Front-door thoracic ultrasound in patients with community-acquired pneumonia to diagnose and predict pleural infection: a prospective study

Extract

Pneumonia remains a significant public health concern, affecting a substantial portion of the UK adult population annually. This study investigated the utility of front-door thoracic ultrasound (TUS) in the early identification of parapneumonic effusion (PPE) and its subsequent progression to pleural infection in patients with community-acquired pneumonia (CAP).

Shareable abstract

The incidence of parapneumonic effusions is higher than previously estimated when thoracic ultrasound is used as a diagnostic tool. TUS may have a role in prognostication and prediction of pleural infection, and this requires further evaluation. https://bit.ly/3UkV7nQ

To the Editor:

Pneumonia remains a significant public health concern, affecting a substantial portion of the UK adult population annually. This study investigated the utility of front-door thoracic ultrasound (TUS) in the early identification of parapneumonic effusion (PPE) and its subsequent progression to pleural infection in patients with community-acquired pneumonia (CAP).

41 CAP patients were enrolled at a District General Hospital in the UK, and TUS was utilised within 24 h of admission.

The study revealed a higher incidence of PPE (63%) than previously reported. Forty-six percent of patients with PPE had effusions detected solely by TUS, which were not visible on chest X-rays (CXR). Additionally, all patients who died during admission had TUS-detected effusions, with three of them not showing effusion on the corresponding CXR. However, there were no specific TUS features that correlated with pleural infection development or resolution. The study also indicated higher mortality in patients with pleural effusion, although statistical significance was not reached due to the sample size.

This pilot study underscores the potential of TUS in identifying previously undetected PPE in CAP patients. However, further research is needed to evaluate the role of TUS in pneumonia management to better understand the natural history of PPEs and their implications for patient outcomes.

Approximately, 0.5–1% of the UK adult population develops pneumonia each year [1]. Of these, it is estimated that 20–57% have an associated parapneumonic effusion (PPE) at presentation [2], of which 5–10% progress to pleural infection [3]. Both PPE and pleural infection are associated with high morbidity and mortality and carry a significant health burden [4, 5].

The data on incidence of PPE is derived from old studies that used chest radiograph for diagnosis, which is known to have poor sensitivity for pleural effusion [6]. Thoracic ultrasound (TUS) is better at detecting pleural effusion [6], but its role in early identification of PPE to predict the development of subsequent pleural infection has not previously been studied. We explored the use of TUS in the front-door assessment of patients with community-acquired pneumonia (CAP), particularly if this aided earlier identification of pleural infection.

Over a 4-week period, consecutive adult patients (age>18 years) admitted with CAP to a District General Hospital in Northern England underwent TUS within 24 h of admission. CAP was defined as an acute infective illness associated with new consolidation on CXR or computed tomography (CT) scan, and not attributed to COVID-19. TUS findings such as size of effusion (if present), echogenicity, presence or absence of lung consolidation or collapse, presence of septations and loculations, and diaphragmatic features (if any) were documented. Pleural effusions measuring 2 cm or more were sampled [7]. The demographic, clinical, microbiological and biochemical blood parameters were recorded at admission and every 48 h during the in-patient stay until the point of discharge. Clinical progress was followed through the electronic patient record for the duration of the in-patient stay, including radiographic development or worsening of pleural effusion detected by CXR. Clinically relevant outcomes including in-patient mortality and length of hospital stay were recorded from the hospital record.

Over 4 weeks, 41 patients with CAP were admitted (age range 40–90, median age 74). The results are shown in figure 1. In total, 26 out of 41 patients (63%, 95% CI 47%–78%) had a pleural effusion at presentation detected by either TUS alone, or TUS and CXR. Of these, 19 (46%, 95% CI 31%–63%) had an effusion detected on TUS alone which was not visible on the corresponding CXR.

FIGURE 1.

Results from the prospective study on front-door thoracic ultrasound in patients with community-acquired pneumonia to diagnose and predict pleural infection. TUS; thoracic ultrasound: CXR: Chest radiograph; CAP; Community-acquired pneumonia.

Most effusions (20 out of 26, 77%) were <2 cm in depth and were deemed not amenable to safe sampling by the operator. This is also in accordance with previously known data that PPEs measuring <1 cm on CXR [8] and 2.5 cm on CT [9] tend to resolve without need for drainage. 6 out of 26 (23%) patients had effusions more than 2 cm depth. Three out of 26 (11.5%) of these were sampled, of which two were proven to be pleural infection. All three of these effusions were visible on CXR at admission. Of the remaining three patients that did not have a diagnostic pleural aspiration, one died prior to sampling of fluid whilst another was discharged home without sampling (figure 1).

In total, four out of 41 (9.7%) patients died during admission. All four had effusions detected on TUS at admission, of which three did not have an effusion on the corresponding CXR. The mortality in those without effusions (none out of 15, 0%) compared to those with effusions (four out of 26, 15.4%) in this cohort was not statistically significant (χ²=2.6 with two degrees of freedom, p=0.11). One of these four patients was proven to have pleural infection and required further pleural intervention (chest drain insertion) for management. None of the patients were referred for surgical intervention. The median length of hospital stay for patients with and without PPE was 6.5 and 6.0 days respectively.

The evolution of pneumonia to pleural infection is thought to occur progressively in three stages: The first or “exudative” stage characterised by the development of an uncomplicated PPE; the second or “fibrinopurulent” stage wherein septations and loculations might develop along with formation of “complicated PPE”, or, in some cases, frank pus in the pleural space; and the third or “organising” stage where a “pleural peel” may form causing impaired lung function [7]. The natural history of this disease, however, varies between patients as some are more likely to develop multiloculated effusions whilst others develop frank pus, thereby calling its “linear” nature into question.

There are currently no validated clinical risk prediction tools that predict the development of PPE and pleural infection from pneumonia. A large prospective cohort study, which recruited 1269 patients, identified seven clinical factors that were independent predictors of development of pleural infection from CAP [3]. These include low serum albumin, low serum sodium, elevated platelet count, high C-reactive protein, history of alcohol abuse and intravenous drug use. Another study prospectively analysed 4715 patients with CAP and determined five clinical factors that could predict the development of pleural infection, namely young age (<60 years), alcoholism, pleuritic chest pain, tachycardia and leukocytosis [10]. Neither of these studies evaluated radiological features as potential predictive tools, and neither of the scores from these studies has been prospectively validated.

TUS is a readily available bedside radiological modality that is used in guiding safe sampling of pleural fluid and has a role in predicting a diagnosis of pleural infection prior to fluid sampling, with TUS features such as septations, loculations and echogenic swirling suggestive of the diagnosis [6, 11]. However, its role in predicting the development of and outcomes in pleural infection has not previously been studied. This pilot evaluation revealed a higher incidence of PPE than previously recorded in cohorts where TUS was not used as the diagnostic modality. The presence of PPE has been linked to poor outcomes [4], and therefore, using front-door TUS instead of CXR alone may have a role in prognostication in patients with CAP. In this cohort, we could not ascertain a correlation between TUS features such as presence of septations with the development of pleural infection. Moreover, no specific TUS characteristics were found to corelate with either resolution or progression to pleural infection.

Mortality in our cohort was higher in those with pleural effusion than those without, with the result not reaching statistical significance likely due to the small sample size. Previous studies have suggested that the presence of pleural effusion with a pneumonic illness, regardless of whether the fluid is frankly infected, is associated with a higher mortality [4]. The mechanism of this important finding requires further large-scale studies.

Conclusion

The incidence of PPE may be higher than previously estimated in cohorts where TUS was not used for initial assessment. Our pilot study did not identify TUS features that could predict the development of pleural infection from pneumonia. However, the TUS characteristics that were associated with resolution or progression to pleural infection are not well understood. A detailed prospective evaluation of the use of TUS in patients with pneumonia to characterise the natural history of PPEs and ascertain its role in predicting outcomes is required.