Evaluation of Post-COVID-19 Chest Pain and Dyspnea in Outpatients Using Speckle Tracking Echocardiography

Clinicians are frequently facing patients complaining of post-COVID-19 chest pain and dyspnea. However, it remains to be seen if these symptoms indicate pathology of the cardiovascular system. We aimed to evaluate heart functions in outpatients with post-COVID-19 chest pain and dyspnea, using 2D-speckle tracking echocardiography (2D-STE). This cross-sectional study recruited consecutive patients who presented to cardiology outpatient clinics between June 15 and July 15, 2021. Subjects had recovered from COVID-19 1-2 months prior to admission. ECG, echocardiography including 2D-STE images, were obtained for all patients. Findings were compared with sex and an age-matched control group consisting of 67 healthy adults. A total of 78 patients were included. The median age was 38 (IQR, 34-45) years, and 64.1% were female. There were no signicant differences between the patients and control group regarding laboratory, ECG, and echocardiography ndings. Moreover, left ventricle global longitudinal strain (LVGLS) measurements in both patient and control groups were within the normal ranges and did not show a signicant difference [-20.5 (-21.8- -17.9) vs. -19.8 (-21.4- -18.9), p=0.894]. Post-COVID-19 chest pain and dyspnea are unlikely signs of cardiovascular involvement in outpatient young adults who had not been hospitalized with COVID-19.


Introduction
Post-Acute COVID19 Syndrome (PACS) is de ned as persistent symptoms 3-4 weeks longer than the initial onset of COVID-19 symptoms [1,2]. Two of the most common symptoms, suggestive of cardiovascular involvement, present in PACS are chest pain and dyspnea, affecting 20% of patients recovering from COVID-19 [2]. However, whether these symptoms are a component of the non-speci c PACS milieu or indicate pathology in the cardiovascular system has remained elusive. This uncertainty adds to patients', as well as clinicians' concerns: clinicians, particularly family physicians, are increasingly confronting a signi cant number of patients with post-COVID-19 chest pain and dyspnea [3].
The literature lacks detailed data investigating post-COVID-19 chest pain and dyspnea in outpatients from a cardiological perspective exclusively. To better understand this issue from a cardiological view of point, we aimed to evaluate heart functions using electrocardiogram (ECG) and echocardiography, including 2D-STE, in outpatients presenting with chest pain and dyspnea and in those recovered from COVID-19. Thereby, we could provide informative data to clinicians who are increasingly facing many so cases.

Materials And Methods
Consecutive patients who presented to the cardiology outpatient clinic with persistent chest pain and dyspnea were included in this cross-sectional study. The subjects had recovered from COVID-19 1-2 months before admission. Emergency conditions, and comorbidities that may cause deterioration in LVGLS analysis, were determined as exclusion criteria. Acute coronary syndrome, heart failure (left ventricle ejection fraction <50%), atrial brillation, severe cardiac valve disorders, renal failure (eGFR<30ml/min/1.73m2), severe chronic obstructive pulmonary disease, anemia (hemoglobin level under 11.9 for females and 13.6 for males) and being under 18 years old were the exclusion criteria.
Demographics and laboratory results on admission were recorded. Findings were compared with sex and an age-matched healthy control group consisting of 67 adults. ECG analysis was performed by a blinded cardiologist (M.K), using a standardized comprehensive ECG reading protocol [6]. It included intervals, rate, QRS morphology, premature atrial and ventricular contract, and T-wave abnormalities. Corrected QT interval (QTc) was calculated using the Fridericia formula [7].
Echocardiographic images were obtained using Philips Epiq7 (Philips Healthcare, Inc., Andover, MA, USA) and recorded by standard techniques. According to the American Echocardiography Association guidelines [8], left ventricular end-systole and end-diastolic diameters, left atrium diameter, interventricular septum thickness, left ventricular posterior wall thickness, right atrium and ventricle diameters were measured. Measurements of mitral in ow included the peak early (E-wave) and late (A-wave) diastolic lling velocities and calculation of E/A ratio. Early diastolic mitral, septal and lateral annular velocities (e') were measured in the apical four-chamber view [9]. Left ventricular ejection fraction (LVEF) was measured using the modi ed Simpson's rule [10].
Left ventricular global longitudinal strain (LVGLS) was analyzed by another cardiologist (D.İ), blinded to study data, using the Qlab13 (Philips Healthcare, Andover, Massachusetts) program ( Fig. 1). While the end-diastole is regarded as the peak R wave of the electrocardiogram, end-systole was estimated aortic and LV myocardial Deformation Imaging [11,12]. Demographic, clinical and laboratory characteristics are summarized in Table 1. Of 78 patients, only three had been hospitalized for COVID-19, and the high sensitive troponin T (hs-TnT) level of two patients had been elevated (33 pg/ml and 42 pg/ml, respectively, reference limit < 14 pg/ml) during hospitalization.  The distribution of the symptoms on examination was as follows; atypical chest pain in 59, atypical chest pain and dyspnea in 5, typical chest pain in 3, typical chest pain and dyspnea in 4, and dyspnea in 7 patients. Treadmill exercise test of those with typical chest pain was negative. Besides, no pathological nding was found on the chest x-ray of patients with dyspnea.
There were no signi cant differences between the patients and control group regarding demographic characteristics and comorbidities, including hypertension (HT), diabetes (DM), smoking, hyperlipidemia (HPL9, and asthma. Laboratory ndings were within the normal range and were similar between the groups, except for platelet and C-reactive protein (CRP) levels, which were within normal ranges also (Table 1).
Regarding echocardiographic characteristics, all parameters, including left and right side functions, were found within normal ranges in both groups and did not show a signi cant difference (table). Moreover, LVGLS measurements in both patient and control groups were within the normal ranges and did not show a signi cant difference [-20.5 (-21.8--17.9) vs. -19.8 (-21.4--18.9), p=0.894] ( Table 2).

Discussion
Our study provides data that laboratory and cardiological features, including LVGLS of young adults suffering from post-COVID-19 chest pain and dyspnea, were ordinary and similar to healthy populations'.
Of note, our study comprised only outpatient subjects and young adults beyond crucial cardiovascular risk factors and comorbidities. Moreover, only three had been hospitalized with COVID-19, and only two had been suggestive of cardiac involvement [elevated hs-TnT level) during hospitalization.
To our knowledge, the literature lacks data investigating post-COVID-19 chest pain and dyspnea from a cardiological perspective; in this context, our study is preliminary. On the other hand, many papers are available investigating post-COVID-19 cardiac involvement irrespective of symptoms and reporting discordant results. According to a study [5], in 80 COVID-19 survivors, most heart functions, which had been impaired during hospitalization, improved three months after discharge. However, a quarter of patients still had abnormal LVGLS compared with the initial analysis. Similarly, Özer et al. [13] demonstrated impaired LVGLS values in over half of those who had a myocardial injury during hospitalization and in one-third of all at 1-month follow-up. Dissimilarly to our study, both studies involved an older population (57.7 and 59.9 years, respectively). Besides, they included hospitalized patients, indicating more severe COVID-19 infection, and those with more frequent cardiac risk factors and comorbidities. Another point worth mentioning is that the follow-up analysis was performed irrespective of the presence of symptoms.
On the contrary, there are also studies showing the opposite. A study demonstrated no proof of persistent cardiac dysfunction on echocardiography performed 40 days after hospital discharge following recovery from COVID-19 [14]. Of note, this study did not involve an STE study to identify more subtle myocardial changes. Another research, which is very similar to ours regarding population demographics, is a prospective study of 149 healthcare workers. There were no differences in cardiac magnetic resonance (CMR) characteristics, including cardiac functions, hs-TnT level, and N-terminal pro-BNP at six months post-infection versus age, sex, and ethnicity matched seronegative controls [15]. Similar to our study, the population had relatively fewer comorbidities, and only one patient had severe COVID-19. This report did not include ECG and LVGLS analysis, however.
One of the primary aims of this study was to raise awareness amongst primary care professionals, who are increasingly confronting with many patients with post-COVID-19 chest pain and dyspnea. In reference to the current study, post-COVID19 chest pain and dyspnea in young adults who have not required hospitalization and those without comorbidity and laboratory abnormalities are unlikely due to cardiovascular pathology. Nevertheless, it should be kept in mind that PACS may have signi cant cardiovascular manifestations, particularly in patients who have been hospitalized with COVID-19, the elderly, and those with signi cant cardiovascular risk factors and comorbidities. Further studies compromising elderly and heterogeneous populations with cardiovascular risk factors and comorbidities are needed in this eld.

Limitations
The study was a single-center, and the number of patients included in the study was relatively small. The study population lacked an older population; thus, it was without signi cant cardiovascular risk factors and comorbidities. In addition, only three patients had been hospitalized because of COVID-19, and only two had been suggestive of cardiac involvement (elevated hs-TnT level) during hospitalization. In this regard, the study population did not include patients who had suffered from severe COVID-19 infection.

Conclusion
This study could give helpful insights into the currently mostly enigmatic issue, that post-COVID-19 chest pain and dyspnea are unlikely signs of cardiovascular involvement in outpatient young adults who had not been hospitalized with COVID-19. Figure 1 An example of left ventricle global longitudinal strain speckle tracking of a patient from apical fourchamber view.