Highlights
- •COVIDLUS is the first study to investigate the association of lung ultrasound findings with physiological recovery in post-COVID patients.
- •Lung Ultrasound is a useful tool to monitor the resolution of COVID-associated lung inflammation in an outpatient setting.
- •CT features of lung fibrosis correlate poorly with lung ultrasound findings at 12 weeks following COVID pneumonitis.
- •Many patients in COVIDLUS report poor quality-of-life despiteresolution of COVID-related CT and lung ultrasound changes.
Abstract
Background
Long-term respiratory effects can occur after COVID-19 pneumonia (CP). The COVID Lung Ultrasound Study (COVIDLUS) aimed to investigate the utility of serial lung ultrasound (LUS) to track functional and physiological recovery after hospitalisation in patients with CP.
Methods
Between April 2021 and April 2022, 21 patients were recruited at discharge (D0). LUS was performed on D0, day 41 (D41) and day 83 (D83). CT Thorax was performed on D83. Lymphocyte count, Ferritin, Lactate Dehydrogenase, Troponin, CRP, and D-dimers were measured at D0, D41 and D83. 6 minute walking test (6MWT) was performed on D83 and quality of life questionnaires and spirometry completed on D41 and D83.
Results
19 subjects completed the study (10 males [52%]; mean age: 52 years [range:37–74]). 1 patient died. LUS scores were significantly higher at D0, compared to D41 and D83 (Mean score:10.9 [D0]/2.8 [D41]/1.5 [D83]; p < 0.0001). LUS scores correlated poorly with CT at D83 (Pearson r2 = 0.28). Mean lymphocyte counts were lower at D0 but increased at D41 and D83. Mean serum Ferritin was significantly lower at D41 and D83, as compared to D0. The mean 6MWT distance was 385 m (130–540 m). Quality of life measures did not differ at D41 and D83. Lung function increased between D41 and D83 with mean increase in FEV1 and FVC of 160 ml and 190 ml respectively.
Conclusion
LUS can monitor the early recovery of lung interstitial changes from CP. The utility of LUS to predict development of subsequent lung fibrosis post-COVID deserves further study.
Keywords
Abbreviations
6MWTSix Minute Walk Test
ARDSAcute Respiratory Distress Syndrome
ATSAmerican Thoracic Society
CPCOVID-19 Pneumonia
CPAPContinuous Positive Airway Pressure
CRPC-Reactive Protein
CTComputerised Tomogram
D0Day 0 (day of hospital discharge)
D41Day 41
D83Day 83
ERSEuropean Respiratory Society
FEV1Forced Expiratory Volume in 1 s
FVCForced Vital Capacity
GGOGround glass opacification
LDHLactate Dehydrogenase
LUSLung Ultrasound
NHSNational Health Service
RT-PCRReverse Transcriptase Polymerase Chain Reaction
SARSSevere Acute Respiratory Syndrome
TLCOTotal lung gas transfer
VO2maxMaximum oxygen uptake
1. Background
Novel coronavirus SARS-CoV-2 is responsible for the COVID-19 pandemic. The disease causes a viral pneumonia, which can lead to respiratory failure, with or without intensive care support. In the acute phase, CT Thorax has a high sensitivity for detecting the early lung changes of COVID-19. The typical CT appearances seen in COVID-19 pneumonia (CP) have been described previously and consists of predominantly peripheral changes with a bilateral but patchy distribution with interspersed spared areas [
[1]
,[2]
].The availability and remoteness of CT can at times limit patient access. Conversely, Lung Ultrasound (LUS) has been shown to be well tolerated at the bedside and can be performed repeatedly with no ionising radiation exposure to the patient. LUS features of acute CP have been reported by many authors in recent months, with findings ranging from single and coalescing B lines to subpleural consolidations and pleural irregularities [
[3]
]. LUS has a higher sensitivity than Chest X-ray [[4]
] and correlates well with CT findings for diagnostic accuracy of CP [[5]
]. Furthermore, LUS has been shown to be useful for monitoring patients in the acute and short-term phase of CP [[5]
].Many patients experience long-term respiratory effects such as interstitial lung disease and pulmonary vascular disease after CP [
[6]
]. The role of LUS monitoring the continued progress of individuals following recovery from CP is unclear. The prospective COVID Lung Ultrasound Study (COVIDLUS) aimed to investigate the utility of serial LUS to track functional and physiological recovery after hospitalisation in patients who had recently contracted COVID-19. The study also assessed the correlation between LUS with CT Thorax at D83 following development of CP.2. Study design & methods
2.1 Patients
Between April 2021 and April 2022, 21 sequential consenting patients were enrolled in the study from a single Acute NHS Trust. All patients were aged 18–80 years with a positive sample for SARS-CoV-2 (as confirmed by RT-PCR from a nasopharyngeal swab) on admission to hospital. All patients received assisted ventilation (CPAP) or high flow nasal oxygen and Dexamethasone during their inpatient episode, as per local guidelines. Patients who were unable to mobilise independently, with visual or cognitive impairment, or who were pregnant were excluded. The study had ethical approval from the NHS Health Research Authority and the Health and Care Research Wales Board of the United Kingdom.
2.1.1 Lung ultrasound
The LUS protocol was adapted from Lichter et al. [
[7]
] and Soldati et al. [[8]
]. LUS was performed using a 12-zone method, with the subject in an upright position. Images from the anterior, lateral, and posterior aspects of both lungs were captured using a portable LUS probe (Butterfly IQ®, Guilford, CT, USA), by 3 trained operators (TKL/NE/AB) on Days 0 (D0), 41 (D41) and 83 (D83) of the study. Each LUS took between 5 and 10 min to perform, with images stored digitally for subsequent analysis.A scoring system was employed for each LUS zone, ranging from 0 to 3, with the total score of 36 per time point. A score of 0 was given in the context of normal lung: a continuous regular pleural line, normal lung sliding and the presence of A-lines. A score of 1 was given when B-lines were seen, and a score of 2 where B-lines were seen coalescing, indicating more severe lung aeration loss. The presence of subpleural consolidations equated to a score of 3. (Figure A). The interobserver agreement in LUS scores at D0 was quantified using the interclass correlation method. The mean of three independent LUS scores was calculated for each subject on D0, D41 and D83 of the study.
2.1.2 CT scanning
Non-contrast high resolution CT Thorax was performed on D83 using a multidetector CT scanner (Toshiba Aquilon One). Images were reconstructed with a 1 mm slice thickness and both soft tissue and lung kernels. Two radiologists (NS, NW) independently scored the CT scans on the extent of involvement using an established semi quantitative scoring system used for COVID-19 proposed by Pan et al. [
[9]
]. Each lobe was scored for the extent of anatomic involvement as follows: 0 (no involvement), 1 (<5%), 2 (5–25%), 3 (26–50%), 4 (51–75%), 5 (>75%). The sum of each lobe then produced a total CT extent score (0-25). Additionally, the presence or absence of features which could represent ‘fibrotic-like changes’ (parenchymal bands, traction bronchiectasis, honeycombing and/or distorted interfaces) was recorded.2.1.3 Inflammatory markers
Full blood count, serum Ferritin, C-reactive protein (CRP), D-dimers, Lactate Dehydrogenase (LDH) and Troponin were obtained on D0, and thereafter at 6 [D41] and 12 [D83] weeks respectively.
2.2 Markers of functional recovery
A 6-minute walk test (6MWT) was performed at D83 with a rating of perceived exertion (modified Borg scale) obtained immediately following the 6MWT.
A health-related quality of life questionnaire (EQ-5D-3L) was completed by each subject on D41 and D83 of the study and daily oximeter readings were measured using a fingertip pulse oximeter (Creative Medical). Subjects were instructed to report any new symptoms at each clinic visit. Spirometry was obtained from subjects on D41 and D83 of the study using BodyBox 5500 (Medisoft, Belgium) and performed according to ATS/ERS guidelines.
2.2.1 Statistical analyses
Correlation of mean LUS scores with measured blood parameters and lung function was performed, with Spearman's rank correlation used to analyse non-parametric data. The correlation between subject CT and LUS scores at D83 was analysed by Pearson's coefficient. The association of lung fibrosis on CT with mean LUS scoreat D83 was further evaluated by Student's t-test. A p value < 0.05 was used to denote statistical significance.
3. Results
3.1 Patients
The baseline demographic data for the recruited patients is provided in Table 1 (Supplemental Data). Of the 21,19 patients completed the study. A 69-year-old male was re-admitted to hospital 15 days following initial discharge with a bacterial pneumonia, and subsequently died from ARDS. One patient did not attend for CT at Day 83 and was excluded from the study.
3.1.1 LUS scores and correlation with lung function and serum biomarkers
The mean LUS scores of patients (n = 19) on D41 and D83 were significantly lower, as compared to D0 of the study (2.9 ±2.1 [D41] and 1.3 ±1.3 [D83]) vs 10.7±3.3 [D0]; p < 0.001) as shown in Fig 1. Similarly, lung function increased significantly between D41 and D83; with an average of 160 mL and 190 mL increase in FEV1 and FVC respectively at these timepoints (Fig 2). There was evidence of a small negative association between TLCO and mean LUS scores at D41, but this association did not reach statistical significance (r2 = −0.43; p = 0.07). Mean lymphocyte counts were lower at D0 but increased significantly post hospitalisation (Mean ± SD: 1.49 ±0.90 × 10 [
[9]
]/L [D0] vs 1.91±0.45 × 10 [[9]
]/L [D41]; p = 0.02) – Fig 3A. Mean serum Ferritin was significantly lower at D41 and D83, as compared to D0 (mean +/SD: 274 ± 218 μg/L [D41]/194 ± 140 μg/L [D83] vs 1272 ± 984 μg/L [D0]; p < 0.005), whereas changes in the concentration of other serum biomarkers (Troponin/LDH/D dimers/CRP) did not change significantly during these time points.3.1.2 LUS interobserver agreement
Inter-observer agreement was high across all LUS scoring time points, with Pearson's correlation coefficient confirming reproducibility and agreement between observers (r2 ≥ 0.87 for D0, D41 and D83 respectively between observers 1&2, 2&3, and 1&3).
3.1.3 Six-minute walk tests and quality of life measures
6MWTs were completed in 18/21 subjects at D83. The mean distance attained by subjects in this study (range) was 385 m (130–540 m). Data for the 6MWT was not available in 3 subjects (2 declined; 1 died). Quality of life measures did not differ significantly between subjects at D41 and D83 of the study; a score of 75 or more on the Visual Analogue Scale of Health (EQ-5D-3L) was observed in 12 (63%) subjects at D83. None of the subjects reported new symptoms at D41 or D83 of the study. Daily resting oxygen saturations were above 94% throughout the study for each subject.
3.1.4 LUS SCORE and CT scores
The mean CT score for subjects at D83 was 7 (Range:1–13). Ground-glass opacity was the most common feature noted on CT (Figure B). No significant correlation was observed between mean LUS scores with CT scores at D83 (r2 = 0.28; p = 0.25) – Fig 4. Whilst the presence of a feature suggesting ‘fibrotic-like changes’ was reported in up to 35% of cases in this study, the changes were usually mild and represented by distorted interfaces. Furthermore, the presence of ‘fibrotic-like changes’ on CT was not found to be correlated with higher LUS scores at D83.
4. Interpretation
In this study, we prospectively evaluated a cohort of carefully characterised patients who had presented with CP, for up to 83 days following hospitalisation. LUS scores in these patients were observed to be significantly higher on the day of hospital discharge using our protocol, as compared to at 6- and 12-weeks post discharge. The reduction in LUS scores following hospital discharge in these subjects was generally seen to mirror improvements in their lung function and serum biomarkers, in particular serum Ferritin. In contrast to the study by Hernández-Píriz and colleagues who found that only 68% of the patients followed up prospectively following CP had full resolution of LUS changes, our study showed resolution in nearly 90% of cases by 3 months (as defined by a LUS score <5) [
[10]
]. The LUS appearances at D83 of our study were mainly characterised by ‘A’ lines, denoting normal pleural artefact. We believe the differences observed in our study compared to others reflect a milder study population in COVIDLUS, as exemplified by the low numbers of study patients who developed ARDS from CP. This may explain why a correlation between LUS score and either extent on CT or the presence of a finding suggesting ‘fibrotic-like changes’ at D83 was not seen at D83 of our study. LUS scores may have correlated better with CT extent of ‘fibrotic-like changes’ at D83, had subjects in our study developed more extensive lung fibrosis following their acute CP.As the COVID pandemic is a recent phenomenon, longer-term sequelae of CP to the lung parenchyma have yet to be established. The temporal changes within the lung parenchyma on CT in acute CP are well-documented [
[11]
,[12]
]. Basal predominant, peripheral ground glass opacification (GGO), consolidation, and linear opacities are the most common findings observed. Several studies have considered these temporal changes over time and question the cause of the CT appearances. Vijaykumar and colleagues prospectively studied a cohort of 80 patients, who were admitted to hospital following CP, with CT at 3 months and 12 months following discharge [[13]
]. GGO and fibrotic-like linear bands were noted on CT at 3 months. Significantly, improvement in these abnormalities was reported at 12 months. The authors of this study as well as others [[14]
], have questioned the validity of these ‘fibrotic-like changes’ that have been observed in follow-up CT studies to date.In COVIDLUS, both radiologists independently agreed on the presence of at least one finding suggestive of ‘fibrotic-like change’ on CT in 18/21 patients. Given the improvements over time in other studies, we are unable to confirm the significance of these features or whether LUS scores may in time prove more useful for the temporal assessment of CP. Based on our findings, we would propose prompt CT evaluation for lung fibrosis in post-COVID patients with a high LUS score after 12 weeks, although clearly more work needs to be done to assess whether any such correlation exists between LUS and CT at a later interval in a more severe cohort.
Whilst many of our subjects in our study reported near complete recovery of their respiratory symptoms by D83 of the study, we were surprised to find 37% of subjects reporting a score of less than 80 by Visual Analogue Scale of Perceived Health on the EQ-5D-3L. The mean observed distance attained by subjects on the 6MWT in this study (385 m) was also far lower than would have been expected for healthy aged-matched individuals without disability [
[15]
]. Our findings indicate that the level of functional disability experienced by patients with CP may be far greater than predicted, based on lung function alone. The findings are consistent with other studies of Post-Acute Covid Syndrome (‘Long-Covid’) where a significant number of individuals with PACS had reported reduced exercise capacity compared to those without. Durstenfeld and colleagues performed a meta-analysis of patients with persistent symptoms at least 3 months after contracting Covid-19 and confirmed a difference of 4.9 mL/kg/min in VO2max in individuals with symptoms consistent with “Long-Covid”, as compared to asymptomatic individuals [[16]
]. Causes for exertional intolerance in these individuals (other than deconditioning) such altered autonomic function, endothelial dysfunction, and possible muscular or mitochondrial pathology were proposed by the authors of this study.The one patient who died shortly following enrolment to our study recorded the highest LUS score on D0, despite other clinical and blood parameters being in the normal range on the day of discharge. We suggest that high LUS scores in this case may represent severe interstitial inflammation caused by COVID-19, or evidence of unresolved infection. Despite early intubation and supportive measures, this patient died shortly after re-admission. Whilst other studies have quantified the extent of LUS findings to support decision-making in acute CP; to our knowledge none of these studies had continued to evaluate the sonographic changes following the resolution of CP.
Despite COVIDLUS being the first study to prospectively investigate the association of lung ultrasound findings with lung physiology and serum biomarkers in post-COVID patients, it has several limitations. Patients in our study were recruited at the height of the COVID pandemic in 2021/22 when UK hospital admission rates for this condition were high; however the number of subjects recruited into this study was relatively small and from a single centre only. The findings from COVIDLUS therefore be less generalisable to other patient groups, and may also have less applicability for newer variants of COVID-19 [
[17]
,[18]
]. In addition, our results need to be interpreted with caution as COVIDLUS was not statistically powered to delineate differences in lung function or CT parameters relative to LUS scores. Nonetheless we suggest that LUS can be used in an outpatient setting, to evaluate serial changes in the lung/pleura in patients following CP. Our adapted LUS protocol demonstrated excellent interobserver agreement at various time points throughout the study and subjects found LUS acceptable as a point-of-care test, without appreciable adverse effects. LUS images can be stored digitally by using the Butterfly IQ+® probe, thereby facilitating remote or retrospective analysis, without appreciable loss of image resolution to identify the key LUS features of this condition.WHO Update on omicron.
https://www.who.int/news/item/28-11-2021-update-on-omicron
Date accessed: November 28, 2021
In summary, our study has shown that LUS can be used repeatedly to track the early recovery of lung interstitial changes from COVID-19 pneumonia without significant adverse effects. The utility of LUS to predict development of subsequent lung fibrosis post-COVID deserves further study.
Declaration of competing interest
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Tuck-Kay Loke reports financial support was provided by Kent Surrey and Sussex NIHR Clinical Research Network, United Kingdom.
Acknowledgements
The authors would like to acknowledge the contribution of the patients and staff of Tunbridge Wells Hospital, without whose support this project would not have been completed.
Appendix ASupplementary data
The following is the Supplementary data to this article.
- Multimedia component 1
APPENDIX 1. FIGURES
Fig. 1LUS scores on D0, D41 and D83 (N = 19).
Show full caption
Mean LUS scores decreased significantly by D41 & D83, as compared to D0 of the study
(Mean LUS scores: 10.9 [D0] / 2.8 [D41] / 1.5 [D83]; p < 0.0001)
Fig. 2Lung function FEV1 (1A), FVC (1B) TLCO (1C) and KCO (1D) measured on D41 and D83 of the study (N = 18).
Fig. 3Lymphocyte count (3A) and Serum Ferritin (3B) from 19 subjects on D0/D41 and D0/D41/D83 respectively
Show full caption
Ferritin was significantly lower at D41 and D83 compared to D0.
Fig. 4Scatterplot of CT scores and LUS scores at D83 (N = 19)
Show full caption
LUS scores correlated poorly with CT scores at D83
(Pearson's correlation coefficient r2 = 0.28).
APPENDIX 2. IMAGES
Fig. ALUS showing (1A) normal appearances, (1B) discrete B lines, (1C) coalescing B lines, and (1D) subpleural consolidation (Images captured using Butterfly IQ+ ®).
Fig. BCT features of ground glass change (indicated by arrows) at D83 following CP.
References
- Radiological findings from 81 patients with COVID-19 pneumonia in Wuhan, China: a descriptive study.Lancet Infect. Dis. 2020 Apr; 20: 425-434
- Temporal changes of CT findings in 90 patients with COVID-19 pneumonia: a longitudinal study.Radiology. 2020 Aug; 296: E55-E64
- Chinese Critical Care Ultrasound Study Group (CCUSG). Findings of lung ultrasonography of novel coronavirus pneumonia during the 2019-2020 epidemic.Intensive Care Med. 2020 May; 46: 849-850
- Point-of-care lung ultrasound is more sensitive than chest radiograph for evaluation of COVID-19.West. J. Emerg. Med.: Integrating Emergency Care with Population Health. 2020; 21
- Diagnosing COVID-19 pneumonia in a pandemic setting: lung Ultrasound versus CT (LUVCT) A multi-centre, prospective, observational study.ERJ Open Res. Jan. 2020; 00539
- Respiratory Follow-up of patients with COVID-19 pneumonia.Thorax. 2020; 75: 1009-1016
- Lung ultrasound predicts clinical course and outcomes in COVID-19 patients.Intensive Care Med. 2020 Oct; 46: 1873-1883
- Proposal for international standardization of the use of lung ultrasound for patients with COVID-19: a simple, quantitative, reproducible method.J. Ultrasound Med. 2020 Jul; 39: 1413-1419
- Chest CT patterns from diagnosis to 1 Year of follow-up in patients with COVID-19.Radiology. 2022; 302: 709-719
- Importance of lung ultrasound follow-up in patients who had recovered from coronavirus disease 2019: results from a prospective study.J. Clin. Med. 2021 Jul 20; 10: 3196
- Six-month follow-up chest CT findings after severe COVID-19 pneumonia.Radiology. 2021; 299: E177-E186
- Organizing pneumonia of COVID-19: time-dependent evolution and outcome in CT findings.PLoS One. 2020; 15 (Nov 11)
- CT lung abnormalities after COVID-19 at 3 Months and 1 Year after hospital discharge.Radiology. 2022; 303: 444-454
- Interstitial lung disease after COVID-19 infection: a catalog of uncertainties.Radiology. 2021; 299: E216-E218
- Six-minute walk distance in healthy subjects: reference standards from a general population sample.Respir. Res. 2022; 23: 83
- Use of cardiopulmonary exercise testing to evaluate long COVID-19 symptoms in adults: a systematic review and meta-analysis.JAMA Netw. Open. 2022; 5
- Assessing transmissibility of SARS-CoV-2 lineage B.1.1.7 in England.Nature. 2021; 593: 266-269
- WHO Update on omicron.https://www.who.int/news/item/28-11-2021-update-on-omicronDate accessed: November 28, 2021
Article info
Publication history
Published online: March 02, 2023
Accepted:
February 27,
2023
Received in revised form:
January 30,
2023
Received:
December 10,
2022
Footnotes
☆All authors listed contributed substantially to the study design, data analysis and interpretation, and the writing of the manuscript.
Identification
Copyright
© 2023 Elsevier Ltd. All rights reserved.
00064-1&id=fx1.jpg){kind=link}
00064-1&id=fx2.jpg){kind=link}
00064-1&id=fx3.jpg){kind=link}
00064-1&id=fx4.jpg){kind=link}
00064-1&id=fx5.jpg){kind=link}
00064-1&id=fx6.jpg){kind=link}