Abstract
Background
One of the most prevalent causes of emergency room visits is acute dyspnea. Several etiologies, including cardiac, pulmonary, metabolic, psychogenic etc… may be involved. Acute heart failure (AHF) is among the most common causes. This study aims to evaluate, in patients presenting with acute dyspnea to the emergency departement (ED), the accuracy of a diagnostic approach combining Lung ultrasonography (LUS) and clinical assessment as compared to the traditional AHF diagnostic work-up.
Methods
This is a bi-centric cross-sectional observational study, conducted at the Emergency and Cardiology Department of both the Hedi Chaker Hospital in Sfax and the Habib Thameur Hospital in Tunis for the period extending from 01/07/2022 to 30/09/2023. The diagnostic performance of pulmonary ultrasonography was studied and the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were compared with those of clinical examination, chest X-Ray, NT-pro -BNP and the Transthoracic echocardiography (TTE) which was the reference exam.
Results
The most common cause of acute dyspnea is acute heart failure (79.3 %). LUS had a sensitivity of 94,2 % in diagnosing AHF and a specificity of 77,5 %. Its PPV and NPV were respectively 92 % and 81 %. The area under curbe (AUC) of B-Lines required for the diagnosis of interstitial pulmonary syndrome was excellent (92 %). There was a moderate significant positive correlation between the number of B-Lines and NT-Pro-BNP levels r = 0.51, P < 0.001. Also, there was a very strong significant positive relationship between the pulmonary congestion assessed by LUS and Left atrium – pressure r = 0.788, P < 0.001
Conclusion
LUS is an excellent test both to confirm and exclude the diagnosis of AHF in patients consulting the emergency room for acute dyspnea and therefore deserves to be performed systematically.
Introduction
It is estimated that 26 million people worldwide suffer from heart failure (HF). Its prevalence is increasing In the United States and Europe, HF is responsible for a significant amount of morbi- mortality. ,
Because of the severe dyspnea it generates, Acute Heart Failure (AHF) is among the most common causes of emergency visits
To establish its diagnosis, the European Society of Cardiology suggests a method that includes performing a clinical assessment first, followed by a battery of routine tests, including biological tests, an electrocardiogram (ECG), and a chest X-ray (CXR).
Following this minimal assessment, if the diagnostic suspicion proves to be justified, it is recommended to measure the biological markers of heart failure, namely BNP or NT-pro-BNP. Their elevation should prompt us to perform a trans-thoracic echocardiography, which is the “Reference Exam” for the diagnosis of heart failure, while if they are negative, it would allow us to rule out this diagnosis
This approach, although classic, has its advantages: It has the merit of being systematic, making it possible to consider more differential diagnoses. Nevertheless, it also has limits: the interpretation of the BNP is sometimes difficult because it exposes to many false positives, it is expensive, requires a fairly long waiting time in an environment where the speed of execution and results are critical.
The ideal would therefore be to adopt a diagnostic technique that is more accurate, less expensive, non-invasive and feasible at the bedside. Finally, the technique should combine execution speed and availability.
It is to these requirements that we are going to try to respond by proposing a new diagnostic approach based on the detection of B lines, using pulmonary ultrasonography.
Several observational studies and a recent meta-analysis have suggested that pulmonary ultrasonography has higher diagnostic performance (in terms of sensitivity, specificity, negative predictive value, positive predictive value) for the diagnosis of acute heart failure than standard clinical work-up, CXR, and natriuretic peptides.
Thus, we conducted this study, that aims to evaluate, in patients presenting with acute dyspnea to the ED, the diagnostic performance of pulmonary ultrasonography for the diagnosis of acute heart failure.
Methods
Study design
This is a bi-centric cross-sectional observational study conducted at the Emergency and Cardiology Department of both the Hedi Chaker Hospital in Sfax and the Habib Thameur Hospital in Tunis for a period of 15 months extending from 01/07/2022 to 30/09/2023, which aims to evaluate the diagnostic performance of pulmonary ultrasonography in the diagnosis of acute heart failure.
We included in this study patients with the following criteria: All adult patients (age ≥ 18 years old) consulting for acute dyspnea defined as either: a sudden onset of shortness of breath or on increase in the severity of chronic dyspnea in the previous 48 h.
A probable origin of this dyspnea related to a newly discovered acute heart failure or to a decompensation of an already known heart failure.
A cardiac origin of dyspnea is considered if there was at least one physical sign in favor: sinus tachycardia, crackling rales, distended jugular veins, hepatojugular reflux, hepatomegaly, lower limb edema.
We did not include in this study patients with the following criteria: Dyspnea following trauma, Body Mass Index >40. In fact, for such patients it’s technically challenging to obtain good acoustic windows.
History of pneumothorax, lobectomy or lung cancer; history of severe interstitial lung disease that can lead to misinterpretation of pulmonary ultrasonography.
Study protocol
Consecutive sampling was the method used in our clinical trial. Each patient who consulted in the emergency room for acute dyspnea and who responds to the inclusion criteria, underwent a minimal battery of tests, which included a thorough clinical examination (Blood pressure, pulse rate, oxygen saturation, respiration rate, Body Mass Index), an ECG, a standard biological evaluation (Creatinine, urea, C-reactive protein, complete blood count, electrolytes. D- dimer, arterial blood gas analysis if needed, ultra-sensitive cardiac troponins), and a chest X-ray.
If the patient was eligible for inclusion in the study, in addition to the previous examinations, he would benefit from a biological sample for NT-pro-BNP, a pulmonary ultrasonography and a transthoracic echocardiography using A « General Electric 6S-D » sector array probe with a frequency range from 3 to 8MHz in the Hedi Chaker Hospital and a « Sonosite M-Turbo 2D Echo » sector array probe with a frequency of 7.5 MHz in the Habib Thameur Hospital.
The operator performing the pulmonary ultrasonography was blinded to the results of the X-ray, NT-pro-BNP, and transthoracic echocardiography in order to avoid influencing his judgment.
The patients were subsequently divided into two groups depending on the presence or not of pulmonary congestion (defined by 3 B-Lines or more) on LUS.
For the patients included in the study, we collected from medical records standard demographic data, Cardiovascular risk factors, history of cardiac, pulmonary or renal disease, history of cardiac surgery or cardiac implantable devices, medications.
Pulmonary ultrasonography
Pulmonary ultrasonography should be performed as soon as possible within 2 h of the patient’s admission to the emergency room.
It was requested only if the diagnostic of acute heart failure was suspected and the operator did not have to be aware of the other clinical and para-clinical information.
B-lines were described as distinct, respiration-synchronized vertical hyperechoic reverberations that resembled laser beams and extended from the top to the bottom of the screen. ,
The same probe used for echocardiography was utilized to perform pulmonary ultrasonography on each subject while they were semi-supine according to international recommendations
We evaluated eight chest zones, two parasternal chest scans and two scans of the anterior and lateral basal chest was obtained on the right and left hemi thoraxes.
We adopted the 8-zone protocol as it is recommended by the 2012 International evidence-based recommendations on pulmonary ultrasonography suggesting performing LUS with either 8 or 28 chest zones Furthermore, the 8 zone- protocol was more suitable and adapted for patients admitted in the emergency department as they were most of the time on acute dyspnea.
A positive region was defined by the presence of three or more B-lines in transverse intercostal plane. , Acute interstitial pulmonary syndrome was defined, according to the International Consensus Conference on LUS, by the presence of two or more positive regions in each hemithorax .
Trans-thoracic echocardiography
A cardiologist performed an echocardiography within 12 h of the patient’s admission. End systolic and end diastolic diameters as well as left ventricular (LV) ejection fraction (EF) were determined using apical two-and four chamber views by Simpson biplane formula.
Diastolic function was assessed from the pattern of mitral inflow by pulsed wave Doppler.
Mitral annular early diastolic velocity (e’) was assessed at the septal and lateral sites of the mitral annulus using tissue Doppler imaging. E/A ratio, e’ wave peak velocity, E/e’ were calculated.
Pulmonary arterial systolic pressure (PASP) was estimated using tricuspid regurgitation velocity, and Inferior Vena Cava (IVC) diameter and collapse which were analyzed by epigastric scan according to recent guidelines Tricuspid Annular Plane Systolic Excursion (TAPSE) was obtained from M-mode recordings of the lateral tricuspid annulus. Results were normalized for body surface area (in square meters) where appropriate.
Patients were classified according to LVEF, reduced EF (LVEF ≤ 40 %) midly reduced (40 %<FEVG<50 %) and preserved EF (LVEF ≥ 50 %)(5).
Following the collection of the various echocardiographic parameters, the most recent American Society of Echocardiography recommendations for the evaluation of left ventricular filling pressures were used.
Statistical analysis
Statistical analyses was carried out using SPSS software version 23 (SPSS Inc., Chicago. Illinois, the USA). We expressed categorical variables as percentages, and continuous variables as mean values (±standard deviation [SD]) when the distribution is normal or medians with semi-interquartile ranges (SIQR) when it is not normal. Normally distributed continuous variables were compared using the Student t-test (independent sample t-test for comparison between the 2 groups, paired sample t- test for self-comparison); non-normally distributed continuous variables were analyzed by non-parametric test (Mann-Whitney U test for independent series and Wilcoxon test for paired series). When the application conditions is validated, categorical data was analyzed using the Chi2 test of Pearson, otherwise Fisher exact test.
The Phi Coefficient was used to understand the strength of the relationship between two binary variables.
We estimated the sensitivity (SE), specificity (SP), positive and negative predictive value (PPV and NPV, respectively) for dichotomous tests and by the Area Under Curve (AUC) for quantitative tests.
They were calculated using test for the presence and absence of heart failure. They were also calculated for each kind of non-invasive methods namely pulmonary ultrasonography NT-Pro-BNP X-ray and trans-thoracic echocardiography.
The Pearson’s correlation coefficient was the test used to measure the statistical relationship between two continuous variables.
P-value of 0.05 was considered significant.
Ethical consideration
The medical secrecy regulation as applied to all data acquired. Also, no additional costs were imposed on the patient to carry out this study. The practitioner taking charge of the patient had access to all the clinical and paraclinical data. As a result, this could in no way influence the treatment of the patient. Last but not least, this study was conducted in agreement with the ethics committee of the hospital and the university.
Results
Descriptive study
Baseline characteristics
Over a 15-month period, a convenience sample of 180 patients were identified and eligible (120 at the Hedi Chaker Hospital and 60 at the Habib Thameur Hospital), 20 of these were excluded leaving 160 for analysis.
The median age was 68.88 ± 12.77 years with a minimum of 22 and a maximum of 90.
The proportion of male patients in our population was 53.75 % while the percentage of female patients was 46.25 %.
The most common cause of acute dyspnea is acute heart failure with a percentage of 79.3 %.
Analytic study
Patients characteristics
Patient characteristics were studied depending on the presence or not of pulmonary congestion (defined by 3 B-Lines or more) on LUS
Continuous variables’ description, divided into patients with B-Lines ≥ 3 and B- Lines <3 group, is reported in Table 1 .
| All ( n = 160) | B-Lines ≥3 n = 122 (76.3 %) | B-Lines <3 n = 38 (23.8 %) | P Value | |
|---|---|---|---|---|
| Age (years) | 68.8 ±12.7 | 69.63± 12.6 | 67.08±13.3 | 0.7 |
| Male | 86 | 71 (58.2 %) | 15 (39.5 %) | 0.062 |
| Diabetes mellitus | 65(40.6 %) | 51 (41.8 %) | 14 (36.8 %) | 0.706 |
| Dyslipidemia | 62(38.8 %) | 50 (41 %) | 12 (31.6 %) | 0.344 |
| Hypertensio n | 89(55.6 %) | 73 (59.8 %) | 16 (42.1 %) | 0.063 |
| Smoking | 38(23.8 %) | 30 (24.6 %) | 8 (21.1 %) | 0.828 |
| CAD | 37(23.1 %) | 32 (26.2 %) | 5 (13.2 %) | 0.123 |
| HF | 70(43.8 %) | 57 (46.7) | 13 (34.2 %) | 0.194 |
| SVA | 77(48.1 %) | 67 (54.9 %) | 16 (42.1 %) | 0.195 |
| VHD | 52(32.5 %) | 42 (34.4 %) | 10 (26.3 %) | 0.430 |
| COPD | 24 (150) | 15 (12.3 %) | 9 (23.7 %) | 0.116 |
| CRF | 16(10.1 %) | 8 (6.6 %) | 8 (21.1 %) | 0.02 |
| Chronic Renal Failure | 44(27.5 %) | 34 (27.9 %) | 10 (26.3 %) | 1 |
Clinical and paraclinical characteristics
Table 2 compares the clinical and paraclinical indicators that could lead to the diagnosis of acute heart failure decompensation.
| All ( n = 160) | B-Lines ≥3 n = 122 (76.3 %) | B-Lines <3 n = 38 (23.8 %) | P value | |
|---|---|---|---|---|
| Clinical Congestion | 147 (91.9 %) | 119 (97.5 %) | 28 (73.7 %) | 0.000 |
| Pulmonary Congestion in CXR | 108 (67.5 %) | 98 (80.3 %) | 10 (26.3 %) | 0.000 |
| B-Lines | 25.55 ± 15.58 | 31.13±12.5 | 7.63±9.86 | 0.000 |
| Elevated LA- Pressure | 117 (73.1 %) | 113 (92.6 %) | 4 (10.5 %) | 0.000 |
| NT-Pro-BNP (pg/mL) | 14125±10629 | 15584±9799 | 9442±11912 | 0.002 |
| Troponins (ng/l) | 1254±4545 | 1179±4175 | 1493±5631 | 0.712 |
| Hb (g/dl) | 11.44±2.37 | 11.46±2.29 | 11.39±2.64 | 0.883 |
| WC (mm3) | 8950±4099 | 9015±4040 | 8740±4330 | 0.718 |
| CRP (mg/L) | 49.99 ±78.48 | 49.82±77.21 | 50.55±83.53 | 0.960 |
| Urea (mmol/L) | 12.20±9.16 | 11.53±8.33 | 14.37±11.29 | 0.157 |
| Creatinine (µmol/L) | 136.59±111.81 | 139±118 | 128±85 | 0.620 |
| LVESD | 38.20 ±10.94 | 40.43 ±10.96 | 31.05±7.23 | 0.000 |
| LVEF | 44.09 ±14.47 | 41.11±14.31 | 53.66±10.36 | 0.000 |
| LAA | 29.10 ±14,91 | 30.42±16.32 | 24.87± 7.76 | 0.045 |
| SPAP | 57.37±15.80 | 58.58±13.15 | 53.26 ± 22.39 | 0.188 |
| Dilated IVC | 87 (54.4 %) | 73 (59.8 %) | 14 (36.8 %) | 0.016 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
