Quality of life 1 month after acute pulmonary embolism in emergency department patients

Objective: The Pulmonary Embolism Quality- of-Life (PEmb-QoL) questionnaire assesses quality of life (QoL) after pulmonary embolism (PE). We aimed to determine whether any clinical or pathophysiologic features of PE were associated with worse PEmb-QoL scores 1 month after PE. Methods: In this prospective multicenter registry, we conducted PEmb-QoL questionnaires. We determined differences in QoL domain scores for four primary variables: clinical deterioration (death, cardiac arrest, respiratory failure, hypotension requiring fluid bolus, catecholamine support, or new dysrhythmia), right ventricular dysfunction (RVD), PE risk stratification, and subsequent rehospitalization. For overall QoL score, we fit a multivariable regression model that included these four primary variables as independent variables. Results: Of 788 PE patients participating in QoL assessments, 156 (19.8%) had a clinical deterioration event, 236 (30.7%) had RVD of which 38 (16.1%) had escalated


INTRODUC TI ON
Pulmonary embolism (PE) is the third most common cardiovascular disease and it has an annual incidence of 0.6 to 1.0 per 1000 in Europe and the United States. 1,2 When a patient presents to the emergency department (ED) or to their primary care physician with a PE, the physician ideally will use a validated risk stratification tool to determine how best to treat the patient based on disease severity. Most risk stratification strategies, however, do not consider early quality of life (QoL; within weeks of patient disposition) as an outcome. 3,4 Functional limitations are common after venous thromboembolism (VTE) and likely impair QoL. [5][6][7] The literature, however, is limited on patient QoL shortly after a PE diagnosis.
Although concerns about PE severity and acute clinical deterioration influence risk stratification, it is unclear if early clinical deterioration impacts subsequent QoL. The presence of abnormal right ventricular (RV) physiology is an important marker of the acute physiologic cardiac burden of PE. The persistence of RV abnormalities and development of chronic thromboembolic pulmonary hypertension can be associated with decreased cardiopulmonary fitness. However, this complication is uncommon. It is reported in just 2%-4% of those with a previous PE. 8 Cardiopulmonary fitness may also be impacted, though, by the treatment delivered. In one study, 35% of intermediaterisk PE patients treated with anticoagulation monotherapy had elevated RV pressure measurements at 6 months versus 7% of those treated with anticoagulation and systemic thrombolysis. 9 The anticoagulation monotherapy cohort had worse 6-month cardiopulmonary fitness compared to the anticoagulation and thrombolysis cohort.
In this study, our primary objective was to determine if any clinical or pathophysiologic features of PE detected during the index PE presentation were associated with patient QoL 1 month post-PE. We used the validated Pulmonary Embolism Quality-of-Life (PEmb-QoL) questionnaire to measure this primary outcome. PEmb-QoL was developed and validated in 2009. 10 The validation study was conducted on patients who had PE diagnosed between 2001 and 2007 and completed the PEmb-QoL between 10 and 91 months after the index PE diagnosis. 11 Our secondary objective was to determine if there was a difference in domain QoL at 1 month between groups of PE patients with an abnormal RV who did or did not receive escalated PE interventions.

Study design and setting
This prospective, observational, multicenter study is a planned analysis of two registry databases that were previously reported. 12  The study was approved by each institution's review board. Initially, informed consent was required for telephone contact. Later the central site's institutional review board (IRB) approved a waiver of written informed consent and allowed patients to be contacted for voluntary participation in the PEmb-QoL survey. With federal funding, a single IRB approved the protocol with waiver of informed consent.

Study population
Each site enrolled men and women 18 years or older, who had acute PE diagnosed within 12 h of ED presentation and were willing to interventions. For those without and with clinical deterioration, social limitations had mean (±SD) scores of 2.07 (±1.27) and 2.36 (±1.47), respectively (p = 0.027). For intensity of complaints, mean (±SD) scores for patients without RVD (4.32 ± 2.69) were significantly higher than for those with RVD with or without reperfusion interventions (3.82 ± 1.81 and 3.83 ± 2.11, respectively; p = 0.043). There were no domain score differences between PE risk stratification groups. All domain scores were worse for patients with rehospitalization versus without. By multivariable analysis, worse total PEmb-QoL scores with effect sizes were subsequent rehospitalization 11.29 (6.68-15.89), chronic obstructive pulmonary disease (COPD) 8.17 (3.91-12.43), and longer index hospital length of stay 0.06 (0.03-0.08).
Conclusions: Acute clinical deterioration, RVD, and PE severity were not predictors of QoL at 1 month post-PE. Independent predictors of worsened QoL were rehospitalization, COPD, and index hospital length of stay. complete the PEmb-QoL questionnaire 1 month later. To be included in the study, the PE had to be confirmed by (1)

Measurements
We studied four primary independent variables: (1) acute clinical deterioration, (2) right ventricular dysfunction (RVD), (3) initial PE risk stratification, and (4) subsequent rehospitalization. Below we define data collection specific to each variable in the registry databases analyzed for this study. As in previous studies, acute clinical deterioration was defined as a composite of death or other predefined adverse events occurring within 5 days of PE diagnosis in the ED. 12,14,15 The adverse events included cardiac arrest, respiratory failure, hypotension requiring fluid bolus or catecholamine support, and new dysrhythmia.
RVD was determined based on point-of-care goal-directed echocardiography (GDE) performed during the index PE ED evaluation by emergency medicine (EM) attendings credentialed in point-of-care ultrasound and EM residents in training. GDE digital video images of the parasternal long-and short-axis apical fourchamber and subcostal four-chamber views were archived. All six study sites had academic EM residency programs with advanced emergency ultrasound fellowship programs directed by the site investigator for this study. As independent adjudicators, site investigators interpreted the archived GDE images using guidelines for RV assessments in PE. [16][17][18] Prior reports have demonstrated the accuracy of our GDE in PE approach compared to comprehensive echocardiography, sharing both high inter-and high intra-rater agreement. 16 Severe RV dilation was determined by an absolute RV basal diameter of >41 mm or, relatively, as RV diameter greater than or equal to the LV basal diameter. Severe RV systolic dysfunction was determined by visual estimation, tricuspid annular plane systolic excursion less than 10 mm, or interventricular septal position being flat or bending toward the LV base. RV dilatation was considered a requirement for determining RVD. A GDE score was assigned 0 points if no severe RV dilatation was found. A score of 1 point was assigned for severe RV dilatation alone, 2 points were assigned for severe RV dilatation combined with septal deviation or severe RV systolic dysfunction, and 3 points were assigned for severe RV dilatation and both septal deviation and severe RV systolic dysfunction. For this study, RVD was a binary variable and present if GDE score was greater than 0. PE risk stratification was determined using the previously reported PE short-term clinical outcomes risk estimation (PE-SCORE) model, which was developed and validated in the registry database. 12 19,20 Subsequent rehospitalization was defined as a hospitalization after the index PE hospitalization but within 30 days of PE diagnosis. We considered rehospitalization to be a surrogate for impaired cardiopulmonary functioning or unresolved acute PE-related cardiopulmonary dysfunction.

Outcome measures
The primary outcome of this report was patient reported QoL. To detect cardiopulmonary fitness and psychosocial impact of recent PE diagnosis, we decided to administer the PEmb-QoL questionnaire assessment 30 ± 3 days after each patient's enrollment date. Reports on QoL and PE have ranged from 30 days to years after the index PE. [21][22][23] We suspect assigning QoL to the end of the expected therapeutic anticoagulation period limits the assessment of PE related QoL to its physiologic treatment. Conducting the QoL assessment 30 days after index PE provides a temporal association of recent PE and QoL and an early opportunity to understand the impact of the PE diagnosis on a patient's well-being and experiences. Furthermore, a 30-day QoL assessment offers an opportunity for providers and patients to be attuned to issues of compliance and the psychosocial stressors and limitations to well-being, which may be addressed with patients and their support systems. We conducted 30-day QoL assessments prospectively.
A dedicated bilingual research coordinator contacted enrolled patients by telephone and administered the 42-item PEmb-QoL questionnaire. English-and Spanish-speaking patients were enrolled using either the English or Spanish version of the informed consent and the PEmb-QoL participation script, its questions, and its responses. Following a standardized introduction, the sequence and wordings of the questions were systematically performed. There are nine main questions on the original validated PEmb-QoL. We preserved the exact wording of questions from the original PEmb-QoL. We reversed the scores of questions 1, 4, 5, and 9 so that increasing scores were associated with worsened QoL. Questions 2 and 3 were descriptive and therefore not used in scoring reports. Questions 2 and 3 focus on the intensity of lung symptoms and the comparison of past versus current lung symptoms. The absence of current lung symptoms was coded as a missing response. For the domain of work-related problems (question 5), a missing score was used if limitations were not specifically due to lung problems.
The PEmb-QoL was organized into six domains with the accompanying score ranges: frequency of complaints (8 to 40), activities of daily living (12 to 39), work-related problems (4 to 8), social limitations (1 to 5), intensity of complaints for pain (1 to 6) and for breathlessness (1 to 6), and emotional complaints (10 to 60). Total PEmb-QoL score was 37 to 164. Each domain score was also transformed into a 100-point scale. Higher scores signified worse QoL.
As reported by Rochat et al., 24 we created a summary PEmb-QoL score by averaging the six domain-transformed scores. We used the raw domain and overall scores for univariate analysis. For ease of interpretation, we used rescaling along 100 points for multivariable analysis. This method is often done and has no effect on the conclusions from statistical inference.

Data analysis
Descriptive statistics, including counts (percentages), means (SDs), and medians (interquartile ranges [IQR]) are reported as appropriate. Univariable analysis of QoL domain and summary scores were compared by predictors, demographic clinical features, comorbid conditions, and echocardiography findings at presentation. We compared the QoL of survivors with abnormalities by GDE who did and did not have reperfusion interventions. Missing data were relatively scarce; we imputed for missing data using imputation by random forest. Imputed values were not used in presenting univariate means and frequencies. We reported missingness for each variable to provide context for how much data was imputed for each variable.
We used raw scores for univariate statistics for each PEmb-QoL domain by our four primary predictors of interest: acute clinical deterioration, RVD (with or without reperfusion intervention within 5 days), PE risk stratification (low-, intermediate-, or high-risk based on PE-SCORE), and subsequent rehospitalization. We used a twosample t-test to compare outcomes by acute clinical deterioration.
We used ANOVA to compare outcomes by RVD and by PE-SCORE risk group. The analysis of ordinal variables using statistical methods for continuous variables yields valid inference. 25,26 We fit independent multivariable linear regression models for each PEmb-QoL domain outcome, as well as a multivariable linear regression model for PEmb-QoL score (the average score for each patient across standardized domain scores on a scale from 0-100).
We then fit multivariable regression models for each domain outcome using a more complete set of variables and used least absolute

RE SULTS
We approached 1823 PE patients of any acuity level between August 2016 and November 2020. As shown in Figure 1, complete PEmb-QoL data were available for 788 patients. Of these, 20 patients (2.5%) did not have interpretable GDE images.  Table 1 shows there were only a few statistically significant differ- For RVD by GDE with or without reperfusion (reference group: no RVD by GDE), Table 2 Table 3 shows PEmb-QoL domain score stratified by initial PE-SCORE risk classification. None of the domains had significant differences between low-, intermediate-, and high-risk PE-SCORE groups. Table 4 shows all PEmb-QoL domain scores and total scores were significantly worse for patients with subsequent rehospitalization versus those without.

Multivariable analysis results
Tables S4-S9 show multivariable linear regression effect size estimates for each PEmb-QoL domain score (transformed on a scale from 0-100). Subsequent rehospitalization and length of stay were significantly associated with higher QoL scores.  Compared to the earlier year, our patients reported reduced QoL 1 month after the index PE. We analyzed the association between clinical testing and clinical outcomes, which are considered pertinent to the early phases of PE care, to subsequent patient-reported QoL.
We found a strong association between comorbidities and reduced QoL. We also found an insignificant relationship between markers of PE severity, such as abnormal RV and subsequent clinical deterioration, with reduced QoL. Therefore, it is plausible that a PE diagnosis may be associated with reduced QoL for any patient regardless of the degree of PE severity.
Our report suggests psychosocial stressors should be considered as both predictors and outcomes by health care providers. With TA B L E 3 Univariate statistics for each domain score by PE-SCORE risk class. Follow-up questions and responses allow patients to expound their responses, potentially revealing nuances and richer information that structured responses may miss. 33 The structured PEmb-QoL questionnaire does have some advantages, though, over semistructured interviews, including objectivity, better time to completion, and the avoidance of biased ratings by interviewers. These advantages are helpful in clinical trials and large observational studies. 5,6 Sixth, we assessed the association of early echocardiography findings of RV dilatation or systolic dysfunction with QoL. We did not repeat echocardiography assessments of survivors. It is plausible that subsequent abnormal findings may have persisted, worsened, or improved as the condition evolved. In addition, the magnitude and direction of any changes from baseline echocardiography findings may be associated with QoL experiences reported by the patient.
Seventh, the strong associations of length of stay and subsequent rehospitalization with PEmb-QoL scores may not be causal.
It is plausible that deconditioning may occur with longer hospitalizations or comorbid conditions like COPD that lead to increased susceptibility to PE recurrence or prolonged recovery from the index PE. It is plausible that length of stay and subsequent rehospitalizations are surrogate predictors of worse QoL.
Finally, our study did not include data collection on the presence or absence of conversations between health care providers and patients (during index PE hospitalization or postdischarge follow-up) or patients' understanding about the illness, its treatment, and guidance with monitoring of fitness. These conversations may have a psychosocial impact on QoL and subsequent rehospitalizations.

CON CLUS IONS
Thirty days after the index pulmonary embolism, our patients re- Anthony J. Weekes takes responsibility for the paper as a whole.

ACK N OWLED G M ENTS
The authors thank Pilar Tochiki, our research coordinator, for scheduling and conducting the quality-of-life questionnaire and for maintaining the database.

FU N D I N G I N FO R M ATI O N
This was supported by the Agency for Healthcare Research and Quality (grant R01HS025979). The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality.

CO N FLI C T O F I NTE R E S T S TATE M E NT
JTN is an ultrasound consultant for Philips. The other authors declare no conflicts of interest.