreting our results. For example, the retrospective data were only obtained from single center and, therefore, cannot be generalized. In addition, the group of patients over 80 years of age is comparatively small, especially when looking at BGAs. Although this analysis is based on a large and well-characterized cohort of patients with suspected PE, larger prospective studies are needed to validate our proposed score in patients with suspected PE and low PTP. In addition, the cause of death during hospitalization was not recorded. Therefore, it remains unclear whether this was related to PE or another cause. Moreover, patients with lung disease were also included in our study (table S1). These study data, therefore, represent a clinically relevant but also heterogeneous patient population. However, an influence of the pulmonary diseases on the results cannot be excluded with certainty. Further studies with larger numbers of patients are necessary to investigate this aspect. The present study also represents a significant improvement in the management of patients with suspected pulmonary embolism. However, it is important to recognize the potential for selection bias due to the retrospective nature of the data analysis and the initial exclusion process.
Declarations
The study was approved by the local ethical committee (reference number: 19-318A).
The supplement containing this article is not sponsored by industry.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.